Island Connections


Material Information

Island Connections Mobile Phones and Social Change in Rural Fiji
Physical Description:
1 online resource (398 p.)
Peseckas, Ryan J
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Committee Chair:
Committee Co-Chair:
Committee Members:


Subjects / Keywords:
cellular -- communities -- community -- development -- exchange -- fiji -- internet -- islands -- kinship -- mobile -- network -- networks -- oceania -- phone -- remittance -- remittances -- telecommunications
Anthropology -- Dissertations, Academic -- UF
Anthropology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation


Mobile phones have become nearly ubiquitous around the world. Enhanced connectivity implies socioeconomic changes in communities that until recently had limited access to telecommunications. I conducted ethnographic research in three rural Fijian villages during 2011-2012 to document how inhabitants use mobile phones and internet, and the social implications of enhanced connectivity in the Fijian context. I focused on three domains: the configuration of calling networks, the use of mobile phones for economic exchange, and the role of mobile phones in romantic relationships. I collected data via participant observation, interviews and surveys. Nearly all households owned at least one phone. The majority received their phones as gifts. Inequalities in phone access are partially overcome through borrowing. Owners had a mean of 62+/-52 contacts. Phone contacts were overwhelmingly indigenous Fijian kin, but contact lists include individuals of diverse sex, age, and marital status. Patterns of phone use are heavily shaped by economic considerations and vary according to promotions. Phones are used extensively to arrange exchange of money and goods with urban kin. Phone ownership and use is positively associated with household income, but frequency of phone interaction is not associated with higher value or volume of exchange. This indicates that variation in phone access is primarily a symptom of economic inequality, rather than a cause. I assessed accuracy of responses about calling frequency by comparing 360 reciprocal response pairs. The intraclass correlation coefficient for these pairs was .471, indicating moderate agreement. Fijian youths use mobiles extensively to network with members of the opposite sex, and phones are also blamed for marital infidelity. Texting is used primarily for romantic relationships, and is more common among youths than adults. Overall my data indicate a shift toward geographically extensive, diffuse social networks, and intensified interaction between urban and rural community members in Fiji.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by Ryan J Peseckas.
Thesis (Ph.D.)--University of Florida, 2014.

Record Information

Source Institution:
Rights Management:
Applicable rights reserved.
lcc - LD1780 2014
System ID:

This item is only available as the following downloads:

Full Text




2 2014 Ryan Peseckas


3 To my parents who have given me love and unconditional support in all of my pursuits


4 ACKNOWLEDGMENTS I wish to acknowledge the a ssistance of a number of individuals, communities and organizations, without whose generous help and cooperation I could not have successfully completed this research. At the Unive rsity of Florida I thank first and foremost my committee chair, Gerald Murra y. Jerry took me on as a graduate student in 2006 despite my lack of undergraduate training in anthropology, and has offered helpful advice and encouragement at each step of the way Peter Collings served as a research and lik ewise has provided much advice, as well as assistance with reading dra fts and writing letters of reco mendation. I also wish to thank the members of my committee, C.K. Shih, Abdoulaye Kane, and Michael Bannister, for agreeing to oversee my p roject and for providing valuable input throughout Also at the University of Florida I am grateful to Christopher McCarty, who went out of his way to assist me with my data analysis, and Paul Richards, who supervised me as a teaching assistant and has be en very supportive in my search for post graduate employment. Finally, I thank my fellow graduate students at the University of Florida who have been my friends and colleagues during the 7 memorable years I have spent in Gainesville. I was assisted in my research by a great many people in Fiji. At the University of attachment to the Social Sciences Faculty at USP, and took care of various tasks related to my visa. I could not have undertaken my research without the kindness and patience of my Fijian hosts in Cagi Village, Toba Village, and Veidogo Village. In particular I am extremely thankful to the families that took care of me: Manu Vueti and his wife Makata, Tevita Suraki and his wife Elenoa, Seveci Ravetau and his wife Sala,


5 and Reverend Mosese Vakatale and his wife Raijeli. These families opened their homes to me, cooked delicious Fijian food, and always made me feel welc ome to stay as long as I liked. The rest of the inh abitants of Cagi Toba and Veidogo were equally welcoming and hospitable and I count many of them as friends Finally, I am very grateful for the o rganizations and individuals that generously funded my research. A preliminary field season in 2009 was mad e possible thanks to (TCD), as well a field research scholarship established by UF Professor Emeritus Paul Doug hty My fieldwork from July 2011 to 2012 was funded by a Ful bright Scholarship.


6 TABLE OF CONTENTS page ACKNO WLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ .......... 12 LIST OF FIGURES ................................ ................................ ................................ ........ 14 LIST OF ABBREV IATIONS ................................ ................................ ........................... 18 ABSTRACT ................................ ................................ ................................ ................... 19 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 21 Inspiration ................................ ................................ ................................ ............... 21 New Perspective on a Changed Fiji ................................ ................................ ........ 22 Mobile Revolution ................................ ................................ ................................ ... 24 The In formation Age Goes Global ................................ ................................ .......... 25 A Network Approach to Community ................................ ................................ ........ 27 Culture and Identity ................................ ................................ ................................ 29 Technology and Socio cultural Change ................................ ................................ .. 31 ICTs, Access, and the Communicative Ecology ................................ ..................... 33 Summary ................................ ................................ ................................ ................ 37 2 RESEARCH QUESTIONS, METHODS, AND SITE SELECTION .......................... 41 Overview ................................ ................................ ................................ ................. 41 Resea rch Questions ................................ ................................ ............................... 41 Site Selection ................................ ................................ ................................ .......... 42 Study Population ................................ ................................ ................................ ..... 43 Data Coll ection ................................ ................................ ................................ ....... 44 Participant observation ................................ ................................ ..................... 44 Interviews ................................ ................................ ................................ ......... 45 Copying local records ................................ ................................ ....................... 47 Surveys ................................ ................................ ................................ ............ 47 Household surveys ................................ ................................ .................... 47 Phone surveys ................................ ................................ ........................... 48 Survey logistics and compensation ................................ ............................ 48 Equipment ................................ ................................ ................................ ........ 49 Hardware ................................ ................................ ................................ ... 49 Software ................................ ................................ ................................ ..... 49 Data Analysis ................................ ................................ ................................ .......... 49 Reporting of Results ................................ ................................ ............................... 50


7 Reporting of Statistics ................................ ................................ ...................... 50 ................................ ................................ .... 50 Anonymity of Par ticipants ................................ ................................ ................. 50 3 RESEARCH SETTING ................................ ................................ ........................... 52 Overview ................................ ................................ ................................ ................. 52 Fiji ................................ ................................ ................................ ........................... 52 Geography and Environment ................................ ................................ ............ 53 Economy ................................ ................................ ................................ .......... 54 Demography ................................ ................................ ................................ ..... 55 History ................................ ................................ ................................ .............. 56 Western contact ................................ ................................ ......................... 56 Colonialism 1874 1970 ................................ ................................ .............. 5 7 Independence, political strife, and the uncertain future .............................. 58 Fijian Culture / Itovo Vakaviti ................................ ................................ ............ 60 Telecommunications in Fiji ................................ ................................ ............... 64 Colonial telecommunication developments ................................ ................ 64 Radio and television ................................ ................................ ................... 67 Mobile phones ................................ ................................ ............................ 68 Internet ................................ ................................ ................................ ....... 71 The Communicative Ecology of a Rural Fijian Community ................................ ..... 72 The Village and the Community ................................ ................................ ....... 73 Characteristics of the Study Communities ................................ ........................ 78 Aspects of Fijian Kinship ................................ ................................ .................. 85 Kinship, Communication and Etiquette ................................ ............................. 87 Social Structure ................................ ................................ ................................ 96 Leadership ................................ ................................ ................................ ...... 103 Implications of Social Structure for Communication ................................ ....... 107 Overlapping constraints ................................ ................................ ........... 107 Crossing social boundaries ................................ ................................ ...... 110 Mobile phones and transcending barriers to communication ................... 111 Technological and Material Aspects of the Communicative Ecology .............. 114 Telecommunications ................................ ................................ ................ 115 Transpo rt ................................ ................................ ................................ .. 117 Energy ................................ ................................ ................................ ..... 121 Predictions for Mobile Use in the Study Communities ................................ .... 122 4 PHONE OWNERSHIP AND USE IN RURAL FIJI ................................ ................ 124 Overview ................................ ................................ ................................ ............... 124 Distribution of Phones in Rural Fijian Communities ................................ .............. 125 Telecom Technology in Rural Fiji ................................ ................................ ... 125 Distribution of Phones among Households ................................ ..................... 129 Distribution of Mobile Phones among Age Groups ................................ ......... 131 Distribution of Phones by Sex ................................ ................................ ........ 133 Phone Ownership ................................ ................................ ................................ 136


8 Acquiring a Phone ................................ ................................ .......................... 136 Gifts ................................ ................................ ................................ ......... 137 Purchases ................................ ................................ ................................ 139 Other means of obtaining phones ................................ ............................ 139 Borrowing and Sharing ................................ ................................ ................... 140 Sharing within the household ................................ ................................ ... 140 Borrowing outside the household ................................ ............................. 141 Life Cycle of a Phone ................................ ................................ ..................... 143 T ime between Phones ................................ ................................ .................... 147 Phonelessness ................................ ................................ ............................... 149 Aesthetic and Symbolic Aspects of Phone Ownership ................................ .......... 151 Ownership vs. Non ownership ................................ ................................ ........ 152 Handset Model ................................ ................................ ............................... 154 Personalization of the Handset ................................ ................................ ....... 156 Carrying and Displaying the Phone ................................ ................................ 158 Using the Phone ................................ ................................ ................................ ... 159 Voice Communicatio n ................................ ................................ .................... 160 Phone conversation: purpose, content and style ................................ ..... 162 Cost considerations ................................ ................................ ................. 170 Special purpose Calls ................................ ................................ .................... 172 Emergencies ................................ ................................ ............................ 172 Health ................................ ................................ ................................ ...... 174 Crime ................................ ................................ ................................ ....... 176 Pranks ................................ ................................ ................................ ...... 177 Text ................................ ................................ ................................ ................ 178 Internet ................................ ................................ ................................ ........... 181 Secondary Uses of Mobile Phones ................................ ................................ 184 Clock ................................ ................................ ................................ ........ 185 Radio and music ................................ ................................ ...................... 186 Games ................................ ................................ ................................ ..... 188 Other uses ................................ ................................ ............................... 189 Summary ................................ ................................ ................................ .............. 189 5 PATTERNS OF TELEPHONE COMMUNICATION IN RURAL FIJI ..................... 191 Introduction ................................ ................................ ................................ ........... 191 Data and Analysis ................................ ................................ ................................ 192 Profiles of Village Phone Users ................................ ................................ ............ 192 Peceli: A Call to Service ................................ ................................ ................. 192 Elisape ci: Reluctant Romantic ................................ ................................ ........ 195 Tevita: No nonsense ................................ ................................ ...................... 197 Etika: Relationships Up in Smoke ................................ ................................ .. 199 Talica: A Crutch for Old Age ................................ ................................ ........... 200 Analyses of Phone Survey Data ................................ ................................ ........... 201 Characteristics of Phone Owners ................................ ................................ ......... 201 Frequency/Intensity of Phone Use ................................ ................................ ........ 204 Calling and Texting Frequency ................................ ................................ ....... 205


9 Amount of Phone Credit Used ................................ ................................ ........ 208 ................................ .......... 211 Saved in ................................ ................................ ................................ ...... 214 Individuals Identified as Heavy Phone Users in Surveys ................................ 216 Relationships Among Calling F requency Variables ................................ ........ 217 Relationships between Demographic Subgroups and Calling Frequency Variables ................................ ................................ ................................ ..... 217 Ranking Individuals Based on a Composite Measure of Phone Use Frequency/Intensity ................................ ................................ ..................... 219 Characteristics of Phone Contacts ................................ ................................ ........ 221 An Overview of the Phone Cont acts Data ................................ ...................... 221 Sex ................................ ................................ ................................ ................. 226 Marital Status ................................ ................................ ................................ 229 Ethnicity/Native Villag e ................................ ................................ ................... 232 Occupation ................................ ................................ ................................ ..... 233 Relationships between Phone Owners and Contacts ................................ ........... 2 36 Friends ................................ ................................ ................................ ..... 236 Business relations ................................ ................................ .................... 237 Kin ................................ ................................ ................................ ............ 238 Temporal and Spatial Aspects of Phone Communication ................................ ..... 244 Comparison of Community Calling Networks ................................ ........................ 250 Summary ................................ ................................ ................................ .............. 254 Community Level Differences ................................ ................................ ............... 254 Subgroup Level ................................ ................................ ................................ ..... 255 Individual Level ................................ ................................ ................................ ..... 256 6 MOBILE PHONES, EXCHANGE AND INEQUALITY WITHIN FIJIAN COMMUNITIES ................................ ................................ ................................ .... 258 Overview: Economic Benefits of Mobile Phones ................................ ................... 258 The Rural Fijian Economy ................................ ................................ ..................... 260 The Household Economy ................................ ................................ ............... 261 Land ................................ ................................ ................................ ............... 261 Communal Obligations ................................ ................................ ................... 263 Sources of Household Income in Rural Fiji ................................ ..................... 266 Farming for subsiste nce consumption and for the market ....................... 267 Handicrafts ................................ ................................ ............................... 269 Small business ................................ ................................ ......................... 273 Government assistance and pensions ................................ ..................... 275 Household Income in Toba, Cagi and Veidogo ................................ .............. 275 Sources of income in the study v illages ................................ ................... 277 Income inequality between village households ................................ ........ 279 Household income, lifestyle, and phone use ................................ ............ 282 Kinship Exchange and Remittances ................................ ................................ ..... 284 Exchange within the Village ................................ ................................ ............ 286 Exchange outsid e the Village ................................ ................................ ......... 288


10 Local exchange ................................ ................................ ........................ 289 Exchange with partners in distant areas of Fiji ................................ ......... 290 Overseas exchange ................................ ................................ ................. 295 Balance of reciprocity ................................ ................................ ............... 296 Characteristics of exchange partners ................................ ....................... 297 Comparing exchange in Toba, Cagi and Veidogo ................................ .... 298 Phone Use and Exchange ................................ ................................ .................... 300 E xchange and Inequality among Households within Rural Fijian Communities .... 303 Summary ................................ ................................ ................................ .............. 304 7 TECHNOLOGY, LOVE AND MARRIAGE IN RURAL FIJI ................................ .... 307 Overview ................................ ................................ ................................ ............... 307 Methods ................................ ................................ ................................ ................ 309 Opportunity ................................ ................................ ................................ ........... 310 Childhood and Youth ................................ ................................ ...................... 311 ................................ ................................ ... 313 No Such Thing as a Wrong Number ................................ ............................... 315 Mobiles and Mobility ................................ ................................ ....................... 316 Spy Games at the Suva Bus Stand ................................ ................................ 318 Courtship ................................ ................................ ................................ ........ 319 Mobiles and Elopement: A Means of Escape? ................................ ............... 321 The Emergence of Internet ................................ ................................ ............. 323 Deception ................................ ................................ ................................ .............. 324 Expectations and Reality ................................ ................................ ................ 324 Mobiles and Marital Strife ................................ ................................ ............... 327 Impacts on Family and Community Life ................................ ................................ 329 Norms and Realities Surrounding Sex in Fiji Today ................................ ....... 330 Love and Marriage ................................ ................................ ......................... 331 Separation and Divorce ................................ ................................ .................. 335 Summary ................................ ................................ ................................ .............. 336 8 CONCLUSIONS ................................ ................................ ................................ ... 338 Economics ................................ ................................ ................................ ............ 338 Social life ................................ ................................ ................................ .............. 340 I deology ................................ ................................ ................................ ................ 343 APPENDIX A PRONUNCIATION OF FIJIAN WORDS ................................ ............................... 347 B COPIES OF SURVEY FORMS ................................ ................................ ............. 348 Household Survey Questions ................................ ................................ ............... 348 Phone Survey Questions ................................ ................................ ...................... 354 SIM Survey Questions ................................ ................................ .......................... 358


11 C COLLECTING DATA FROM SIM CARDS ................................ ............................ 360 SIM Cards and SIM Card Readers ................................ ................................ ....... 360 Copying SIM Da ta in the Field ................................ ................................ .............. 362 Data Management ................................ ................................ ................................ 364 The Scientific Value of Phone Contact Lists ................................ ......................... 367 Logistical Considerations ................................ ................................ ................ 367 Methodological Considerations ................................ ................................ ...... 368 Phones span multiple levels of analysis ................................ ................... 368 Phone directories are not a complete record of calling activity ................ 371 Inaccuracy of recall ................................ ................................ .................. 373 Assessing informant accuracy about frequency of phone interactions ..... 375 An alternative approach to testing response accuracy ............................. 379 Ethical Considerations ................................ ................................ .................... 380 D ANALYSES OF CALLING NETWORKS ................................ ............................... 383 Phone Use Frequency/Intensity ................................ ................................ ............ 383 Contents of Phone Contact Lists ................................ ................................ .......... 384 E CALCULATIONS OF THE MONETARY VALUE OF EXCHANGED GOODS ...... 385 LIST OF REFERENCES ................................ ................................ ............................. 387 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 397


12 LIST OF TABLES Table page 3 1 A summary of the basic characteristics of the three study villages ..................... 83 4 1 Percentage of village populations owning mobiles, Easytels, and SIMs ........... 128 4 2 Mean phones per household in the study villages ................................ ............ 130 4 3 Mean number of months between loss of previous phone and acquisition of current phone ................................ ................................ ................................ ... 148 4 4 Domestic calling and texting prices for Fiji mobile providers (March 2012) ...... 170 5 1 Correlations between phone use frequen cy variables ................................ ...... 217 5 2 Subgroups and phone use frequency/intensity variables ................................ 218 5 3 Analysis of differences in calling network co mposition for individuals in various social subgroups ................................ ................................ .................. 225 5 4 Chi square test for differences in calling network composition for groups of different marital status ................................ ................................ ...................... 230 5 5 Characteristics of community calling networks ................................ ................. 252 5 6 Aspects of community networks ................................ ................................ ....... 253 6 1 Mean household income over the past 12 months in the study villages ........... 276 6 2 Gini coefficients for household income (without exchange) in study villages .... 279 6 3 Correlations between household income and socioeconomic variables ........... 282 6 4 Correlations between socioeconomic variables and phone use variables ........ 283 6 5 Details of exchanges ................................ ................................ ........................ 289 6 6 Median test of significance of differences of exchange volume by village ........ 299 6 7 Correlations between phone use and income/exchange variables ................... 300 6 8 Correlations between calling frequency and exchange variables ..................... 302 6 9 Comparing inter household income inequality with and without exchanges ..... 304 C 1 .CSV file ................................ ................................ ................................ ........... 365


13 C 2 Survey spreadsheet (simplified) ................................ ................................ ....... 366 C 3 Community directory spreadsheet (simplified) ................................ .................. 366 C 4 List of al l phone interactions (simplified) ................................ ........................... 366 C 5 Example of an entry removed from analysis of respondent accuracy ............... 377 C 6 Calculating d ivergences in responses about calling frequency ........................ 377 D 1 ....................... 384


14 LIST OF FIGURES Figure page 1 1 Telecom development in Fiji 2000 2012 ................................ ............................. 26 3 1 Regional map with Fiji highlighted ................................ ................................ ...... 54 3 2 A Cagi villager poses with an old phone. ................................ ............................ 66 3 3 The entrance to Korovou Prison in Suva, with Vodafone advertisement. Photo courtesy of author. ................................ ................................ ................... 69 3 4 Satellite photograph of a Fijian village (Image DigitalGlobe, Google 2013) .... 74 3 5 Map of Fiji with study villages labelled. Copyright Daniel Dalet. Labels added by author. ht tp://d ..................... 79 3 6 Population tree diagrams for Toba, Cagi, and Veidogo villages ......................... 84 3 7 Overhe ad diagram of a Fijian house ................................ ................................ ... 90 3 8 Members of a patrilineage at a Christmas meal. ................................ ................ 91 3 9 Fijian social structure (simplified di agram) ................................ .......................... 98 3 10 of author. ................................ ................................ ................................ .......... 105 3 11 A group of men gat her for an afternoon kava session. Photo courtesy of author. ................................ ................................ ................................ .............. 110 4 1 ....... 128 4 2 Number of phones per household, by village ................................ ................... 130 4 3 Age distribution of mobile phone owners vs. overall age distribution ................ 131 4 4 Percentage of each age cohort owning mobile phones ................................ .... 132 4 5 Sex of mobile phone owners in the study communities ................................ .... 134 4 6 Relationships of mobile phone owners to the heads of their households ......... 135 4 7 How respondents obtained their mobile phones ................................ ............... 136 4 8 Sources of gifted phones ................................ ................................ .................. 137 4 9 Number of months owning current mobile phone (n=146) ................................ 144


15 4 10 What happened to your last mobile phone? (n=109) ................................ ........ 145 4 11 How many mobile phones have you owned before your current phone? ......... 145 4 12 Frequency of secondary uses of mobiles ................................ ......................... 185 4 13 Mobile phone with plastic bottle used as an amplifier for music. Photo courtesy of author. ................................ ................................ ............................ 187 4 14 Phone with wires installed, so that it can be hooked up to stereo speakers. .... 188 5 1 A young woman takes a break from housework to chat on the phone. Photo courtesy of author. ................................ ................................ ............................ 196 5 2 Characteristics of phone owners in the study villages ................................ ...... 202 5 3 Age distribution of mobile phone owners. Percentages indicate what p ercentage of each age cohort in the study communities own mobile phones. 202 5 4 Number of calls and texts on a typical day ................................ ....................... 206 5 5 Mean calls and texts per day by subgroup (error bars= +/ SE) ....................... 206 5 6 Mean calls and texts per day, by village (error bars= +/ SE) ........................... 207 5 7 Money spent on phone credit during a typical week (error bar= +/ SE) ............ 209 5 8 Money spent on phone credit during a typical week (villages) .......................... 210 5 9 Scatterplot of number of saved phone contacts, by sex and age ..................... 212 5 10 Mean number of saved phone contacts by subgroup (error bars= +/ SE) ........ 212 5 11 number was saved ................................ ................................ ........................... 215 5 13 Composite ranks for phone use frequ ency Cagi ................................ .............. 220 5 14 Composite ranks for phone use frequency Veidogo ................................ ........ 220 5 15 Characteristics of all phone contacts and ties. ................................ ................. 222 5 16 Characteristics of individuals found in phones ................................ .................. 223 5 17 Characteristics of phone based relationships ................................ ................... 224 5 18 Sex of phone owners and phone contacts ................................ ........................ 227 5 19 Age/sex of phone owner vs. sex of phone contacts ................................ .......... 228


16 5 20 Calling frequency between same sex and opposite sex individuals ................. 228 5 21 Marital status of phone owners and contacts ................................ ................... 230 5 22 Age of phone owner vs. marital status of phone contacts ................................ 231 5 23 Ethnicity of phone contacts by village ................................ ............................... 232 5 24 Mean number of village phones within which contacts from different occupational groups were found (non village dwellers only) ............................ 234 5 25 Calling frequency to contacts of different occupational statuses ...................... 235 5 26 Age of each phone owner and the percentage of their contacts who are related to them as friends (excluding owners with <10 contacts) ..................... 237 5 27 Proportions of kin types in Fijian communities. ................................ ................. 239 5 28 Kin relationships of phone contacts ................................ ................................ .. 240 5 29 Calling frequencies between parallel and cross kin in the same generation. ... 242 5 30 Geographic location of phone contacts by village ................................ ............. 245 5 31 Mean number of in ties per node by geographic location ................................ 246 5 32 Location of phone contacts vs. calling frequency ................................ ............. 247 5 33 Calling frequency vs. meeting frequency ................................ .......................... 248 5 34 Percentage of contacts living in same village as phone owner ......................... 249 5 35 Scree plot of proportion of phone contacts saved in proportion of each ................................ ................................ ......................... 253 6 1 Toba village men form a bucket line to clean out the village bathing pool du ring a communal work day. Photo courtesy of author. ................................ .. 264 6 2 Toba men working together to cut copra. Photo courtesy of author. ................ 268 6 3 Cagi women working together to plait a pandanus mat ( ibe ). Photo courtesy of author. ................................ ................................ ................................ .......... 27 0 6 4 Box and whisker plot of household income by village ................................ ....... 276 6 5 ................................ ................................ ............... 278 6 6 Lorenz curves for study villages for comparison. ................................ .............. 280


17 6 7 Excha nges in the study villages, by percentage of total monetary value .......... 289 6 8 The men of a Cagi itokatoka load cassava and taro into bags. ........................ 292 6 9 Relationships between exchange partners ................................ ....................... 297 6 10 Number of exchange partners and exchange events per household ................ 298 6 11 Value of exchange events and total income from exchange ............................ 299 6 12 Net overall household income and net household income from exchanges ..... 299 6 13 Comparison of calling frequency for exchange and non exchange relationships ................................ ................................ ................................ ..... 302 C 1 SIM card and mobile phone. ................................ ................................ ............. 361 C 2 Consistency of responses about calling frequency: ................................ .......... 378 C 3 Inconsistency between responses about calling frequency. ............................. 379 D 1 Scatterplot of calls per day, by age and sex of phone owner ........................... 383 D 2 Scatterplot of texts per day, by age and sex of phone owner ........................... 383 D 3 Scatterplot of weekly phone credit expenditures by age and sex ..................... 384 D 4 Percentage of respondents who named each individual as a heavy phone user ................................ ................................ ................................ .................. 384


18 LIST OF ABBREVIATIONS 2G 2 nd Generation mobile technology. This standard switched to digital, as opposed to the analog 1G phones. 3G 3 rd Generation mobile technology. Features mobile broadband of at least 144 kbps. 4G 4 th Generation mobile tec hnology. Standards for 4G have not been set, but the best known variety is LTE. $US United States Dollar, the unit of currency used in the U.S. $FJ Fiji Dollar, the unit of currency used in Fiji. Worth $US 0.54 dollars on 12/31/2011, roughly the midpoint o f my fieldwork. CDMA Code Division Multiple Access EDGE An upgraded version of 2G network technology GSM Global System for Mobiles: a standard for 2G mobile networks that has become the dominant system in most countries of the world. ICT Informatio n and c ommunication technology Kg kilograms, a unit of mass Km kilometers, a unit of distance MVNO Mobile Virtual Network Operator: a mobile service provider that does not have its own network infrastructure, instead leasing access from another provider (e.g. I nkk Mobile) SIM Subscriber Identity Module: the chip in mobile phones that identifies subscribers on a mobile network as well as holding phone contacts and other saved information UMTS Universal Mobile Telecommunications System, a 3 rd generation mobile sy stem for GSM networks.


19 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy ISLAND CONNECTIONS: MOBILE PHONES AND SOCI AL CHANGE IN RURAL FIJI By Ryan Peseckas May 2014 Chair: Gerald Murray Major: Anthropology M obile phones have become nearly ubiquitous around the world. Enhanced connectivity implies socioeconomic changes in communities that until recently had limited a ccess to telecommunications. I conducted ethnographic research in three rural Fijian villages during 2011 2012 to document how inhabitants use mobile phones and internet, and the social implications of enhanced connectivity in the Fijian context I focused on three domains: the configuration of calling networks, the use of mobile phones for economic exchange and the role of mobile phones in romantic relationships. I collected data via participant observation, interviews and surveys Nearly all households owned a t least one phone The majority received their phones as gifts. Inequalities in phone access are part ial ly overcome through borrowing. Owners had a mean of 62 +/ 52 contacts Phone contacts were overwhelmingly indigenou s Fijian kin, but contact lists include individuals of diverse sex, age, and marital st atus. P atterns of phone use are heavily shaped by economic considerations and vary according to promotions. Phones are used extensively to arrange exchange of mone y and goods with urban kin P hone own ership and use is positively associated with household income, but frequency of phone interaction is not associated with higher


20 value or volume of exchange This indicates that variation in phone access is primarily a symptom of economic inequality, rather than a cause. I assessed accuracy of responses about calling frequency by comparing 360 reciprocal response pairs. The intraclass correlation coefficient for these pairs was .471, indicating moderate agreement. Fijian youths use mobiles extensively to net work with members of the opposite sex and p hones are also blamed for marital infidelity. Texting is used primarily for romantic relationships, and is more common among youths than adults. Overall m y data indicate a shift toward geographically extensive, d iffuse social networks, and intensified interaction between urban and rural community members in Fiji


21 CHAPTER 1 INTRODUCTION Inspiration The call ended abruptly. I was cut off awkwardly in mid sentence as my $10 phonecard expired, without even th e chance to say goodbye. I hung up the receiver of the roadside payphone with a feeling of helplessness My clothes were damp with sea spray and the pavement seemed to roll beneath my feet like an ocean swell. I tried not to cry, but the tears came anyway and dissolved the salt that encrusted my eyelashes, causing my eyes to burn uncontrollab ly. I winced as a bus lurched by, belching exhaust and blaring a remix of a Hindi pop song. With the roar of traffic around me, I had barely been able to understand what my girlfriend was saying as she broke up with me over the phone. What a way for a relationship to end. It was April, after a wave tossed 3 hour journey in a small open boat. I had spent the past six weeks working on a tiny i sland called Mamaca one of the more than 300 islands in the Fiji archipelago. As a Peace Corps volunteer, my assignment to promote marine resource management had brought me to many out of the way vi llages, where transport and tele phone communication were inconvenient, unpredictable and costly. On Mamaca the radio tele phone had been disconnected by Telecom Fiji more than a year previously. One of the local men, responsible for colle cting payment from villagers for their use of the phone, had embezzled the money and the bill r emained unpaid. Because of this I had had to journey to a neighboring island by boat each time I wanted to make a call. As pathetic as it sounds, I only


22 managed to make contact once with my girlfriend in the United States during the ent ire six weeks I was on Mamaca Frustrated and disheartened, I slogged down the street toward the Suva bus station. In the subs equent days I was able to reflect more philoso phically on my A more long distance Th e unravelling of my relationship with an ex girlfriend may se em like a frivolous introductio n to a study of mobile phone use and social change However, it is the memory of first hand experiences like these, as a Peace Corps volunteer i n rural Fiji from 2003 to 2005, that gi ve me faith in the fundamental importance of my research. O ur access, or lack of access to telephones and other communication technologies has real materi al and social consequences; a single phone call can literally alter the course life. This story is just an isolated incident from my past brought about through a con flue nce of unusual events I n rural Fiji however the pervasive difficulties of long distance communication have fundamentally shape d social horizons and life aspirations for millenia This dissertation is an exploration of the adaptations that occur in a socio cultural system when constraints on communication undergo rapid and radical change. New Perspective on a Changed Fiji I returned to Fiji as a PhD student in the summer of 2009 to explore a research project dealing with rural urban migrati on. Just 6 months previously, Digicel, a provided a competitive alternative to Vodafone, the longstanding incumbent service provider, thus putting an end to the monopoli stic arrangements that had for years stifled


23 a revelation to me mobile phone service had finally come to the villages, and rural Fiji seemed like a different place than I had experienced just 5 years previously. In light of these exciting developments in telecommunications, I reevaluated the goals of my research project. As an admirer of the work of Marvin Harris and other Cultural Materialists, I had embarked on my trip that summer mentally preparing myself to measure c oconut groves and count calories. After all, a key premise of Cultural Materialism is that the material base of society tends to determine social structure and ideology, rather than the other way around. In stead of a study focusing on the rural economy, I became fascinated by a more ethereal question: What role does communication play in community life, and through what processes does access to communication technologies lead to changes in the material, soci al, and ideological domains of a socio cultural system? I returned to graduate school at the University of Florida that f all and came up with a proposal to study the relationships between mobile phone use and social change in Fijian villages. I returned t o Fiji for a third time in May 2011 and spent a year conducting an ethnographic study of mobile phone use in three vil lages Toba, Cagi, and Veidogo. I focused particular attention on the role of mobile communication in three domains: the formation and mai distance exchange of material resources, and romantic relationships. This dissertation is the culmination of that research project. In this introductory chapter I discuss the theoretical issues underlying my research project and the approaches that I adopted. I begin with a brief description of


24 the global proliferation of mobile technology. Next, I discuss concepts of community, and two ways of understanding culture. I then present a causal model that relat es technological change to changes in other socio cultural domains. In the final section of the introduction, I outline the chapters of this dissertation, and describe the ways in which I hope that my research can contribute to the discipline of anthropolo gy. M obile Revolution In recent years, many of the most salient changes to our lifestyles have stem med from the proliferation of information and communication technologies (ICTs ). The technologies through which we communicate influence the content of our m essages the manner in which our messages propagate, and the networks of people with whom we interact. The cumulative impact at the societal level is that new communication media imply cultural change (McLuhan 1962). We increasingly use ICTs in all aspec ts of our lives: to access and exchange information to work and play together and to share and consume ideas, images, art, and entertainment We surround ourselves on a daily basis with televisions, radios, mp3 players, mobile phones tablets, and laptop s and spend a large portion of our time text messaging surfing the internet and checking and updating our various online social networks. As ICTs become incr easingly ubiquitous, so too are they capable of an ever increasing array of functions. A basic s martphone can be used for calling and texting; it serves as a clock, a video and still camera, a calculator, a GPS navigator, a calendar, a radio and music player, a gaming platform, not to mention the myriad uses and possibilities that come with internet access: search engines, social networking, and entertainm ent, just to name a few. Mobile phones have become one of the three items


25 that people carr y on their bodies at all times. The other indispenable tools of modern life are a wallet and keys alt hough m obiles may soon usurp their functions too. We say that we live in an Information Age. T his trope is not just descriptive; the fact that we self identi fy as citizens of a networked society affects the way we think, speak, and act and the way we approach t echnology Our discourse is filled with references to ho w interconnected we have become. S ome times our techno integratedness makes us feel smug and sophisticated at other times overwhelmed and paranoid. Some among us are optimistic increa singly connected future : p erhaps technology will lead us t o greater understanding, allow us to transcend discrimination and violence, or even enable us to tap into a universal human consciousness. O ther s despair that with our increased focus on communicati on devices, we have neglected our families, neighborhoods, and other local groups, and for gotten how to have to face social interactions. There is a fear that by plugging in to the network, we surrender our individuality, and expose ourse lves to surveillance, abuse and control by governments or other powerful interests. The Information Age Goes Global In the developed world, where telephones have been commonplace for more than a century, we tend to take instantaneous, long distance commun ication for granted. In remote and developing regions, however, mobile phones are a recent, revolutionary development with far reaching, and only partially realized social and economic consequences. Had this brie f introduction been written a decade ago, it might have seemed t riumphalistic, overblown, or willfully ignorant of global technological disparities. After all, t he 24/7 connectedness already enjoyed by Japanese teenagers or American business


26 executives had l ittle bearing on the lives of farmers in t he Sahel, poor who probably lacked electricity and landline phone service, let alone mobile phones and internet access. These g eographic or socioeconomic disparities in access to information technologies referred to as the Digital Di vide have meant that the impacts of recent developments in ICT have been very uneven across regions and across socioecono mic strata. However, this state of affairs has changed considerably in the past decade with the rapid proliferation of mobile telepho ne access in developing countries In 2005, the global mobile penetration rate was only 34%, and only 23% in developing countries (ITU 2013). Thanks partly to universal access policies, and partly to the liberalization of telecom markets, by 2011 90 % of th e world population lived in areas with access to 2G mobile service and 3G service is now being extended to a growing proportion of these areas There are currently an estimated 6.8 billion mobile subscriptions worldwide, and mobile penetration rates in developing countries have reached 89%. In 2014, the number of mobile phones will surpass the human population Figure 1 1. Telecom development in Fiji 2000 2012


27 broad global trends. Mobile network co verage in Fiji remained limited through the early 2000s, in part due to the subscription rates finally exploded in 2008 with the liberalization of the Fijian telecom market, and the entry of Digicel as a competitor for the incumbent Vodafone. Liberalization was accompanied by expansion of network access to rural areas and outer islands, and by falling service prices. Today, the Fijian mobile m arke t is approaching saturation and increasing numb ers of customers are beginning to take advantage of the next generation of connectivity: mobile based internet. Technologies are intertwined in complex ways with social processes, such that a drastic technological change often is associated with socio cul tural changes. The remainder of this chapter discusses mechanisms through which the rapid global proliferation of ICTs, such as mobile phones and internet, might affect the structure and functioning of social groups A Network Approach to Community Severa l stu dies have focused on the implications of mobile phone use for human soc ial interaction and for economic development ( e. g. Ling 2004; Castells 2007; Donner 2008 ; and Katz 2008) Naturally, most of the early mobile phone research focused on the urban, d eveloped contexts in which mobiles were first adopted More recently, emphasis has shifted to the developing and/or rural regions which have gained mobile access during the past few years This technological transition (and transition in academic focus ) i s of interest to anthropologists because small, rural communities, often described particular domain of study since the


28 birth of our discipline A long standing assumption has bee n that traditional communities, characterized by dense, close knit, locally bounded social networks, differ in fundamental ways from the diffus e, loosely knit communities of modern industrialized societies. Accord ingly, researchers in the past have often r estricted their studies of communities to the interactions observed within village s neighborhood s or towns (Wellman 1999). pointed out, the boundaries of communities are de fined by social relationships, whereas neighborhoods (or villages) are defined by geographic boundaries (1999). In other words, from the network perspective, communities are really a series of overlapping c boundaries and the confines of social groupings such as family, lineage, or clan Thus, my research sites are Fijian villages, but I use the contents of telephone contact lists to study the configuration of geographically scattered networks that make up the socially defined communities of the individuals who reside within the villages. Historical research suggests that these geographically scattered ties have in fact industrial so cieties. Nevertheless, t he introduction of ICT s greatly facilitates the proli feration and maintenance of long distance social ties and broad, diffuse networks Rainie and Wellman refer to this process as a transition to networked individualism (2012). With technological and social conditions in rapid flux, anthropologists could benefit from the adoption of t his network perspective to analyze the evolving relationship between local groups and broader global processes.


29 Culture and I dentity Individual l networks are a key unit of analysis in my research, but I recognize the overarching importance of culture i n shaping these networks. By culture, I mean the socially learned complex of ideologies and behaviors common among a group of people. Two divergent approaches to culture provide a starting point for an investigation of ICTs and culture change : diffusionism and boundary maintenance D iffusionism was a dominant paradigm during the early decades of academic anthropology Its main premise, which has been discredited as overly simplistic is that differences betwe en cultural groups are due to mutual isolation while similarities are due to historical episodes of interaction The logic of the diffusionist approach is as follows: i n isolation, separate group s develop divergent adaptations t o their particular environments. Subsequent c ontact between different groups leads to the exchange of ideas and customs a process known as cultural diffusion. A shared cultural trait, such as a certain style of pottery, fo und among two different groups most likely indicates that sometime in the past, these groups came into contact. In reality many institutions, such as bride price or the incest taboo, have been developed independently by multiple societies. When cultural d iffusion does occur, p ower dyna mics influence the balance of cultural exchanges the common experience of recent centuries being that small scale societies tend to succumb to the domination of the market and the state. The result of contact between groups can therefore be considered as a form of acculturation, or erosion of cultural attributes. Rapid acculturation was the primary rationale for so of the early 20 th century ; urgent studies of small scale societies carried out on t he assumption that increasing global integration s pel led the beginning of the end of cultural variation If


30 global integration means the death of cultural difference, then perhaps the mobile phone and other information technologies are the coup de grace F rom the diffusionist standpoint ICT access will inevitably accelerate the penetration of global flows in to the remote c ommunities, leading to the establishment of a universal mono culture. A second theoretical approach to culture ethnic bou ndary maintenance, assert s that we generate cultural difference s as part of the process of group identity formation (Barth 1969) The need to draw line s between is a timeless human imperative Even w here no appreciable environmental, behavi oral, or ideological difference s exist the m embers of separate groups invent boundaries by which they may distinguish themselves from one another From this standpoint, mobile phones and other ICTs do not spell the end of cultural diversity. Indeed, ICTs are just one more medium through which novel kinds of cultural difference are expressed; users might project their group orientation through unique styles of phone ownership, use, or ornamentation. M obile and internet b ased communication networks could f orm the basis for new virtual communities, dispersed in space but tied together by shared interests, identitie s, beliefs, ways of communicating and thinking From this perspective, cultural diversity is not threatened by technology; rather, technology is j ust one more avenue through which diversity will be channeled and expressed in the future In fact, there are concerns that internet users tend to seek out like minded individuals online, and mostly consume media that confirm their pre existing biases. Thi s potential for ideological segregation of the internet is sometimes


31 Neither diffusionism nor ethnic boundary maintenance can fully explain the nature of culture, but each e ncompasses an important facet of the process of culture change. In Chapter 3 I describe Fijian culture, and how aspects of diffusionist and boundary maintenance theories can be particularly useful for understanding change in the Fijian context. However, it is necessary first to establish the particular role of technology in socio cultural change. Technology and S ocio cultural C hange Social scientists have long suspected that technology plays an important role in socio cultural change. Indeed, s ome promine nt anthropologists, such as Leslie White, have assigned technology the status of a prime mover in cultural processes (White 1959 ) In the 1980s a n academic crit ique emerged regarding the deterministic role that technology had played in the social scienc e literature up to that point. As Pfaffenberger described it, many social scientists had a master narrative of modern culture, which sees history in terms mastery over nature (1992) and t echnological development as a natural and inevitable process Part of the furor over technological determinism focus ed on indelicate turns of phrase in particular, the matter of fact discussion of technological change i n terms of its ch seems to imply that technology is an autonomous entity divorced from society. The debates over technology and social change mirror broader late 20 th century debates in the social sciences over structure and agency ( Ortner 1984; Rasmussen 2000). Structu ral functionalist approaches, like technological determinist approaches, treat social structures as sui generis having real causal efficacy, whereas in theories of


32 practice structures are only internalized patterns or suggestions for action ; for example, habitus memory traces (Bordieu 1977; Giddens 1984). Out of the critique of determinism a number of more balanced relationist approaches have been developed which emphas ize that the de sign and use of technology is itself heavi ly shaped by economic, social, and ideological fac tors. Among these critical approaches are the Social Shaping of Technology (SST) (MacKenzie and Wajc man 1985 ) the Social Construction of Technology ( SCOT) (Bijker, et al. 1987) and Actor Network Theory (L atour 2005) This dissertation focuses on processes of socio cultural change that are causally related to the use of mobile phones. In order to build a coherent argument, we will need a model for how information and communication technologies (ICTs) artic ulate with other cultural domains and processes. This model should incorporate the insights of Social Shaping and practice theories, while acknowledging that with technology in particular, there are objective material constraints (resources) that in additi on to virtual rules, shape individual courses of action (Rasmussen 2000) What is a technology? A technology has three related facets or components ( Bijker, et al. 1987). First a device, such as a mobile handset is a physical artifact and object of value. We must consider how this artifact is adopted within the system of social relations, conceptions of property, and exchange. Second, there are activities or processes that surround the device in the case of a mobile phone, recharging the battery, carrying the phone, dialing numbers, checking text messages, and commun icating through the phone become part of the everyday routine of life. Third, there is knowledge surrounding the technology. In the case of mobile phones, this


33 includes the process of design ing the phone and the mobile network but also knowledge of how to use the phone how to operate the menus, save a contact in the phonebook, or send a text. Each of t hese aspects of technology can be unevenly distributed among the members of a commu nity or society. For example, women, children, or some other social sub group might have restricted access to phones, or knowledge of the operation of the phone might be d isproportionately concentrated among youths or individuals with extensive formal educa tion The particular distribution of technological devices, use behaviors, and knowledge is a function of local infrastructure, economics, and cultural dependent and con tingent upon local material conditions, configurations of power and knowledge shared ideologies and individual practices New technologies might enable or encourage the disintegration of traditional social structures, but they might also serve to bolster existing social relations and forms. ICTs, A ccess, and the C ommunicative E cology Information and communication technologies (ICTs) are a special subset of technolog ies that serve as conduits for the storage, manipulation, and communication of information. Due to their mediating role, ICTs are involved in a vast array of social inter actions and processes in fact, ICTs may mediate any kind of social interaction or process IC therefore be more subtle and complex than, for ins tance Mobile phones have many poten tial functions. For example, individual s might use their phone s almost exclusively to play games, or as an alarm clock. However, it is my assumption that the use of mobile phones for long distance communication holds the


34 most important implications for society To be precise, mobile phones enable instantaneous, point to point communication at a distance through two basic modalities: voice and text. Mobile based internet is a third application that may incorporate a variety of modes of communication and information exchange for instance web pages, chatrooms, and social networking sites. From an evolutionary perspective, technologies allow humans to overcome the physical limitations of our bodies Simple tools like the hammer multiply the amount of force we can exert over a given area; the pulley allows us to lift heavy weights Mobile phones transmit the sound of our voices over a distance roughly the equivalent of allowing us to wh isper into the ear of someone standing on the other side of the world. This ability represents a relaxation of space time constraints on human interaction. This relaxation is analogous to what Giddens in discussing modernity, referred to as the the lifting of institutions out of local contexts and their expansion and generalization in time and space (1991). Giddens wrote that in modern 19). T here is disa greement about the extent of this relaxation of social processes due to ICT development The most extreme position is taken by transcending capabilities of recent ICTs render obsolete the notion of distance as an organizing principle of human behavior and social structure (Cairncross 20 01). While rejecting this extreme position, ot her researchers agree that ICTs do, to a degree, enhance spatial


35 and temporal flexibili ty (Stephen and Marvin 1 996; Aoyama and Sheppard 2003). T his relaxation remains contingent however, up on the smooth functi oning of material spatially fixed infrastructure such as towers, transmitters and power plants (Couclelis 2004) Thus, society has not (yet) transcended the limitations and dictates of time and space W ith strictures of time and distance relaxed to a considerable extent, how will human societies function? In recent decades a body of theories has emerged, collectively termed theories of the Information Society which speculate that ICTs will lead to broad based changes in social and economic organiza tion. Castells, a prominent theorist, argued that we are moving toward a Network Society, in which key social structures are organized around digital information networks. Castells argued that the role of space and time in human interactions will become pa d that among other things, the new temporal spatial regime have major implications for production, consumption, and social organization (2004). Drucker speculated that there will be a transition from an economy based on material goods, to one based on intangible goods, information, and intellectual property (Drucker 1969) These are compelling perspectives on the macro implications of ICTs, but they provide little guidance for analyzing social change at the micro scale At the micro scale the most direct socio cultural implication of mobile phones (and similar ICTs) is that they reconfigure access to information, people, services, and other information technologies (Dutton 2005 ). I ndividuals, households, a nd higher level social groups can use ICTs in strategic ways to pursue social and economic resources and objectives. These objectives need not conform to the model of


36 culturally and from individual to indi vidual, and may have little to do with the accumulation of material wealth. degree of access to ICTs is conditioned by material, social, and ideolo gical factors. More so than other technologies ICT access is conditioned by netwo rk effects. In a social network when one entity (node) becomes connected to another, it simultaneously affects the status and opportunities of all other entities in the network A concrete example will illustrate this concept. W hen Semesa, a Fijian villag er, acquires a mobile phone, it indirectly exposes his household members and neighbors to new sources of information, and potentially expands their social and economic possibilities subsequently buys a phone then Semesa becomes connected via the phone to her, and indirectly, to her circle of contacts opening a new series of conduits for information and socioeconomic resources Such network effects on communication are not restricted t o ICTs, nor even to tec hnologies per se generator, it may allow Semesa to charge his phone more often and perhaps to receive an important time sensitive call from Vani that he otherwise would not have received if his phone battery had been flat. If Semes a were subsequently to have a quarrel with Vani, or with Ropate, then perhaps those ties would be temporarily closed to him, and indirectly, closed also to his circle of kin and friends. As illustrated by this example, d etermining the nature and extent of any tele access requires a systems approach. The communicative ecology is a useful conceptual tool in this regard. The communicative ecology is the pattern of communication in a soc ial group based upon the total system


37 of social and technological actors, linkages, and constraints. In Chapter 2 I use this ecological systemic metaphor to describe how Fijian culture and social structure, as well as material conditions in Fijian communit ies, tend to shape mobile phone access and use. Summary To summarize the preceding theoretical discussion, mobile phone access has increased rapidly throughout the world, most recently in rural, developing regions that formerly had limited telecommunicati ons services Such regions are the site of many small scale, kinship oriented communities which have been a central topic of anthropological study up to the present and remain important repositories of human socio cultural variation In this dissertation I explore the implications of mobile phone access for social life in kinship organized communities in Fiji. I approach the concept of community from a network perspective, in which communities consist of overlapping ego networks of individuals, which tran scend the boundaries of social and geographic units such as clans and villages. The nature of ICT access, and social network formation in any society is shaped by local cultural factors. I take to mean the socially shared (and socially learned) behaviors and ideologies held by members the group. I have outlined two basic approaches to understanding culture change: diffus ionism and boundary maintenance. I believe that both of these approaches have some value in unders tanding the case of ICT use and social change in Fiji. I propose to take a balanced approach to examine the causal relationship between socio cultural and technological change in Fijian communities. I assume at the


38 outset that the use of technologies can effect socio cultural change; however, I also acknowledge that technological design, access, and use patterns are heavily shaped by socio cultural factors. ICTs are a special subset of technologies that are distinguished by their ability to process informa tion and mediate communication. In this sense, the impacts of ICTs are particularly contingent on network effects. I assume an overarching systems perspective to understand mobile phones role within the system of material and social opportunities and cons traints in Fijian communities. For a community level study I propose that the most fruitful approach to studying the impacts of ICT is to study how they reconfigure access to information, people, services, and other technologies (Dutton 2005 ). Access, bo th to ICTs themselves, and secondarily to the in formation, people, and services made accessible via ICT use, is contingent upon local material, social, and ideological factors. Thus, the implications of ICT are context dependent. My choice of Fiji as a re search site was partly serendipitous, and partly due to my personal history there. However, there are also sound methodological reasons for conducting this research in Fiji. The impacts of technological change tend to be most pronounced where the change is most recent and rapid. Rural Fijian mobile access increased very suddenly in 2008 with the deregulat ion of the local telecom market. B efore 2008 many rural Fijian villages still relied on communal radio telephones or just a few shared between households Thus, individualistic mobile phone use is only now becoming integrated within rural Fijian presence is still remarkable enough to excite comment and controversy.


39 The remote island setting of many Fijian vil lages is a context in which distance transcending technologies such as mobile phones may have an especially fundamental impact The island setting is methodologically advantageous in terms of reducing the number of variables under consideration, due to isl connections and undiversified economies. Oceania closely related cultural histories raise the potential for comparative studies on islands with very different ecological and economic conditions. It is for this reason that Oceania has been characterized (Sahlins 1958) I have attempted to structure this dissertation so that the chapters flow in a logical p rogression However, each chapter should also be complete and com prehensible in itself. Chapter 2 explains my research questions, methods, and the basis upon which I chose my research sites. Chapter 3 focuses specifically on the Fijian context of my research : I describe Fijian history, culture, and social struct ure, and use the concept of a communicative ecology to understand how material and cultural factors influence communication p atterns in Fijian communities. Chapter 4 provides mostly qualitative descriptions of phone ownership and use in rural Fiji: how pho nes are acquired and become part of property relations, how they are used, and what purposes or functions they fulfill. In Chapters 5 and 6 I analyze in detail the socioeconomic data I derived from household surveys, and the phone network data, which I cop ied from calling networks, and discusses how network configuration relates to the rural Fijian communicative ecology described in Chapter 2. Chapter 6 focuses on the role of mo bile phones in exchange of material resources, with implications for household income


40 sexual relationships in Fiji. Chapter 8 is my conclusion, in which I tie together the various observations and concepts developed in the preceding chapters.


41 CHAPTER 2 RESEARCH QUESTIONS METHODS, AND SITE SELECTION Overview In this chapter I describe and explain my research questions as well as how I chose my research sites, defined the boundaries of the study population, selected research participants, and collected data. Copies of my survey forms are included in Appendix B and an in depth explanation of collection of data from SIM cards in Appendix C Research Questions My resear ch questions can be formulated broadly, and then broken down into more specific sub questions based upon which falsifiable hypotheses may be generated. The fundamental question underlying my research is phone use lead to socio cultural c literature review as well as first hand observation, I focused my inquiry on mobile phone use in t hree domains of social life : orks among kin, friends, and business associates distance exchange of economic resources For each of these three domains, I list several relevant questions: Social networks (Chapter 5) 1. Is the size or composition of phone based social network related to the the frequency or intensity of phone use, or to other social or demographic c haracteristics of t he phone user ? 2. Do phone based social networks differ in composition from their social networks built around face to face interaction? 3. What are the characteristics of the individuals who are most central within each k?


42 4. How do the calling networks of the three communities differ in size or demographic or economic characteristics? Exchange (Chapter 6) 5. What goods are exchanged, what is their monetary value, and how does exchange fit into the broader household economy? 6. Are phone ownership, frequency/intensi ty of phone use, or the size of an 7. Does long distance exchange serve to increase or decrease inter household income inequality in the study communities? Romantic relationships (Chapter 7) 8. How are mobile phones used by Fijians to seek out and maintain romantic relationships? 9. What individual and community level advantages or di sadvantages do Fijians 10. Does the behavior of Fijians in romantic relationships, and the outcome of relationships involving mobile phone use depart from pre existing behaviors or relationship outcomes (i.e. cohabitation, sex, pregnancy, marriage, or divorce) ? Site Selection Although Fijian communities vary broadly in terms of their geographic, social and economic characteristics I hope that some of the conclusions of my research will be broa dly applicable to the rural Fijian context I therefore decided to study several communities in order to have a broader basis for making comparisons and generalizations. I ultimately decided upon 3 communities because this was the maximum that my time and funding constraints permitted; also, 3 communities allows me to triangulate in order to better understand the relationships among community level vari ables. My initial interest in Cagi and Toba stemmed from the fact that these villages had been the focus of previous studies, in 1958 and 1974 5, permitting an analysis of change over time (Bayliss Smith 1976; Bedford 1978; Watters 1969) I visited Cagi and


43 Toba in 2009, and spent a combined five weeks in the two villages, which facilitated my research activities in 2011 2012. In 2011 I returned to Fiji to co llect data for this dissertation. I decided to revisit Cagi and Toba and also to incorporate a 3 rd village, Veidogo into my research. I conducted fieldwork in these three villages from Ju ly 2011 to May 2012 (a span of 10 mo nths), living with local families. I was also guided in my site selection by the theoretical idea that the uses and impacts of telecommunications technologies are heavily shaped by the local communicative ecology, including the availability of transport technologies (Altheide 1994) I therefore sought three communities that had varying degrees of access and ban centers on Viti Levu Island. In this regard, Toba village is the most remote, r elying on small open boats for transport to urban areas. Cagi has an intermediate level of access, with a ferry arriving twice a week from the capital c ity. Veidogo has the most conven ient transportation access, as the village is only a 30 minute bus ride away from Suva, Cagi Toba and Veidogo also differ along other socioeconomic dimensions: access to telecom services, population size, agricultural potential, income sources social structure and homogeneity of kinship groups. I have l isted the characteristics of these three communities in a table to facilitate comparison (Table 3 1 in the next chapter ). Study Population I define the boundaries of the study population based on both geographic and social criteria. For the purposes of my day to day ethnographic research, the study population consisted of all the residents o f the three villages: Toba, Cagi and Veidogo d the members c alling network s. These calling networks extend


44 geographically beyond the village and local area, to distant areas of Fiji, and ultimately span the entire world. I analyze some variables, such as income, at the household level, but other variables, such as telephone use, at the individual level. Therefore the units of analysis of my study can be defined either in terms of individuals or in terms of households. Data Collection I collected data via participant observation and informal and semi structured inte rviews, a household survey with demographic and economic components, and a phone survey of telephone owners. I also photographed and transcribed written records: the ledgers of sales transactions in village shops, and the minutes of community meetings as a means of supplementing and cross checking my survey results Participant observation In 2009 and 2011 2012 I spent a combined 4 months living in Cagi, 3 months in Toba, and 3 months in Veidogo. During this time I lived as a guest in the home s of local fam ilies, who kindly made room available in their houses for me to sleep and work and treated me as a member of their households The particular households in which I lived were recommended to me through acquaintances or proved appropriate due to the leaders hip roles of the household heads in the community. For example, I came to Cagi via an acquaintance in a neighboring village; my eventual Cagi host was the man who served as the traditional liaison with the village from which I had come and so naturally wa s appointed to welcome me to Cagi In Toba I stayed with the family of the government headman ( turaga ni k oro ) wh o had in the past hosted other visiting researchers. I n Veidogo I stayed with a family recommended to me by the local Methodist m inister, with whom I was acquainted via my host family in Toba


45 In each of my host households I ate my meals together with the household members and attended church with them at least once a week. My daily schedule varied. On some days I accompanied household members or friends from the village to the ir gardens or fishing in the sea. If there was a community event such as a meeting or wedding I usually attended. Most days, however, I spent walking around the village, visiting and conducting surveys and interviews. Duri ng the evenings after dinner I joined in the village social life, which typically involved either kava drinking or watching DVD movies ( when electricity was available). Interviews I conducted a variety of interviews in the study communities. Unstructured and key informant interviews were part of my daily participant observation activities in Cagi Toba and Veidogo In each community I also conducted broad ranging interviews with local leaders including the hereditary chief ( turaga ni vanua ), elected headma n ( turaga ni koro ) and religious leaders (Catholic catechis t or Methodist ministers and preachers ( vakavuvuli talatala and vakatawa ). These k ey informant interviews dealt with general issues affecting the study communities, recent and upcoming development projects, local history and the structure of local kinship groups. I asked ordinary community members a variety of questions about their economic activities, their interactions with migrant kin, topics related to telephone use, and general questions about Fijian culture and language. It was easy to engage villagers in conversation. Fijians are very hospitable hosts and attempt to put their visitors at ease. Many were curious about my life in the United States and this serv ed as a launching point for conve rsations comparing l ife in Fiji and the U.S. K ava drinking is an excellent opportunity for lengthy conversations, as several


46 people are seated together for hours on end, and the mild narcotic effect of kava l ends itself to open, fluid conversation. Nowaday s, Fijians spend many evenings watching DVD movies, which generally hampers opportunities for conversation. In this sense, the weeks long power outages in Cagi and general lack of electricity in Toba were a boon for my research, as many evenings without el ectricity were spent either drinking kava or simply laying on the floor of a house, chatting in the dark. I conducted semi structured interviews in conjunction with the household and phone interviews. As a consequence I mainly conducted these interviews wi th adult heads of household (male or female) and with phone owners. The household interviews incorporated questions about household economy (allocation of labor and resources within the household) and exchanges of goods and money carried out with kin and f riends living outside the community. Interviews conducted during the phone survey dealt primarily with phone use: the contexts of phone based interactions, purposes of phone calls, etiquette of phone conversation, technical issues with phones or network se rvice, and the forms and quality of relationships sustained through the phone. All conversations and interviews were conducted in the Bau dialect of the Fijian language, with the exception of two interviews involving individuals who were very fluent in Eng lish. Casual, unstructured interviews often took place around the kava bowl, or in outdoor settings, and were not recorded. I tried to write down the key details of the conversation s as soon as possible after concluding such informal conversations I recor ded semi structured and key informant interviews using a digital voice recorder (with the prior consent of the participant). I roughly transcribed my recorded interviews


47 (not word for word) and imported all interviews and field notes into MAXQDA to facilit ate analysis. Copying local records I was permitted to photograph a variety of local records in the study communities I have not systematically analyzed these records, but in the case of Cagi I have transcribed a portion of the store records in order to s erve as a check on estimates of household income elicited during the household interviews. In Cagi I was permitted to photograph records from the cooperative store (daily purchases by each household, sales of copra and kava by each household to the coopera tive, and purchases of bulk goods by the cooperative). In Cagi worth of village meetings, and records of sales of phone credit to villagers by a local entrepreneur. In Toba I was permitted to photograph the few extant sales records from a defunct cooperative store, as well as some post office records of money transfers. In Veidogo I photographed documents describing recent and ongoing development projects in the community. Surveys Copies of my survey forms a re included in Appendix B. Household surveys In each village, I first conducted a census survey of all households. This census incorporated a demographic and economic questionnaire that I carried out with either the male or female head of household. In ad dition to household members and economic immediate kin and about individuals outside the village with whom they had exchanged money or goods during the preceding year (ru ral urban exchange or remittances) I also


48 asked which household members owned phones, which permitted me to construct a community wide list of phone owners, with whom I subsequently conducted the phone surveys. I ended the household survey with a brief in terview to better understand the household economy, and to clarify any inconsistencies on the questionnaire. Phone surveys Based on the information I derived from the household census, I attempted to survey every phone owner residing in the study communit ies. I was largely successful: I separately surveyed 170 phone owners from 99 different households in the three villages. Only one phone owner declined to be surveyed. There were a few individuals whose status either as owners of functioning phones, or as current village residents, was questionable; nevertheless I tried to survey as many of these individuals as possible. In phone surveys I asked the phone owner questions about his/her history of phone ownership and typical patterns of phone use. I then copi list of phone contacts to my computer using the SIM card reader, and elicited information about each of the phone contacts. I ended the phone survey with an interview about the use of the phone, and the phone s significance in his/her everyday life. I provide a detailed description of the process of copying data from the SIM card in Appendix C. In Appendix C I also discuss logistical, methodological, ethical issues concerning this data collection technique. Survey logistics and compensation I offered p articipants in both the household s urvey and the phone survey $FJ 5 (equi valent to about $US 2.5 0 ) as compensation for their participation This is approximately equivalent to the wages for 2 hours of unskilled labor in Fiji. Househ old


49 surveys varied in duration based upon the number of household members, kin, and economic and exchange activities, but generally lasted from 30 minutes to 1 .5 hour s Phone surveys varied in duration based on the number of contacts in the phone directory but also tended to range from about 30 minutes to 1.5 hours. The longest interview was conducted with a man who had over 400 phone contacts. It was completed during 3 separate sessions (all while drinking kava together ) I offered this long suffering man double compensation for his participation in the study Equipment Hardware I used a laptop computer, a handheld digital voice recorder, and a Dekart SIM card reader to aid in data collection. I used an external hard drive to back up my data daily in cas e of computer problems. Software I used Microsoft Word and Excel to record field notes and as a spreadsheet to record survey responses. I imported interview transcripts and field notes into MAXQDA text analysis software to facilitate analysis. I imported phone directory data into UCINET and Netdraw software in order to visualize and analyze calling networks. I used SPSS to aid in statistical analysis. Data Analysis I analyzed network data in UCINET and Netdraw. I imported household survey data from Excel s preadsheets into SPSS for statistical analysis. When analyzing network data, I generally treated telephone ties as symmetric (undirected) ties, by symmetrizing my calling matrices. If two responses about a particular tie differed, then I symmetrized on the


50 directories. Respondent A said that they call each other weekly. Respondent B said Reporting of Result s Reporting of S tatistics In this dissertation I report the means of population distributions along with their standard deviations. Error bars in charts represent the standard error of the mean. Telecommunications technolo gy the central topic of this dissertation, is undergoing rapid changes. By the time this dissertation is read, the technological situation in Fijian villages is likely to have evolved considerably Nevertheless, I adopt the common ethnographic convention of writing in the present tense. This is to make my prose easier to read, and hopefully to more fully immerse the reader in the atmosphere of the Fijian village s which I experienced. The reader should not assume on this basis that the social practices and conditions I report in this dissertation persist unchanged into the present day. Anonymity of P articipants M obile phone research inhabits an ethical gray area As a society we are still struggling to decide what kinds of information we are comfortable sha ring wit h researchers. We feel uneasy about the fact that when we use a phone or log on to the internet, our activities may be monitored by companies or governments, often without our knowledge or consent. Everyday mobile phone interactions are typically mundane, but may also convey embarrassing, personally damaging, or even incriminating information. Sometim es, the


51 very fact that a connection exists between two individuals can be socially or legally unacceptable. Studies of mobile communication can be et hically complex because phone communication is two way. By asking one phone owner about their phone use and phone contacts, it necessarily reveals something about the people with whom they make contact. These individuals not have given their explicit permission to be research subjects. For all of these reasons, it is very important to pr otect the anonymity of research participants in a ny study that deals with mobile communication. To this end I have altered the names of my individual research participants as well as the names of the Fijian communities in which I worked. The details of stories and anecdotes have also been subtly altered in many cases, so that they cannot be clearly attributed to actual individuals. In prin ciple, even if a member of one of the study communities were to read this dissertation, they should not feel that their privacy or the privacy of their family and friends has been infringed upon. I believe that I have taken sufficient precautions to compl y with this principle.


