Studying Japan: Handbook of Research Designs, Fieldwork and Methods 3848750856, 9783848750856

Citation preview

Nora Kottmann | Cornelia Reiher [eds.]

Studying Japan Handbook of Research Designs, Fieldwork and Methods

Nora Kottmann | Cornelia Reiher [eds.]

Studying Japan Handbook of Research Designs, Fieldwork and Methods

BUT_Kottmann_5085-6.indd 3

16.11.20 13:18

© Coverpicture: Robin Weichert

The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at http://dnb.d-nb.de ISBN

978-3-8487-5085-6 (Print) 978-3-8452-9287-8 (ePDF)

British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library. ISBN

978-3-8487-5085-6 (Print) 978-3-8452-9287-8 (ePDF)

Library of Congress Cataloging-in-Publication Data Kottmann, Nora | Reiher, Cornelia Studying Japan Handbook of Research Designs, Fieldwork and Methods Nora Kottmann | Cornelia Reiher (eds.) 501 pp. Includes bibliographic references and index. ISBN

978-3-8487-5085-6 (Print) 978-3-8452-9287-8 (ePDF)

Onlineversion Nomos eLibrary

1st Edition 2020 © Nomos Verlagsgesellschaft, Baden-Baden, Germany 2020. Printed and bound in Germany. This work is subject to copyright. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage or retrieval system, without prior permission in writing from the publishers. Under § 54 of the German Copyright Law where copies are made for other than private use a fee is payable to “Verwertungs­gesellschaft Wort”, Munich. No responsibility for loss caused to any individual or organization acting on or refraining from action as a result of the material in this publication can be accepted by Nomos or the editors.

BUT_Kottmann_5085-6.indd 4

16.11.20 13:18

Table of contents List of figures ..............................................................................................

11

List of tables ................................................................................................

12

Foreword Franz Waldenberger ...............................................................................

13

Ilse Lenz ..............................................................................................

14

Acknowledgements .......................................................................................

17

Introduction: Studying Japan ........................................................................... Nora Kottmann and Cornelia Reiher

19

Chapter 1

29

How to begin research: The diversity of Japanese Studies ....................... Roger Goodman

1.1 Positioning one’s own research in Japanese Studies: Between Area Studies and discipline ............................................................................................ Verena Blechinger-Talcott

40

1.2 Let the field be your guide ....................................................................... Daniel P. Aldrich

43

1.3 Studying marriage in Japan: A social anthropological approach ....................... Joy Hendry

47

Further reading .....................................................................................

51

References ...........................................................................................

51

Chapter 2

How to ask: Research questions ....................................................... Gabriele Vogt

53

2.1 Your research questions may change and that is ok ....................................... Nicolas Sternsdorff-Cisterna

65

2.2 Studying Japanese political behaviour and institutions ................................... Kenneth Mori McElwain

68

2.3 Capturing social change in Japan .............................................................. David Chiavacci

72

Further reading .....................................................................................

76

References ...........................................................................................

76

Chapter 3

How to organise research: Research designs ........................................ Kaori Okano

3.1 Developing a comparative study: Single women in Hong Kong, Tokyo and Shanghai ............................................................................................. Lynne Y. Nakano

78

88

5

Table of contents 3.2 Contained serendipity as fieldwork in Japan: Studying Chinese people in Japan ... Jamie Coates

91

3.3 The universe of cases: Agricultural cooperatives in Japan as a case study ........... Kay Shimizu

95

Further reading .....................................................................................

99

References ...........................................................................................

99

Chapter 4

How to identify relevant scholarly debates: Reviewing the literature ......... Urs Matthias Zachmann

102

4.1 Looking for sources in all the right places ................................................... Patricia L. Maclachlan

117

4.2 Ambiguity and blurred boundaries: Contextualising and evaluating heterogeneous sources ............................................................................................... Sonja Ganseforth

121

4.3 Doing migration research in Japan: The roles of scholarly literature .................. Gracia Liu-Farrer

125

Further reading .....................................................................................

130

References ...........................................................................................

130

Chapter 5

How to collect data: An introduction to qualitative Social Science methods ...................................................................................... Akiko Yoshida

132

5.1 Participant observation and interviews: Going with the flow and dipping in and out .................................................................................................... Emma E. Cook

142

5.2 Transnational research in Japan Studies—an oxymoron? Studying cross-border labour mobility in globalising Japanese production organisations ..................... Karen Shire

146

5.3 ‘Bullseye view on happiness’: A qualitative interview survey method .................. Barbara Holthus and Wolfram Manzenreiter

151

Further reading .....................................................................................

155

References ...........................................................................................

155

Chapter 6

How to do fieldwork: Studying Japan in and outside of Japan ................. Levi McLaughlin

6.1 The cosmology of fieldwork: Relationship building, theoretical engagement and knowledge production in Japan Anthropology ............................................. Nana Okura Gagné

6

157

169

Table of contents 6.2 A mobilities approach to ‘Japan’ fieldwork .................................................. James Farrer

173

6.3 Building arguments on national policies from everyday observations ................. Hanno Jentzsch

177

Further reading .....................................................................................

181

References ...........................................................................................

181

Chapter 7

How to interview people: Qualitative interviews .................................. Nora Kottmann and Cornelia Reiher

184

7.1 The empire of interviews: Asking my way through Japan ............................... Christoph Brumann

196

7.2 The art of interviewing: A Japanese perspective ............................................ Tomiko Yamaguchi

200

7.3 Talking through difficult topics ................................................................ Allison Alexy

204

Further reading .....................................................................................

208

References ...........................................................................................

208

Chapter 8

How to observe people and their environment: Participant observation ..... Christian Tagsold and Katrin Ullmann

8.1 Of serendipities, success and failure and insider/outsider status in participant observation ......................................................................................... Susanne Klien

211

223

8.2 Doing and writing affective ethnography ..................................................... Akiko Takeyama

227

8.3 Reflections on fieldwork in post-bubble Japan: Gender, work and urban space .... Swee-Lin Ho

231

Further reading .....................................................................................

235

References ...........................................................................................

235

Chapter 9

How to access written and visual sources: Archives, libraries and databases .................................................................................... Theresia Berenike Peucker, Katja Schmidtpott and Cosima Wagner

238

9.1 Clever approaches to tricky sources: How to extract information from business archives and war memorials .................................................................... Katja Schmidtpott and Tino Schölz

248

9.2 Writing transnational history through archival sources .................................. Sheldon Garon

252

7

Table of contents 9.3 Accessing quantitative data for qualitative research: White Papers, official statistics and micro datasets .................................................................... Shinichi Aizawa and Daisuke Watanabe

256

Further reading .....................................................................................

261

References ...........................................................................................

261

Chapter 10 How to combine methods: Mixed methods designs .............................. Carola Hommerich and Nora Kottmann

264

10.1 Reflections on multi-method research ......................................................... Robert J. Pekkanen and Saadia M. Pekkanen

283

10.2 Texts, voices and numbers: Using mixed methods to sketch social phenomena .... Laura Dales

287

10.3 Examining facts from different angles: The case of the deregulation of employment relations in Japan ................................................................. Jun Imai

292

Further reading .....................................................................................

297

References ...........................................................................................

297

Chapter 11 How to analyse data: An introduction to methods of data analysis in qualitative Social Science research ..................................................... David Chiavacci

300

11.1 Negotiating the ethics of gathering research data in a subcultural context .......... Katharina Hülsmann 11.2 Researching sex and the sexuality of Japanese teenagers: The intricacies of condom use ......................................................................................... Genaro Castro-Vázquez

310

313

11.3 Studying economic discourse ................................................................... Markus Heckel

317

Further reading .....................................................................................

321

References ...........................................................................................

321

Chapter 12 How to make sense of data: Coding and theorising ............................... Caitlin Meagher

323

12.1 Cresting the wave of data ....................................................................... Nancy Rosenberger

335

12.2 Lost in translation? Grounded theory and developing theoretical concepts .......... Celia Spoden

339

12.3 Coding: Mapping the mountains of ethnographic post-disaster data ................. Julia Gerster

343

8

Table of contents Further reading .....................................................................................

347

References ...........................................................................................

347

Chapter 13 How to systematise texts: Qualitative content and frame analysis ............. Celeste L. Arrington

349

13.1 Qualitative content analysis: A systematic way of handling qualitative data and its challenges ........................................................................................ Anna Wiemann

363

13.2 Analysis of biographical interviews in a transcultural research process ............... Emi Kinoshita

367

13.3 Qualitative content analysis and the study of Japan’s foreign policy .................. Kai Schulze

371

Further reading .....................................................................................

375

References ...........................................................................................

375

Chapter 14 How to understand discourse: Qualitative discourse analysis .................. Andreas Eder-Ramsauer and Cornelia Reiher

377

14.1 Media buzzwords as a source of discourse analysis: The discourse on Japan’s herbivore men ....................................................................................... Annette Schad-Seifert

389

14.2 Analysing affect, emotion and feelings in fieldwork on Japan .......................... Daniel White

393

14.3 From buzzwords to discourse to Japanese politics ......................................... Steffen Heinrich

397

Further reading .....................................................................................

402

References ...........................................................................................

402

Chapter 15 How to finish: Writing in a stressful world ......................................... Chris McMorran

405

15.1 Training your ‘writing muscle’: Writing constantly and theoretically ................. Aya H. Kimura

414

15.2 Writing stories ...................................................................................... Christian Tagsold

418

15.3 Writing about Japan .............................................................................. Richard J. Samuels

422

Further reading .....................................................................................

426

References ...........................................................................................

426

9

Table of contents Chapter 16 How to conduct reliable and fair research: Good research practice ........... Cornelia Reiher and Cosima Wagner

428

16.1 Fairness in research and publishing: The balancing act of cultural translation ...... Isaac Gagné

442

16.2 Digital oral narrative research in Japan: An engaged approach ........................ David H. Slater, Robin O’Day, Flavia Fulco and Noor Albazerbashi

446

16.3 Writing for publication: Eight helpful hints ................................................. Christopher Gerteis

450

Further reading .....................................................................................

452

References ...........................................................................................

452

Chapter 17 How to present findings: Presenting and publishing .............................. James Farrer and Gracia Liu-Farrer

455

17.1 Finding an audience: Presenting and publishing in Japanese Studies ................... Scott North

466

17.2 Ethnographic film and fieldwork on active ageing in rural Japan ...................... Isabelle Prochaska-Meyer

470

17.3 Weird and wonderful: Popularising your research on Japan ............................ Brigitte Steger

474

Further reading .....................................................................................

478

References ...........................................................................................

478

Notes on contributors ....................................................................................

480

Index .........................................................................................................

493

10

List of figures Figure 1.1:

Heuristic overview of sociological theory

33

Figure 1.2:

Heuristic model of the relationship between structuralist and interpretative theories and the methods they use

34

Figure 1.3:

Some heuristic dichotomies for thinking about research in Area Studies

36

Figure 5.1:

Bullseye chart in practice, three examples

154

Figure 10.1: Three core mixed methods designs

269

Figure 14.1: Three concepts related to discourse

379

Figure 14.2: Key steps in discourse analysis

381

Figure 14.3: Questions in Social Science discourse research

383

Figure 14.4: An overview of Critical Discourse Analysis

384

11

List of tables Table 5.1:

What to consider in choosing data collection methods

138

Table 9.1:

List of main White Papers in Japan

257

Table 12.1:

The process of coding

324

Table 12.2:

Elicited and extant text

327

12

Waldenberger

Franz

Foreword ‘Anything goes, as long as it is relevant and convincing.’ This guidance by my supervisor sounded like an invitation to confidently rely on my curiosity and creativity when doing research for my PhD back in the late 1980s. But I soon learned to translate the statement into ‘Anything goes, as long as it complies with the rules.’ The rules set by the academic community defined what was relevant and convincing. Methods form an integral part of this. They are the tools and rules of the trade of scholars: as tools they enhance our abilities to explore, test and verify, as rules they constrain what is acceptable. German Japanese Studies mostly differs from the more traditional Japanology with regard to its focus on subjects beyond culture, literature and language. When the new academic community started to establish itself at German-speaking universities in the 1980s, it had no genuine methodology. Instead it borrowed from the so-called Methodenfächer (method subjects) like Sociology, Political Science or Economics. But how could methods developed by disciplines that favour theories which are abstract from time and space be usefully applied to academic enquiry interested in phenomena that are defined by specific time-space constellations, like the family in post-war Japan or Japanese firms in the 1990s? Anthropology provides a solution as it offers a methodology which explicitly honours timespace contingencies, and some of the best research on Japan, like Ronald Dore’s classic British factory—Japanese factory (1973), has been achieved by applying anthropological methods. However, not all issues in the realm of management, the economy, politics and society lend themselves to anthropological methods. So, scholars in the field of Japanese Studies continue to be confronted with the tension between research interests about phenomena specific to Japan and research methods not primarily concerned with specifics. The handbook Studying Japan does not resolve this tension, but it does provide a pragmatic way of coping with it. And it does so in a comprehensive and systematic manner. By making the various methods of the Social Sciences accessible and by offering guidance on how to apply these tools and rules during the different stages of a research project, this handbook will prove highly valuable for those who study, teach and do research on Japan. Given its pluralistic approach, the handbook does not proclaim that there is only one right way to conduct research. It has no intention of being the Bible of Japanese Studies, but it certainly has the potential to become The book of recipes on how to make one’s research both relevant and convincing. The editors deserve both thanks and respect for taking up the challenge of embarking on this project as well as for what they already accomplished with the conference in 2019 and now with the timely publication of this handbook. The German Institute for Japanese Studies (DIJ) in Tokyo is very happy and proud to have been part of this endeavour. Franz Waldenberger Director, German Institute for Japanese Studies Tokyo, July 2020

13

Lenz

Ilse

Intercultural research, methodology and the emerging space of transnational knowledge When people from other corners of the world do qualitative research on Japanese contexts, they engage in an intercultural enterprise. I am not speaking of closed national cultures in terms of methodological nationalism. In this globalising world, the mass media, personal travel and capitalism have contributed to opening up and interlinking cultures: people in many places watch anime on the Internet, eat sushi of diverse quality and wear trousers produced by low-paid female workers from the Global South. But this has not resulted in a globalised, flattened world culture. Rather, cultures have been and are thriving as contradictory complex configurations of meaning and practices, and they blend elements from what is seen as home or far away. In this sense, those not socialised in the Japanese context and language start on an intercultural tour when they decide to do research on social or cultural issues focusing on Japan. This approach of intercultural interaction, communication and interpretation can bring new perspectives to the study of Japan, which of course is already comprehensively covered by Japanese researchers. This book is a detailed, diverse and extremely useful travel guide and companion on the road to reflexive and successful intercultural research in and on Japan. I want to congratulate the editors for this constructive and timely collection. They belong to the middle generation of researchers and thus show rich expertise in identifying and handling the various challenges of qualitative research on Japan. Like other pioneers in Germany, I had to find my way through the confusion, traps and thickets on this road mainly on my own with some support from advisors in Japan and elsewhere, when researching gender in industrialisation and later in industrial computerisation in Japan from the 1970s. Therefore, I find it extremely gratifying that younger generations can refer to this compendium on the why, how and where of doing research in Japan. Let me go on with the why, how and where: intercultural and transcultural research is an urgent issue for Cultural and Social Sciences in globalisation (Gerharz 2021; Rosenthal 2018). However, it is charged with tensions which are also present in the national context but less visible. Let me touch on some basic issues while drawing on the rich suggestions from the articles in this volume. The first is the relationship between the researcher and the researched subjects: the main aim of qualitative research is to bring to light and to interpret how actors as subjects see and construct sociocultural contexts and themselves (Rosenthal 2018). As researchers often used to see themselves as the main subjects of their projects, this creates tensions which have been debated as the representation problem or crisis in intercultural research (Gerharz 2021). Researchers and actors enter interactions in qualitative research as a process of cocreation (see Bruman, Ch. 7.1). As many contributors highlight in this handbook, (self-)reflexivity is an indispensable compass or everyday eyeglasses for researchers on the intercultural research road. They need to reflect on their own interest in the research issue and on the interaction, including its ethical and power dimensions. How am I ‘pre-formed’ and pre-informed by my social position ac-

14

Foreword cording to class, gender, minority/majority status or world region? Researching about gender in education, will I ask only women or also men or queer people? And will I interview migrant men and women as well as ‘ethnic Japanese’? So researchers have to reflect on whom they include or exclude through their concepts (e.g. of gender) and selection of interview partners. This also applies to interpretation: Will I accept the fact that mothers make lunchboxes (bentō) for school children as something natural (as some interviewed mothers might say) or will I look for contradictions and ambivalences in the interview texts? Researchers do not have to belong to the group they do research on; the contributions in this volume rather suggest that crossing borders of age, gender or nationality may add value to both the interviewer and the interviewed. But they will have to reflect on their own position, experiences and potential power. The second issue are the hermeneutic dynamics in qualitative cocreated research or how to create and interpret meanings in an intercultural process. The first obvious barrier is the Japanese language, which in my view can be only overcome by using it. Expert interviews with international actors may be done in English or German. But for interpretative qualitative research this may not work. Having tried it at the request of my interview partners, I found that at least the semantics are different in the end and thus qualitative substance may suffer. Also, many Japanese appreciate the outside researcher taking the trouble to learn their language, with the result that the interview situation becomes more like an everyday interaction. But reflexivity is also needed in intercultural qualitative research as a continuous exchange process of meaning between cultures or intercultural hermeneutical dialectics. In which ways can researchers craft their theoretical and empirical framework so that it does not follow Eurocentric (or ‘Nipponcentric’) codes and is open to articulation and interpretation by the actors? Asking why mothers make a bentō-box for schoolchildren makes sense in Japan but not so much in Germany, and may also involve new stereotyping. Doing research on otaku, would one translate the term and look for English equivalents or start from the fact that it is now an international term explained in various national Wikipedias? Referring to these examples, I want to argue that intercultural hermeneutical dialectics are not simply a matter of translation but rather of reflecting the ongoing cocreation of meanings between researchers and actors/ research subjects. Doing intercultural qualitative research in Japan implies that the actors articulate their meanings and constructions and have an open space for this. The researchers will have to understand these meanings and then go beyond them in their own interpretation, while keeping the trust of their interview partners. Intercultural qualitative research in this sense is evolving in many world regions. Thus, new spaces of transnational knowledge creation are emerging (Gerharz 2021) and Area Studies like Japanese Studies can play a key role in this. Let me raise some questions to conclude: Will these spaces still be centred on Japanese Studies outside Japan and research inside Japan? Or will mainstream Cultural or Social Sciences in the ‘West’ overcome their tendency towards exoticising or singularising Japan and (finally) join in creating these spaces, thus opening themselves up to comparative and reflexive universal research (Lenz 2013)? With more intercultural research covering shared problems, will the circulation of knowledge still be a one-way road

15

Foreword between ‘the West’ and Japan or become a truly transnational exchange (see, for example, Ochiai 2012–)? And how will the emerging transnational academic spaces recognise and negotiate the deep inequalities in the postcolonial world of academia?