52 CHAPTER 3 RESEARCH SETTING Overview I n this chapter I describe the rural Fijian context in which I conducted my research and how local history, c ulture, social structure, and material conditions have shape d rural Fijian communicati on practices I begin with a ge neral overview of Fiji My accounts of Fijian geography and history are superficial, and meant only to orient and acquaint the reader I n the second part of the chapter I use the unifying in order to describe how patterns of communication are influenced by Fijian social structure, as well as the technolog communities In the final section of this chapter, I describe the three communities in which I cond ucted ethnographic research: Toba Cagi and Veidogo These descriptions are likewise superficial and meant only to facilitate broad comparisons and contrasts between the communities Based on the characteristics of the study communities, as well as the n ature of the rural Fijian communicative technology, I will generate predictions about mobile phone communication in Chapters 5, 6, and 7 and test them using data collected in Toba, Cagi and Veidogo. Fiji M any people know Fiji only as a stereotypical island paradise with friendly, exotic natives and tropical beaches based company, also capi talizes on these utopian


53 images exporting bottled Fijian spring water to wealthy overseas consumers for a hefty profit (Connell 2006) Behind the idyllic faade Fiji is a n ethnically diverse, rapidly changing society ; an influential country in the Pacifi c island region, and a regional hub for transport, communications and tourism. Fiji still grap ples with issues that have their roots in the colonial past and faces complex emerging challenges in the realms of ethnic politics, human rights and global clim ate change In the wake of a series of coups de etat the country is seeking a sustainable path back to democratic government The following overview of Fijian geography and history will serve to orient the reader as to the context in which I carried out m y research on Fijian communication practices. Geography and E nvironment Fiji is an archipelagic country in the tropical Southwest Pacific Ocean, consisting of 332 islands, 110 of which are inhabited. The total land area of Fiji is 18 270 km 2 slightly smal l er than the U.S. state of New Jersey The largest island, Viti Levu, accounts for nd largest island, Vanua Levu, an relatively small. Much of consists of ancient volcanic rock which has broken down over millions of years into fertile soils. The peaks larger islands block the mo isture bearing Southeast trade winds, creating a humid tropical climate on the wi ndward (southeastern) sides of islands, while the leeward (northwestern) sides of the larger islands are drier (Neall and Trewick 2008) On smaller lower islands, this climate pattern is less pronounced. fringed with coral reefs that teem with many species of marine life errestrial flora and fauna island hopped via Southeast Asia over the


54 course of millions of years The first human settlers arriving via Melanesia on outrigger sailing canoes about 3000 years ago carried with them domesticated plants and animals including the Oceanic triumvirate of dogs, pigs and chickens S ince European contact in the late 18 th century many other species have been introduced to the archipelago. Some introduced species such as the cane toad and mongoose, hav e proven to be ecological nuisance s while others, such as cows and cassava, have become economically important. Figure 3 1. Regional map with Fiji highlighted in blue; with inset map of the Fiji archipelago. Maps courtesy of Economy Fiji is considered a developing country, although it is among the most developed of the Pacific Island nations with a per capita GDP of $4,900 (CIA World Factbook 2013 ) Compared to other Pacific island countries, Fiji has a relatively diversified economy based around sugar exports, tourism, and remittances from Fijians working abroad. Other industries include clothing, copra, gold and silver mining, and lumber. However, 70% of the labor force is engaged in agriculture, a large pro portio n of which are subsistence or semi subsistence farmers.


55 Demography Fiji has a n ethnically diverse population of about 890,000 ( FBS 2007 ) The population growth rate is 0.73% and overall life expectancy is 72 years (CIA World Factbook 2013). Approximately 57% of the population are indigenous Fijians 37% are Indians, and 6% are of European, Chine se, Oceanic, and other descent. 52% of the population dwells in urban areas In recent decades there has been a politically sensitive debate about the terms that a re used to refer to non indigenous citizens of the Fiji Islands. For example, ethnic Indians living in Fiji have been variously referred to as Indo Fijians, Fiji Indians, un ilateral decree that all citizens of the Fiji Islands are to be referred to henceforth as opulation to be referred to as iTaukei (the indigenous party that was deposed in the 2006 coup, oppose this inclusive nomenclature. They prefer to indigenous citizens This dissertation focuses particularly on the comm unities and cultural practices of indigenous Fijians ( iTaukei ). For the sake of brevity and clarity refer exclusively to these Indigenous Fijians. When referring to members of other ethnic i citizens of Indian descent, and refer to other groups as Chinese, European, or other citizens of Fiji. This is not intended as a political It is simply a convention I adopt so that my writing is less cluttered, and easier to understand.


56 History ethnic diversity is a result of successive waves of settlement. The first human inhabitants of Fiji were the Lapita people, who sailed to Fiji via Melanesia in outrigger canoes approxima tely 3000 years ago (Nunn, et al. 2004) Settlements of these ancient sea faring people are characterized by their eponymous dentate stamped pottery. the descen da nts of these original settlers with subsequent cultural influen ces and genetic infusions derived from inter island contact with Vanuatu, Tonga and Samoa Indigenous Fijians today display a diverse blend of the physical and cultural traits that ethnologists typ ically ascribe to Melanesia and Polynesia, a legacy of Fiji postion at the crossroads between these two regions Over the course of 3 millenia, indigenous Fijians develo ped a hierarchical society of chiefdoms in which distant communities were bound together through ties of marriage, kinship, war, trade, and tri bute. Fijians of the pre contact era were horticulturalists and fishermen, but were also masters of woodworking, shipbuilding and navigation, combat and dance. Fijians never developed a system of writing, but had a rich oral tradition through which legends and histories were passed down. Western contact European and American explorers and traders began visiting Fiji regularly in the early 1800s. They eagerly sought Fijiian beche de mer and sandalwood, which were valuable commodities in the China trade. At t he time of European contact, Fiji was in the midst of a process of political centralization. Rival indigenous confederacies ( matanitu ) dominated the islands surrounding the Koro Sea. The arrival of Europeans, and


57 particularly trade in guns, accelerated and intensified the wars between confederacies. A simultaneous struggle was being waged in the realm of religious ideology. Methodist and Catholic missionaries gained a foothold i n Fiji in the 1830s, and labored zealously to convert the Fijians from t heir indigenous religion Missionaries contributed to end ing customs such as cannibalism and wife strangli ng, and encouraged an end to the wars that were rending Fijian society. confederation converted to Christiani ty in 1854 firmly establishing the legitimacy of the new religion. The following two decades saw a spate of land alienation in Fiji (fuelled by a blackbirding Melanesian labor for Fij multiplying plantations and political jockeying chiefs and the European powers natural resources and geo strategic position Eventually the indigenous state of Bau prevail ed in th e tribal wars, and held sway over a Fij i that was, for the first time, largely unified However, u nder threat of imminent annexation by the United States, Fiji was voluntarily ceded by its chiefs to Great Britain in 1874 Colonialism 1874 1970 first colonial administrators instituted a paternalistic system of indirect rule in Fiji that sought to preserve indigenous leadership structures and soc ial institutions. Fiji was surveyed by a Lands Commission and 87% of all land set aside under the communal ownership of corporate descent group s ( mataqali ). A set of Native Regulations tied Fijians to their native villages, where they lived a largely su bsistence


58 lifestyle. Villagers were legally obligated by the colonial regulations to plant a yearly quota of crops, and to assist in communal task s such as house building. The colonial regulations denied Western commercial interests in Fiji a suffici ent pool of proletarian labor ers To alleviate the labor shortage about 60,000 indentured Indians were shipped from the subcontinent to Fiji fr om 1879 1916 to work in the sugarcane fields The majority of these Indians settled permanently in Fiji after their period of indenture and became tenant farmers or small b usinessmen. Later, wealthier merchants from the subcontinent came to Fiji of their own acc ord to cater to the needs of the burgeoning Indian community Chinese also arrived in Fij i in smaller numbers to establish businesses. The descendents of these Indian and Chinese immigrants today form an important segment of lasses. Fiji was used as an Allied base during the Second World War. Many Fijians had their first extensive interactions with Westerners while working as laborers on military bases, or by fighting alongside Allied soldiers in the jungles of Melanesia Afte r the war, t he regulations stipulating that indigenous Fijians remain in their village s and c ontribute to communal projects were abolished, p art of a series of reforms aimed at encouraging entrepreneurialism among indigenous Fijians and economic developmen t for the country Urbanization accelerated and modernization efforts geared up for independence. Independence, political strife, and the uncertain future On the eve of independence in 1970 Fiji ans looked optimistically towar d a prosperous future Throughout the 1970s and early 1980s, Fiji was a model of progress and ethnic tolerance in Melanesia, a region otherwise beset with political instability. However, unresolved tensions simmered beneath the surface in Fiji Many indige nous


59 Fijian s fear ed that Indians, by then the demographic majority, would seize political power and ultimately wrest control of most valued asset, their land. Underlying the ethnic tension s were complex factors, including emerging class distinctio ns associated with urbanization a nd a long running power struggle among the indigenous chiefly elite The militar y and the Methodist C hurch were also key players in the bid to identity and future Was Fiji to be a Christian nation with poli tical control by indigenous chiefs guaranteed in perpetuity, or was Fiji to be come a modern, multi cultural democratic state? In 1987, the election of a multi ethnic L abor coalition prompted a military coup that aimed to restore indigenous political and cu ltural primacy in Fiji. Subsequent coups occurred in 2000 and the latest in 2006. The government of Fiji has been controlled by the military since the 2006 coup although new elections are slated for 2014. In the meantime there has been ongoing censorship of the press and suppression of political opposition raising serious concerns about the likelihood of truly open elections coups have been relatively free of vio lence, but their cumulative effect has created a climate of uncertainty in the countr y. Indians perhaps face the greatest risk s from political instability and many of the Indians who have been able to, have migrated overseas. Th e loss of tens of thousands of well educated, skilled members of the labor force has affected the Fijian economy (Reddy, et al. 2004 ) However it has also allowed indigenous Fijians to regain their status as the clear demographic majority allaying their immediate fears of an Indian political and cultural take over Beyond politics, Fijian society has underg one fundamental modernization since independence ucture have been upgraded including trans port


60 and telecommunications. Greater ease of movement and communication has broadened the economic and social horizons of rural Fijians w ho were for much of the colonial period confined by law to their villages After decades of rural urban migration, the majority of the population now lives in urban areas The government has struggled to keep pace, and the Suva Nausori corridor is garlanded with a sprawl of squatter settlements The quest for better opportunities has led many Fijians to emigrate overseas, and sizeable Fijian communities (both indigenous and Indian) exist in England, Australia, New Zealand, and on the West Coast of the U S. Despite these extensive changes, Fijian society still bears the imprint of the early co lonial policies. Indians, Chinese, and Europeans continue to hold sway over a dispropo rtionate amount of the urban business sector while most of the indigeno us population remains rooted in rural areas Villages serve as a social and symbolic anchor not only for rural residents, but also for Fijians in urban areas and overseas, who look back to the ir native village s as a source of cultural id entity, stability a nd security. Li fe in Fijian villages today is similar in many ways to the way i t was decades ago. However, tec hnological ch ange opens avenue s for rural Fijian culture to take on new form s of expression Fijian Culture / Itovo Vakaviti The diffusionist and boundary maintenance approaches to culture, briefly outlined in Chapter 1, are each useful for understanding aspects of Fijian society and Fijian identity. From the diffusionist standpoint, Fijian society can be understood as the result of historical waves of immigrants and exogenous influences, breaking in


61 people on their outrigger sailing canoes some 3000 years ago. Fundamental aspects of the Proto Polynesian cultural comp communities in the present day. Among these traits are a fishing/horticulture/husbandry subsistence complex, patrilineal descent with an emphasis on primogeniture, and an Austronesian language closely related to the languages of neighboring Oceanic archipelagoes. developments, and overlain by diverse exogenous influences. For centuries before European contact, Fijians voyage d to neighboring Tonga and Samoa to marry, trade and wage war. This legacy of inter island voyaging is manifested in the diverse body types, dialects, and customs found in different parts of the Fiji archipelago today. The arrival of Europeans brought even more drastic changes to Fiji: written language, Christianity, a host of new technologies, immigrant neighbors from the Indian subcontinent, and fundamentally different political forms. From this standpoint, the mobile phone is just the latest in a long li ne of acculturative influences that have increasingly integrated Fiji within the global mainstream of behaviors and ideas. The fact that Fijians have derived many of their practices and ideas as diffusions from overseas is undeniable. However, as an ethnog rapher speaking to rural indigenous Fijians, it quickly becomes evident that this is not the way that Fijians see their world. Fijians keenly feel their separateness as a cultural group, and draw sharp recent ly arrived ethnic minorities: Even as rural Fijians watch the latest Hollywood movies or adopt the latest model of


62 mobile phone, they interpret these cultural accretions th rough the lens of their Fijian traits that they consider alien or degrading. The specific contents of Fijian culture, then, are part of an ongoing discourse over values and tradition what is authentic, what is spurious, what is Fijian, and what is the Other. In this sense Fijian culture is a reactive mechanism for drawing distinctions from other groups. Nicholas Thomas argues that in Fiji this identity differentia tion began Samoans faka Tonga (Thomas 1992). The Europeans, Indians and Chinese with whom Fi jians came into contact after circa 1800 were far more different from Fijians than were their Samoan and Tongan neighbors. The need for cultural boundary maintenance accordingly became that much stronger during the colonial and postcolonial eras. Indeed, to, and evaluation of, institutions that originated during the colonial era. Fijian value judgments about culture are typically expressed through a discourse that opposes Light (r aram Pre colonial heathen practices such as inter tribal warfare, wife strangling and cannibalism are now considered evil and backwards, butobuto The arrival of Christianity is considered as the arrival of rarama (the Light). Benevolent autochthonous v alues such as generosity to kinsmen have been used as a way to understand how, in a sense, Fijians ha ve been Christians all along.


63 Other post contact developments such as capitalist social relations, ar e held in rhetorical opposition to what is authentically Fijian. Thomas (1992) noted that many of the attributes which Fijians ascribe to themselves, suc h as an aversion to commerce, are specifically the opposite of the cultural values of Fiji Indians perhaps emphasized as a form of boundary maintenance tied to broader ethnic political rivalry. Although Fijians assert the moral superiority of their cultur e, they also emphasize that their inability to fully embrace the market is the reason why Fijian communities are relatively themselves. Fijians today remain curious about the potential benefits of increasing engagement with the world, while simultaneously hesitant that excessive engagement might lead to a degradation of the lifestyle and sense of uniqueness that they cherish. Returning to our discussion of mobile phones, it would require a willfully blind eye to deny the importance of technology in the changes that have occurred in Fijian society since the early 1800s. However, the role of socio nological colonial history and peripheral position in the postcolonial global economy. The tim ing of technological br eakthrough, was determined whol ly by the regulatory actions of the Fijian government. Thus, technological and social factors, in Fiji as elsewh ere, are intertwined in processes of socio cultural change.


64 T elecommunications in Fiji The story of telecommunications development in Fiji takes place a gainst this broa der historical and cultural backdrop. At the time of European contact, Fiji ans already h ad a millenia long histor y of oceanic voyaging and interaction with neighboring island groups. Since at least the 1500s, chiefly elites in Fiji, Tonga and Samoa intermarried and exchanged canoes and other valuables Wooden double hulled sailing canoes the largest of which reached lengths of 30 meters, were used for long distance voyaging between Fiji, Samoa and Tonga Fijians had no system of writing but they did have ext ensive oral traditions, which became enshrined in legends and in m eke, songs and dan ces that are considered the intangible property of the kin groups that originated them Fijians also developed t ools sea shell ( davui ) served as an effective horn, and was used to announce the de ath of a chief, or by fishermen to celebrate a catch of sea turtles as they triumphantly headed back toward land Wooden slit gongs ( lali ) were used to drum out messages between villages. The roots of certain forest trees could also be beaten like a drum t o transmit signals at a distance. A range of different drumbeats transmitted signals for warfare, celebration, retreats, and fi res. Davui and lali are still used in Fijian villages today although their uses have been adapted to contemporary needs: sign al ing the beginning of village meetings and church services. Colonial telecommunication developments Modern transport and telecommunications infrastructure emerged in Fiji as part of the apparatus of capitalist enterprise and colonial administration. After cession in 1874, British authorities embarked on transport infrastructure projects geared toward


65 This included laying rail, roads and bridges linking the main cane (Britton 1980). A doption of Western ships and establishment of commercial ports at Suva, Levuka, Savusavu and Lautoka enhanced the volume of inter island commerce. In the realm of communication, the earliest Wesleyan missionaries in Fiji immediately set t o work on an orthography for the Fijian language and the first Fijian grammar was printed in 1839 (Clammer 1976) running newspaper, the Fiji Times, began circulating in Levuka in 1869. However, transmission of messages between islands was still very boat jo urneys were subject to the vagaries of weather. During the late 1800s, carrier pigeons were pressed into service to deliver me ssages between Levuka and Suva. T h e flight took around 30 minutes ; pigeon borne news was then locally disseminated via newspaper. The first telephones in Fiji were installed by the Colonial Sugar Refining Company in 1898, serving the sugar industry I n 1904 a phone line was laid between Suva and Levuka, and in 1910 the first public telephone system went into service in the new capital city Suva. Around the same time an overhead pole line was built across the main island Viti Levu. This cross island line required constant maintenance and repairs, and was eventually abandoned in 1937 when the Southern co astal road opened (TFL 2013) For years thereafter however local phone lines in many of rural areas still consisted simply of wires strung between trees.


66 Figure 3 2. A Cagi villager poses with an old phone Decades ago, the 14 villages of Koro Island were connected via a wire that was strung between trees along the coast Each village had only one phone, and t he line could only carry one call at a time, which could be overheard at any phone set along the line. Photo courtesy of author. The expa nsion of tele phone access in Fiji after WWII was rapid, but uneven. In y overloaded manual exchange was replaced by an automated exchange. The number of phones in Fiji rose from 2158 in 1945 to 46,583 in 1984 (TFL 2013 ). Th roughout the 19 telephone system continued to be upgraded, and capacity added to exchanges. However, the majority of these developments occurred around urban areas and on the main islands of the archipelago. Until only a few years ago, m ost outer isla nds and remote areas were serviced only by co mmunal radio telephone stations.


67 Radio and t elevision Fiji Broadcasting Corporation (FBC) began public radio broadcasts in 1935. For rural Fijians in the mid to late 20 th century, radio played a more prominent r ole than telephones as a source of news and entertainment During his fieldwork on Moala in the late 1950s Marshall Sahlins observed that most radio programs were in English, and drew few listeners in the village However, mig ht cram into a single house to tune in for one of the occasional Fijian language broadcasts (1962:83). Conducting fieldwork in Lau in 1970, Arno likewise observed that Fijian language radio broadcasts functioned as a valued resource in the flow of convers ation that constituted the major medium of e ntertainment on the isl Fijians, even in remote rural areas, took a keen interest in national, regional, and world affairs stories of conflict among nations in the Middle East, and between the Irish and En glish in Northern Ireland, on the Arno 1992: 13). In 2012, elderly villa gers corroborated to me the accuracy of these ethnographic accounts pointing out the foundations of old houses from their ch ildhood where villagers used to crowd together to listen to one of the few available radios. Radio continues to be a major source of information and entertainment for villagers, and many rural Fijian households today run their battery powered radios both d ay and night. In 1994, Fiji TV began domestic television broadcasts There are now 4 public channels with a variety of local and international programming in English, Hindi and Fijian languages In 2005 Fiji TV introduced Sky Pacific, a paid subscription satellite service In many rural areas public television reception remains poor, and the primary r ationale for television ownership is to tune in for live rugby games (worth watching no matter how grainy the reception) and to play movies on DVD p layers known as


68 Video shops and roadside vendors across Fiji sell and rent che ap pirated DVDs of American movies and professional wrestling bouts Filipino soap operas, and Bollywood musicals These can be bought in Suva at around FJ$1 3 (US$ .50 1.50) per disc, and circulate extensively throughout the villages. Some rural Fijians have an astounding knowledge of Western pop culture gained through watching movies I fielded many questions from villagers about which professional wrestlers and movie stars I had had the good fortune of meeting. V illagers with particular physical or behavior al quirks often are sometimes given nicknames of celebrities the y resemble. For example, I once met Eddie Murphy and Wesley Snipes over bowl s of kava on remote Fiji an island s Mobile phones Fiji acquired its first mobile network in 1994 when Vodafone began local operations. In 2001 less than 10% of Fijians had mobile phones as prices remained prohibitively high and network coverage limited mainly to urban areas. To a large extent, further d evelopment was held back by lack of competition in the Fijian telecom market. Prior to 2008, Fiji had 3 telecom operators each with monopolistic control over its own domain. Telecom Fiji Limited (TFL) was the domestic phone serv ice provider, Fiji International Telecommunications (FINTEL) was the international provider, and Vodafone Fiji was the mobile provider. These incumbents were contractually guaranteed exclusive control over their respective domains until 2015. In 2007, und er government pressure TFL, FIN TEL and Vodafone Fiji signed a Deed of Settlement in which th ey agreed to the phased entry of new competition. This settlement paved the way for the Fiji government to issue a license to Digicel to provide mobile teleco mmunic ations in Fiji. P er the agreement, Digicel would not be allowed to


69 begin service in Fiji until October 2008. Vodafone took advantage of the intervening months to upgr ade its services and prepare itself for the entry of competition As part of its preparati ons Vodafone launched Inkk, a mobile virtual network operator (a separate service provider that operates off the Vodafone network ) in November 2007 to crea te stiffer competition for the incoming rival Digicel Digicel descended upon Fiji in October 2008 in a cyclone of billboards, free t shirts, radio ads, and promotions, and Vodaf one responded in kind. Today these two multinational telecom titans remain locked a n advertising satur at ed competition for the hearts and pocketbooks of Fijians. Digicel sponso rs the Fiji 7s rugby team. Vodafone national rugby stadium is named after TFL. Figure 3 3. The entrance to Korovou Prison in Suva, with Vodafone advertisement Photo co urtesy of author.


70 Even the gothic lookin g Korovou Prison in Walu Bay is painted with cheerful red and white logo. One can hardly walk a few steps in Suva without encountering a billboard or a cluster of the small signs that festoon the fronts of shops, advertising the sale of phone credit Television and radio are inundated with commercials touting the latest promoti ons and call in contests offering phone handse t s and calling credit to the lucky winners Another marketing strategy employed by Vo dafone and Digicel is to send sales representatives directly to rural villages The salespeople often stay overnight with their village hosts, and sell handsets at a di scounted rate. This rural outreach is an effective approach given that many rural Fijian s do not travel frequently to town s where they might purchase a phone. As Semesa, a Toba man recalled he Vodafone reps staying at the local school were selling hand sets really cheap, so I decided to go to the forest that day and cut eno ugh copra to buy Without such aggressive marketing, Semesa might have just waited a few more months until he could obtain a second hand phone from a relative. providers, the telecom industry has spawned independent businesses that offer mobile phone accessories, repairs, SIM unlocking, and sell used phones and generic Chinese made handsets that can accommodate SIM cards from any network provider. According to u nofficial estimates by Telecom regulators, Vo dafone/Inkk now controls approximately 75 80 20 25%. The fixed line market is still controlled exclusively by TFL which caters mainly to urban businesses and government ministries It remains costly to place calls from one network provider to another, which hampers communication between users of different


71 networks High interconnection fees tend to favor the incumbent that possesses the largest network, in this case, Vodafone. There is an ongoing dialogue bet ween service providers and government regulators about adjusting the maximum allowable interconnection fees between service providers, to enhance competition and provide the Internet As the Fijian mobile phone mar ket approaches saturation, int ernet has become the latest frontier for telecom growth and development. Fiji occupies a fortunate position among Pacific Island countries, because it has direct access to the South ern Cross C able Network (SCCN), a 3 rd generat ion submarine fiber optic cable that links Australia to the U.S. with intermediate landings at Fiji and Hawaii. The SCCN gives Fiji the advantage of direct access to high speed internet, whereas many other Pacific Island countries must rely on expensive s atellite connections. In 2012 approximately 34% of Fijians were internet users. A serious barrier to the growth of internet use in Fiji (and other developing countries) is limited computer access and training. Unsurprisingly, the fastest growth in internet use in Fiji has been mobile internet provided by both Digicel and Vodafone. Mobile data speeds are lower than fixed broadband but the advantage is that internet e phone, even in remote areas. In 2012 m any of the youths in T oba Cagi and Veidogo were becoming acquainted to the internet for the first time via their mobile phones I now turn to a discussion of the communicative ecology of a Fijian community how these communication technologies fit into the social and material framework of rural Fijian villages.


72 The Communicative Ecology of a Rural Fijian Community Communication patterns in any social group are shaped by a com plex array of cultural and material pathways and constraints. The context in which communication occurs can be thought of in terms of a communicative ecology in which the various entities interact with synergistic, antagonistic, and network effects (Tacchi, et al. 2003 ). Foth and Hearn conceive of a communicative ecology as having three layers: a technolog ical layer, consisting of the array of communication tools available, a social layer, which consists of people and modes of organizing them and a discursive layer, which is the ideas, themes, and other content of communication (2007). These layers are mut ually shaping, with complex causal interactions occurring both within and between layers From this perspective each communication event can be analyzed at both the individual and community level as part of a complex media environment that stra ddles the ph ysical, social, and psychological/ ideological realms of causation. The ecological metaphor is useful for studying communication because it allows us to make recourse to concepts from ecological and systems theory. As in ecologi cal studies, we define social and/or geographic boundaries for the communication system we are studying, and examine how its constituent entities interact and affect the overall functioning and stability of the system. The systemic emphasis shifts us away from a narrow focus on the im pacts of isolated technologies, to a holistic understanding of how those technologies fit into broader patterns of social interaction. The communicative ecology concept (and associated research strategies) build communication (19 94 ) and draw inspiration from media ecology theorists of the 20 th century, such as McLuhan (1962) and Innis, whose work illuminated the fundamental role of media in shaping social interactions and structures


73 The communicative ecology concept has also been adopted for applied re search that aims to maximize beneficial impacts of ICT s in developing region s (Slater, et al. 2003 ). To analyze mobile phone use and to understand the significance of mobile phones to rural Fijians, then knowledge of local technol ogical conditions as well as an acquaintance with Fijian culture and social structure is necessary ethnography of Moala (1962) useful a s a template for describing Fijian social structure and my di scussion derives many insights from the w ork of Arno (1993), who studied communication pattern s in a Fijian village in Lau in the early 1970s. T his dissertation is not meant as a comprehensive ethnography and I do not attempt to describe every facet of life in Toba, Cagi, and Veidogo R ather, I focus on those aspects of individual experience and community life that most directly relate to communication practices. For readers seeking a broader treat ment of Fijian culture, there are excellent ethnographies of Fiji spanning from the mid 1800s until the present day e.g. ( (Becker 1995; Hocart 1929; Sahlins 1962; Toren 1990) The V illage and the Community As noted by Wellman (1999), a village is a geographically bounded residential grouping whereas a community is defined by social relationships that transcend geographic boundaries as well as the boundaries of social categories or groups. T he geographic and social nucleus of our communicative ecology is a Fijian village ( koro ). A typical Fijian village is small, with about 50 400 inhabitants. The village proper consists of a well groomed grassy field ( rara ) surrounded by clusters of cinderblock, wooden, and corrugated metal houses. Other centrally located buildings include the church and the community hall, the ce


74 sessions are held. Well trodden paths wind between the village houses, and laundry propped up on tall bamboo poles flutters in the breeze. During the day, women are busy around the ir house s swee ping, cooking, a nd washing clothes, while small children wander through the village playing Meanwhile the men are away at their gardens, which are scattered throughout the forested hills surrounding the village. In the evening older children return from s chool, and youths and men return from the gardens. After the youths play a quick game of touch rugby on the village green showers are taken, evening prayers said, and families share the ir main meal of the day. After dinner men, women and youths disperse t o their respective social circles to drink kava and chat often late into the night. Figure 3 4 Satellite photograph of a Fijian village ( Image DigitalGlobe, Google 2013 )


75 Villages may have a small and unprepossessing appearance, but the interaction s of village inhabitants are guided by a complex calculus in which sex, age, rank, and kinship, among other factors, must be considered. These cross cutting and overlapping social structures form a system of invisible constraints and pathways through which communication in the village flows. The social structure of the village comprise s a number of hierarchically nested patrilineal descent groups that are historically related through shared descent, intermarriage, or pre colonial migrations war s, or allia nce s These corporate groups, known as mataqali communally own the farmland and forest that surrounds the village. After a community member is born s/he is inscribed in a government registry ( Na Ivola ni Kawa Bula ) which denotes his/her descent group af f iliation, and guarantees lifelong right of access to the communal landholding. Aside from land tenure, d escent groups have important ceremonial and political/economic functions in the everyday life of the village. The houses of the members of ea ch descent group are typically clustered together so that villagers are surrounded in their everyday activities by their closest kin. descent group and native village is therefore life long and legally recognized, and ocial life and identity. roles are also strongly shaped by sex and age. There is a clear cut sexual division of labor. Men farm, fish, cut copra, build and repair houses ; w omen weave mats, fish, gather firewood, wash clothe s, cook, clean the house, and care for children. The social lives of men and women are likewise segregated to a strong degree. This is partly a function of proximity men work in the gardens each day, while women are generally confined to the village doing domestic


76 work. At community wide events, men and women sit in separate parts of the community hall, and seating in church is likewise usually divided by sex. Even when men and women are co present at social events in the village it would be considered str ange, or improper, for a man to converse extensively with women. social status and economic role, and also defines the cohort of peers with whom a Fijian socializes. Young children run about the village together with their age mates. O more regimented, but after school hours they play primarily with other members of their age cohorts. Younger children attempting to socialize with members of an older age cohort are ridiculed as boci (uncir cumcised penises) and quickly sent on their way After completing secondary schooling, Fijians join the ranks of unmarried youths ( cauravou and gone yalewa ). Male and female youths play an important economic role in the household assisting their parents w ith gardening and domestic chores, respectively. However, youths also have considerable freedom to socialize, and spend the evenings roaming around the village and drinking kava or homebrew with their peers. Youth ends with marriage, at which point a Fijia n becomes a full fledged adult ( tamata uabula ), and abandons the circle of youths for the more respected, if less exciting, role of an adult. These sex and age based social divisions are on display every evening, when after the shared family meal male and female youths, and married men and women each gravitate toward their respective peer groups to drink kava watch movies, chat, play cards, or engage in some other form of entertainment. In addition to these social structures, Fijian villages have well de fined geographic boundaries. The forest, farmland, and inshore fishing areas surrounding each village


77 are communally owned, with specific, clearly demarcated various constituent mataqali Most villages are located adjacent t o a road with access to motorized transport, although in some highland or remote island areas, villagers must travel by footpath or boat to reach neighboring communities. While villages remain geographically circumscribed, in recent decades the groups of community members who are associated with rural village s have become increasingly dispersed. Since the early 1900s there has been an ongoing process of to larger i slands, and from rural to urban areas. Urbanization has been spurred by a variety of factors, including rising rural populations, falling prices for agricultural commodities, and a lack of jobs, schools and services in many rural areas. Increasing exposure to Western goods and media has also fuelled a desire for an urban lifestyle. Some migrations are circular, with individuals returning to their villages after a period of years working in urban areas, either to raise a family or to retire (Chapman and Pro thero 1985) Over time however, many indigenous Fijian families have settled communities have experienced such extensive migration that entire lineages have vacated the village and decisions about village issues are conducted, in part, through consultations between rural and urban based leaders Despite long absence and infrequ ent opportunities to visit, most urban and overseas dwelling Fijians continue to identify strongly w ith their native villages. The time around Christmas and New Years is a favorite time for migrants to visit the village and reunite with rural kin. Throughout the year m igrants may send money or other forms of


78 assistance to close kin in the village, or ass ist rural villagers when they need to come to the city for work, recreation or schooling. Migrants may also contribute material assistance via urban based development committee s that organize fundraisers for village projects. Through such contributions, ma ny migrants maintain close connections to their native villages and receive a warm welcome whenever they return to visit. However, the level of continued social participation and contributions varies widely and the negligence and absenteeism of some migra nts can be subject of pointed commentary in the village. Thus, while villages are neatly bounded in space, the community of individuals associated with the village is widely scattered, extending wherever migrant villagers have settled and wherever marriage communicative ecology, and will be addressed in their turn. Characteristics of the Study C ommunities Cagi Toba and Veid ogo have many fundamental similarities: they are all rural coastal villages, inhabited exclusively by indigenous Fijians who speak the Fijian language in their daily interactions. Every household obtains at least a portion of its livelihood from subsistenc e farming or fishing. Social life in these communities is built around ties of kinship, and inter relations between kin groups are structured by a hierarchy based on traditional roles. The brief foregoing discussion of Fijian community life applies broadly to these three villages. However, I selected Cagi Toba and Veidogo as research sites because they encompass a broad range of the geographic, demographic, and socio economic variation found among Fijian communities. Toba and Cagi are situated on smaller rural


79 islands in the Lomaiviti group, to the East of the m ain island of Viti Levu, whereas e geographic differences fundamentally shape economic conditions in the study communit ies, which in turn influences migration and various other aspects of life. Figure 3 5 Map of Fiji with study villages labelled. Copyright Daniel Dalet. Labels added by author. http://d map Veidogo, situated near Suva, has the highest mean household income level. In addition to wages and subsistence farming, most families in Veidogo also benefit from the proceeds of land leases. Because the village has a main highway, Veidogo is able to lease land to a variety of businesses and private homeowners. Payouts from these leases are distributed among households in the village, and total approximately $FJ 800 annually per household


80 In recent years Veidogo has been hooked up to the municipal electric grid, so that villagers have metered 24 hour electricity in their homes. The only basic need that remains unfulfilled is water: Veidogo lacks a reliable source of clean water, and depe nds on rainwater tanks for drinking water, while villagers bathe and wash clothes in a nearby stream. The village is currently lobbying the government to have municipal water piped in. Occupationally, Veidogoa ns include an eclectic mixture of lawyers, civ il servants, technicians, day laborers and subsistence farmers. Close relatives, some with tertiary degrees, and others with a primary school education, drink kava together each night, wake up in the morning to catch the bus to Suva where they go to their respective places of work. Many of the women are housewives, but some have tertiary educations and wor k in white collar jobs. As Veidogoa ns have become increasingly integrated within the m arket economy and rely upon government services in their daily liv es, there is some awareness of new vulnerabilities: I heard protracted conversations about the high cost of metered electricity, and suggestions that it might be better to switch to solar panels. Veidogo v illagers are reliant on the bus to make the ir daily commute s to work and have keenly felt the recent hikes in bus fares. Overall though Veidogoa ns enjoy mos t of the benefits of urban life while maintaining a village lifestyle, with its options to fall back on subsistence farming and fishing in times of ne ed. Among the 3 study villages, Toba occupies the other extreme in terms of ease of transport and economic development. Toba does not have a regular ferry service, so villagers rely on small, privately owned open boats to leave the island. E ven the most


81 b asic services and facilities require a journey to Levuka or Suva, which are about 1 and econdary school students must board in Levuka or stay with relatives near Suva, while the elderly and infirm tend to relocate perman ently to Suva for access to medical care Transport of agricultural produce by open boat even to Levuka, is prohibitively expensive. Tob ans must rely on sporadic visits by government ships in order to bring their copra to market. Other than copra, agricul ture is mostly limited to subsistence production. Many households depend on the sale of pandanus mats to pay school fees and productive fishery. However, the unpredictable transp ort situation limits the systematic exploi tation of this resource. As may be expected, Toba has fewer amenities and material comforts than Veidogo. There are a few privately owned generators, but electricity is sporadic and does not reach all the houses in the village. At the household level, people have fewer appliances, and fewer telephones than the other two study villages. Ambitious Toba villagers tend to relocate to urban areas because the potential for economic advancement is very limited on the isla nd. is geographically peripheral by good fortune it lies on the main transport route between two largest islands. Large ferries, capable of carrying bulk cargo, trucks a nd hundreds of passengers, stop by the island twice a week. Cagi is additionally blessed with ample land, rainfall and A gricultural productivity and


82 reliable transport allow Cagi villagers to earn considerable income from sale of cash crops predominantly taro, kava and copra. Each of these crops has different seasonalities and markets, providing a degree of resilience in the face of poor weather or shifting market prices. Cagi villagers explained that villagers harvest co pra on a daily basis to earn small amounts of money for food and other everyday household expenses. Vill agers periodically harvest larger plots of taro or kava in order to pay for larger expenses: school fees, church tithes, annual fundraisers, trips to Su va, or perhaps the purchase of household appliances Women contribute to household income by weaving pandanus mats, which are sold to buyers in the urban areas of Viti Levu. The productive agricultural economy on has spawned some secondary income opportuni ties. From its proceeds, the Cagi cooperative pays a shopkeeper and a copra dryer supervisor. A biofuel plant was recently built on village land; this innovative project extracts the oil from dried copra and converts it into usable fuel, and employs severa l village men. In terms of infrastructure, Cagi has a large village generator that provides electricity every night from 7 10pm. Cagi has a good source of springwater, as well as rainwater tanks. However, the system of pipes in the village requires expans ion, one of priority projects. There is a primary school in the village, and the island has a local high school and hospital, which means that relatively few family members need to relocate to the ci ty for education or healthcare.


83 Tab le 3 1. A summary of the basic characteristics of the three study villages, for comparative purposes Variable Toba Cagi Veidogo Population (2011 2012) 9 4 326 124 # h ouseholds 19 57 23 Mean residents/hh 4.8 5.7 5.4 Mean a ge 30.3 27.6 27.7 Median a ge 3 0 22 25.5 Social structure of village 1 yavusa (6 constituent mataqali) 2 yavusa (3 mataqali in one, and 2 mataqali in the other) 1 mataqali Social structural complexity Medium High Low Geographical situation Rural, s mall outer island Rural, m edium siz e outer island P eri urban main island Mode of t ransport to Suva Open Boat Ferry Bus Cost of round trip to Suva ($FJ) 60 (varies) 110 2.40 Mean # of trips away from local area/hh/past year 2.2 0.8 1.4 Main sources of income Copra, Handicrafts Taro, Co pra, Kava Wage work, small business, fishing Relative income level Low Medium High Mean annual hh income ($FJ) 1771 5764 11156 Electricity source Private Generators Village Generator Municipal Grid Electricity hours/day Irregular 7 10pm daily 24 hours Drinking w ater source Spring fed/rain Spring fed/rain Rain Locally available schools Primary Primary/Secondary Primary Tertiary Locally available healthcare Nurse Rural Hospital Urban Hospital Phone networks available Vodafone, TFL Digicel, TFL Vodafon e, Digicel, TFL Mean # phones/person .26 .34 0.46 Mean # phones/hh 1.26 1.96 2.48 *I counted Suva as part of the local area for re sidents of Veidogo. Most Veidogo residents visit Suva at least once a week. Population tree diagrams for the three village s indicate how these geographic and economic factors influence migration and ultimately the village social structure s Toba migration by young adults. Village elders in both Toba and Cagi estim ated that well over half of the


84 community members have yet to step foot even once i n the village. Veidogo, by contrast, h as a more uniform age structure, because young adults can seek higher education and work in the immediately surrounding area. In fact, sever al Veidogo households are recently arrived having settled in the village thro ugh ties with local kin, in order to take advantage of the many opportunities that life near the city offers. Figure 3 6 Population tree diagrams for Toba, Cagi, and Veidogo villages


85 Aspects of Fijian K inship Rural Fijian communities are structured b y kinship. Fijian kinship incorporates two fundamental principles: descent within patrilineal groups, and intermarriage between descent groups. These principles, along with the norms of social interaction associated with each kin relation, play an importan t role in shaping Fijian communication patterns. The Fijian system of kinship can be most easy described in terms of the intermarriage of two patrilines (Groves 1963) Each patriline is a corporate descent group, whose members stand to one another in the r elationship of parallel kin. The terminology and behavioral etiquette for parallel relatives mimics that of the nuclear family. Thus, men in the same generation of a given descent group stand in the relationship of brothers ( veitacini ) and those in the fir st ascending or descending generation are related as fathers and sons ( veitamani/veiluveni ). The Fijian emphasis on primogeniture is reflected in parallel kin terms that differentiate based on birth rank Same sex siblings call each other by different ter ms ( tuaka or taci ), depending upon their relative birth rank. Children also refer to their same sex siblings based on their birth rank relative to the parent. tamana levu (big father) and mo sister is tinana lailai (little mother). Seniority, whether age based, generational or through bi rth rank, confers authority. Junior kinsmen must show respect to their seniors and defer to their judgements and decisions Daughters belong to their natal descent group, but their structural p osition is


86 and her economic contributions and offspring thereafter belong to her affines. Marriage in Fiji is usually (but n ot always) kin group and village exogamous. Only 15 out of 99 (15%) household heads in the study villages had married a partn er from the same village. Even t his figure is probably unusually high for Fiji due to the very high incidence of intra village mar riage in Cagi The norm, formerly more strictly observed, is th at marriage can occur only between cross cousins. If a bride and groom are in fact otherwise related, there is usually an alternative kinship path that allows them to be reckoned as cross cous ins, or they are simply labeled as such after the fact When a man and woman marry, the n the members of their respective patrilines become oriented toward each other in relationship of cross kin. A marriage binds the two patrilines through kinship and cre ates enduring ties and obligations. formalities and group level exchanges leading up to the wedding, and is periodically reaffirmed after the wedding with rites surrounding t he birth of the first child, the kau mata ni gone ) and mutual contributions to the life crisis rites of shared kinsmen ng social connection and privileged ba sed on the special MoBr SiSo tie known as vasu In many areas of Fiji, children are alternately named after paternal and maternal kin. Marriage therefore initiates a multi generatio nal social and economic bond between groups.


87 Cross relatives do not work closely together as a corporate unit, as do members of the same patriline. However, the fusing of separate patrilines, with all of the associated po litical and economic benefits and obligations, makes cross relationships extremely important in their own right. In contrast to the hierarchical parallel relationships, c ross relationships are characterized by egalitarian forms of interaction, which nevert heless are guided by distinct forms of etiquette. Fijian kinship is classificatory, meaning that the categories and terms applied to extended indefinitely to apply to any relative, no matter how distant tacina ) include not only actual same sex siblings, but also FaBrSo, FaFaBrSoSo, or anyone else cousins ( tavalena ) include not only his M oBrCh or FaSiCh, but by extension anyon e else related as a sibling to his MoBrCh or FaSiCh According to the classificator y principle, a Fijian can quickly determine his or her relationship with a ny stranger, as long as at least one kinsman, no matter how distant, is held in common. Although th e same kinship terms are applied to bot h close and distant kin, behaviorally and practically speaking kin are distinguished based on their distance veiwekani dina (true relatives). This term implies an actual ge nealogical connection, perhaps descent from a shared set of grandparents or great grandparents. Other kin are simply veiwekani a term that denotes some sort of connection but not a genealogically close one Kinship, C ommunication and E tiquette When Fijia ns talk about their kinsmen they typically define their relationship s in terms of the nature of communicat ion that is allowed or expected within that form of


88 relationship. For example, a Fijian might say of his vugona (MoBr), a relationship characterized kinsmen have a shared template based upon which they may model their interactions. The idiosyncracies of individual personalities tend to reced e into the background, as Fijian kinsmen beh ave in conventional ways that are appropriate to their shared tie. As Arno pointed out, if the Fijian social system is built around kinship, and kin relationships are defined by rules of communication, then the rules of communication are the key element in the Fijian social system (Arno 1993) Fijian kin relationships can be broadly divided into hierarchical relations hips and egalitarian relationships and sp e cific types of kin ties are additionally defined by joking, respect, or avoidance behavior patterns. These prescribed patterns of behavior tend to be most pronounced with close kin. With more distant kin, greater freedoms are taken with the form of intera ction and relationships more closely approximate generic veiwekani ). Hierarchical relationships predominate among parallel relatives (i.e., members of the same patriline), while egalitarian relationships are characteristic of cross relatives (in laws). Interactions among parallel kin mimic the hierarchical parent child and sibling sibling relationships that structure the nuclear family. Senior individuals may speak freely to their juniors, order them about, and criticize or insult them if nec essary. In contrast, junior individuals are restricted in the kinds of things they may say to their seniors, and anything a junior says must be expressed with the appropriate respectfulness honorific terms of address, and subservient bodily comportment T here is not much latitude for joking and free ranging conversation among closely related


89 parallel kin As Arno and Sahlins pointed out, this does not mean that social interactions within households and within descent groups are necessarily unpleasant. Rath er, it means that much of the interaction within descent groups is often for instrum ental purposes and is respectful in tone. The husband and wife relationship is likewise unequal. Fijian spouses rely extensively on one other to fulfill vital economic rol e s in the household, but t he wife is ultimately subservient, and defer s to day affairs of the household the husb and may order his wife around, issue harsh criticisms or even (to a degree) use physical violence agains t her A woman may flee to live with her may seek a formal divorce. Husbands and wives do not openly show affection in public. However, many Fijian husbands and wives are privately affecti onate, confide closely in each other, and work tog ether harmoniously as a team to provide their plan its future course. The hierarchy that pervades Fijian social relations is communicated both verbally and non verbally through demonst rations of respect ( veidokai ). In old en tim es Fijians had a chiefly language register in which specialized wor ds and euphemisms were used to refer to the bodies and possessions of high ranking individuals (Arno 1993 :xv). The use of this register has large ly disappeared but the emphasis on rank distinctions remains Fijians address elders or high ranking individuals with honorific plurals ( kemudrau/kemud ou/kemuni ) to indicate respect. C eremonial events are the occasion for flowery rhetoric and elaborate de corum including a strong emphasis on rank based seating arrang e ments and order of precedence in kava drinking. Eloquence is a highly


90 valued attribute; v akaturaga (chiefly) speech appropriate to such formal occasions, is delivered in a dign ified tempo and pitch (Arno 1993 :xv) Respect is expressed non verbally through bodily comportment and the use of space within buildings in the village. Every village building has a designated i ra ) and side ( i cake ). These dimensions are in fact situa ted within the horizontal plane Low d promply sit down on the floor to show their respect to the owner of the house. The high side and its side wall doors are reserved for the male owner of the house, and f or honored guests. Within dwelling loga ) is located behind a wall or curtain at the highest end of the house, while food preparation and other mundane activities occur at the low end. Figure 3 7 Overhead diagram of a Fijian house During meals, family members range themselves in order of seniority along the length of the room, with the wif e at the lowest end, serving out food from the pot. During events in


91 the community hall the chief and community elders sit on the high end of the hall and women sit at the low end. Mid ranking men sit ranged along the wall roughly in order of their pre cedence in the community. Figure 3 8 Members of a patrilineage at a Christmas meal. At this shared meal ( kanavata ), the indiv the food. Photo courtesy of author. There is also a vertical component to th ese hierchical spatial relations S tanding while other people are seated is consid ered a prerogative of high ranking individuals; everyone else in the room must remain seat ed. If some one must stand up or walk across the room, they are expected to hunch over while apologizing by repeatedly mutterin g tilou (excuse me). Thus, Fijian social interaction has an inherent hierarchical component t hat is demonstrated by terms of address, word choice and tone of voice


92 the relative locations of individuals in a building and their bodily comportment as they s it, speak, or move about In social interactions, hierarchy manifests itself in the in the constant deferment of juniors to the opinions, judgments, decisions, and requests of their seniors. Community meetings are dominated by the opinions and judgements o f the v illage older men. Within smaller social events or work groups, the senior individual present likewise assumes the mantle of leadership. Younger, lower ranked individuals more often than not simply remain silent during the deliberations of their elders. Considerations of seniority affect the allocation of economic resources, and can override notions of personal property. One day I was walking through Cagi village with Saki, the 17 t a levu toward us his dog loping by Karaik Pita whistled enviously at Saki under his breath. se, p robably off, t ordered, as he motioned to the boy to step behind the corner of a house. Without a old cutoff jeans, while Pita new athletic shorts. In such situations, juniors must simply comply with the demands of their elders. One of the most stinging criticisms that can be leveled at a Fijian is that he is viavia levu overstepping his prope r social role A senior kinsman can easily silence his jun ior by dismissively calling him boci ( un circumcised penis), a shame inducing reference to his immaturity and social inferiority. When a junior


93 individual is being punished or verbally castigated by an elder, he will usually simply be silent and look downward, the characteristic expression of madua ized in rural Fijian society Relationships between fathers and sons, and elder and younger same sex s iblings can become strained over time by the imposition of strict discipline. Later in life, when they no longer live together in the same house, fathers and sons, and elder and younger siblings may still bear grudges and may rarely speak together at lengt h. In contrast to parallel kin, relationships among cross kin are broadly egalitarian in nature. The epitome of the egalitarian principle is found in the cross cousin relationship ( veitavaleni ). Same sex cross cousins should be open and friendly to each o ther at all times, never showing anger. Cross cousins can communicate freely about any topic an d are expected to joke together. For example, cross cousins characteristically rib one another about their shortcomings, sometimes in outrageous ways. They have strong ss cousin requests something, the request is very difficult to deny. As a result of this open, mutually supportive relationship, same sex cross cousins are typically close life long friends and confidants. T hey are favored kava drinking partners in adulthood, because of the wide latitude for joking and ability to presence (Arno 1993 ) Opposite sex cross cousins ( veidavolani ) stand in the relatio nship of potential spouses. Althoug h their socializing is limited in scope because they are opposite sex, they are expected to flirt with each other in a very public, exaggerated, and even formulaic manner. Such flirtation does not imply actual attraction; it is merely the expected form of behavior. Male and female c ross cousins might for example, loudly


94 tease each other with sexual innuendos while in the presence of other villagers this is proper form. If these sorts of flirtations occurred privately, however, it would be inappropriate b ecause it would be assumed that the cousins were actually carrying on a n illicit relationship One time during a village dance, I witnessed an 18 year old youth being summoned to dance by a young married woman. They were both ski llful dancers, and performe d a very flirtatious dance, to the cheers and laughter of everyone in the hall. Puzzled, I whispered to a young woman married to to Yes, but they (the dancers) are cross cousi ns so they can act like In egalitarian relationships, great respect takes the form of avoidance Cross uncles and nephews ( veivugoni because the SiSo represents the fruit of the marital bon d that joins two separate patrilineages. The SiSo must show great respect to his MoBr, and as a sign of this respect these relatives usually avoid speaking altogether. However, the SiSo may turn to his MoBr for assistance if necessary, and the MoBr will do his utmost to honor the request. MoBr SiSo is a relationship of deep mutual obligation. Another such respectful avoidance relationship is that between brothers and sisters ( veiganeni ). After puberty, brothers and sisters avoid speaking to one another, and in some areas of Fiji they may not sleep under the same roof If post pubescent brothers and sisters must convey a message to each other, they typically do so through an intermediary. Furthermore sisters and brothers must not make or overhear references t o sex (even joking ly great embarrassment and one or both siblings would flee the scene.


95 This respectful avoidance associated with veivugoni and veiganeni relationships gives rise to physical feelings of discomfort or embarrassment ( madua ) when in each For example, I was sitting in the kitchen after breakfast one day with In Fijian kinship, a classificatory MoBr SiSo relationship) ap proaching the house with a bag of groceries. Neither of them made eye contact or said a word in greeting ; Taniela simply laid the groceries on the table, got some food from the stove and sat on the floor in the next room to eat his breakfast in silence O n another occasion I was at a kanavata a meal shared among al l of the members of a mataqali to celebrate the birthday of a senior member. As I sat down at usually exte nded to guests. Thinking that I had mastered the conventions of Fijian hospitality, I steadfastly refused their invitations more humble place. It turned out that my hosts really had just wanted me to move to another s eat by sitting where I did, I forced a nother man to sit directly across from his momo (MoBr), an awkward situation for both individuals T he expression of these relationship specific forms of behavior tends to diminish with the distance of the kinship ti e. Thus, the hierarchical behaviors described are most pronounced among close parallel male relatives. The joki ng, avoidance or obligation associated with certain cross ties likewise tend s to be more pronounced with immediate relatives. A few villagers tol some of these behavioral norms o ne young man said that he freely speaks with his


96 cross uncles, because he prefers to be friendly and open with them rather than remaining silent all the time Also if specific individuals share an unusually close bond, their personal feelings may associated with their tie. Not least among these exceptions, today a considerable proportion of Fijian couples marry eac h other despite the fact that they are not cross cousins. Thus the extent to which kin behavioral norms are observed can vary from individual to individual, community to community and region to region. By and large however, these norms of behavior heavily shape social interactions among Fijian kin. Individuals are constrained in terms of whom they may communicate with, what messages may pass between them, and the style or manner in which the message must be communicated. Information flows through appropria te channels, with the etiquette appropriate to the message and the receiver Many messages need to flow through suitable intermediaries in order to eventually reach their in tended target. One of the topics to consider then, is how mobile phones and other ICTs might reroute or social barriers and whether the lack of co presen ce in phone interactions might permit greater latitude in interactions between kin. Social S tructure At the time Great Britain annexed Fiji in 1874, a variety of political forms existed within the archipelago. Some areas with extensive Polynesian influence, suc h as the Lau islands, exhibited distinctly hierarchical social organization, while other areas had the more fluid sort of leadership structures typically asc ribed to Melanesia. Adopting a strategy of indirect rule, the colonial administration sought to preserve aspects of native leadership, while standardizing local political institutions throu ghout the co lony. Colonial Lands Commissioners embarked on a proce ss of cataloguing each


97 conforming to a standardized neo traditional template. In the process complex local realities had to be shoehorned to fit the new model During the past century local rivalries, as well a s natural demographic change, have given rise to structural ambiguities in many communities. Moreover, Fijians often give contradictory accounts of local s ocial structure, and use terms such as mataqali and y avusa in ambiguous ways (Sahlins 1962; Arno 1993 ) Those caveats aside, the following description captures the basic principles of Fijian social organization. Fijian social struc ture is pyramidal in arrangement It consists of a hierarchy of progressively more inclusive patrilineal descent groups, begin ning with households at the base and culminating with very large, inclusive groups that share descent from a mythical common ancestor. These apical groups may be dispersed over a wide geographic area and incorporate individuals with little if any recent g enealogical ties Fijian social structure therefore incorporates aspects of both descent based kinship organization, and regional political alliance.