Ilse Lenz Professor Emerita of Sociology, Ruhr-Universität Bochum Berlin, July 2020 References Gerharz, Eva (2021): Postkoloniale Ethnographie und Indigenous Research Methodology. In: Poferl, Angelika/Schroer, Norbert (eds.): Soziologische Ethnographie. Wiesbaden: Springer. Lenz, Ilse (2013): Differences of humanity from the perspective of gender research. In: Rüsen, Jörn (ed.) (2013): Approaching humankind: Towards an intercultural humanism. Göttingen: V&R unipress, pp. 185–200. Ochiai, Emiko (ed.) (2012–): The intimate and the public in Asian and global perspectives. https://brill.co m/view/serial/IPAP, [Accessed 27 August 2020]. Rosenthal, Gabriele (2018): Interpretive social research: An introduction. Göttingen: Göttingen University Press.

16

Acknowledgements Research can’t be done alone and builds on other people’s support, on critical and oftentimes controversial discussions, on mutual help and the exchange of resources and knowledge. All of the above apply to this handbook. Editing a book across continents and coordinating more than seventy authors requires time, logistics and help from others. We could not have put together this handbook without the close collaboration, effort, support, trust, knowledge and work of numerous people around the world. From the very beginning, we received wholehearted support for this project from Sandra Frey and Alexander Hutzel and their colleagues at Nomos publishing. The same is true for our colleagues at Freie Universität Berlin (FUB), the German Institute of Japanese Studies (DIJ) and the Heinrich Heine University Düsseldorf (HHU). The editors would particularly like to thank Verena Blechinger-Talcott, Susanne Brucksch, Lorenz Denninger, Andreas Eder-Ramsauer, Isaac Gagné, Sonja Ganseforth, Barbara Geilhorn, Markus Heckel, Steffen Heinrich, Barbara Holthus, Katharina Hülsmann, Hanno Jentzsch, Agnes Laba, Elena Meyer-Clement, Hannes Mosler, Theresia Peucker, Richard Samuels, Annette Schad-Seifert, Elisabeth Scherer, Kai Schulze, Christian Tagsold, Julia Trinkert, Katrin Ullmann, Franz Waldenberger, Corey Wallace and Matthias Zachmann for their contributions, their valuable feedback, their trust and/or for providing overall support. Both editors are also greatly indebted to their home institutions, Cornelia to FUB, Nora to the DIJ. We would like to express our sincerest thanks to Cosima Wagner who has been a source of motivation throughout the process that finally led to the completion of this handbook. She will also take on the task of creating a webpage to digitally enhance this book. A lot of inspiration emerged from discussing methods and the handbook with our students. We thank all students who participated in our method courses at FUB, HHU and Musashi University. FUB students’ suggestions for the structure and content of this handbook were presented by Thora Singer and Egor Skripkin to the chapter authors during the conference ‘Studying Japan: The impact of transnationalisation and technological innovation on methods, fieldwork and research ethics’ in July 2019 in Berlin. At the same conference, Susanne Auerbach, Andreas Eder-Ramsauer and Jan Niggemeier presented PhD students’ expectations of what a method handbook for the study of Japan should look like. They inspired many authors to think more about their audience and to reconsider the structure of their chapters. We are greatly indebted to Karin Klose, Brigitte Peek and Thomas Weitner at FUB and Joachim Röhr at the DIJ for their administrative support of the conference. We also thank Andreas Steinhöfel for the wonderful design of the poster and the flyer. Without the commitment and great help by FUB’s student assistants Antonya Schmidt, Maria Natalia Seidel-Hirose, Julia Süße and Toby Wolf and DIJ interns Isabel Schreiber and Marie Ulrich the conference would not have been possible. We also thank the German Research Foundation (DFG), the DIJ, FUB’s Ernst Reuter Society and the Gender Equality Fund at the Department of History and Cultural Studies at FUB for their financial support. After the conference, we were tasked with writing our own chapters, collecting the contributions, editing and formatting manuscripts. For their invaluable help during this process, we cannot thank Marie Ulrich and Isabel Schreiber enough. We are extremely happy and grateful that they both agreed to keep on working with us after their internships to finish the project 17

Acknowledgements together. We thank both of them for formatting all chapters and reference lists, organising all formal matters and index terms and for helping with the final proofs. Of course, we would also like to thank the DIJ for paying them. We are grateful to Furkan Kemik, student assistant at FUB, for his help with creating the handbook’s index and formatting the notes on contributors. Without language support from native English speakers and their copy-editing skills, this handbook would have been less accessible to its readers, thus, we very much thank Martyn Ford, Hilary Monihan and Katrina Walsh for their great work. Last but not least, we thank Robin Weichert who kindly provided the wonderful cover picture. Most of all we thank the authors who contributed to this book, who believed in the project and who shared their knowledge and expertise with us. Their trust and overwhelming positive and encouraging feedback was especially valuable when we were—at times—discouraged by the sheer amount of work and the multiple tasks related to this project. Our families and friends were of invaluable aid and we would also like to thank them for listening to us when we were facing problems, for helping us to think about other (more enjoyable) things and for celebrating (interim) successes with us. With the finished manuscript in our hands we are delighted to present this handbook to the Japan(ese) Studies community and hope that readers find it useful. Nora Kottmann and Cornelia Reiher Berlin and Tokyo, August 2020

18

Introduction: Studying Japan Nora Kottmann and Cornelia Reiher

1. Introduction The handbook Studying Japan emerged—just like any good research project does—from a puzzle. In 2016, we were both teaching in Japanese Studies programmes at German universities where methodological training is often squeezed into the curricula here and there, but generally not taught in a systematic manner. In our courses, we were often confronted with questions from students such as ‘How do I start my research?’, ‘Which methods suit which research questions and designs?’, ‘How should I conduct my research?’ or ‘What should I do with my data?’ This made us wonder how we could teach Social Science research methods to students who want to conduct research in or on Japan in a more systematic way. Lacking a comprehensive handbook on the methods of Social Science research on Japan that we or our students could use in class, we started to think about what such a handbook could and should look like and eventually decided to create one ourselves. Now, more than four years later and after countless discussions, millions of Skype calls, two conferences and numerous encounters with our authors, we are very proud to write this introduction to just such a handbook. For us, this handbook is a milestone that began with (still ongoing) discussions on methodology in Japan(ese) Studies over the course of sharing our experiences teaching research methods to Japanese Studies students at Freie Universität Berlin and Heinrich Heine University Düsseldorf. From these conversations emerged a joint teaching project of (method) courses focusing on Japanese foodscapes in Berlin and Düsseldorf, which resulted in a conference in Berlin in 2017 where students from both universities presented their projects and discussed method education with scholars from Berlin, Düsseldorf and Japan (Reiher 2018a). Around that time, we first talked about the idea of creating a method handbook for a Japanese Studies audience, and in early 2018 we wrote a book proposal and began to recruit authors. From the very start we were (and still are) overwhelmed by the positive feedback from colleagues and everybody else we talked to about this project. We soon realised that there was so much material to discuss with regard to methodological challenges and the method handbook that we decided to invite the authors of each chapter to Berlin for a conference in the summer of 2019. Discussions with the authors substantially shaped some of the common threads that run through almost all chapters of this book: 1. What is specific to research on and in Japan? 2. How do transnational entanglements change the study of Japan? 3. How do technological innovations enable and challenge research on Japan? and 4. What are the ethical implications when studying Japan? This handbook is a collaborative effort, and we are grateful to everyone who supported it.

19

Nora Kottmann and Cornelia Reiher

2. Why this handbook and why now? Why is a handbook of qualitative Social Science research methods for the study of Japan necessary at all, and particularly at this point in time? There are wonderful books on Social Science methods, fieldwork and research designs on the market and for Japan(ese) Studies, the volume Doing fieldwork in Japan, edited by Theodore Bestor, Patricia Steinhoff and Victoria Lyon-Bestor (2003), is certainly the most influential.1 It is widely used by those who plan to or are already conducting fieldwork in Japan. Some other edited volumes or special journal issues have addressed issues related to fieldwork and to ethnography, in particular, in Japan such as reflexivity, responsibility and fieldwork ethics (Alexy/Cook 2019; Furukawa 2007; Hendry/ Wong 2009; Linhart et al. 1994; Reiher 2018b; Robertson 2007). Very few discuss data analysis (Kobayashi 2010; Shimada 2008). Several individual contributions primarily address fieldwork, fieldwork ethics and ethnography in Japan (Aldrich 2009; Gill 2014; Hendry 2015; McLaughlin 2010; Numazaki 2012; Yamashita 2012). Yet, despite the valuable publications this handbook builds on and is indebted to, there is, at least to our knowledge, no comprehensive and coherent handbook on the study of Japan that addresses the whole research process from the first idea to the publication of findings, explains and discusses the most common methods in Social Science research in and on Japan in a ‘how-to’ manner and can be used by students, researchers and teachers alike. Therefore, one motivation for putting this handbook together is to offer a starting point for learning and teaching methods as well as research designs in a Japanese Studies context and beyond. In addition to this relatively pragmatic reason, there are, however, three more reasons why we consider this handbook necessary and timely. First, there is an increasing demand for systematic and transparent research practices in Japanese and Area Studies communities against the backdrop of the increasing marginalisation of Area Studies in academia, particularly in Europe (Basedau/Köllner 2007; Ben-Ari 2020).2 Secondly, the transnationalisation of Japanese Studies as a research field, of Japan as its research subject and of research teams requires researchers to rethink traditional national and disciplinary boundaries. Thirdly, technological innovations provide new and exciting opportunities for research, yet also pose various challenges, including in regard to ethical questions. This handbook is our attempt to address and discuss these and further developments with scholars around the world and contribute to respective methodological discussions. We believe that it is important to strengthen international and interdisciplinary exchange and discussion about how students and scholars of Japan can best conduct research in a transparent and ethical way and produce reliable, comparable and comprehensive research results that scholars from Area Studies and Social Science disciplines alike can relate to.

1 There are many Social Science method books focusing on a range of topics. Thus, in this handbook’s individual chapters, the authors give recommendations and introduce handbooks on the respective topics. Of course, there is also a great variety of method books in Japanese (see, for example, Kishi et al. 2016). We would also like to mention two edited volumes that explicitly address teaching in/for Japan(ese) Studies in a Japanese and a global context, namely Gaitanidis et al. (2020) and Shamoon/McMorran (2016). 2 For an ongoing, interactive discussion on the topic, see Curtis (2020). For an early contribution on the positioning of Japanology in the Social Sciences in a German context, see Lenz (1996) and Seifert (1994).

20

Introduction

3. What this handbook is about Studying Japan mainly targets (PhD) students and researchers who plan to draw on qualitative Social Science methods to conduct research on Japan. It also offers a handy tool for colleagues who teach courses on fieldwork, research designs and methods or want to address specific methodological issues in class in order to prepare their students to conduct their own research projects and write theses. This handbook is about qualitative Social Science research on Japan, focusing on the entire research process that begins with a vague interest in a research topic, which is then developed into a research question and eventually leads to findings presented in a thesis, an article or a book. Since the study of Japan is an interdisciplinary field, research focusing on Japan’s society, politics, culture, economy and history draws on a wide variety of theories and methods from various disciplines. Therefore, throughout this handbook the authors present insights from Sociology, Political Science, Anthropology and History, but also address several recurring themes and challenges. One challenge for both Japanese Studies and Area Studies scholars has been the translation of methods developed in other disciplines (mostly in the West) to specific (often non-Western) field sites and research subjects. One could argue that these translation processes are part of every research project, where methods have to be adjusted to a specific field site or a researcher’s skills or resources. However, there are some issues that are particular to the study of Japan in and outside the country. The most obvious is language. Translation of Japanese sources and data as well as cultural norms is the task of every Japan researcher, regardless of their nationality. Therefore, it is important to be reflexive regarding one’s own positionality, the reciprocity of trust-relations (Takeda 2013), the ways sensitive issues are handled or conventions for encounters in the field. At the same time, an increasing focus on transnational entanglements, mobilities and processes (not only) in Japan-related research challenges traditional national and disciplinary boundaries (Soysal 2016). This implies that research on Japan is not only carried out in Japan anymore (Adachi 2006; Aoyama 2015; Kottmann 2020). It also means that it is important to contextualise findings on Japan in a global context, no matter if a researcher studies Japan’s transnational entanglements or compares Japan with other countries.3 In addition, an increasing focus on the transnationalisation of cultural, social and political phenomena in and beyond Japan involves several methodological challenges. For example, researchers may need to visit multiple sites or be able to conduct multilingual case studies within Japan (Arrington 2016; Avenell 2015; Farrer 2015). Furthermore, the research enterprise itself has become more transnational. In addition to cooperation across the boundaries of individual Area Studies (Middell 2018), research teams are increasingly international and interdisciplinary. This provides new opportunities, but also poses questions with regard to languages, institutional differences or divergent ethical requirements. Transnational collaboration is often enabled through recent technological innovations ranging from online communication tools to software for data analysis or data repositories. Technological innovations provide new tools for getting in touch with informants via social media, 3 For an ongoing discussion on comparisons in Japan(ese) Studies and Area Studies, see Sidaway/Waldenberger (2020).

21

Nora Kottmann and Cornelia Reiher accessing data online, making large sets of data available for other researchers or a public audience or coordinating an international research team. In fact, this very handbook would not have been possible without tools for online communication and for storing data online! But these new technologies also pose challenges to researchers studying Japan and require them to develop new strategies for research. They create new types of reciprocity and demand attention is paid to the impact of social media in the whole research process online and offline (Baker 2013; Danley 2018; Gerster 2018; Postill/Pink 2012). Not only do translation processes have (new) ethical implications, but so does the transnationalisation of Japan research and technological innovations. In fact, ethical issues are of high relevance during the whole research process, ranging from the originality of research questions to ensuring fairness in publishing. While these issues pop up in almost all chapters, we devote a separate chapter to the topic to stress the importance of good research practices, academic integrity and research ethics, such as properly quoting sources, ensuring fairness and respect to research participants and colleagues, and protecting the privacy of interviewees.

4. Editorial decisions This handbook offers a large number of contributions on a variety of topics, but we are aware that we cannot cover everything there is to say about methods and methodology in the study of Japan. Thus, we had to make a number of decisions to limit this handbook’s scope, including the level of detail in the chapters, author selection and the format of the handbook. One choice we made was to focus on qualitative methods because these are the methods we are most familiar with and which our students are most likely to use. Another was to only write short overviews for each topic in the main chapters, although much more could have been said about each of them. To account for this, we provide further reading for those who would like to know more about the specific topics as well as to connect the literature on research design, fieldwork and methods from the Social Science disciplines with the study of Japan. Selecting contributing authors for the handbook was a more difficult process. We planned the handbook as an international collaborative project and sought to balance contributions with regard to disciplines, nationality, gender and career level, but because of our own academic background and the context from which this handbook emerged, many of the handbook’s authors are food, family and gender scholars, and a significant number were educated and/or work in Germany. Nonetheless, we offer interdisciplinary perspectives on each topic, and the handbook unites contributions by anthropologists, political scientists, sociologists and (fewer) historians. In short, the more than 70 selected authors whose contributions are featured in this handbook do not represent the full spectrum of Social Science research on Japan, but rather this selection reflects our own positionality in the field. We are, of course, aware that there are many more wonderful Japan scholars in the world! Finally, and despite a variety of technological innovations, we decided to publish this handbook as a physical, and therefore static, book that might be quickly partly outdated, especially the information on social media, websites and technological tools. Why did we choose a static format like a printed book? The short answer is: we love books and we are sure that at least 22

Introduction some information will remain pertinent. We imagine students and researchers carrying this handbook to Japan and back and having it at hand when they need it, even when there is no internet connection available. Despite these parochial and romanticised ideas about books, we are planning to enhance the printed version of the handbook with a website that features more information on methods and will be updated on a regular basis.4

5. How to use this handbook This handbook offers a starting point for learning, teaching and applying methods in a Japan(ese) Studies context and beyond. It is structured in such a way that it can be used for (self-)studying and teaching alike. The handbook could be utilised for comprehensive reading in order to gain an overview of qualitative methods in Social Science research on Japan as well as to structure one-term method courses. Yet, the handbook’s seventeen chapters can also be read individually; they can be used to learn about a specific method of data collection or analysis, expand one’s knowledge, familiarise oneself with a certain topic or just look up specific information. In addition, the individual chapters can be applied to courses as and when required. The handbook covers the entire research process in seventeen chapters from the outset to the completion of a thesis, paper or book. While this structure and the ‘how-to’ style might suggest that the research process consists of neatly separated steps, in reality, this is not the case. We are aware that the research process is often circular and dynamic and that the individual steps are often not carried out one after another in a linear manner, but sometimes even in reverse order. The blurred boundaries between the different tasks and steps in the research process are also addressed in the individual chapters. Yet due to the limitations of a book, which only allows for linear narration, as well as for reasons of clarity, this handbook is structured to follow the steps of the research process as they are most commonly organised. The seventeen chapters are all structured in a similar and easy-to-access format: a chapter introduction (‘main chapter’) and three short essays with further reading and a joint reference list. The main chapters feature an introduction to key ideas, concepts and practices, point out key terms, address the most important problems and the strategies that can be employed to solve them, present selected case studies and offer further up-to-date reading. While the main chapters address the respective topics in a relatively general way, they always refer to the specific challenges and opportunities encountered when doing research on and/or in Japan. Three short essays written by senior and junior researchers in Japan(ese) Studies from around the world follow the main chapters. There are a total of 51 essays, each offering insights into how 4 A number of smaller decisions were made related to gender-sensitive language, the order of Japanese names, the order of authors and the transcription of Japanese terms. With regard to gender, we decided to use ‘she’ or ‘her’ for female, male and other genders when the gender of the subjects is unclear. This is not meant to be exclusive, but rather to challenge old ways of thinking that took the use of masculine forms to refer to both genders for granted in academia. Japanese names are written in the following—and in Japan unusual—order of first name first and last name second. This is due to criticism from some of our Japanese authors, who did not want to be treated differently from the other authors. Therefore, we decided to deviate from the way of writing Japanese names normally practised in Japan(ese) Studies. In the case of more than one author, names are mentioned in alphabetical order. Japanese terms are romanised based on the modified Hepburn system.