98 Figure 3 9 Fijian social structure (simplified diagram) Social relationships within and between Fijia n descent groups, are ranked according to se niority. Thus, the i tokatoka within a mataqali are accorded seniority based on the relative birth rank s of their founders, often conceptualized as a group of brothers. Within descent groups, members of younger g enerations are subservient to members of older generations, and within each generation, those born later are subservient to those born earlier. D escent groups strive to present a strong, u nified face within the community, to enh ance their political influe nce and prestige and to protect their members If an individual becomes involved in a dispute, or commits a transgression, it reflects upon the entire descent group. S enior members intervene to determine a solution or to make amends to the leaders of the aggrieved descent group. Alternatively, if an individual is being persecuted his descent group through its leadership structure, will work to


99 protect him. The internal hierarchy of Fijian descent groups promotes decisive leadership and efficient cooperat ion, and discourages dissension in the pursuit of these shared goals. The inter relationships of higher order descent groups ( mataqali and y avusa ) are additionally defined by ceremonial roles and political alliances Ceremonial roles are part of a caste li ke system that distinguishes between chiefly lineages ( turaga ) and vanua ) (Hocart referred to these divisions as moieties) In indigenous oral histories, lineages typically represent the indigenous groups that originally inhabited a given territory, while chiefly lineages are said to have arrived more recently often by sea. The land people, in order to achieve harmony, settle pre existing disputes over leadership, or to assist with defense from some outside threat, ceded leadership to the chiefly outsiders. At the base of the pyramid of Fijian society is the household, the fundamental social and pr oductive unit in Fijian villages Just as the village has a chief, a married man is referred to as the chief ( turaga ni vale ) within his o wn household, and his wife is m arama ). H ierarchy permeates all parent child and sibling sibling interactions within the household. The male head of household directs the labor of his sons in the garden The housewife likewise oversees the training of her daughters and supervises their housework. Older siblings direct their younger siblings, and may chastise them for improper behavior. This hierarchy functions to create an unambiguous basis for delegat ing tasks. S ahlins felt that the Fij ian household hierarchy goes well beyond enabling a smoothl y functioning household economy: it is a microcosm of the hierarchies that structure Fijian society at higher levels (1962:105).

PAGE 100

100 Extended patriarchal households were formerly the norm in Fiji. Now adays newly married couples normally establish a n independent considered unusual but not abnormal. In Toba Cagi and Veidogo 10 out of 99 households (10%) featured a husband who had settled i common for a variety of non nuclear kin aunts, nephews, widowed parents to subsequently attach themselves to a nuclear household on a temporary or permanent basis. In the 3 communities in which I work ed, 42 out of 99 (42%) of households were nuclear, and the rest incorporated a variety of kin from outside the nuclear family. A number of closely allied households comprise the base level Fijian descent group, known as the itokatoka ( sublineage). The itok atoka is a group of close patrilineal kin perhaps originally a group of grown male siblings and their families, which over time branches out to incorporate more distant patrilineal relatives. The households in a n i tokatoka frequently work together to acco mplish shared economic and ceremonial objectives. When there is a vi llage wide event requiring contributions from the various kin groups, i tokatoka members cooperate closely and pool their resources Because they are closely related and share the same circ le of close kin, itokatoka members must also p ool ceremonial goods, raise funds, and delegate responsibilities associated with weddings, funerals, and other kin centered ceremonial events. Several itokatoka form a mataqali (subclan). The colonial governmen t early on decided that native land should be owned communally at the mataqali level. Thus po litical and demographic dynamics within mataqali affect the allocation of plots of land, a very important aspect of the economic life of the community. Each m ataqa li has a

PAGE 101

101 traditional, caste like role within the community such as chiefs ( turaga ) spokesmen ( matanivanua ) and warriors ( bati ). On ceremonial occasions in the community, senior representatives from each mataqali assume their respective roles in the proc eedings. While these roles are today largely ceremonial, they occasionally have important economic and political implications, particularly with chiefly succession s The mataqali like its constituent i tokatoka functions as an important corporate group f or marshal l ing communal labor and contributions. For example, the accounting for the annual village fundraiser in many villages is split into mataqali sections, with each mataqali responsible for getting its constituent households to contribute their requi red share.The mem bers of a mataqali are closely related, and as with the itokatoka frequently work together on kinship ceremonial, such as weddings and funerals. The broadest local level group in the Fijian social structure is the yavusa (clan) comprisi ng several mataqali It should be noted that the term yavusa is used somewhat ambiguously by Fijians It can refer to the formally defined groups catalogued by the Fiji government, but the term can also refer to the group of people with whom an individual has kin ties egocentric yavusa as a formally defined descent group, is sufficiently broad that its constituent s may be distributed over multiple villages, although it is also p ossible for one village to contain part or all of one or more yavusa Yavusa membership is based on the principle of share d descent but yavusa may also incorporate historically unrelated mataqali that formed alliances during precolonial days. Hence, the y avusa exists at the boundary between the realm of kinship and the realm of regional politics.

PAGE 102

102 The rule of exogamy within yavusa tends to be absent (Nayacakalou 1978). Unlike the itokatoka or mataqali the yavusa is not frequently used as the basis for orga nizing labor for communal tasks; rather, the main function of the yavusa is as a unit of ceremonial and social identification, or to determine precedence of accession to traditional leadership positions. Fijian village s are comprised of a variable arrange ment of descent groups. S ometimes a village consists of a yavusa in its entirety sometimes a fraction of a yavusa and sometimes portions of multiple yavusa There may be mataqali or itokatoka in a village that joined the community long ago b ecause of war, alliance, or traditional service to a chief. Whatever the configuration of descent groups, they stand to each other in defined ceremonial roles with relative hierarchical ranks. The three study villages have social structures with varying de grees of complexity. Cagi a large village, has a relatively complex social structure, with two clans ( yavusa ) that are subdivided into 5 constituent mataqali T hese mataqali in turn, are subdivided into even greater number of patrilineages ( itokatoka ). T he association of the 5 mataqali in Cagi is partly attributed to shared descent, and partly to historical in migration of kin groups from neighboring villages. In Toba there are 6 mataqali comprising a single y avusa However, three of the mataqali have bec ravi ) on the three more populous mataqali These arrangements are reflected in the v configuration: each of the more populous subclans occupies a different corner of the village along with the sma ller mataqali that leans upon it.

PAGE 103

103 Of the three study villages, Veidogo has the simplest, mo st cohesive social structure. Veidogo comprises a single mataqali (subclan). Decades ago, this mataqali was part of a large, rapidly growing village, which was subse quently divided into 4 smaller villages, each comprising a single mataqali There is consequently less basis for intergroup rivalry within the village and it is relatively simple to obtain consensus in decision making. However, Veidogo is in another sense more diverse than the other villages, because several of the heads of household have origins in other regions of Fiji. Because the village is so favorably situated near the city, several of the village men from other parts of Fiji and s ettled in the village of their wives. Leadership Ratu Sukuna, a famous Fijian statesman, once characterized Fijian society as a three legged stool resting upon the traditional leadership ( vanua ), the government ( matanitu ) and the church ( lotu ) This trip artite administrative structure remains relevant in village life today. Vanua government, and church each have leadership hierarchies within the village, and each hierarchy ascends stepwise to the local, regional and national level. M essages and decisions originat ing at the higher levels filter down to the villag e via the local leaders, through established lines of communication with their parallel and higher level counterparts. Local church and government leaders may derive specialized institutional know ledge and influence through particip ation in regional or national lev el conferences or decisionmaking bodies. Members of the village who are not high ranking by birth may seek to distinguish themselves through service in the church or local government as a means of upward social mobility. C ommunal work in Fijian villages can be distinguished based upon the leadership structures through which labor is marshaled. Cakacaka ni vanua (works of

PAGE 104

104 the land) secure cooperation by order of the chief to the heads of t he various social units through their senior figures An example might be a ceremonial presentation of mats made on behalf of the village at the death of a local official. In cakacaka ni y avusa (works of the kin group or egocentric kindred ), a person seeks assistance through their relatives, the particular configuration of which is unique to each individual (Nayacakalou 1978:14). Such kin centered events include weddings and funerals. A third type of task cakacaka vaka matanitu is required and/or sponsore d by the government, and marshals cooperation via the operation of the village meeting ( bose ) and committee leadership. Such tasks include the labor of constructing new village latrines with funding for materials provided by the annual village fundraiser ( soli ) or the provincial government. Traditional leadership within Fijian villages is provided thro ugh the ranked structure of local descent groups The highest ranking member of the predominant local group is typically considered the turag a ni vanua ). Thus in a one y avusa village, the turaga ni y avusa ( y avusa chief) is considered the chief of the village. He is by the same logic the head of the senior mataqali within the y avusa and the senior itokatoka within that mataqali Leadership of t he various non chiefly mataqali and it okatoka is delegated to their most senior members ( turaga ), who are however ultimately subject to the higher ranking turaga ni yavusa above them. By extension, the chief of the village is subject to higher ran king para mount chiefs, who may live in distant areas of Fiji, but are connected through traditional alliances. Hereditary chiefs are accorded special respect and play an important moral leadership role in exhorting community cooperation and promoting religious obse rvance However, many of the

PAGE 105

105 traditional prerogative s of chiefs to marshal labor ( lala ) receive the first fruits of the harvest ( sevu ) and punish transgressions within the village have largely been usurped by the government. The local representative of t he Fijian government is the turaga ni ko ro (TNK ) a village man elected for a set term of years by his fellow villagers. The quite distinct from that of the turaga ni vanua ( hereditary chief), and in fact the TNK need not be a high rankin g individual at all. The TNK is responsible for chairing the monthly village meeting and the village committee, for overseeing projects and maintenance activities in the village and for liasing with provincial government officials Figure 3 10 The C agi wo meets for tea in the community hall Photo courtesy of author.

PAGE 106

106 The various affairs of the village are run by committees made up of rank and file villagers committee, and so forth. The TNK is the head of the village committee ( komiti ni koro ) annual fundraiser ( soli ) The various other c ommittees may hold smaller fundraisers to raise the money necessary for their respective projects. Many villages also have a government primary school, and/or a nursing station staffed by government employees posted ( lesi ) from other areas of Fiji, with their housing and salaries paid by the Fijian governm ent. However, these employees do not generally take part in the governance of the village that hosts them T he hierarchy of religious leade rship depends on the mix of religious denominations present in the village. Methodists are the dominant Christian den omination in Fiji although Catholics are substantial minority, and in recent decades a variety of evangelical churches have gai ned followers In the case of a Methodist village the Church assign s a resident vakatawa (preacher) to lead the local church. I f the village is the designated headquarters of the local church circuit, it host s a talatala (ordained minister). Vakatawa are typically locals of nearby villages, while talatala during their careers complete tours of service all over Fiji. Among the loca l villagers there are also several lay preachers ( dauvunau ) who attain their position through training and service to the local church Village wide issues are discussed at the monthly village meetings ( bose vakoro ) attended by all villagers The TNK pre sides over the parliamentary style meeting while the hereditary chief provides an opening and closing speech. The head of the local

PAGE 107

107 church leads an opening and closing prayer. Each committee leader in turn presents briefly on the issues or projects they a re currently engaged in. Local officials may attend the meeting to explain new government policies or to inform villagers of relevant public health, development, or legal issues. Villagers in the audience can then ask questions, provide feedback, complaint s, and recommendations. If new initiatives or fundraisers are decided upon responsibilities for funding and labor are delegated and timetables are establi shed for their execution. Often these responsibilities are divided at the village meeting among the various mataqali which later independently hold internal meetings to delegate responsibilities, collect donations, and e stablish targets and timetables for the completion of tasks. Village meetings can also serve as a venue for the airing of grievances a s, who perhaps exasperated by unresolved personal conflict s, submit them before the judgement of the public forum. During my stays in the study villages, m ealtimes that followed the village meetings usually provide d interesting gossip and pointed critiques about the proceedings and the contributions of the various villagers. Implica tions of Social Structure for C ommunication Overlapping constraints Given the preceding description of Fijian social structure, we can c onceptualize how information flows within and across the boundaries of the various social, kinship and administrative structures within a Fijian village and speculate as to how the introduction of mobile phones might influence the nature or direction of t hose information flows. Furthermore, we can formulate predictions, based on the various sex,

PAGE 108

108 age, rank, and kinship based communication. The first key aspect of Fijian social structure is hier archy within each tier or institution of Fijian society, whether household, sub clan, village, or church circuit, there is a well defined leadership structure. L e aders (or senior kin ) are considered the repositories of group histories and expert knowledge and their judgements are r arely publicly challenged or questioned unless by someone of comparable rank Particularly within patrilines, there is a strong pattern of deference to senior members, while junior members provide few suggestions or criticisms. T here is an in grained tendency to defer to the acknowledged authorities in any matter. Thus, information and decisionmaking have a pronounced top down flow. A relatively small group of individuals play a disproport i onate role in making group level pronounce ments and decisions, and those individuals tend to be older, male, and married. A second key aspect of Fijian social structure is the clear division of social and economic roles by sex and age/marital status. Men and women have limited interactions in thei r daily work, and spend most of the daytime hours in separate areas: men in the gardens, women around the house Because women and men have clearly separated economic roles, knowl edge of specific techniques and economic resources is concentrated in one gr o up or the other. Likewise, Fijian villagers tend to socialize nearly exclusively with fellow members of their age/sex peer cohort Every night after dinner youths, and married men and married women gravitate toward their respective social circles to drink kava watch movies, chat, or engage in some other form of entertainment. There is a preference for

PAGE 109

109 g within the wrong group. I nteractions that do occur across age/sex cohort li nes are often constrained by seniority distinctions that prohibit members of younger age cohorts from speaking freely. Beyond the constraints on open interactions between men and women, and junior and senior kin, there are also kin relationships defined b y respectful avoidance in which individuals may not co nverse freely These include the brother sister relationships ( veiganeni ) and cross relationships in the first ascending/descending generation ( veivugoni ). As Arno observed th e co presence of parents and children, siblings, or cross uncles and nephews at a social event puts a one of the people so related, usually the junior of the two leaves the group in order to relieve the generally felt awkwardnes Finally, geographic distance creates hurdles and rifts in a community communicat ion network. The everyday occure nces of life in the village are largely unknown to community members living in urban centers, or overseas. Likewise, events that have occurred among the population of urban based community members may only be learned face to face through the occasional visitor to the village For this reason, the Christmas holidays have always been a time for catching up and visiting between relati many years. After an absence of several years, a city dweller may return to the village, meet for the first time children who have been born in the intervening time, visit the graves of close relatives who have died, and absorb a variety of entertaining shocking, or salacious bits of gossip about local affairs. Overseas Fijians tend to be the most isolated from the flows

PAGE 110

110 of information that circulate among community members, due to their distance and the difficu lty and expense of calling and visiting. Crossing social boundaries Although Fijia n communities are cross cut by a variety of age, sex, kinship and rank based barriers to free communication, t here are some key points of articulation in the social structur e which permit information to spill across these boundaries. In the realm of kinship, cross cousins play an important role in this regard. Cross cousins may joke together about sensitive topics, such as personal foibles or conflicts in the village, in a wa y that defuses the issues or at least allows some degree of dialogue or closure Figure 3 11 A group of men gather for an afternoon kava session Photo courtesy of author. Arno (1992) studied the preferred kava driking partners of individuals in a La uan village and found that there was a marked preference for drinking with cross cousins or

PAGE 111

111 with very distant kin of other categories ( the social distance serving to reduce the strength of behavioral pro scriptions). Among a group of such freely assoc i ating individuals, kava sessions can become an uninhibited forum for joking and debating about village issues that could not be raised in the presence of close parallel kin. Arno found that such joking plays an important role in defusing conflicts, but also in transferring information between different circles of kin. W omen also play a key role in transferring inform ation across the boundaries of patrilineages Women are a point of articulation between their natal kin groups and those of their husbands. Particul have privileged access to i nformation about topics of political or economic interest in the local area Within the village, c riticism s which may not be openly discussed around the i close agnates, may be freely discussed around his wife or the wife of an other agnate (Arno) This is because women are not considered full fledged membe kin group s A woman, perhaps exposed to gossip about her hu sband or his close agnate during a group mat weaving session, can later privately relate this gossip to her husband and the information may then circulate In a more general sense, spouses in the privacy of th eir home can exchange information that was learned in their respective sex segregated social circles Mobile phones and transcending barriers to communication The introduction of mobile phone s may soften or reconfigure some of the aforementioned constrain ts on communication. For example, m obile phones may allow ambitious subordinate members of kinship groups, or of the local church or government hierarchy to overstep their organizational ranks Such an individual might use a mobile

PAGE 112

112 phone to make direct con tact with organizational counterparts in Suva or in other parts of Fiji, thereby pre empting established decisionmaking processes or lines of communication. For example, a n active and ambitious head of a village youth group might directly contact NGOs in t he capital city to seek funding for local projects. This could enable individuals to transcend to a degree, the strictures of the local social hierarchy, and pursue upward mobility Indeed, ICTs may generally serve to undermine hierarchies, because contro l of information is a key as pect of authority. In terms of age and sex cohorts, phones enable men and women (both married and unmarried) to speak more freel y with one another than they could communicate face to face in the atmosphere of constant surveil lan ce of the village A young man and woman might be romantically interested in one another but are unable to meet or speak freely du e to the watchful eyes of disapproving parents, siblings or o ther relatives. They may use mobile phone s to establish contact, flirt, or build rapport, and eventually even to arrange private rendezvous. Outside of sex or romance, certain men or wome n may have a social desire to interact with members of the opposite sex, more so than is considered acceptable by village norms. Such an individual could use the phone to have lengthy conversations with friends or relatives of the opposite sex without attracting attention or censure in the village In terms of kinship mobiles may facilitate communication between individuals standing in avoidance relationships who would not be able to converse freely with one another face to face. Many of my informants said that talking to a cross uncle or an opposite sex sibling was much easie r over the phone, than in person. They stressed that seeing s uch relatives face to face provokes a feeling of shyness ( madua ) that is not

PAGE 113

113 strongly felt over the phone. Perhaps over time phone contact between these relative s will lead to an erosion of avoidance relationship s in Fijian kinship. T he most obvious const raint which the mobile phone helps to overcome is geographic distance. Before mobile phones became widely available, conversation and news trickled irregularly betwee n the village and urban areas. O nly at a time of extensive visiting, s uch as the Christmas holidays, c ould people really catch up on the news and gossip from the other side. Phones have, in a sense, turned every day into Christmas, allowing Fijians to remain in close touch throughout the year with kin living in distant places community Often, villagers migrate to the city or overseas for primarily economic reasons, and after obtaining employment are able to send money and gifts back to the village to assist their families. The economic contributions of migrants are greatly appreciated, but their presence in the village is keenly felt by close kinsmen. In this respect the phone is considered a great blessing by many rural Fijians. One village woman showed me a text message she ha d received from her brother, living in Australia, earlier that day. The contents were nothing special saying hello, asking about the weath er, saying he would call soon. However, t he woman said that when she received the message, after not hearing from her brother f or several months, it made her weep with happiness. However, there a re also negative aspects of the intensified communication between village and city Villagers often noted that phones allow malicious gossip to instantaneously travel from the vi G ossip subsequently circulates within urban kin networks, and ultimately returns back to

PAGE 114

114 the village. Sometimes we (village dwellers) are surprised when we get asked over the phone by someone in Suva about a fight or argument that occurred in the village earlier that same day like someone else in the village told them about it and they want to know more when this happens I just prete nd as if I know nothing about the situation Others, women especially, said that gossips or villagers with grudges might tell slanderous lies to u reputation s Thus, local conflicts leak outside the village boundaries and are disp ersed among a broader aud ience. Such free ranging gossip has little to do with resolving the roots of the conflict, and may serve to exa cerbate village conflict s through the addition of embellishments and rumors. Technolog ical and Material Aspects of the Communicative Ecology Bey ond social factors, communication is shaped by material and technological constraints and opportunities. M obile phones articulate with other locally available technologies in antagonistic or synergistic ways. Technological development in many rural, devel oping regions is highly uneven: r emoteness, or scarcity of key resources and specific forms of knowledge can result in tec N eighboring c ommu nities may vary w idely terms of the availability of basic services such as water, el ectricity, and motorized transport. In this section I focus particularly on the inter relationships between three technological sectors: telecommunications, energy and transport and how they affect communication in Toba, Cagi and Veidogo Any of these t hree sectors might serve as the limiting factor in the smooth functioning of social or economic processes in a village. In Cagi for instance, the village generator broke down for nearly the entire 3 months I was present in the village. This

PAGE 115

115 made it difficu lt for villagers to operate lights and other appliances, but it also affected communication, because phone batteries were often drained, so that outgoing calls could not be made and incoming calls could not be receiv ed. Transport was less impacted, because the trucks and ferries that provided local transport did not depend on go ods, because their phone batteries were often dead. In Toba, villagers frequently experience limitations in energy, transport and telecommunications. Opera infrequent and unpredictable due to fuel shortages. Fuel was often scarce because it had to be brought to the island in small open boats, which in turn needed to be arranged via a series of phone calls. Phone service depended on a weak signal from a mobile tower on a neighboring island, so that phone calls were often cut off in the middle of conversation, and calls had to be placed from specific places in the village where the signal was most reliable. The result was a situation in which the ability to travel, conduct business, and communicate over the phone was very unpre dictable, and often had to be postponed for hours or days. with mobile phones is heavily affected by such local conditions. Telec ommunications Media and communication tech nologies can be distinguished on technical grounds based on whether they are analog or digital, or use earthbound or satellite transmission. Two basic cosiderations for the sociological study of telecommunications are whether a technology transmits point t o point, or broadcast messages and whether the communication is one way or two way (back and forth) Until very recently, one way

PAGE 116

116 radio broadcasts played a key role in conveying information to rural Fijian villages. The messages conveyed through these med ia were often intended to be private, point to point communications, but the limitations of the available technology meant that often messages intended for a private audiene had to be broadcast publicly. One middle aged Cagi woman recalled the days before the radio telephone n those days we found out everything through the radio. If someone in Suva died or something like that, they would announce it ( kacivaka ) on the radio. They would even call out over the public radio if someone had sent some money for you to pick up at the post office ( nomu ilavo sa kau tiko yani oqo, waraka ena posi ) Today these private kinds of messages are transmitted via mobile phone, and the news remains restricted to those to whom it is directly relevant. Broadcast technologies especially rad io, still play a vital role in the communicative ecology however Radios re main source of news and musical entertainment Most village radios run on disposable batteries, meaning that radios can be used all day long, indepe ndently of a generator or other source of electricity. Men relaxing in the village often carry their radio s around with them throughout the day and some even when they work in their gardens As one Toba man Radios tell us the news about what i s going on in the world, and phones allow us to know what is going on with our close kin and friends. Therefore if you have Cheap, reliable, instantaneous mobile communication allows better coordinati on of transport, travel, and exchange of material goods. A Methodist minister and his wife, now living on the outskirts of the capital, re minisced about the years that they spent on

PAGE 117

117 one of the small remote islands in the Lau group where the husband had be en sent on ministerial duties They recalled that there were frequently shortages of basic items in the village store sugar, flour, or batteries for radios might be out of stock for weeks at a time. When batteries were out of stock, then nobody could list en to the radio, and literally no news of the outside world arrived in the village. The minister recalled with sadness that on another occasion a c lose relative of his died in his native village, and the minister did not learn of the death until months lat er, when he finally returned to Suva. Nowadays close relatives are notified immediately about a death, both for personal reasons and to allow preparations for the funeral to begin as quickly as possible. The speed, reliability and convience of sending mes sages via mobile phone represents a qualitative shift in rural Fijian communication. However, not all villages are equally endowed with mobile service. This is a function both of the location of mobile network towers, and local topography, which can block signals. Toba does not have a local mobile network tower, and catches weak signals from towers on nearby islands. As a result, Digicel is virtually unusable in Toba, while Vodafone calls are often dropped, and require villagers to seek out particular spot s in the village where the signal tends to be strongest. Cagi has a dedicated Digicel tower that provides a reliable signal, but similar to Yavu, receives only a weak Vodafone signal from a neighboring island. Veidogo, situated near the capital city, recei ves strong signals from both Digicel and Vodafone. Transport The role of geography and transport in shaping life in island communities cannot be overemphasized. Most small islands have limited resource bases, and depend on

PAGE 118

118 imported goods from a mainland to fulfill even basic needs. Transportation difficulties disrupt all sorts of processes, including resupply of staple foods and fuel, the ability to get medical attention, and the ability attend important social events, such as weddings and funerals. In par ticular, access to urban areas is vital because it allows villagers to seek wage employment or to sell their produce at market, to access urban shops, education and health services, and to generally be exposed to the cosmopolitan environment of the city. Y oung people from remote villages must attend boarding schools or live with urban relatives in order to complete their schooling. Likewise many elderly people in remote villages spend their last years living with urban kin, in order to have access to medica l care. In urban and peri urban areas such arrangements are unnecessary, because people can commute daily to schools and jobs. Toba, Cagi and Veidogo have each undergone socio economic changes in recent decades as a result of transport development. Transpo rt to Cagi was revolutionized through two developments in the 1970s and 1980s a road was built around the perimeter of Koro island, enabling motorized transport to neighboring villages, and a ferry dock was built on the southern tip of the island, allowin g large roll on/roll off ferries to dock safely. Ferries now stop twice a week on their way between Suva and Savusavu, picking up passengers and bulk produce to bring to the urban markets. These combined developments have allowed Cagi to become prosperous by cultivating cash crops: kava, copra, and taro. It has also made personal travel much more convenient and p redictable, and has given rise to local ly owned trucking

PAGE 119

119 companies that efficiently deliver goods between villagers and their urban kin for a small fee. Tob ans, by contrast, rely on privately owned, open fiberglass boat to reach urban areas In fact, Toba inhabitants claim that they face greater transport difficulties today than they did in recent decades. In the 1970s and 1980s the Toba cooperative possessed a large boat that made regular trips to Levuka and Suva to sell fish, and carry copra and passengers. This boat became defunct in the mid 1980s, but its absence was compensated by frequent visits from government ships, part of a national policy that with the recent change of government, and national budget shortages, visits from these government ships have become much less frequent. Copra often sits in the Toba vill age shed for months until a way can be found to bring it to market. During my 2 month stay on Toba I witnessed the diverse inconveniences caused by the local transport situation. My journeys to and from the island were delayed, sometimes by a week, by ba missed his boat back to the island, which resulted in over a week of cancelled classes. Finally, the father of the schoolteacher died, requiring a sudden trip to Suva. As nobody else wished to tra vel that day, the schoolteacher was forced to personally hire a boat at substantial cost. In contrast to the outer island villages, Veidogo is very close to Suva one can sit al acros s the harbor. U ntil about 1990, there was no road connecting this coastal village to the main highway, which is lo cated just a few miles inland. In those days, r eaching the highway required an

PAGE 120

120 hours long trek through the forest. As a consequence, most peop le preferred to travel to Suva by boat, which was costly and contingent up on weather. These factors made daily commuting to urban jobs very difficult, so most villagers adhered to a semi subsistence fishing and farming lifestyle. After the dirt road was b uilt, urban bu ses began incorporating the village in their routes Suddenly commuting to the city for work or school became a very realistic option, and today, nearly every village household has at least one regular wage earner. Veidogoans can shop in urba n shops and markets daily if need be, and enjoy lower prices and wider selection than outer island villagers. Veidogoans can also choose from and medical facilities located in Suva Rather than needing to travel elsewhere for sch ool or healthcare, some Veidogo households actually host kin from other parts of Fiji, who are in need of these urban based s ervices. Rural Fijians tend to travel long distances for a limited set of social and practical purposes. The main reasons for trav el are special visits to relatives (typically for weddings, funerals or around the Christmas holiday), student sports events, medical treatment, specific shopping needs or bureaucratic tasks. In other words, people either travel to visit with distant kin, or they travel out of practical necessity. In this way we can understand why Toba ns, who have the least convenient transport options, still tend to travel long distance more frequently than Veidogoa ns and Cagi ans. It is because Toba ns must travel away from their small island in order to obtain basic necessities of life: education, medica l care, and store bought goods these necessities cannot be acquired locally.

PAGE 121

121 Energy Modern communication technologies rely on a source of electrical power for their operat i on. In that sense, the lack of an ample, dependable source of electricity is a communicative ecology. Many peri urban villages, such as Veidogo are linked to the city energy grid, and receive dependable 24 hour electricit y. The only disadvantage of this arrangement is the need to pay for metered electricity. Other villages, such as Cagi have a single village generator that operates on a set schedule perhaps 7 10pm daily. This setup is economically advantageous in rural a reas because each household does not need to acquire its own separate generator. The major disadvantage s of a village generator are the inflexibility of the fixed hours of operation, and the fact that a malfunctioning generator may plunge the whole village into darkness for weeks, (or in the case of Cagi 3 months) until repairs can be made. A further disadvantage of this arrangement is that certain individuals abuse the system by obtaining appliances with high energy requirements, such as refrigerators. Th is can community members when there is insufficient energy for all households A third situation that of Toba is where there is no central g enerator in the village T hose households that can aff ord to, may obtain a small private generator. At the level of the household th is requires extra expenditures to obtain a generator, and to provide fuel which discourages daily use of electricity The disadvantage is that electricity is not used on a daily basis. At the community level, t his arrangement provides a basic form of resil ience, however Any given generator may malfunction on a given day, but people may borrow to operate lights for a special event.

PAGE 122

122 Alternative energy sources are now being harnessed in some Fijian communities. Two Toba inhabitants had small solar phone chargers, which allowed them to keep their phones running despite days without electricity. Other villages have small hydro or wind power generators. It seems that in most villages solar, hydro or wind power may might provide a stopgap source of energy, to at least operate basic lighti ng and recharge phone batteries should conventional generators fail. Predictions for Mobile Use in the Study Communities I n this chapter I used the concept of a communicative ecology as a way of understanding how material conditions, social structure and c ultural norms in the study communities shape communication. Mobile use may not conform to the same kinds of social constraints that govern face to face communication. However, we can use these constraints as a starting point to formulate predictions about the nature of Fijian mobile communication. Based on the varying level of material infrastructure and economic opportunity in the three villages, Veidogo residents should have a higher rate of mobile ownership and more intensive phone use than Cagi and Cag i more than Toba This is not only because Veidogo residents have more money to spend on phones and calling credit, but also because Veidogo r eceives the strongest network signal (from both Digicel and Vodafone) has 24/7 electricity to ensure phone batter ies are charged, and convenient access to transportation. By the same logic, Toba should have the lowest levels of mobile ownership and use, and Cagi intermediate levels. Following from the discussion of the social aspects of the communicative ecology, men in each study community should communicate disproportionately with

PAGE 123

123 other men over the phone, and women with other wom en. Likewise, phone users would be expected to communicate disproportionately with individuals in the same age/marital category. In terms of kinship I expect phone users to communicate disproportionately with same sex cross cousins, and to a lesser degree with opposite sex siblings and cross relatives in the first ascending and descending generations. It is difficult to predict the extent o f communication with parallel relatives, because although they should be the focus of much instrumental communication, the hierarchical limitations on free discourse may limit the amount of calling between such relatives for purely social reasons. The act ual reality of rural Fijian mobile communication may depart from these predictions or be considerably more complex, influenced by local patterns of mobile ownership or cultural factors as yet unexplored. For this we must move on to discuss the actual data I collected.

PAGE 124

124 CHAPTER 4 PHONE OWNERSHIP AND USE IN RURAL FIJI Overview In this chapter I describe patterns of phone owne rship and use in Toba Cagi and Veidogo villages. This is a broad topic, ranging from the distrib ution of phones in communities, and n otions of ownership and borrowing, to the way phones are used to communicate through voice, text and internet as well as the uses of secondary features of phones, such as games and calculators. I begin by describing the way phones and other telecommunicat ions technologies are distributed among the households and demographic subgroups in the study communities. I then describe how residents of the study commu nities acquire their phones, the range of phone lifespans, and the variety of ways in which mobile ph ones ultimately meet their end in rural Fiji I explain the socio economic consequences of phonelessness, and how unequal access to phones is partly remedied through extensive sharing and gifting of phones in Fiji I also describe some of the aesthetic and symbolic aspects of phone ownership and use, which relate to image and social position in the community The second part of the chapter is a descriptive account of the various ways that rural Fijians use mobile phones. The primary functi on of mobiles is communication at a distance either through voice, text, or internet Each of these com munication modalities is used for a variety of goals and purposes and in a variety of styles. M obile phones incorporate an ever increasi ng array of sec ondary functions such as clocks, games, and music regarding the use s of which I also present data

PAGE 125

125 Distribution of Phones in Rural Fijian Communities Telecommunications technology in Fiji has evolved rapidly in recent decades and is still in flux. As la ndlines and radio phones rapidly become obsolete, m obile phones are becoming increasingly ubiquitou s. However, mobiles remain unevenly distributed among the individuals and households in Fi jian communities. As a scarce and coveted uneven distribution reflects how socioeconomic inequality operates in the rural Fijian context and affects the way in which the socio economic benefits of connectedness are distributed among individuals and households. Telecom Technology i n Rural Fiji Res idents of Toba, Cagi and Veidogo own three type s of tele com technologies: home phones (Easytel) mobile phones, and SIM cards. M ost rural Fijian communities do not have acc ess to fixed line telephone networks For decades rural Fijians have relied on publ ic radio telephones, often necessitating that villagers travel to neighboring communities or government station s in order to make or receive a call. In the year 2000, Telecom Fiji introduced its Easytel service, a wireless CDMA network that facilitated com munication in rural areas that do not have access to fixed cables. Easytel provides better service than radio telephones, because it allows simultaneous communication on a single frequency band, so that multiple calls do not interfere with each other ( an i nherent limitation of radio telephones). Easytel also enabled rural Fijian families, for the first time, to own private telephone sets in their homes. Easytel phones resemble landline home telephones, with a handset connected by a cord to a base that has a numeric keypad. However, Easytel phones differ from actual landlines in that they are not connected via a phone jack to a telephone cable. Instead, Easytel runs off a wireless network similar to those used for mobile phones. An

PAGE 126

126 Easytel phone must be plug ged into a power supply, but has a battery so that the phone can run for several hours in the absence of electricity. Easytel phones were popular in Fijian villages throughout the early 2000s when mobile networks remained geographically limited and mobile service was prohibitively mobiles have become the preferred choice in most areas of Fiji and Easytel phones are rapidly becomin g obsolete The main rationale for ownin g an Easytel phone nowadays is if there is an immediate family member or kinsman (such as a parent, child or sibling) in another part of Fiji who also happens to have an Easytel so that cheap in network calls can be made to that person Another benefit of owning an Easytel is that most urban businesses and government ministries are still on the TFL network. Thus, Easytel phones are economical only as long as one makes frequent calls to businesses, government offices, or other TFL users. Many of the aging Easytel phones found in villages are now poorly functioning, or have already become non operational. A common problem is that after several years of hour electricity, the Easy until the electricity is switched on at night to turn the phone on again. Several households I visited had Easytel phones where the battery had been dead for days or weeks, and nobody had even bothered to try to recharge it. Other households had functional Easytel telephones, but they had at some point in the past accumulated a very large phone bill (I am referring h ere to post paid subscribers) Rath er than pay the

PAGE 127

127 phone bill, the househo ld simply allowed their phone service to be switched off by TFL and household members transitioned entirely to using mobiles. Beginning in 2008 with the liberalization of the Fijian telecom market, mobile service became available to most rural communities and the cost of mobile phone use fell dramatically. Mobiles are almost universally preferred to Easytel phones because they are portable and they are more personal and personalizable. Mobiles may include features that Easytel phones do not, such as game s, music, a light, and internet access. forethought a mobile owner can communicate extensively without spending much money on credit. In response to these market trends, TFL has released the Handiphone, a mobile handset w network. However, TFL still has a ers: Digicel, Vodafone and Inkk. Th obile calling network is one factor discouraging new customers from choosing TFL. Finally, there are some individuals who own only a SIM card, but no mobile phone handset. These tend to be younger people without the means to buy a phone, but it also happen keep s the SIM card until able to acquire another handset. The limited benefit s of owning a SIM are that one can use it to store c ontacts, and also may plug the SIM into another enabling a degree of independence when borrowing and using other phones.

PAGE 128

128 The following charts show overall ownership rates of Easytel phones, mobile phones, and SIM cards in the three stu dy communities. An evident pattern is that mobiles are far more common than Easytel phones in each of the study villages. Ownership of a SIM ( without a corresponding phone) is comparatively rare in all of the villages. Table 4 1. Percentage of village populations owning mobiles, Easytels, and SIMs Item Toba Cagi Veidogo Total Count F unctioning mobile phones 15 (16%) 101 (31%) 53 (43 %) 169 6 (6 %) 15 (5%) 3 (2%) 24 SIM card (but not a functioning mobile) 2 (2%) 10 (3%) 1 (1%) 13 Each field contains the actual count, followed by the per that item (in parentheses) Figure 4 1. Phone own ership rates total population The rate of mobile ownership increases from Toba to Cagi to Veidogo Meanwhile, the rate o f Easytel home phone ownership follows an opposite trend, decreasing from Toba to Cagi to Veidogo Of the three study villages, Toba is the most remote and rural, while Veidogo is the closest to an urban area. Cagi occupies an

PAGE 129

129 intermediate position. A poss ible explanation for this pattern of phone ownership is that Toba has relatively unreliable mobile reception, and Tob ans have less disposable income to spend on mobile phones. They therefore continue to rely on older Easytel phones that they have owned for years. In contrast, Veidogo and Cagi residents have more disposable income and better mobile reception than Toba does. Veidogo (located right outside Suva) was the earliest of the study villages to acquire a mobile signal, and continues to have the most r eliable service of any of the villages. The near complete transition to mobile phones in Veidogo seems to portend remote rural villages will likewise switch over completely to mobiles. Ownership of a SIM (without a mobile phone) is comparatively rare and plays an insignificant role in terms of communication in the study communities. After all, the SIM alone does not permit one to call or text, but merely serves to store contacts and phone credit. The disproportionately large numb er of Cagi residents with SIMs is due to a recent visit to the village by a Digicel promotional team, which sold a bundle of 100 SIMs to the village rs Distribution of Phones a mong Households The great majority of mobile phones in t he study villages are o wned exclusively by single individual s Out of 146 mobile owners, 83% reported exclusive ownership, 12% reported jointly owning the phone with their spouse, and 5% reported that the phone was owned collectively by the household. A single rural Fijian hous ehold may have multiple mobile phone owners. This permits members of the household to contact each other throughout the day, or when a household member takes a trip to another part of Fiji. Another benefit of having multiple mobiles in a household is that each mobile owner can be responsible for his/her own

PAGE 130

130 calling costs. In contrast, phones shared among household members can cause arguments, because certain individuals may overuse the phone and incur large bills, or hone, as Horst and Miller observed in Jamaica (Horst and Miller 2007). Households within the study villages have unequal endowments of telephones. The total number of functioning phones per household (including both mobiles and Easytel) ranged from 0 to 6, with an overall mean and median of 2. The mean number of phones per household varied between 1.1 for Toba to 2.4 for Veidogo with Cagi in between at 2.0 phones per household. In Toba 6 households (32%) had no phone of any kind; in Cagi only 2 households (4%), and in Veidogo every household had at least one mobile or Easytel phone. Table 4 2. Mean phones per household in the study villages Village Mean phones/hh Toba 1.1 Cagi 2.0 Veidogo 2.4 Figure 4 2. Number of phones per household, by village

PAGE 131

131 In the three villages combined, o nly two households with Easytels lacked any mobile phones. The rest of the households with Easytel phone s also possessed one or more mobiles. In fact, households with Easytels have more mobile phones on average than househ olds without Easytels (1.9 mobiles/hh compared to 1.8). H ouseholds with Easytel phones are more likely to have greater financial resources, which is also positively associated with mobile ownership. The mobile phone therefore does not seem to be treated as a replacement for the Easytel phone; rather, mobiles serve as a compliment to Easytel home phones in households that can afford both Distribution o f Mobile Phones a mong Age Groups Phones are not evenly distributed among the different age groups in Fiji an communities. Young people below age 16 and older people above age 60 have much lower mobile ownership rates than people aged 16 60. Figure 4 3. Age distribution of mobile phone owners vs. overall age distribution

PAGE 132

132 Figure 4 4. Percentage of eac h age cohort owning mobile phones In rural areas at least, there is widespread agreement that children should not own mobile phones until they finish secondary school. Students are considered too immature, and mobile phones are thought to be a distraction from their studies. Furthermore, students are generally discouraged from having the kind of private relationships or independent social lives that are facilitated by mobile phone s In practice, however, some rural students do come into possession of mobi le phones before they finish secondary school. Such s tudents often acquire their phones as gifts from indulgent relatives such as uncles or grandparents, rather than from their Emele, a Cagi woman, reported that her 15 year old daughter was given a phone as a gift by her m omo (classificatory MoBr), a gift that Emele felt Now my daughter just uses h Emele sighed in mock exaspe ration. There are also exceptions : one 13 year old girl in Cagi was given a mobile phone, because she lived with foster p arents and used the phone every day to

PAGE 133

133 call her biological father on Viti Levu. Most villagers believe that parents should be allowed to check through the phones of their school age children to make sure they are using the phone properly. In practice, however, this sort of monitoring may be difficult to carry out as young people are quite savvy about using passwords and other privacy featu res. Once a youth leaves or completes secondary school, he or she has practically the same rights as an adult, including freedom to own and use a mobile phone. Mobile ownership rates remain high throughout the adult years, until around age 60, a t which poi nt ownership declines There were 8 individuals in my sample in the 76 80 age range, none of whom owned mobiles. Several of the 60+ year old phone owners were given their mobile phones by adult sons and daughters living in other areas of Fiji, as a means t o stay in contact. Often, older people have a very limited understanding of their mobile phones, and use them merely to answer calls from their grown children. Very old people in Fijian villages typically have limited mobility and are largely confined to t he household. This leads to a gradual wit hdrawal from social interaction in the community Paradoxically, while such elderly people generally have reduced use for mobiles, the ability to even occasionally receive a call can take on great importance Distr ibution of Phones by Sex Men and women have unequal rates of phone ownership in Toba Cagi and Veidogo Overall, 56 % of phones in the three study v illages are owned by men, and 44 % by women. The degree of inequality varies between villages Veidogo the vi llage with the highest ove rall rate of phone ownership (43 %), also has the most equitable distribution among the sexes ; in fact in Veidogo slightly more phones are owned by

PAGE 134

134 women than by men Toba the village with the lowest overall rate of mobile ownersh ip (18 %) has the most un balanced distribution among the sexes. Figure 4 5. Sex of mobile phone owners in the study communities One way of explaining this pattern is that in villages where phones are scarce, such as Toba they tend to be monopolized by individuals in positions of power. In other words, if a household only has one mo bile phone, it is likely to be controlled by the male head of household, who has the final say over the allocation of household resources. When the household manages to acqui re a second mobile phone, then the wife of the head of household may be able to assert ownership over it. This interpretation fits with data about the ownership of phones among various family members in a household. It would seem that adults have preferent ial access to phones over their children, and in each generation, males have preferential access over females (Figure 4 6 ).

PAGE 135

135 Figure 4 6. Relationships of mobile phone owners to the head s of the ir household s An alternate way of explaining the differing patterns of ine quality in the three villages, is that wo men in Veidogo tend to be better educated than their outer island counterparts, and are more likely to have a job in the city. A woman can usually justify requiring a mobile phone if she commutes to town to work every day, because it allows her to stay in contact with her husband and children. A woman with an independent source of income may also have a greater say in household financial decisions and purchases. Most married couples prefer to each sep arately own a mobile phone, so that they can stay in contact with their spouse throughout the day. The main concern with owning multiple phones in a household is economic: the more phones, the higher the total household expenditure on calling and texting t ends to be. Although there is widespread concern over mobile phones and marital infidelity in F iji, this does not seem to be considered a Furthermore, respondents were roughly evenly divided on t he question of whether

PAGE 136

136 husbands and wives have the right to monitor the phone contents of their spouse to ensure fidelity mobiles should be respected as the private property of the ind ividual, married or not. Phone Ownership Acquiring a P hone A striking aspect of phone ownership in rural Fiji is the high proportion of villagers who receive their phones as gifts, rather than purchasing them. This echoes ethnographic observations in the Caribbean and Africa, where people hand down their old phones to impoverished relatives when a newer model is acquired (Horst and Miller 2007; James and Versteeg 2007; Aker and Mbiti 2010) Out of 146 Fijian mobile owners surveyed, 58% received their phone s as gifts. Two thirds of these gifted phones were already used, and only one third were new when they were given. Only 39% of villagers purchased their mobile phones, and 3% obtained their phones by other means (found or traded). Figure 4 7. How resp ondents obtained their mobile phones

PAGE 137

137 Gifts Gift exchange is a quintessential aspect of the kinship economy, and mobile phones are no exception. Fijians seeking a mobile phone are very proactive about asking potential donors if they might give them a used phone, or even buy them a new one Villagers do not rely merely on fellow villagers for such gifts but rather request phones from a wide variety of close and distant kin, friends and business associates living in Fiji and overseas I personally was asked by several villagers if I might give them my mobile pho ne when I left Fiji, or whether I might purchase and mail them a new handset after I returned to the United States. Figure 4 8. Sources of gifted phones Mobile phones are most commonly received as gifts from kin. These kin are not necessarily members o f the immediate family: 32 respondents received phones as gifts from cousins, brothers in law and sisters in l aw, aunts and uncles, and more distant relatives. 11 respondents received their phones as gifts from friends. Among these, 4 were given by girlfriends/romantic partners, and 3 were from schoolteachers with whom they had established a friendly relationship. 5 respondents received their phones for

PAGE 138

138 free from companies for which they work. For inst ance, Maikeli works as a security guard in Suva, and was given a mobile phone by the security company so that he can arrange his nightly shifts, and so that he can call the police in the case of an emergency while he is on duty. The geogr aphic location of phone gift givers also varies widely. 20 of the mobile phones were given as gifts by fellow villagers in Toba Cagi or Veidogo 32 of the gift givers live in the Suva area, 27 live in various other parts of Fiji, and 5 reside overseas. Respondents cited a wide variety of reasons for the occasion of the gift of the phone. Among these, 13 phones were given as presents on birthdays or holidays. 53 respondents said there was no particular occasion associated with the gift. Usually, this means that the recipien t simply asked a friend or relative to give them a spare used phone. An opportune occasion for such a request is when a friend or kinsman acquires a new phone; the us ed handset is often given to whichever individual who asks first, or asks most persistentl y. Another common tactic is to ask a relative who is traveling overseas to purchase a phone; it is widely assumed among villagers that phones overseas are cheaper and of better quality than those available for sale in Fiji. Other occasions for receiving a phone are serendipitous : for example, one respondent got his friend drunk and thereby successfully persuaded him to hand over his new phone (which he had bought earlier that same day)! Another mobile was given to a woman skilled in medical massage as a gi ft from one of her grateful patients. Two phon es were given to youths by girlfriends they met during trips to distant parts of Fiji, in order to help them remain in touch and keep their relationships alive

PAGE 139

139 Purchases Prices for a pre paid mobile phone han dset in Fiji range from about $ FJ 50 for a very basic model, to hundreds of Fiji dollars for high end smartphones that include internet capability, cameras, and touchscreens. This is a major expenditure for a typical Fijian villager without a steady income The fact that many rural Fijians do not habitually save money makes such purchases even more difficult hence, the common reliance on receiving phones as gifts In Waidroka a village near Suva, local villagers each receive a land lease payout of severa l hundred Fiji dollars per year this money derives from village land rented to businesses along the main road to Suva. A local villager explained that directly after these payouts, villagers are often seen walking around with fancy new mobile phones. In o ther words, purchase of a new mobile is typically only affordable to rural villagers in the ca se of such a sudden windfall. It also indicates that nowadays when Fijians suddenly acquire a large amount of money, a new mobile phone is among the first items t hat they wish to buy. Other means of obtaining phones One respondent claimed to have found her mobile phone lying in the grass in the village. Four respondents said that they acquired their present phone by trading phones with someone else. One village you th traded his phone to a cousin who was receiving persistent calls from an ex calls, the cousin begged the village youth to trade phones with him. Given frequent accounts of phone loss and thef t, no doubt some villagers acquire their phones by finding them, or stealing them Mobiles have become a prime target for theft because

PAGE 140

140 they are valuable and easy to swipe The most common safeguard Fijians take against phone theft is simply to carry their phone s with them at all times. Borrowing and S haring The concept of ownership in Fiji differs somewhat from that which Westerners typically may exert a right to borrow or us e the item. The extent of this right of usufruct depends on the item in question, and the nature of the relationship and relative social standing of the individuals. The strongest indicators of phone ownership in Fiji are the exclusive righ t to carry the phone around, and the exclusive right to address book. Beyond these prerogatives that are generally reserved to the owner, b orrowing is far more common in Fiji than in most Western countries. Indeed, extensive sharing of mobile phones may be a general characteristic of developing contexts, or any context in which phone ownership is not universal. Sharing within the household The most extensive phone sharin g in Fiji occurs within the household. Married adults usually share their mobile phone s freely with their spouses. Youths living within more possessive or sensitive about sharing their phones, but nevertheless most yout hs said that they allow other household members make calls when they need to. Phones that are shared extensively within a household may have a hodgepodge of contacts entered b y the various household members. This is an indication that mobile phones, at lea st for now, are for some purposes treated as household goods in rural Fiji.

PAGE 141

141 Children may be es and use them to play games. Through such fiddling, p re adolescent children often know more about the functioning of th sometimes ask their children directions for how to change a phone setting, or how to One elderly Fijian with whom I spoke believe s that Fijian childr en today have a natur al proficiency with technology which, he speculated, may be absorbed subliminally while the children watch TV. Borrowing outside the household To a more variable extent, phones are also shared and borrowed by friends and relatives liv ing outside the household. Of 146 mobile phone owners, 75% responded that individuals from outside the household have borrowed their phone during the last month, while 25% said borrowing outside the househ old had not occurred during the past month Of thos e who lent their phone during the past month, they lent to a mean of 2.5 different individuals during that time. These borrowers included a variety of kin in the village, as well as friends or co workers outside the village. Sera, a Cagi tend to borrow phones from the same people from whom you borrow other things like sugar, salt or kerosene. And in return, the same people you borrow from tend to borrow back from you. Everyone in the village knows which houses they can turn to for kereker e (borrowing). together ( sota na yalona ). This statement reflec ts the importance of confidence between partners in informal exchange networks (Espinoza 1999, Schmink 1984).

PAGE 142

142 Villagers usually borrow phones from among their closest neighbors, simply out of convenience. However, close relatives who do not live nearby may also be good despite the fact that he lives on the other side of the village, because he is her classificatory brother and they habitually lend each other various kinds of as sistance Borrowing a phone for the purpose of calling is usually done in the following manner: the borrower seeks out somebody that currently has a FreeCall promotion, so that the call does not cost anything to either the owner or the borrower. It is ther efore ve Iko Freecall tiko? e tawa tiko na nomu talevoni? ( ) Failing this, the borrower may ask someone without FreeCall to use their phone, either making the call using the prior to making the call. Another alternative, if the borrower owns a SIM card, is to insert the SIM card in the he credit is then deducted from the es not cost the lender anything Fijian villagers also often receive calls on the phones of other villagers Villagers who have no phone tell their urban kin and friends the phone numbers of their closest neighbors i n the village. That way if someone wishes to reach the phoneless villager, they can call the neighbor and ask the neighbor to hand the phone to their friend so they can talk. This arrangement does not cost the neighbor any calling credit, but repeated day after day, it can become tiresome. Living in Cagi I mysteriously began receiving calls from people completely unknown to me, asking to speak to my various village neighbors. I always cooperated, but one occasion actually became embroiled in an

PAGE 143

143 argument ov er the phone with an unknown Fijian woman who refused to identify herself to me, and yet demanded that I give my mobile to a neighbor so she could talk to him. These sorts of situations are common in the village, and are one of the reasons that so many Fi jians desire new SIM cards so t hat they can acquire a fresh phone number that has not been widely disseminated among kin, friends and strangers and with any luck, keep the number secret from unknown third parties Lending a phone is the sort of small fav or that Fijian villagers grant each other on a daily basis without complaint. However, a few villagers express a general displeasure at lending out their phone, even if it involves no financial cost on their part. Lending a small annoyance b ecause the borrower is draining the battery, and also might use the occasion to play a game or two, or snoop through the address book or the saved the borrower leave their sight they sit there as the borrower makes their call, then take the phone back immediately. Tawake, a Batiki youth, said that his previous ph one had a memory card which could store and play music Other villagers frequently borrowed the phone to listen to songs a nd wo uld bring the phone back only once the battery was completely drained When Tawake was later shopping for a new phone he purposefully sought out a basic top borrowing his phone so much Life C ycle of a P hone In a sense, rural villages are the graveyards where old phones in Fiji go to die. The very nature of the village environment is hard on phones they are frequently doused or dropped into water, and roughly handled by stray toddlers and children. Many hand me down phones are already on their last legs when they arrive in the village,

PAGE 144

144 batter ies barely holding a charge, or have some idiosyncratic trai t, such as a cracked screen or missing number key s on the keypad. Such used phones might continue to operate for only a few months before some terminal malfunction or trauma befalls them. Only about 50% of village phones last beyond one year. Among 146 mobile phone owners surveyed, the mean number of months that the owner has possessed his/her current p hone is 15.6 months. The longest anyone claimed to have owned their present phone is 6 years (72 months). The following chart gives a rough idea of the longevity of the typical mobile phone in rural Fiji. Figure 4 9. Number of months owning current mob ile phone (n=146) I asked 109 respondents what happened to their last mobile phone, which yiel category. This category incorporates a broad array of problems and folk diagnoses, often resulted from some form of prior damage or water exposure, so these categories may

PAGE 145

145 as the one ph one that was sat upon or another that was slammed in a door, or another hurled against a wall by an enraged husband (the wife had been caught gossiping with relatives in the city). Figure 4 10. What happened to your last mobile phone? (n=109) Figur e 4 11 How many mobile phones have you owned before your current phone? Respondents ranged from those still using their first mobile phone, to those who had previously owned 5 other phones. Respondents had owned a mean of 1.7 mobiles before attaining the ir current mobile phone. The largest group (31%) had owned two mobiles previous to the current phone. Phones are commonly lost during trips to the city. Often they are left in a taxi or seen again Several

PAGE 146

146 phones disappeared inexplicably during a night of heavy drinking in Suva. Perhaps because they are accustomed to the forgiving village environment, rural Fijians are somewhat careless about how they handle their phones when they visit the c ity The most surprising statistic is the large number of phones that are destroyed by water For many mobiles, the final resting place is the bottom of the Pacific Ocean, when the phone falls out of a pocket during a boat ride or fishing trip It is tempt ing to imagine, a Tempest these phones lying full fathom five, undergoing a sea change into something rich and strange, with sea nymphs to hour ly ring their knell phone s are simply soaked by rainwater, tumb le into a wash basin, or are even dunked into cup s of tea. Men who take their phones with them daily to the gardens usually wrap them up in plastic bag s to avoid accidental water exposur e. When a phone breaks or stops working, the owner typically removes t he SIM, in order to save the list of contacts in the address book, and in the hope of finding another phone in which to use the still functional SIM. Often though, a new phone is not procured anytime soon; the SIM expires after 3 months, and the contacts a re lost anyway. Most villagers do not prepare themselves for this eventuality by writing out their contacts. However, a couple of informants had indeed meticulously copied their contacts into notebooks for safekeeping. Etika a young married man, was job l ess for several months in Suva. During that time top it up, but before it expired he copied out all of his contacts, all 50 or 60 of them (several of which were girlfriends) into a notebook. Before he was able to obtain money

PAGE 147

147 to buy a new SIM, his jealous wife discovered his notebook and burnt it in the kitchen fireplace Etika such a big deal, because you can get their numbers again from other relatives. But many of my friends in Suva, friends who worked in the market, again. They are total ly lost ( yali vakadua Time between P hones In most developed countries, losing or breaking a phone is a temporary inconvenience. T ypically we quickly acquire a new phone and within a few days we are once again in full communication. This is not the case in rural Fiji, where the loss of a phone may lead to months or even years of phonelessness. Buying a new phone must wait for an infrequent trip to the city, and at any rate acquiring enough money to buy a new phone is ver y difficult. Most villagers simply wait until someone gives them another phone as a gift. Finding such a donor can be a long and uncertain process. The case of Ruci, a young single mother in Cagi is illustrative of the circuitous process of obtaining a phone nd mobile phone vau (hibiscus) branches to put them to soak in the ocean in order to make skirts for the sc hoo l of the year meke (traditional dance) As I was putting the h ibiscus in the ocean to soak, I forgot I had my phone with me and it fell out of my pocket. Even after I dried it out, the phone d SIM was still goo a year, which really hurt ( sa mosimosi saraga ). I still had my old SIM at that point, but I took a trip to Suva and my Auntie there asked me for the SIM so I gave it to her. Later, my Auntie gave the SIM to Kusitino (a

PAGE 148

148 cousin in the village), wh o is still using that SIM today. While I was phoneless I my father let m e keep the phone for myself, because he had gotten a new phone during a trip to Suva. I asked 104 phone owners (who had previously owned another phone) how much time they spent phoneless, between their previous phone and their current phone. Answers ranged from none (they had already acquired a new phone before losing or giving away the old phone, or immed iately acquired another phone) to 5 years. The mean period of time between phones was 6.5 months. Table 4 3. Mean number of months between loss of previous phone and acquisition of current phone Men Women Toba Cagi Veidogo 6.1 7.1 14.6 4.1 7.9 Compari ng the length of time between phones between genders and villages yields unexpected results. One might predict that men, who have greater control over household finances and priority in access to household goods, would obtain new phones much more rapidly t han women after the loss of a previous phone However, the mean time between phones for men and women is very similar a difference of only one month In terms of the different villages, Toba the most remote and poorest village, does in fact have the long est waiting period between phones 14.6 months However, Veidogo the wealthiest village and closest to Suva, has nearly twice the waiting period as Cagi which economically speaking is the am unsure how to explain this, at least on the basis of economic factors.

PAGE 149

149 Phonelessness When a technology such as the mobile phone is acquired by the great majority of people in a society, then those few who remain phoneless seem even poorer by comparison. Horst and Miller observed in Jamaica that the lack of a mobile phone 2007:59). However, t his is not really the case in the Fijian villages in which I conducted research. In rural Fiji, phone ownership is not a reliable indicator of soci al status. For example, the turaga ni vanua (the chief and highest ranking member of the village) in both Cagi and Toba did not personally own mobile phones (although other members of their households did) I did not encounter any evidence that being phone less detracted at all from the respect paid to these men. The general disconnect between phone ownershi p and social status makes sense in the Fijian context. First, Fijian culture places less emphasis on wealth as a measure of social worth, than do most m arket oriented societies. In addition, m any rural Fijia ns obtain their phones as gifts Because they are not usually purchased by the actual owner, phones are not strongly associated with personal wealth or achievement in the minds of Fijian s one is simpl y fortunate enough to have a relative capable of giving one a gift Although there is no stigma of poverty associated with lack of a phone, those who are poor and/or socially stigmatized are certa inly less likely to possess a phone. In other words, p oorer households are more likely to be phoneless, but phoneless households are not necessarily seen as poor. P honelessness can, however, economic opportunities Rusiate a Toba youth, worked in Suva for several y ears as a painter. While in Suva he had a mobile phone and used it to contact lots of girlfriends.

PAGE 150

150 One day he was on a ladder painting a house and his phone fell from his pocket and shattered on the floor. He removed the SIM and put it in a drawer for safe keeping, but it expired and he lost the numbers in his address b ook. Shortly after that, Rusiate retur ned to live in the village, where he has remained for the past 3 years without a establish contact with any of his former girlfriends, and meeting girls in the village is much more difficult than in the city he mournfully noted Another informant, Josevata, e xplained how phonelessness can reduce a remittance income. He had had a phone until a few weeks be fore, when his to make calls, b ut noted that he makes and receives calls far less frequently than ings and goings of kinsmen in the city This is important because one of the best opportunities to ask for some money, or some other favor, is when a kinsman comes to visit the village or is passing through the city so that it is possible for them to visi t a bank, wire some money, or drop some sugar or flour off at the ferry to be delivered to you. Since transport is a key limiting factor in all sorts of enterprises in Fiji mobile phones are ntage of social and economic opportunities. As elderly Fijians become too infirm to engage in work or to leave the house, they are often left alone for much of the day. Eseta, an elderly woman in Cagi lives in a house on top of a steep hill, and she canno t descend to the village without great physical difficulty. Her grown son recently departed to work in the city for an extended

PAGE 151

151 period, leaving her alone with her deaf husband and an 11 year old adopted ghly critical of many of the villagers, and does not feel welcome when she goes down into the village. She claims that she is satisfied and resigned to her lonely way of life, and is not interested in travelling or hearing news. However, she said she cried a lot when her son left the village, and wishes she could get in touch with him more easily. Without a phone, or a way to regularly borrow one, this is difficult. A few respondents seemed unconcerned about the loss of their phone s One woman who now has a phone, but was phoneless for the entire pre vious year, said that concern her ( au vawelewele ga I would just borrow other only if there is an eme rgency ( leqa ) or if I want to call Most Fijian villagers value phones and consider them very useful. However, there was no clear consensus on whether or not phonelessness is a source of hardship for villagers. Some think th at not having a phone is difficult, because you miss out on social opportunities and borrowing the phones of others is inconvenient. Others feel that it is not a big problem, and that the basic need for communication can be met simply by borrowing phones f rom others. Ironically, it seemed that those who currently owned phones were more emphatic about the hardships of phonelessness, than those who were in fact phoneless themselves. Aesthetic and Symbolic Aspects o f Phone Ownership Mobile phones are physical artifacts that, much like clothing, shoes, or jewelry, can

PAGE 152

152 Mobile phones as artifacts, image and social identity on at least five le vels: 1. Ownership vs. non ownership 2. Handset model 3. Personalization of the handset 4. Carrying and displaying the phone 5. Using the phone I will describe each of these expressive aspects of phone ownership and use in turn, and how they operate to shape individuals self image and the way the individual is perceived by others in the context of a rural Fijian village. Ownership vs. N on ownership The very act of owning of a phone indicates at least a basic level of social competence Phone ownership implies that one ha s people to contact, and important things to talk about. As a valued commodity, phones also represent wealth or material attainment. The degree of this association is inversely related to the rate of mobile ownership in a society. In the 1980s, mobiles wer e luxury items in the United States, and ownership of a mobile was associated with prestigious, hi gh paid jobs. Today, mobiles have become commonplace even in rural Fijian villages. As Horst a nd Miller within a very short time, th e ce ( 2007: 64). Nevertheless, ownership of a mobile phone still indicate s a basic threshold of social involvement and economic attainment. M obile phones and other material possessions are less important a measure of social standing in rural Fiji, than in m any market oriented societies. In Fiji, l ack of a phone could possibly indicate immaturity, economic hardship, a paucity of social contacts, or technological aversion or incompetence. However, v illagers know each ies very well, having lived together in clos e quarters for years. The

PAGE 153

153 underlying reasons for any successes or shortcomings are well known to all in the village In this sense the lack of a mobile does not produce a Fijian social identity, but could serve as just one indication, or means of reinforcing an existing peripheral role within the community. For example, Akuila, a youth in Cagi is a clever and socially active person. He s technical institutes. Akuila has a high end smartphone which he uses to listen to music and to surf the internet on a daily basis This phone use pattern fits with existing image within the village as an engaged and capable individual. In co ntrast, Neumi is an unmarried, middle aged man who lives with his aging mo ther. He is known as a good for contribute his annual soli and at any rate who would call him? For Neumi, phonelessness just serves as one more indication of a general ineptness and social incompetence. However, in rural Fiji it is also quite possible for a well respected and socially competent individu al to be phoneless In this sense, it is better to think of phone ownership or phonelessness in Fiji and similar contexts as reflecting, rather than molding, personality and reputation. In fact, a few well respected villagers claimed to have no desi re to own a phone: these tended to be middle aged, married men of a hardworking demeanor. These men considered mobile phones to be fri volous and a waste of money, and pointedly stated that the purpose of a phone should be to transmit important message s no t to make idle chatter. They said that in the rare case that they needed to make a phone call they could just borrow a phone from a neighbor This viewpoint toward phones is certainly

PAGE 154

154 not the norm, but it is worth noting that there is a counter current to the general desire to own phones among rural Fijians. Handset M odel Beyond mere phone ownership, the model and style of phone also say something about the owner. Owning the latest model, with the most advanced functions, could indicate that the owner is te chnologically savvy, aware of the latest developments in technology and has the financial means to purchase a high end handset model (or at least prosperous kin capable of giving it as a gift) rural Fijia ns seem to pay attention to specific brands or models. Villagers are generally aware of the differences between the most common types of handsets, and their differing capabilities r craze to acquire the latest or greatest phone model. main concern seemed to be whether a phone featured a memory card, a camera, internet access, or had slots for multiple SIMs. Fijians would think it bizarre to spend the night outside a store in order to buy the latest iPhone. Indeed, most villagers receive their mobiles as second hand gifts from relatives and so must make do with wha tever phone they are given. Because villagers have persistent problems with phones breaki ng or malfunctioning they sometimes express a preference for cheaper phone models, believing they are sturdier and break less easily than more expensive models. Several villagers proudly boasted about the longevity of their battle scarred, entry level Nokia or Motorola hands ets explicitly valuing toughness and basic functional ity over advanced capabilities.

PAGE 155

155 The majority of villagers I surveyed had basic models that enable only calling, texting, a clock/alarm, and games. However, phones with multiple SIMs, cameras, memory car d (which enables storing and playing music) and internet are becoming increasingly popular (and common) in villages Phone lights are a coveted feature, and are us ed on a nightly basis to find One villager explained that while Fijians can get around the village quite comfortably by the light of the moon, the phone light is very useful for cloudy or moonless nights. I have also seen phone lights used for catch ing land crabs at night ( cina va lairo ) and for reading B ibles during evening church sermons. Ownership of an advanced model phone, and conspicuous use of features such as music and internet may in indirect ways serve to enhance in the village context Such features are most important among te enagers and youths, who often use their mobiles as a source of entertainment when they socialize. Youths spend most of their free time hanging out together, often simply relaxing on the floor of an empty house, with nothing to do. The ability to play music is greatly a ppreciated in such situations, because it helps to pass the time, and gathering. these idle times, and use Bluetooth to trade songs between phones. A few youths have learned how to open their phone casings and attach wires, so that the phone can be connected to a stereo. This allows them to act as DJs for village dances, something that definitely enhances social prestige in youth soci al circles. A few youths in the study villages have internet enabled phones and extensive knowledge of the internet. These

PAGE 156

156 individuals sometimes entertain their friends while drinking kava, by sh owing them funny Youtube videos, dirty pictures, or jokes fro m the internet S uch technological savvy is less common, and less highly valued among village adults. Musi c plays less of a role in adult social gatherings, which usually revolve around kava drinking accompanied only by quiet conversation Several adults have music and internet enabled phones, but profess a general ignorance of their functioning. Adults do appreciate the abilit y to play music on their phones typically religious music or Fijian si gidrigi tunes, as opposed to the contemporary pop and regg ae preferred by the youth. However, most adults through Bluetooth so they must ask their tee naged sons and daughters to install music on their phone s for them. Personalization of the H andset An extensive literature has de veloped around the topic of mobile phones, fashion and the body. This literature grew out of the fascination with the portability of phones, and how we sometimes experience mobiles as literal extensions of our bodies, or at least as an integral part of our bodily fashion (Fortunati, et al. 2003; Katz and Sugiyama 2005; Oksman and Rautiainen 2003) For example, a recent Verizon commercial In the commercial, a man sits in a futuristic chamber, passively receiving various technological ( ). The vision promoted in the ad is of a phone that operates so seamlessly and int uitively, that it becomes almost like an extra

PAGE 157

157 organ or appendage. This body/phone synthesis progresses as we move toward ever more compact phones that feature an ever expanding array of tools. Decoration and personalization is one aspect of this synthesis Common forms of personalization include custom handset casings, which come in a wide array of colors and designs, mobile phone charms, which are small decorative beads or figurines that display, and use of a unique ringtone or ringback tone. In Jamaica, some women color coordinate their endowed assistant who always kept her phone firmly lodged between her very eviden Another Jamaican woman described dressing in clothes such that the style and quality In Fijian villages such flaunting and displays were not much in evidence; ostentatiousness is gen erally frowned upon, and would likely expose one to gossip and phones. The only forms of ornamentation I encountered were the occasional sticker on the phone casing (usually religious in nature), or setting a family picture as the screen background. This is possibly due to lack of availability: phone decorations such as The gener observation in Jamaica that people in rural areas were less concerned with style than were city dwellers. Paradoxically Fijians seem to decorate their phone chargers with more gusto than the phones themselves. This is a strategy to prevent theft: there is always a

PAGE 158

158 shortage of functioning chargers in the village, so they are often swiped when left unattended. As a consequence chargers are plastered with unique stickers, painted with so that the owners can unambiguously identify whose charger is whose One form of embellishment that is common among youths (but not meant for public display), is to write the names of phone contacts in a stylized font. For example, address book, which might serve a fu nction of protecting privacy, but as one secondary schooler explained, there is no real purpose to it, its just so that it looks good ( rairai vinaka ) which play a popular so ng either when your phone rings, or for the person on the other end of the line while they are waiting for you to pick up the phone. M purchase these, instead simply alternating among the various tinny ringtones pre programmed in to thei r handsets One exception, a giant of a man who hung his phone around his neck on a string, Cagi received so many phone calls He had a customiz ed ringtone which was a Fijian sigidrigi chime out repeatedly again during the village kava sessions. Carrying and Displaying the P hone Fijian villagers tend to carry their phones with them most places they go. For the most part, Fijian men carry their phones in their pants or shirt pockets. Women in the village typically wear shorts underneath a n i sulu (sarong) and so carry their phone

PAGE 159

159 hidden in their shorts underneath the outer i sulu One young woman I met kept her phone inside her bra, and a few villagers hang their phone s around their necks by lanyards. One of the oft stated reasons for carrying phones everywhere is that an unattended phone is at risk of being stolen. More likely though, a nosy villager might pick up the phone and look through the contacts, recent calls a nd messages. Such snooping is tolerated to some degree within the village, despite the fact that some Fijians are very sensitive about the privacy of their phones. For that reason, people tend to immediately erase any text messages, and often prefer to mem orize (cram) sensitive numbers rather than have them visible within their address book. Phone calls are allowed almost everywhere and anytime. The major exceptions are church, school, and formal meetings, for which most people leave their phones at home, or at least put them on silent mode. There is some debate about the politeness of making phone calls during kava sessions, and some members of kava parties express annoyance at ringing ph ones. For this reason men at kava sessions sometimes get up and go ou tside to answer a call, which one villager app The general ambiguity toward the use of technology during kava drinking make s sessions (incl uding the sevusevu ceremony) phones, cameras, and other technology are generally frowned upon, whereas during casual kava drinking it is not a problem. Using the P hone Mobile phones feature an ever expanding array of functions. For the purposes of this stu dy, the primary functionality is communication at a distance, through either voice, text, or internet. People may use phones for explicitly goal oriented communication, but

PAGE 160

160 social interactions through the phone may also serve unintentional, and even unnoti ced In this subsection I describe in general terms the range of purposes, content and styles of phone communication. I also present some basic data such as the clock, games, music, and so forth Documenting the purpose, content, and style of phone based communications can be difficult, because phone users do not necessarily want to share their conversations with resear chers. Even if a researcher manage s to fortuitously overhear a phone call, s/he is usually only privy to one side of the conversation Texts are often very private in nature, and likewise are not generally shown to others. For these reasons, I did not attempt to systematically record the content of calls or texts. I instead spontaneously observed phone interactions in the course of daily life in the villages, and recorded my qualitative impressions I supplemented these observations with data from interviews with community members about their uses and perceptions of mobile p hones. Finally, I possessed a Fijian mobile phone throughout the period of my fieldwork, and freely gave my number to any acquaintance who asked for it. I was therefore able to participate personally in many phone i nteractions with Fijians and get a feel for the purposes, styles and etiquette of phone communication Voice C ommunication Many phone users frame the benefits of phone ownership in terms of the accomplishment of concrete goals: security, work, or socializing. A single phone call can simultaneous ly fulfill various functions reestablishing contact with an old friend, relief of stress or boredom, rel ating gossip, apologizing for a transgression or asking a favor.

PAGE 161

161 Furthermore, these functions may arise from pre planned goals or they may be fortuitous side effects of spontaneous, free flowing interactions. Horst and Miller noted in their study of mobile communication in Jamaica that a surprising number of calls ar e extremely short and devoid of specific content (2007:17), consist ing entirely of an exchange of pleasantries This observation dovetails nicely with s admission that the average len gth of a cell phone call is a mere nineteen second s (Hors t and Miller 2007:96). Horst and Miller refer extensive, fluid phone based a way to maintain a wide array of potentially useful relationships, which might someday pay off in terms of companionship, sex or material assistance, but which in the immediate term simply provided the benefits of social contact. The frequent use of phones for apparently non instrumental purposes in Jamaica runs up against the perceived virtu es of phones cited by users in many studies, such as safety in emergencies, or for conducting business while on the go. Given that so many users describe their mobile use in instrumental terms, the scarcity of instrumental content in phone calls is something that begs explanation. Perhaps we can turn to Malinowski theory that much of the communication that goes on between humans is not for the purpose of transmitting information; rather, the very act of communication reinforces social bonds Malinowski dubbed this seemingly idle com ( 1923) If a large proportion of p hone calls are phatic in nature then a tie in a calling network might simply indicate that two actors periodically engage in such small talk over the telephone; perhaps no information of major educational, political or economic

PAGE 162

162 importance passes between t hem. Rather, the would represent the group with whom one engages in small talk as a means to reinforce relationships, strengthen solidarity, and provide mutual emotional support. The opposing view of phone use wou ld be that it is purely instrumental. In this view, perhaps, communication through the medium of the telephone lacks much of the affect of face to face conversation. Our e veryday needs for small talk and phatic communi on are largely satisfied face to face with those in our immediate surroundings while the phone is mostly used to transmit messages with an overt purpose piece s of in some socially or materially advantageous way For example someone drifting at sea in a broken down boat can call to shore to arrange a rescue. Someone in need of $50 to pay their school fees may call a relative in the city to arrange a money transfer. In reality, communication over the phone can conform to eith er of these modes but usually incorporates aspects of both T he relative frequency or emphasis placed on phatic or instrumental communication may varies according to the individual phone owner and depend upon whom one is calling the time of day, season of the year, cost of calling and other factors In order to better understand the variety of Fijian phone interactions and their underlying motivations I will discuss my observations from Toba Cagi and Veidogo Phone conversation: purpose, content and st yle In the village households in which I lived, many phone calls were made with explicit goals in mind. I know this, because many calls were preceded by a short discussion We should call Eseta (our daughter) today and see if sh e put that package on the ferry Thus, the immediate

PAGE 163

163 purpose for many phone call s is of the call is typically interwoven within a free ranging conversation that involves the exchange of a broad variety of information, as well as expressions of sentiment that reinforce social relationships and influence the emotional state of the call participants In fact w hen I asked Fijian villagers in interviews why they call people, a frequent respon va bula ga, vakatataro tu ga ). Observations confirmed this: nearly every phone conversation I overheard included an exchange of greetings, and questions about the well being and everyday lives of imme diate family members. The following example is a realistic (although invented) account of such a mixed purpose call. O ne afternoon Raijeli who lives in the village, calls her sister Asenaca, who lives in the city. In the course of the conversation, Raij eli and Asenaca chat about their health, the weather, what they ate for dinner last night, and accounts of some recent events in the village and the city. Parts of their conversation involve loud ice descends nearly to a whisper, so that passersby cannot overhear the salacious tidbits she is relating to her sister. This small talk while superficially useless, serves to keep both Raijeli and Asenaca informed about the general goings on in the geog raphically dispersed community. Raijeli and Asenaca can use this knowledge to contribute meaningfully to future conversations in their respective face to face peer groups. In fact, Raijeli will share some gossip she gleaned from this phone conversation lat er that same day, when

PAGE 164

164 her role as a competent and knowledgeable woman in the community, somebody who Discrete pieces of mo re immediately might enable one or the other of the sisters to initiate preparations for future needs, or to avoid potential social faux pas or economic mishaps. For example, Raijeli tells Asenaca that one of th eir kinsmen in the village is planning a wedding. In response, Asenaca can begin to acquire the necessary resources to make a contribution to the wedding well in advance. Asenaca update s Raijeli on the health of their distant cousin, who is convalescing in an urban hospital. Based upon this information, Raijeli can call the has passed away. This small talk is perhaps unplanned, arising fortuitously from the flow of the c onversation. Raijeli did not consciously plan to ask her sister about these topics when she decided to make the call, but the information is sufficiently useful, in and of itself, that it would have justified the call. In fact, Raijeli initially decided t o make the call because she needs some colored yarn to finish the fringe of a decorative mat she is weaving. After a half hour of small talk, she finally asks Asenaca if she can send some yarn on the next ferry. Asenaca sometimes feels burdened by the freq uent requests of her sister in the village, but after to our cousin e assures Raijeli. Asenaca see s this as a good opportunity to put in a counter kakana dina starchy root crops that are the foundation of every Fijian meal ) here at our house and

PAGE 165

165 responds. When Raijeli hangs up, she feels a vague sense of well being that is difficult to de scribe. She may attribute it to the fact that she has solved the problem of obtaining yarn, but the feeling may equally derive from reconnecting with her sister, whom she only gets to see once a year during Christmastime. The call also yielded some interes ting bits of information (a few of which may be useful in a practical sense). Although Raijeli now is obligated to send her urban sister a bag of taro, she can probably leverage that gift to make a future request of Asenaca maybe $50 toward the annual fun draiser ( soli ) that is coming up in 2 months. I invented this conversation between Raijeli and Asenaca, but modeled it after the pattern of many actual phone interactions I overheard in Fijian villages. The point is that the instrument al functions of the p hone call exchange of economic resources, requests for assistance, relation of discrete pieces of important information tend to be embedded in a matrix of small talk. This small talk in itself fulfills important social and emotional functions, and may le ad to altered courses of behavior on the part of the call participants as they purse their everyday social, economic or political goals Generally speaking, Fijians may use their phones to discuss the full range of topics, and to express the full range of emotions that they express in face t o face conversations. As with face to face communication, much of Fijian phone conversation consists of greetings, pleasantries, and semi formulaic j okes that are typical among certain types of kinsmen For example, cros s cousins ( tavale ) or people from

PAGE 166

166 communities that share a teasing relationship ( tauvu ) will likely open and end any phone conversation with jokes and playful insults. One of the patterns I noticed was that Fijians talk extensively about food over the pho ne. This is not surprising : Fijians are passionate about food, and meals serve as a comforting reminder of home and togetherness. However, long distance conversation about food between people living in different areas of the country also serve s as a sort o f economic and social condition and celebrates the divergent lifestyles, climates, and cuisines found throughout Fiji. In this connection, Fijians do not consider it rude to answer the phone when a relative calls during a mealtim e Quite the opposite the phone call often becomes a and passed around to various family members so that they can personally say hello. Behaving as if the caller were present in the room, the pe ople eating invite the person on the other end of the line to come and join in the meal: The person on the other end thanks them for the invitation, and detailed description of the meal ensues, often with expressions of approval or jealousy on the other end of the phone. inned fish the usual you know Fijian conversational emphasis on food is deeply ingrained, and is part of the structure of relationships between kin groups and regions in Fiji. Many villages have a typical food for which the village is famous. For exam ple, if a certain village is known for seaweed ( nama nama

PAGE 167

167 as a way of expressing that they were traveling to that village. Certain regions of Fiji stand in a customary relationship called veita bani or veibatiki In such relationships, there is a pair of taboo foods, which people from the respective regions may not eat in when Wai droka villagers in Rewa and natives of Namosi prov ince eat together, the Waidroka people must not eat fish, and the Namosi presence, they will choke on the bones ( ora ). This is the basis of much good natured teasing that serves to break the ice at so cial events. This teasing also has a socioeconomic dimension. Thus, in Cagi someone on the phone might emphasize the fact that they are eating t aro leaves, a village staple food. The reference to taro leaves can evoke warm associations of home or childhood or it can be used by a villager in conversation with an urban kinsman as a subtle way to express their relative poverty and lack of culinary options. Conversely, those in the city talk about eating bread, or noodles, or canned food, which are store bough t foods associated with city life. These dietary discussions can take the form of playful boasting, complaints, or wistful pining for a favorite food that is locally unavailable. One time we were having a lunch of shellfish ( sici ) in a neighboring village and a Cagi woman received a phone call from her husband who was chopping firewood in the forest. She held the phone up next to a pile of empty sici On one occasion I overheard the contents of a meal intentionally misrepresented over the phone, I suspect as a lead

PAGE 168

16 8 you know how life is in the villag Cagi housewife mournfully told her daughter who works in the city. In fact we were sitting in front of a sumptuous spread, which even included a chicken purchased at the village store. Divulging the true contents of this ample meal would not fit well with a request for money, however, which should in principle only be made if one is truly in need. Given that the content of phone calls is so rich and varied, the ir ultimate is For example, phone calls between adult kinsmen tend to be family affairs ( ka ni veiwekani ) in which the conversation revolves around the a ffairs and well be ing of household members and shared kin. Among my host families, Sunday meals were a favorite time for phone calls to grown siblings, grown children, and their families who might speak on only a monthly or yearly basis In such contexts the phone is often set simultaneously in the c onversation. Child are prodded by their parents to greet their kinsman in turn through the phone, while th e adults conduct the bulk of the substantive conversati on. incorporate requests for money, goods, or farm produce. Calls between same sex youths tend to be less concerned with family matters, and more with personal acti viti es and social goings on in the area Unmarried youths call friends in other parts of Fiji (or even in the local area) and may talk about recent or upcoming social events or brag about exploits with

PAGE 169

169 members of the opposite se x These calls, much like face to face interactions, often feature friendly insults, teasing, and dirty jokes. Calls between unmarried Fijians of opposite sexes are usually made with t he intent of flirting, romance or courtship. A distinguishing feature o f these conversations is that they are typi cally private in nature. If a youth receives a call from a girlfriend while engaged in kava drinking, he may step outside, but I have also seen youths carry on such conversations while sitting in the kava circle, murmuring so softly into the phone that not even people sitting a few feet away can discern what is being said Generally though, youths use text messages for private communications with their girlfriends and boyfriends. Some calls involve gossip ( kakase ) about other villagers ; married women in particular are often accused of using telephones for gossip. S uch gossip can come around full circle when the object of ridi coconut This can lead to silent antagoni sm or even open confrontations in the village that threaten the peace and solidarity of the community. Two village husbands whom I phone gossip. Some of the most extensi ve phone users in the communities I studied were middle aged men in positions of responsibility in the church or village government. These include ministers and preachers, the government headman ( turaga ni koro ), and members of the school committee. Mobile phones allow these civically engaged individuals to arrange meetings and weekly rotations of preachers for church services, and to contact government agencies about requests for funding for development

PAGE 170

170 projects. Outside of calls to kin for various kinds of assistance or event coordination, these local administrative calls formed the bulk of instrumental phone communication in the village. Cost considerations Whatever the purpose of a call, one of the key considerations that shapes the course hone conversation s is the availability of phone credit. The desire to save money was a consistent theme in interviews with Fijian phone owners C alls about important matters must sometimes be postponed or cut short due to the lack of credit Conversely, Fi jians are usually eager to an swer incoming calls because it involves no Table 4 4 Domestic calling and texting prices for Fiji mobile providers (March 2012) In Network Peak/min In Network Off Peak Out of Network (Peak) O ut of Network (Off Peak) To Landlines (Peak) To Landlines (Off Peak) Text in Network Text to other Network Digicel 0.36 0.36 0.4 0 0.4 0 0.4 0 0.4 0 0.12 0.15 Inkk 0.3 0 0.3 0 0.45 0.39 0.44 0.22 0.2 0 0.2 0 Vodafone 0.44 0.36 0.45 0.41 0.45 0.3 0 0.15 0.2 0 All prices in $FJ. Peak hours vary by provider 7am 7pm for Digicel, 7am 5pm for Inkk network as the caller. Interconnection fees for calls ac ross different networks are hi gh in Fiji. In fact, most respondents told me that as a matter of policy they never call anyone on a different phone network. For this reason, v illagers often try to get a phone from the same company as the phone of a significant other/family member. Thus, a woman in Batiki sought out a TFL Handi mobile the only one in her village, simply so t hat she can call her daughter at her boarding school dormitory, which only has a TFL phone.

PAGE 171

171 When Fijians are being charged based on the duration of a call, they tend to keep conversations brief. The briefest of all conversations occurred between subscribers of the Inkk mobile network. Unlike Vodafone and Digicel, which charge in 30 second increments, Inkk charges f or each second of call duration. Thus, one can save mon ey by using Inkk by having very short conversations. Some c onversations on Inkk phones that I overheard involved only a very brief greeting, quickly relating a key piece of information, and hanging up with hardly a goodbye. The contrast in conversational s tyle was especially strong when comparing paid calls with calls made during a FreeCall promotion FreeCall, which cost $FJ 1.50, provided a user with cost free in network calling during a 24 hour period The ability to make protracted phone calls enables t thout any need to feel rushed due to the mounting cost of a call Ofte n, the recipient of a call ask s up fro nt if the caller has FreeCall and gauge s the pace of their conve rsation accordingly. I witnessed on several oc casions people during such calls placing the phone on about some other task. These conversations proceeded with long natural pauses, almost as if the person on the oth er end of the line were present in the room.FreeCall thus plays an important role in communication between relatives and loved ones living in distant parts of Fiji. In summary, the general sort o f conversations between kinsmen and friends weave together i dle conversation, socially useful information, cooperative planning of events, and requests for assistance. The extent to which conversation s meander is heavily shaped by cost considerations which in turn depend on promotional offers

PAGE 172

172 Special purpose C al ls There are a few special types of phone interactions tha t do not conform to the mixed phatic instrumental model of conversations between kin and friends: emergencies, health related calls, calls made to assist with committing crimes, and pranks. E mergenc i es Early adopters of mobile phones in Western countries often cited personal safety as one of the principle reasons for obtaining a mobile phone (Ling 2004). The portability of the mobile phone means that when one meets with an unexpected problem, one can safety related functions multiplies. For example, during the riots of 2013, man y Egyptians have used an online p), which allows users to rate, on a color coded sca le from green to red, the level of violence occurring in each neighborhood of Cairo, so that other users may avoid danger ous areas ( ). One type of emergency ( leqa tubu kos o ) frequently mentioned by villagers was maritime disasters. L ike other Pacific islanders, Fijians travel extensively between islands, often in small open boats ( fiber ). Very few of these small boats have a VHF radio s or emergency equipment on board. These boats frequently have engine trouble, and may drift for days or weeks before encountering land. Fijians recognize the life saving potential of mobile phones in such situations. Mobiles can be used to summon emergency transport, so that ill or injured peo ple can reach a hospital quickly. While m ost rural villages are in close proximity to a nursing station, these nurses have only limited expertise and equipment. R eaching a hospital capable of dealing with life threatening conditions may require a journey o f

PAGE 173

173 several hours in a truck or open boat In such cases the availability of a mobile phone may be the difference between life and death. Other medical situations are not urgent or life threatening, and the phone merely serves as a convenience in accessing h ealthcare Manoa a Cagi village man, has a sickly wife and children who often have minor illnesses He sometimes calls the village nurse who lives just a few hundred meters away, to ask if he can bring his sick wife over for a checkup. Manoa emphasized t hat the phone makes such communication physically easier, especially if he wants to contact the nurse after dark. I did not hear any first hand accounts of phones being used in medical emergencies, but simply having a mobile phone provides Fijians peace of mind in this regard. The most common place in rural Fijian households is economic villagers sometimes need a quick infusion of cash to pay school fees, participate in a village fundraiser or to fund an unexpected funeral. Few village rs have sufficient savings to deal with such situations. For example Jioji, a Cagi man, fell ill shortly before the yearly fundraiser. Unable to go to his farm to harvest kava to sell in order to fulfill his obligation Jioji called his brother in the city and asked if he could send some money on short notice I describe the use of phones in remittance sendi ng in detail in Chapter 6 While the incidence of violent crime is low in Fijian villages, rural informants value th e feeling of safety they derive from the ir mobile phone s. This is frequently out of concern for the safety of their children as they traveled to and fr om school. For this reason some parents, particulary in peri urban areas, give their children phones to carry to school.

PAGE 174

174 Health Studies have f ound at least three areas in which mobile phones can improve health service delivery. Mobiles can increase efficiency of direct contact between service providers and patients, allow patients to self monitor their conditions (via phone based consultation wi th doctors), and give isolated and remote groups increased access to health services (Mechael 2008). At the present time in Fiji mobiles play a beneficial, although indirect role in facilitating contact with health service providers although there are mobile based programs in place to provide diagnoses, health advice and warnings about epidemics, Fijians in the communities in which I worked did not use their phones in this way to directly acce ss information from health providers (Valemei 2013). In fact, r adio and TV are much more intensively used to disseminate health information to the Fijian public. For instance, Fijian TV has public service announcements nearly every commercial break that d iscourage binge drinking, suggest that food tastes better without salt, and even a music group strumming guitars and s Wash your ha Other than helping Fijians to access transport during medical emergencies (described in the previous section), mobiles assist Fijians in gaining access to traditional remedies Fijians make extensive use of herbal medicines and usually attempt to cure themselves before resorting to prescription drugs Fijian tradit ional healing tech niques include massage, and use of a w ide variety of medicinal plants which may be ingested, rubbed on the skin, or otherwise administered (Cambie and Ash 1994). My Cagi host mother, for example was an expert in massage ( veibobo ). This ty pe of massage was

PAGE 175

175 used in the case of minor maladies or aches, and involved massage of the stomach and back. Patients would typically come to the house for massage for a series of 4 nights, giving small gifts to the masseuse for her services. The mobile ph one allows long distance consultation with experienced herbal practitioners in order to determine the appropriate cure for a malady. I overheard numerous phone discussions, in which the affliction of a villager was being described in detail to a knowledgab le relative in a distant part of Fiji who then prescribed a particular plant infusion or emetic. Because many Fijians now live in urban areas without easy access to forest land the phone serves as a conduit for obtaining medicinal plants from relatives in the village. Bags full of tree bark, leaves, and other medicinal herbs are sent to urban kin by boat or bus in the event of illness This assistance may eventually be gratefully reciprocated with gifts of bags of flour, or some other item that is scarce in the village. Medical problems in rural Fiji are often (at least partly) attributed to evil spirits (Becker 1995:104 106) Many Fijians believe that such spiritual afflictions can be diagnosed or healed through prayer. For example, a young woman in one of my host families had nightmares, and frequently woke up in the middle of the night wailing or crying. After one such episode, her father woke everyone in the household, and they sat together and p rayed together in the darkness. act list of her children is sick, the woman calls this man, who prays all night long. In the morning he calls her back to inform her of the source of the illness, and the correct course of spiritual/medical treatment to address it.

PAGE 176

176 Finally, mobiles help to alleviate the longer term financial and social burdens of illness and injury. For example, the husband of a married Cagi woman had his arm badly injured in a farming accident. He spent several months in Suva convalescing, during which time the phone provided a means of communication and emotional support between them. In Western countries there have been concerns that mobile use can damage our health due to radiation emitted from the handset next to the head ( Hossmann and Hermann 2003 ). Fijians seemed generally unaware o f these issues. Only one man, Mikaele, ever mentioned the topic to me. A n avid consumer of news, Mikaele had heard on the radio about research linking health problems to mobile phones; he interpreted Mikaele came up w ith a creative way to combat this virus: during a phone conversation he frequently switches the phone between his left ear and hi s right ear, to reduce the likelihood of infection on either side of his head. Another emerging mobile safety related issue in many countries is texting while driving (Wilson and Stimpson 2010). In rural Fiji there are few vehicles, and I did not note any conversations among villagers on this topic. Crime Serious criminal activity is rare in Fijian villages. However, many villagers urban areas. One commonly described crimin al use of phones involves an accomplice who spots a target visiting an ATM or walking out of a shop. The lookout calls ahea d to a group of friends, who intercept the unsuspecting target and steal the freshly withdrawn money. Mobiles can also be used by loo kouts while a robbery is in progress, to give a warning to those inside the building if the residents or the police are approaching.

PAGE 177

177 Concerns about mobile orchestrated crimes are mostly relevant for those who live in urban areas. In rural areas, mobiles ar e more commonly the targets of theft, rather than being used as an accessory in thefts. Several i nformants mentioned having had their phone stolen in the village. Often these thefts seem opportunistic and unplanned: someone enters a village house looking f or somebody, and finding nobody around, sees an unattended mobile phone and swipes it. Such a crime is hard to get away with in the village, unless the phone is promptly given or sold to someone in a distant place. More frequent targets of theft are phone chargers, which are often left plugged into wall sockets, unattended. There is always a shortage of functioning chargers in the villages, and it is n a daily basis. This serves as ample incentiv e for theft. As a precaution, Fijians often ornament their phone chargers with stickers and designs, so that they could be identified in the case of loss or theft. Pranks M ischievous villagers have developed telephone prank calling into an art form by usi make a call phone number on their screen call people who would probably not otherwise answer the phone if they knew the true identity of the caller. One might ask why such a phone setting exists at all; or why anyone should want to answer fter all, the intentions of any caller should be questioned if ow who is calling. The reason, as it was

PAGE 178

178 explained to me by a villager, is for their own protectio n ll into the wrong hands S ome Fijians have a policy of never answering calls from private numbers. If they do answer, and it turns out to be a friend, they may affectionately complain, bastard!). friend repeatedly, hanging up each time before the friend can answer. This is a form of teasing ( vakatikati a village youth, and impersonating his girlfriend (or just flirting with him anonymously). The girls try to get the youth to respond to the flirtations, and to say something embarrassing which they can use later to tease him. One Cagi youth described a mor e elaborate prank in impersonates a relative in Suva. The prankster tells the victim that he has sent him money via TMO (a telegraph money order). The victim eagerly makes the t wo hour walk to the Nasau post office to pick up the gift, only to find t here is no money waiting there. Text Fijian villagers overwhelmingly pre fer voice communication to text, and i n fact many villagers do not know how to text. This text aversion is not universal across cultures. For example, text has acquired great popularity, and even become a preferred mode of communication in the Philipines, Japan, Finland, and many other dev eloping and developed contexts (P ertier ra et al 2002; Ito et al 2005). O ne of the primary rationales for texting is cost on many prepaid service plans, texting is considerably cheaper than calling, appealing especially to young people or low income phone users.

PAGE 179

179 communicat ing unobtrusively in places such as church or school. Text allows users to send basic messages while avoiding the pressure, awkwardness, or unpredictability of verbal conversation. The asynchronous nature of text communication allows one to carefully compo responding. Finally, text messages, unlike calls, leave a lasting record, so that m read over and over again. Text ers have developed conventionalized styles of expression to partially overcome the un expressiveness of the medium. For example, joke or a flirtation and that offen se should not be taken. Given all of the potential advantages of text, it is perhaps surprising then that mo st rural Fijians rarely use it When explaining their aversion to texting, Fijian villagers told me that when they use the phone they want to veirog oci domo I am sad (e mosi vei au) said Aqela, a Cagi housewife whose husband was convalesci I need (se bula ti ko se mate!) she said half jokingly. Many respondents also be excessively difficult and time consuming. A few older respondents said that their eyes are bad, which makes texting difficult. In fact, a large proportion of villagers do not know how to text, s o these explanations may in part be an evasive way of admitting to a lack of technological savvy. A few villagers claim to be avid texters, but qualify this by saying that they only use text in lieu of calling when text is free of charge due to a promotion. Livai, a young

PAGE 180

180 married man, said that when he bought his Digicel phone it came with 1000 free texts. When texts cost money (albeit still cheaper than a phone call), calling is almost universally preferred. Text messaging is used most extensively among youths in rural Fiji for a narrow range of purposes. The main reasons for texting are to send a surrept itious message to a lover, to send a message about a private kava drinking party, or to participate in a called areas, probably because most phones are not equipped with cameras. The fact that texting is typically used for romantic communications in Fiji, coupled ever the opportunity arises, militates against the saving of personally meaningful texts in phones. I n fact, i t is very typical for Fijians to erase a text immediately after reading its contents, for fear that someone else might read it. Survey responses about frequency of texting confirm the general observation that text is primarily used by young, single people to communicate with boyfriends and girlfriends: t he mean age of those who send 6 or mo re texts a day is 24 years old, while many older phone user s do not know how to text. When asking villagers about their texting habits, I often encountered some confusion. That is because Fijian phone pro viders have created popular text services texts, and phone credit requests. These cost free services allow users who do not have any phone credit to enter a numeric code that send s a text to

PAGE 181

181 another phone, ormulating a unique message, but rural Fijians use these free text based services much more frequently than they send ordinary texts One of the factors preventing more general use of text in Fiji is that predictive text is not helpful for texting in the Fijian language. A couple of villagers actually sought out my help to turn off the English language predictive text on the ir phones, which had accidentally been activated and thoroughly puzzled them. Also technology, in that one only t ends to text with other friends who habitually use text. Because the pool of avid texters in rural Fiji is very small, it discourages the widespread sending of texts as a general practice. Once text does eventually catch on in rural Fiji, it may become pop Internet Internet, particularly mobile broadband, is rapidly gaini ng popularity in Fiji. A t the time of research however, internet was just beginnin g to catch on in rural villages. O n ly 30 out of 146 (21%) of mobile phone owners in Toba, Cagi and Veidogo claimed to have ever used the internet on their mobile phone or otherwise For many of these villagers exposure to the internet was been very limited perhaps only accompanying an ur ban dwelling cousin to an internet caf once or twice. Only 14 (10%) use d internet regularly on their mobile phones. Several other respondents noted that their mobile phones were capable of accessing the internet, but that they do not know how to use this feature. The mean age of villagers who said they had ever used the internet was 35; relatively young. Moreover, those few villagers who frequently use mobile internet are

PAGE 182

182 almost all unmarried youths. Most explained that they learned how to use the interne t during computer class in secondary school. The rest were shown by a friend or relative during a visit to an internet caf. 24 reported having used Facebook, 16 reported having used email, and only 4 Skype. A few mentioned using the internet to read the n ews, check sports scores, or watch videos on YouTube. Facebook is definitely the most commonly known website among rural Fijians, as a multifaceted communication tool. It enables rural Fijians to contact friends and family both in Fiji, and overseas through both synchronous (chat) and asynchronous (posts, comments) modalities. Facebook also facilitates posting and looking at pictur es of other people. In the context of youth and relationships, Facebook is a good way to find pictures of attractive potential partners. Fijian youths often add a n attractive stranger as a Facebook friend, chat with them online and then ask them for their phone number. A few village youths said they had partaken in long distance relationships initiated through Facebook. The internet, and Facebook in particular, could play an important role in tying together the Fijian tran snational community in the future provided that there are enough outgoing, internet savvy individuals present in the village s Pelasio, one of the male youths in Cagi was remar kable in this regard; he uses his smartphone to access and post on Facebook on a daily basis. Whenever there is a c ommunity event, such as the construction of a new house or a wedding or funeral, he takes multiple pictures with his phone and uploads them to Facebook the same day.

PAGE 183

183 Pelasio is Facebook friends with many of the Cagi villagers living overseas and in Su va, and they frequently comment on the photos he posts. These interactions strengthen community bonds by keeping migrants informed of village events, problems, and celebrations in real time For example, one time Pelasio posted pictures after a heavy flood in which the village road was washed away. For urban or overseas community members, such photos contextualize international news about disasters in Fiji, and invest it with personal significance. Over the long term those living outside the village can k eep tabs on births, deaths, and marriages so that they can stay better informed about the evolving social structure of the community during the years between visits to the village use th e internet, were aware of the possibilities that Facebook presented One man asked me to photograph them doing renovation work on the local primary school and (I to ok this t In Suva and other urban areas, internet cafs are popular. Some cafes are known for their low price s (ab out FJ$1/hour) while others boast colder A/C or a wider range of games. Cafes tend to be packed throughout the afternoon and eveni ng with schoolkids in uniform, hooting and cursing as they play first person shooter games like local network. Some of the internet cafes offered special all night gaming deals specially catering to th is preteen crowd leading to po lice scrutiny (Rawalai 2013) 20 something Fijians expressed embarrassment to me about using one Internet cafes are also popular for using

PAGE 184

184 Facebook or Skype to contact friends and relatives, and for typing and pri nting out resumes and other official documents. monitoring and attempted censorship of anti government blogs on the internet. Since the 2006 coups, a number of Fijians, both domestic and overseas, have run vocal anti government blogs such as coupfourpointfive ( ) and Intelligentsiya ( http://intelligentsiy attempted to block access to these blogs (Anonymous 2007). Secondary U ses of Mobile P hones Horst and Miller observed a difference between phone ownership in developed countries such as England, and in d eveloping cou ntries such as Jamaica. For residents of wealthy countries they asserted, a phone is a tool largely used for calling and texting. (2007:65). One of the underlying factors was a Jamaican value for exploiting every available resource to the fullest getting the most ars, ringtones, and so forth. Indeed, f or people in developing countries, a mobile phone maybe one of just a few prized possessions, so it makes sense to exploit it to the fullest. Many with an important exception : many older vi llagers in particular, have a limited knowledge of phone operation, and therefore are uncomfortable exploring the various submenus on the phone. As Horst and Miller observed in Jamaica (2007: 66), in Fiji many parents rely on their children to figure out various aspects of their phone, such as saving contacts in the address book, sending texts, or resetting the clock. For those with

PAGE 185

185 a good grasp of phone operation ( mostly young people), fiddling with mobile phone s is a favorite pastime. Figure 4 12. Fr equency of secondary uses of mobiles Clock Prior to arriving in the field, I erroneously assumed that Fijians would not make much use of the clock or alarm features on their phones. Many events i n villages are scheduled by general times of day (morning, a fter lunch) rather than at a set hour As a Peace Corps volunteer, my exper ience of Fijian village life had been one of soft schedules and flexible deadlines. After all, m ost village work is independently carried out a farmer may go to the gardens, and re turn home when ever he chooses. Women have a regular routine of housework, but likewise complete it as they see fit during the course of the day. Village events that do run ac cording to a specific schedule, such as church and community meetings, are typical ly an nounced by the village crier, or the beating of a slit gong (lali) which makes a personal timepiece less necessary.

PAGE 186

186 Despite these cultural differences Fijians frequently expressed the importance of features on a daily basis, and in fact note that the two primary uses of the phone are for calling, and for telling the time. Aside from church and other official functions, villagers really value the ability to structure th eir daily work activities. For example, men liked to know how many hours remain until sunset, and women like to know when the It would be interesting to examine whether widespread use of mobiles coincides with gr eater specificity in the scheduling of events, or greater emphasis on promptness. Radio and music Many rural Fijians still have basic mobile phone models that are not capable of playing music. Newer models with memory cards that can store mp3 songs and wit h Bluetooth to allow transfer of songs between phones, are in high demand and are growing more common. The use of phones to play music has become an important aspect of youth culture in the villages. Many youths, both male and female, do not own radios or cd players. Phones fill this gap by providing musical accompaniment for kava sessions, or for just lounging around with friends. Popular songs are downloaded from the internet, typically in an urban internet caf, and diffuse their way across rural Fiji b y successive transfers via Bluetooth from phone t o phone. Village y ouths can often be seen huddled in small groups After the arrival of a new hit song, it quickly makes its way onto all of the village phones (at least those with memory cards), and is played repeatedly throughout the village ad nauseum until a new song arrives to replace it.

PAGE 187

187 In formal gatherings such as community kava sessions, meetings, or even in the household with their parents, youths are often passive spectators in the deliberations. E arbuds (earphones) are much sought after because they allow youths to continue listen ing to music thro ughout long, tedious village event s that are dominated by the speeches and affairs of the community elders. To a le sser extent than downloaded songs villagers also use phones to tune into radio stations. This is not always straightforward, as according to villagers some areas. One young woman taught m e a trick she had learned to improve radio reception on the phone: she takes the shiny metallic paper from the inside of an empty Benson and Hedges cigarette package and carefully rolls it into a long tu be. She inserts this makeshift antenna into the phone which supposedly cuts down on the static. Figure 4 13. Mobile phone with plastic bottle used as an amplifier for music Photo courtesy of author.

PAGE 188

188 Many youths are adept at these sorts of technical improvisations I encountered anothe r youth whose phone had makeshift electrical wires protruding from the casing. When installed properly in the phone the wires can be inserted into a pair of speakers in order to amplify songs played on the phone. A nother a n stereo amplifier consists of the sawn Figure 4 14. Phone with wires installed, so that it can be hooked up to stereo speakers. Not likely sent to the village as a gift from a relative working as a security contractor in the Middle East. Photo courtesy of author. Games A s children most villagers were never expos ed to gaming consoles, computer gam es, or video arcades; thus, rural F ijians do not have a very long experience with, or sentim ental attachment to video games. Nevertheless many villagers are enthusiastic

PAGE 189

189 players of mobile phone based games. Even most basi c mobile handset s have 3 or 4 simple games pre programmed, such as Many children also whenever they are allowed to Although games are considered something for children, I also witnessed adults playing on occasion Sometimes adults and children alternate turns playing, hovering pandanus mat in the evening. For most Fijians, mobile based games are nothing more than an occasional diversion, but it is worth noting that these games represent many doors for the consumption of other forms of media including the inte rnet Other uses 12 respondents reported using their clock alarm on a daily basis, 15 reported frequent use of the calculator on the phone, 4 use the calendar, 21 use the camera. Summary Mobile phones are increasingly common in Fijian communities, but no t every individual, n or even every household, has a telephone. There is an element of contingency and unpredictability associated with phone ownership in Fiji : handsets are often hand me downs that do not last very long. If a phone is lost or malfunctions, it may be many months before the individual can acquire another phone. Lack of a phone is not strongly indicative of poverty, but most villagers recognize that having a phone can bring social and economic benefits. The disadvantages of phonelessness are p artially offset by frequent borrowing of phones among household members, neighbors, kin and friends.

PAGE 190

190 Phone communication is used to accomplish a broad array of purposes, some of which are planned, some unplanned, and some entirely unnoticed by the user. Purpose, content, and style of phone interactions vary between users but one commonality is that the instrumental aspects of phone communication are usually wever, as it can serve se condary instrumental, social, or emotional functions T he length of phone calls important role in allowing conversation uninhibited by monetary constraints. Text is consid erably less popular than voice communication, and many Fijians do not in fact know how to use it. The primary use of text is among youths, who use it to surreptitiously contact their romantic interests. Mobile internet is growing in popularity, but knowled ge of its use is constrained to a small percentage of community members, mostly youths. Facebook is the most commonly used website among village youths, and seems to have considerable potential for tying together the geographically scattered members of the village community.

PAGE 191

191 CHAPTER 5 PATTERNS OF TELEPHONE COMMUNICATION IN RURAL FIJI Introduction he electro Galaxy 126). Although this evocative quote from Marshall McLuhan preceded the advent of mobile telephones by decades it aptly expresses how communication techn ologies such as radio, television and mobile phones have reconfigured the spatial and temporal constraints on human interaction. With the emergence of mobile phones, Internet, and social networks, theorists of the Information Society describe a shift towa rd networked individualism, in which s ocial life revolves less around local, bounded social groups such as lineages clans or villages and increasingly revolves around unique, geographicall y dispersed communities of friends, kin, and other as sociates (Rainie and Wellman 2012). The implications of this shift are particularly fundamental in rural, kinship organized communities, in which relationships rela tive to industrial societies, tend to be more geographically bounded and pre determined by group affiliation. According to a well regarded theory, then mobile phones should permit rural Fijians to maintain larger, more socially diverse, and more geographically dispersed social networks than was possible before the advent of mobile phone s. One way t o analyze the nature and extent of these social changes is to examine the calling networks which egocentric social networks It is through these networks that ideas, information and material resources flow in and out of

PAGE 192

192 c ommunities. Comparing these egocentric mobile calling networks with face to face and other pre mobile communication forms allows us to evaluate how social practices and social structures are evolving with the adoption of ICTs in rural Fijian communities. This chapter has three overlapping goals: to describe patterns of telepho ne communication in rural Fiji, to determine how, and to what extent access and use patterns vary among Fijian individuals, demographic subgroups, and communities, and to understand h ow these patterns relate to the broader Fij ian patterns of social interaction described in Chapter 3 Data and A nalysis data derived from phone surveys I conducted in Toba Cagi and Veidogo in 2011 and 2012. I briefly de scribe the survey in Chapter 2, and full copies of survey forms can be found in Appendix B P rofiles of V illage P hone U sers not do much justice to the unique personalities or personal histories of the Fiji villagers with whom I lived. Often, individ patterns reflect aspects of their personal ities worldview s and social role s ; even in a small Fijian community the sheer variety of ways in which people approach their phones can be surp rising. I therefore op en this chapter with a few brief profiles of phone users in Toba Cagi and Veidogo to illustrate the diversity of approaches habits, and opinions toward mobiles. Peceli: A C all to S ervice Peceli is a 55 year old married man and one of the community leaders in Cagi The senior member of his subclan, he also holds an important leadership role as the manager of the copra mill outs ide the village. I use my phone a lot for bu sine

PAGE 193

193 I talk to co op officials in Suva as well as our customers over the phone I always keep my phone battery charged up and I purchase FreeCall 7 days a week, so I never hesitate to make a call if I need to. Peceli carries his mobile phone on a lanyard aro und his neck wherever he goes, and receives a seemingly endless stream of business and personal calls, which are always announced by his unique ringtone: a Fijian sigidrigi version of the John Denver massive fist entirely conceals the handset whenever he picks up his phone to answer a call. I used to turn my phone off frequently and did no Peceli continued. Then, one day my cross cousin ( tavale ) Peni, who works for Dig icel, kakua ni off taka na nomu talevoni ). If a person calls you to ask how to get in touch with someone, help him out ( vakacegui koya ka bibi ) w hen someone calls and needs a ssistance with something, and maybe you are in a position to help help, you will feel bad ( e na mosi vei iko day So now I accept every call I receive, and am happy about the fact th at someone thought to call me. I receive calls at midnight, calls in the early morning, from overseas. E ven if someone calls me with a private number (which many Fijians refuse to answer), always answer it. This is the right thing to do, b ecause someday I may need One day a woman called me Peceli it was a wrong number. But we jumped right into a

PAGE 194

194 conversation anyway ( keirau v eitalanoa sara ). As we chatted I discovered that this Lauan lady had kin n umber is stored extended his handset out for me to inspect. store my number in your ph I work at the Koro Biofuel plant. So just by talking over the phone I was able to help her out ( vakacegui koya Peceli is distinguished by the fact that he has more contacts saved in his phone (318) than anyon e else I su rveyed in the study villages. I automatically store the number of anybody who calls me his number right in front of me. Because of this m y phone serves like a community telephone director y whenever someone in Cagi or even from another village on the island, and chances are I can tell them the number. If someone from outside Cagi wants to get in touch with our turaga ni koro (v illage headman) call me first. For example, a taro buyer called me once and asked me to announce at the village meeting that he would be coming to the village that week to purchase taro. Just now Apete (ano ther Cagi want ed to call someone in Nabasovi, a village on the other side of the island, know their number. I gave him two numbers for other people in Nabasovi if he calls one of those people, maybe they in turn can give him the number for the person he needs to Cagi people living in Suva. If there is any kind of problem or obligation within my family or mataqali I just pick up the

PAGE 195

195 phone and call our relatives in Suva. Whenever we need something, our kin in Suva help us out era tou sa sema mai, na family mai S on mai, o iratou mai S uva ). People in Suva also call me to ask for the numbers of other people in Cagi benefits. ne, Ryan, the way I see it, is making the whole world smaller ( sa vakavolekataka na talevoni ). I sit here and I can call America and Australia the wide world. It makes this easy. Before, if you wanted to make a call you had to walk all the way to Nasau t wo hours away But now you can call anywhere in the world from this Elisapeci: Reluctant R omantic Elisapeci is a 31 year old woman who settled in Cagi aft er her recent divorce, moving in with an elderly widow to whom she is distantly related. Elisapeci, in fact, blames her divorce on the mobile phone. H er husband was secretly using his phone to contact other women: husband on his pho know about it ; when I finally discovered his cheating I confronted him and we fought and split up Since her divorce, Elisapeci has been receiving lots of calls from men in various parts of Fiji m e ; we just talk on the phone. They say they want to meet me and marry me exclaimed incredulously vinakati au mai Cagi even call me, asking

PAGE 196

196 me to go drink kava with them at night. But if I were to go over there and drink kava with them, who knows Cagi gi ve out my phone number to their friends in Suva as a joke. Then these strangers even want to pick the phone up, and I refuse to also circumspec ven when Figure 5 1. A young woman takes a break from housework to chat on the phone. Photo courtesy of author.

PAGE 197

197 Elisapeci uses her phone primarily to contact friends rather than the kin in her natal village, with whom she has largely lost contact in Lau; I here anymore. My old phone had some numbers from Lau in it, but I lost them all when I lost my old phone. If I had only crammed (memorized) the ir numbers then overall view of mobile phones is ambiguous. useful because it enables me to ask relatives for help, or to send food to my relatives in whether the food that we send to our relatives has arrived at the proper place or not Mobile phones are also really their phones properly i n accordance with Fijian customs So, the phone has both good ia na talevoni talega vaqo e vaka me e seg a ga kina na itovo vakaviti. Sa sega na itovo, eh. Sa tiko na kena vinaka, sa tiko na kena ca ). In light of her own recent divorce Elisapeci half jokingly suggested a solution to the problem of mobile phones and marital infidelity in Fiji neither a man nor his wife own mobile phones Tevita: No nonsense Tevita is 48 year old married m an, soft spoken and with an understated sense of humor. He sees his mobile phone primarily as a functional tool for emergencies, rather than for recreational ask for things from relatives, like people in Suva ( kerekere ). Money, topup, tea, money for a soli call if there is a real problem or some important topic I need to call somebody about In my opinion the amount of kerekere on the part of those in the village can cause family

PAGE 198

198 problems with those living in Suva, because in Suva they need money to live; everywhere they go, bu s or lorry, money, money money But, our kin in Suva also ask for food from us living money is on its way Tevita has only 28 saved con just save any number, only the numbers of a few relatives ( wekaqu of my sisters is married in Bua; and so I have her children as contacts in my phone; another sister is married in Tail e so sara ga na noqu leqa dina qai qiri al kava in the rve the battery need to make calls anyway. Sometimes the phone battery goes dead and I just leave it at home rawarawa ) because if t all know who is using it to call whom, or if someone calls us then ll known incident in Cagi in which one of the village youths was fooled into pursuing a relationship with a n older married

PAGE 199

199 the phone is used carelessly ( vakaveitalia Etika: Relationships Up in S moke Etika, a 27 year old married man, exemplifies how the SIM has only 14 numbers saved in it, but he previously used to have many more contacts. Etika told me the story of how he lost his phone contacts months ago. I was liv ing in Suva at the time and my part time job as a security guar d was slack (I to spend on phone credit However, my SIM at that time had 1000 free texts (a promotion on new Digicel SIMs), so I just texted all the time, sent callback messages, and waited for other people to call me. I got a few warning texts from Digicel that my these numbers belonged to various girlfriends of mine, and my wife was jealous. One day when I was out of the house my wife found my that she burnt a difficult thing losing your contacts, because of the friends hips you lose. For relatives you can just call another family member and ask for th friends it can be very difficult. S ome of the people friends who work in the market, or with whom I studied in secondary school. Some contacts like that are very h ard to get back again. M

PAGE 200

200 know my new number, so they are lost for good ( yali saraga, yali vakadua regain some of the contacts has such and e ti ko vei iko na naba nei ka? But like I said, Talica: A Crutch for Old A ge Talica is a very old w oman. She is bright and outgoing but her mind and body are in decline. Talica lives with her gran She i s u sually confined to the house during the day doing small chores Ana gave Talica a mobile ph two daughters, one in a neighboring village, and one living in Australi a, can call her directly Also, when Talica makes a rare unaccompanie d trip on the bus to Suva, she carries her phone with her so that Ana can call her to check that she is safe In fact, the only thing that Talica knows how to do is to answer a call she In the house sh e occasionally uses the light on her phone, or listens to the radio, but she mu st ask Ana to help her to to do it herself For Talica, the phone is a limited purpose tool ; still the contact it facilit ates with her immediate family provides her with a degree of independence and security. For the elderly in rural Fiji, the mobile phone can be the sole remaining means of engaging w ith life outside the household, and e ve n occasional calls can help to ove rc ome loneliness. As these profiles suggest, mobile phones can serve very different purposes for different people; some of these differences are related throughout the life cycle, while others are based on idiosyncratic differences in personality or the fulfillment of special positions in the community

PAGE 201

201 Analyses of Phone Survey Data I will now present the data I collected from the phone surveys, and discuss what the data indicate about mobile based social i nteractions I break the analyses down into broad categories : the characteristics of phone owners, the frequency/intensity of phone use, the characteristics of people contacted through the phone, the nature of the relationships between village phon e owners and the individuals they contact through the phone, temporal and spatial aspects of phone communication, and a comparison of the calling networks of the three study communities. For some of the analyses in this chapter I simply provide a brief sum mary and include the detailed statistical data and charts in Appendix D for readers who are interested. Characteristics of Phone O wners Mobile phone ownership is becoming increasingly widespread in rural Fiji Nearly every household in Toba Cagi and Veid ogo had at least one telephone. However, as long as some individuals and household s remain without, then phones are scarce resource s It would seem likely that under conditions of scarcity, individuals with dominant social standing would have disproportion ate access phones. In Fiji older, married men are considered the masters of their households, and also have the most influence in community decisions It would stand to reason, then, that mobile phone ownership rates would be highest among older, married men Males do in fact possess a slight majority (56%) of all phones in Toba Cagi and Veidogo P hone ownership rates do not increase directly with advancing age; however. I ndividuals aged around 16 60 have the highest ownership rates, while children and the elderly have low rates of ownership.