23

Nora Kottmann and Cornelia Reiher individual scholars actually deal with their respective method in practice. The authors share their experiences, offer concrete advice on and precise insights into their fields of interest, and elaborate on their perspective(s) and individual way(s) of studying Japan both in and outside the country. Yet, the essays are not only illustrations of research experiences but also give insights into a wide range of topics in the study of Japan, including nuclear power plants, single women, families, food safety, Japan-China relations, condom use, social inequality, host clubs, party politics and agriculture. In so doing, the essays celebrate the diversity and plurality of scholarship on Japan. Furthermore, the essays show that there is no ‘right’ or ‘wrong’ way of doing Japan research, but that research always reflects the researchers’ positionality and that it is necessary to make thoughtful decisions and explain them well.

6. Structure and content of this handbook The first four chapters set the context for Japan research and address fundamental steps that often take place at the beginning of the research process. In chapter 1, Roger Goodman provides an introduction to the diversity of Japanese Studies and to research on Japan in the Social Sciences. Against this backdrop, Goodman provides advice on finding a research topic and explains how a researcher’s biography and theoretical (pre-)assumptions affect this choice. The importance of research questions as well as the actual process of finding and asking questions is the focus of chapter 2 by Gabriele Vogt. In chapter 3, Kaori Okano addresses (case study) research designs and touches upon the discussion of theory building and testing as well as inductive and deductive processes. Urs Matthias Zachmann discusses the importance of reviewing scholarly literature and the need to identify and position oneself in relevant debates in chapter 4. He also explains the challenge of balancing debates from Area Studies, the Social Sciences as well as debates from Japan.5 The subsequent chapters focus on data collection. Chapter 5 by Akiko Yoshida starts with an overview of the most common qualitative data collection methods used in Social Science research. Yoshida explains different types of methods and comparatively discusses their respective characteristics, which is followed by chapters that each introduce and discuss one specific method in more detail. Levi McLaughlin addresses fieldwork—physical and virtual as well as in and outside of Japan—in chapter 6, Nora Kottmann and Cornelia Reiher introduce and discuss the world of qualitative interviews in chapter 7 and Christian Tagsold and Katrin Ullmann elaborate on observational research with a focus on participant observation in chapter 8. Finally, in chapter 9, Theresia Berenike Peucker, Katja Schmidtpott and Cosima Wagner deal with the collection of written and visual sources in archives, libraries and Japanese online databases.6

5 The essays in these chapters are written by Verena Blechinger-Talcott (Ch. 1.1), Daniel P. Aldrich (Ch. 1.2), Joy Hendry (Ch. 1.3), Nicolas Sternsdorff-Cisterna (Ch. 2.1), Kenneth Mori McElwain (Ch. 2.2), David Chiavacci (Ch. 2.3), Lynne Nakano (Ch. 3.1), Jamie Coates (Ch. 3.2), Kay Shimizu (Ch. 3.3), Patricia Maclachlan (Ch. 4.1), Sonja Ganseforth (Ch. 4.2) and Gracia Liu-Farrer (Ch. 4.3). 6 The essays in these chapters are written by Emma E. Cook (Ch. 5.1), Karen Shire (Ch. 5.2), Barbara Holthus and Wolfram Manzenreiter (Ch. 5.3), Nana Okura Gagné (Ch. 6.1), James Farrer (Ch. 6.2), Hanno Jentzsch (Ch. 6.3), Christoph Brumann (Ch. 7.1), Tomiko Yamaguchi (Ch. 7.2), Allison Alexy (Ch. 7.3), Susanne Klien (Ch.

24

Introduction Chapter 10, by Carola Hommerich and Nora Kottmann, focuses on mixed methods research, and it connects the chapters on data collection and data analysis. It serves a somewhat special role, as it provides a basic introduction to key terms and concepts of quantitative methods. The chapters that follow are devoted to data analysis, which may occur during and/or after the data collection process. In chapter 11, David Chiavacci addresses the importance of data analysis for the whole research process, introduces the main analytical approaches and discusses the use of computer-assisted qualitative data analysis software. The subsequent chapters each address specific analytical methods. In chapter 12, Caitlin Meagher focuses on (modified) grounded theory designs, the process of coding, the development of concepts and, ultimately, theory. Following this, in chapter 13, Celeste Arrington introduces content and frame analysis, and discusses their similarities and differences as well as each method’s strengths and weaknesses. In chapter 14, Andreas Eder-Ramsauer and Cornelia Reiher discuss various forms of discourse analysis, define basic concepts and explain individual steps in analysis.7 Finally, the last three chapters of the handbook deal with finishing one’s research projects and address basic cross-cutting issues like ethics and writing. In chapter 15, Chris McMorran writes about the importance of successfully completing one’s research project(s) despite the various obstacles in researchers’ private and professional life. Furthermore, he encourages researchers to demystify the writing process. In chapter 16, Cornelia Reiher and Cosima Wagner address the importance of following good and fair research practices throughout the whole research process and introduce new trends, such as open scholarship. In the final chapter 17, James Farrer and Gracia Liu-Farrer introduce various oral and written forms of presenting one’s findings for both an academic and a wider audience. In this context, the authors stress the importance of carefully thinking about the audience one wants to reach.8 Throughout the handbook, all the authors write as concretely as possible and in an easy-to-access manner. They summarise key points, highlight key issues, define key terms, include visual models, offer lists of important journals, provide links to important webpages and introduce helpful tools (digital and analogue). While all the authors write from their respective perspective—as novice or established researchers; as Japanese, European, Australian or American citizens; as sociologists, political scientists, anthropologists, human geographers or economists; as people of a specific gender and age—they provide information that is helpful and applicable for students, researchers and colleagues from different national contexts and academic cultures.

8.1), Akiko Takeyama (Ch. 8.2), Swee-Lin Ho (Ch. 8.3), Katja Schmidtpott and Tino Schölz (Ch. 9.1), Sheldon Garon (Ch. 9.2) as well as Shinichi Aizawa and Daisuke Watanabe (Ch. 9.3). 7 The essays in these chapters are written by Robert J. Pekkanen and Saadia M. Pekkanen (Ch. 10.1), Laura Dales (Ch. 10.2), Jun Imai (Ch. 10.3), Katharina Hülsmann (Ch. 11.1), Genaro Castro-Vázquez (Ch. 11.2), Markus Heckel (Ch. 11.3), Nancy Rosenberger (Ch. 12.1), Celia Spoden (Ch. 12.2), Julia Gerster (Ch. 12.3), Anna Wiemann (Ch. 13.1), Emi Kinoshita (Ch. 13.2), Kai Schulze (Ch. 13.3), Annette Schad-Seifert (Ch. 14.1), Daniel White (Ch. 14.2) and Steffen Heinrich (Ch. 14.3). 8 The essays in these chapters are written by Aya H. Kimura (Ch. 15.1), Christian Tagsold (Ch. 15.2), Richard J. Samuels (Ch. 15.3), Isaac Gagné (Ch. 16.1), David H. Slater, Robin O’Day, Flavia Fulco and Noor Albazerbashi (Ch. 16.2), Christopher Gerteis (Ch. 16.3), Scott North (Ch. 17.1), Isabelle Prochaska-Meyer (Ch. 17.2) and Brigitte Steger (Ch. 17.3).

25

Nora Kottmann and Cornelia Reiher

7. Summary and future perspectives In a nutshell, the handbook Studying Japan provides an overview of and hands-on advice for the individual steps in the research process and discusses methodological opportunities and challenges brought about by the transnationalisation of research subjects, research practices and research groups as well as by technological innovations and the digital revolution, while paying attention to good research practice and ethics. It enables students and teachers to study, teach and apply methods and to develop research designs and strategies for fieldwork in Japan. The challenge of producing both an area-sensitive yet academically sound study is a problem not only for scholars and students of Japanese Studies but also for researchers from all Area Studies. Thus, this handbook is a valuable tool for both the international Japan(ese) Studies community as well as for all Area Studies scholars who take the local characteristics and languages of ‘their’ areas seriously. At the same time, scholars from the Social Sciences who plan to study Japan in more depth can use this book to engage with Japan more deeply. We hope this handbook inspires further reflection on the conducting and teaching of research in and beyond Japan. We think that the discussion of the methodological and ethical challenges arising, in particular, from transnationalisation and technological innovations in Social Science research in and on Japan should be continued. We are looking forward to future discussions, possibly an interdisciplinary handbook on quantitative methods in the study of Japan and to enhancing this book through a website that could serve as a means to connect researchers internationally who would like to share their experiences of using and teaching methodology in a Japan(ese) Studies context. Meanwhile, we hope that you find this book useful in facilitating your research or teaching. It might help to keep in mind this advice: while there is no single ‘right’ or ‘wrong’ way of studying Japan, work as precisely and reliably as possible, be critical and pragmatic and, most importantly, have fun, follow your curiosity and don’t lose your fascination with your research. References Adachi, Nobuko (ed.) (2006): Japanese diasporas: Unsung pasts, conflicting presents, and uncertain futures. London: Routledge. Alexy, Allison/Cook, Emma (2019): Reflections on fieldwork: Exploring intimacy. In: Alexy, Allison/Cook, Emma (eds.): Intimate Japan: Ethnographies of closeness and conflict. Honolulu, HI: University of Hawai‘i Press, pp. 236–260. Aldrich, Daniel (2009): The 800-pound gaijin in the room: Strategies and tactics for conducting fieldwork in Japan and abroad. In: PS: Political Science & Politics 42, No. 2, pp. 299–303. Aoyama, Reijiro (2015): Introduction: Japanese men and their quest for well-being outside Japan. In: Asian Anthropology 14, No. 3, pp. 215–219. Arrington, Celeste L. (2016): Accidental activists: Victims and government accountability in Japan and South Korea. Ithaca, NY: Cornell University Press. Avenell, Simon (2015): Transnationalism and the evolution of post-national citizenship in Japan. In: Asian Studies Review 39, No. 3, pp. 375–394. Baker, Sally (2013): Conceptualising the use of Facebook in ethnographic research: As tool, as data and as context. In: Ethnography and Education 8, No. 2, pp. 131–145. Basedau, Matthias/Köllner, Patrick (2007): Area Studies, comparative Area Studies, and the study of politics: Context, substance, and methodological challenges. In: Zeitschrift für Vergleichende Politikwissenschaft 1, No. 1, pp. 105–124. Ben-Ari, Eyal (2020): Area Studies and the disciplines: Japanese Studies and Anthropology in comparative perspective. In: Contemporary Japan 32, No. 2, pp. 240–261. Bestor, Theodore C./Steinhoff, Patricia G./Lyon-Bestor, Victoria (eds.) (2003): Doing fieldwork in Japan. Honolulu, HI: University of Hawai‘i Press.

26

Introduction Curtis, Paula (2020–): Virtual roundtable: The ‘rebirth’ of Japanese Studies. http://prcurtis.com/events/AA S2020/, [Accessed 14 August 2020]. Danley, Stephen (2018): An activist in the field: Social media, ethnography, and community. In: Journal of Urban Affairs 7, No. 1, pp. 1–17. Farrer, James (ed.) (2015): The globalization of Asian cuisines: Transnational networks and culinary contact zones. New York, NY: Palgrave Macmillan. Furukawa, Akira (ed.) (2007): Frontiers of social research. Melbourne: Trans Pacific Press. Gaitanidis, Ioannis/Shao-Kobayashi, Satoko/Yoshino, Aya (eds.) (2020): Kuritikaru nihongaku: Kyōdōgakushū o tsūshite ‘nihon’ no sutereotaipu o manabihogusu. Tōkyō: Akashi. Gerster, Julia (2018): The online-offline nexus: Social media and ethnographic fieldwork in post 3.11 Northeast Japan. In: ASIEN—The German Journal of Contemporary Asian Studies 149, pp. 14–32. Gill, Tom (2014): Radiation and responsibility: What is the right thing for an anthropologist to do in Fukushima? In: Japanese Review of Cultural Anthropology 15, pp. 151–163. Hendry, Joy/Wong, Heung Wah (eds.) (2009): Dismantling the East-West dichotomy: Essays in honour of Jan van Bremen. London: Routledge. Hendry, Joy (2015): The state of Anthropology in and of Japan: A review essay. In: Japan Forum 27, No. 2, pp. 121–133. Kishi, Masahiko/Ishioka, Tomonori/Maruzama, Satomi (2016): Shitsuteki shakaichōsa no hōhō. Tōkyō: Yuhikaku. Kobayashi, Tazuko (2010): Preface for the papers on different perspectives on biographies. In: Newsletter RC 38, pp. 11–12. Kottmann, Nora (2020): Japanese women on the move: Working in and (not) belonging to Düsseldorf's Japanese (food) community. In: Matta, Raúl/de Suremain, Charles-Édouard/Crenn, Chantal (eds.): Food identities at home and on the move: Explorations at the intersection of food, belonging and dwelling. London: Routledge, pp. 293–318. Lenz, Ilse (1996): On the potential of Gender Studies for the understanding of Japanese society. In: Kreiner, Josef/Ölschläger, Hans Dieter (eds.): Japanese culture and society: Models of interpretation. Tokyo: Deutsches Institut für Japanforschung, pp. 267–289. Linhart, Sepp/Pilz, Erich/Sieder, Reinhard (eds.) (1994): Sozialwissenschaftliche Methoden in der Ostasienforschung. Wien: Institut für Japanologie. McLaughlin, Levi (2010): All research is fieldwork: A practical introduction to studying in Japan as a foreign researcher. In: The Asia-Pacific Journal 8, No. 30, https://apjjf.org/-Levi-McLaughlin/3388/article.h tml, [Accessed 14 August 2020]. Middell, Matthias (2018): Transregional Studies: A new approach to global processes. In: Middell, Matthias (ed.): The Routledge handbook of Transregional Studies. London: Routledge, pp. 1–16. Numazaki, Ichiro (2012): Too wide, too big, too complicated to comprehend: A personal reflection on the disaster that started on March 11, 2011. In: Asian Anthropology 11, No. 1, pp. 27–38. Postill John/Pink, Sarah (2012): Social media ethnography: The digital researcher in a messy web. In: Media International Australia, Incorporating Culture & Policy 145, pp. 123–134. Reiher, Cornelia (2018a): Japanese foodscapes in Berlin: Teaching research methods through food. In: ASIEN—The German Journal of Contemporary Asian Studies 149, pp. 111–124. Reiher, Cornelia (ed.) (2018b): Fieldwork in Japan: New trends and challenges (Special Issue). In: ASIEN —The German Journal on Contemporary Asia 149. Richards, Lyn (2015): Handling qualitative data: A practical guide. London: Sage. Robertson, Jennifer (ed.) (2007): Politics and pitfalls of Japan ethnography: Reflexivity, responsibility, and anthropological ethics (Special Issue). In: Critical Asian Studies 39, No. 4. Seifert, Wolfgang (1994): Zum Stand der sozialwissenschaftlichen Japanforschung im deutschsprachigen Bereich: Ein kurzer Bericht. In: Doitsu kenkyū 18, pp. 52–61. Shamoon, Deborah/McMorran, Chris (eds.) (2016): Teaching Japanese popular culture. Ann Arbor, MI: Association for Asian Studies. Shimada, Shingo (2008): Die ‘dichte‘ Lebensgeschichte: Überlegungen zu den Methoden der empirischen Sozialforschung im interkulturellen Kontext. In: Cappai, Gabriele (ed.): Forschen unter Bedingungen kultureller Fremdheit. Wiesbaden: VS, pp. 265–280. Sidaway, James/Waldenberger, Franz (2020–): Comparing comparisons. TRAFO—Blog for Transregional Research. https://trafo.hypotheses.org/category/comparing-comparisons, [Accessed 14 August 2020]. Soysal, Yasemin Nuhoḡlu (ed.) (2016): Transnational trajectories in East Asia: Nation, citizenship, and region. London: Routledge. Takeda, Atsushi (2013): Reflexivity: Unmarried Japanese male interviewing married Japanese women about international marriage. In: Qualitative Research 13, No. 3, pp. 285–298.

27

Nora Kottmann and Cornelia Reiher Yamashita, Shinji (2012): The public Anthropology of disaster: An introductory note. In: Asian Anthropology 11, No. 1, pp. 21–25.

28

Chapter 1 How to begin research: The diversity of Japanese Studies Roger Goodman

1. Introduction The single most important decision for any research project is where to start: what question to examine and how to address it. This chapter sets out some of the key processes that researchers should consciously and conscientiously go through in making these decisions and attempts to turn them into a set of explicit and transparent steps to help those who are about to begin their own research projects. These principles apply at any level, from an undergraduate dissertation through to a major new project by a senior professor. They are built around the very simple premise that, in all research projects, the researcher is the main research tool. Just as any workman needs to know their tools, the researcher of Japan needs to know themselves. This chapter, therefore, looks at the importance of interrogating the personal biography and theoretical assumptions that all researchers bring to their work before they decide upon a research topic and research puzzle. In doing so, it also provides a guide to reading research which has already been undertaken by others in any field of Japanese Studies, from Natural and Medical Sciences through to the Social Sciences and Humanities.1

2. The importance of personal biography As the accounts by Daniel Aldrich, Verena Blechinger-Talcott and Joy Hendry in the essays following this chapter show, every research project starts with the researcher. We study—or we should study—things that we know about and things that interest us. We tend, however, to be very bad at acknowledging this fact. Until the 1970s, indeed, most social scientists failed to acknowledge in more than the most superficial way their own role in their studies. They felt that to do so was in some way not scientific. They presented themselves as objective researchers who collected data in a value free manner through robust methodologies which they then analysed using the latest theoretical models available.

1 The ideas in this paper were first explored when the author was looking for a topic for his doctoral thesis (Goodman 1984) and were developed in articles which reflected on the relationship between how that project and a number of subsequent projects were designed and the conclusions which were drawn from them (Goodman 1990a; 2000a; 2006).