PAGE 202

202 Figure 5 2. Characteristics of phone owners in the study villages Figure 5 3. Age distribution of mobile phone owners Percentages indicate what percentage of each age cohort in the stud y communities own mobile phones.

PAGE 203

203 School aged c hildren, in rural Fiji at least, rarely own phones because use of phones is seen as improper or unnecessary Fijian c hildren do not have any sources of income, so unless a parent or other relative gi ves them a phone as a gift, they are unable to acquire one independently. Elderly villagers are free to own and use phones, but several factors tend to limit their engagement with the technology. First many elderly level of social engagement d eclines over time as their circle of peers dwindles and begins to die. S ome el derly people have poor eyesight or other reduced capacities, which detracts from their ability to see the screen or to operate efficiently. As Fijians become e lderly, they often move into the households of their married adult children, and may have a limited say over the allocation of household economic resources. Finally, some aging Fijians feel a degree of discomfort with new technology, and are of the opinion that mobile phones are meant for the younger generation. There are a few notable differences in phone ownership patterns among the study communities A disproportionate percentage of Toba male farmers. This fact, a s well as Toba p mean age (41 years old) may be explained by the fact that mobile ph ones are relatively scarce in Toba and so tend to be controlled by the married male heads of household. Conversely, Veidogo is unusual in that women actually own a slight percentage of Veidogo a n women have jobs in the city. Their independence, mobility and earning power may give Veidogo women a greater say in household economic decisions, or at least a greater rationale for needin g a mobile phone.

PAGE 204

204 Although observed patterns of phone ownership in Toba Cagi and Veidogo largely conform to expe ctations regarding age and sex, there are many deviations from the expected pattern. F or instance several village households had a single mob ile phone, owned and primarily used by a housewife or an unmarried youth rather than the male head of household In such cases, the household unique economic situation, as well as the personalities and social habits of the individuals may be important f actors. Frequency/Intensity of Phone U se One of the key aspects of phone use is how often, or for what duration of time, individuals use their phone s for communication In the ab sence of a way to directly measure this, I did so by asking phone users during surveys to report on their own use frequency/intensity. Self reports about behavior can be problematic, because respondents often have trouble accurately rec alling past or habit ual actions (see Appendix C ) I therefore decided to ask several survey questions relating to phone use frequency/intensity, to see whether the resulting data portr ay a consistent overall pattern. These questions included self reported number of calls and texts sent in a typical day, self reported spending on phone credit in a typical week, number of contacts saved individual in thei r phones, and number of times each indi vidual was named as a their peers in the village. I first discuss each of these measures separately. Then, at the end of this section I combine these variables into a single composite measure in order to rank individuals against th eir fellow community members

PAGE 205

205 on the basis of their levels of phone use frequency/intensity. This provides insights about the relative levels of phone use among villagers with different characteristics. Calling and T e xting Fr equency I asked each phone owner that the amount of calls or texts an individual sends varies greatly depending on whether or not there is a prom o tion i n effect on a given day. F reeC all is a popular promotion in whi ch the phone user pays $FJ 1.50 in exchange for 24 hours of free in network calls and texts Rural Fijians tend to make many calls and send many texts on days when they have F reeC all but when they d o not have F reeC all they generally just wait t o be called or texted by others. Thus, a typical survey re ree C all, I can make 10 calls in ree C ev er I received such a response, I followed up by asking the respondent how many days a week they typically purchase F reeC all, in order to estimate an average daily calling/texting figure. About 90% of respondents reported calling and texting fewer than 6 ti mes each in a typical day. There were a few outliers who claimed to call and text much more frequently the maximum responses were 50 texts and 30 calls per day. Rather than removing these outliers, I decided to incorporate them in the analysis so as to be able to cross check with other phone us e variables, to see whether outlying responses are consistent overall with very heavy phone use.

PAGE 206

206 Figure 5 4. Number of calls and texts on a typical day Figure 5 5 Mean calls and texts per day by subgroup (error bars= +/ SE) On average, respondents make voice calls more frequently than they send texts (a mean of 3.82 calls per day vs. 2.52 texts). Calling and texting levels are relatively similar bet ween sex and marital subgroups, but there ar e broader differences in texting than calling. This is partly due to the fact that over 40% of respondents said that they

PAGE 207

207 never send texts, whereas almost every phone owner claimed to make at least one call a day. C alls and texts per day (combined) are ne gatively associated with age ( r= 0.21 p=.005 ). Y ounger people text frequently ; one of the main uses of text is for private communication between lovers, an activity mainly confined to unmarried youths. In contrast, many older Fijian adults do not know how to text, or choose not to. Figure 5 6 Mean calls and texts per day, by village (error bars= +/ SE) Reported calling and texting frequency appear to vary among the three villages. However, the inter community differences are likely exaggerated by outliers, as the median figures for calling and texting in the three villages do not vary signi ficantly (median test p=.854). The overall picture painted by these data is that the amount of calling and texting that individuals engage in varies greatly from day to day, based on promotions, but overall calling and texting levels are very similar between individuals

PAGE 208

208 with different characteristics The only notable exception is texting, which is largely confined to younger people. Amount of Phone Credit U sed Fo r cash strapped villagers, the expense of phone credit is one of the main factors limiting the number of calls and texts they send. The amount of phone credit used therefore Each of The villagers I surveyed all used the prepaid plan, in which the user periodically adds Purchase of phone credit in rural Fi ji, like calling behavior itself, is heavily shaped by the schedule of promotions. Most villagers are very strategic with the money they use on the ir phone s and generally try to spend the least amount possible. Many respondents report ed habitually purchas ing credit only o nce a week, during the weekly usually every Thursday Saturday The 5 UP promotion gives a user five times the credit of their purchase, so for most respondents it was worth it to put off a credit purchase for a few days until the promotion came into effect Otherwise villagers usually buy FreeCall Petero a Cagi youth, told me that when he adds credit t o his phone he always buys FreeC all, then makes calls a ll night long until the phone Calling all night like that to save money t he joked. Bolou a young unmarried woman, F reecall with the first $ 1.50, and use s the remaining $.50 for texting after 9 pm, whe n texts cost only 1 cent. typical week? How much of that is purchased with your own money, and how much is

PAGE 209

209 credit transferred to you by friends or The mean reported weekly amount spent on phone credit was $FJ 10.77, of which a mean of $6.67 was personal spending, and $3.53 was credit sent or gi fted by friends or relatives. Figure 5 7 Money spent on phone credit during a typical wee k (error bar= +/ SE) It would be expected that members of households with high income, and in particular adult males who typically have a greater say over household finances, would spend the most on phone credit. However, the survey data indicate that bo th men and women claim to spend about $FJ 10 per week on phone credit, of which about 2/3 is their own money, and about 1/3 consists of credit sent by relatives or friends. Reported spending was lowest among divorced individuals, largely because they repor ted receiving less transfers of credit from friends or famil y. A ge and spending on phone credit are not significantly correlated: r (170) = .023, p= .768. T here are small, non significant differences in mean reported spending on phone credit among the thr ee villages. Residents of Toba claim to spend the most, but they also receive the most transferred credit from friends and relatives. Veidogo residents

PAGE 210

210 spend more of their own money on phone credit than residents of Toba or Cagi but receive less credit fr om friends or relatives. These figures make sense from the standpoint of the relative economic situations of the three villages Tob ans rely more on aid from urban kin, while Veidogo a ns, many of whom have regular jobs in Suva, are the most financially self sufficient. Figure 5 8. Money spent on phone credit during a typical week (villages) In retrospect it seems apparent that many survey responses about phone credit expenditures were exaggerated. An expenditure of $10/week would equal $520 per year, ne arly 10% of the mean annual household income in the study communities (and that would be for just for one mobile phone many households have multiple phones!). As another means of cross checking the accuracy of survey responses, I asked Vilivo, the main s eller of phone credit in Cagi how much credit he sells to villagers during a typical week. Vilivo est imated weekly sales of about $FJ 100, and that the amount of credit sold in any week has never exceeded $FJ 300. Divided among the nearly 100 mobile users in Cagi village, it would seem that in Cagi at least, actual mean individual expenditures on phone credit would be closer to $1/week, than the $6.67

PAGE 211

211 indicated by the survey data. It is also possible that respondents systematically overestimated their pers onal expenditures, while perhaps underestimating the amount of credit transferred by friends and relatives. Whatever the case, although inaccurate in an absolute sense, the data on phone credit expenditures may still be useful in terms of comparing the re lative expenditures of different subgroups and villages. The overall pattern shown by the data is that phone credit expenditures among individuals in different villages or age/gender/marital subgroups are remarkably similar. Number of C onta cts Saved in an hone frequency of phone use. Countervailing features of Fijian social structure make predicting the size of calling network s difficult Because of virilocal postmarital residence, Fijian men tend to have more extensive local kinship ties than do women who have married into a village. Fijian m en also te nd to have more spatial mobility than women do and may interact with a broader variety of individuals in the course of their daily work than do women On the other hand, married women have close ties to both their natal kin as wel l a kin group, which creates th e potential for women to have large, di spersed calling network s V ariation in calling network size t hroughout the life cycle is also difficult to predict on the basis of the discussion of Fijian social relations in Chapter 3 Fijian y ouths do not tend to h ave very deep relationships with distant kin, but they do active ly socialize with youths from other villages, and often save the phone numbers of new ly met friends. Older people, especially those with grown children, have extensive and growing

PAGE 212

212 network s of affines nieces and nephews, and grandchildren Some married adults also maintain business, church, or other institutional contacts. From the outset then, sex, age and marital status calling netwo rk size. Figure 5 9 Scatterplot of number of saved phone contacts, by sex and age Figure 5 1 0 Mean number of saved phone contacts by subgroup (error bars= +/ SE)

PAGE 213

213 The number of saved phone contacts found in village phones ranged from 4 to 318. T he overall distribution of the number of phone contacts saved in phones is positively skewed most individuals do not have more than 50 contacts, while a few have far more contacts. contacts in their phones. The overall age/co ntacts distribution (Figure 5 9 ) is roughly parabolic very young and very old users tend to have fewer contacts, and those in the 20 60 age range tend to have more. However, the general pattern seems to be that people gain contacts as they get older, get married and start a family. After the age of about 50 the number of contacts tends to decline, perhaps as peers and relatives die, and the individual becomes less mobile and socially active. The mean number of contacts is relatively similar among men (66 contacts) and women (57) This similarity between the sexes in Fiji is noteworthy when compared to s in Jamaica, where a s ample of 25 users had a mean of 127 contacts for men, and o nly 74 for women. Horst and Miller speculated that Jamaican women tended to have smaller address books than men, and a greater Meanwhile in Norway Ling found that women dominate phone use because women play a primary role in maintaining family related social networks (1998). Compared with age and sex, t here are greater differences in number of saved contacts among marital subgroups in Fiji. D ivorced villagers had the hig hest mean number of contacts (78 contacts) while widowed individuals had the least, with less than half as many contacts on average as divorced individuals (only 38 contacts) In

PAGE 214

214 retrospect it makes sense for divorced Fijians to have many phone contacts because they occupy an ambiguous social position that allows them to mingle within both the circles of the youths and the married adults. Divorced individuals have the extensive kin networks acquir ed with adulthood and marriage, but they may also freely us e their phones as a means of socializing with and courting members of the opposite sex. Overall the data indicate that in size. At the level of age/gender/marital subgroups or villages, however, the mean networ k sizes are quite similar. Number of aved in r in their phones. The advantage of this measure is that it less heavily affected by individual istic differences in habits regarding t he saving of phone contacts ( see Appendix C ) A major limitation of this measure however, is that it does count the number of people living outside the village that communicate with the villager in q uestion. In other words, a villager might rarely call her fellow villagers over the phone but might communicate very actively with people outside the village. T hi s measure of calling frequency/intensity cannot detect such village external relationships, only phone ties within the village There is no clear correlation ( r= 0.008) between age and the number of other Once again there is a roughly parabolic shape to the age distribution, with very young and old phone owners featured in fewer other telephones.

PAGE 215

215 Figure 5 11 Scatterplot of percentage of other villagers within which each phone was saved Men on average are saved in slightly more other village phones than women: each male phone owner was saved in an average of 13.4 other village phones, and each female phone owner only 12.0 other village phones. The average male individual in m y sample has a 20.8% chance of being saved in each other the average female has only a 14.7% chance. However, there are several females who also rank highly on this measure. Indeed, Figure 5 11 exaggerates the preponderance of men found in village phone contact lists. This is because s everal of the men featured in the highest percen tage of other village phones were from Toba which is such a small village that being written in a high percentage of the village phones is more lik ely t han in a large village The overall pattern is more or less balanced between men and women In terms of age, the usual parabolic pattern is observ ed: very young and elderly phone owners

PAGE 216

216 tend to be found in less phones, while the most active ph one users ar e found in the 20 55 age range. phone contacts) described in the previous sub section. Ind ividuals Identified as Heavy P hone U sers in S urveys I asked telephone owners to name four people in their community (other than themselves) who talk on the phone a lot. Many respondents found this question humorous, and gleefully rattled off the names of several chatty neighbors. Althou gh the level of consensus varied in e ach com munity there were a few individuals widely thought to talk on the phone a lot. The top 3 vote recipients in Toba received 43% of the total votes, in Cagi the top 3 received 37% of the votes, and in Veidogo the top thr ee received 25% of the total votes. Meanwhile t he pr oportion of owner s named at least once varied widely: in Toba 67%, Veidogo 57%, and in Cagi only by at least one survey respondent as a heavy phone user. It is diffi cult to associate the perceived heaviest phone use rs with any particular demographic or social category. To give an idea of the variety of individuals named as heavy phone users the 3 individuals in Toba married men. The top 3 in Cag i were all married women. The top three in Veidogo included a married man, a married woman, and a widowed woman. In each of the study communities, then, there are certain individuals widely perceived to be heavy phone users. H owever these heavy phone users do not neatly fit any specific profile; they may be male or female, younger or older.

PAGE 217

217 Relationships A mong Calling Frequency V ariables Each of the five variables I have just discussed captures some aspect of phone use frequency in the study villages. Howev er, it remains to be seen whether the survey data for the five variables paint a consistent overall picture Logically, high scores on one variable should be predictive of high scores on the other four variable s As i t turns out each pair of phone frequen cy/use variables does have a significant positive correlation, with the exception of only one variable pair : daily c alls/texts and # of votes as a frequent phone user That correlation is positive, but lacks s ignificance at p < .05. Table 5 1 Correlatio ns between phone use frequency variables Variable 1 Variable 2 N Correlation Sig nificance # of daily calls+texts Amount spent on phone credit in a typical week 159 +.183 .021 # of daily calls+texts # of phone contacts 170 +.169 .033 # of daily calls +texts # of phones saved within 159 +.316 .000 # of daily calls+texts # of votes as a frequent phone user 159 +.114 .152 Amount spent on phone credit in a typical week # of phone contacts 170 +.202 .008 Amount spent on phone credit in a typical week # of phones saved within 170 +.201 .009 Amount spent on phone credit in a typical week # of votes as a frequent phone user 170 +.172 .025 # of phone contacts # of phones saved within 170 +.490 .000 # of phone contacts # of votes as a frequent ph one user 170 +.340 .000 # of phones saved within # of votes as a frequent phone user 170 +.355 .000 Relationships between Demographic Subgroups and Calling Fr equency V ariables There seem to be few meaningful differences between socio demographic su bgroups within the study communities in terms of phone use frequency/intensity Among the major demographic variables ( village, age, sex, marital status, ethnicity and occupation) and the phone use frequency /intensity variables, the only significant relat i onship is the negative correlation between respondent age and the reported number of calls and texts sent in a typical day. This relationship as mentioned before, is

PAGE 218

218 largely due to the fact that young people text frequently (especially between lovers), whi le many older Fijians have never learned how to text. Table 5 2. Subgroups and phone use frequency/intensity variables Variable 1 Variable 2 N Test df Correlation Sig nificance Age $ spent on phone credit 170 Correlation .023 .768 Age # of calls+texts/ day 159 Correlation .223 .005 Age # of saved contacts 170 Correlation .134 .082 Sex $ spent on phone credit 170 Median test 1 .920 Sex # of calls+texts/day 159 Median test 1 .641 Sex # of saved contacts 170 Median test 1 .501 Marital status $ spent on phone credit 169 Median test 1 .662 Marital status # of calls+texts/day 158 Median test 1 .092 Marital status # of saved contacts 169 Median test 1 .387 Village $ spent on phone credit 170 Median test 2 .222 Village # of calls+texts/day 15 9 Median test 2 .854 Village # of saved contacts 170 Median test 2 .403 Ethnicity $ spent on phone credit 170 Median test 1 .560 Ethnicity # of calls+texts/day 159 Median test 1 .099 Ethnicity # of saved contacts 170 Median test 1 .353 Occupation $ spent on phone credit 170 Median test 3 .226 Occupation # of calls+texts/day 159 Median test 3 .127 Occupation # of saved contacts 170 Median test 3 .058 Mood the null hypothesis, H 0 that the there is no difference between the medians of the samples. I used this non parametric test because my phone use frequency variables are non normally distributed and could not be successfully relatively robust to outliers, of which there a re several for each of my variable s ass umes independent random samples, and that population distributions have the same shape. My distributions, which are all positively skewed conform to these assumptions Note: Due to small samp le

PAGE 219

219 Ranking Individuals Based on a Composite Measure of Phone Use Frequency/I ntensity I have so far discussed five dif ferent variables related to an overarchin g concept: frequency/intensity of phone use. of these five variables equally, and rank all the individuals in each community according to their score on this composite measure. The goal is to derive a we ll rounded picture of phone use frequency/intensity, and to see if there are any meaningful patterns according to which individuals of differening demographic characteristics use their phones. In the following scatter plots the highest data points (in the vertical sense) are the individuals with the greatest overall intensity/frequency of use within their communities. Figure 5 12. Composite ranks for phone use frequency Toba seem to be meaningful patterns in these data. Although the top ranked individuals in In fact, the mean rank for men (26) is only slightly higher than the mean rank for women (29). In terms of age, the youngest and oldest phone users tend to have lower

PAGE 220

220 composite ranks, whereas the highest ranked phone users are distributed among users 50. Figure 5 13. Composite ranks for phone use frequency Cagi Figure 5 14. Composite ranks for phone use frequency Veidogo Based on the lac k of clear patterns in the data, I conclude that differences in phone use frequency/intensity are primarily explained by individual difference s in personality, family history or economic factors rather than socio demographic factors

PAGE 221

221 Characteristics of Phone C ontacts In this section I analyze the characteristics of the contacts saved in phone The data permit analys is of the extent to which phone communication is bounded within, or permeates across the social divisions of age, sex, kinship, rank, ethnicity, and occupation. These data also allow me to determine what sorts of individuals play the most import ant roles within their roles. Finally, we can use these data to compare the calling network s of Toba Cagi and Veidogo as a way of exploring how community level geographic, te chnological, or social structural differences influence patterns of communication. To obtain my data, I copied the contacts saved in each address book, and then asked the a series of questions about each saved contact (see Appendix C for a more detailed explanation) minimum of 4 and a maximum of 318 unique saved contacts, with a mean of 62 saved contacts. I asked about each phone location, and ethnicity. I then asked relationship specific que stions about the nature of the tie and frequency of phone based and face to face interaction between the phone owner and the contact An Overview of the Phone Contacts D ata The resulting data can be analyze d using conventional statistical methods, or analyzed a s a sociogram using methods from social network analysis From the social network approa ch, e ach phone number represents an individual, depicted as a node in the social network. I copied phone numbers belonging to 4649 unique individuals from the 170 phones I surveyed. Phone numbers can also be aggregated and analyzed at

PAGE 222

222 the household level, but I do not do so in this chapter. Each time I found a particular number in a particular phone, then that pair of numbers represents a relationship between the two phone owners, depicted as a tie between the two nodes. After cleaning up my data and removing invalid phone numbers, I have a total of 10,537 of these ties. Figure 5 15 : Characteristics of all phone contacts and ties. This graph presents the characteristics of both nodes (individuals) and ties (relationships) for comparative purposes. The blue distribution (in the background) summarizes the characteristics of all 10,537 ties. The red distribution (clo ser to the reader) represents only the 4649 unique individuals (nodes) found as contacts in village phones. In Figure 5 15 if the red bar for a given characteristic is shorter than the blue bar, that means that individuals with that characteristic are f ones.

PAGE 223

223 Thus, there is a high absolute number of phone based relationships involving married people (tall blue bar), and fewer phone based relationships involving divorced people (short blue bar). However on average each divorced individual was found in m ore village phones than each married individual We know this because of the relative sizes of the red and blue bars for each characteristic. Figure 5 16 displays the average number of ties for a node with each characteristic. This is important because it indicates which kinds of individuals are more networks. (average individual saved in 12.5 p hones), being a native of the same village as the phone owner (regardless of current residence) (5.2 phones), and divorced marital status (6.7 phones). Figure 5 16 : Characteristics of individuals found in phones

PAGE 224

224 Figure 5 17 : Characteristics of phon e based relationships Figure 5 17 displays the characteristics of the 10,537 ties (phone based relationships) for which I have data. A couple of notable features that stand out: 74% of all phone ties are between kinsmen, and 59% of contacts meet face to f ace once a year or less. Most of the remaining contacts meet daily, and are likely residents of the same village. This indicates that rural Fijian phone communication is, in a sense, bifurcated : most calls are between kin who live in distant areas of Fiji, and usually meet each other only once a year. The other major category of phone based relationships is between residents of the same village, who see each other every day. A key question is whether the members of the various socio demographic subgroups in rural Fiji vary in terms of the composition of their calling networks. I used chi square tests to determine whether there were significant differences in the composition of contact lists based on sex, marital status, residence location, ethnicity and occu pation.

PAGE 225

225 In fact, there are statistically significant differences (p<.05) in the distribution of phone contacts for all variables in Table 5 3. However, the effect sizes are mostly small. Only a few pairs of variables displayed a medium effect size, in ter ms of differences in the composition of contact lists, and none displayed a large effect size. Table 5 3 Analysis of differences in calling network composition for individuals in various social subgroups Owner variable Contact variable Df Valid N Pears on Chi square value p (2 tail) Phi V Effect size Sex Sex 2 10407 103.782 .000 .100 .100 small Sex Marital 3 10066 8.796 .032 .03 .03 small Sex Location 3 10394 55.185 .000 .073 .073 small Sex Ethnicity 3 10087 41.399 .000 .064 .064 small Sex Occupation 3 9586 95.978 .000 .100 .100 small Sex Relationship 2 9963 132.228 .000 .115 .115 small Sex Kin type 6 9941 576.336 .000 .241 .241 medium Sex Call Frequency 3 9876 66.371 .000 .082 .082 small Sex Meet Frequency 3 9939 59.839 .000 .078 .078 small Marital Sex 6 10407 45.174 .000 .066 .047 small Marital Marital 6 10066 263.129 .000 .162 .093 small Marital Location 9 10394 21.068 .012 .045 .026 small Marital Ethnicity 9 10087 36.514 .000 .060 .035 small Marital Occupation 9 9586 174.980 .00 0 .135 .078 small Marital Relationship 6 9963 350.041 .000 .187 .133 small Marital Kin type 18 9941 241.813 .000 .156 .09 small Marital Call Frequency 9 9876 464.248 .000 .217 .125 medium Marital Meet Frequency 9 9939 99.482 .000 .100 .058 small Locat ion Sex 4 10407 33.687 .000 .057 .040 small Location Marital 6 10066 76.673 .000 .087 .062 small Location Location 6 10394 401.924 .000 .197 .139 medium Location Ethnicity 6 10087 1043.025 .000 .322 .227 medium Location Occupation 6 9586 228.520 .000 154 .109 small Location Relationship 4 9963 386.397 .000 .197 .139 small Location Kin type 12 9941 428.530 .000 .208 .147 medium Location Call Frequency 6 9876 4447.283 .000 .213 .150 medium Location Meet Frequency 6 9939 981.131 .000 .314 .222 medium Ethnicity Sex 4 10407 28.977 .000 .053 .037 small Ethnicity Marital 6 10066 32.379 .000 .057 .040 s mall Ethnicity Location 6 10394 35.448 .000 .058 .041 s mall Ethnicity Ethnicity 6 10087 211.131 .000 .145 .102 s mall

PAGE 226

226 Table 5 3. Continued Owner variabl e Contact variable Df Valid N Pearson Chi square value p (2 tail) Phi V Effect size Ethnicity Occupation 6 9586 38.535 .000 .063 .045 small Ethnicity Relationship 4 9963 35.636 .000 .060 .042 small Ethnicity Kin type 12 9941 261.857 .000 .162 .115 small Ethnicity Call Frequency 6 9876 108.308 .000 .105 .074 small Ethnicity Meet Frequency 6 9939 110.067 .000 .105 .074 small Occupation Sex 6 10407 207.655 .000 .141 .100 small Occupation Marital 9 10066 170.420 .000 .130 .075 small Occupation Location 9 10394 80.384 .000 .088 .051 small Occupation Ethnicity 9 10087 253.066 .000 .158 .091 small Occupation Occupation 9 9586 544.948 .000 .238 .138 medium Occupation Relationship 6 9963 602.913 .000 .246 .174 medium Occupation Kin type 18 9941 782.565 .000 .281 .162 medium Occupation Call Frequency 9 9876 241.195 .000 .156 .090 small Occupation Meet Frequency 9 9939 345.382 .000 .186 .108 small similar ip between two variables. Cohen established widely used effect size standards: .10= small effect, .30= V, the thresholds decrease somewhat for large co ntingency tables. Calling network composition differences with m edium, rather than small, effect sizes are found between members of different communities, and between surveyed villagers with different occupations. In other words, a greater amount of the v ariation in can be explained by the village in which they live ( Toba Cagi or Veidogo occupation s rather than the demographic characteristics (age/sex/marital status) This wou ld indicate that socio economic conditions of the community, as well as household to household socioeoconomic differences have a larger influence on the contents of mobile social networks than do intra community demographic divisions (age, sex, marital sta tus). Sex Based on my description of Fijian com munication patterns in Chapter 3 it would be e xpected that men and women communic ate over the phone primarily with other

PAGE 227

227 members of the same sex. After all, extensive private socializing between men and wom en is unusual in rural Fiji In fact it might be considered improper for a married individual to socialize frequently with an opposite sex peer who is not close kin Figure 5 18 Sex of phone owners and phone contacts The data indicate that mobile social networks are actually not segregated to a considerable degree by sex. Men do have significantly more male 2 (1, N=10104) =59.99, p=.001 However, in practical terms the difference is not large: 57% % of fem The preponderance o f male phone contacts among phone owners of both sexes is p ossibly due to the fact that males own t he majority of phones in Fiji. However, I do not have country wide data to confirm this suspicion Among both male and female phone owners, advancing age is positively associated with a higher proportion of male phone contacts. This may be a generational phenomenon, rather than something that changes throughout the lifespan. After all, it would be expected that among older Fijians there is a disproportionatel y high rate of

PAGE 228

228 male phone ownership, while phone ownership is more gender balanced among younger Fijians. This is both because of a progressive relaxing of gender/sex norms in recent decades in Fiji, and because as phones become less scarce, women and othe r subordinate groups gain increasing access. Figure 5 19 Age/sex of phone owner vs. sex of phone contacts Figure 5 20. Calling frequency between same sex and opposite sex individuals

PAGE 229

229 Reported calling frequency between the sexes was somewhat simila r to frequency of calling to same calling frequency. The most visible difference is that females were about 10% more likely to respond that they call a given man on a daily basis, than for a man to re port calling a woman on a daily basis (Figure 5 20 ) Overall though men and women do not differ significantly on any of the phone us e frequency/in tensity variables (see Table D 1 in Appendix D ). This is interesting given that women are perceived by many v illagers as talkin g more on the phone than men do. One middle aged widow, Joana, speculated that women use the phone more heavily than men because when grown children call home, they want to talk to their mother, not their fat her ( Ni sa qiri mai na gone, s a via vosa ga vei Na ). This fits with Fijian cultural concepts of the father as a stern disciplinarian, and the mother as more of a nurturing The father, just goes around the village and drinks lots of kava; he family life ( Tata, sa levu na nona lako vaqo, vaqo, gunu yaqona sega ni kila na bula ni vuvale ) Another respondent speculated that because men are typically away from the home, in the gardens, and if the phone is left in the house during the day it is usu ally the woman who answers it. Marital S tatus I did not ask respondents to estimate the ages of their phone contacts. It seemed that such a difficult question would yield inconsistent and inaccurate data. Instead, I asked respondents to report the mar ital status of contacts, which is a much more straightforward question to answer.

PAGE 230

230 Figure 5 21 Marital status of phone owners and contacts Overall, about 2/3 of all phone contacts were currently married or had been married at some point in the past and about 1/3 were single. Divorced and widowed contacts made up only a small percentage of the entire sample, but are likely under reported because I suspect that respondents often simply There are sign ificant differences in calling network composition related to the marital status of owners and contacts; however the effect size is small. Table 5 4 Chi s quare test for differences in calling network composition for groups of different marital status Ch i square statistic N Df p Phi 263.129* (1 cell had expected count <5 ) 10066 9 .000 .162 .093 Married and widowed phone owners had a higher proportion of married phone contacts saved in their phones, while single and divorced phone owners had a higher proportion of single contacts saved in their phones. This makes sense, since single phone owners are predominantly youths, who socialize extensively with other youths. The fact that divorced and single individuals frequently contact each other on the phone

PAGE 231

231 is also expected, given the important role that phones play in meeting and courting unmarried members of the opposite sex. It is even probable that some of the e xtramarital romance ; however, I could not think of a graceful way to ask such a question in a survey. Figure 5 22 Age of phone owner vs. marital status of phone contacts This scatter plot reinforces the conclusion drawn from the preceding bar graph t hat young rural Fijians tend to talk over the phone with other young (unmarried) people, while older rural Fijians tend to talk more with older (married) people. There seems to be a slight represent the addition of unmarried grandchildren nieces and nephews to the calling networks of aging Fijian grandparents

PAGE 232

232 Ethnicity/Native Village I asked phone owners about the ethnicity/native village for each of their phone contacts. These data indica te the degree to which communication is restricted within the network of individuals belonging to a particular village These data are also interesting because they indicate the extent to which rural indigenous Fijians interact with members ethnic groups (Indians, Chinese, Europeans, etc.) Figure 5 23 Ethnicity of phone contacts by village Among indigenous Fijian phone contacts, t he calling networks of residents of the more remote rural villages ( Toba and Cagi ) have a higher percentage of contacts native to the same village (whether or not these contacts actually reside d within the village). This indicates a lower degree of interaction with people from other regions of Fiji, which is to be expected in remote rural areas. Veidogo residen ts, by contrast, called natives of a greater variety of communities and regions. ethnically ho mogeneous. Overall, 97% of contacts were indigenous Fijians. Such

PAGE 233

233 homoge neity might be expected in Toba and Cagi where opportunities to interact with people of other ethnic groups are limited. On the other hand, most Veidogo a ns interact with Indians, Chinese, and members of other ethnic groups on a daily basis in Suv a. Despit e this, only 6% of the phone contacts in Veidogo phones are non indigenous Fijians. Among non Fijian contacts in the three communities, 41% were business relationships, 54% friends, and only 16% kin. These data underscore a few general truths about Fijian society. In remote rural areas, contact between indigenous Fijians and members of other ethnic groups tends to be very limited. Many of the non Fijian contacts in Toba and Cagi phones were Indian school teachers, nurses, and other civil servants posted to rural areas for their jobs Inter ethnic interaction in urban and peri urban areas is much more frequent, but is often superficial and work or business related. Overall these data emphasize the fact that social life for most Fijians remains centered a round kinship. Most of the indigenous Fijian contacts, even those not native to the same village, are in fact kinsmen and kinswomen related to the phone owners through marriage. Miscegenation among Indians and Fijians remains uncommon and as such interact ion with members of ethnic minorities remains limited to business and casual socializing. Occupation Job opportunities in Fiji are heavily influenced by whether one live s in a rural or urban/peri urban location. In rural Fijian villages, most household s d epend on farming for their livelihood, and with few exceptions, rural women are homemakers. In urban areas and peri urban villages such as Veidogo there are broader opportunities to acquire wage paying jobs, and both male and female villagers may work ful l or part

PAGE 234

234 time in the urban area while supplement ing their wages with a degree of subsistence gardening and fishing. My phone survey included the following options occupation: student, farmer, household, employed, or reti red. A reasonable prediction would be that formally employed individuals would be found in more phones on average, than unemployed individ uals, farmers or homemakers. The idea underlying this prediction is that employed people tend to have more economic r esources, which would a llow them to buy more phone credit and make more calls. A more secure financial status would also enable someone to financially assist kinsmen in need potentially making them a useful person to keep as a phone contact However, the data show that in fact among non village residents, employed individuals are found in fewer phones (on average) than are homemakers, farmers and retired people. Figure 5 24 Mean number of village phones within which contacts from different occupation al groups were found (non village dwellers only)

PAGE 235

235 Moveover, the average calling frequency is very similar for farmers, homemakers, employed workers and retired people. Students are the only occupation al category with a very unique profile for calling frequ ency This can be explained by the biological children attending boarding school in other parts of Fiji. This explains the high rate of between villagers and student s Figure 5 25 Calling frequency to contacts of different occupational statuses The data imply that occupational status plays a n egligible role in determining a non village dwelling ity within the community calling network. Inste ad, geographic location and kinship ties are more important in determining the centrality of an individual within the calling network, and the frequency with which a pair of individuals interact on the phone On the other hand, rural villager dwellers with important occupational positions (shopkeeper, truckdriver) tend to have central network positions. This is probably because they provide key services within the community, and are often contacted by fellow villagers for practical or business reasons.

PAGE 236

236 Rela tionships betw een Phone Owners and C ontacts For each saved contact, I asked phone owners whether they are related as lationship the phone owner and contact were related In Fiji there is some blurring of categories among kin, friends, and business partners In the Fijian classificatory system of reckoning kinship, even very distantly rel ated people can be considered kin, and often distant ly related kinsme n conduct business with each other Conversely, individuals with a close business relationship might begin to reckon one another as kin as a consequence of the bond they have formed. If t here was any doubt about a specific response, I asked the respondent to choose the relationship category that best described the most common type of interaction for the relationship in question. For examp le Ropate, the local taxi driver, might be stant cross cousin, b ut they primarily interact when Jone pays Ropate to drive him in the taxi The appropriate survey response should therefore have been business relationship Friends In Fiji, a friend is someone generally outside the sphere of close rel ations, with whom one socializes in an egalitarian manner, free of the formalities of rank and obligations of kinship. Friends therefore often come from disparate geographic areas, and are acquired in places in which people from different parts of Fiji min gle extensively, such as cities, urban workplaces, or boarding schools.

PAGE 237

237 Figure 5 26 Age of each phone owner and the percentage of their contacts who are related to them as friends (excluding owners with <10 contacts) Youths, more so than any other ca tegory of Fijians, use their phones predominantly to contact friends. Most secondary schoolers with phon es have only a small number of contacts in th eir a ddress books, of which many are friends they met at school or age peers living in nearby villages Th e relatively high proportion of friends in be attributed in part to the way Fijians evolve socially as they grow older Adult Fijians, who have extensive knowledge of local genealogies, can confidently reckon in which degree of kinship they stand, even with distantly related individuals from other villages. Youths, with only a hazy knowledge of local bloodlines and marriage histories often just fa because they are unsure in which ki n relation they stand Business relations account for a very sm all percentage of contact lists. However, males have more than

PAGE 238

238 twice as many of these contacts as women (3.7% for men vs. 1.7% for women). This reflects the fact that in rural Fiji men tend to hold most of the local leadership positions that involve interactions with urban based business, church, government and non governmental organizations. Many of the business con tacts belong to local transportation and government service providers. These include teachers, doctors and nurses, police posts, taxi and truck drivers, and produce buyers. Other business contacts are services located in central i zed urban areas (mostly Suva) including government offices banks, and B usiness contacts included radio stations (to participate in on air contests), or numbers that vil lager s had saved after listening to a TV or radio ads for health and cosmetic products, such as beauty creams or medicines. Kin Kinship is the principle around which Fijian communities are organized, and predominantly consist of inter actions with kinsmen As described in Chapter 3 in F iji there is a fundamental dist inction between parallel and cross kin. Because of complex marriage histories, there are usually multiple options for reckoning a particular kinship tie. Sahlins (1962) th eorized that in ambiguous cases, relationships than the more restrictive parallel relationships Arno (1992) noted a preference for drinking kava with cross relatives (rather than parallel) in or der to be able to engag e in open dialogue and joking. Sahlins further observed that the preponderance of cross ties could also be partly explained by intra community marriage s: Fijian spouses by definition adopt a cross cousin relationship. If parallel relatives marry, their rel ationship is converted to that of cross

PAGE 239

239 cousins relatives to similarly redefine their relationships from parallel to cross. Both Sahlins and Arno recorded the distribution of kinship categ ories with in the Lauan communities in which they worked. Sahlins (1962) and Arno (1993) asked male heads of household in 4 villages to list their kin relationships with other male heads of household. I did not exhaustively catalog all of the kin relationsh ips among community members in Toba Cagi and Veidogo in order to analyze local genealogies. However, my phone contacts survey yielded a large dataset that indicates the categories of kin with whom Fijian villagers socialize over the phone. The following c hart compares the proportions of different kin categories found among the phone contacts in Toba Cagi and Veidogo Figure 5 27 Proportions of kin types in Fijian communities. Frequency of incidence of different kin relati onships among male household heads in 4 Fijian villages compared to their incidence in my phone contacts data from Toba Cagi and Veidogo Arno asked 4 male heads of household to name their kin relationship with 85 other local heads of household in Yanuya nu In Moala, Lau, asked 16 male heads of household in Naroi village, 6 in Nuku village, and 12 in Keteira village about their relationships with other men in their communities.

PAGE 240

240 small, non probability sample s, their d ata do seem to confirm a systematic preference for cross ties over parallel ties. The proportions of kin relationships I found among Toba Cagi and Veidogo phone owners and phone contacts are in fact very similar to the proportions of kin relations that Sa hlins and Arno found in the communities in which they studied. In particular, in the 0 generation, there is a mark ed preponderance of cross ties over parallel ties. My data diverge somewhat in the +1/ 1 generation, in which my samp le featured more parallel than cross ties This opens up an interesting cross relatives called for social purposes, whereas phone contacts in ascending/descending generations are primarily f or parallel relatives called for family/kinship obligations within the des cent group In practical terms, this means that adult Fijians are more likely to save their nephews and nieces. Figure 5 28 Kin relationships of phone contacts

PAGE 241

241 village data would seem to indicate that 1) a tendency to ties, which Sahlins and Arno noted, appears also to be in effect in Toba Cagi and Veidogo and 2 ) interacting through the telephonic medium does not seem to make Fijians favor or disfa vor any particular kin category. I n other words, the proportions of different kin categories found in phon egocentric kindreds Fijians are not particularly biased toward or against talking with any kind of kin throug h the telephone, vis vis face to face interaction I was particularly interested i n how interactions mediated through the telephone might affect the expression of avoidance, joking, or respect that is required in face to face interactions in specific kinship degrees. Fijians often say that speaking face to face to cross uncles or sister s provokes a feeling of unease, shame or embarrassment ( madua ). I was interested in whether, by talking through the phone, these feelings of madua could be avoided and more open interaction would be possible. By and large, villagers claimed that they obser ve the same types of joking, avoidance or respect over t he phone as they do in person. Thus p hone conversations between cross cousins are typically jovial and jocular, while if a sister or cross uncle calls (kin categories which practice avoidance), then F ijians usually pass off the phone to someone else, so that they can avoid the embarrassment or awkwardness of talking to these k in. Paradoxically then, it is wise to save the numbers of sisters and cross uncles, simply so that if such a relative calls, one can recognize their number and pre emptively pass the phone off to someone else to do the talking!

PAGE 242

242 S uch avoidance is practiced by many villagers in both face to face and telephone interaction; however, several villagers emphasized that they feel more comf ortable most informants stressed, was that over the phone was does not have to look the uncl e/sister in the face. If this truly is the case, then increasing frequency of phone interaction in the future might contribute to a weakening of avoidance taboos and Fiji and other kinship comfortable communication over the phone, it could be a gateway for more regular face to face interaction s Figure 5 29 Calling frequencies between parallel and cross kin in the same generation Husbands and wives are technically considered cross cousins, but I intentionally omitted these pairs from this analysis because of their dispropo rtionately frequent communication. A demonstrated in Figure 5 29, there are evident differences in the average reported frequency of phone interaction between cross cousins and parallel cousins ( i.e., siblings ). Cross cousins are nearly twice as likely as parallel relatives to talk on the phone on a daily basis. This is despite the fact that parallel relatives ( veitacini ) are

PAGE 243

243 typically more closel y related in a biological sense. This would indicate that contacts who are called on a daily basis are primaril y contacted for so cial, rather than for instrumental purposes. On a side note, I was initially confused because the names of kin saved in Fijian phone address books are typically reckoned children rather than the p hone owner him/herself. For example, if a man with children saves the phone number of his own father, the name entered in the phone is not usually T a (Dad), but Tai (Grandpa). A few p hone owners indicated that this was to help children to learn about their generally use phones, though, reluctance to address or refer to people directly by their given name s Fijians most com monly address and refer to each o ther b y their kinship degree, i.e., tavale (cousin); or they use teknonymy: i.e., tinai Samu ( s mother) This is also true when they are talking about kinsmen in the 3 rd person. In daily interaction, Fijians often use to avoid directly addressing passersby. For example, when a man with a toddler in his a rms sees his cross cousin approaching on loudly say to the toddler (within earshot of the cousin) M omo (c ross uncle ) It may be that these respectful forms of indirect address are carried over into phone contact lists and phone conversation. An alternate explanation phones, perhaps because the parents do not know how. In summary, t hese observations are interesting from the standpoint of the internal dynamics of Fijian kinship, and how kinship communication adapts to the uptake of new

PAGE 244

244 technologies However, it may be difficult to make valid cross cultural com parisons regarding kinship and phone use For example, Horst and Miller studied the contents of address books finding an average of 95 contacts, o f which an average of only 13 contacts (14%) were kin I n Jamaica, formal marriages and two parent households are rare, and relationships between baby mothers and baby fathers are fluid. These adaptable social relations encourage Jamaicans to form broad social network s in which formal rules of kinship are not very meaningful (2007:91) Despite the incompatibility of Fijian and Jamaican kin categories, many of these Jamaican relationships probably functioned similarly in a social, economic, an d psychological sense, to Fijian kinsmen. Temp oral and Spatial Aspec ts of Phone C ommunication A primary impact of mobile phones is their ability to soften the spatial and temporal restrictions on social interactions. As such it is interesting to analyze to what extent people use their phones for communication at long dista nces, vs. short distances, and how these interactions relate to the frequency of face to face interactions between individuals. In this, the geographic situation of villages as well as available transport options may play a determinative role. Veidogo and Toba have fairly similar distributions of ge ographic locations of contacts. Cagi mainly from the other two villages in that 30% of Cagi phone contacts actually reside in the village itself (compared to less than 15% each in Toba and Veidogo ). A possible explanation is that Cagi is a larger village, with roughly three times as many inhabit ants as either Toba or Veidogo Another contributing factor may be that nearly everyone in Cagi uses the same network (Digicel), facilitating cheap i ntra village calling.

PAGE 245

245 Figure 5 30. Geographic location of phone contacts by village The majority of contacts found in Toba Cagi and Veidogo reside in other areas of Fiji, outside the local area. The relative proportions must be treated cautiously however for each village I had to define the local area based on unique criteria. For example, for Batiki and Cagi the local area comprised their respective islands which were very different in size ( Cagi is on Koro island, which has 14 villages, while Toba is on Batiki, with only 4 villages ) areas, especially in and around the capital Suva. It would have been interesti ng to distinguish the location of phone contacts as either rural or urban. However, the names of cities such as Suva and Lautoka are often applied loosely by villagers, towns a re surrounded by ambiguous peri urban areas. Thus it would have been difficult for survey respondents to differentiate between urban and rural dwelling contacts. Less than 3% of contacts in each village reside overseas. The paucity of overseas contacts indicates the fact that relatively few rural based Fijians migrate

PAGE 246

246 overs eas, but also the fact that villagers almost never call overseas, instead simply waiting for their overseas kin to call them (due to high international calling costs). Therefore some people occasionally receive calls from overseas kin, but simply do not sa ve the number in their phone because t back. Figure 5 31 Mean number of in ties per node by geographic location Geographic t (i.e., the number of village phones that contact is found within). Contacts living in the local area, in m ore distant parts of Fiji, and o verseas are all found in approximately two phones in each of the study villages. The only major difference is that i ndividuals residing in the same village are found in many phones: Cagi contacts on average a re found in 13.7 other Cagi phones, vs. 5. 4 phones for Toba and 6.5 for Veidogo This is once again likely due to village size; Cagi simply has many more resident p hone owners who are likely to save a contact info. This effect is felt much less strongly for contacts living outside the village, because individiduals, even within the same village, tend to have different circles of kin and friends.

PAGE 247

247 Ne vertheless, Cagi village dwelling contacts are still found in slightly more phones than Toba or Veidogo contacts. Figure 5 32 Location of phone contacts vs. calling frequency There are noteworthy differences in responses about calling frequency with contacts located in different geographic areas. There is a general pattern of calling people who live nearby more frequently than those who live farther away. In particular, for contacts that live in the same village as the phone owner, nearly 40% ar e called on a daily basis (according to survey responses) These varying periodicities of communication may serve very different purposes. In interviews, respondents stated that calls to co villagers, though more frequent, are often brief and involve mun baby this afternoon while I made on regular occas ions such as birthdays or holidays. These calls involve exchanges of family news and expressions of sentiment that reinforce kinship ties, and

PAGE 248

248 may also involve requests for assistance, or cooperative planning for upcoming family obligations such as wedding s and funerals. Toba daily weekly monthly yearly Total daily 1% 0% 2% 6% 9% weekly 4% 1% 13% 6% 24% monthly 12% 1% 10% 36% 59% yearly 1% 0% 0% 7% 9% Total 18% 2% 25% 55% 100% Cagi daily weekly monthly yearly Total daily 9% 1% 2% 15% 26% week ly 9% 3% 5% 11% 28% monthly 7% 1% 6% 25% 39% yearly 1% 0% 0% 5% 6% Total 25% 5% 13% 56% 100% Veidogo daily weekly monthly yearly Total daily 6% 1% 2% 1% 9% weekly 10% 7% 7% 2% 26% monthly 11% 4% 26% 10% 51% yearly 1% 0% 1% 12% 14% Total 28% 13% 35% 25% 100% Figure 5 33 Calling frequency vs. meeting frequency Figure 5 33 portrays the relationship between the frequency with which phone owners call the individuals on their phone contacts lists (y axis), and the frequency with wh ich t hey meet the same individuals (x axi s). Cells are shaded darker, when a higher percentage of con tacts falls within that category. A few broad patterns stand out. In Toba and Cagi the two outer island villages, the majority of phone contacts are only seen face to face on a yearly basis. Most of these annually meeting contacts are called on a monthly basis. In Veidogo which has a more central peri urban location, there is a more even balance among contacts met on

PAGE 249

249 a daily, weekly, monthly and yearly basis. Communities which are close to an urban hub, and which have good transportation options, tend to have more frequent face to face contact with kin and friends living in other communities. This probably is partly because relatives from distant parts of Fiji combine trips to Suva with visits to their peri urban kin, and indeed on such trips visiting kin smen may stay as household guests of their periurban kin for days or even weeks. For many rural villages this makes sense one tends to meet co villagers everyday, but those from other villages, even nearby villages, are rarely met as regularly as once a week. By the same logic few phone contacts are called on only a yearly basis it seems that most people only enter a contact if they call the person at least monthly. Figure 5 34 Percentage of contacts living in same village as phone owner One might expect that as an individual ages, his or her mobility patterns and proportion of local vs. distant social relationships might evolve. The percentage of

PAGE 250

250 phone contacts who are co villagers is one way to gauge these changes. There is a significant negative correlation between age and percentage of phone contacts that live in the same village (N= 170, .196, p=.010). This seems to indicate that as Fijians get older, they accumulate more geographically distant kin and friends; marriage to a native of a distant area of Fiji would be one major cause. I t migh t also mean that Fijians simply become less interested in calling their fellow villagers as they get older Overall, these spatio temporal data point to a situation in which phone communication serves diverse social purposes. On the one hand, monthly phone calls serve to sustain relatio nships with kin who meet perhaps only once a year at Christmas. On the other hand, phones are useful for making local calls about everyday, mundane needs and social events. These calls tend to be more frequent, and may occur even between residents of neigh boring houses. Comparison of C ommunity Calling N etworks The largest unit of analysis in my study is the village community itself the calling networks for Toba Cagi and Veidogo can be compared and contrasted to analyze how geographic, economic and socia l factors shape group level communication. In social network analysis there is a basic distinction between ego networks and complete networks. However, my data are neither ego networks nor complete networks: rather, they are a sort of hybrid between the tw o The reason that my data fall short of being complete networks is that for logistical reasons, I only surveyed phone owners resident in the three study villages, and was unable to survey anyone living outside the villages. The average village resident h as 62 ties. However, for individuals who live outside the village, the only ties I know about are those to village residents. On average, each non village resident in my networks has

PAGE 251

251 only 1.78 ties to village residents No doubt these non village residents have many more phone ties to other people living outside the study communities, but I do not possess these data. Social Network Analysts have developed a variety of measures to describe both the overall nature of a network, and the relative connectedness of the node s and subparts (components) of the network. For example, centrality measures quantify each the concepts of degree, closeness, and betweenness. Degree is the id ea that actors with many connections have many options for actors to contact, assist, or influence (and in turn may be contacted, assisted or influenced by many other actors). Closeness is more potent and allow the actor to exert more leverage. Betweenness is the idea that interactions of others in the network. These concepts transfer to a tel ephone network, in that the direct phone contact with another individual may reinforce social relationships, serve as a source of useful information, and a potential source of aid or social/economic resources when they are needed. All three of my village n etworks consist of a single component; in other words, every participant in the network is connected to every other participant, however distantly. In terms of comparing the structural properties of the three networks, I am limited by the fact that these a the networks are of very different sizes Cagi

PAGE 252

252 nodes and seven times as many ties as Toba Still, some basic network meas ures can be presented (Table 5 5 ). Table 5 5 Characteristics of community calling networks Village Toba Cagi Veidogo Total nodes 593 2416 1701 Total ties (calling ties) 1005 6800 2731 Avg Degree (village residents) 62.13 67.18 51.53 Avg Degree (non village residents) 1.53 2.10 1.41 Avg Degree Overall ( calculated by UCINET) 3.275 5.288 3.135 Principal Eigenvalue (from centrality analysis) 20.39 45.39 24.08 H Index 14 55.000 33.000 Density 0.006 0.002 0.002 Components 1 1 1 Component Ratio 0 0 0 Connectedness 1 1 1 Fragmentatio n 0 0 0 Closure 0.037 0.095 0.043 Avg Distance 2.774 3.307 3.382 SD Distance 0.6 0.683 0.652 Diameter 6 6 5 Breadth 0.62 0.682 0.691 Compactness 0.38 0.318 0.309 Measures derived from UCINET analysis: Network >Multiple Measures >Network Level (eg Co hesion)*measures for whole calling networks; unweighted, reciprocal (undirected) ties Another way to compare the community networks is to determine what proportion indicat e the degree of community cohesiveness ; the extent to which kin/social relations are shared among multiple community members The slope of the line in Figure 5 35 is one For example, in Toba a g reater proportion of the total contacts are found in a greater percentage of the Toba share a greater Cagi or Veidogo do.

PAGE 253

253 Figure 5 35 Scree plot of proportion of phone contacts saved in proportion of each Table 5 6 Aspects of community networks Village Mean # ties/conta ct (overall) Mean#ties/conta ct (non village resident contacts only) Likeli hood that any given contact will be found in any given phone Proportion of in village nodes to total nodes Percentage of all ties that are in village ties Out village nodes per in village node Toba 1.715017 1.536673 0.096042 4.6% 15% 20.96 Cagi 2.818069 2.103799 0.02083 0 6.3% 30% 15.00 Veidogo 1.632019 1.435025 0.027076 3.5% 15% 27.83 wise, has twice as many in 6). Also, for every in village node in Cagi there are only 15 out village nodes, compared to 20.96 for Toba and 27.83. The overall picture that emerges is that there is extensive intra village calling in Cagi, compared to the other two villages. In addition, members of the Cagi calling network who reside outside the comm unity tend

PAGE 254

254 to be connected to more village residents than in either Veidogo or Toba again, likely a function of village size. Summary In this chapter I presented data regarding phone use patterns in Toba Cagi and Veidogo While the sheer quantity of da ta can be overwhelming, we can derive some broad conclusions about the dynamics of communication in these communities Community L evel D ifferences Toba, Cagi and Veidogo have very different endowments of infrastructure. In particular, the reliability of t ransport, electricity and phone network service in the communities varies. B etter electric/ mobile network access and higher mean household income help to explain the higher rates of phone ownership in Veidogo and Cagi However, the overall frequency/intens ity of phone use among phone owners in the three communities is very similar this includes the average number of phone contacts, frequency of texting/calling, and amount of money spent on phone credit. The greatest community level differences in the calli ng networks of the three villages seem to derive from village size. In a large village such as Cagi there is extensive intra village calling while in smaller villages contact between villagers is mostly face to face. Furthermore, Cagi individuals have lar ger numbers of non resident Cagi natives saved in their phones, reflecting the more extensive urban based community associated with the village. Another noteworthy community level difference concerns the heterogeneity of phone contacts. In all three villag indigenous

PAGE 255

255 hub for migrants from all areas of th e country. Subgroup L evel Fijian communities are very clearly structured by sex, age, marital status, rank and kinship. It might be expected that phone communication would likewise be structured and segregated along these lines. The most important conclus ion of this chapter is that phone use patterns as well as the size and characteristics of ego centric calling networks are broadly similar between members of these various socio demographic categories Rather than the expectations of sex or age segregat ed communication that we derived from the discussion in Chapter 3 it seems as if there is extensive communication between members of different sexes and age cohorts. The most noteworthy differences, at the level of demographic subgroups, are that young pe ople tend to contact more other young people (and older people, other older people). In a related sense, single (and divorced) people disproportionately contact other single people. The magnitude of these trends is so slight, however, that the ove rall comm unication network is not in a meaningful sense by age, sex or marital status. T he calling network s are dominated by kin relationships, although young people o verwhelmingly other indigenous Fijians, with members of other ethnic groups mainly contacted for business reasons. Among kinsmen, the majority of ties in the same generation are cross relatives, while the majority of ties in the first ascending/descending generation are parallel relatives. Based on the kinship etiquette discussed in Chapter 3 this might seem to indicate that phones are primarily used to

PAGE 256

256 contact age peers for social purposes, while contact with members of different generations is usually fa mily related, and perhaps more instrumental in purpose. Furthermore, frequency in calling between same generation cross relatives is markedly higher than parallel relatives. Individual Level Although there are broad similarities in the calling patterns be tween demographic subgroups as a whole, the second key observation of this chapter is that there are substantial differences in phone use among individuals, both within and across these sociological categories This is true of both phone use frequency/inte nsity, and in terms demographic and socioeconomic variables (age, sex, etc.) play a less important role tories, and specialized social/economic roles within the community. In terms of spatial and temporal dynamics, m ost calling patterns are highly uneven; people tend to make many calls when they purchase a 24 hour FreeCall promotion, and otherw communication might occur during one day of the week. Thus, in rural Fiji phone use patterns are heavily shaped by the pricing and schedules of promotions.There is a tendency to call contacts who live nearby (in the same village, or the local area) more often than contacts who live far away. Many phone contacts fall into a group that calls each other monthly, but only meets on a yearly basis. The overall conclusion of this chapter is that phone s fulfill multiple, diverse functions. Any individual may use their phone primarily to contact close kin, friends, or romantic interests. An individual might mainly use his/her phone for two unrelated

PAGE 257

257 purposes: on the one hand to call close kin for instrum ental purposes, and on the other communication may be very different. Thus, phone use is closely tied to individual, idiosyncratic needs and personality traits. This is refl ected by broad variation in phone use patterns at the individual level; however, these differences fade away at the level of socio demographic subgroups and whole communities.