29

Roger Goodman From the late 1970s these assumptions of ‘scientism’ began to be challenged by what some called the ‘reflexive turn’ (O’Reilly 2009, pp. 187–93). Increasingly, not just social scientists but even medical and physical scientists began to realise that, consciously or unconsciously, they brought with them a personal perspective on an issue which might influence not only why but also how they asked a particular question and how this might indeed affect what they saw and concluded.2 By the mid-1980s, as ‘reflexivity’ increasingly became intertwined with various debates about ‘post-modernism’ in Social Science, some researchers began to question whether it was possible to examine anything objectively and whether every research project was nothing more than a reflection of the cultural and political prejudices of the individual researcher. To some extent, this denial of objective truth was linked to and pushed by those whose beliefs in the ‘certainties’ of Marxism had been crushed by the crumbling of the former Soviet Union. One response to this collapse in faith in the scientific method was to turn the researchers’ microscope on to the researchers themselves. What did they discover about themselves as a result of looking at the other? Examples of this in the case of Japanese Studies can be seen in the works of Brian Moeran (1985), Matthews M. Hamabata (1990) and Dorinne Kondo (1990).3 Most researchers in the 1980s took a less extreme position which took into account three elements of any research project: the researcher, the research and the reader (Okely/Callaway 1992). They argued that it was sufficient to give the reader ample autobiographical information and a detailed account of how a project was set up to allow them to judge the research they produced against their background knowledge of the researcher. What was some of the personal information which researchers felt was important to share in the case of research on Japan? Gender (as exemplified in Hendry’s account, see this chapter, Ch. 1.3) was one. Women had a very different experience of Japan from men (Roberts 2003). Indeed, the fact that there are such strong gender divisions in Japan often leads to different forms of study, for example, with a tendency for men to study the public sphere and women the private sphere. Sexuality was another variable which was increasingly made explicit in studies of Japan in the 1980s, as indeed it was elsewhere as the study of identity politics and gender more generally became a global focus for research. This was most clearly expressed by Western authors who felt that the public expression of their sexuality was important since they did not want to separate their sense of self (which included their sexual orientation) from their role (as a researcher). An explicit example of this is the autobiographical account by John W. Treat (1999), but the importance of sexuality in giving access to certain worlds in Japan is also acknowledged in the work of Mark J. McLelland (2000) and Wim Lunsing (2001), who were among the first scholars to provide deep ethnographic accounts of the experience and worldviews of homosexuals in Japan. 2 Different disciplines have their own key figures in the ‘reflexive turn’ movement, but history will probably suggest that the single most influential figure was Pierre Bourdieu and the single most influential book was his Outline of a theory of practice (1977) with its notion that all researchers need to ‘objectify their own objectifications.’ Other important figures in these debates were Mikhail Bhaktin, Jacques Derrida and Michel Foucault. 3 Ostensibly, Moeran’s ethnography is on a rural community in Kyushu, Hamabata on family businesses and Kondo on small manufacturing firms in Tokyo. In practice, each of them is also an account of what they discovered about themselves through their encounters with Japan.

30

Chapter 1 How to begin research A variable which could be inferred from these personal accounts, though not always stated explicitly, is age (Smith 2003). This, of course, affects the researcher’s ability to empathise with and access different generations of Japanese. Age, gender and sexuality, of course, all interact. If one accepts that Japan is still a very patriarchal and gerontocratic society (and the make-up of the Diet and company boards would suggest that it is), then there is an argument that young women make the best researchers since they are the most likely to have the basic categories of how Japanese society operates ‘explained’ (‘mansplained?’) to them. Older, more experienced male researchers may be expected to ‘know’ these things. The ideal scenario for a social scientist is to be ‘patronised’ since that is when people reveal what they think are the basic underlying assumptions of their worldview. This is one reason (along with, ironically, the fact that their Japanese is too good) why it is often more difficult for native anthropologists to undertake research on their own society than it is for foreigners (see Yamaguchi, Ch. 7.2). At the end of the 1980s, Harumi Befu and Josef Kreiner (1992) carried out an interesting project which explored the impact of national background on the way that overseas researchers approached the study of Japan. They argued that researchers with different nationalities and different ethnic backgrounds bring with them ‘cultural baggage’ which impacts (generally unconsciously) on the type of questions they ask about Japan. North American scholars have a cultural predisposition when they look at Japanese society to focus on ‘race’, Koreans and Chinese on blood ties, Indians on minority and outcaste status, the Soviets (at the time) on collectivism, Germans on social democracy and the English on social class since these are the ‘key’ social variables in their own societies.4 Another issue which is rarely discussed in the personal introductions to accounts of Japan is politics, either personal or national. As Sheila Johnson (1975) has shown, the U.S. view of Japan between the 1940s and 1970s was largely determined by U.S. relations with China. That is almost certainly still the situation today. Further, within societies, right-wing commentators have generally had a more sympathetic view of Japan in the postwar period—because of its economic success and high levels of social stability—than left-wing commentators, who have been concerned about the lack of national unions to protect and fight for workers’ rights.

3. Interrogating the relationship between the person and society The above are all personal biographical details which may be pertinent to understanding the position which a researcher brings to their study of Japan. There are two other sets of assumptions which are actually much more significant, but which are rarely, if ever, discussed explicitly, although they can be gleaned by an astute reader simply by looking at the bibliography and acknowledgments of any academic book on Japan. These two sets of assumptions 1. about the relationship between the person and society (see sections 3 and 4) and 2. about the distinction between Japanology and Japanese Studies (see section 5) overlap to a considerable degree. Moreover, they are essentially independent from any of the other variables that have been ex4 It was during their workshop that I realised for the first time that the way I was looking at the issue of returnee children (kikoku shijo) in Japan was so strongly driven by my interest in the class effects of education as a result of my own experience of the highly class-divided English education system.

31

Roger Goodman amined previously; they have no relation to gender, ethnicity sexuality and educational background and, since the collapse of Communism in 1989, there is no reason why they should have a connection with nationality, age or class. The first of these sets of assumptions relates to the very nature of what constitutes academic research in the Social Sciences and Humanities. Put at its most simple, the Social Sciences and the Humanities can be defined as the study of the relationship between the person and society, depending on how ‘person’ and ‘society’ are defined in a particular place or time. This simple formulation is what disciplines as apparently varied as Archaeology and Psychology, Law and Economics, History and Literature, Linguistics and Business, and Education and Sociology all share, even as they invent their own special language for describing these key variables. In Anthropology, what is termed the study of the person or personhood lies at the very core of the study of any society, but an understanding of personhood is key in other disciplines too. While every society makes a distinction between self (ego) and role (persona), the relationship between the two varies over time and space. In Western societies, post-Enlightenment ideologies have seen the conflation of the two as leading to healthy ‘individuals’—and their separation as problematic. Western ideas of Freudian and Jungian psychoanalysis, for example, seek to find out how the role that a person is forced to perform (mother, worker, student) constrains their sense of self and how the self can be allowed to express itself fully again. In most societies, however, it is the ability to separate the two which is seen as essential to a healthy lifestyle. In Japan, when the self and role become overly conflated, the person may be perceived, or perceive themselves, as ‘selfish’. Distinction (kejime) is the skill that all small children develop that enables them to separate their sense of self from any role they need to perform and anyone who is unable to do so may be perceived as immature. Naikan and Morita therapies (Reynolds 1989) are focused on meditating on how one’s sense of self has got in the way of good role performance. In Social Science disciplines, the study of society can most simply be described as the examination of how rituals and symbols have been used, and by whom, to construct a sense of community. As Cohen (1985) shows, this can be either internally or externally generated; people can construct their own sense of who they are or they can be defined by others. Who does the constructing is a question of political and economic power—domestic or extra-domestic—in both cases. As Figure 1.1 shows, there are two ways that the relation between the person and society can be examined. Structuralist approaches look at how society constrains the actions of the person. Interpretative or social action approaches look at how the person constructs society.5 Structuralist approaches in turn can be broken down into two traditions: those which assume that society is essentially based on consensus and those which assume it is based on conflict. The former used to be described as functionalist and the latter as Marxist. Both terms have increasingly come to be used in a derogatory fashion (functionalist for being too ‘conserva-

5 Karl Marx, Max Weber and Émile Durkheim, generally credited as the ‘founders’ of these three approaches in the Social Sciences, are today often dismissed as ‘dead white men’, but it can be argued that virtually all current Social Science theory is either derived from, or developed in opposition to, their seminal work.

32

Chapter 1 How to begin research tive’, Marxist for being too ‘socialist’), but in purely analytical terms they remain the most useful way to think of work which looks at how society constrains the person. Figure 1.1: Heuristic overview of sociological theory Sociological theory Structuralist

Interpretative/agency

↓

↓

Society based on

Society based on

Society based on

consensus

conflict

competition

↓

↓

↓

Functionalism

Marxism

Social Action

↓

↓

↓

Emile Durkheim

Karl Marx

Max Weber

1858–1917

1818–1883

1864–1920

Functionalist and Marxist approaches have very different underlying assumptions. This can be seen very easily in work on education systems.6 Both functionalists and Marxists see education as effectively a black box in to which are fed the raw material of pre-school children. It is the outcome of the educational experience over which they disagree. Functionalists describe relations between the educational system and other institutions; Marxists explain why these relations exist and change over time. Functionalists see the socialisation process as a common value which holds society together; Marxists examine interests underlying those values and how socialisation differs systematically by social class. Functionalists see the education system as offering opportunities for mobility; Marxists see the role of education as maintaining structured social inequality (reproducing social class through reproducing social capital).7 In opposition to the structuralist theories of the Marxists and functionalists who see education as a black box, interpretative or social action theorists are more interested in what happens inside the black box of education. They want to know how the participants—the teachers, parents, policymakers and children among others—construct the society that makes up the school. Unlike structuralist theories, which assume these participants are passive in the face of societal rules and norms, the assumption in social action theory is that the participants have a level of agency, even if they cannot all express it equally. Students can conform to the goals and the methods for achieving those goals that the school has set, but they can also rebel, retreat, ritualise, colonise or innovate, to use some of the categories identified in the classic work of Robert K. Merton (1938). As Peter Woods (1979) has shown, teachers can also take a number of different roles and positions in relation to the curriculum and school rules. It is reported, for example, that left-wing teachers in Japan have sometimes supported adopting rightwing history textbooks as exemplars for their students of the dangers of the state getting involved in controlling the messages of history (Goodman 2020).

6 For an overview of these theories, see Sever 2012. 7 For probably the best-known analysis of the difference between functionalist and Marxist interpretations of Japanese society, see Ross Mouer and Yoshio Sugimoto (1986, chapters 2 and 3). A classic functionalist account is the work of Chie Nakane (1970); a classic Marxist account can be found in the work of Rob Steven (1982).

33

Roger Goodman Figure 1.2 suggests that the choice that a researcher makes between a structuralist or a social action approach to studying a particular problem can also influence the methodologies that they need to use. While it is not always the case, very often structuralist theories require quantitative research methods since they set out to measure the extent to which society constrains the activity of the person. Interpretative theories, on the other hand, often require qualitative research methods since they set out to examine how persons construct the world around them. Figure 1.2: Heuristic model of the relationship between structuralist and interpretative theories and the methods they use

Structuralist theories

Interpretative theories

e.g. functionalist (Durkheimian), conflict (Marxist) theories

e.g. social action (Weberian) theories

↓ measure the extent to which society constrains the activity of the person and tend to use

↓ examine how persons construct the world around them and tend to use

↓

↓

quantitative methods

qualitative methods

e.g. questionnaires, structured interviews, big data sets.

e.g. participant observation, unstructured interviews.

The best research should take into account both structuralist (functionalist and Marxist) and interpretative social action theories. They should also draw on both quantitative and qualitative research methods—in what is sometimes called ‘mixed methods’—since the difference between structuralist and interpretative theories can not only push researchers towards different methodologies (see Hommerich/Kottmann, Ch. 10). This difference can also explain why they may end up with very different conclusions when looking at apparently the same phenomenon, as the following example suggests.

4. Example of the impact of theoretical assumptions on research on contemporary Japan During the 1990s, Joshua Roth (2002) and Takeyuki Tsuda (2003) undertook detailed anthropological fieldwork among the nikkeijin (Latin Americans of Japanese descent) community who were invited, in large numbers, to come and work in Japan in the late 1980s which was then facing severe labour shortages in the country as the economy boomed (see Gagné, Ch. 6.1). Roth and Tsuda’s subsequent ethnographies agreed on almost all points in their account of this community. In particular they agreed on the fact that the nikkeijin, who had been so

34

Chapter 1 How to begin research proud of their Japanese ancestry when in Latin America, were disappointed on the reception they received in Japan and, in the process, ‘rediscovered’ their ‘Latin Americaness’. Where Roth and Tsuda differed was in the conclusions they reached for the future of nikkeijin in Japan. While Roth believed Japan would be able to contain within it minority groups, like the nikkeijin, as ideas of ‘Japaneseness’ became more broadly defined, Tsuda believed the nikkeijin identity would disappear inside the boundaries of an increasingly tight definition of ‘Japaneseness’. The reason for their different conclusions lay not so much in their views of the nikkeijin community as in their views of Japanese society and in particular their underlying assumptions of the relationship between the person and society. Tsuda (2003) saw Japan in very functionalist terms. He believed that the intrinsic nature of Japanese culture meant that anything coming from outside was perceived as potentially contaminating and, hence, in need of either rejection or purification before it could be accepted into society. Such an approach saw society functioning like a self-contained, biological organism with clearly defined boundaries and mechanisms for dealing with anything polluting from outside. Roth (2002), conversely, saw Brazilian Japanese ethnic identity coming from interaction with the political and economic structures within which the nikkeijin were forced to operate in Japan. It was not Japanese culture as such that was responsible for the rejection of the nikkeijin, but interest groups within Japan—such as employers, politicians, journalists and, particularly, labour brokers (hence the word ‘brokered’ in the title of his book). These groups, he said, used the language of culture and history to legitimise the marginalisation of the nikkeijin group for their own economic (cheap labour) and political (reinforcement of Japanese ethnic identity) ends. It was in opposition to this marginalisation that the nikkeijin had been constructing their own cultural forms (drawing on ideas of ‘Brazilianness’). As their class position strengthened in Japanese society, so the Brazilian nikkeijin would be able to exert economic and political pressure that would lead to their cultural lifestyles being accepted as part of the definition of ‘Japaneseness’. Compared to Tsuda, Roth’s view of society was much more flexible in terms of the power (‘agency’) that it gave to the different actors, even though he recognised that these same actors were themselves constrained by the political and economic realities of the contexts in which they moved. It was his (what might be termed ‘social action’) assumptions about the way societies operated that explained the very different conclusions he reached from Tsuda’s functionalist approach. Twenty years later, what can we now say about the situation of the nikkeijin in Japan? To a certain extent, we can say that neither Tsuda nor Roth was correct in identifying the future for the nikkeijin. The bursting of the economic bubble in the early 1990s meant that many of the nikkeijin were forced to return to Latin America and those who did stay often ended up as distinct but marginalised communities who operated as a peripheral and insecure workforce for sections of Japanese industry. In short, the functionalist and social action theories of Tsuda and Roth needed to be complemented with insights from Marxist thinking.

35

Roger Goodman

5. Japanology versus Japanese Studies The former section on the importance of the structure/agency dichotomy, with its references to some of the major intellectual traditions in Social Science, might have appeared rather abstract for a chapter on starting research on Japan, even if it has been spiced with some examples from the study of Japanese society. This has been on purpose because it is easy to lose sight of some of these big questions in the excitement of commencing a new project. Not taking them into account can have major ramifications for the project. The advice to look at such issues from the very beginning applies to anyone undertaking any project on any topic in the Social Sciences, but perhaps it is particularly important in the case of those in Area Studies. This is because, as Verena Blechinger-Talcott (see this chapter, Ch. 1.1) points out, Area Studies researchers can suffer from being considered ‘less rigorous or theoretically sophisticated’ than their disciplinary colleagues. There is a second set of assumptions which also needs to be taken into account at the very beginning of a research project which relate specifically to those doing research in Area Studies. These might be characterised broadly as ‘Area-ology’ versus ‘Area-Studies’ approaches or, in the case of Japan, ‘Japanology’ versus ‘Japanese Studies’. While the former long predates the latter, these two approaches have existed alongside each other in almost all Area Studies communities since the 1950s. In many parts of the world, however, they inhabit virtually parallel universes, publishing in different journals, attending different conferences and, sometimes, even being placed in different departments within the same institution. Figure 1.3 sets out, very simply, some of the key differences between these two communities. The core intellectual difference between them is whether a society is best studied in its own terms (an emic approach) or through a comparative lens (an etic approach). The former sees History as the key discipline and Philology as the key tool; the latter sees Sociology (in the broadest sense) as the key discipline and the use of universally applicable theory as the key tool. The former focuses on, and looks, for continuities; the latter discontinuities. The former assumes a society can only be studied in its own right; the latter that it should be judged by universal normative values. In general, the former has a view of society as essentially based on consensus; the latter sees society as more conflict-ridden. Even more broadly, the former is often associated with the Humanities; the latter with the Social Sciences. Figure 1.3: Some heuristic dichotomies for thinking about research in Area Studies

(Area)-ology

(Area)-Studies

Approach

Emic

Etic

Reference point

Internal comparison

External comparison

Key disciplines

History

Sociology

Key tools

Philology

Theoretical terms

Assumptions

Continuities

Discontinuities

Moral universe

Relativistic

Universalistic

36

Chapter 1 How to begin research Human behaviour

Society based on harmony

Society based on conflict

(functionalist)

(social action theories, Marxism)

University depart- Humanities ments

Social Sciences

As with the relationship between the person and society described above, the significance of taking a Japanological or Japanese Studies approach to a project is rarely explicitly addressed even if its impact is potentially considerable. To give just one example, whether we believe it is the past (‘history’) which determines the present (the Japanological approach) or the present which writes the past (the Japanese Studies approach) leads to a very different view of how we should think about contemporary Japan. Since they are social scientists, the three contributors of case studies to accompany this chapter all work in the Japanese Studies rather than Japanological tradition. All of their work is explicitly or implicitly comparative; they are interested in how the examples they look at in Japan shed light on the experience of similar phenomena in other countries—particularly their own (U.S., U.K. and Germany)—and vice versa. In order to do this, they all draw on theoretical ideas from their disciplines, which have a common currency, at least in the English-language Social Science literature on Japan. While they all place their studies in a historical context (Hendry’s study of marriage in rural Japan in the 1970s has a detailed analysis of historical antecedents, see this chapter, Ch. 1.3), they are sceptical about narratives which suggest Japan is somehow unique because of its distinctive history or topography. Where they do come across narratives of uniqueness—such as Japan being a society based on ‘natural’ consensusseeking harmony and group-mindedness—they question the source of such narratives and ask whose interests they serve—as in Blechinger-Talcott’s analysis of political corruption in Japan (see this chapter, Ch. 1.1). They all encourage the use of multiple theoretical perspectives and mixed methods in order, as Daniel Aldrich neatly puts it, ‘to convince skeptics that our findings are not an artefact of the way that we approached the problem’ (see this chapter, Ch. 1.2). Blechinger-Talcott most clearly picks out the distinction between Japanological and Japanese Studies approaches in her essay (see this chapter, Ch. 1.1). This is not surprising, since the philologically-based Japanological approach to Japan is still strong in continental Europe and the Social Science community tended until relatively recently to see the study of Japan as somehow ‘exotic’. The Japanese Studies community has had to fight hard in the past two decades to create a distinctive voice in Continental European institutions but, having done so, it possibly now enjoys a better, more mutually respectful, relationship with its Japanological colleagues than almost anywhere else.

6. Practical steps for beginning graduate research on Japan It is because so few researchers begin with analysing their intellectual assumptions that the majority of this essay has emphasised that element when beginning research. Most researchers do, however, start with themselves when they look for a topic to study in that they generally

37

Roger Goodman understand that they need to build a project around their own skills, ideally around a topic that they are in a position to study better than anyone else. Ironically, by identifying their socalled ‘unique selling points’ and designing a project around them, most researchers then discover that this is what they really do want to research because it is a topic they already know something about. Finding a topic is relatively easy compared to nailing down a puzzle within that topic which is going to keep the researcher engaged for months (in the cases of master’s students) and years (in the case of doctoral students). Put simply, research projects need a ‘research itch’. A research itch is a puzzle to which the researcher genuinely does not know the answer but the search for which will keep them intellectually challenged for the length of the project. The importance of the ‘research puzzle’ is that—even if the researcher never actually finds an exact ‘answer’ to the puzzle—it becomes the researcher’s ‘elevator pitch’ and sets the boundaries to the project and gives it an overall shape (see Vogt, Ch. 2). Most research puzzles are centred on specified data sets which appear to be counterintuitive or social institutions which cannot be explained in one’s own cultural terms. Examples of research puzzles which have guided my own research (and the publications which then appeared) over the past three decades include: • •

•

Why is Japan the only country in the world where the government has established special institutions for children who have returned from living overseas (Goodman 1990b)? Why was there such anxiety among the heads of children’s homes in Japan in the early 1990s around the reduction in the number of children needing to be taken into care (Goodman 2000b)? Why, when it was widely predicted in the mid-2000s that the number of private universities would fall over the following decade by between 15–30%, did the actual number increase by 15% (Breaden/Goodman 2020)?

Having found a research puzzle to which they genuinely do not know the answer and to which they cannot find an obvious answer in the research literature, the researcher needs next to undertake the hypothetical exercise of how a Marxist, Durkheimian and a Webarian scholar would approach this topic and what the implications of each of these approaches are for their methodology.8 The researcher also needs to run the project through the heuristic Japanological–Japanese Studies dichotomy since it is likely that much of the literature that they use (especially the literature in Japanese) will also be divided along these lines. Finally, the whole project is turned on its head, so that what is presented is not only an important puzzle which needs to be solved, but also one which the researcher is particularly well placed to tackle. Classically, therefore, the best research proposals—and certainly the ones most likely to win research funding—generally looks something like the following: •

This is the ‘research puzzle’ (written to catch the attention of the reader; the first two sentences are the most crucial of any research funding application).

8 All new graduate students in the Oxford School of Global and Area Studies are required to undertake such a hypothetical exercise as part of their first-year methods training course. The assignment that they are set is: ‘Take a research topic to do with contemporary Japanese, Chinese, Indian or any other society and describe and analyse what would be the different assumptions that a Marxist, Weberian and Durkheimian researcher would bring to such a topic—and how those assumptions might affect both their research questions and their research methodologies. The word limit for this exercise is 1000 words.’

38

Chapter 1 How to begin research • • • •

This is what has been done in this general area before (preferably not too much but also not nothing). This is why the research is so important (these are the theoretical, methodological, applied, ethnographic, data gaps it seeks to fill). This is how I am going to tackle it (an account of theory and methodology—where; how long; how). Just by chance: I happen to be particularly well qualified to address this puzzle because of my background, networks, language, research skills.

7. Summary This chapter on ‘how to begin research’ has focused more on the researcher than what they study since, as stated earlier, the researcher in the Social Sciences and the Humanities is the main research tool. Every researcher brings with them a bag of skills and strengths as well as biases and weaknesses which will, necessarily, affect the way that any research project is approached. These all need to be acknowledged before the project can even begin. If they are fully accommodated, then the project will be able to take an intellectual puzzle, examine it from all angles and make a serious contribution to our understanding of Japan. Indeed, the skill of the researcher to incorporate multiple theoretical positions and research methodologies in addressing important questions about Japan is what will distinguish them as an academic scholar.

39

1.1 Positioning one’s own research in Japanese Studies: Between Area Studies and discipline Verena Blechinger-Talcott

In Social Science research on Japan, and especially in the field of Japanese politics, identifying a good research topic often presents itself as a major challenge for young scholars. While Political Science usually expects scholars to develop research projects based on theoretical considerations, for example the relationship between two variables, and to identify cases for study according to features relevant to theory and related hypotheses, most students in Japanese Social Science research are genuinely interested in studying empirical phenomena in Japan. Scholars from the discipline might thus consider Area Studies (and Japanese Studies) less rigorous or theoretically sophisticated, more interested in thick description than in ‘relevant’ contributions to the field of Political Science. Traditional Japanologists, on the other hand, may challenge social scientists working on Japan for over-theorising or oversimplifying and over-reducing actual complexities in the interest of theoretical models. While there is no one-size-fits-all recipe with which to overcome these challenges, in my experience, it is helpful to base one’s own research on genuine empirical research based on phenomena in Japanese politics and/or society, while at the same time placing Japan in a broader comparative context. An active search for interdisciplinary debate is important, as is a true passion for one’s topic. My first research project on corruption in Japanese politics started out with a keen interest in institutions in Japanese law, politics and society (Blechinger 1998). I had just finished my MA in Japanese Studies with a thesis on the relationship between social practices and changing legal norms after 1945, looking at ways in which family law and new legal norms such as gender equality and individual freedom affected family relationships. I wanted to understand how normative change affected social behaviour, and how actual social practices affected the ways in which norms were shaped and implemented. For my MA, I had studied Civil Law, Political Science and Japanese Studies. Realising that I had acquired knowledge about Japan and the Japanese language, but was lacking the analytical and methodological tools to answer my questions, I enrolled in a PhD programme in Political Science, where I focused on institutions and the relations between politics and law. By the time I had completed my course work, debates in German and Japanese politics centred on issues of political finance and corruption, and I was puzzled by the differences in both debates. While the German debate about illegal party donations to the Christian Democratic Party (CDU) under the then chancellor Helmut Kohl was strongly shaped by arguments about personal misconduct and individual misconduct, arguments about systemic corruption and deeply entrenched practices of bribery were shaping debates in Japan. German newspaper reports about corruption in Japan also pointed to aspects of Japanese culture, such as gift-giving relations, to explain the assumed intrinsic nature of corruption in Japan. I was not only puzzled, but had also found my research topic: How could we explain the prevalence of corruption in political systems, and in which ways could legal reform, for example in the electoral system, affect corrupt phenomena? What 40

Chapter 1 How to begin research makes corruption systemic? What are the incentives for politicians and bureaucrats, but also for private sector actors, to engage in corrupt behaviour even in the face of highly negative sanctions? As corruption is a phenomenon that takes place in secrecy and usually only comes to light when it is exposed in a scandal, I had to spread my research further in order to find answers to my questions. In the following years, I analysed how politicians in Japan were financed, where they received funds for their work and how they defined the boundaries between legitimate behaviour and corruption. I also looked at the legal norms and related discourses in Japanese politics and its bureaucracy to prevent or at least reduce corruption. I interviewed politicians and spoke with political secretaries who administered politicians’ accounts and were involved in fundraising. I spoke to business representatives about their experience with political contributions, and I spent many hours studying records of parliamentary debates about political and campaign financing reform. I also participated in regular study groups on campaign financing in Tokyo and discussed the state of political financing with Japanese journalists who had followed campaigns and exposed (or decided not to expose) corruption scandals. I also followed politicians on the campaign trail to learn how they spent their funds and where they felt the pressure that might have made them inclined to take the risk of engaging in corrupt behaviour. The research led to my dissertation and my first book. Throughout that time, I worked in academic contexts on Japanese politics both in Japan and elsewhere, but I also formed a network with scholars working on corruption elsewhere. The comparative perspective, and also the questions asked by non-Japan specialists shaped my work and stimulated further research. At the same time, through exchange with ‘general’ political scientists and non-Japan area specialists, I learned how to position Japan as a case in a broader comparative context—which also was useful for countering arguments that focused on culture and gift-giving as the main cause of corruption in Japan. Through the work on my first project, I developed a keen interest in the relationships between business and politics as well as the state and the market, which has since shaped my academic work. Having studied the relationship between politics and money, I started to become interested in the role of companies as political actors, both at the domestic and the international level. In later projects, I looked into patterns and strategies used by Japanese (and international) firms to affect political decision-making processes. In my current research project, I am interested in the role of politics in globalised markets and especially in the governance of global value chains. When I started my dissertation, the field of Japanese Studies in Germany was just changing from scholars using a predominantly historical and philological approach to a more diverse field including Social Science approaches. At the same time, Political Science in Germany was very strongly focused on Germany, Europe and the U.S., and non-Western cases were not common. In Germany, my dissertation research was thus considered ‘exotic’ both for my colleagues in Japanese Studies and those in Political Science. I benefitted greatly from cooperation with scholars and colleagues in Japan, especially at the Institute of Social Science at the University of Tokyo, where I was able to learn from highly empirical political scientists with a strong comparative focus. I also encountered international graduate students working on Japanese politics. The study groups at the Institute of Social Science (and later at the German Institute for Japanese Studies, where I initiated the Social Science Study Group) and also the Japan Politics Colloquium at the University of Oxford, led by Arthur Stockwin at that time, 41

Verena Blechinger-Talcott provided a network of like-minded social scientists working on Japan and a forum for constructive criticism and exchange. Moreover, working together with doctoral students from Japanese universities who were interested in similar issues helped me to reflect on my basic assumptions and expectations in a double way—on the one hand, these discussions often challenged my somewhat German perspective on the ways of politics and, on the other hand, they allowed me to revisit my theoretical literature based on the empirical evidence from Japan. In two cases, we also did joint interviews. My presence as a German researcher allowed me to ask questions that would have been more difficult for my Japanese research colleague to ask. Afterwards, we compared notes about linguistic aspects of the interview. In summary, researchers studying Japan often face the challenge of balancing disciplinary and Area Studies’ demands. This will affect researchers’ choices of research topics as well as the ways in which they conduct and present research to appeal to different audiences. In order to perform this balancing act successfully, I suggest that young researchers start out with empirical research on Japan, but put their empirical findings in a broader comparative context and reach out to interdisciplinary debates—theoretical ones and debates that discuss Japan as a case among others. Academic debates often vary in different national contexts, as was the case with my research on corruption. These differences can pose puzzles and thereby motivate research projects. In this sense, researchers should always keep their eyes open for contradictions and differences in public and scholarly debates within and across national borders. Forming networks with scholars outside one’s discipline or national academic context as well as with those studying the same phenomenon in a different setting is another piece of advice I can offer to young scholars. This will help researchers to position their research in broader comparative contexts (see Kimura, Ch. 15.1). Collaboration with colleagues from both other disciplines and Area Studies, and particularly with colleagues from Japan, will provide researchers with inspiration and networks they can draw on in the future—both intellectually and professionally. In particular, I recommend that researchers make themselves familiar with academic debates on Japan in the Anglo-American community and reach out to colleagues from the U.S. and the U.K. This will help to produce research that the global community researching Japan will perceive and to which scholars from Area Studies and Social Science disciplines alike can relate to.

42

1.2 Let the field be your guide Daniel P. Aldrich

There is no single way to begin research, nor is there any sure-fire strategy to ensure that topics evolve into successful publications. Nevertheless, I am a big fan of several approaches, including building up interesting puzzles from real world empirical examples, being flexible when in the field, avoiding using culture as a catch-all explanation and writing about your interests and passions.

Puzzles from the real world Almost all of my research projects began as puzzles that I observed in the real world while spending time in the field, whether Japan or North America, and not from reading peer-reviewed articles, books or political theory. My first book project grew out of the failure of another, more standard Political Science project that I began while a graduate student at Harvard University. The abandoned project focused on the electoral strategies used by a Liberal Democratic Party (Jiyūminshutō, LDP) politician who was running for office. I hoped to follow in the footsteps of past social scientists like Gerald Curtis at Columbia University and Richard Fenno of the University of Rochester, both of whom ‘soaked and poked’ in the lifestyles of their subjects. Rather than writing articles and books from the comfort of a library carrel, these political scientists shadowed politicians, watching them on the campaign trail and talking with them after a day of glad-handing and baby kissing. My own dreams of success evaporated after several weeks of shadowing and ringing ears from the ‘nightingales’ (female announcers who used microphones to speak to crowds as their buses passed by) when my candidate abruptly lost the election and told me to get lost. Stuck in Japan for several more months without a viable project, I remembered a question that had come to me when thinking about Japan’s scientific progress following World War II: How did the only country in the world that experienced the horrors of nuclear weapons end up developing one of the most advanced commercial nuclear power programmes in the world? I wondered what the Japanese government had done to assist private utilities as they sought to promote atomic energy after going through the shock of the Hiroshima and Nagasaki bombings. With time on my hands, I wangled an interview with Tokyo Electric Power Company (Tōkyō Denryoku kabushiki-gaisha, TEPCO) executives at TEPCO HQ in Tokyo through some cold-calling to the phone number listed on their website. After I had pestered them a number of times, they invited me in, and I began to ask them about how they sited their nuclear power plants. Engineers and bureaucrats at that firm spoke of the ways that they sought to induce compliance through a variety of side payments and benefits. Then I began speaking with anti-nuclear activists at local organisations such as the Citizens’ Nuclear Information Center (Genshiryoku Shiryō Jōhōshitsu, CNIC). After some soaking and 43

Daniel P. Aldrich poking, I discovered a whole system of benefits and incentives offered to host communities in rural, coastal communities that were willing to have a nuclear facility in their backyard. The Japanese government had been far more than a passive umpire in the field of energy as some might envision. Instead, it took a side early—supporting the growth of the field in the late 1940s—and sought to support private energy firms throughout the nuclear power plant. This initial foray grew into several articles and the book Site fights: Divisive facilities and civil society in Japan and the West (Aldrich 2008).

From personal experience to a research project Where my first project sprang from the collapse of my intended research, my next major research project came from going through an actual disaster. As I was finishing up my dissertation on controversial facilities like nuclear power plants and turning it into a book, my family and I moved to New Orleans, Louisiana. There we settled into a short-lived but comfortable existence in the neighbourhood known as Lakeview, just south of Lake Pontchartrain. Within seven weeks of our arrival that name became all too real with the arrival of Hurricane Katrina and the collapse of the levees that held back Lake Pontchartrain and the Mississippi River. Lake water flooded our neighbourhood, with twelve feet of water destroying everything in our home, including my hard drive, all of our clothes, toys, books, records and material possessions. We got out alive, but evacuating and then trying to rebuild showed me how misconceived my vision of recovery was. My vision of disaster response involved U.S. government agencies like the Federal Emergency Management Agency (FEMA) swooping in to support survivors in a government-led process. Alternatively, the private market, such as homeowners’ insurance, was able to help rebuild. But rather than the government (through FEMA) or the market (through insurance), all of the aid, assistance, offers of places to stay and schools for our children came through friends, friends of friends and our social network. Instead of such aid coming from the market or the state, social capital and social ties proved to be the engine of resilience. I wondered if my own experiences might be similar to those of survivors from other major catastrophes around the world. With the support of the Abe fellowship, I spent time poking and soaking in disaster-affected communities around the world. This research grew into a comparative research project on disaster recovery in India, North America and Japan called Building resilience: Social capital in post-disaster recovery (Aldrich 2012). My most recent project came from watching Japan experience the triple disasters of March 11, 2011—the 9.0-magnitude earthquake, tsunami and nuclear power plant meltdowns (sadly enough at the Fukushima reactor near the town of Futaba, where I had done fieldwork for my first book). While many observers argued that cultural factors (Japanese stoicism, etc.) could account for the initial signs of recovery, and others claimed that it was going to be a function of damage (e.g. how high the tsunami was when it came ashore to each village), it was immediately obvious that certain locations were bouncing back faster than others. Some cities, towns and villages had higher levels of mortality, while others were decimated; and in the years since the events, some have rebuilt and increased in population, while others have only brought back the elderly. After I had spoken to several mayors, local administrators and NGOs, it was obvious that some had built strong and broad networks of assistance, while oth-

44

Chapter 1 How to begin research ers had far more limited ties. During the two years I spent in the field between 2011 and 2018 through a series of short, medium and longer stays (one funded by a Fulbright fellowship), this investigation became my newest book Black wave: How networks and governance shaped Japan’s 3/11 disasters (Aldrich 2019).

The challenge of Japan’s presumed uniqueness But studying Japan also comes with some difficulties. One of the regular challenges about using cases from Japan has been the belief in the uniqueness of Japan. If, as many Japanese and non-Japanese observers like to claim, Japan is indeed unique, then it is very hard to apply lessons and best practices from Japan to other settings. For example, if Japanese citizens are indeed the only ones in the world who say one thing and do another (tatemae and honne), or if there are esoteric aspects of the Japanese aesthetic that cannot be easily captured (wabi sabi), then there are few conversations we can have with scholars and observers of other nations and systems. It would be hard for a scholar of American or African politics, for example, to have a useful exchange of ideas with those of us studying Japan (see Kimura, Ch. 15.2; McElwain, Ch. 2.2). However, if Japan has institutions, incentives and policy arenas like other countries—and I’ve enjoyed scholarship from Hayden Lesbirel (1998), Steven Vogel (1996) and Richard Samuels (2003) that exactly builds on this approach—then we should be able to learn something from its experiences. One of the reasons I have followed the work of these scholars is because they begin by recognising explanations for empirical outcomes that stretch beyond those built on a belief in a nativist or unique culture. Instead, organisations and rules can in turn change behaviour and create new outcomes and norms. In this sense, we have seen more ‘mainstreaming’ of work on Japan, especially Japanese political outcomes, from various scholars (Catalinac 2016; Ono/Yamada 2018; Pekkanen et al. 2006; Rosenbluth/Thies 2010; Saito/ Horiuchi 2003) who use a variety of tools to demonstrate the broader lessons from events in Japan to events and phenomena far outside it.

Be flexible! I know from experience that it’s great to have a clear plan in mind before beginning a research project, whether one in a library carrel in Berlin or in the agricultural areas around Rokkasho. But I have also learned how important it is to be flexible and open to the realities of the field. Too often students may feel trapped by their proposal or by existing theories rather than feeling free to go off list and try out new approaches. I provide a long list of specific advice about beginning research in Japan in my article ‘The 800 pound gaijin in the room’ (Aldrich 2009), including suggestions on going with letters of introduction, business cards, thank-you gifts and an affiliation with a Japanese institution. I also strongly suggest using multiple methods. That is, if you enter graduate school able to carry out a regression analysis using quantitative data, then you should leave with a new skill set, such as the ability to carry out focus groups, interviews and participant observation. If, on the other hand, you’re only comfortable using qualitative methods, then you should take courses in social network analysis, regression analysis

45

Daniel P. Aldrich and geographic information systems (GIS) analysis to expand your toolkit. Our research should always be driven by a problem, not by our methods. If we can only carry out one type of analysis, we miss the chance to study a phenomenon from other angles and to convince sceptics that our findings are not an artefact of the way that we approached the problem. Having a broader toolkit means that when you tackle a new problem you’ll be able to come at it convincingly from multiple angles (see Hommerich/Kottmann, Ch. 10). In this sense, once you’ve found an interesting puzzle, I would encourage students to think about the different ways to empirically understand it, running from direct talks with relevant actors to a map of their social network, to a survey of communities in which they operate. One lesson from my own career has been that projects need to develop organically from empirical observation. I would encourage graduate students to keep their eyes open to the reallife puzzles that are constantly emerging around us and to think through ways to study those outcomes methodically and systematically.