PAGE 258

258 CHAPTER 6 : MOBILE PHONES, EXCHANGE AND INEQUALITY WITHIN FIJIAN COMMUNITIES Ov erview : Economic Benefits of Mobile P hones With the rapid adoption of mobile phones in developing countries, researchers and policymakers have been eager to determine the extent to which mobile phones facilitate various aspects of social and economic devel opment. A popular narra tive depicts mobile opportunities for inhabitants o f remote and underdeveloped region s (Corbett 2008; Sullivan 2007) Th is optimistic narrative has resonated strongly in Pacific nations, such as Fiji, where development initiatives have long been hindered by the difficulties of inter island communication and trans port (Belshaw 1964; Spate 1959) Althoug h it is generally acknowledged that access to ICTs such as telephones, mobiles, and internet can provide economic benefits at both the macro and micro scales few studies have documented in detail the actual mechanisms through w hich these benefits materialize. Furthermore, it is unclear how evenly the economic benefits of connectedness are distributed within communities and whether this tends to reduce or increase pre existing levels of soc ioeconomic inequality Several mechanisms have been proposed throu gh which mobile phone access can lead to economic benefits National level studies have correlated rising mobile ownership with increased GDP and inflows of foreign direct investment (Waverman, et al. 2007) At the individual level, researchers have theorized that mobile related economic benefits can derive from increases in social capital due to enhanced networking capabilities (Goodman 2007) the advantages gained from better access to market information (Jensen 2007) the opportunities mobiles provide to start new

PAGE 259

259 businesses (Samuel, et al. 2007) and through the sending and receiving of remittances (Horst 2006), (Skuse and Cousins 2007) This chapter focuses particularly on the connection between mobile communication and remittances, and more broadly, the exchange of econo mic resources among Fijian kin. Mobi le phones facilitate remittance sending by allowing individuals to promptly communicate time sensitive eco nomic needs (such as payment of hosp ital bills or school fees) and by assisting with the micro coordination of remittance transactions. Mobile phones can now even be used to instantly transfer money between users (Mobile Money or M Paisa) a service available in Fiji since 2010. If it gain s a sufficiently broad user base, mobile money transfers may partiall y fill the role of brick and mortar banks, which are absent in many rural developing areas. In Fiji mobile phones play a facilitating role in a ll varieties of exchange For e xample, the task s of accumulating ceremonial goods and making wedding arrangements are made much easier with the mobile phone. M obiles enable Fijians to quickly notify bereaved kin i n the event of a death, so that funeral attendees can amass their funerary gifts and t he burial can be carried out in a timely fashion At an interpersonal level, mobile phones assist Fijians to instantly communicate requests f or assist ance to their kin and to arrange the logistics of transfer s of money, goods or services. If mobile phones enhance and other forms of exchange then village households with the greatest degree of phone access may also derive the bulk of the economic benefits of telecommunications development phones and exchange must be contextualized within the broader village economy. I first describe rural Fijian productive

PAGE 260

260 activities and sources of income. I then describe Fijian exchange practices. Finally, I address the topic of economic inequality in Fij ian communities. For ea ch section of the chapter I provide a descriptive overview of conditions and practices in Fiji, and then present the relevant data that I collected in Toba Cagi and Veidogo The Rural Fijian Economy Fiji is considered a developing c ountry, although it is among the most developed of the Pacific Island nations with a per capita G DP of $4,900 (CIA 2013 ) Compared to other Pacific island countries, Fiji has a relatively diversified economy based around sugar exports, tourism, and remitt ances from Fijians working abroad. Other industries include clothing, copra, gold and silver mining, and lumber. 70% of the labor force is engaged in agriculture, a large portion of which are subsistence or semi subsistence farmers. The economy of a Fijia n village is a hybrid of communal/ kinsh ip oriented and individualistic/ market oriented relationships and processes Although land in rural Fiji is communally owned, much agricultural production is geared toward sale in the local or export market. Economic e xchange among villagers likewise has mixed features of kinship buying and selling has become a vital aspect of the rural Fijian economy, as villagers need money to pay school fees, church tithes, contribute to village fundraisers, travel to weddings and other events, and to buy clothing, food, and a growing range of household appliances. Money can be incorporated within Fijian gift exchange: in fact, the cru cial distinction between kinship exchange and the market is not whether or not money is involved, but rather the nature of the relationships, sentiments and obligations invoked in the exchange (Toren 1999).

PAGE 261

261 The Household E conomy Most economic activities i n the village are organized and car ried out at the household level. As described in Chapter 2, relationships within Fijian households are hierarchical in nature. The married male head of household directs the labor of other household me mbers, and in partic ular directs sons and other junior males in the household as to which tasks they should accomplish that day in the gardens such as clearing plots, weeding, planting or harvesting The wife of the head of household is ultimately subservient to her husband, but she oversees the completion of household tasks such as cooking, cleaning and childcare, and supervises the labor of daughters and other junior females.The male head of household is the final arbiter in decisions about the allocation of household goods and resources In practice however, many Fijian wives have a considerable say in household financial planning, and most Fijian couples arrive at economic decisions through open discussion. Youths occupy an ambiguous position in the household economy: the y have a degree of freedom and self determination, which is balanced with an expectation that they contribute sufficiently to the household. Male youths are active farmers, and after completing their schooling usually begin cultivating their own plots of l and. When youths obtain money from the sale of crops, copra or fish at the market, they are expected to contribute part of this income to th eir parents in the form of cash or storebought goods. Th contributed to h is parents varies from household to household, but generally seems to be in the rang e of to 2/3 of the total. Land Land ownership is one of the points of articulation between the household and the collective modes of production in Fiji. Land in Fiji is h eld in communal ownership by

PAGE 262

262 subclans ( mataqali ), whose members by law have lifelong usufruct rights. In practice, however, specific plots of land are controlled by the various sublineages ( itokatoka ) and the plots are further subdivided at the household level. Typically a household maintains control over a plot of land as long as it is kept in continuous cultivation, and plots are passed down from father to son across generations Within a land holding there can be shifting of plot bo undaries over time, as demographic ambitions and personalities exert themselves. For example, if all the members of a given itokatoka have died out ( kawaboko ), or have migrated to urban areas, their land may be encro ached upon by the remaining sublineages Hardworking farmers may establish control over additional land by clearing long fallow plots and by planting fruit or nut trees. Taking cr without first asking permission is considered theft ( butako ) an d may be punished by law There are certain degr ees of kinship which allow bending of these property rights however. O ne day as I walked down the road with a group of young women we drank green coconuts and picked pumpkins f or dinner from a roadside garden ; they explained that this was acceptable because one of the women had a special cross kin relationship known as vasu with the landowner and therefore had privileged access to his goods Closely related villagers might si milarly the owner of the garden only after the fact. Village politics surrounding land are shaped by the relative resources and needs of the kin groups within the village. Cagi has plentiful arable land. While the land re sources of Cagi mataqali are not equal, ambitious farmers in any mataqali

PAGE 263

263 an other mataqali in order to plant crops. In Toba by contrast, available arable land is insufficient to fulfill the economic aspirations of all villagers. S ome Toba mataqali have much more extensive coconut groves than others ; and some lineages have expanded in population while others have dwindled, causing land imbalances One Toba man is the last surviving representative of his mataqali and solely controls the large mataqali landholding. He is actively seeking to lease this unused land to a tourism developer as a way to earn extra money. Such adverse conditions in villages can encourage rural urban migrati on, or perhaps uxorilocal postmarital residence if a man marries a woman from a village with greater economic opportunities. Communal O bligations Fijian v illage life demands significant contributions of labor and economic resources from households, in orde r to accomplish collective goals. Some of these resources and labor are mobilized through kinship ties while others are mobilized through t he village government structure. The annual fund raiser and monthly village maintenance tasks fall under the purview of village gover nment, and are overseen by the turaga ni koro (elected village head man) and the village committee. The first week of each month is designated the macawa ni koro (village week) in which all villagers are expected to pitch in to cut the grass dig garbage pits, clean the community hall, and other upkeep and maintenance tasks. Alternatively one day out of every week may be reserved for such communal work After dawn the turaga ni koro blows the davui and when breakfas t is finished the men straggle out of their houses to trim the grass, while women rake up fallen leaves and burn piles of garbage.

PAGE 264

264 Figure 6 1. Toba village men form a bucket line to clean out the village bathing pool during a communal work day Photo c ourtesy of author. C ommunal labor demands can lead to tensions in any village and grumbling that some indi viduals do not do their fare share of the work As one Toba The way we do things here gether as a villag e to cut copra to pay our village debt, not everyone pitches in. Our lack of productivity and adequate preparation is always explained with excuses ( iulubale A key civic obligation is the annual village fundraiser ( soli ni estimedi ). The purpose of thi s fundraiser is to secure the money needed to fin ance village projects and upkeep The village committee arrives at an estimate of the funds needed for the upcoming year and the responsibility for contributing money is typically either divided

PAGE 265

265 among each of the household s adult males, or the individual residents of the village As villages have developed more complex needs, the cost of annual fundraisers has tended to climb. In Cagi in 2011 the soli was set at $ FJ 300 per adult male, a daunting sum for ma ny rural Fijian households to accumulate The day of the annual soli is festive, and everyone turns out in their brightest jamba ( dresses ) and bula shirts to drink kava in the community hall. The total amount contributed by each individual is displayed on a large chalkboard for all to see and to comment upon This is an occasion for households and kin groups to display their power ( kaukaua ) and social worth in the village. The atmosphere at such events is always one of levi ty, but beneath the surface, har sh judgments and evaluations are being made Those who have contributed their full share sit confidently and contentedly, while those who have come up short cower uneasily around the edges of the gathering; some laggards skip the event entirely out of sham e Households that have shirked on their contributi ons will be the subject of poin ted gossip, and likely feel social repercussions until they belatedly complete their solis perhaps months later. Many village households request financial assistance from m igrant kin around the time of the soli so that they are able to complete their contribution on time. Fundraisers are also a mechanism through which migrants and urban community members can periodically re assert their active membership in village life. No n residents who contribute to the annual soli are publicly acknowledged during the proceedings. Residents, whose own financial burdens are made lighter by voluntary migrant

PAGE 266

266 contributions may demonstrate their appreciation by s howing such non residents spe cial hospitality and respect whenever they return to the village for a visit. Other communal obligation s derive from ties of kinship. The marshaling of labor and resources for these obligations is overseen by hereditary heads of kin groups. For example, t he amassing of funds and traditional valuables ( iyau ) for weddings an d funerals are overseen by the turaga (senior members or chiefs of sublineages) for the itokatoka or mataqali involv ed in a ceremonial event. A s the wedding of an itokatoka member approac hes, each household within the itokatoka might be tasked with providing a given number of bags of root crops for the feast During the wedding itself (a traditional Fijian wedding lasts 4 days) the men and youths of the itokatoka work together to butcher livestock and prepare the earth ovens ( lovo ) Each household teeth), which are pooled together in order to make a group level presentation to the note is kept of which household contributed which items. Later, those households that contributed valuables receive a proportional share of the valuables received in the exchange. Sources of Household I ncome in R ural Fiji Fijian villages vary widely in t erms of the extent and productivity of land and fishing resources, and in terms of local employment opportunities. In c ommunities located close to urban areas or tourist zones such as Veidogo residents are able to c ommute daily to wage paying jobs Low s killed jobs in Fiji typically pay only about $FJ 2 3 per hour. Neve rtheless, such a dependable source of income is usually considered preferable to a household relying entirely on farming for its livelihood

PAGE 267

267 For rural and outer island village residents, s teady wage paying jobs are usually not available in the local area People in remote communities may either migrate seasonally to work in the cane fields of western Viti Levu or work for a period of years in an urban area before return ing to the village perhaps with accumulated savings invested in a new house or a vehicle Instead of wages, rural v illage dwellers rely upon periodic sale of handicrafts and cash crops such as kava, copra, and taro in order to obtain money Some are able to bring their produ ce directly to an urban market, while many others sell locally to middlemen or cooperatives Farming for subsistence consumption and for the market Gardening with hand tools remains the economic ma instay of most Fijian villages; v irtually every rural house hold maintains some sort of a garden or farm. Sizes of household gardens, the variety of crops planted, and proportion of crops used for personal consumption (versus sale) vary based on availability and quality o f local land, the amount of able bodied labo r a household can muster, and local transport and access to markets. For example, Toba farmers gro w crops mainly for subsistence consumption because land on their island is not very fertile or abundant, and because the nearest market is on a neighboring i sland. Ther e is no ferry service to Toba and transport of bulky root crops on small open boats is prohibitively expensive As a consequence Tob ans depend upon handicrafts, copra and the occasional sale of fish to obtain money. Weaving mats for sale is the exclusive domain of women, but forms such an important part of Toba household income that men also contribute to the process by planting pandanus trees, and harvesting and processing the leaves.

PAGE 268

268 Figure 6 2. Toba men working together to cut copra Pho to courtesy of author. Cagi in contrast, has ideal conditions for raising cash crops There is abundant, very f ertile land, regular rainfall, and ferries from Suva visit the island twice a week. Consequently Cagi households pl ant large gardens which ampl y provide for subsistence needs as well as yielding cash crops (mainly taro an d kava ) which are sold to middlemen working on the island. The village has a cooperative that b uys green copra di rectly from villagers and dries it before selling it to urban bu yer s. Farming can be so lucrative, that Cagi ans occasionally pay one another in cash in exchange for a gardening work. In particular, if a man has a large task such as clearing a new plot out of the forest he may pay other villagers about $FJ 15 do llars per day for their labor

PAGE 269

269 The timing of the harvest of the various crops allows for the provision of different household needs. During the season, copra can be harvested everyday, and is a steady source of income used to pay for daily expens es such a s food from the store. By contrast, kava takes at least 3 years to mat ure, so villagers tend to use a kava harvest to finance large, one off expense s such as house repairs. Cagi villagers closely follow the price o f copra, taro and kava in the Suva market and try to time their harvests for periods of high prices. For communities such as Cagi that rely on sale of crops in the market, national and international price f luctuations play a major role in shaping cycles of economic hardship and prosperity. Veido go does not have especially fertile or abundant land, but it is blessed with a very convenient location near the capital city Suva and with ample marine resource s. Most Veidogo households earn money through daily wage labor in the city, and supplement th is income by gardening and fishing Subsistence activities in Veidogo reduce household expenditures on store bought food and if necessary seafood or cassava can also be sold at the Suva market to earn extra money. Veidogo households therefore have a high degree of flexibility in pursuing diverse income sources and resilience against economic hardship Handicrafts Certain Fijian villages mass in these villages a wide variety of wooden bowls, masks, weapons, an d other memorabilia, many of them non traditional in design are produced and sold directly to retailers, or sold through a cooperative arrangement. In most communities however, handicraft production is geared for local use or sale, and centers upon pandan us mats ( ibe ), plaited from the tough, leathery leaves of a coastal tree ( Pandanus spp. ).

PAGE 270

270 Figure 6 3. Cagi women working together to plait a pandanus mat ( ibe ) Photo courtesy of author. Ibe have important practical and ceremonial role s in Fiji. Fijia ns sit on the floor s of their houses and every floor surface of a house apart perhaps from the kitchen, is covered in ibe Ibe may last for years, but eventually become dirty and tattered and must be replaced. These mats are also a key item in ceremonial exchanges associated with weddings, funerals, and other life crisis ceremonies. Ibe for ceremonial occasions can be very ornate, with a fringe of colorful yarn ( kula ), and interwoven patterns of black dyed pandanus ( somo ). Some of these patterns are the c ultural property of

PAGE 271

271 specific villages or regions, and are only pr oduced by women in those areas, as with the distinctive vakadivilivili pattern of Batiki mats On islands without local employment opportunities ibe are an import ant source of household inc ome, particularly because they are a large store of value. Fijians are rarely able to save significant amounts of money, so the sale of an ibe can be timed in order to pay a large exp ense, such as school fees or a household contribution to the annual fun draiser. Depending on size and design, ibe can fetch anywhere from $FJ 50 to upwards of $FJ 500. Plaiting ibe is the exclusive domain of women, although men may assist w ith other parts of the process, such as planting pandanus trees, harvesting the leathe ry leaves ( voivoi ), and drying or cutting off t he Depending on the size and design of the mat s and the skill of the woman one woman might plait 2 or 3 ibe in a year. In Cagi the women form cooperative work groups to plait ibe tog ether. They wor k together to rapidly complete mat, then move on to the next, and the next. This arrangement encourages women to work efficiently, makes the work more enjoyable by making it a social event, and allows the quick completion of an i be for sale. Such c ooperative workgroups are also used to plait ibe he ibe together in order to raise money for a chicken coop, which they then planned to use to start an egg business. Plaiting of ibe is an integra l part of village social life, as w omen spend much of their day plaiting in the company of other women. The work is very repetitive, and the

PAGE 272

272 women spend mo st of the time chatting, joking, and exchanging news. The weaving of ibe is therefore an important context for information exchange among village women. Urban women do not typically plait ibe becau se they are unable to cultivate pandanus, they may have full time wage employment, and they often live in cramped housing without sufficient floor space t o plait mats. Village women say that many of the women who have been raised ( susu madrai ) ( i.e., in urban areas ) have never even learned how to pla it ibe Because urban people do not create their own ibe they often buy th em from villagers. Sometimes ibe are specially commissioned the urban b uyer describe s the exact size and design, and the villager will create the mat for an agreed upon price. The re involved haggling is considered demeaning by Fijians, and at any rate the prices for various mat sizes are fairly standard Village women do not wait for such orders however, and work continuously on plaiting ibe in the expectati on that the resulting mats will eventually find a buyer If a household has an extra ibe on hand they may use it in the household, or save it for exchange at an upcoming ceremonial obligation ( soqo ). Typically a sale is arranged when in Suva passes on the rural sel needs an ibe Tara a Toba woman, plaited an ibe last year and was looking for a buyer. Sera a kinswoman of who works in a resort in the tourist zone of Western Viti Levu told some acquaintances there there that she knew of someone with spare ibe for sale and they arrived at a deal Toba w as taking a trip to go visit Sera

PAGE 273

273 and so he carried the ibe along to deliver to the buyer, and collecte d the money to give to Tara when he returned. Usually the buyer of an ibe is not directly related to no r acquainted with the seller, and there are few repeat buyers. Phones are therefore very useful to rural women as a means of getting in contact with urba n buyers. As one Toba the amount of money we spend on phone credit is small compared to the money we get from selling an ibe so we see the phone as a big economic benefit in our household. Small business Island economies are characterized by a high degree of openness (reliance on trade) and hyperspecializatio n (economic development based on one or a few products and services) (Bertram and Poirine 2007) These characteristics are a function of small size and remoteness: most islands have few natural resources and a wide variety of products must be imported. Furthermore, island economies face diseconomies of scale, high transport costs, and limi ted social capital F r om the standpoint of island communities then, any small business opportunities enabled by the mobile may be disproportionately beneficial Enterprising villagers with some savings or disposable income often run small them in the village at a su bstantial markup Typical items sold in village canteens include kava, cigarettes, tinned fish, rice, sugar, oil, noodles, popsicles (where there is electricity), ph one cards, or homemade sweets. Most s uch household businesses are very casual and informal, and are run directly out of the living room, with the store wares perhaps locked in a cabinet. Most household businesses do not last very long before unpaid debts accumulate, or the stock runs out, or is consumed by the household

PAGE 274

274 members them selves. One T oba couple confessed, When we want to top up our phone we just take it out of our store b a tin of fish off the store shelf Lako ga! ( Just g o for it! ) Often, villa gers will venture to such a home cash in hand sa cegu ). Because these businesses are so ephemeral, and most do not keep books, it was difficult for many villagers to estimate the profits they had earned from their business es I n addition to such small, temporary canteens most Fijian villages also h ave at least one permanent shop that is either run by a local family or by a cooperative. Such stores sell a broader range of goods rice, flour, sugar tobacco, kava, as well as thin gs like toilet paper, soap and fishhooks. These store s usually keep formal records of inventory, purchases, and sales. The owner of a store in Toba explained to me that t he phone is vital to his store operations H e needs to keep track of the comings and g oings of boats so that he can order new inventory and have it promptly transported to the island. To obtain new inventory he calls his brother in Suva, who buys goods in bulk and arranges to drop them off at the wharf when ever the next boat is departing f or Batiki. His urban brother also transfers credit to his phone so that he can make the necessary calls. I did not encounter many innovative or new forms of business that explicitly relied on mobile phones. The only possible candidate was a fish marketing arrangement on B atiki. Batiki has rich fishing grounds but access to the market is difficult due to remoteness and a lack of regular ferry transport. Batiki village youths have teamed up with a boat owner from a neighboring island, who calls ahead of time when ever he is

PAGE 275

275 making a trip to Suva. With this forewarning, the youths can fish all night long and transport the catch fresh to the Suva market Without the microcoordination enabled by mobile phones, the delivery might be delayed and the f ish would quic kly spoil. Government assistance and pensions The Fijian government provides various forms of welfare and social security to widows and other eligible individuals. This is typically a means tested old age and disability assistance of $FJ 60 110 per month. Formally employed workers pay a percentage of their wages to the Fiji National Provident Fund (FNPF) which the person may receive upon retirement as a lump sum, or as a monthly pension. Out of 99 surveyed household s 26 (26%) had received pension payments or government assistance in the previous 12 months. The majority of these village households received monthly disability assistance on behalf of an elderly widow living in the household. Household I ncome in Toba Cagi and Veidogo I measured household inco me via a survey, in which I asked the head of each household to estimate the total amount of money the household acquired during the previous year from various sources: sales of crops (with separate categories for the major cash crops) copra, fish, livest ock and handicrafts, wage labor, small business earnings, and income from pensions and government assistance programs. Some respondents had difficulty estimating the amount earned in a year from certain sources; in these cases I asked them to estimate the amount earned during a typical we ek or month during the past 12 months and calculated the annual total based on that estimate. income categories were designed to encompass all local monetary income sources Only o ne significant aspect of hou sehold income was not addressed in

PAGE 276

276 the survey subsistence income accurately would have been difficult given the variable proportion of storebought vs. farm grown food consumed among village h ouseholds and the complexity of estimating monetary equivalents of food consumed Nevertheless, these income data provide a basis for comparing the relative income levels of village households. Figure 6 4. Box and whisker plot of household income by v illage (exchange excluded). The box and whisker plot displays the minimum, 1st quartile, median, 3rd quartile and maximum for household income for each village. This figure gives an idea of the large differences between the villages, and large differences among households within villages. Table 6 1 Mean h ousehold income over the past 12 months in the study villages Village I ncome ($FJ) (exchange excluded) Net income from exchanges ($FJ) Net household income ($FJ) Toba 1771 +/ 1080 250 +/ 801 1522 +/ 1053 Cagi 5764 +/ 5262 90 +/ 910 5855 +/ 5253 Veidogo 11137 +/ 12068 832 +/ 2550 11969 +/ 12503 M ean income figure s are reported along with their standard deviations The data indica te that there is substantial variation in household income, bo th between villages and among households in the same village. Veidogo household mean reported income is over 6 times that of Toba households while Cagi

PAGE 277

277 mean reported income is roughly in middle of the two other villages. This is cert ainly in dicative of unequal economic opportunities for residents of the 3 study villages. I f subsistence farming and fishing were incorporated into the income calculation, then the differences between communities would likely be less pronounced Tob ans and Cagi an s derive a higher percentage of their diets from subsistence than do Veidogo a ns. Suffice it to say though, that cash is far more plentiful in Veidogo than in Toba households, while Cagi has an intermediate economic position. Sources of income in the study villages Although Cagi mean household income is more than three times that of Toba households the relative proportions of various income sources in Toba and Cagi are quite similar. In both villages roughly of total household income derives from wages (whether from local employment, earned by current residents and carried back to the village, or sent back by household members working in urban areas), and 10% of total income derives from government assistance welfare or pensions. In both Toba and Cagi roughly 20% of total household income derives from copra sales. In Cagi sales of taro and kava account for a further 35% of total income, whereas in Toba fish and handicrafts (likewise harvested from the environment) account for 32% of income. In contras t to these two outer island villages, residents in peri urban Veidogo derive 85% of their income from wages and a further 7% from small businesses. Only around 5% of income is derived from sale of fish and crops at the Suva market. Farming in Veidogo is ca rried out on a more limited scale and nearly all crops that are harvested are consumed by villagers rather than sold. Still, a high proportion of Veidogo a consist of storebought food.

PAGE 278

278 Figure 6 5

PAGE 279

279 Income inequality between village households In addition to the inequalities between communities, income is also unevenly distributed among households within the study villages The Gini coefficient is a standard measure of inequality, in which a coefficient of 0 represents perfect equality (all households have the same income), and 100 represents perfect inequality (one household earns all the income, the rest earn none). The Gini coefficients for household income (not including exchange income), for Toba Cagi and Veidog o are 32.2, 38.8, and 45.1 respectively might expect a small commu nity to have significantly less inequ ality than an entire country, based on the similarity in lifestyle and income s ources among village households but the inequalities in the 3 study villages are comparable to that of Fiji as a whole. Table 6 2 Gini coefficients for household income (witho ut exchange) in study villages Village Gini coefficient (hh income without exchanges) Toba 32.2 Cagi 38.8 Veidogo 45.1 Villagers did not openly complain about these inequalities at least in my presence, nor did they seem to consider them a major issu e. Villagers explained inequality primarily in terms of personal responsibility those who work hard on the farm make lots of money, and those who are lazy make little money. Lack of farming productivity is often attributed to drinking too much kava, which produces a hangover that predisposes a man to sleep away the morning. There is a healthy respect for hard work in rural Fiji which is understandable given the communal nature of many tasks

PAGE 280

280 and obligations S hirkers directly affect everyone else in the vi llage, who must pick up their slack; this may be a contributing factor to the prevalence of the discourse about personal responsibility. Figure 6 6 Lorenz curves for study villages for comparison. Lorenz curves provide a visual representation of the d istribution of income in a population. The line of equality represents a theoretical distribution in which income is totally evenly distributed. The greater the area between the curve and the line of equality, the higher the degree of inequality (as measur ed by the Gini index). A few inf ormants provided more nuanced commentary on the underlying reasons for inequality in Fijian villages. Several noted that while some households receive substantial remittances from ur ban or overseas relatives, oth ers do not Josefa, a Cagi man, described how the land that is available to the various households in the village is not equal, either in extent or in productivity. He noted that in the cyclone of 2010, villagers with low lying gardens saw their crops flooded and de stroyed, while others with gardens in the hills did not suffer the same losses Land Josefa explained other villager still such arrangements are not alw ays made Anoth er man described how some

PAGE 281

281 villagers inherit large coconut groves ( loga ni niu ) that were planted by their fathers and grandfathers, while others inherit none. This is still not a good excuse the man noted, given that a man may plant his own coconut tre es and they will become productive with in just a few years. In practice, though, many people lack the drive or initiative to improve their prospects. Finally, another informant emphasized the role of education in the economic well being households, and th e village more generally He noted that some villagers hav e only 6 (or even fewer) years of formal education. In his opinion, t here is a big difference in the mindset and capabilities of such people when compared to people who complete secondary school. In he lack of education On a day to day basis financial hardship in rural Fiji can stem from various causes Illness, particularly in the adult, productiv e members of a household, is an unforeseen cause for hardship. Another is inclement weather. In Cagi for example, most households rely on copra to earn m oney on a daily basis. During prolonged periods of rain the copra in the village dryer does not dry su fficiently. Instead of using the sun and wind to dry the copra, villagers must rely on firewood which is much slower. Once the dryer fills up with green copra the local cooperative no longer purchases copra until the existing stock is dried. In such time s, households may resort to borrowing money, or just eat produce from the farm. Households that run short of daily essentials like soap and kerosene will kerekere these from neighbors who have not yet run out. On rare occasions the store may announce that it is allowing $10 of store credit per family, to be paid back when the weather improves and the dryer reopens.

PAGE 282

282 Another situation that can lead to financial hardship is unfavorable market conditions. Taro is ordinarily sold to middlemen until the first wee k of December, but the market closes until the first week of January, because the export boats to New Zealand g the Christmas holiday. In addition, the market price for kava fluctuates considerably : i n 2010 the price of kava fe ll as low as $FJ 18/kg. By S eptember 2011 it had risen again to $FJ 30/kg. The price also fluctuates seasonally it falls at the beginning of the school holidays because lots of people harvest at that time, he upcoming term; hence, t he market becomes glutted and the price falls. When the price is low, most villagers wait for the price to rise again before harvesting. Sometimes, though farmers get desperate and are forced to sell their kava at a low price just in order to live ( me rawa kina na bula ). Household in come, lifestyle, and phone use ificant impacts on lifestyle, the quality of housing and range of material possessions In the 3 strudy villages t here ar e significant positive correlations between household income and the education level level of the head of household, the number of household appliances and telephones, and the amount of travel engaged in by household members (see Table 6 3 ). Table 6 3 C orrelations between household income and socioeconomic variables Annual household income (exchange excluded ) S ocioeconomic variable s N Statistic Sig nificance A nnual hh income HH head education (years) 99 .454 .000 A nnual hh income Household appliance count 99 .477 .000 A nnual hh income # functioning phones/hh 99 .423 .000 A nnual hh income Trips away from local area/y ea r ( hh head ) houjljhou household) 99 .260 .009 Furthermore, household income and certain other socioeconomic variables have s ignificant positive correlation s with heavy phone use. The num ber of functioning household is positively correlated with calls and texts per

PAGE 283

283 day, phone credit used in a week, and composite phone use rank. The number of trips a pho ne owner took away from the local area in the past 12 months is positively correlated with the number of saved con tacts in their phone, and composite phone use rank. Household income (without remittances) is positively correlated with composite phone use r ank. Table 6 4 Correlations between socioeconomic variables and phone use variables Socioeconomic variables Phone use variables N Statistic Sig. (2 tail) Annual hh income # of contacts saved in phone 170 .050 517 Annual hh income Calls+texts per day 161 .135 .087 Annual hh income Phone credit used in a week 161 .041 .609 Annual hh income # of other phones saved in 170 .093 .230 Annual hh income % of surveys named as frequent phone user 170 .088 .256 Annual hh income Composite phone use rank 170 .207 .007 HH head education (years) # of contacts saved in phone 170 .060 .436 HH head education (years) Calls+texts per day 161 .018 .820 HH head education (years) Phone credit used in a week 161 .105 .187 HH head education (years) # of other phone s saved in 170 .011 .888 HH head education (years) % of surveys named as frequent phone user 170 .067 .385 HH head education (years) Composite phone use rank 170 .103 .180 Household appliance count # of contacts saved in phone 170 .054 .481 Household appliance count Calls+texts per day 161 .179 .023 Household appliance count Phone credit used in a week 161 .132 .095 Household appliance count # of other phones saved in 170 .100 .197 Household appliance count % of surveys named as frequent phone u ser 170 .130 .092 Household appliance count Composite phone use rank 170 .146 .057 # functioning phones/hh # of contacts saved in phone 170 .019 .806 # functioning phones/hh Calls+texts per day 161 .315 .000 # functioning phones/hh Phone credit used in a week 161 .164 .037 # functioning phones/hh # of other phones saved in 170 .032 .677 # functioning phones/hh % of surveys named as frequent phone user 170 .110 .153 # functioning phones/hh Composite phone use rank 170 .156 .042 Trips away fr om local area/year # of contacts saved in phone 170 .194 .011 Trips away from local area/year Calls+texts per day 161 .033 .681 Trips away from local area/year Phone credit used in a week 161 .041 .605 Trips away from local area/year # of other phones saved in 170 .023 .769 Trips away from local area/year % of surveys named as frequent phone user 170 .182 .018 Trips away from local area/year Composite phone use rank 170 .210 .006

PAGE 284

284 It seems that material prosperity (both high income and ownership of many household appliances) is associated with heavy phone use. This makes sense, from the standpoint that money allows one to purchase mobile phones and calling credit. It seems, then, that households with higher income do have more telephones and use their phones more frequently/intensively. Kinship Exchange and R emittances Indigenous Fijians, like members of other kinship oriented societies, frequently exchange goods and services through non market based mechanisms B uying and selling between Fijian k insmen is generally discouraged; instead, they exchange economic resources as reciprocal gifts. Aside from their functions in reinforcing social solidarity, reciprocal exchanges play an important role in addressing economic needs and hardships in Fijian communities. Some Fijian gift exchanges are large scale ceremonial events associated with weddings, funerals, or other group level rituals. These large scale exchanges involve the poolin g of resources within kinship groups, followed by ceremonial ex change between separate groups, and subsequent redistribution of the acquired resources within the respective groups. Other exchanges are small scale and lacking in ceremony and involve reciprocal transfers of goods and services among close kin smen Recip rocal exchange events can vary widely in terms of the volume of goods or mone y (in terms of value), and the period of time that elapses between the initial gift, and its counter gift. T he essential et hic underlying Fijian kinship exchange is helping a kinsman in need Thus, when a kinsman makes a request, it is very difficult to deny, unless fulfilling the request would place the donor herself in need. Assuming the request is granted, it

PAGE 285

285 gives rise to an implicit obligation on the part of the recipient to reciprocate in the future should the donor need something in return. el for generalized reciprocity (1972) gifts between Fijian kinsmen need not be reciprocated in equa l measure n or wi thin a specified time frame. A series of unbalanced exchanges may even be used repeatedly in order to effect a flow of resources from the affluent to those who are in ne ed. However, if a gift remains unreciprocated, it implicitly elevates over that of the recipient. Consistent with the Fijian association of chiefliness with prosperity and generosity, repeated giving brings social prestige. Thus, in Fijian communities exchange can serve to reduce material inequaliti es by converting them into social inequalities (Sahlins 1962). Fijian physical presence. In such cases the sentiment of kinship embodied in the exchange is reinforced through body l anguage, elaborate speeches and perhaps a shared bowl of kava T hese expressive aspects of exchange may b e absent when using the telephone as a medium of communication. N evertheless Fijians seem generally accepting of the use of mobile phones to arrange l ong distance exchanges Several respondents noted however that more important requests are properly made face to face, rather than over the phone. For example, a request to a skilled for gardening these weighty requests should be made in person not over the phone and accompanied by a traditional offering such as a or bundle of kava roots

PAGE 286

286 We can devise a typology of exchanges based on the geographic proximity of the parties involved. There are exchanges that occur between households within villages, exchanges with people in nearby villages, exchanges with people living in other areas of Fiji, and exchanges with people living overseas. These categories are distinct bec ause the kinds of goods exchanged, frequency and value of exchanges, and degree to which exchanges are balanced tend to vary based on the relative locations of the exchange partners. As such, exchanges with kin in different geographic locations fulfill dif ferent functions in the household economy Exchange within the V illage S mall scale exchanges of food and goods among households are a pervasive aspect of village life in Fiji. These exchanges are known as kerekere which literally means The prac tice of borrowing hbors is so common that colonial authorities tried to discourage kere kere believing that it made capital accumulation impossible, and therefore served as a hindrance to economic development (Spate 1959) However, k erekere plays an important role in circulating ec onomic resources among households, and in enabling households to overcome episodes of scarcity in order to meet day to day needs. Because these exchanges are conducted within the confines of the village, ph ones play a negli gi ble role in intra village kerekere However, kerekere exchanges are still relevant to our discussion because they demonstrate the kinds of needs that can be satisfied locally, and the kinds that require assistance from kin living outside the village I did not collect comprehensive survey data on within village exchanges, because they are so frequent and yet relatively insignificant in terms of monetary value.

PAGE 287

287 However, I gave diaries to five households in each of the three study villages in which household members recorded all kerekere events in which they participated du r ing the span of a week. The data from these diaries are only peripher ally relevant to our discussion of mobile communication but suffice it to say that village househo lds averaged from 2 10 kerekere transactions per day, borrowing and lending out a wide variety of food, tools, and other goods to th eir neighb ors and kinsmen. A few typical examples of kerekere diary entries are: a cup of flour soy sauce, 5 spoons, pen an d paper, a fishing spear a spring for a generator, washing detergent, and hot peppers. by kerekere but only an unspoken expectation that the giver might sometime in the future make a similar request in return. Small amounts of money (around 1 5 dollars) may also be but unlike food or borrowed items, there is a general expectation that money be paid back promptly Kerekere is not carried out randomly: e very household in the village has certain other househ olds to which it turns for kerekere These ties of mutual suppor t are guided, but not predetermined, by kinship. For instance one Cagi man noted that requests to close relatives are easy, while requests to more distant relatives tend to be more difficult. Out of logistical convenienc e, kerekere partners often live in neighboring houses, but villagers also kerekere from close relatives who live in more distant corners of the village Specific households build up a working relationship over time. As one villa ge lady described, it is important that those who kerekere together get along well together ( sota na yalona ). Some villagers acquire a bad reputation because they frequently ask for things and seldom reciprocate. P eople usually avoid kerekere

PAGE 288

288 might make an excuse or tell a white lie if such a person makes a request help you also all Outright refusal of a kerekere is very bad etiquette however as i t is tantamount to denying a kinsman in his time of need There was one man in a village I visited who was mentally ill, and extremely persistent in begging for food and cigarettes He was such an outcast that people thought nothing of telling him to just go away. He would however enter the house and sit by the door, plaintively requesting food and remarking on how hungry he was. After e a ch sharp denial, he would remind the household residents to their dismay, of his kin ship ties to them. Inevitably the man would get what he came for, which illustrates how difficult it can be Exchange outside the V illage As part of my household survey, I asked heads of household to recall all individuals (not resident in the sam e village) who had sent goods or money to the household, and all of the individuals to whom the household had sen t goods or money during the preceding 12 months. I then asked the head of household to enumerate how many exchange events occurred with each pa rtner, and what goods or money had been exchanged. I n total I collected data about 1768 exchange events in the three study villages. I then used current market prices published by the Fiji government in order to calculate the value of exchanged produce an d to assign each transaction a monetary value. I n the following analysis I divide exchanges between villagers and kin living outside the village into three categories: loca l exchanges, exchanges to distant parts of Fiji, and overseas exchanges.

PAGE 289

289 Figure 6 7. Exchanges in the study villages, by percentage of total monetary value Table 6 5 Details of exchanges Exchange category Items sent % of all events total # partners Total # of events Total value ($FJ) Value per partner ($FJ) Value per event ($FJ) Local Area to Village Farm produce/ services 1% 3 14 220 73 16 Village to Local Area Farm produce 8% 9 146 4032 448 28 Rest of Fiji to Village Store bought food and goods; money 36% 171 633 39190 229 62 Village to Rest of Fiji Farm produce, handicrafts, money (rarely) 49% 183 862 48619 266 56 Overseas to Village Money 6% 27 107 32891 1218 307 Village to Overseas Clothing, toiletries <1% 2 6 118 59 20 Local exchange Exchanges with partners in nearby villages represented only a small proportion (9%) of the 1768 total exchanges carried out by Toba Cagi and Veidogo residents, and accounted for only 3% of the total monetary value of exchanges. Many of the inhabitants of neighboring villages are close relatives, due to frequent intermarriage in the local area. However, exchange among kinsmen in neighboring villages is not so

PAGE 290

290 important from an economic standpoint: Villagers can turn to their immediate kinsmen in their own village for small everyday needs (i.e. kerekere ), while the y tend to rely on urban or overseas kin in order to acquire money and storebought goods, which are similarly scarce in nearby villages. Most of the local area exchanges I recorded in the surveys were between a small number of Veidogo village households and their close kin l iving ne arby in Suva, who often host Veidogo children while they attend urban secondary school s The village rs regularly send farm pro duce and fish to the host families as a form of compensation. This food might be carried by the donor directly to the recipient, w ho only lives a bus or truck ride away. These exchanges were typically small and frequent perhaps just a few baskets of root crops or a string of fish each week and averaged only $FJ 28 in value per transaction. Exchange with partners in distant areas o f Fiji The bulk of all recorded exchange events (85%) occurred between vi llagers and kin living in distant parts of Fiji. This category is, for most intents and purposes, rural urban exchange: the majority of long distance exchange partners live in cities, primarily Suva and Lautoka However, a few exchange partners also live d in distant rural villages. Like kerekere between households in the village, rural urban exchanges are based ultimately on the principle of helping a kinsman in need and are not n ece ssarily equally balanced In practice, though, a shrewd sort of accounting often underlies these exchanges, and in due time a request from one end is very likely to be met by a counter request. Paulo, a Cagi man, said that when he and his friends in Suva t alk on the phone his urban friends often ask (e sega ni dua na ka o vinakata ? ) tell them if ther e

PAGE 291

291 return: ( qai dua mai n a mequ kava se ke qu dalo ). each other first though of each other and send Exchanges of food, goods or money are usually arranged between the married adult heads of households, on a household to household basis. I did on on e occasion see an exchange pooled between several related households. The wives of four brothers a ssembled together the morning a crate arrived from Suva, and measured out the rice and sugar out precisely into their respective plastic containers. Youths ma y independently conduct some exchanges on the side with friends or kin in the city. in contrast to those of adults, seem to be geared more toward recreational or personal items a bottle of liquor, a few packs of duty free cigarettes, sh oes, make up or cologne. In long distance exchanges, each side contributes those resources which they have in most abundance, in order to obtain the resources that they lack. R ural people send farm produce to urban relatives taro, cassava, and other root crops packed into old sugar sacks but also a variety of fruits, vegetables, kava and seafood and occasionally woven pandanus mats ( ibe ) This produce can be a sort of nostalgic om the native village, but is also a real financial boon to urban households because store bought food is so expensive. Urban kin send cartons of flour, sugar, rice, and other store bought foods, which are cheap er in the city than in the village or send cloth, clothing, tools, or other manufactured goods whic h are unavailable in the village store or they may transfer cash to a rural post office via TMO

PAGE 292

292 The mean monetary value of these trans actions turned out to be fairly balanced in both directions $FJ 56 from villagers to kin in other parts of Fiji, a nd $FJ 62 for goods or money sent to Toba Cagi and Veidogo villagers. The number of transactions sent from each side was also comparable. Transactions with distant dwelling kin, then, are usually reciprocal and balanced in nature. Figure 6 8 The men of a Cagi itokatoka load cassava and taro into bags, to be sent to relatives in Suva for a wedding feast Photo courtesy of author. The nature of exchange in any community is heavily shaped by local transport options. Cagi residents are able to send bags of produce (for a small fee) on truck s that board the weekly ferry to Suva. Relatives in Suva can then pick up the ir goods from the Suva office of the truck ing business. As such, Cagi ans can conveniently send fo od or other items to urban relatives anytime on only a few days notice Decades ago, b efore

PAGE 293

293 the bui lding of the road and the wharf on Koro Island, Cagi villagers depended on passing copra boats in order to travel or to send goods to Suva. These boats, without a fixed schedule, would appear on the h orizon unannounced and villagers quickly scrambled to their gardens in order to harvest taro bananas, or whatever else they wanted to send to Suva. According to villagers, the improvement and regular scheduling of ferry transport has enabled a far greate r flow of rural urban exchange. In Toba shifting fortunes have left the village in some ways more isolated today, than it was decades ago. Because Batiki is not visited regularly by a ferry, Tob ans must load any cargo into small open boats. This form of t ransport is inconvenient and costly, so Tob ans tend to just bring along a couple of bags of produce or land crabs with them when they go to visit their relatives in Suva or Levuk a rather than engaging in regular exchanges Part of Tob economic difficu lties stem from the ir inability to promptly acquire cash via telegraph money order (TMO) post office, which provided money transfer services, has been closed for the past two years due to mismanagement. V illagers tasked with running the money transfer service have repeatedly embezzled money After each scandal the post office has been closed until the debt to the Fiji postal service could be paid off by fundraisers in the local villages. Sin ce few people send letters, the loss of mail servic e has not been strongly felt. However, the inability to receive money transfers in the village has caused real hardship. Relatives still send money, but it mus t be retrieved at Levuka on the neighboring island of Ovalau. Most Batikians only visit Ovalau ev ery few months, perhaps once or twice a year, and the trip to Levuka can be cost ly, eating into the

PAGE 294

294 benefits any money received there. Also, the inability to receive money in a timely fashion diminishes s As a result o f the diminished inflow of cash, Toba two village stores must accept payment in copra, or simply provide credit. In the words of one Toba villager, it hurts a lot ( mosi saraga ) for the village to be in this cash strapped state of affairs. Aside from material exchanges of produce, money and goods, rural and urban kin provide important non material services for each other. When rural kin visit the city, either for recreation or business, they stay at the homes of their urban kin. These stays may be for a period of weeks or even months, and often require urban dwellers to make room for extra bodies in already cramped housing conditions. When rural kin arrive in the city, they rarely possess the resources to pay for the daily costs of city life such as bus and taxi fare s and storebought food. As such rural visitors may strain the financial means of their urban hosts. Urban households also host the children of rural relatives who come to town to attend secondary school. In the case of such long term hosting arr angements, the rural families send farm produce on a regular basis to help offset the costs of feeding their child ren When they visit town, r ural people alleviate the burden s on their urban hosts by bringing bags full of produce along wi th them. They also reciprocate by playing host to their urban kin whenever they visit the village usually around the Christmas holidays. Finally, rural residents perform important practical and ceremonial functions in the village on behalf of their ur ban kin. Village men often look after the farms and houses of the ir absent brothers and cousins. Some urban and overseas Fijians pay their village kinsmen to keep their land under continuous cultivation. The migrant profits from the

PAGE 295

295 sale of the crops, and thro ugh this arrangement the migrant also maintains control over the land in case they should eventually return to live in the village. Absentee villagers also prefer that their houses be constantly occupied, and will enlist a village kinsman to live in and ma intain their house Overseas exchange Exchanges with relatives living overseas accounted only 6% of the total number of exchanges, but 26% of the total monetary value of all exchanges in Toba Cagi and Veidogo These exchanges overwhelmingly consist of mo ney transfers via Western Union or another transfer service, and are very unbalanced: overseas relatives sent goods or money to the villages 107 times, with a combined value of $FJ 32,891, while villagers reciprocated on only 6 occasions, with a total valu e of only $FJ 118. In this sense overseas exchanges closely resemble one way remittances. This lack of reciprocity is justifiabl e based on the need centered logic of kinship exchange. Fijians living overseas tend to be well off financially compared to the ir village kin and t herefore capable of lending assistance with fundraisers, school fees, or other large expenses. Also villagers reciprocate in a sense by providing important services for their overseas kin such as looking after their houses and garden s, or hosting overseas relatives when they return to Fiji for a vi sit Most v illagers with overseas kin do not tend to correspond with them often by phone mainly initiating contact when they need monetary assistance with a fundraiser, school fees or an ot her expense. I spoke to Vili a Fijian soldier who serves in the British Army, who returns to Fiji twice a year for vacation. Vili said that his village kin rarely call him when he is overseas because of the expense. More typically villagers send him a

PAGE 296

296 text him and ask him to call salary he has the m eans to help them. In re turn, his kin look after his home and hi s land in his native village in Kadavu. When Vili returns to the village he is treated as an honored guest in recognition for his frequent contributions. He did complain mildly that when he returns to Fiji many casual friends and acquaintances in Suva tr y to get him to buy alcohol for them. With limited economic opportunities in rural Tonga, many village households there rely on remittances from overseas relatives to main tain their standard of living. Utilizing the local resources available to them industry selling traditional Tongan goods (mats and bark cloth) to migrant s in the U.S (Small 1997) This sort of exchange was not in evidenc e in Toba Cagi or Veidogo but could c onceivably emerge in the future with falling mobile calling costs, and enhanced communication possible through the internet. Balance of reciprocity There were a total of 301 exchange relationships between village hou seholds and people residing outside the study villages. Of these 301 relationships, 94 (31%) featured reciprocated exchanges in the past 12 months. The other 69% of exchange relationships featured one side or the other sending, but not receiving anything i n return. By location, 35% of the 262 exchange relationships with people living in other areas of Fiji were reciprocated, 9% of the 11 local area exchange relationships were reciprocated, and only 4% of the 28 overseas exchange relationships were recipr ocated.

PAGE 297

297 Characteristics of exchange partners 65% of exchange relationships were between heads of household and their children or siblings. The most common categories (relative to the male head of household) were So, Da, Br, Si, WiBr, and WiSi. The remainin g 35% of exchanges were carried out with a wide array of predominantly close kin, including the sons and majority were closely related in a genealogical sense or through marriage Figure 6 9 Relationships between exchange partners One notable as pect of these data is that a large proportion of exchanges are between people standing in cr oss relations, and in particular between village men and the Fijian cross relation, but also that marriage ties can be used as a means to access useful economic resources. I t is also noteworthy that little exchange was carried out with parents or cross or parallel aunts and uncles; perhaps this is because most

PAGE 298

298 relatives in the ascending generation are either already, deceased or share residence in the same village as the hea d of household Comparing exchange in Toba Cagi and Veidogo Given their different economic situations and access to transpor t, it might be expected that exchanges in the study villages would vary in frequency The median number of exchange partners and ex change e vents for each household diffe red significan tly between villages, but in a practical sense, the figures a re somewhat similar (Figure 6 10 ). Figure 6 10 Number of exchange partners and exchange events per household While the median total net m onetary value of the household exchanges did not differ significantly between the three villages, there were wide variations between households in the same village. In particular, several Veidogo households received large amounts of money from relatives wo rking in overseas countries. One retired Veidogo man with a son serving in the British army, cfor example, claimed to receive $FJ 1000 per month from his son. Meanwhile, the exchanges reported by Toba households resulted in a net overall loss for the villa ge.

PAGE 299

299 Figure 6 11 Value of exchange events and total income from exchange Table 6 6 Median test of significance of differences of exchange volume by village Variable Test N Chi square statistic Df Significance (2 tailed) # of remitters to village hh Median 99 10.13 2 .006 # of exchange events received/hh Median 99 7.527 2 .023 # of partners remitted to Median 99 5.40 2 .067 # of exchange events sent/hh Median 99 12.63 2 .002 Net hh income from exchange Median 99 2.294 2 .318 Figure 6 12 Net overall household income and net household income from exchanges

PAGE 300

300 Phone Use and E xchange One of the key questions of this chapter is whether mobile use bring s economic benefits via enhanced access to remittances. It is diffi cult to establish causality based on survey data alone, but in order for this hypothesis to be viable, high household remittance income (or at least a high volume of exchanges) should be associated with heavy phone use. Accordingly, I test the correlations between the phone frequency /intensity variab les introduced in Chapter 5, and household exchange and income variables. It turns out that none of these variable pairs are significantly correlated. Thus, i t seems that while phones facilitate exchange the amount of phone use by village rs is not correlated with the amount of exchange they engage in. Table 6 7 Correlations between phone use and income/exchange variables Phone use variable Exchange variable N Statistic Significance # of contacts saved in phone Value of exchange receip ts 170 .012 .876 # of contacts saved in phone Value of exchange outlays 170 .017 .823 # of contacts saved in phone Net exchange income 169 .022 .775 # of contacts saved in phone Absolute value of exchanges 170 .004 .962 # of contacts saved in phone # of exchange events (total) 170 .061 .429 # of contacts saved in phone # of exchange partners (total) 170 .120 .118 Calls+texts per day Value of exchange receipts 170 .001 .995 Calls+texts per day Value of exchange outlays 170 .009 .908 Calls+texts p er day Net exchange income 160 .006 .944 Calls+texts per day Absolute value of exchanges 170 .010 .899 Calls+texts per day # of exchange events (total) 170 .021 .788 Calls+texts per day # of exchange partners (total) 170 .008 .922 Phone credit used in a week Value of exchange receipts 170 .121 .127 Phone credit used in a week Value of exchange outlays 170 .071 .372 Phone credit used in a week Net exchange income 160 .060 .451 Phone credit used in a week Absolute value of exchanges 170 .119 .133 Pho ne credit used in a week # of exchange events (total) 170 .070 .376 Phone credit used in a week # of exchange partners (total) 170 .032 .688 # of other phones saved within Value of exchange receipts 170 .023 .765 # of other phones saved within Value of exchange outlays 170 .033 .671 # of other phones saved within Net exchange income 169 .001 .989 # of other phones saved within Absolute value of exchanges 170 .039 .616 # of other phones saved within # of exchange events (total) 170 .002 .977

PAGE 301

301 Table 6 7. Continued Phone use variable Exchange variable N Statistic Significance # of other phones saved within # of exchange partners (total) 170 .118 .127 % of surveys named as frequent phone user Value of exchange receipts 170 .105 .172 % of surve ys named as frequent phone user Value of exchange outlays 170 .093 .227 % of surveys named as frequent phone user Net exchange income 169 .032 .676 % of surveys named as frequent phone user Absolute value of exchanges 170 .131 .088 % of surveys named as frequent phone user # of exchange events (total) 170 .064 .409 % o f surveys named as frequent phone user # of exchange partners (total) 170 .080 .298 Composite phone use rank Value of exchange receipts 170 .007 .929 Composite phone use rank Value of e xchange outlays 170 .018 .818 Composite phone use rank Net exchange income 169 .018 .818 Composite phone use rank Absolute value of exchanges 170 .005 .954 Composite phone use rank # of exchange events (total) 170 .002 .980 Composite phone use rank # of exchange partners (total) 170 .131 .088 # of phones/hh Value of exchange receipts 170 .050 .518 # of phones/hh Value of exchange outlays 170 .037 .635 # of phones/hh Net exchange income 170 .020 .793 # of phones/hh Absolute value of exchanges 170 .0 52 .503 # of phones/hh # of exchange events (total) 170 .070 .362 # of phones/hh # of exchange partners (total) 170 .067 .384 Although overall phone use frequency/intensity does not correlate with high levels of exchange, perhaps villagers have partic ularly strong phone ties with those with whom they conduct exchanges. If so, this might indicate that more frequent contact with particular individuals leads to increased exchange between them On the whole, pairs of individuals who engage in exchanges tal k more frequently on the phone, on average, than pairs who do not exchange (a mean score of 3.03 vs. 2.44 on the scale from 0 However, most exchange pairs are also immediate relatives, so it is uncl ear whether or individuals.

PAGE 302

302 Figure 6 13 Comparison of calling frequency for exchange and non exchange relationships From my phone network data, I identified 268 calli ng relationships between villagers and individuals living outside the village who had sent them goods or money during the past 12 months. I identified 256 such calling relationships with the exchange of goods flowing in the opposite direction. Although ex change pairs tend to talk more frequently than non exchange pairs (Figure 6 13 ) among exchange partners more freque nt contact does not correlate with more frequent exchange, nor a greater total value of goods and money exchanged Table 6 8. Correlation s between calling frequency and exchange variables Calling frequency (01234) Exchange variable N Statistic Sig. # incoming exchange events from 268 .023 .709 Frequency of calling with Total v alue of incoming exchange events f 268 .042 .495 # outgoing exchange events to 256 .065 .300 Total value of outgoing exchange 25 6 .082 .193

PAGE 303

303 Exchange and I nequality among H ouseholds within R ural Fiji an C ommunities In the beginning of this chapter I presented data that indicated substantial community level income differences between Toba Cagi and Veidogo as well as considerable i ncome inequality among the households within these communities. A central question of this chapter, is whether long distance exchange (mediated through mobile phones) serves to ameliorate or exacerbate these inequalities. If wealthier households have bette r access to telephones, and telephones enable exchange, then pe rhaps wealthier households reap a disproportionate share of the benefits of exchange serving to increase levels of inequality in communities. Researchers have proposed various models for how r emittances a ffect income inequality within communities. Fundamentally, if lower income households receive a disproportionate share of remittances, then levels of inequality in the community should decrease, and if higher income households receive a disprop ortionate share of remittances, then income inequalities should increase. Lipton (1980) reasoned that re mittances typically compound existing disparities in rural communities, by disproportionately accruing to those households with the most favorable mig ration opportunities (i.e., the wealthy, well connected and well educated). ct on inequality depended on a opportunities a re diffused among households of unequal socioeconomic sta tus within the community. Massey, Goldring and Durand (1994) analyzed simultaneous data from 19 Mexican communities, surmising that over tim e migration opportunities tend to spread through various st rata of a community. Thus,

PAGE 304

304 tends to improve over time as migration opportunities become more widely available to the households in the community (Massey, et al. 1994; Star k, et al. 1986) It should be noted that much of the exchange that occurs in rural Fiji is reciprocated, with resources flowing both in and out of the village, rather than one way remittances. As I noted, the ethic underlying Fijian gift exchange is assis ting a kinsmen in his time of need Thus, local and rural urban exchange in Fiji tends to be more balanced than overseas exchange, because Fijians residing overseas are seen as more capable of assisting their rural kin in Fiji. Depending upon how the goods and money from exchanges accrue among community households, exchange may act to either increase or decrease levels of material inequality in the community. A standard measure for expressing the degree of inter household income inequality in a community is the Gini index (Stark, et al. 1988 ) By comparing the Gini index in each community for non exchange income, and for total income (including exchanges) we can determine whether exchange in that community servest to increase or decrease inequality. The res u lts are mixed: In Toba exchange has the effect of slightly increasing income inequality; in Veidogo exchange slightly decreases income inequality, and in Cagi there is virtually no difference. Table 6 9 Comparing inter household income inequality with a nd without exchanges Village Gini coefficient (hh income without exchanges) Gini coefficient (hh income including exchanges) Difference (Gini before exchange and after) Toba .322 .384 +.062 Cagi .388 .383 .005 Veidogo .451 .420 .031 Summary There ar e extensive differences in income both between Fijian communities, and between households within Fijian communities. These differences have real impacts on

PAGE 305

305 the lifestyle and well being of household members. Fijians engage in various forms of gift exchange, which, in principle, involve kinsmen providing material assistance to those who are in need. However, different households are unequally positioned in regard to their ability to access the resources of kinsmen. First, some households have more kinsmen tha n others. Second, some households have more affluent kinsmen, who have sufficient economic resources to share with their rural village kin. Third, and central to this chapter, some individuals may have better access to their kinsmen than others, via mobile phones or other ICTs. The phone serves as a tool that facilitates long distance exchange between kinsmen. If phone access is unevenly distributed within communities, and particularly if wealthier individuals have a greater ability to afford phones and pho ne credit, then the material benefits of exchange may also disproportionately accrue to the wealthy, exacerbating inequality. My data indicate that there are indeed broad differences in income between Toba Cagi and Veidogo as well as broad differences i n income between households in those communities. These income differences are correlated to household appliances. Furthermore, wealthier households in these communities have more telephones, and household members tend to score higher on several of the phone use frequency/intensity variables described in Chapter 5. It therefore seems that having more money (by whatever means) not only enhances material aspects of life in the househ old, but also leads to more phone communication relative to other community members.

PAGE 306

306 However, higher levels of phone use do not seem to be related to the amount, frequency, or monetary value of exchanges, nor the net income derived from exchanges. Furtherm ore among pairs involved in exchanges, more frequent phone contact does not correlate with more frequent exchanges, or a h igher total value of exchanges. This raises two possibilities. First, the ethic underlying Fijian kinship exchange is need. For a wea lthier household to be receiving more financial assistance than a poor household runs counter to this ethic. Therefore, the system has a sort of built in cultural homeostat that limits inequality. Second, although levels of phone access and use are not equ al between households, perhaps some critical threshold has been reached, in which everyone can access phones (if only through borrowing) sufficiently in order to arrange whatever exchanges are necessary. Despite the outcome of this analysis, it should not be inferred that mobile phones play an insignificant role in exchange. In fact, nowadays nearly every exchange in Fijian villages is arranged through mobile phones. Villagers pointed this out as one of the most important benefits of phones. The results of this analysi s mean simply that more phone use does not imply more exchange (or more valuable or profitable exchanges). Finally, the net impact of exchanges on income inequality in the study villages is unclear. In Toba exchanges served to increase inter ho usehold income inequality slightly, in Veidogo, to decrease inequality slightly, and in Cagi the effect was negligible. In any community these affects likely depend upon the specific migration histories of local households.