46

1.3 Studying marriage in Japan: A social anthropological approach Joy Hendry

My choice of marriage as a subject to study in Japan was made rather easily, for I spent six months prior to my formal study sharing a room in a house with five young Japanese girls, for whom the subject was far and away the most discussed. This period was largely for language study, and I was enrolled at the time in an intensive course at a language school in Shibuya/ Tokyo. Fortuitously I saw an advert offering accommodation in a house with ten young Japanese who wanted to have a native English speaker in their midst, and we spoke in English every evening over dinner. Otherwise they chatted to each other in Japanese, which not only helped my own language acquisition but gave me a wonderful insight into issues interesting to young people at the time. In the case of my female roommates, this was definitely marriage, especially whether such a thing should be based on love or arranged by their elders. My topic was decided then (Hendry 1981/2011 for the outcome).

Finding a field site Starting fieldwork in the discipline of Social Anthropology requires planning, of course, but serendipity is also useful for a successful study (see Coates, Ch. 3.2; Gagné, Ch. 6.1; Klien, Ch. 8.1), as we have seen in the way my attention was drawn to the topic. When I did that first fieldwork many years ago now, we were actually given little preparation, but it was usual to expect to spend at least one year in the same place. The idea was to get to know all the people in a chosen area—it could be a geographical area, a community around a common interest or perhaps an enterprise of some sort. Whichever area was chosen, spending a year with people enables several things. First, it sees through a full annual cycle of events—understanding the seasons and attitudes to them and witnessing the enactment of all the annual rituals. A year also gives the researcher time to get to know people well, and indeed, the people to know the researcher and to understand what they are about. In Japan, first questions may elicit answers that the interlocutor thinks the researcher may want to hear; with time, an in-depth response is more likely to be revealed. A year also allows the researcher to become used to the local dialect and linguistic idiosyncrasies. These vary greatly throughout Japan, even more so in Okinawa, for example, and failure to take them into account could result in severe misunderstandings. With these issues in mind, and my choice of marriage as a topic, I set out to find a suitable location. As it happened, I thought it would be good to work in a village. It was common practice in those days, and I resolved to look at the rituals involved in building a relationship and expectations for the future, as well as the various ways of meeting a suitable partner that my Tokyo friends had been debating. Within one village, I would be able to place the subject 47

Joy Hendry of marriage within a broader social context. It didn’t really matter to me where the village would be, or what would be the local economic base; there is quite a bit of Japanese literature on marriage and I was able to compare my findings with those picked up elsewhere. I simply needed to find a place to live for me and my husband, among a manageable number of houses, and I thought it would be good to find a beautiful spot. The most important plan then—a vital plan for an anthropologist or indeed any researcher—is to start out with some good introductions to the people with whom you will work. When I set out to do my first fieldwork, my supervisors were in Oxford, and at the time there was no one there who had worked in Japan, so I asked a Japanese scholar who had visited my department in Oxford to help. He introduced me to a senior Japanese anthropologist who has helped me all my life, and I realised that it is always a good plan to have a supervisor in Japan wherever a student’s university is based. They can provide a great deal of local assistance unavailable at home, and mine was able to tell me about the related fieldwork his colleagues and students were doing. This introduction also gave me a university attachment in Tokyo, which I think inspired more confidence in the people I approached than my Oxford one did. So I have tried hard ever since to procure the same facility for all my PhD students. We discussed various possibilities for locations, and I spent some time visiting a selection of them. In every case, I would need somewhere to live, so this was an important consideration within a relatively small community. For the first village I tried, I only had a personal connection through a friend, and people seemed suspicious. For the second, I was introduced by an English teacher to the local education office as I had heard that they had houses for teachers. They did indeed and were kind enough to take me out to see some. It was a delightful area in rural Shikoku, and stunningly beautiful, but my project seemed likely to fail because they revealed that, sadly, all the young people were leaving. In the end, the village I found was a thriving community in Kyushu (Hendry 2021), and I found it through a Japanese anthropologist who had worked in the area—a student of my supervisor, as it happened. He not only found me an empty house, but took it upon himself to introduce me to all the important people in the area and to make sure they knew who I was and what I was planning to do. That was wonderful, for the head of the village immediately invited me to his son’s wedding, where I met and shared sake with almost all his neighbours, who were happy to help me with my research afterwards. A first stroke of serendipity then, because they asked me (and my husband) all sorts of questions about our marriage, so I assume they then felt some obligation to reciprocate. I also learned that weddings are a great time to discuss details about marriages, and fortunately I was invited to many more (Hendry 2003).

Settling in My new next-door neighbour explained another Japanese custom, which I would recommend to all those who plan to live in Japan, anthropologists or otherwise. Later, I learned and could identify all the important divisions for sharing community tasks, but for the time being my neighbour took me to the other houses in the immediate vicinity, where I introduced myself and handed over a small gift. ‘Not too big,’ he said, ‘they won’t want to be obligated to you.’ So I gave them a few postcards from Oxford. My house was actually over the border from the village I had chosen as my focus, so I didn’t see those neighbours much over the year, but some

48

Chapter 1 How to begin research 40 years later, when I approached one of them for business purposes, he remembered me, and it was quite helpful. These people were also then able to explain to curious strangers who the ‘funny foreigners’ were—we were rather rare in those days—certainly the only two in our immediate vicinity, possibly the only two in a town of 35,000. Settled in, I then had to work out a way to approach people, and to start the inquiry. An advantage of having a year to spend is that there is no need to immediately impose a list of questions on people. Of course, it is useful to have an idea of the questions you want answered, but I found that I learned a lot more if I was able to insinuate myself into open situations where people were already talking, and gradually steer the conversation around to my subjects of interest. In the village, my first task was to identify times and places where I would naturally meet people going about their everyday lives. There were three shops, and these were always good locations, the two fish shops attracting two different generations, which helped me to understand in-law issues, and a tobacconist’s, which I later discovered was a favourite place for outsiders to ask about local families (with a view to arranging a marriage). There was also a village hall where meetings took place, but most helpful of all in those days was the village bath house. Almost everyone went there: the older women first, then the younger ones, and finally the housewives, so I could choose my time depending on what I was after, and people were wonderfully talkative soaking in the hot water! Another good approach to learn about the villagers was through the local policeman. He lived with his family in a nearby police house, and he kept a detailed list of all the occupants of his patch, together with a record of particularly valuable property. Probably because of the appropriate introduction from the Japanese anthropologist, he was willing to share all this with me. It was a perfect introduction because I created a notebook, which I use to this day, in which I entered the names of the residents of each of the 54 houses in the village, and then called on them in turn to verify his record, and to ask about how their marriages were arranged. It sounds pretty cheeky, and I am not sure it would work everywhere, but the people of this village cooperated. I also approached local policemen in later research projects, but in some cases you needed an introduction. My introduction from Kyushu to other far distant places actually worked better than a letter from my own university. In one area, they filled me in on all the yakuza families in the area, and I discovered that the son of one of them had become best friends with one of my sons at school.

Some final thoughts on taking notes In that village notebook, I eventually collected the names and dates of death of all the ancestors remembered in the Buddhist altars, and I made a detailed diagram of all the families and their relationships. It was very useful to see who was related to whom and how that affected their invitations to weddings, introduction to potential spouses (still common in those days) and other life-cycle events including funerals. I have returned to this village many times over the forty years since I first worked there. The notebook offers me a great opportunity to go around updating it with new births, marriages and deaths, asking to pray at the Buddhist altar to say goodbye to those with whom I had worked when they were alive, and generally keeping in touch. Early on, I also made a detailed map of the houses, which I numbered, and this helped me to find my way around. I recommend that both these tasks are undertaken at the

49

Joy Hendry very start of a research project, though of course the notebook may be a computer file these days, but together they become a superb investment for all subsequent activities (see Kottmann/Reiher, Ch. 7). A general diary is also crucial, for things observed early on are only properly explained much later, and small things may be forgotten if not recorded. I almost never used a tape-recorder, although I know others do, but I found that people would often elaborate on what they first said while I was writing, and things would come out that I had never thought to ask.

50

Further reading Befu, Harumi/Kreiner, Josef (eds.) (1992): Othernesses of Japan: Historical and cultural influences on Japanese Studies in ten countries. München: Iudicium. Bourdieu, Pierre (1977): Outline of a theory of practice. Cambridge: Cambridge University Press. Goodman, Roger (1984): Is there an ‘I’ in Anthropology? Thoughts on starting fieldwork in Japan. In: Journal of the Anthropological Society of Oxford 15, No. 2, pp. 157–168. Lury, Celia/Fensham, Rachel/Heller-Nicholas, Alexandra/Lammes, Sybille/Last, Angela/Michael, Mike/ Uprichard, Emma (2018): Routledge handbook of interdisciplinary research methods. London: Routledge. Okely, Judith/Callaway, Helen (1992): Anthropology and autobiography. London: Routledge. O’Reilly, Karen (2009): Key concepts in ethnography. Thousand Oaks, CA: Sage.

References Aldrich, Daniel P. (2008): Site fights: Divisive facilities and civil society in Japan and the West. Ithaca, NY: Cornell University Press. Aldrich, Daniel P. (2009): The 800-pound gaijin in the room: Strategies and tactics for conducting fieldwork in Japan and abroad. In: PS: Political Science and Politics 42, No. 2, pp. 299–303. Aldrich, Daniel P. (2012): Building resilience: Social capital in post-disaster recovery. Chicago, IL: University of Chicago Press. Aldrich, Daniel P. (2019): Black wave: How networks and governance shaped Japan’s 3/11 disasters. Chicago, IL: University of Chicago Press. Befu, Harumi/Kreiner, Josef (eds.) (1992): Othernesses of Japan: Historical and cultural influences on Japanese Studies in ten countries. München: Iudicium. Blechinger, Verena (1998): Politische Korruption in Japan: Ursachen, Gründe und Reformversuche. Hamburg: Institute of Asian Affairs, Vol. 291. Bourdieu, Pierre (1977): Outline of a theory of practice. Cambridge: Cambridge University Press. Breaden, Jeremy/Goodman, Roger (2020): Family-run universities in Japan: Sources of inbuilt resilience in the face of demographic pressure, 1992–2030. Oxford: Oxford University Press. Catalinac, Amy (2016): Electoral reform and national security in Japan: From pork to foreign policy. New York, NY: Cambridge University Press. Cohen, Anthony (1985): The symbolic construction of community. London: Tavistock Publications. Goodman, Roger (1984): Is there an ‘I’ in Anthropology? Thoughts on starting fieldwork in Japan. In: Journal of the Anthropological Society of Oxford 15, No. 2, pp. 157–168. Goodman, Roger (1990a): Deconstructing an anthropological text: A ‘moving’ account of returnee schoolchildren in contemporary Japan. In: Ben-Ari, Eyal/Moeran, Brian/Valentine, James (eds.): Unwrapping Japan: Society and culture in anthropological perspective. Manchester: Manchester University Press, pp. 163–187. Goodman, Roger (1990b): Japan’s ‘international youth’: The emergence of a new class of schoolchildren. Oxford: Oxford University Press. Goodman, Roger (2000a): Fieldwork and reflexivity: Thoughts from the Anthropology of Japan. In: Dresch, Paul/James, Wendy/Parkin, David (eds.): Anthropologists in a wider world: Essays on field research. New York, NY: Berghahn, pp. 151–165. Goodman, Roger (2000b): Children of the Japanese state: The changing role of child protection institutions in contemporary Japan. Oxford: Oxford University Press. Goodman, Roger (2006): Thoughts on the relationship between anthropological theory, methods and the study of Japanese society. In: Hendry, Joy/Wong, Dixon (eds.): Dismantling the East-West dichotomy: Views from Japanese Anthropology. London: Routledge, pp. 22–30. Goodman, Roger (2020): Education and the construction of Japanese national identity: Rhetoric and reality. In: Almqvist, Kurt/Duke Bergman, Yukiko (eds.): Japan’s past and present. Stockholm: Bokförlaget Stolpe. Hamabata, Matthews M. (1990): Crested kimono: Power and love in the Japanese business family. Ithaca, NY: Cornell University Press. Hendry, Joy (1981): Marriage in changing Japan: Community and society. London: Croom Helm.

51

References Hendry, Joy (2003): From scrambled messages to an impromptu dip: Serendipity in finding a field location. In: Bestor, Theodore C./Steinhoff, Patricia G./Lyon-Bestor, Victoria (eds.): Doing fieldwork in Japan. Honolulu, HI: Hawai‘i University Press, pp. 55–70. Hendry, Joy (2011): Marriage in changing Japan: Community and society. London: Routledge. Hendry, Joy (2021): An affair with a village. Stirling, Scotland: Extremis Publishing. Johnson, Sheila (1975): American attitudes toward Japan, 1941–75. Washington, D.C.: AEI Press. Kondo, Dorinne (1990): Crafting selves: Power, gender and discourses of identity in a Japanese workplace. Chicago, IL: Chicago University Press. Lesbirel, Hayden (1998): NIMBY politics in Japan: Energy siting and the management of environmental conflict. Ithaca, NY: Cornell University Press. Lunsing, Wim (2001): Beyond common sense: Sexuality and gender in contemporary Japan. London: Kegan Paul International. McLelland, Mark J. (2000): Male homosexuality in Japan: Cultural myths and social realities. Richmond: Curzon Press. Merton, Robert K. (1938): Social structure and anomie. In: American Sociological Review 3, No. 5, pp. 672–682. Moeran, Brian (1985): Okubo diary: Portrait of a Japanese valley. Palo Alto, CA: Stanford University Press. Mouer, Ross/Sugimoto, Yoshio (1986): Images of Japanese society: A study in the structure of reality. London: Kegan Paul International. Nakane, Chie (1970): Japanese society. Harmondsworth: Penguin Books. Okely, Judith/Callaway, Helen (1992): Anthropology and autobiography. London: Routledge. Ono, Yoshikuni/Yamada, Masahiro (2018): Do voters prefer gender stereotypic candidates? Evidence from a conjoint survey experiment in Japan. In: Political Science Research and Methods. DOI: 10.1017/ psrm.2018.41. O’Reilly, Karen (2009): Key concepts in ethnography. Thousand Oaks, CA: Sage. Pekkanen, Robert/Nyblade, Benjamin/Krauss, Ellis (2006): Electoral incentives in mixed-member systems: Party, posts, and zombie politicians in Japan. In: American Political Science Review 100, No. 2, pp. 183–193. Reynolds, David K. (1989): Flowing bridge, quiet waters: Japanese psychotherapies, Morita and Naikan. New York, NY: SUNY Press. Roberts, Glenda S. (2003): Bottom up, top down and sideways: Studying corporations, government programs and NPOs. In: Bestor, Theodore C./Steinhoff, Patricia G./Lyon-Bestor, Victoria (eds.): Doing fieldwork in Japan. Honolulu: University of Hawai‘i Press, pp. 294–314. Rosenbluth, Frances/Thies, Michael (2010): Japan transformed: Political change and economic restructuring. Princeton, NJ: Princeton University Press. Roth, Joshua (2002): Brokered homeland: Japanese Brazilian migrants in Japan. Ithaca, NY: Cornell University Press. Saito, Jun/Horiuchi, Yusaku (2003): Reapportionment and redistribution: Consequences of electoral reform in Japan. In: American Journal of Political Science 47, No. 4, pp. 669–682. Samuels, Richard (2003): Machiavelli’s children: Leaders and their legacies in Italy and Japan. Ithaca, NY: Cornell University Press. Sever, Mustafa (2012): A critical look at the theories of sociology of education. In: Journal of Human Sciences 9, No. 1, pp. 671–650. Smith, Robert J. (2003): Time and Ethnology: Long-term field research. In: Bestor, Theodore C./Steinhoff, Patricia G./Lyon-Bestor, Victoria (eds.): Doing fieldwork in Japan. Honolulu, HI: University of Hawai‘i Press, pp. 352–366. Steven, Rob (1982): Classes in contemporary Japan. Cambridge: Cambridge University Press. Tsuda, Takeyuki (2003): Strangers in the ethnic homeland: Japanese Brazilian return migration in transnational perspective. New York, NY: Columbia University Press. Treat, John W. (1999): Great mirrors shattered: Homosexuality, orientalism, and Japan. New York, NY: Oxford University Press. Vogel, Steven (1996): Freer markets, more rules: Regulatory reform in advanced industrial countries. Ithaca, NY: Cornell University Press. Woods, Peter (1979): The divided school. London: Routledge.

52

Chapter 2 How to ask: Research questions Gabriele Vogt

1. Introduction We constantly ask questions, but we rarely ever think about the very process of doing so. We wonder, we ponder and we eventually pose questions that we strive to explore. Will it be raining today? Should I bring an umbrella? What are the chances I will end up soaked if I decide to leave the house without it? This chapter sets out to highlight the process of asking questions, not just any questions, but research questions. A research question will be your companion for a couple of weeks or months or years, depending on the scale of the research you embark on. No matter how long you plan on sticking to each other, it is of utmost importance that you choose your companion wisely, and not only because you will need to face them day in and day out (and in your dreams or nightmares). Your research questions will also determine what kind of theories and methods you will need to familiarise yourself with, what kind of data you will need to gather and analyse, and, finally, what the punchline of your research project will be. Let me try and give some structure to what, in the beginning, might seem to be an impenetrable jungle: How can you carve out your research question from what usually starts as a broad field of interest? This chapter, in particular, follows three lead questions: What are the specific steps in designing a research question? What difficulties will you most likely face in the process? And what strategies are there to overcome those difficulties? Before discussing this process of asking a research question, I will provide an overview of the types of research questions and address the elements generally deemed essential to any good research question. Moreover, throughout this chapter, I will highlight aspects that I deem important or at least peculiar when formulating research questions in Japanese Studies. I do so as a long-time student of Japanese politics and society myself, and as a teacher who has supervised a hundred MA theses and two handfuls of PhD theses. Disclaimer: While experience surely does no harm, every new endeavour of embarking on a research project follows its own set of rules, and your supervisor will only be able to assist you up to a certain point. From there on you (and your research question) will need to fly on your own.

53

Gabriele Vogt

2. What is the core of a research question? Every research question is based on an idea, a genuine or vague interest in a phenomenon and curiosity. Yet, in Social Sciences not every idea that pops into your mind will eventually develop into a research question. Firstly, we need to make sure that nobody else has had and developed the same idea before (i.e. we need to do our literature review, see Zachmann, Ch. 4; for good research practice see Reiher/Wagner, Ch. 16), and, secondly, we need to embark on the often painful process of pairing an idea that we feel passionate about with social scientific rigour (i.e. we need to be thorough in our research design, see Okano, Ch. 3). And then, of course, there is the question of what to do when no idea pops up. Let yourself be consoled by the experience of Max Weber: ‘ideas come when we do not expect them, and not when we are brooding and searching at our desks. Yet ideas would certainly not come to mind had we not brooded at our desks and searched for answers with passionate devotion’ (Gerth/ Mills 1948, p. 136). So, reaching the idea that will eventually turn into our research question is equally a matter of hard work and patience. For Japan scholars, in particular when you study contemporary Japan, these ideas often originate while you are in Japan. You may find yourself riding a train in rural Japan and notice how you feel increasingly annoyed by many passengers staring at you or calling you gaijin (foreigner) behind your back. You may feel intrigued by a personal story a Japanese friend shares, or you may stumble across an exciting picture in a museum—anything that sticks with you has the potential of awakening your curiosity. The train story is actually my own: I experienced the gaijin calls in the late 1980s, and while I was still in junior high school back then, these train rides may have sparked my general interest in studying international migration to Japan. I did not find ‘my’ research question, though, until many years later—and after reading widely on Japanese politics and society, thoroughly familiarising myself with the numbers and laws in Japan and in Germany (adding a comparative matrix focuses your view onto the actual case study; see Nakano, Ch. 3.1), debating my hinges with friends and colleagues, and diving deeply into migration literature at the same time.

3. Types of research questions: The common denominator and specific forms Let us now think about the general types and characteristics of research questions first before addressing the more hands-on issues of how to turn our idea into a research question. Karen Mattick, Jenny Johnston and Anne de la Croix (2018, p. 104) provide one of the most straightforward and, at the same time, simplest definitions of what a research question is: ‘a research question is a question that a research project sets out to answer.’ While we can agree that this is indeed the common denominator of research questions, it is also important to acknowledge that research questions take on different forms depending on the specific research project. Are you engaged in a descriptive study or an analytical study? Or maybe in exploratory research of a subject that has hitherto not been studied? In a study that aims to compare, 54

Chapter 2 How to ask e.g. across time or across countries or cultures, a study that explains or evaluates, e.g. the effect of certain policies? In a study that provides a close-up on one case, or a study that aims to contribute to theory building? Are you proceeding inductively or deductively (see Okano, Ch. 3)? Moira Kelly (2018, p. 82) provides a set of six types of research questions, which reflect this variety of approaches—some of them being more suitable for qualitative research, some of them rather for quantitative research designs aimed at testing variables (see Hommerich/ Kottmann, Ch. 10). I will list them here and, in order to clarify the scope of the various research questions with some concrete examples, I will add some mock questions, thematically connected to my research on the migration of health-caregivers from South East Asian countries to Japan (Vogt 2018). Types of research questions 1.

2.

3.

4. 5. 6.

Describing and finding correlations: What are the characteristics of the newly introduced migration avenue for health-caregivers from South East Asian countries to Japan? How is the recruitment of health-caregivers connected to developments in Japan’s labour market? Examining an aspect of an issue in detail: What is the role of language training in the newly introduced migration avenue for health-caregivers from South East Asian countries to Japan? Drawing on theory to examine an issue: Does the push/pull-model of international labour migration (Hollifield 2000) explain the small scope of health-caregiver migration to Japan? Comparing attributes: Do gender, ethnicity or age influence the likelihood of potential migrants applying for health-caregiver migration to Japan? Explaining: Why do only few health-caregivers come to Japan under the current models of labour migration? Assessing whether an intervention works: Would a reform of language education in the sending countries help raise the number of international health-caregiver migrants to Japan?

While I framed my mock questions according to Kelly’s model (2018, p. 82), I should add two aspects that readers might want to understand as warnings regarding these very categories: first, I firmly believe that, whenever possible (and it would easily be possible in the above examples), we should stay clear from yes/no questions. This is because 1. they are less exciting than any other question you could ask,1 and 2. because social reality is never black or white, which means that the answer to yes/no questions will always be a ‘to such and such a degree’ rather than a simple ‘yes’ or ‘no’. Secondly, most research projects will have components of more than just one of these categories inherent in their design. In particular, I would hope that no piece of research stops at a simple description of a phenomenon (category 1 question) without developing the knowledge gained by describing this phenomenon into further questions that are designed to strengthen 1 Andreas Sebe-Opfermann (2016, p. 26), for example, states that yes/no questions are too simple and not worthwhile pursuing at all.

55

Gabriele Vogt the analytical component of the research. Also, projects tend to consist of more than just one research question. Often you see a lead question supplemented by two or three sub-questions. Either way, part of the process of developing your research question is to decide which kind of research question to pursue. This decision is strongly connected to the question of how to design your overall research project. I will address this interwoven process in more detail in the following subsection. Before I do so, let me lay out some of the core characteristics that all research questions share. Key issues Your research question is the linchpin of your research project. Creativity and hard work interplay when you are carving out your research question. Good research questions stress the analytical component over the descriptive side.

4. The characteristics of research questions: Relevance, originality and rigour Relevance, originality and rigour are the three core characteristics of any well-designed research question (Mattick et al. 2018, pp. 105–107).

4.1 Relevance A research question should address something relevant. This immediately opens up the question of relevance to whom? When you think about the relevance of your research question, it is important to be aware of the audience you are writing for. Is it your thesis supervisor, the editorial board of a journal, a committee deliberating approving or declining a research grant proposal? Is it a researcher from your discipline or another? Or is it the general public, the media or policymakers? In more general terms, it is important that you are able to explain the relevance of your research question to any given audience (see Farrer/Liu-Farrer, Ch. 17). So be prepared to deliver the core content of your research question (and by extension your research project) to a variety of interested people, i.e. in scholarly terms as well as in lay terms (Mattick et al. 2018, p. 105). Also, observe your audience when talking about your research question, since you will be able to draw insights from their reactions as well. Is this project equally relevant to others? Are you getting to your point quickly enough? I.e. have your ‘elevator pitch’ prepared: the content and relevance of one PhD thesis in 30 seconds! I have always found it extremely painful to squeeze years of work and thinking into the blink of an eye, yet let me assure you that it has been extremely useful at times: as an opener to an interview with a politician who is juggling meeting with you between a campaign speech and a strategy meeting in her office, or a coffee break conversation after a panel with business representatives who grant you just enough time for the ‘elevator pitch’ and an exchange of name cards before floating on to the next person. The

56

Chapter 2 How to ask quality of your ‘elevator pitch’ will determine whether you will be able to land a follow-up meeting and an actual interview appointment (see McLaughlin, Ch. 6). Moreover, as a researcher on Japan, be prepared to deliver the core content of your research question and project not only in your mother tongue, but also in English and Japanese. Additionally, be prepared to answer questions regarding the relevance of your research question to the country you are from. For example, I found it extremely useful to study the history and scope of international labour migration to Germany before/while studying this issue in Japan, since I was frequently asked about this comparative dimension during interviews, in the question and answer sessions following presentations, etc.

4.2 Originality ‘[E]nsure that the research question will lead to original work that generates new insights and does not duplicate previous research’ (Mattick et al. 2018, p. 106; see Reiher/Wagner, Ch. 16). There are essentially two ways to ensure the originality of your research, and I strongly advise you to see both of them through: a thorough literature review, as well as multiple and intense debates about your research question with peers and teachers (see Zachmann, Ch. 4). The how to of a literature review is dealt with in chapter four of this book,2 but at this point, let me address two aspects that are of particular importance with regard to the relevance of the literature review for ensuring the originality of your research. As Japan scholars, you must not ignore the probably vast research literature on your chosen research area that has been published in the Japanese language. I speak from own experience when I stress that in some cases the research literature in Western languages may cover one or at best two dozen volumes (which you can go through relatively speedily), while the research literature in Japanese can fill a whole section in any major bookstore or library (which will take you much more time to work your way through). I certainly found this to be the case with literature on the Okinawan protest movement against the continuous system of forward deployment of U.S. military units in the prefecture (Vogt 2003). Also, while doing your literature review, revisit your research question numerous times (Browne 2013, p. 103). In fact, I suggest you engage in some sort of ping-pong game between reading what has been done in your research field, and refining your original question, until you are confident that you have reached and are able to articulate the research gap that your work will be able to fill (see Zachmann, Ch. 4). Next to the rather quiet work of reading, debating your research question with peers and teachers and basically anybody who would take the time to listen to your early ramblings, and later on more refined thoughts is equally important in ensuring the originality of your research. You may notice that all three essays paired with this chapter—by David Chiavacci, Kenneth Mori McElwain and Nicolas Sternsdorff-Cisterna—refer to this very point. Kelly (2018, p. 85) calls this process the ‘“friendly” version of peer review’. In this process you look for both academic and practical advice. The academic advice will essentially be directed at the quality of your research question. It can trigger you to rethink your perspectives, and—in a best-case scenario—it can reopen your eyes to patches to which you have gone almost blind because of the overfamiliarity you have already acquired with your research subject. Also, very 2 Also, I found Anthony Onwuegbuzie and Rebecca Frels (2016) to be very useful as a book to be used in the classroom when teaching the path to a comprehensive literature review.

57

Gabriele Vogt importantly, choose your debate partners from backgrounds as diverse as possible. McElwain (see this chapter, Ch. 2.2) explicitly refers to this point when explaining how he improved his own research projects. And, again, make sure you also choose Japanese colleagues for your debate group. You can initiate some of these encounters yourself: volunteer to present your project in study groups and workshops designed for junior scholars; go lunching with your supervisor; attend conferences and make sure you meet new people during coffee breaks and social events. Some of these encounters will just happen to you: the person in the seat next to you on a plane ride might turn out to be full of practical advice regarding the research question you have in mind. On a flight from Dubai to Manila, I once talked for hours with a registered Filipina nurse working abroad. By sharing some of her life course events, she enlightened me with hints at where my research question was still insufficient.

4.3 Rigour Finally, rigour is among the core characteristics of any well-designed research question. Rigour speaks directly to the development of a research question. Mattick et al. (2018, p. 107) highlight several points ensuring rigour in research questions, but most prominently they stress the need to align the research question with methods of data collection and data analysis that you can apply. This is a valid point that I will come back to in the following subsection, and also discuss the somewhat extreme case of this alignment, i.e. cases when research questions are ‘shaped by researchers’ preferred methodological tools, which are inevitably linked with the way they see the world’ (Mattick et al. 2018, p. 107; see Goodman, Ch. 1). One final word of advice regarding the rigour of research is to handle research questions that are close to your heart with even more carefulness than any other. If you feel too passionate, too emotionally involved with a research question, you may lose the objectivity you need to do research (Sebe-Opfermann 2016, p. 23). The degree of personal involvement with a certain topic is something that ultimately everybody needs to judge for themselves. How research disciplines position themselves to this question, where the disciplines draw their boundaries and what tools they suggest applying in order to handle emotion in research varies largely. While, for example, anthropologists to some degree need and want to immerse themselves fully into the field, many political scientists shy away from even becoming a member of a political party as they strive for a neutral and detached position within the area they study. Just some months ago, while studying political participation in Okinawa, I attended a citizens’ rally and found myself struggling over whether to join the concluding chants and whether to hold the protest posters up high in the air or not. Where is the point where I lose my distanced view of the research object, and ‘go native’ instead? To what degree can I actually keep a distance from an issue I have been following for a quarter century and feel deeply sympathetic about? There is no one-size-fits-all answer to this question. I would, however, like to urge you to keep in mind your role as a researcher, to be reflexive about your research methods and activities throughout the process of conducting research, and to disclose them without holding back on any of the applied methods and activities (see Alexy, Ch. 7.3). As a supervisor I have, however, come to dread theses that reflect a student’s personal life course events, political engagement, sexual orientation, etc., as very often these theses trade in a rigorous research design for emotional involvement and an activist tone, and eventually re-

58

Chapter 2 How to ask semble policy papers rather than academic theses. A more experienced researcher, however, may well be equipped with enough coolness to tackle issues close to their heart in a rigorous manner nevertheless. However, do not let yourself be fooled: a research project that you feel passionate about will, on the one hand, help you pull through the times when you find it hard to motivate yourself to keep working on your project; on the other hand, it is always more difficult to pull through with such a project, as the constant efforts to stay away from any subjectivity—which will be needed in data collection, data analysis and in the process of writing up your study—can be quite tiring. Key issues Relevance, originality and rigour are the core components of any good research question. Make sure you develop your research question in close dialogue with the expertise present on your topic in Japan. Always keep in mind who your audience is. Be aware of the difficulties of emotional engagement with your research topic.

5. How do I develop my research question? Now that the general types and characteristics of research questions have been laid out, how do we develop our research question? Think about the following two approaches. Firstly, to borrow the words of Kelly (2018, p. 81): ‘We ask questions all the time, so how hard can it be? Developing good research questions, however, involves a level of craft.’ Secondly, in the words of Doris Leung and Jennifer Lapum (2005, p. 63) while reflecting on their own research process: ‘Poetry has taken them beyond the traditional limits of knowing and allowed them to conceptualise their research questions by situating and locating their selves within their research.’ I argue that we need in fact both: we do need the craft component in order to design our research question, and we do need a bit of art in order to transform our research question into something enlightening. The good news is you can learn craft, and you can hope for arty inspiration. Actually, there is no bad news to this story. (Please do not despair if writing poems has so far never inspired you to draw up a research question!) Now let us focus on the steps you can acquire through training.

5.1 The process: Ping-ponging back and forth When discussing the originality of your research question in the subchapter above, I have already sketched out the idea of a ping-pong game: your research question will be refined from a broad idea to actual ‘research question[s] [that] will lead to a project that aims to generate new insights’ (Mattick et al. 2018, p. 104) by going back and forth between the initial idea, the relevant secondary literature and debates with peers. Each round you play of this game, your question will become more refined. You may need to take a detour once in a while, and be assured that this is rather the rule than the exception. Let me stress two aspects in particular.

59

Gabriele Vogt Detour: While ping-ponging, be prepared not only to go multiple rounds, but also to ‘consider alternative questions’ (Kelly 2018, p. 84) altogether. Obviously, this will be necessary when you happen to figure out that your initial research question is lacking in relevance, originality or rigour (see above). Also, if you feel that your initial idea is not going anywhere, such as, for example, when a policy reform your hypothesis is based on does not come through (see this chapter, Chiavacci, Ch. 2.3), or when social reality is changing quickly, which might make your research question significantly less interesting or even obsolete (see this chapter, McElwain, Ch. 2.2). Sometimes you cannot help but ‘[g]o back to the basics’ (Bodemer/Ruggeri 2012, p. 1439). It always hurts to make a radical cut like that, but be aware that following through with a research question that, for valid reasons, you can no longer fully support is the significantly worse choice than starting from scratch again. Most likely, by the way, you will not even need to start from scratch, but will be able to build on your previous work in ways that might surprise you. In 2005, when I started researching labour migration to Japan, I basically started to wrap my head around a ‘non-case study’. Little did I know that in 2006, a new government initiative would open up a sector-specific migration avenue to Japan. Needless to say, I needed to completely rethink the story and refine my research question. However, I was still able to use what I had studied on migration policymaking in Japan, on the stakeholders, the laws, the numerical development and diversity of the migrant population and much more. On track: ‘Define your terms and identify assumptions that underpin the question(s)’ (Kelly 2018, p. 84). Once you feel confident that you can settle on a research question, i.e. one you are on track with, be prepared for some serious brainwork as the process of ‘back and forth’ is about to really kick in. Let me quote Kelly’s (2018, pp. 84–85) list of bullet points to illustrate this thought. The back and forth process of defining research questions • • • • • • • • • • • • •

60

Identify possible research topics. Identify possible questions. Consider alternative questions. Break down your proposed research question(s). Define your terms and identity assumptions that underpin the question(s). Decide on a question. Check the literature to see whether the chosen question (in the form it is in) has already been answered. Refine your question, if necessary. Develop a project proposal around your research question. Seek feedback on your draft proposal from supervisors and peers. If necessary, reframe the research question in the light of issues raised by constructing the proposal. Carry out your study. Go back to your question from time to time to check that you are still on the track that you started on.

Chapter 2 How to ask I think it might be helpful to think about the process of developing your research question, not in the form of a list but as a circle or a spiral: most work packages will reappear frequently and ask you to revisit and rethink them. Also, when you respond negatively to the final check of whether or not you are still on track, go back and refine your question, or come to terms with the fact that your research has obviously started to take a different route. While you proceed through this spiral, your research question will naturally change. Sometimes there will be little changes; sometimes the changes will be quite profound. As an anthropologist embarking into the field without a clear-cut question in the first place, you actually hope and strive for the research question to emerge based on the empirics you encounter (see this chapter, Sternsdorff-Cisterna, Ch. 2.1). Also, for researchers in other disciplines, to experience a change in research questions is not only ok, but it is actually a really good thing! It gives proof of your project development. Just make sure your overall research design still corresponds to your questions, and in case adjustments should have become necessary, implement them as soon as you can. A word of caution is in order here: an important part of the art of developing your research question is to know when to break out of that circle! When has the time come to stop revisiting and refining your question? If you feel insecure about making that decision by yourself, talk to your supervisor, other researchers who have more experience than you and to peers or researchers from other disciplines you possibly can relate to. If you still feel insecure, another approach might be to just do it: break out of the cycle, stop wondering and pondering for a while and do some empirical work for a change. This will help you to find out how far your research question will take you, and most likely you will be pleasantly surprised. Key issues Be prepared to rethink your research question multiple times before reaching a solid level of confidence with your questions. As Sternsdorff-Cisterna (this chapter, Ch. 2.1) points out, your research question will change eventually during your project. That is ok!