PAGE 307

307 CHAPTER 7 TECHNOLOGY, LOVE AND MARRIAGE IN RURAL FIJI Overview long distance relationship vi a phone and internet with a young woman named Lennay sequent death, were highly Lennay Kekua had never actually existed. Lennay Kekua was a hoax perpetrated by a female voice and persona over the phone and Internet. with the risk s and rewards of pursuing online relationships. An ever increasing proportion of our social interactions are mediated, partially or wholly, through communication technologies. Examples of this trend include the popularity of Facebook, and the tremendous r ise in online dating (Guadagno, et al. 2012; Rege 2009) The use of communication technologies to pursue romantic relationships is by no means limited to Western, industrialized countries. Fijians use mobile phones extensively for this purpose. Indeed, f or many Fijians romance is the aspect of life upon which mobiles have had the greatest impact and may even be the primary rationale for obtaining a mobile phone in the first place. Although my initial research focus in Fiji was on the economic aspects of mo bile phone use, I realized that a ny well rounded account would require a detour into the topic of romantic relationships

PAGE 308

308 The ability to communi cate instantaneously (and affordably) over a distance influences how people seek out romantic partners and how they behave within relationships. Communication technologies reshape relationships in two fundamental meeting new people, for sustaining long d istance relationships, and for diversifying or intensifying the forms of interaction within a relationship. Second the characteristics and behavior of romantic partners. In other wo rds, these technologies facilitate deception. Couples who have never met face to face often lack an in depth knowledge of their partner. Even basic characteristics such as age, physical appearance, relationship status, and occupation can be hard to verify thro ugh a phone call, online chat room, or dating website profile. Deception is in many ways related to opportunity. Individuals may lie about their identity or personal characteristics in order to appear more attractive to a partner. People may also use d eception to simultaneously explore relationships with multiple partners, without any of the partners knowing. Another facet of the deception involves the face to face relationships of the mobile/internet user. The privacy afforded by mobile and internet co mmunication relationships unbeknownst to close friends, spouses or children. Such infidelity, when eventually discovered, can seriously disrupt family and community life. In this chapter I us e the concepts of opportunity and deception to understand how Fijians use mobiles (and to a lesser, but growing extent, t he Internet) to seek and carry out

PAGE 309

309 these uses of mobi les, and how these behaviors fit in with existing social expectations and institutions surrounding romance and sexual behavior in Fiji, both within and outside of marriage. of mobile phones and internet in romantic relationships is in many w ays similar to how these technologies are being used in the United States and other Western countries. However, technologically mediated relationships have unique implications in the context of Fijian culture and institutions, and particularly in the cont ext of rural, kinship based communities in which practices such as online dating are unprecedented. The goal of this chapter is to determine how the use of communication technologies in relationships shapes the broader functioning of Fijian households and communities. Methods I did not systematically collect survey data about the prevalence of relationships initiated or sustained through mobile phones in Toba, Cagi, and Veidogo nor the prevalence of marital strife and divorce attributed to illicit use of m obiles in Fiji. However, I was able to record many stories that were told to me by rural villagers. I collected stories from men and women, middle aged people and youths, however, I naturally had better access to male youths and adults, and so stories rela ted to me by males are over represented in this chapter. I solicited some of these sto ries during interviews, but many others were related to me spontaneously during casual conversations in the village, or around the kava bowl. The frequency with which thi s topic was voluntarily raised, and the sheer number of individuals implicated in such stories, suggests that Fijians of both sexes and a broad

PAGE 310

310 range of ages use their mobile phones extensively to initiate and to maintain romantic relationships. Some of th ese stories concerned local events, while other stories took place in distant parts of Fiji. Many of the events related in this chapter are sensitive, and would be considered personally embarrassing by the individuals involved. I have subtly altered the de tails of the stories so that the actual individuals involved cannot be identified. However, the basic substance and cultural import of the stories remains unchanged. Opportunity Mobile phones facilitate romantic relationships in rural Fiji in several way s. First, mobiles enable communication at a distance, and allow people to contact each other much more frequently and affordably than they could in the recent past. For the owner of a mobile phone, carrying on a long term, long distance relationship become s a realistic possibility. Mobile phones broaden the potential pool of mates, and geographic distance becomes less of a consideration in selecting a partner. A young man and woman from distant islands, who may have met only briefly at a dance or a sports e vent, can maintain contact until the next opportunity to meet presents itself. facilitates the storage and exchange of names and phone numbers. Many youths maintain a mobile doss ier of phone numbers of members of the opposite sex that they can turn to at opportune times. If a Fijian youth breaks up with a girlfriend, or gets the opportunity to travel to Suva, he can scroll through his list of contacts, make some calls, and explore what new possibilities might open up. Third, although Fijian youths talk frequently about girls, in the actual presence of girls they can be very shy ( madua ). Based on evidence from interviews and as well as

PAGE 311

311 direct observation, it seems that communicatin g through the phone allows Fijian youths to overcome much of this shyness. This perhaps in part due to the medium (talking through the phone does not require looking each other in the face), and partly because youths can talk privately over the phone, away from other villagers, who might overhear the conversation and gossip, or otherwise embarrass the youth.The ability to flirt privately, at a distance through text messages and phone calls thus allow s some relationships to blossom which might never otherwi se have taken root There are three approaches to relationships that I will discuss in this chapter: formal marriage, which is carried out with the support and approval of kin and the Church, informal marriage (elopement), and casual sex, in which there i s no immediate intention of marriage. In the following sections I explain how communication technologies open up opportunities for Fijians and affect the course of relationships in each of these domains. Childhood and Youth As long as they are students ( g one vuli ), young Fijians are expected to devote themselves to their studies. They are subject to strict discipline from their parents at home, and from their teachers at school. In many households parents enforce nightly study after dinner and an early cur few. Students are not allowed to drink kava or to smoke, which excludes them from the nightly social gatherings of the youths around the kava bowl. When they finish secondary school, usually around the age of 16 19, Fijians join the ranks of the village y ouths ( cauravou (male) or gone yalewa (female)). Youths comprise a major social category within Fijian communities. Youths make important economic contributions through farming an d household labor, but have very limited

PAGE 312

312 input into community decision making processes which are dominated by the married heads of household. Youths have most of the same freedoms as married adults, but have few of the social responsibilities. Youth is therefore considered an exciting, enjoyable and carefree time of life. One of the prerogatives of youth is ownership of a mobile phone. Mobiles are a symbol of independence and autonomy, and are an important tool for youths to coordinate their active social lives. Male youths play rugby together in the afternoons after they return from the gardens, and drink kava together at night. Often they sleep away from home in the houses of their friends or in empty houses that are used as youth hangouts. Female youths help out with household chores and childcare, but similarly spend their fre e time socializing with friends attending local church youth events and dances Romantic relationships are one of the primary preoccupations of Fijia n youths. Opportunities to mingle and socialize with members of the opposite sex are a source of great an ticipation. Such events include weddings and funerals, sports tournaments, fundraisers and church youth groups. Youths carefully bathe and groom themselves and wear their best outfits to even seemingly mundane gatherings, in the hope that they might get lu cky or catch the eye of a romantic interest. Romantic relationships among youths may be informal, carried on without the knowledge or approval of parents and kin, or they may take the form of formal courtsh ip with the intent of marriage. Youth ultimately e nds with marriage, after which a Fijian is considered a full fledged adult ( tamata uabula ) and responsible member of the community.

PAGE 313

313 etwork One of the ways that mobiles create opportunity is that they are a convenient way to exchang e and store contact information. Fijians generally do not hesitate to ask new acquaintances for their phone number s it is an accepted way of acknowledging a social tie. Many youths maintain extensive lists of phone numbers of members of the opposite sex. For example, while I was in Cagi one of the young female schoolteachers three week stay in the village was a great source of excitement for the local youths who teased each other consta ntly about their failed attempts to attract her attention. Evidently many of them asked Marica for her phone number, because when I subsequently conducted veipasai ) among young men and women in the village who, in turn, try their luck with various partners. Epi, a Cagi youth, said that if he has been talking with a girl for a while, but loses interest in her ( me sa cegu na veitalanoa ya ), he will pass her number on to one of his other male friends so that he can give it a try. On the other side of the coin, Pita said tha t he sometimes gets a call from an unfamiliar girl, and upon asking her from whom she acquired his number, she names one of his other phone first girl to whom he was previously talking has lost in terest in him and has moved on to better things Most village youths with mobile phones seem to be carrying out at least one phone based relationship, and some youths carry on several relationships at once. A young man and woman living in distant villages might call each other on a weekly basis,

PAGE 314

314 whenever one of them manages to scrape together money to buy phone credit. These relationships are for the most part low key, with low expectations. The young man and woman have vague intentions to meet someday, but no specific plan. As one youth put keirau se veitalanoa tu ga ). Often the relationship peters out as one of the partners loses interest, loses their phone, or moves on to a more serious and substantive relationship. At tim es, phone based courtship in Fiji can resemble a feeding frenzy. Amelia, a young woman in a village near Suva, called into a radio station one day to participate in one of the many on air promotional contests. As part of the contest she had to sing a popul ar Fijian song on the air. She won the contest, and the DJ asked her for her phone number so that she could claim her prize. In the process, the DJ accidentally broadcast her phone number live over the radio. Within minutes, Amelia said, she was receiving calls from men all over Fiji, who told her that they had been enchanted by her voice. Because youths use phones to such an extent to pursue relationships, they tend to be more sensitive than married adults about the privacy of their texts or showing others the contents of their phone address books. Such snooping is common occasionally if I carelessly left my phone around the house, members of my host families would inform me about the contents of text messages I had received during my absence. One young wo man stores her SIM cards in her bra, both for convenience and security. A young man in Cagi sleeps with his mobile inside his pillowcase at night, lest a family member try to use it or look through his text messages. Many village youths set passwords to lo ck their SIMs, so that neighbors or parents cannot casually scroll through the contents of their phone.

PAGE 315

315 No Such Thing as a Wrong N umber When we misdial a call in the United States we usually just apologize and immediately hang up. A misdialed call in Fij i, at the bare minimum, leads to a conversation about the respective locations of the two callers and an account of the weather in each place. Sometimes it can lead to much more. Adriu, a village youth, told me one day he received a misdialed call from an Indo Fijian woman who lives in Australia. The woman had been trying to call an Indian relative in Navua (another part of Fiji). Instead of hanging up, Adriu and the Indian woman started chatting and became Another woman had several contacts saved in her p hone that were labeled simply with the names of Fijian provinces. When I asked her about these contacts, she replied that these were peopl e from various parts of Fiji who had misdialed her phone number. She had struck up conversations with them, and saved their numbers in her phone so that if someday she should happen to need assistance in a given province of Fiji, she wo uld have a friend th ere to call. One of the most popular phone p romotions in Fiji is FreeCall in which the p hone user pay a small fee ( $FJ 1.50) to obtain 24 hours of free in network calling. This promotion is ideal way to make lots of phone calls without paying the normal r ates Many Fijians ad mitted that when they have Free Call and have already run through their entire list of phone contacts, they simply dial random numbers in the hope of striking up a new friendship. In particular, if a member of the opposite sex picks up the phone, it is a great opportunity to flirt and joke. At the time of research, s everal of my informants

PAGE 316

316 were currently carrying on phone based relationships with people they met in this haphazard manner. Mobiles and Mobility Many Fijian youths take advan tage of their freedom and relative lack of responsibilities, and travel to other parts of Fiji. The only resources necessary to gade wavoki (take a trip) are money for boat or bus fare, and a kinsman or friend willing to take one in as a guest. Youths are homes by helping out in the gardens and the household chores. Itinerant youths may therefore be welcome to stay for weeks or even months among their hosts. Mobiles are valuable companions on such ad ventures: mobiles allow youths to call up existing friends to announce their arrival, and to network and maintain contact with the new friends they meet in the areas that they visit. Such travels are a means by which mobile numbers are traded far and wide among youths for the purpose of making me rawa na friends ) in distant parts of Fiji. For example, Josefa, a youth from the Viti Levu highlands, is able to travel extensively around Fiji using money that he earns from planting and selling kava. His most recent journey was to Batiki, where a female cousin of his was making her kau mata ni gone ). As Josefa and I sat on the Levuka seawall and waited for the ferry to Suva, he showed me pictures of one o f I complimented him. We met through the telephone. I visit her sometimes in Vanua Levu, or we meet when she comes to

PAGE 317

317 conspiratorially Josefa has multiple girlfriends (one of whom is married) living in his highland village, and also a girl in Lau. Josefa met his Lauan girlfriend through adept social networking: a La uan friend instructed him that if he should meet with a good looking girl there, to give her his phone suitable girl, and she and J osefa began talking over the phone. They called each other back and forth for three years before finally meeting. When we finally met in Suva, oh, it was on sara ga Josefa clapped his hands enthusiastically at the memory of their fir st meeting that entire time, did you e No, he said, but we fel l in love through the telephone ( veidomoni ena talevoni ) ranging relationships are facilitated by his ability to travel, and to make new connections through the phone. His ability to generate disposable income from kava sales certainly helps, but ultimately his freewheeling lifestyle hinge s on the connectedness and mobility enabled by buses, ferries, mobile phones, and the inte rnet. Often youths are phoneless when they depart on their journeys, but return home months later with a second hand phone given to them by a relative, or by a new found boyfriend or girlfriend hoping to stay in touch and keep the relationship alive. One T oba Lomaiviti Provincial festival a few weeks ago. I was sitting under a vakatunuloa (arbor/festival stall) when my female cousin walked up and handed me this mobile phone She said the phone belonged to a friend of hers who wanted to meet me, and that the friend wanted me to have this phone so that we can call each other. I called the

PAGE 318

318 girl up using the phone and we met a few times while I was in Suva. A few days later I ha d to come back to the village on the boat, but my new girlfriend and I have been keeping in touch by phone ever since. My eventual plan is to marry her and bring her back to Toba e se bera ni vakadeitaki ). When I ask e neirau ga, e sega ni ka nodra ). Spy Games at the Suva Bus S tand Phone based relationships can go on for months or ev en years. Travel within Fiji co nvenient opportunity to meet in person. For most rural villagers, the ideal place for such meetings is Suva. Not only is Suva the capital and b usiness hub of Fiji, it is also a large enough city that Fijians can escape from the watchful eyes of village kin. The best place for a rendezvous is the Suva bus stand. Bustling with villagers hauling bags of produce to the market day laborers commuting to their urban jobs, and hawkers selling pineapple and Indian sweets, the Suva bus stand offers anonymity among the throngs. Eroni, a Toba youth, met with a girl he had been talking to over the phone when e called each other back and forth for about a week. Finally one day she told me to meet her at the bus stand at 5:30 after she got off work, and told me what she would be wearing so that I could recognize her. I got to the bus stand early and it was full of people. I hid myself in a place out of view. I she told me, and right at that moment I saw her. We were both standing in the middle of the street, talking to each other on the phone So

PAGE 319

319 sota ena talevoni ) in every sense of the Amelia, the young woman whose number was accidentally broadcast over the radio, eventually arranged to meet u p with several of the male admirers who called her after her singing performance. Each time she arr ived at the Suva bus stand well in advance, in order to stake out a hiding place from which she could secretly observe the man and see what he looked like. T hat way, she could decide whether she wanted to actually meet him face to face ( sota va mata ). If the man was old or ugly, she just (ran away without saying hello). Amelia eventually eloped with one of these men from the radio contest, with whom she corresponded by phone for a full thre e years before finally meeting. Amelia reflected on how her life has changed since getting her first mobile in 2004. Before getting a phone, she and another single female cousin used to go out a lot to the bars in Suva to meet men. After getting her phone, she no longer felt the need to go out to bars she could just use her phone to meet new people. Amelia therefore feels that her mobile phone has played an important role in broadening her social horizons while allowing her to avoid the unhealthy atmosphere of bars Courtship Several of the young people I interviewed were earnestly pursuing their phone based relationships with the intent of marrying. Ruci, a young single mother in Cagi described to me he r rel n February of this year my boyfriend and I started talking. tabatabaka na naba ). He introduced himself, and we proceeded from there. He asked if I was single ( galala vosa

PAGE 320

320 va veidomoni, vaya ( vinaka tik o ga na yalo ). We talk over the phone about the Bible, Maria, Jesus, Divine Even though also sent each o ther pictures over the phone i My boyfriend lives on Taveuni ( an island near Vanua Levu ). He was supposed to come to Cagi planning for me to go to Tav euni and go see him on Christmas, and after that he will come to Cagi to ask my father for my hand in marriage. ( Lako mai lakovi au o koya vei T a ). Anyway I prefer that my boyfriend and I meet first somewhere else bec ause the people in this village gossip said I could just go and stay in his village for good; but my father would not allow this officially and then I says Taveuni is too far away ; he would prefer that I get married to someone in Cagi so that I would be close by to them after I get married, but I told my sa noqu digidigi ). Because postmarital residence in Fiji is usually virilocal, some Fijian parents seek to influence choice of spouse because for either emotional or economic reasons they do away. In this sense mobile phones can empower young women to pursue their independent desires regarding marriage.

PAGE 321

321 Meeting and marrying through the mobile phone may also provide an avenue of escape for young people in unhappy or oppressive situations especially for young women who envision start ing a new life by marrying into another village. One Methodist minister was reluctantly recruited to preside over the wedding of a couple who had met through the phone. The minister was reluctant because he said that the couple was mismatched ( veicalati ) the boy was unattractive ( very black ), while the girl was rumored to be quite pretty. They had met over the phone and after only one month of talking back and forth, sight unseen, the young couple had agreed to marry. The minister explained that the girl was the daughter of a Pentecostal minister. She therefore had a very strict upbringing, in which typical youth recreations like kava drinking and dancing were off limits. The minister conjectured that the girl was unworldly, because of her sheltered upbringing, but also might be willing to overlook the shortcomings of her unattractive fiance because she wanted so desperately to escape from under the thumb of her oppressive father. The minister had to accompany the veicurumi a ceremonial marriage proposal that great shame ( madua Mismatched looks notwithstanding Mobile s and Elopement: A Means of E scape? A Fijian wedding ( va kamau ) is a costly, complicated, and drawn out affair. The such a wedding is prestigious, but costly in terms of the

PAGE 322

322 economic resources used, the social obligations incurred, and the length of time required to plan and to carry out the event. Many young Fijians are unable to marshal the resources for such a weddi ng, or are simply too impatient to wait. together ( dro marriage which Fijians call vakatevoro (married in the way of the devil) a union which has not received legal, ceremonial, and religious sanction. Traditionally, eloping with a woman without obtaining the consent of her kin is considered a form of theft ( butako ). After a suitable period to let eventually extend a traditional apology ( bulubulu Unbeknownst to me, an elopement was being plotted via secret phone calls as I was carryout out my field work in Cagi Maraia, a 20 year old girl in the village, had been talking to a young man in Suva over the phone for the past 6 months. The girl had kept her relationship secret from everyone in the village except for one close male cousin. I did not have a chance to interview Maraia, but according to other sources in the village, she had grown to resent her life at home. Her unmarried elder sister had 3 young children, and Maraia was often enlisted to do the housework and to take care of the children The m an Maraia was talking to over the phone represented, among other things, a means to escape this situation. Eventually an opportunity arose for the lovers anged to pick her up at the Suva wharf. During her weeklong stay in the city, they secretly decided to marry. Maraia returned to the village for just a few days then once again boarded the late night ferry to Suva without telling anyone except

PAGE 323

323 her truste in the morning that she was gone, but when she called him a few days later he forgave her for running away In accordance with custom, t tooth a nd other traditional valuables, to daughter and to make the marriage official The Emergence of I nternet Internet use among rural Fijians is still uncommon, but quickly gaining popularity. Most v illagers I i nterviewed used internet exclusively on their mobile phone, but some had had previous exposure at an internet caf in Suva. Villagers still have limited knowledge of how to use search engines or other internet functions; many exclusively use the internet t o access Facebook. In fact for many villagers Facebook is virtually synonymous with the Internet: i n response to my interview questions regarding internet use, Fijians are increasingly usi ng Facebook to keep in touch with relatives and to meet new friends and romantic interests I spoke with several young Fijian women who chat with men overseas via Facebook. One girl said that she enjoys chatting with men over Facebook, but gets worried bec ause sometimes they ask for her phone number. One time an American soldier in Afghanistan called her home in the village. Her father answered the phone, which created a very embarrassing situation. Two rural village girls used Facebook to chat with a coupl e of Ghanaian engineers, who were about to move to Fiji on a job contract. The Ghanaians were looking for potential Fijian girlfriends arrival in Fiji the village girls went to live with them in Suva for s long however, and the girls ended up returning to the village, where one of them is now chatting via

PAGE 324

324 Facebook with a young African American man in the U.S. As she described it, their conversations (carried out over Facebook) mostly consist of flirtations and sexual fantasizing; she has invited the young man to come to Fiji and he told her that he is currently saving up for a plane ticket. Deception Aside from opening up new opportunities for relationships, com munication technologies also facilitate deception. These deceptions have two main aspects. One aspect is that it is much easier for individuals to misrepresent themselves to a romantic partner when communicating over the phone or i nternet, than it is in pe rson. Dishonest individuals can lie about t heir personal characteristics, habits or activities and it is difficult for the partner on the other end of the line to verify their claims. The second aspect of the deception is that the people with whom individ uals are surrounded in their everyday lives parents, spouses, children, neighbors have limited ability to oversee the activities in which a family member is engaged over the phone or Internet. Some ver the phone and internet unbeknownst to spouses family members and neighbors. Such relationship s, due to their purely mediated nature, can seem di seemingly not subject to the same norms or ethical standards that guide face to fa ce interactions. Expectations and R eality One of the most common sorts of deception in Fiji an mobile based relationships i s intentional misrepresentation of personal characteristics After a lengthy telephone based courtship, the two expectant lovers fina lly meet in person, only to be disappointed that the person with whom they are standing face to face does not even remotely resemble the description provided over the phone.

PAGE 325

325 The discrepancy between expectation and reality can be shocking. One village yout h had carried out a relationship over the phone with a supposedly young woman. He invited her to come to the village to meet his family, at which point he planned on broaching the topic of marriage. The woman arrived in the village, and much to the villager), and as he later learned, she already had several children in Suva. The youth and his family promptly sent the woman packing, but the story was an ongoing source of embar rassment. Another story was told to me as a way of teasing Vatu, a man from a neighboring had conducted a months long relationship over the phone with a woman in Taveun i. The relationship had reached the point where his son was regularly sending money to his girlfriend via the post office. Vatu became disgusted with this state of affairs and told his son to go on and consummate the relationship already, before he wasted any more t even married. The son agreed and bought a ferry ticket. As the ferry approached the shore of Taveuni called the girlfriend and she waved from the dock. Her face materialized in the distance an d the young man realized instantly that he had made a big mistake. He frantically called his father as the ferry approached the dock, ped from the uncomfortable situation as quickly as he could, by fleeing Taveuni on the next ferry to Suva, where he has been staying ever since.

PAGE 326

326 Another tale was set in Suva, and was related to me by a friend of the protagonist. In this case, Samu, a bache lor in his 40s, was searching through Facebook and found the profile of a girl, aged 17. She had posted pictures of herself, as well as and calling her. He lied about his age, but because their phone interactions were going very well he did not want to break off the relationship or reveal that he was much older than she was Finally they agreed to meet one day at the Suva bus stand. Samu called and told the girl that he wa s standing outside the MH supermarket, (even though he was really beside the police post), so that he could observe her without being seen. He saw teller of the stor y embellished), that Samu fle d the scene out of shame With a heavy of his, who was only 16 years old. He instructed the young man to adopt his name (Samu), and filled h im in on some of their past conversations and jokes so that the girl would not realize that the switch had occurred. The ruse actually worked, and after a few phone conversations the young boy and girl arranged to meet. They quickly consummated their relat ionship and within a year, they had a child together. At that point the ir deception finally came to light, however, because when the marriage papers were signed the girl learned that her l she had been led to believe. Other stories, rather than overt deception, merely involve disappointed expectations. One youth on Taveuni started talking over the phone with a girl from

PAGE 327

327 Yasawa (in Western Fij i). They got along well in their phone conversations, and the Taveuni youth thought about pursuing the relationship. However, he assumed that because the girl was from Yasawa that she would not be good looking (this is a regional stereotype held by some Fi jians ). So he broke off their blossoming relationship but did village friend also hit it off with the young woman from Yasawa. After a brief phone courtship, his fr travel from Yasawa. She arrived in the villa ge on Taveuni, and to the dismay of her first phone boyfriend (the one who had rejected her), the Yasawa girl was extremely attractive. He cou ld only look on with jealousy at the happiness of his friend and his beautiful wife to be. Mobiles and Marital S trife When I first arrived in Cagi and tried to explain my project to study the use of mobile phones, I was greeted with solemn pronouncements f rom village elders about the evils of the misuse of mobile phones. Unbeknownst to me, my arrival had coincided with a scandal in a neighboring village. A married woman the re had been carrying out an affair over the phone with a man on Viti Levu. She secret ly sold her possessions to get money for a boat ticket, and then unannounced, abandoned her husband and 4 children to go live with her lover. Her husband was determined to get her back though, and enlisted some policemen to accompany him to the house where she was staying with her lover. Ultimately the errant wife was forced by the police to return to the village with her husband. Villagers assigned a heavy share of blame in the situation to the mobile phone, which is considered particularly dangerous in th e hands of unfaithful housewives.

PAGE 328

328 In Toba in the previous year there had been an incident where a youth and a married woman from the neighboring village had engaged in an affair, using mobile phones to arrange trysts in the woods. They were eventually disc overed, and after a period of strained relations between the villages, the problem was eventually resolved through a formal apology ( kena veivosoti ). Local villagers were unanimous in the belief that such a relationship could never have happened without th e facilitation of mobile phones, because the lovers would have been discovered immediately. Another story I heard suggested that sometimes even the mobile phone allow cheaters to evade justice A woman in Western Viti Levu had a job working with Dig icel. Her husband was a carpenter, and had a contract to do some work on a house with a stay at home wife. The carpenter and the housewife began to have an affair. arrange romantic liaisons to suspect something, and through her job at Digicel was able to pull up the text records of her husband. She printed these out and brought them to the home of the other husba nd (whose wife was cheating with the carpenter). She showed him the text records and explained to him that she thought their spouses were having an affair. This revelation led to the dissolution of both marriages. However, the story did have a surprise hap py ending: the cheating couple ended up marrying each other, but so did the husband and the Digicel worker who had been cheated on. In general I was surprised at the forthrightness with which many of my respondents admitted their own marital issues related to use of mobile phones. One of

PAGE 329

329 through the mobile phone. Another woman said that her husband does not store any of his phone contacts in the address book, and always del etes the numbers he dials from woman said that one of the major marital problems they have with the phone is that ngs to her when she picks up the phone, so that her husband is led to suspect her of being unfaithful. This can lead to accusations and household arguments. The acc ounts I provided in this section focus on opportunistic and deceptive uses of mobile phones. I do not mean to imply that all, or even most phone or internet based relationships in Fiji involve deceit. Aside from a generally negative, reactionary discourse about technology, many Fijians see mobiles as a real boon to their personal lives. Mele, a Batiki youth, illustrates how mobiles can strengthen committed relationships. He contacts his girlfriend, a nurse on a neighboring island, twice every day. Apete, a young widower, is engaged to be remarried with a woman in a nearby village. He says that t he mobile phone has been a great help in maintaining their Impacts on Family and Community L ife Mobile phones provide Fijian villagers with unprecedented freedoms and possibilities in the realm of romance as well as exposure to new forms of risk and deception Despite the positive connotations of personal liberation, Fijians take a very ambivalent view of the use of mobi les to seek out ro mantic and sexual partners. Individual sexual freedom often conflicts with social harmony and established norms surrounding sex and rel ationships. Mobile phones are widely blamed for a perceived

PAGE 330

330 increase in premarital and extramarital sex in Fiji, along with the associated problems of teen pregnancy and marital strife. Paradoxically, many of my Fijian informants publicly condemned these social problems, while privately confiding the personal fulfillment and excitement they derive from thei r own mobile based romantic relationships. The question is whether mobiles have qualitatively changed Fijian society through their uses in romantic relationships. Do mobile phones simply facilitate behaviors that previously existed, or do they fundamental ly change the way that relationships are experienced and the way Fijians approach courtship, sex and marriage? The answer to this question may be found by comparing current trends against the ethnographic literature on sex and marriage in Fiji. Norms and Realities Surrounding S ex in Fiji T oday Most Fijians are quite religious, attend church more than once a week, earnestly study Scripture, and take the word of the Bible seriously, if not literally. Fijian sexual norms are heavily shaped by the Christian id eal of monogamous marriage: premarital and e xtramari tal sex are considered Biblical sin s ( valavala ca ) Much of the Fijian public discourse and hand wringing surrounding extramarital sex is accordingly framed in religious terms. For example, a 2008 story i n the Fiji Times quoted a 20 year old Fijian woman exhorting her peers at a Church conference: (if you practice premarital sex) y ou won't be in God' s favour and you will be cursed ( Fiji Times 2008) As in most so cieties, however, Fijian attitudes and behaviors surrounding sex are complex and often contradictory. Actual sexual behavior does not necessarily conform to Biblical canon or cultural norms. There is a spectrum of illicit relationships, which if discovered may result in varying degrees of disapproval or punishment. For example, prostitution is common in the urban areas of Fiji, and occasionally public (and

PAGE 331

331 ecclesiastical) outrage leads to a campaign to clean up the streets C asual sex also occurs surreptiti ously among villagers and secondary students In 2009 the Fijian Ministry of Education released a statement tracing teenage pregnancy to school the inevitable to eventuate girls fall pregnant during the second term holidays because of the festivals and other ( Fiji Times 2009) The prevalence of premarital sex and the potential instability of marriage are evidenced by the presence of more than a few unwed mothers in Fijian villages. There are various paths to single motherhood. An elopement might no t end in stable married life, and the woman subsequently returns to her kin, either pregnant or with young children in tow. An official marriage, or unofficial vakatevoro union might end in separation, in which case the woman often returns to her kin along with her children. Finally, women (usually young women) who are impregnated as a result of casual sex often live with parents or other male kin as a means of sustaining themselves and their young children. Love and M arriage Mobiles allow Fijians to purs ue relationships while escaping the constant surveillance and gossip that normally characterizes village life. As one elderly informant house because you were afraid o rumors Today, text messages slip silently past the eyes and ears of spouses, family members and neighbors, even when they are sittin g in the very sam Text

PAGE 332

332 messaging enables lovers to secretly arrange to meet each other at a time and place where they will not be discovered. How ever, the ethnographies from the early to mid 1900s hint that Fijian lovers foun d creative ways to hook up together long before mobile phones and text messaging existed. Thompson provided a detailed description of communication and interactions mor ning church services by means of covert signals across the aisle which separates the men from the women. beloved is indoors at night, the lover w ill pass near the house and cough laugh, break a branch, or talk as if someone were with him. Or he may enter her house when all is dark and make a noise like a rat. A man or a girl occasionally has a go between, who may be of either sex. He is chosen for his loyalty, but receives a small gift so that he will no t 49) ngande a term which when used in this sense means to take a walk in order to have sexual intercourse). If he is repeatedly repulsed he occasiona lly resorts to a socially condemned practice called ndarandara He enters the house of his beloved at night and attempts to s according to one informant, she talks nicely to him, says a tabu word in fun, and asks him for tobacco. Then she can see whether or not he likes her. If the man does not care for the girl and she desires him, she tries to arouse him by making lascivious movements or by approa 1940a: 50).

PAGE 333

333 Moonlight, which is very brilliant in Lau, is naturally disliked by lovers Occasionally late in the evening my house would suddenly become crowded with young people. It took me some time to discover that this inf lux of guests corresponded have intercourse at night under the palm trees fringing the beach or in an empty house in the village. If a house is already occupied they try to find another place, perhaps the shadow of a canoe. Occasionally they meet during the day in the bush. Married people have connection in the sl 50). young mbure 1929: 1929: 155). emphasis on chastity has tended to break down in relation to the degree of Ton gan and a : 48) cation) ( 1929: man commits ad it is not so serious if she is the wife of a tabu on sexual intercourse between cross cousins, and, in spite of the ideals of premarital chastity for girls and of strict monogamy, there is considerable freedom between classificatory cross cousins both

PAGE 334

334 before and after marriage. In order to avoid gossip and marital jealousy, an affair with a cross on 1940b: 58). Rules against sex between forbidden kinship categories seem to have been more strictly observed than marital vows between cross cousins : according to relationsh one infringement of the brother imprisoned by the colonial government. Informants agree that in olden days the offenders would have bee n exiled from the island (Thompson 1940b: 60) ( vakatevoro ) is likewise not a new phenomenon in Fiji. love ma tches and runaway (1929: 156), while Sahlins found that about 1 0% of conjugal units on Moala in 1955 were living in unofficial vakatevoro unions (Sahlins 1962) Thompson wrote that if a pecially of what abated. At first they are berated in the village, but gossip soon dies away. When the pair return to the village they are accepted as man and wife. A wedding fea 53) the other hand, it was formerly considered a good thing to have bastards, if the father 1929: numero 55). Records on Moala for 1947 54 indicated that nearly 15% of children were born outside of legal wedlock. Sahlins speculated that most of these

PAGE 335

335 ha ve been born of casual sex (Sahlins 1962:93) Thompson further described how couples who gave birth to a child were supposed to refrain from sex until their child was weaned, but that the husband usually has clandestine affairs with other women during this time of enforced abstinence (Thompson 1940a: 33) Separation and D ivorce Until recently legal divorce in Fiji was difficult, because either the husband or wife had to prove wrongdoing on the part of their spouse. The most common legal grounds for divorce were 2 years desert ion, 5 years separation, and cruelty. Hence, unhappy marriages often simply ended in permanent separation. In 2003 Fiji adopted no fault divorce laws with the only precondition for divorce being that the marriage has broken down irretrievably (Taylor 2003 ) This makes actual legal divorce a more viable option rather than just resorting to separation. The ethnographic record presents mixed opinions on the frequency of divorce in colonial and pre colonial Fiji and exact figures are lacking. Hocart judged t people. If a woman marries into another clan, then runs away to her own clan, the 1929: 158 9). According to T house. Then the people will talk about the husband If he wants her back he presents a re 1940b:

PAGE 336

336 common Arno 1992:4). Thus it seems that in rural Fiji in the mid 20 th century at least, marital strife and temporary separations occurred, but rarely resulted in a permanent split. Summary While technologies are used in culturally specific ways to attain cultu rally specific goals, the use of mobiles to pursue romantic relationships seems to be a global phenomenon. Studies ranging from urban Jamaica to rural Papua New Guinea, to the Sudan, suggest that in this regard mobiles are used in similar ways (Horst and M iller 2007:82; ( 2011) In some societies this behavior has given rise example, in the Sudan there are widespread fears of increasing indecency and immorality as a result of mobile phone use (Brinkman, et al. 2009) In the United States, (Mitchell, et al. 2012) Many older Fijians I interview ed were quite earnestly concerned that phones enable the corruption of Fijian youth, as well as leading to marital strife and divorce. As documented by Thompson, Hocart and Sahlins, Fijians in the pre mobile era engaged in similar pursuits, although secre t communication between lovers r equired more tact, creativity, and luck Instead of a text message, lovers gave secret signals in church, or through a trusted go between (Thompson 1940) I have even heard verbal accounts of Fij ian villagers of yesteryear using messages in bottles to make contact with friends in distant villages. In recent decades, some intrepid villagers managed to pursue relationships through the radio telephone, despit e its lack of privacy. One man described h ow he carried on an affair with a woman by radio phone by calling in the middle of the night wh en nobody was around. A popular

PAGE 337

337 over the telephone). Thus it seems that Fijians have always enthusiastically embraced whatever means of communication are at their disposal. The extensive use of mobiles in the present day to initiate relationships, and to arrange romantic liaisons, may be considered the fulfillment of a gener al desire that had always existed, but was often impossible to fulfill without the privacy and secrecy use of new technologies to fulfill pre existing, but previous ly unattainable desires (2000). Based on the evidence presented in this chapter, it is probably inaccurate to say that mobile phones have re volutionized romance and sex or qualitatively changed social relations in Fijian society. However, the greater eas e of communication might mean that a greater proportion of youths will engage in premarital sex, and married men and women in extramarital sex. Certainly too, mobile p hones have increased the viability of long distance relationships. The resulting inter is land marital ties can forge new relations between villages and kin groups that previously had little contact. With the increasing use of mobile internet, these long distance relationships may increasingly occur with Fijians or foreigners living overseas. Many Fijians invest a high degree of personal effort and expenditure into their phone based relationships because it allows them to transcend a village lifestyle which is circumscribed by limited mobility and heavy kinship obligations. Such transnational relationships will likely serve to accelerate processes of global integration in Fiji and elsewhere.

PAGE 338

338 CHAPTER 8 CONCLUSION S In this dissertation I have assessed of the various implications of mobile technology (and to a lesser extent, internet and other IC Ts) in rural Fijian communities. I n this conclusion I synthesize my findings in Fiji, and discuss their significance in the broader context of processes of globalization and technological change Economics I initially gravitated toward the topic of mobile phones in terms of their role in economic development Mobile phones are not (yet) universally owned or uniformly distributed in Fijian communities, nor can all Fijian phone users afford to use their phones as much as they would like to. Howe ver, there is extensive lending and borrowing of phones in rural Fiji which reduces the impacts and disadvantages of phonelessness. FreeCall promotions which allow free calling during a span of 24 hours, greatly facilitate borrowing because they allow calls to be made at no additional cost to the phone owner. From an economic standpoint phone ownership, and heavy phone use are associated with high household income. However, heavy phone use (at least in Toba, Cagi and Veidogo) does not bring disproportionate economic benefits through exchange. This seems paradoxical, because my observations, and the statements of villagers, make clear that phones play a key role in arranging remittance transactions. The explanation lies in the fact that exchange in Fiji is p rimarily based on need, by which logic the economic benefits of exchanges s hould accrue to the needy. Also, phone access has become sufficiently universal in Fiji that, through borrowing or otherwise, anyone can gain access to a phone when they really need to even those

PAGE 339

3 39 without phones are able to borrow phones in order to make critical requests for assistance In fact, in small communities such as Toba, Cagi and Veidogo, many of the economic benefits of m obiles accrue at the community, rather than individ ual or household level. It only takes one phone call (from anyone in the community) to arrange emergency transport, or to find out a key piece of information that later circulates throughout the village by word of mouth. Similarly, an important remittance transaction mobile phones (at least at near universal ownership levels) seem to provide community wide economic benefits, and do not seem to contribute in an immediate sen se to inter household income inequality. Given that many of the economic benefits of connectedness accrue to the community at large, then communities with limited or no mobile access may suffer relative to well connected communities. I saw indications of this when comparing conditions in Toba and Veidogo. Intermittent or weak network coverage has a detrimental affect on the smo oth functioning of all sorts of processes. Even more important than the quality of network coverage is the availability of affordab le, reliable arrange for the transport of actual, material goods. In the absence of such transport, the potential economic benefits of mobile access are greatly reduced The force of mobile phones as a social status symbol is also much reduced in small, rural communities. In the Fijian context, material wealth is not as important an indicator of status as it is in urban, market based societies. A majority of mobile owne rs

PAGE 340

340 in the study communities received their phones as gifts, diminishing any connection between wealth and phone ownership. Furthermore, in small communities everyone knows ever yone else so well that they tend to judg e each other on more fundamental merits. Social life As described in Chapter 2, Fijian villages have a hi gh degree of structure: every individual occupies a specific position within a hierarchy o f descent groups and a web of kin relations, and a specific role defined by sex, age, rank and marita l status. Interpersonal behavior is heavily shaped by norms of interaction associated with kin relations, as well as the strictures on interaction based on unequal rank, seniority and sex. ocentric phone networks display a surprising amount of social diversity. T rue, t here are tendencies for young, single people to have other young, single people in their phones (and for older, married people to have other older, married people) but generall y speaking egocentric networks contain a mixture of ages, sexes, and kin ship degrees. This can be explained in part by the mix of instrumental and social purposes for which the phone is used. kin in the same generation (whom they contacted with disproportionate frequency) and a preponderance of parallel kin in the 1 st ascending/descending generation. Given the distinct roles of cross and parallel kin, it would seem that Fijians predominantly contact age peer s for social purposes, and to a lesser degree contact close parallel kin, probably for more instrumental purposes associated with obligations within the household or lineage

PAGE 341

341 Fijian commu nication over the phone broadly resemble s the forms and styles of fa ce to face communication. However, there is also communication, in that not being co seems to enable Fijians to overcome the feelings of embarrassment or hesitation associated with taboo kin, and for youths the shyness of interacting with members of the opposite sex. This contributes to diminishing the force of social restrictions associated with rank, kinship, or sex. Village affiliation is an important aspect of identity for Fijians. As an increasing proportion of social interaction is carried out at a distance, the community associated network or field of communication, rather than a geogr aphic location. P hones play an important role in overcoming distance and bridging the gap between rural and urban based members of the village community involvement in village life was usually limited to occasional urb an fundraisers, and infrequent visits to the village, perhaps each Christmas. Near universal mobile phone ownership has meant that the everyday occurences of life in both the village and the city are communicated in real time. In a general sense this may s trengthen community bonds. However, it may have detrimental consequences in terms of conflict resolution. within the village and resolved through conversation (Arno 1992) or formal face to face apology ( isoro ) now become grist for the rural/urban rumor mill. shyness, partly explain why mobile phones are so widely used in the domain of

PAGE 342

342 romantic relationships. For many individuals, the grea test impact of mobile phones has been through the r omantic relationships that the mobile phone has enabled (or perhaps, contributed to the dissolution of). These relationships are usually contracted on an individual to individual level, and often reach an advanced stage without the knowledge of close kin, including parents. This brings me to another fundamental impact of mobile phones on Fijian social life: privacy. This is ironic, perhaps, since one of the primary concerns about information and communica tion technologies in Western societies has been increased surveillance of the public. Life in small communities has always been characterized by pervasive nosiness and the near impossibility of privacy or casual nonconformity with social norms Everyone knows nearly everything about the act ivities of everyone else. Surveillance and gossip a re key mechanism s through which community norms are enforced. Aberrant behavior of individuals is detected, swiftly broadcast throughout the village, and dealt with eit her through shunning, ridicule or punishment. The mobile phone provides an oasis of privacy for individuals with unconventional lifestyles, habits or views, and for people to engage in (possibly illicit) social interactions without public knowledge. Thus, m obile phones subvert instututions, developed over millennia, through which norms are applied and enforced. In terms of the structure agency duality, this seems to be a swing of the pendulum toward individual agency. Fijians have an ambivalent view of thes e developments : for example, in the realm of romance individual villagers expressed the excitement of privately pursuing relationships with a variety of partners; however, many decried the overall effects on

PAGE 343

343 community life especially the dissolution of ma rriages associated with mobile infidelity. Even the ability to escape the mundane nosiness of neighbors has value some women said they prefer to call friends in the village rather than walk across the village green and be the object of scrutiny and curiou s questioning. Ideology tiko na kena vinaka, ia e tiko talega na kena ca aspe that they allow kin to stay in closer touch, a key cultural value of Fijians. A second major benefit is that mobile phones allow people to obtain help quickly in times of need, whether the need relates to monetary assistance, medical treatment, or rescue from physical danger. The negative aspect, many Fijians explained, is that phon es can detract from social solidarity bula vakaveiwekani ) that Fijians hold dear as one of the key traits that distinguish them from non Fijians. This concern relative to ICTs is not unique to Fiji norms place far more emphasis on personal freedom. This angst can be understood as a symptom of a shift in the basis of relationships; the kinds of socia l networks we participate in, the kinds of expectations we have of others, and what in turn we can depend on others for. Rainie and Wellman Internet, mobile phones, and soci al networking). The mechanism through which this revolution is occurring, is networked individualism, the increasing ability or tendency of individuals to form diffuse, geographically scattered, highly individualized networks

PAGE 344

344 based on shared interest, rath er than based on shared membership in families, villages, or or other collective groups (2012) Networked individualism is a radical departure from Fijian social relations, in which individuals locate themselves within a matrix of pre define d groupings and kin relationships. The data I collected in Toba, Cagi and Yavu show some indications of a move toward this new form of networked sociality. Particulary in the realm of romantic relationships, young Fijians are creatively extending their net works into unknown territory, both geographic and figurative. On a societal level, one cannot help but conclude that mobile phones paired with mobile internet, are a potent force for cultural diffusion, through which flows of ideas and media from the glo bal mainstream will accelerate. For s mall scale, indigenous societies, diffusion often equates to a form of acculturation. But what of Fijian cultural identity, which constantly reinvents itself in opposition to the outside world? Shortly after I arrived i n Fiji in 2011 I was earnestly asked a strange question by a boy as we hiked through the forest Ryan, is it true, tha After asking for clarification, it became apparent that the boy was referring to a sort of computer chip, ins talled in my wrist or forehead In the ensuing months I was asked this question multiple times by villagers in Toba, Cagi and Veidogo. It turns out that this belief in computer chip implants has its roots in a verse in the Biblical Book of Revelations ( 13:15 17): And he causeth all, both small and great, rich and poor, free and bond, to receive a mark in their right hand, or in their foreheads: and that no man might buy or sell, save he that had the mark, or the name of the beast, or the number of his name."

PAGE 345

345 Fijians, most of whom are deeply religious, had interpreted this enigmatic verse through the lens of the bewildering technological changes occurring around them. In the imagination of believers, a demonic force (perhaps The One World Government) is implanting people with computer chips through which they are ident ified and controlled, and without which they are in fact unable to buy and sell items in the market In other words, chips are required for participation in the new globalized economy. I se e reflected in this story many of the doubts and fears that Fijians, and indeed people around the world, face in an unknown future in which our social lives and everyday activities are increasingly mediated and recorded by information technologies. Perhap s, however mediated communication and networked individualism will not entirely supplant older structures, and multiple forms of sociality can coexist together. Toward the end of my fieldwork, I received word from my host family in Cagi that a friend of m ine in the village had died. Peceli, the self described S whose profile I presented in Chapter 5 had passed away after a sudden illness. Recalling our interviews (and the massive task of surveying him about his 300+ phone contacts), I was truly saddened by the passing of this friendly giant of a man. body was transported from the urban hospital back to the village for burial. At each step of journey grieving relatives and funeral attendees pho tographed the unfolding evens with their mobile phones, and Akuila, the tech love, support and commemoration from Cagi communi ty members around the world. Peceli had died, but in the ethereal world of online networks, Fi jian culture was alive and well. Our social

PAGE 346

346 lives may be mediated by and suffused with technologies, but these technologies ultimately serve very human purposes.

PAGE 347

347 APPENDIX A PRONUNCIATION OF FIJIAN WORDS The Fijian orthography was developed in the year 1839 by the Wesleyan missionaries William Cross and David Cargill, as part of their effort to translate the Bible into the Fijian language. The Fijian orthography us es 23 letters of the Latin script. The pronunciation of Fijian words is highly consistent with their spelling. However, several of the letters have pronunciations unlike those found in English: B is pronounced like th C is pronounced like English G is pronounced like English in the English word Vowels are roughly pronounced as they are in Spanish:

PAGE 348

348 APPENDIX B COPIES OF SURVEY FORMS Household Survey Questions 1. Please name all people currently living in this household. 2. What is the sex of each household member? 3. What are the current ages of household members? 4. What is the marital status of each household member? 5. What is the occupation of each household mem ber? 6. What is the native village of each household member? 7. What is the relationship of each household member to the head of household? 8. Please name all siblings and non resident children of the head of household and his wife. a. Where do are these individuals l ocated and what are their occupations? 9. How many times has each household member traveled away from the local area during the past 12 months? 10. For each trip away from the local area, please describe the destination and purpose of the trip. 11. Does anyone in you r household have a bank savings account? 12. Does your household currently have a functioning: a. Kerosene stove b. Gas stove c. Home phone (Easytel) d. Radio e. TV f. DVD player g. Computer h. Generator i. Refrigerator j. Washing machine k. Car/truck l. Boat 13. Which household members have mobile phones? 14. During the past 12 months, please estimate how much your household has earned from: a. Copra b. Taro c. Kava d. Other crops

PAGE 349

349 e. Fish/seafood f. Livestock sales g. Handicrafts h. Small business i. Local wages j. Non local wages k. Government aid/pensions 15. During the past 12 months, has anyone living outside the village sent goods or money to your household? a. If yes, who are they and what is your relationship to each individual? b. What is each individuals sex? c. If yes, where does each individual live, and what is their occupation and na tive village? d. How many times did you receive money or goods from each individual? e. For each time the individual sent money or goods, exactly what did they send? 16. During the past 12 months, have members of your household sent any goods or money to people livi ng outside this village? a. If yes, who are they and what is your relationship to each individual? b. What is each individuals sex? c. If yes, where does each individual live, and what is their occupation and native village? d. How many times did you send money or goo ds to each individual? e. For each time you sent money or goods, exactly what did you send? 17. Please name 4 residents of this village, who know a lot about events that occur outside the village. Tukuna mada na yacadratou e va na lewe ni koro e (x), ka eratou d au kila e levu na itukutuku me baleta na veika e yaco tiko e na taudaku kei (x). 18. Please name 4 residents of this village who talk a lot on the telephone. Tukuna mada na yacadratou e va na lewe ni koro e (x) ka eratou dau vosa vakalevu ena talevoni. 19. Pleas e name 4 natives of this village, who live outside the village, who have good jobs (high or respected professional positions). Tukuna mada na yacadratou e va na gone ni (x) ka eratou dau tiko ena taudaku ni (x) ka eratou cakacaka vinaka eratou tamata va ka itutu lelevu ena vuku ni nodratou cakacaka.

PAGE 350

350 20. Please name 4 natives of this village, who live outside the village, who often assist with things in the village (either send money or lend their assistance to projects). Tukuna mada na yacadratou e va na gon e ni (x) ka eratou dau tiko ena taudaku kei (x), ka eratou dau veivuke mai vakalevu ena koro oqo. 21. In your view, are the income differences between households in this village large (3), substantial (2), small (1), or are there no differences at all (0)? En a nomuni irai, e levu na duidui, se lailai na duidui ena rawa ka vakailavo ena loma ni koro oqo? Kena ibalebale, e so na vuvale e (x) e levu na nodratou ilavo se iyaya ni vakatautauvatataki kei na so tale na vuvale? 0= sega na duidui, 1= e lailai na duidui 2= e so tiko na duidui, 3= e levu na duidui 22. In your view, is the current level of difference in household income good or bad? Why? Ena nomuni irai, e ka vinaka se sega na levu ni duidui e tiko ena rawa ka vakailavo e na koro oqo? BALETA? 0=no response, 1= e ka vinaka 2= neutral 3= e ka ca 23. If there are differences in household income in this village, in your opinion what is the cause of those differences? Kevaka e dua na duidui ena rawa ka vakailavo e laurai vei iko ena loma ni koro oqo, na cava, ena no mu vakasama, e vakavuna na duidui koya? 24. In your own household, how important for everyday life are the money and goods sent by relatives who live outside the village? (1= not important, 2= a little bit important, 3= somewhat important, 4= very important) E ka bibi ena bula ni veisiga vei kemudou e tiko e vale na veivuke eratou dau vakauta mai na wekamu eratou tu ena taudaku ni Navakavu, se sega ni ka bibi? 1= sega ni ka bibi 2= vakalailai ga 3= e bibi toka 4=e ka bibi sara 25. Can you compare life now, with l ife 5 years ago. From that time until today, has the amount of money and goods sent to your household from relatives lving outside the village: 1) decreased significantly 2) decreased 3) not changed 4) increased 5) increased significantly E rawa ni o ni v akatautauvatataka mada na bula ena gauna oqo, vata kei na bula ena lima na yabaki sa oti. Mai na gauna koya ki na gauna oqo, e sa veisau vakacava na levu ni nodratou dau veivuke main a wekamu e tu ena taudaku kei (x)? 1= toro sobu sara 2= toro sobu 3= sega na veisau 4= toso i cake 5= toso i cake sara 26. If there has been a change, why?

PAGE 351

351 KE VEISAU, ena vuku ni cava beka? 27. Can you compare life now, with life 5 years ago. From that time until today, has the frequency of your travel: 1) decreased significantly 2) decreased 3) not changed 4) increased 5) increased significantly E rawa ni o vakatautauvatataka mada na bula ena gauna oqo, vata kei na bula ena lima na yabaki sa oti. Mai na gauna koya kin a gauna oqo, e sa veisau vakacava na levu ni nomu veilakoyaki? 1= toro sobu sara 2= toro sobu 3= sega na veisau 4= toso i cake 5= toso i cake sara 28. If there has been a change, why? KE VEISAU, ena vuku ni cava beka? 29. How important to you are the following kinds of information to you in your everyday life? (0= not importa nt, 1= a little important, 2= very important) Na mataqali itukutuku cava e dau uasivi vei iko ena nomu bula ni veisiga? 0= sega ni uasivi 1= vakalailai ga 2= uasivi sara Where do you usually get this kind of information? (0=N/A, 1= conversation, 2= villag e leadership, 3= radio, 4= TV, 5= newspaper, 6= advertisements, 7= village meetings, 8= telephone, 9= internet, 10= text, 11= letters (mail)) Iko dau rawata mai vei na mataqali itukukutuku oqo? 0=N/A, 1= na veivosaki, 2= na veiliutaki ni koro, 3= retio, 4= TV, 5= niuspepa, 6= adverts, 7= bose vakoro/tikina 8=talevoni, 9= internet, 10= na text, 11= ivola ena post a. Information about farming or animal husbandry Itukutuku baleta na teitei se na susu manumanu b. Information about the price of produce or goods in th e market Itukutuku baleta na isau ni kakana se iyaya ena makete c. Information about transport schedules (bus or ferry) Itukutuku baleta na veisala ni veilakoyaki, me vaka na basi se na waqa d. Information about government policies or projects Itukutuku baleta na veiqaravi ni matanitu e. Information about politics Itukutuku baleta na politiki f. Information about education or schools Itukutuku baleta na vuli se na koronivuli g. Information about the weather Itukutuku baleta na draki h. Information about world news and even ts Itukutuku raraba baleta na veika e yaco tiko e vuravura i. Information about the things that are happening in the local area

PAGE 352

352 Itukutuku baleta na veika e yaco tiko e (x) j. Information about your relatives who live outside the village Itukutuku baleta na wekam u eratou tu ena taudaku ni (x) 30. Please compare your life now, with your life 5 years ago. From that time until now, how has the amount of information you receive from the following sources changed? (1) decreased significantly 2) decreased 3) not changed 4 ) increased 5) increased significantly) E rawa ni o vakatautauvatataka mada na gauna oqo, vata kei na gauna ena lima na yabaki sa oti. Mai na gauna koya ki na gauna oqo, e sa veisau vakacava na levu ni nomu dau vaqara itukutuku mai na: 1= toro sobu sara 2 = toro sobu 3= sega na veisau 4= toso i cake 5= toso i cake sara a. Government employees or offices Na vakailesilesi se tabacakacaka ni matanitu b. Market vendors O ira na dauvolivolitaki ena makete c. NGOs Na veisoqosoqo ena taudaku kei na matanitu d. Radio Na retio e. TV Na TV f. Newspapers Na wili niuspepa g. Telephone conversations Na veivosaki ena talevoni 31. Is a member of this household a member of a local committee or organization? E dua na lewe ni vale oqo e lewena tiko e dua na soqosoqo se komiti e (x)? a. Who? O cei? b. Which committee? Na komiti cava? 32. If yes, how do members of the committee contact each other? Is the telephone used for any committee business? Kemuni na lewe ni soqosoqo dau bose se veitaratara vakacava? E dau yaga vei kemuni na qiri ena talevoni ena cici vaki ni nomuni komiti se soqosoqo? 33. Ena gauna oqo, iko dau vakayagataka vakavica na sala ni veitaratara se ivurevure ni itukutuku oqo? 0= sega vakadua 1=vakadua/yabaki 2= vakadua/vula 3=vakadua/macawa 4=vakadua/siga

PAGE 353

353 a. Nomudou talevoni dabe e vale (tfl) b. Na ta levoni dabe (tfl) ena dua tale na vale ena koro c. Nomu mobile d. E dua na mobile e taukena e dua tale na lewe ni vale e. E dua na mobile e taukena e dua tale na lewe ni koro e ______ f. Na RT g. Na talevoni saumi mai tauni h. Na text ena mobile i. Na fax j. Na internet k. Na Retio l. Na TV m. Na niuspepa n. Vakauta e dua na ivola ena post 34. Telephone Timeline a. Kemudou e vale a rawata main a nomudou matai ni talevoni dabe (tfl) ena vica na yabaki sa oti? b. Kemudou e vale a rawata mai na matai ni nomudou mobile ena vica na yabaki sa oti? 35. A. E vic a na ilavo kemudou e vale dau vakayagataka ena topup se na recharge ni mobile ena loma ni dua na macawa? b. E vica na ilavo kemudou e vale dau vakayagataka ena topup se na recharge ni talevoni Easytel ena loma ni dua na macawa? 25. Kevaka e sega ni dua na nomudou talevoni e vale, iko dau qiri vakacava? a. kerea mai e dua na talevoni, qai vakayagataka ga vakai iko b. kerea mai e dua na talevoni, qai kerea vei itauk ei ni talevoni me sa tabaka na n aba iko via qiritaka c. vakayagataka e dua na talevoni saumi d au sega ni dau qiri ena talevoni 26. Kevaka o kerea mai e dua na talevoni mo qiri kina, iko dau sauma vakacava na itaukei ni talevoni? a. Au volia mai e dua na noqu flex se recharge card qai vakayagataka ena nona talevoni b. A solia e dua na ilavo vei taukei ni talevoni c. Au dau kerea mai na talevoni kevaka e freecall tiko d. Au sega ni dau solia e dua na ka vei taukei ni talevoni au dau qiri ga sega ni saumi. 27. In your opinion, does the telephone help or hurt the a. economic life of the househol d? Why?

PAGE 354

354 b. relationships within the household? Why? c. relationships among villagers? Why? d. relationships with relatives living outside the village? Why? e. the education or knowledgeability of Fijians Ena nomu vakasama, na talevoni e dau vukea se vakaca cana na: a. bula vakailavo vakavuvale? Baleta? b. veimaliwai vakavuvale ena loma ni vale? Baleta? c. Na veimaliwai vakaveiwekani ena loma ni koro? Baleta? d. Na veimaliwai vakaveiwekani vata kei iratou na wekamu era tu ena taudaku i na koro? Baleta? e. Na vuli se se na ivakatagedegede ni kila ka e Viti? Baleta? Phone Survey Questions 1. Where did you study for primary school? Kemuni a vuli primary mai vei? 2. Where did you study for secondary school? Kemuni a vuli secondary mai vei? 3. During your life, have you ever lived outside this local area for more than a year? If so, where? Ena nomu bula taucoko, e so na vanua ena taudaku kei (x) kemuni sa tiko kina, e siviata e dua na yabaki na nomuni tiko kina? 4. During your life, how many years in total have you lived ou tside of this area of Fiji? Ena nomu bula taucoko, e vica na yabaki isoqoni na nomuni tiko ena taudaku kei (x)? 5. Who owns this phone you individually, or you and your wife, or some other group of people? O cei taukena na talevoni oqori kemuni ga, se kemu drau kei watimuni, se kemudou vakavuvale, se e dua tale e taukena tiko na talevoni oqo? 6. Who typically uses this phone to make calls? O cei e dau vakayagataka na talevoni oqori me qiri kina? 7. Who carries this phone just you, or does someone else also carr y the phone around? O cei e dau kauta na talevoni oqori kemuni ga se e so tale eratou dau kauta wavoki talega? 8. Did you buy this telephone, or did someone else give it to you as a gift? Kemuni a volia mai na talevoni oqori, se e dua a solia vei kemuni vak aloloma?

PAGE 355

355 a. If a gift, from whom? KE LOLOMA, loloma mai vei cei? b. If a gift, how are you related to the giver? KE LOLOMA, kemudrau veicavani kei _______? c. If a gift, where did the giver reside at the time of the gift? KE LOLOMA, o _________ dau tiko mai vei? d. If a gift, did the giver give you this phone for a birthday or other occasion, or not? KE LOLOMA, o _______ a solia vei kemuni na talevoni ena vuku ni dua na siga ni sucu se olodei, se solia ga? e. If a gift, when the phone was given to you was it new or used a lready? KE LOLOMA, ena gauna e sa solia kina vei kemuni, a ka vou se vakayagataki oti? 9. A. How many mobile phones have you owned before this mobile phone? Na talevoni oqori e kavica ni nomuni mobile? B. How many home phones have you owned before this home phone? Na talevoni oqori e kavica ni nomu talevoni dabe/tfl? 10. How many months have you owned your current phone? E vica na vula na nomuni taukena tiko na talevoni oqori? 11. What happened to your last phone? A cava a qai yacova na nomuni iotioti ni talevoni? 12. How long was it between your last phone and when you acquired your current phone? A cava na kena dede main a ca ni nomuni talevoni e liu moni qai rawata tale e dua tale na nomuni talevoni? 13. Do some other villagers in this village (other than members of y our household) sometimes use your phone? Can you tell me who has used your phone during the past month? E so na lewe ni koro e (x) eratou sega ni tiko ena nomu vale e dau vakayagataka na nomu talevoni ena so na gauna? E rawa ni tukuna o cei soti sa vakayag ataka na nomu talevoni ena vula sa oti? a. How are you and (borrower x) related? Kemudrau veicavani kei _______? b. Does (borrower x) use your phone to make calls, receive calls, or send texts? O ___________ dau vakayagataka na nomu talevoni me qiri yani, me sau ma e dua na qiri, se me vakauta e dua na text? 14. Can you estimate how many texts you send in a typical day?

PAGE 356

356 E rawa ni estimedi e vica na text kemuni dau vakauta ena loma ni dua ga na siga? 15. Can you estimate how many calls your make in a typical day? E rawa ni estimedi, kemuni dau qiri yani vavica ena loma ni dua ga na siga? 16. Do you use your phone to listen to the radio? If so, how often? (DWMYX) Kemuni dau vakarorogo retio se sere ena nomu talevoni? 17. Do you use your phone as a clock? If so, how often? (DWM YX) Kemuni dau vakayagataka na nomuni talevoni me kaloko? 18. Do you use the light on your phone? If so, how often? (DWMYX) Kemuni dau vakayagataka na cina ena nomuni talevoni? 19. Do you use your phone to play games? If so, how often? (DWMYX) Kemuni dau qito e na nomuni talevoni? 20. Is there something else that you use your phone for? (DWMYX) E dua tale na ka kemuni dau vakayagataka kina na nomuni talevoni? 21. Do you use your phone for your job, or a small business, or is there some other way that you use your pho ne to make money? Kemuni dau vakayagataka na nomuni talevoni ena dua na job, se bisinisi lailai, se e dua na sala moni rawata kina e dua na ilavo mai na nomuni talevoni? 22. Do you use M Paisa or Mobile Money on your phone? Ko ni dau vakayagataka na M Paisa s e na Mobile Money ena nomuni talevoni? a. If yes, do you use these services to top up your phone? KEVAKA IO, kemuni dau vakayagataka me vakatawana na nomuni talevoni? b. If yes, do you use these services to send money or receive money from others? KEVAKA IO, kem uni dau vakayagataka me vakauta se ciqoma e dua na ilavo? c. If yes, do you use these services to pay bills, such as electricity or water bills? KEVAKA IO, kemuni dau vakayagataka me sauma e dua na bili (bili ni wai se livaliva)? d. If yes, do you use these serv ices as a way to store your own money, like a bank account? KEVAKA IO, kemuni dau vakayagataka me m aroroya kina e dua na ilavo, vaka e dua na baqe? 23. Have you ever used the internet? Kemuni sa bau vakayagataka oti na internet?

PAGE 357

357 a. If yes, how often do you use e mail? (DWMYX) KEVAKA IO, kemuni sa vakayagataka oti na email? b. If yes, how often do you use Facebook? (DWMYX) KEVAKA IO, kemuni sa vakayagataka oti na Facebook? c. If yes, how often do you use Skype? (DWMYX) KEVAKA IO, kemuni sa vakayagataka oti na Skype? d. If yes, how often do you use Instant Messenger? (DWMYX) KEVAKA IO, kemuni sa vakayagataka oti na Instant Messenger? e. If yes, how many years ago did you first use the internet? KEVAKA IO, e vica na yabaki sa oti na nomuni matai ni vakayagataka na internet? f. If yes, who taught you how to use the internet (and how are you related)? KEVAKA IO, o cei a vakavulica se vakaraitaka vei kemuni na internet? (kemudrau veicavani?) g. If yes, how often do you use the internet? (DWMYX) KEVAKA IO, e vakacava na levu ni nomuni da u vakayagataka na internet? h. If yes, in which places do you use the internet? KEVAKA IO, ena gauna oqo, kemuni dau vakayagataka na internet ena vanua cava? (noqu talevoni, talevoni nei itokani, noqu vale, vale nei itokani, vale ni internet, other) 24. Can you estimate how much money you use in a week to top up your phone? E rawa ni estimedi e vica na ilavo kemuni dau vakayagataka mo vakatawana na nomuni talevoni ena loma ni e dua na macawa? 25. Can you estimate how much money relatives or friends transfer to you w ithin a typical week to top up your phone? E rawa ni estimedi e vica na ilavo eratou na wekamuni dau vakauta mai ena talevoni ena loma ni dua na macawa moni vakatawana kina na nomuni talevoni? 26. What kind(s) of SIM cards do you own? Vodafone/Inkk/Digicel? N a mataqali SIM cava kemuni taukena tiko? 27. The numbers that are stored in your phone who stored them, just you? Or did some other people store numbers? Na naba e storetaki ena loma ni nomuni talevoni o cei a storetaka? Kemuni ga, se e so tale sa store ta ka e so na naba ena nomuni talevoni? 28. internal memory? Na naba e storetaki ena loma ni nomuni talevoni e storetaki ena SIM, se ena talevoni, se e so ena SIM, e so ena talevoni?

PAGE 358

358 29. phone/SIM? I mean, are there some numbers that you memorize or write on paper? E so tale na naba e tiko ka kemuni dau qirita, ia e sega ni storetaki ena nomu ni talevoni? Kena ibaleb ale, e tiko e so na naba ka kemuni dau cramtaka ga, se vola ena pepa? 30. What is your current phone number(s)? A cava na nomuni naba ni talevoni ena gauna oqo? 31. Are there any other phone numbers that you used before, that you no longer use? E so na naba ni talev oni o ni a dau vakayagataka e liu, ia, oqo kemuni sa biuta, se vakayalia na naba? SIM Survey Questions 32. Is the owner of this number male or female? 33. What is the marital status of the owner of this number? 34. What is the current location of residence of the o wner of this number? 35. What is the occupation of the owner of this number? 36. What is your relationship to the owner of this number (kin/friend/business) a. If kin, in which degree of kinship do you stand? 37. How often do you and the owner of this number speak on the phone? (DWMXY) 38. How often do you and the owner of this number meet face to face? (DWMYX) 39. Are any of the numbers on this list owned by the same individual? Which numbers? 40. Are any of the numbers on this list owned by members of the same household? Whi ch numbers? 41. When we were doing the household survey, you said that (x) sent goods to your household in the past 12 months. If you want to call (x), which number on this list do you call?

PAGE 359

359 Ena gauna daru a cakava kina na vakatataro ni vuvale, kemuni a tukun a ni sa vakauta mai ena yabaki sa oti o ________. Ni o via qiri yani vei _______, naba cava ena list oqo kemuni na qirita? 42. Ke sega ni volai ena list oqo na naba nei _________, kemudr au dau veitaratara vakacava? E kilai beka na nona naba ni talevoni?

PAGE 360

360 APPENDIX C COLLECTING DATA FROM SIM CARDS how I managed these data and prepared the data for analysis. I des cribe logistical, methodological and ethical issues regarding this data collection technique. As part of my discussion of methodological issues, I assess the accuracy (or at least consistency) of survey responses regarding calling frequency. SIM Cards and SIM Card Readers A SIM (Subscriber Identity Module) is a chip that stores data that identify a subscriber on a GSM mobile telephone network enabling the phone to receive a network signal. A SIM can be removed from one GSM handset and put into another. Th is provides several advantages for phone users, allowing them to switch between different network providers, therefore reducing interconnection fees for calls to different networks. A GSM phone owner can travel to another country with his/her phone, buy a local SIM card and insert it in the phone. This way one can travel internationally but pay local rates for calls In addition to enabling network access, a SIM can store phone contacts text messages, and other information entered by the user This allows a phone user to remove her SIM and insert it in another phone, and have acce ss to all of her contacts withou t the need to laboriously copy them onto paper. SIM storage capacity varies, but most SIM cards can hold up to 250 contacts (stored as simple name/ number pairs) and up to 30 text messages. In addition to the SIM, phone contacts and text messages can be stored in the internal memory of the phone. Typically, there is a phone menu that gives the user the option of whether to store contacts and/or texts on the SIM or in the

PAGE 361

361 phone memory. The internal storage capacity for contacts and texts varies depending on the phone model. SIMs have password feature s that are designed to discourage theft of phones : a personal identification number (PIN) for ordinary u se and a personal unblocking code (PUK) for PIN unlocking. The PIN allows users to set up a 4 digit password so that other people cannot use the phone. If an incorrect PIN is entered 3 times, then the SIM must be unlocked using the 8 digit PUK. The PUK cod e is printed on the card certificate attached to the SIM card at the time of purchase, or can be provided by the phone company in the case of loss. Figure C 1: SIM card and mo bile phone. embedded on a plastic card that slides into a slot in the phone behind the battery and can be removed, reinserted, and swapped between different mobile handsets. There ar e various sizes of SIM cards. All of the SIMs that I measuring 25 mm by 15 mm. Photo used with permission from Zephyris (Richard Wheeler) via Wikimedia Commons.

PAGE 362

362 SIMs are typically sold separately from phone handsets, and are not expensive : in Fiji in 2011 new SIMs cost $FJ 5 10, depending on whether there was a promotion in effect Thus, a sing le person may own several SIMs, each corresponding to a different phone number If a SIM card is not used for a long time (typically 3 months) it may expire and become useless. A SIM card reader is a tool that quickly and efficiently copies the list of contacts ( in name/ number pairs) from the SIM to a computer. The reader itself resembles a thumb drive, which plugs into a computer USB port. Some SIM readers include proprietary software into which the SIM data are copied; the data can then be transferred into a spreadsheet program such as Microsoft Excel for analysis. SIM card readers can also copy text messages, but I did not use this feature (doing so would have infringed on the privacy of participants, and exceeded the guidelines of my IRB clearance) SIM card readers can be acquired online or at electronics stores, are inexpensive (typicall y costing between US$10 and $30) and do not require specialized training or equipment to use. I used a Dekart brand SIM card reader, but most other brands would probably also be adequate for collecting data in the field. Copying SIM Data in the Field The duration of my phone survey s varied widely usually between 30 60 minutes, depending on the number of contacts stored in the SIM/phone. It is a good idea to schedule the survey in advance with the participant, during a window of time in which they are free It is also advisable to conduct the survey in a quiet place away from onlookers and children, who can be a distraction. Before beginning the process, I explain ed to each phone owner the overall purpose of my research, the process of collecting data from the SIM, and the way the

PAGE 363

363 data will be stored and published in order to maintain anonymity and confidentiality. This is best explained in a straightforward and nontechnical way. Some respondents were initially curious or ev en somewhat suspicious about my de sire to access data from their mobile phones. Thorou ghly explaining the process put the participant s at ease. Some participants were actually eager to do the survey, in order to see the operation of t he SIM reader and computer. Several of my surveys were a ccompanied by requests for an impromptu demonstration my laptop computer, which was a novelty in rural Fiji. After the initial explanation of my research, I obtain ed the consent of the participant, whether verbal or writt en. I offer ed the participant compe nsation for participation ( $FJ 5), and thank ed them in advance for their assistance with my research. Before actually copying contacts, I determined all of the places in which the participant stored their phone contacts. I ask ed the phone owner whether th ey memorize any phone numbers, or have any contacts that they write on paper or in a were memorized or written on paper, I type d as many of these as possible into the survey spreadsheet. I t hen ask ed if the participant if s/he had any additional SIMs, and if so, if s/he could please bring all of the SIMs for analysis. I then determine d if any contacts we re stored in the phone memory. I ask ed the phone owner to quickly scroll through his/he r list of contacts while watching the screen. Usually there is a small icon next to the contact name which indicates whether it is saved in the SIM or the phone. If some contacts are saved in the phone, they must either be typed out by hand, or transferred t o a SIM so that they can be copied by the

PAGE 364

364 SIM reader. If there were only a few contacts in the phone memory (~0 20), I usually just type d them out by hand. If there wer e many contacts in the phone memory, I ask ed the permission of the phone owner to copy t hese contacts to a blank SIM ( in my possession), so that I could use the SIM reader to copy them to the computer. After I determine d whether contacts were in the phone memory, I ask ed the participant to remove the SIM from the phone for analysis. Many phon e owners are accustomed to frequently removing and replacing their SIM cards, in order to make calls on different networks. However, a few participant s required an explanation of how to open their phone handset and remove the SIM, and some asked me to remo ve it for them. Removing the SIM card from the phone, and copying data from it using the SIM car d reader, does not damage a mobile phone n or alter the contents of the SIM card. of which the phone owner should be informed beforehand. Finally, I insert ed SIM cards in turn into the SIM card reader, insert ed USB port, and save d the data to the CSV spreadsheet file (Table C 1 ) Data Management I designed my spreadsheets and files with the goal of making the survey process as quick and efficient as possible. The tables below are simplified versions of my survey spreadsheets, which serve to illustrate how I managed my data. For the purposes of this example I will pretend to go through the steps of a survey interview with a fictional SIM, elicit and enter information about his contacts, and ultimate ly convert information

PAGE 365

365 I used a series of four linked spreadsheets to accomplish th is. T he .CSV file (Table C 1) is the format in to which the raw data from the SIM are saved to the computer by the SIM reader. The survey spreadsheet (Table C 2) is the spreadsheet about his phone contacts. The community directory spreadsheet (Table C 3) is a list of all phone numbers found in all of the contact lis of the individuals associated with those phone numbers. This file corresponds to a No delist for purposes of social network analysis. Finally, the list of all phone interactions (Tabl e C 4) is a combined list of every phone people who contact each other over the phone), along with the characteristics of each of the se relationships. This file corresponds to an Edgelist for purposes of social network analysis In Tables C 1 to C contacts, as a means to illustrate how information about the phone contacts transfers between the different spreadsheets. I used UCINET and Netdraw software to analyze the data from my phone survey spreadsheets The Community directory (Table C 3) corresponds to a nodelist, and the List of All phone interactions (Table C 4) corres ponds to an edgelist. Table C 1. .CSV file Name Phone # Amy 3928381 Bill 9380193 Charles 3829219 The SIM card reade r copies the list of contacts from the SIM and saves it to the computer as a CSV spreadsheet file with paired names and phone numb ers. I immediately copied and paste d these names and numbers into the first two columns of another file, t he Survey spreadsheet (Table C 2 below ).

PAGE 366

366 Table C 2. Survey spreadsheet (simplified) Name Phone # Sex Location Relationship Call Frequency Olav 60 39291 Bill 9380193 M Cagi Theresa 7294920 I bega n by pasting the data from the CSV file into the first two columns of the survey spreadsheet, which mimic the columns of the CSV file. The third and fourth columns (Sex and Location) represent fi xed characteristics of the person associated with a phone number. The fifth and sixth columns (relationship and calling frequency) represent characteristics that are specific to the relationsh ip between the phone owner and each stored contact. Take note of the spreadsheet ( Table C 3 below). This is a list of all phone numbers found in every phone cont act list in that 9380193 belongs a man who liv es in Cagi for that information again I will simply ask Roger about his relationship with Bill, and how frequently they talk to each other on the phone (Columns 5 and 6). This lookup function reduces the duration of surveys. It also ensures that each node in my eventual network only has one set of characteristics. Table C 3. Community directory spreads heet (simplified) Phone # Sex Location 4450203 F Suva 8291039 F England 9380193 M Cagi 2948482 M Suva The Community di rectory spreadsheet contains information about all of the contacts found in all of the phones in the entire community After completi ng each phone survey, I paste d contacts into this spreadsheet. The list therefore gr ew progressively as I continue d to conduct surveys i n a community. This spreadsheet was he Survey Spreadshe et (Table C 2 above ), so that in formation for any number that was in the Community directory a utomatically populated the Survey spreadsheet. In this table, the phone number 9380193 is highlighted. This one contact list above. When the number 9380193 is pasted into the Survey Spreadsheet, the corresponding data for Sex and Location for that phone number appear automatically in the Survey Spreadsheet. Table C 4. List of all phone interactions (simplified ) Ego Phone # Alter Phone # Relationship Call Frequency 9113949 3928381 Mother/Son Weekly 9113949 9380193 Brothers Daily 9113949 3829219 Cousins Monthly The List of all phone interactions words, the phone numbers for all of the phone owners in the community (1 st column), are paired with all of the numbers found in their contact lists (2 nd column). Beside each pair of numbers is the information elicited about the relatio nship during the survey (frequency of calling, kin relationship, etc.). After each phone survey, I paste d dsheet. The list therefore grew progressively as I continue d to conduct su rveys in a community. N otic e that the numbers in the second Edgelist for th e purposes of Social Network Analysis.

PAGE 367

367 The Scientific Value of Phone Contact Lists Using phone contact data from SIM cards communication network has logistical, methodological, and ethical conveniences and inconveniences In this section I will discuss these issues relative to other existing techniques for studying phone communication networks Logistical Considerations Copying phone contact lists, as I did, to analyze a calling network is far more time consuming than using actual calling records provided by a network service provider. Access to actual calling records obviates the need to conduct surveys, while providing a precise record of calls placed among a population of phone numbers over a given span of time Using such data from phone service providers, researchers have been able to analyze calling networks for entire countries (e.g., Onnela et al. 2007) However, using actual calling records requires special access to these data, which are typically closely guarded by servic e providers, and therefore inaccessible to most independent researchers. Because I did not ha ve access to actual calling records I was forced to use phone contacts as a proxy for the actual calling network. Using phone contacts for a network study is labo r intensive and time consuming. In the case of a community whole network study, every phone owner in the community must first be identified through a household census. Each phone owner must then be surveyed individually about their contacts. These phone su rveys are tedious and lengthy the phone contacts portion alone typically ranged between 30 minutes to more than one hour. The tediousness of these surveys necessitates payment of research participants. The unusual and invasive nature of these surveys requ ires an in depth

PAGE 368

368 explanation before the beginning of each survey, which adds to the time and labor demands on the researcher and participant C onducting surveys in rural developing villages presents further logistical challenges. Many such villages lack reliable electricity, so bringing an independent e laptop is advisable Scarce or unreliable electricity batteries are dead at the t ime you have pre arranged to conduct the survey. A dead battery makes it impossible to see the screen in order to check if any contacts are This may require t he interview to be rescheduled for another day. The SIM card reader is ultimately used as a shortcut to save time. Without the screen and typed out one by one, which would be very laborious and time consuming. The SIM card reader (or some alte rnative time saving technique) is necessary in order to make this approach logisti cally feasible, because if contacts had to be typed one by one, the average interview might take several hours. Although collecting phone contacts for network studies is very labor intensive, this approach also has certain methodological advantages, which for certain research questions makes it superior even to the use of actual calling records. Methodological Considerations Phones span multiple levels of analysis Analyzing a telephone communication network is a complex problem because phones may be owned by individuals, married couples, or households; furthermore, a phone may be used or borrowed by a much broader group of individuals. The degree to

PAGE 369

369 which landline and mobile ph ones are jointly o wned or shared is shaped by cultural and economic circumstances. The extent of sharing may vary widely even within a single community. Landline (Easytel) phones in Fiji tend to be owned by the family unit, and used extensively by other v illagers. Sharing of mobile phones is more variable. In wealthy societies mobile phones tend to be highly personal devices, and are rarely jointly owned or shared with others. In developing countries such as Fiji mobiles are scarce r and tend to be ex tens ively shared and borrowed by friends, kinsmen and neighbors. In Fiji, e ven individuals who own their own phones, out of convenience, or because of a dead ba ttery or lack of phone credit. The interchangeability of SIM cards adds another layer of complexity to the problem of analyzing calling networks One phone owner might own several SIM cards, each of which is associated with a separate phone number. In contexts such as Fiji, it is necessary therefore to dist inguish between the registered owner(s) of a mobile phone (or SIM), and the group of people who actually use the phone or SIM to make or receive calls. A couple of examples illustrate the importance of these issues in obtaining accurate research results T ake for instance an individual who owns 3 different SIM cards. In a set of act ual calling records obtained from a network provider these phone numbers would each be treated as a separate individual, whe n in reality it is just one person The resulting net work would have an exaggerated number of nodes, and the ties for each node would paint an incomplete picture of the true range of relationships engaged in by each individual. In Veidogo village, a single household was associated with 23 separate phone numb ers. If each of

PAGE 370

370 these phone numbers were equated with an individual person, it would yield a very inaccurate picture of reality! Another example that illustrates the problems of treating phone numbers as individuals, regards the issue of borrowing. A man might purchase a phone, but the wife might actually use the phone more often than he does The calls made on that phone number will be attributed to the man (to whom the SIM is registered) but in reality it is usually the woman using the phone. To m ake things even more complicated, the woman might occasionally lend the phone to her neighbor s, who make calls to a range of people who do not belong to the social network of the phone owner or his wife Finally, in Fiji phones and SIMs are frequently giv en outright to friends and family members as gifts. In fact, the majority of my respondents received their phones as gifts (mostly used) rather than purchasing them. The SIM card remains registered to the original purchaser, who perhaps has very different characteristics (sex, age, etc.) than the new owner. None of these nuances are reflected in calling data collec ted by mobile network providers. A call between two numbers is simply assumed to be a conversation between the two individuals to whom these numb ers are registered usually at the time of purchase For these reasons, in developing contexts particularly, a survey based study that copies contacts from address books c an yield more accurate and meaningful results than a study that uses actual calling records Surveys and observation can establish which phones and phone numbers belong to which individuals, and the degree of sharing or borrow ing associated with each phone number. Multiple numbers associated with one individual or household can later be aggregated into a single node for the

PAGE 371

371 purposes of network analysis. D etailed d ata about phone owners and their households can be correlated with their phone use behavior to learn about patterns of phone use among different demographic groups These kinds o f analyses are not possible with sets of calling data, because detailed infor mation about the individuals or households making calls on a given phone number is usually unknown. Phone directories are not a complete record of calling activity A methodologica l limitation of my approach is that p hone address books are an imperfect proxy of calling activity, and represent only those numbers which an individual h as actually saved into the phone, rather than numbers that have actually been called Individuals var y in their propensity to save a number into their phone directory. Some make a habit of saving every number they call, or from which they receive calls. Others rarely save numbers, instead writing them on paper, rely ing on memorization asking neighbors, or other strategies. As one Toba woman put numbers rather than saving them just too lazy to e nter them into the address book Crammed numbers might be those numbers that are most frequently ca lled and hence committed to memory ( i.e., a spouse or close relative), or they might be numbers which the phone owner does not wish to enter into the address book, for fear that they will be discovered; for example, an illicit lover. Some phone owners sim ply do not know how to save contacts in their phone address book; these people rely entirely on cramming or writing phone numbers on paper. It is fairly common for a Fijian to call somebody simply in order to ask for the phone number of some third individ ual, whose number has been lost or forgotten. Alternatively, w hen someone from outside a village attempts to contact someone living in the village he or she might just call

PAGE 372

372 any other village whose number is known, and a sk them to pass the message on to its intended recipient. may also vary with duration of phone/SIM ownership. Over time more and more contacts tend to accumulate in the address book S ome people are di ligent about era sing outdated or unused numbers, while other phone address books are like messy attics filled with numbers of deceased relatives, ex girlfriends, or people briefly befriended and never contacted again. In fact, quite a few numbers which a re saved in phone directories are never actually used. Some phone relationships are very one sided, with one party always calling and the other always receiving. In some such cases the receiving party never saves the numbers in the directory and simply answers when they receive a call. This seems particularly to be the case with overseas kin. Entering and erasure of numbers can be influenced by cost considerations one fone numb ers into her address book, because they are too expensive to call from her Digicel phone, and she is afraid if she enters the number she will dial it inadvertently and lose credit For all of these reasons, the number of contacts in an address book is an i mperfect indicator of actual calling network. In rural Fiji, people who borrow phones owned by other villagers usually do not frequency of borr owing and the relationship between the phone owner and borrower. Members of the same household sometimes save their contacts in the mobile phone owned by their father, brother, or daughter, which they borrow on a daily basis.

PAGE 373

373 Sometimes such contacts are co mpletely un recognizable to the phone owner during a survey. The fact that out of household borrowers do not typically save their contacts in borrowed phones is helpful in that the list of numbers saved in the address book can usually be assumed to have bee n entered by the phone owner himself. The downside is that the network data collected from village phones do not reflect the calling networks and calling activities of non phone owners within the village. This relates to a larger methodological issue: eve ryone in a Fijian community uses phones to some extent, but not everyone (or even every household) possesses a phone. Thus, simply surveying phone owners about their calling habits leaves out the contacts and calling activities of non phone owners. A possi ble solution is to survey everyone in a village about their calling activities and contacts, and not simply interview phone owners. However, this greatly increases research costs and time requirements. Given global trends in telecommunications development prospects for using saved phone contacts as a source of data for network analysis should improve. As phone ownership becomes increasingly universal, the prevalence of joint ownership and phone borrowing and lending will likely decrease. Likewise, as the older generation dies and the younger, tech savvy generation takes their place, it will increasingly be the case that phone owners will use the directory feature exclusively to store their numbers, and that the phone directory will be a more accurate repre sentation of the numbers called by the phone owner. Therefore, conditions in the near future should be increasingly conducive to use of this method. Inaccuracy of recall My approach to data collection relied on the ability of respondents to accurately asse ss the frequency with which they call specific individuals. The problem is that

PAGE 374

374 people are generally not very good at recalling the absolute or even the relative frequency of their inte ractions with other individuals. Bernard, Killworth and Sailer studied the accuracy of informant reports about past interactions in a series of experiments with teletype users, ham radio operators, and users of an early version of email, and compared their responses to actual records of their interactions. Regardless of commu nication medium, they found that people could recall less than half of their interactions measured in terms of count or in terms of frequency. They found that accuracy was only about 6% better regarding recall of recent interactions, compared to interacti ons that occurred longer ago. Furthermore, individual differences in accuracy could not be accounted for by personal characteristic s of people, such as sex or age. Bernard, Killworth and Sailer concluded on the basis of these experiments that what people s ay about their past interactions bears no useful resemblance to their actual behavior. (Bernard and Killworth 1976; Bernard and Killworth 1977; Bernard, et al. 1984; Bernard, et al. 1980; Bernard, et al. 1982) Sev eral studies have compared self reports of mobile phone use to actual calling found that correlations between actual phone use and recall of phone use (duration and n umber of calls) 6 months later were moderate to high (0.5 0.8) across countries, and that overall the self reported duration of calls was overestimated by 40%. Tokola et al. found a correlation of 0.71 between records from network operators and self report s of average calling time among 70 volunteers, who on average overestimated average call duration by 46% (2008). Another study in the U.S. assessed the accuracy of recall by comparing with billing records, yielding a correlation of 0.74.

PAGE 375

375 In this section I assess the accuracy of my respondents in Fiji in regard to the frequency of t heir phone interactions. In many ways my Fiji data are similar to those of Vrijheid, et al. (2006) and Tokola, et al. (2008). I asked respondents about the frequency of their tec hnologically mediated interactions with specific individuals. However, the conditions of my survey s also differ ed in important ways At the outset, I presented my research participants with the list of contacts that they had entered into their phone addres s book, and then asked about the frequency of interaction with each contact. This contact list served as a prompt which may have helped to jog the ons. Indeed, Bernard, et al. noted that recall accuracy may improve wh en informants are given such prompts (for example, a checklist rather than a series of open ended questions) (1984) Studies of response accuracy also show that people are more accurate at recalling long term patterns or tendencies (such as overall freque ncy of communication), rather than specific events or relative frequencies (Freeman, et al. 1987) Consistent with this approach, I asked respondents about the general frequency of their phone interactions during the past year, rather than the specific number of phone interactions wit hin a short time frame (i.e., last week). Assessing informant accuracy about frequency of phone interactions While conducting how frequently do you and daily, wee kly, monthly, yearly, or never? I only int erviewed residents of villages, and did not interview any of their phone contacts residing outside the village. Therefore, m ost of my data are one sided reports of relationships However, my data do include 468 reciprocal ties which represent t he phone interactions of pairs of residents of the same village I can compare the

PAGE 376

376 consistency of the responses of these reciprocal pairs as a way of assess ing the overall frequency. If the responses of tw o phone owners match, that woul d seem to imply accurate recall of calling frequency, their recall. It is important to note that I am not testing the accuracy of informants against actual behavior (i.e. calling records), as Bernard, et al. did. Rather, I am simply testing the rate of agreement betwee n informants in their responses. Thus, even if two responses match, it does not strictly imply that the responses we re accurate relative to actual calling frequency Within the 3 study villages there were a total of 468 such reciprocal pairs ; pairs of village residents who had entered one another in their phone address books Out of these pairs only one of the respondents actually provided a survey response about calling frequency, or one of the respondents was confused about the identity of the owner of the other phone, or because the respo ndents were clearly referring to different individuals when giving their responses (as evidenced by the different kinship terms they used). Such confusion is particularly understandable with phone ties in the same village, because in Fiji names are recycle d generation after generation in the community, and there may be several individuals with the same name simultaneously residing in a village. The table below illustrates one such flawed pair removed from analysis For the 361 valid response pairs, I then

PAGE 377

377 between responses along the ordinal scale of possible responses about phone calling frequency: daily, weekly, monthly, yearly, never. Table C 5. Example of an entry removed from analysis of respondent accuracy Phone o wner Contact name in phone Kin Relationship Calling Frequency John Ruth Sibling W eekly Ruth John Cross cousin M onthly An example of one of the pairs I removed from my analysis of respondent accuracy. In this case, it appears that John was not giving inf ormation about his calling interactions with Ruth, but rather with interactions with John who according to Fijian kinship is her cross cousin through marriage Table C 6. Calculating divergences in responses about calling frequency Phone Owner #1 Phone Owner #2 Response of Owner #1 Response of Owner #2 Difference in degree Abel Karen Weekly Weekly 0 Samuel Franklin Daily Monthly 2 In order to assess respons ordinal variable and calculated the degree to which each response pair differed along the following ordered scale: Daily, Weekly, Monthly, Yearly, Never. In the following table, Ab about the frequency of their phone contact matched, so there was a difference in degree of 0. However, ies in between). rater agreement used for categorical variables. A weighted version of the Kappa coefficient can be used for ordinal variables. I used an SPSS analysis to determine the intraclass correlation coefficient (ICC), which is equivalent to a quadratic weighted Kappa test, to assess the consistency of responses for my 361 response pairs (Fleiss and Cohen 1973) The Average Measures ICC/weighted Kappa for 360 cases was .471 with a 95% confidence inter val ranging from .349 .570. According to the guidelines established by Landis and considered fair to good agreement. Perhaps a more intuitive way to assess response accurac y is to compare the actual distribution of responses to the distribution of responses that would be expected by chance, if informants were randomly guessing their responses. The figure below

PAGE 378

378 shows the distribution of the degrees of difference for the 361 p airs of respondents. In terms of the distribution of differences between responses, 37% of all response pairs matched exactly 44% differed by one d egree 15% differed by 2 degrees, 2% by 3 degrees and 2% by 4 degrees. In Figure C 2 the blue distribution (in front) indicates the discrepancies in the actual measured responses about calling frequency in 361 pairs of phone owners The green distribution (in back ) is the distribution of error we would expect if people were giving completely random guesses when answering the question about calling frequency. The theoretical (chance) and actual distributions of response error differ sign 2 (4, N=361) =167.47, p=.000. This chi square test indicates merely that my respondents performed better than pure chance in terms of the accuracy (consistency) of their responses. Figure C 2. Consistency of responses about calling frequenc y: Differences between responses in reciprocated pairs 37% of my response pairs matched perfectly in their responses, and 44% differed by only one degree (daily vs. weekly, or monthly vs. yearly, for example). Only

PAGE 379

379 19% of pairs had a discrepancy greater t han one degree This is certainly not perfect or as accurate as using actual calling records, but in practical terms, my question about calling frequency seems to have yie lded useful (and usable) data This measure of calling frequency can serve as a meas ure of tie strength for purposes of network analysis. Figure C 3. Inconsistency between responses about calling frequency. The distribution of degrees of difference between pairs of responses (n=361) about calling frequency along the scale daily, week ly, monthly, yearly, never. An alternative approach to testing response accuracy An alternative method of assessing response accuracy would be to rather than an ordinal variable In this approach value of 1. Analysis of calling frequency as a ratio variable yields a mean difference between re spondents of 200+/ 149 days per year for the 361 response pairs. By contrast t he median difference between respondents is only 40 days per year, and the mode is 0. Shading denotes degrees of difference between each pair of responses about calling frequency. Possible answers were: daily, weekly, monthly, yearly, never.

PAGE 380

380 Th is highly skewed distribution is due in part to the non linear progression of the units have a magnitude of 7, 4, and 12, respectively. This is one justification for tre ating More importantly perhaps, in the survey questionnaire, the question about calling frequency was not present ather, respondents probably interpreted this q uestion in terms of an ordinal scale with the intervals between the choices roughly equal in value. I therefore argue that my initial assessment of respondent accuracy (as an ordinal variable) is the more appropriate approach. Ethical Considerations Copyin g mobile phone directories for research purposes is a novel technique. It is important that any ethical concerns surrounding this technique be addressed in order to protect the rights and privacy of research participants. The primary concern is that many p eople consider the contents of their mobile phones to be highly personal, or even sensitive, information. Data must therefore be collected, analyzed, and presented in a manner that maintains the anonymity of research participants as well as the anonymity o f all of the contacts found in every telephone in the survey. Privacy safeguards must be implemented from the point that each survey begins, and continue even after the research is published. An effective way to ensure that participants can provide truly informed consent is to explain your research project as thoroughly and as often as possible. I began my stay in each community with a public presentation in the community hall, in which I explained my research and data collection techniques to the communit y at large. I again explained my research individually to each participant prior to obtaining informed consent at the

PAGE 381

381 beginning each survey. Part of this explanation included a mock demonstration of ho w the SIM reader works, and an explanation of the forma t in which their contacts would be copied to the computer screen. I also showed participants an example of a social network diagram to explain the underlying reason why I was interested in copying their phone contact data. As part of the process of obtaini ng consent, it should be stressed to the participant that their participation is entirely voluntary, and that they may opt out of any part of the survey, or the entire survey if they so wish. Before copying contacts to the computer, there are a few steps o f the process that require manipulating the phone and checking information on the phone screen. Whenever possible I asked the participant to manipulate their own phone, to press any necessary buttons themselves and to remove their own SIM card. Letting th e participants hold and manipulate their own phones provides them with a sense of control over the process and also protects the researcher from potential misunderstandings. After copying the contents of the SIM card to the computer, I adhered to the follo wing routine : 1. Before I looked at the computer screen, I turned the screen toward the participant, and asked them whether they wished for me to remove any of the displayed contacts from the spreadsheet. This was in case the participant did not want me to se e certain names or numbers, or did not wish for certain information to be included within the purview of my study. 2. If the participant wished to remove contacts I asked him/her to list the spreadsheet entry numbers of any of their contacts that they wish me to remove. 3. Before deleting the specified entries, I minimized the spreadsheet so that only names or numbers of the contacts to be deleted. 4. I then deleted these entries fro m the spreadsheet, once again showing the screen to the participant to demonstrate that these entries have been removed.

PAGE 382

382 5. I then save the contacts list to my computer as a password protected Excel CSV file. This last step deserves emphasis, because it is important that files being viewed or copied by individuals who are not associated with the research project. Finally, when the project data are presented at conferences or publ ished, the real names and/or phone numbers of participants should not be used. As an added precaution it is also advisable to use a fictitious name for the study community, although this can be left to the discretion of the researcher, and is ultimately su bject to the wishes of the community members. In practical terms, pursuing this research approach requires a degree of rapport and trust between the researcher and members of the study community. I had previously visited Toba and Cagi in 2009, and had est ablished relationships with many of the inhabitants. When conduct ing surveys, I attempted to visit a house only after I had been previously introduced to th e inhabitants in another, sociable setting. Finally, I tried to schedule my surveys in advance, so t hat participants could plan to be in a place where they would feel comfortable answering questions about their phone use. I conducted my surveys during the span of a number of weeks, which allowed me to increase the degree of trust and rapport in the commu nity as time progressed

PAGE 383

383 APPENDIX D ANALYSES OF CALLING NETWORKS This appendix includes a few charts that provide a more in depth look at analyses of calling networks. These charts are referenced in the text of Chapter 5. Phone Use Frequency/Intensity Figure D 1 Scatterplot of calls per day, by age and sex of phone owner Figure D 2 Scatterplot of texts per day, by age and sex of phone owner

PAGE 384

384 Figure D 3 Scatterplot of weekly phone credit expenditures by age and sex Figure D 4 Percentage o f respondents who named each individual as a heavy phone user Contents of Phone Contact Lists Table D 1 Variable Test Test Statistic Number of saved contacts Independent Samples Median test .604 # of calls and texts Independent Samples Median test .759 $ spent on phone credit Independent Samples Median test .920

PAGE 385

385 APPENDIX E CALCUL ATIONS OF THE MONETARY VALUE OF EXCHANGED GOODS All prices and monetary values in the preceding chapters are listed in Fiji Dollars ($FJ). On December 31, 2011 (the approximate midpoint of my field research) the exchan ge rate was $FJ 1= $US 0.54. I collected survey data about goods exchanged during the previous 12 months in each of the 99 households in the study villag es. I calculated a monetary value for these goods based on market prices of produce For m ajor cash crops (such as taro) I derived price data from Fiji Government publications, and for minor crops (such as tomatoes, hot peppers or eggplant) I personally ch ecked the prices in the Suva market. The market prices of the main cash crops (such as taro, cassava, and yams) can fluctuate considerably throughout the year according to the harvest seasons. Because I did not have precise dates for the exchange events, I calculated the value of these crops based on an average of published monthly market prices for 2011 Using this approach I arrived at the following prices: Taro ( dalo ): $FJ 0.95/kg Cassava ( tavioka ): $FJ 0.77/kg Yams ( uvi ): $FJ 1.27/kg Coconuts ( niu ): $FJ 4.60/dozen Kava ( yaqona ): $FJ 31.91/k g (this assumes kava consisting of 50% chips ( lewena ) and 50% roots ( waka ) by weight) S urvey responses rarely estimated exchanged produce quantities in terms of weight however Instead, respondents quoted the quantit y of units of each item exchanged. Units include d bundles ( ivesu ), stringers ( itui ), stalks (of bananas or plantains), woven palm leaf baskets ( isu ), poly bags of various sizes ( taga ), and plastic buckets ( vokete ). Fortunately, buckets and poly bags come i n a few standard sizes. I

PAGE 386

386 asked several salesmen in the Suva market for estimates of the capacity (by weight, count, or monetary value) of these various containers as the basis for my calculations. Bundles (typically of taro) weigh about 10kg Stringers (o f fish) vary in contents and price but are typically $20 in the market Stalks (of plantains) cost roughly $10 in the market and represent of a bag ful Small (square) bucket: used to transport small tree fruits. One bucket of tangerines sells for $FJ 10. L arge (round) bucket: likewise used for small tree fruits. One bucket of tangerines sells for $FJ 15. Poly bag: holds 35 kg of taro or yams, or 40 kg of cassava, or 36 husked coconuts. (this is a middle ground estimate derived by combining the capacity of t he more common 25 sugar bags typically used in the market) Woven basket (isu): According to market salesmen, the capacity of a typical isu is approximately that of a poly bag. Clearly, the resulti ng figures for value of exchanged produce are only estimates based on calculations employing rounding and generalization. However, I was able to use these calculations for all three study sites, providing a basis for compari s ons between households and comm unities.

PAGE 387

387 LIST OF REFERENCES 2008 Premarital sex 'a curse'. In Fiji Times Online. Suva: Fiji Times. 2009 Teen pregnancy worries. In Fiji Times Online. Suva. 2013 CIA World Factbook: Fiji Islands, Vol. 2013. Adamic, Lada and Natalie Glance 2005 The Poli tical Blogosphere and the 2004 US Election: Divided They Blog. Proceedings of the 3 rd International Workshop on Link Discovery: 36 43. Ake r, Jenny C. and Isaac M. Mbiti 2010 Mobile Phones and Economic Development in Africa (June 1, 2010). Center for Globa l Development Working Paper No. 211. ( ). Altheide, D.L. 1994 An ecology of communication: Toward a mapping of the effective environment. The Sociological Quarterly 35(4):665 683. Anonymous 2007 Government warned of harm in blocking blog spots. In Fiji Times Online. Suva: Fiji Times Online. Aoyama, Yuko and Eric Sheppard 2003 The dialectics of geographic and virtual space. Environment and Planning A 35(7):1151 1156. Arno, Andrew 1993 The world of talk on a Fijian island : an ethnography of law and communicative causation. Norwood, N.J.: Ablex Pub. Corp. Barth, Fredrik, ed. 1969 Ethnic Groups and Boundaries: The Social Organization of Culture Difference. Oslo: Universitetsforlaget. Bayliss Smith, T., Bedford, R., Brookfield, H., and M. Latham 1988 Islands, Islanders, and the World : The Colonial and Post colonial Experience of Eastern Fiji. Cambridge Cambridge University Press. Bayliss Smith, T.P. 1976 Koro in the 1970s: Prosperity through Diversity? Canberra: Australian National University. Becker, Anne E. 1995 Body, self, and society : the view from Fiji. Philadelphia: University of Pennsylvania Press.

PAGE 388

388 Bedford, R., ed. 1978 The Small Islands and the Reefs. Melbourne: UNESCO/UNFPA. Belshaw, Cyril S. 1964 Under the Ivi Tree: Society and Economic Growth in Rural Fiji. Berkeley: University of California Press. Bernard, H. Russell, and P.D. Killworth 1976 Informant accuracy in social network data. Human Organization (35):269 286. 1977 Informant accuracy i n social network data II. Human Communication Research (4):3 18. Bernard, H.R., et al. 1984 The Problem of Informant Accuracy: The Validity of Retrospective Data. Annual Review of Anthropology 13:495 517. Bernard, H.R., Peter Killworth, and Lee Sailer 1980 Informant Accuracy in Social Network Data, IV. A Comparison of Clique level Structure in Behavioral and Cognitive Data. Social Networks 2:191 218. 1982 Informant Accuracy in Social Network Data: An Experimental Attempt to Predict Actual Communicati on from Recall Data. Social Science Research 11:30 66. Bertram, G., and B. Poirine 2007 Island Political Economy. In A World of Islands. G. Baldacchino, ed. Malta: Agenda. Bijker, W., T. Hughes, and T. Pinch, eds. 1987 The Social Construction of Techno logy. Cambridge: MIT Press. Bordieu, Pierre 1977 Outline of a Theory of Practice. Cambridge: Cambridge University Press. Brinkman, Inge, Mirjam de Bruijn, and Hisham Bilal 2009 The mobile phone, 'modernity' and change in Khartoum, Sudan. In Mobile Phon es: The New Talking Drums of Everyday Africa. M. de Bruijn, F. Nyamnjoh, and I. Brinkman, eds. Pp. 69 91: Langaa & African Studies Centre. Britton, Stephen G. 1980 The evolution of a colonial space economy: The case of Fiji. Journal of Historical Geograp hy 6:251 274.

PAGE 389

389 Cairncross, Frances 2001 The Death of Distance: How the Communications Revolution is Changing Our Lives. Cambridge: Harvard Business School. Cambie, R. C., and J. Ash 1994 Fijian medicinal plants. [Australia]: CSIRO Australia. Castells M., et al. 2007 Mobile Communication and Society: A Global Perspective. Cambridge: MIT Press. Castells, Manuel 2004 The Network Society: A Cross Cultural Perspective. Northampton: Edward Elgar. Chan, W. and B. Wellman 2004 The global digital divide Within and between countries. IT & Society 1(7):39 45. Chapman, Murray, and R. Mansell Prothero 1985 Circulation in Population Movement : Substance and Concepts from the Melanesian Case. London: Routledge & Kegan Paul. Clammer, J. R. 1976 Literacy and social change : a case study of Fiji. Leiden: Brill. Connell, John 2006 'The Taste of Paradise': Selling Fiji and FIJI Water. Asia Pacific Viewpoint 47(3):342 350. Corbett, S. 2008 Can the Cellphone Help End Global Poverty? In New York Times. New York City. Couclelis, Helen 2004 Pizza over the Internet: e commerce, the fragmentation of activity and the tyranny of the region. Entrepreneurship & Regional Development: An International Journal 16(1):41 54. Donner, J. 2008 Research approaches to mobile use in the developing world: A review of the literature. Information Society 24(3):140 159. Drucker, Peter 1969 The Age of Discontinuity. London: Heinemann.

PAGE 390

390 Dutton, W. 2005 The Internet and Social Transformation: Reconfiguring Access. In Transforming Enterprise: Economic and Social Implications of Information Technology. W. Dutton, Kahin, B., O'Callaghan, R., Wykoff, A., ed. Pp. 375 394. Cambridge: MIT. Espinoza, Vicente 1999 Social Networks Among the Urban Poor. In Networks in the Global Village. B Wellman, ed. Pp. 147 184. FBS 2007 Census 2007 of Population and Housing. F.B.o. Statistics, ed. Suva, Fiji. Fleiss, JL and Cohen, J. 1973 The equivalence of weighted kappa and the intraclass correlation coefficient as measures of reliability. Educat ional and Psychological Measurement 33:613 619 Fortunati, Leopoldina, James Katz, and Raimonda Riccini, eds. 2003 Mediating the Human Body: Technology, Communication, and Fashion. Malwah: Lawrence Erlbaum Associates. Foth, M, and Greg Hearn 2007 Net worked Individualism of Urban Residents: Discovering the communicative ecology in inner city apartment buildings. Information, Communication and Society 10(5). Freeman, Linton C, A. Kimball Romney, and Sue Freeman 1987 Cognitive Structure and Informant A ccuracy. American Anthropologist 89:310 325. Giddens, Anthony 1990 The Consequences of Modernity. Cambridge: Polity Press. Giddens, Anthony 1991 Modernity and Self Identity. Cambridge: Polity Press. Goodman, J. 2007 Linking Mobile Phone Ownership and Use to Social Capital in Rural South Africa and Tanzania. In The Vodaphone Policy Paper Series, Vol. 3. Groves, Murray 1963 The Nature of Fijian Society. Journal of the Polynesian Society 72:272 291. Guadagno, Rosanna E., Bradley M. Okdie, and Sara A. Kruse 2012 Dating deception: Gender, online dating, and exaggerated self presentation. Computers in Human Behavior 28(2):642 647.

PAGE 391

391 Hocart, A. M. 1929 Lau Islands, Fiji. Honolulu: The Bernice P. Bishop Museum. Horst, H. 2006 The Blessings and Burdens of Communication: Cell Phones in Jamaican Transnational Social Fields. Global Networks 6(2):143 159. Horst, Heather A., and Daniel Miller 2005 From Kinship to Link up: Cell Phones and Social Networking in Jamaica. Current Anthropology 46(5):755 778. Horst Heather A., and Daniel Miller 2007 The Cell Phone : An Anthropology of Communication. Oxford ; New York: Berg. Hossmann, KA and DM Hermann 2003 Effects of electromagnetic radiation of mobile phones on the central nervous system. Bioelectromagnetics 24 (1):49 62. Ito, M. 2005 Mobile phones, Japanese youth, and the re placement of social contact. In Mobile Communications: Re negotiation of the Social Sphere. R. Ling and P. Pedersen, eds. Pp. 131 148. New York: Springer Verlag. ITU 2013 World Telecommu nication/ICT Indicators Database Online, Vol. 2013. Geneva: International Telecommunications Union. James, Jeffrey and Mila Versteeg 2007 Mobile phones in Africa: how much do we really know? Social Indicators Research 84:117 126. Jensen, R. 2007 The Dig ital Provide: Information (Technology), Market Performance, and Welfare in the South Indian Fisheries Sector. Quarterly Journal of Economics 122(3):879 924. Katz, J.E., ed. 2008 Handbook of Mobile Communication Studies. Cambridge, Massachusetts: MIT Pres s. Katz, James E., and Satomi Sugiyama 2005 Mobile Phones as Fashion Statements: The Co creation of In Mobile Communications: Re negotiation of the Social Sphere. R. Ling and P.E. Pedersen, eds. London: Springer Ver lag.

PAGE 392

392 Kema, I. 2005 Globalization and the development of underdevelopment. Journal of Third World Studies (Spring). Latour, B. 2005 Reassembling the Social: An Introduction to Actor network Theory. Oxford: University Press. Ling, Richard Seyler 1988 S and mobile telephony for social networking among Norwegian parents. R&D Report 33/98. Kjeller, Norway, Telenor. Ling, Richard Seyler 2004 The Mobile Connection : The Cell Phone's Impact on Society. San Francisco: Morgan Kaufmann. Lipton, M. 1980 Migration from Rural Areas of Poor Countries Impact on Rural Productivity and Income Distribution. World Development 8(1):1 24. MacKenzie, D. and J. Wajcman 1985 The Social Shaping o f Technology. Buckingham: Open University Press. Malinowski, B. 1923 The Problem of Meaning in Primitive Languages. In The Meaning of Meaning. C. Ogden and I. Richards, eds. London: Routledge. Massey, D.S., L. Goldring, and J. Durand 1994 Continuities in transnational migration: an analysis of nineteen Mexican communities. American Journal of Sociology 99(6):1492 1533. McLuhan, Marshall 1962 The Gutenberg Galaxy: The Making of Typographic Man. Toronto: University of Toronto Press. Mechael, Patricia 2008 Health Services and Mobiles: A Case from Egypt. In Handbook of Mobile Communication Studies. J. Katz, ed. Pp. 91 104. Cambridge, Massachusetts: MIT Press. Miller, D., and D. Slater 2000 The internet: An ethnographic approach. Oxford: Berg. Mitchell Kimberly J., et al. 2012 Prevalence and Characteristics of Youth Sexting: A National Study. Pediatrics 129(1):13 20.

PAGE 393

393 Nayacakalou, R. R. 1978 Tradition and Change in the Fijian village. Suva: South Pacific Social Sciences Association. Neall, Vincent E. and Steven A. Trewick 2008 The age and origin of the Pacific islands: a geological overview. Philosophical Transactions of the Royal Society B 363:3293 3308. Nunn, Patrick D., et al. 2004 Early Lapita Settlement Site at Bourewa, Southwest Viti Levu I sland, Fiji. Archaeology in Oceania 39(3):139 143. Oksman, Virpi, and Pirjo Rautiainen 2003 "Perhaps It is a Body Part": How the Mobile Phone Became an Organic Part of the Everyday Lives of Finnish Children and Teenagers. In Machines that Become Us: The Social Context of Personal Communication Technology. J.E. Katz, ed. Pp. 293 New Brunswick: Transaction Publishers. Onnela, J., et al. 2007 Structure and tie strengths in mobile communication networks. Proceedings of the National Academy of Sciences of the United States of America 104:7332 7336. Ortner, Sherry B. 1984 Theory in Anthropology since the Sixties. Comparative Studies in Society and History 261(1):126 166. Pertierra, R., et al. 2002 TXT ING Selves: Cellphones and Philippine Modernity. Manil la: De La Salle University Press. Pfaffenberger, B. 1992 Social anthropology of technology. Annual Review of Anthropology 21:491 516. Rainie, Lee and Barry Wellman 2012 Networked: The New Social Operating System. Cambridge: MIT Press. Rasmussen, Terje 2000 Social Theory and Communication Technology. Burlington: Ashgate. Rawalai, Luke 2013 Police to monitor internet cafes. In The Fiji Times Online. Suva: Fiji Times Online. Reddy, M., M. Mohanty, and V. Naidu 2004 Economic cost of human capital loss from Fiji: Implications for sustainable development. International Migration Review 38(4):1447 1461.

PAGE 394

394 Rege, Aunshul 2009 What's Love Got to Do with It? Exploring Online Dating Scams and Identity Fraud. International Journal of Cyber Criminology 3(2):494 512. Rice, Ronald E. and James E. Katz 2003 Comparing internet and mobile phone usage: digital divides of usage, adoption, and dropouts. Telecommunications Policy 27:597 623. Sahlins, Marshall David 1958 Social stratification in Polynesia. Seattle,: Un iversity of Washington Press. 1962 Moala: Culture and Nature on a Fijian island. Ann Arbor: University of Michigan Press. 1972 Stone age economics. Chicago,: Aldine Atherton. Samuel, J., N. Shah, and W. Hadingham 2007 Mobile Communications in South Africa, Tanzania, and Egypt: Results from Community and Business Surveys, Vol. 3: Vodafone Policy Paper Series. Schmink, Marianne. 1984 Household Economic Strategies: Review and Research Agenda. Latin American Research Review 19:87 101. Skuse, A., and T. Cousins 2007 Managing distance: Rural poverty and the promise of communication in post apartheid South Africa. Journal of Asian and African Studies 42(2):185 207. Small, C. 1997 Voyages: From Tongan Villages to American Suburbs. Ithaca: Cornell Univ ersity Press. Spate, O. H. K. 1959 The Fijian People: Economic Problems and Prospects: A Report. Suva. Stark, O., J. E. Taylor, and S. Yitzhaki 1988 Migration, remittances and inequality a sensitivity analysis using the extended Gini index. Journal of Development Economics 28(3):309 322. Stark, O., J. Taylor, and S. Yitzhaki 1986 Remittances and Inequality. The Economic Journal 96(383):722 740. Stephen, G, and S Marvin 1996 Telecommunications and the City Electronic Spaces, Urban Places. London: R outledge.

PAGE 395

395 Sullivan, Nicholas P. 2007 You Can Hear Me Now : How Microloans and Cell Phones are Connecting the World's Poor to the Global Economy. San Francisco: Jossey Bass. Tacchi, Jo, Don Slater, and Greg Hearn 2003a Ethnographic Action Research. UN ESCO, ed. New Delhi: UNESCO. Tacchi, Jo, Don Slater, and Peter Lewis 2003b Evaluating Community based Media Initiatives: An Ethnographic Action Research Approach. In IT4D. Taylor, Robin 2003 The Family Law Act Resourcing Implications for Social Welfar e: Internal Report for the Ministry of Women, Social Welfare and Poverty Alleviation. Suva: Ministry of Women. TFL 2013 History of TFL: Development of Telecommunications, Vol. 2013. Suva: Telecom Fiji Limited. Thomas, N. 1992 The Inversion of Tradition American Ethnologist 19(2):213 232. Thompson, Laura 1940a Fijian frontier. San Francisco, New York American Council, Institute of Pacific Relations. 1940b Southern Lau, Fiji: an ethnography. Honolulu, Hawaii,: The Museum. Tokola, Kari, Kurttio, Pai vi, Salminen, Tiina, and Anssi Auvinen 2008 Reducing Overestimation in Reported Mobile Phone Use Associated with Epidemiological Studies. Bioelectromagnetics 29:559 563. Toren, C. 1999 Mind, Materiality, and History. London: Routledge. Toren, Christina 1990 Making sense of hierarchy : cognition as social process in Fiji. London [England] ; Atlantic Highlands, NJ: Athlone Press. Valemei, Ropate 2013 Mobile app for health. In The Fiji Times. Suva. Van Dijk, D. 1999 The network society: Social aspects of new media. Thousand Oaks, CA: Sage.

PAGE 396

396 Vrijheid, M., Cardis, E., Armstrong, B K ., Auvinen, A., Berg, G., Blaasaas, K.G., Brown, J., Carroll, M., Chetrit, A., Christensen, H.C., Deltour, I., Feychting, M., Giles, G.G., Hepworth, S.J., Hours, M., Iavarone, I., Johansen, C., Klaeboe, L., Kurttio, P., Lagorio, S., Lonn, S., McKinney, P.A., Montestrucq, L., Parslow, R.C., Richardson, L., Sadetzki, S., Salminen, T., Schuz, J., Tynes, T., and A Woodward. 2006 Validation of short term recall of mobile phone use for the interphone study. Occup Environ Med 63(4):237 243. Watters, R. F. 1969 Koro: Economic Development and Social Change in Fiji. Oxford: Clarendon P. Waverman, L., M. Meschi, and M. Fuss 2007 The Impact of Telecoms on Economic Growth in Developing Nations. In Moving the Debate Forward: The Vodafone Policy Paper Series #3. Wellman, Barry, ed. 1999 Networks in the Global Village: Life in Contemporary Communities. Boulder: Westview Press. White, Leslie A. 1959 The evolution of culture; the develop ment of civilization to the fall of Rome. New York,: McGraw Hill. Wilson, FA and JP Stimpson 2010 Trends in fatalities from distracted driving in the United States, 1999 2008. American Journal of Public Health 100(11):2213 2219.

PAGE 397

397 BIOGRAPHICAL SKETCH Rya n Peseckas is a cultural anthropologist who specializes in issues surrounding technology and social change. He has conducted ethnographic research in the Caribbean and South Pacific and has a particular interest in issues r elated to islands. Ryan earned hi s PhD from the University of Florida in May 2014 Ryan received a Bachelor of Science degree in Marine Science from Eckerd College in 2003. He spent the next two years in rural Fiji as a Peace Corps volunteer, working on marine resource conservation projec ts with indigenous Fijian villagers. Upon his return to the U.S. Ryan entered the Anthropology graduate program at the University of Florida. nological disparities influence Haitian and Dominican fishi ng activities on the southern Haitian Dominican border, and how these factors ultimately impact fisheries management efforts. In 2009 Ryan spent two months in Fiji and was struck by the impacts of telecommunications reforms that had occurred since his tim e there in the Peace Corps. He returned to Fiji in 2011 on a Fulbright Scholarship, attached to the Social Sciences faculty at the University of the South Pacific. Over the course of 10 months he conducted fieldwork in three rural villages, examining the u ses and impacts of mobile technology and internet in the context of kinship, exchange of material resources, and personal relationships. He collected social network data via an innovative technique: Aside from a Fulbright Scholarship, Ryan has also obtained a number of grants via the University of Florida, including a four year Alumni Fellowship, the Center for

PAGE 398

398 and he was a two time recipient of the Doughty field research grant from the Department of Anthropology. At the University of Florida Ryan has designed and taught three undergraduate courses: Technology and Society, Island Studies, and Peoples and Cultures of Oceania, and assistant instructed a variety of courses including Applied Good Life?