5.2 Narrowing down your research question: Don’t bite off more than you can chew The process of ‘back and forth’ can easily make you feel overwhelmed or at least cause some significant headaches. Another strategy to deal with this—apart from breaking out of the (increasingly vicious) cycle—is to ‘[b]reak down your proposed research question(s)’ (Kelly 2018, p. 84). While many of our research questions focus on social change (see this chapter, Chiavacci, Ch. 2.3), it is also important to acknowledge that we will probably face limitations in what we can do in one project. We will not be able to explain social change as such, but maybe only specific changes occurring in a certain area at a certain time. It is helpful to bear in mind the picture that Nicolai Bodemer and Azzurra Ruggeri (2012, p. 1439) use. ‘Research today resembles a relay race: We focus on a small part of a larger question and then pass the baton to the next scientist.’ Specifically, we should try to separate our research question into ‘manageable “sub-problems”’ (Browne 2013, p. 103). Not only will this strategy be helpful in conducting the actual research, it also will make it easier for our audience to follow our thoughts. If you find it hard to divide your grand ideas into small packages, try to work graphically: use a simple sheet of pa61

Gabriele Vogt per, a flip chart or a white board, and draw out your idea with all its components (see this chapter, Sternsdorff-Cisterna, Ch. 2.1). You may also add colours, lines and arrows or anything else that highlights important details. Skilfully chosen sub-problems may serve well as the consistent thread running through our work. These sub-problems should be logically connected to the lead question! The master of sub-problems, of course, is James S. Coleman (1990). The Coleman Diagram, also known as Coleman Boat or Coleman Bathtub, suggests that when studying social change on a macrolevel, you should do so by choosing and studying processes on a micro-level that will eventually allow you to make causal connections to the macro-level (Institute for Analytical Sociology 2016; see Jentzsch, Ch. 6.3). So try not to bite off more than you can chew, but rather approach your project in multiple bites at a time. In other words, it is always a good idea to break down your research question into multiple sub-questions. I have found it useful to do so by identifying the actors that are central to my research project. When I first came to Okinawa in the autumn of 1995, the whole island population seemed to be on their feet demonstrating against the national government’s base siting policy. However, I could hardly ask ‘What is going on here?’, even if that probably was the first question that popped into my head. So I tried to disentangle the social reality in front of me by creating manageable sub-problems using an actor-centred approach: Which movements are part of this protest wave, and what exactly do they demand? Who are the movement leaders, what is their background, and why are they stepping up right now? Who are the political leaders in Okinawa, how do they position themselves towards the protest wave, and how are they trying to make use of the island-wide protest in their bargaining with national level politicians? Who are their exact target actors in Tokyo and Washington, and what do they stand for? Next to this kind of actor-centred approach, you could also design sub-problems by focusing, for example, on a certain period of time, on a certain event, on a certain set of sources such as a law revision or testimonies of eye-witnesses to an event. There are multiple ways of cutting out your analytical packages. While doing so, you should bear in mind the size of the project you are setting out to conduct. What is the word count you can fill with your analytical work? Be careful not to bite off more than you can chew. Key issue Narrow down your research question by addressing manageable sub-questions rather than trying to explain the world.

6. How the research question affects your methodological choices: Quantitative, qualitative and mixed method approaches You will need to decide on the methodological approach to your study at the latest once you have a manageable package at hand. Realistically, however, the question of which research methods to apply is one that will linger in the back of your head from the very onset onwards. Frankly, there are research methods that you feel familiar with and that you tend to use (Kelly 2018, p. 82). Or, maybe you deliberately fancy venturing out and trying some new approach62

Chapter 2 How to ask es. Collaboration—as McElwain points out in this chapter (see Ch. 2.2)—can be a fruitful path to choose in this case. Either way, sometimes—Alan Bryman (2007) argues that more often than not—your preference for research methods also impacts on the research question you will eventually settle on. It is important that the approach you choose—quantitative, qualitative or mixed method—is ‘appropriate to the question asked’ (Kelly 2018, p. 82). The different approaches to data collection and data analysis that follow this choice are dealt with in other chapters in this volume (see Ch. 5–14). Let me at this point just sketch out in broad strokes that the research literature generally refers to one core difference between quantitative and qualitative research that has a direct impact on the research question (assuming that you decide on the methodology before the question): While quantitative research is said to follow deductive reasoning, qualitative research is often associated with induction. While quantitative researchers will have identified all the relevant variables before data collection begins (which is a precondition if you aim at doing surveys or experiments), ‘some qualitative researchers may be resistant to setting out a formal question and specific details of how to answer it at the start of a project’ (Kelly 2018, p. 83). Please note that Kelly (2018, p. 83) talks about ‘some’ qualitative researchers. The open approach is particularly common in ethnographic research and is associated with methods like grounded theory (Sebe-Opfermann 2016, pp. 30–32; see Meagher, Ch. 12). Not all qualitative researchers shy away from identifying variables. The disciplinary background in which a study is situated generally determines the research approach, and political scientists and some sociologists who apply qualitative methods would surely deem it necessary to lay out a question and/or a hypothesis, and concrete variables to study and/or test before embarking on data collection. Another alternative to deciding for a quantitative or qualitative approach is to combine elements from both and embark on a mixed method study (see Hommerich/Kottmann, Ch. 10). Through data gathered in interviews with fellow researchers, Bryman (2007, p. 5) argues that the ‘traditional view, whereby mixed-method research is viewed as only appropriate when research questions warrant it’, is gradually pushed into the background. Instead the new view on mixed method research claims that it is superior because it tends ‘to provide better outcomes more or less regardless of the aims of the research’ (Bryman 2007, p. 8). This view is based on the fact that we trust data more if it generates the same outcome via different paths, and thus prevents critics from dismissing it as coincidental. According to Bryman (2007, p. 10), there are, however, more reasons that explain the rise in mixed method studies, beyond what you could argue is a ‘task-driven’ decision. These are ‘to secure funding, to get research published or gain the attention of policymakers’ (Bryman 2007, p. 14). More often than the other way around, it is the qualitative researchers that set out to incorporate quantitative data collection and analysis into their research. This is due to the fact that quantitative data is favoured by funding agencies, journal editors and policymakers as seemingly more reliable data. Bryman (2007, p. 18) argues that ‘a widely held principle of social research—that decisions about research methods and approach are subservient to the research questions that guide them—is questionable as a representation of social research practice.’ Bryman challenges the textbook approach, according to which the research question comes first, and theory, methodology, data, etc. fall into place afterwards. He claims that these days a researcher’s approach to the design of their topic is driven more by pragmatic considerations such as: Where do I get funding for this project? Where can I publish my results? How will I most likely be able to have an impact (academic and/or public) with my research? 63

Gabriele Vogt As a junior scholar reading this, you may feel disillusioned (or you are well aware of the determining power of research funding as you may hold such a position yourself); as a senior scholar, you will know that Bryman has a valid point and maybe feel called upon to rethink the essence of a research question for your next proposal. But, of course, in an ideal world it is the enhancement of knowledge that we strive for when formulating our research questions. Key issue On the path from developing your research question into an overall research design, ideally value a task-driven approach over pragmatic necessities of the contemporary research landscape.

7. Summary This chapter set out to discuss the steps involved in developing a research question, the potential difficulties faced and possible strategies with which to overcome these difficulties. While there are different types of research questions—ranging from descriptive to analytical questions—any good research question will be characterised by three components: relevance, originality and rigour. When you develop your research question, make sure you are prepared for a lengthy process of back and forth between your initial idea, a thorough literature review and debates with peers and teachers. Also, make sure you divide your larger question into manageable sub-questions, as this will make it easier for you to follow through with your research and for your audience to understand your approach. Keep in mind that once you develop your research question into a research project, you will need to settle on the methodological approach, most likely according to your disciplinary background. Developing a research question can be a daunting task for any scholar. Do not despair, and talk to your peers and teachers as much as you can throughout this process. Remember, good research always happens in dialogue.

64

2.1 Your research questions may change and that is ok Nicolas Sternsdorff-Cisterna

My training is in Cultural Anthropology, a discipline that uses ethnographic methods to learn about the culture of the places where we work and the lifeworlds of the people we meet. While ethnographic methods are central to Cultural Anthropology, a number of other disciplines use them as well. When conducting ethnographic fieldwork, I have experienced that research questions often change as the project progresses. It is difficult to predict how fieldwork will unfold and thus it is important to have some flexibility about one’s approach. The social world is not static; circumstances change, new connections emerge, etc. Over the course of a project, events can change the conditions that led to an initial research question and the project will need to be adapted. Many colleagues with whom I have discussed fieldwork have also relayed their experiences of adapting their projects to what happens in the field. Given all the contingent ways in which fieldwork can develop, my suggestion is to remain flexible so that the researcher can respond to the actual experience of fieldwork and make the necessary adjustments to their project. Furthermore, ethnographic fieldwork hinges on the person conducting it. The ethnographer is a vehicle of sorts through which interactions happen and data is gathered. Researchers, however, are not neutral figures. They carry with them identities that include their gender, class, race, age, nationality and others, which shape how they are perceived by the people they work with. As such, even if two ethnographers worked on the same topic, there is a good chance that their positionality and individual interests would lead them in unique directions. These factors shape the body of data they collect and the types of questions they can answer. What an ethnographer envisions in the design phase of a project does not always work out, and thus it is important to have some flexibility to adapt one’s research to the vagaries of fieldwork. I have also found that I often need to revise my research questions as I become more familiar with a topic. What seemed like a good question at the beginning later reveals itself to be too broad or in need of adjustment. There are areas of inquiry that I may not have initially considered but later learned about, and they become central to the project. I usually ask at the end of an interview if there was anything that we did not discuss that the interviewee considers important (see Alexy, Ch. 7.3). Sometimes I have been told that we covered all the bases and I am reassured about my line of questioning. Other times, however, interviewees have pointed out aspects that they see as important that I did not address. These moments have been tremendously useful in broadening my perspectives and have sometimes resulted in revisions to my research questions, so I can incorporate these new angles. In my first major project (Sternsdorff-Cisterna 2019), I shifted the focus of my research in response to events that took place a few months before I planned to begin my main period of fieldwork in Japan (see this chapter, McElwain, Ch. 2.2). I entered graduate school with the 65

Nicolas Sternsdorff-Cisterna intention of studying food safety and quality in Japan. I was particularly interested in the distinction between domestic and imported food products; consumer surveys showed that given a choice, many consumers in Japan preferred domestic products, which they perceived as safer compared to imported alternatives. In the spring of 2011, I was in the United States preparing to begin my fieldwork on this topic. Then, Japan was struck by the triple disaster of March 11, 2011. I followed with alarm the news about the disaster and learned from these reports about the pressing concerns regarding food safety because of the radiation released from the Fukushima Daiichi nuclear power plant. From these reports, I learned that discussions about food safety were shifting and I began to consider how I could adapt my research to incorporate these new developments. A few months later, I left for Japan and, with my committee’s blessing, I decided to use my preliminary research on food safety as background information which I could utilise to focus my attention on food safety in light of the accident at the Fukushima Daiichi nuclear power plant. After I began my fieldwork, I learned about topics that I was not initially familiar with, such as the technical aspects of measuring radiation. Radiation does not have a colour, smell, texture or visual cues that reveal its presence. Rather, detectors are needed to measure how much radioactivity is present in a food sample. During my fieldwork, I met people who ran testing centres and learned from them about the different types of detectors available and how testing procedures can affect the degree of confidence in the results. In a nutshell, the longer a sample is tested, the more fine-grained the results become, and, as such, the length of the test affects the degree of confidence with which it can be said that a product has undetectable levels of radiation. As I learned about how these detectors operate, I realised that their technical specifications played a role in the production of data about the effects of the nuclear accident on food safety. From these insights, I developed new lines of inquiry that allowed me to include a discussion of measurement as part of my overall project. I have thus far stressed the importance of remaining flexible so that the ethnographer can adapt to new circumstances and what they learn during fieldwork (see McLaughlin, Ch. 6). Given the fluid nature of this process, it is important to have moments of reflection in order to evaluate what one has done thus far and where the project is heading. One of my mentors, anthropologist David Slater, introduced me to an exercise halfway through my fieldwork that allowed me to do this. In this exercise, we divided a blackboard in half. On one side, I summarised all the data I had collected thus far and the places I was gathering it. On the other side of the board, I listed the research questions that I wanted to answer by the end of my fieldwork. With everything laid out in front of us, we began matching my data to the questions listed on the other side. During this process, we discussed which questions looked like ones I could already address, which questions I could not yet answer, whether there was data that related to questions I had not listed, and also found connections between topics that I did not notice until everything was diagrammed in front of us. This exercise created a moment in which I was able to take stock of my fieldwork; it also gave me an opportunity to brainstorm how I might change my research strategies to ensure that I collected data with which to address the questions I could not yet answer. Through this exercise, I also began drafting an outline of what my ethnography would eventually look like and how I might organise the various research questions I was working with into an overall narrative. My next suggestion comes from advice I was given when I was preparing applications for graduate school. At the time, I was living in Tokyo and anthropologist Anne Allison was 66

Chapter 2 How to ask teaching a seminar in which we read ethnographies about Japan and discussed our individual research interests. As Allison listened to our ideas, we discussed their merits and whether they were feasible. In addition, she often asked us to answer what seemed like a difficult question: why now? Asking a compelling research question was a first and crucial step, but by asking the ‘why now’ question, Allison prompted us to think about how our projects would speak to the contemporary moment. Different projects will have different answers to this question. For some, the answer may hinge on events of contemporary significance or social issues that merit further attention. In other cases, a project is designed to intervene in theoretical debates and its results will move disciplinary knowledge forward. I have also worked on projects for which it seemed difficult to explain what was unique about the topic at that moment in time. Nevertheless, even when it seemed difficult to answer this question, reflecting on it helped me to think about how to position my project and articulate the ways in which it would contribute to our knowledge about contemporary Japan and/or debates in Anthropology. In addition, I realised that thinking about the timeliness of the project became useful down the line when preparing funding applications. A good project will hopefully be competitive for funding, and it is important in these applications to make a compelling case for the significance and urgency of the research. In my current project, I am examining the role of artificial intelligence (AI) and sensors in the makings of a ‘super-smart society’ in Japan. AI as a field dates back to the 1950s and sensors also have a long history, so when designing the project, I took care to note recent developments that make the contemporary moment unique. In 2016, the Japanese government introduced a vision to move towards ‘Society 5.0’, a super-smart society that relies on AI, the Internet of things, big data and robotics to optimise social processes and create a smart society. In 2020, Japan will also introduce 5G cellular networks, which will dramatically expand bandwidth and the speed of data transmission. As 5G networks become available, it is forecast that many objects and places that are currently not connected to the cloud will become ‘smart’ and transmit digital data; AI will play a crucial role in analysing these increasing quantities of data. When I conceptualised the project, I highlighted these developments to explain its contemporary relevance and why it was necessary at this moment in time to explore the role of AI and sensors in Japan. Lastly, I would like to introduce something that may seem obvious but is worth emphasising, and that is to choose a research subject that one is passionate about. Depending on the scope of the topic, a researcher can spend a substantial amount of time working on it; it can take years to decades to see a project through from conceptualisation to publications. Beyond disciplinary concerns, we have our own sense of curiosity, past experiences, political commitments and numerous other variables that shape our interests. The answer to why we are committed to a topic may not always be immediately clear or can change over time, but I would recommend taking a moment to reflect on what drives our desire to learn more about it.

67

2.2 Studying Japanese political behaviour and institutions Kenneth Mori McElwain

Finding a ‘good’ research topic or puzzle, and more importantly, avoiding a ‘bad’ one, is hard to do on one’s own. We are often drawn to questions and cases that interest us personally, but as a professor told us in graduate school, ‘everybody has the right to study what they want, but nobody has the right to get paid for it’. I often bear this aphorism in mind when advising young scholars: there is a distinction between a hobby and a profession. In Political Science, research agendas are inevitably shaped by real-world events. Over the last 30 years, scholarship on populism, climate change and terrorism has grown rapidly, even as interest in the inner workings of the Soviet era Kremlin has waned. The same holds true for the study of Japanese politics. The end of the Cold War and the rise of China have altered the salience of the U.S.–Japan Alliance or Article 9 to East Asian security. Following the collapse of the economic bubble, research on the supposed merits of the developmental state has given way to understanding the political causes of the Lost Decade. After electoral reform in 1994, many scholars began to work on their consequences for party system change and public policy outcomes, while abandoning further analysis of the old electoral system. In some ways, Japanese politics has become less exceptional since the 1990s, thereby challenging researchers who wish to publish in international journals or university presses to justify their case selection in new ways. In comparative politics, the value of studying a specific country—especially if the researcher is based outside that country—is often defined by its ability to explain events in other countries. This manifests in two ways. First, is the case an ‘exception to the rule’, whereby its internal operations or political outcomes force us to re-evaluate accepted theories? Second, is the case ‘typical’, in that understanding its inner mechanisms have direct implications for explaining empirical outcomes in a wide range of countries? Like most scholars who began their careers outside Japan, my research on political institutions uses Japan as a comparative case to explain broader political patterns. Having been trained in the United States after the Liberal Democratic Party’s (Jiyūminshutō, LDP) first defeat in 1993, ‘single-party dominance’ was no longer a sufficient hook to attract attention from my peers, advisors or hiring committees. However, Japan has a treasure trove of data that is unavailable in other countries, making it a useful case in which to (re)test (un)conventional theories empirically. In the rest of this chapter, I will discuss how research on Japanese political institutions and behaviour has changed over the last two decades, through the lens of my own work on electoral and party politics.

68

Chapter 2 How to ask

Studying electoral politics and LDP single-party dominance A central tenet of democracy is electoral accountability, wherein voters have the opportunity to throw out governments that enact unpopular policies or lack administrative competence. One expectation is that we should observe periodic turnovers in governing parties; exceptions to this rule, such as the LDP’s electoral supremacy between 1955 and 1993, attract interest as outlier cases (Pempel 1990). There are multiple explanations for the LDP’s success, particularly under the 1955 System. First, the LDP’s popularity was buoyed by rapid postwar growth and rising living standards (Miyake et al. 2001). Second, the LDP distributed pork-barrel projects to compensate for slower growth in rural areas, which were also their electoral strongholds (Calder 1988; Ramseyer/Rosenbluth 1993). Third, opposition parties failed to coordinate effectively against the LDP, and were particularly hampered by their inability to attract quality candidates (Scheiner 2006). Fourth, the electoral system was biased in favour of rural incumbents, through a combination of seat malapportionment and restrictive campaign regulations, thereby insulating the LDP from challengers (McElwain 2008). These explanations are not exhaustive, mutually exclusive or equally relevant to all time periods. However, the rarity of single-party dominance drew significant academic interest in the determinants of the LDP’s success and its implications for electoral accountability. After the LDP’s first defeat in the 1993 House of Representatives election, academic attention pivoted towards the effects of electoral reform. In 1994, the lower house system was changed from multi-member districts with single non-transferable votes (MMD-SNTV) to a mixedmember majoritarian (MMM) rule, which combined single-member districts with a proportional representation tier. The early 1990s saw electoral reform in a number of other countries, notably Italy and New Zealand. These offered opportunities to test long-standing theories about the relationship between 1. the electoral system and the number of parties, as well as 2. the effects of new electoral incentives on public policy outcomes. On the first point, the predominance of single-member districts under the new MMM system led to predictions that the fragmented opposition parties would gradually merge into a viable alternative to the LDP (Christensen 1994; Curtis 1999). This transition was messy, as both governing and opposition parties experienced a series of defections by Diet members (Kato 1998; Reed/ Scheiner 2003). On the second point, the elimination of intra-party competition over votes, which was endemic under MMD-SNTV, led to a greater focus on policy-based, programmatic competition over clientelistic redistribution (Catalinac 2016; Noble 2010). Since 2005, election outcomes have hinged on national swings in voter sentiment, rather than district-by-district characteristics, suggesting that elections have become ‘nationalised’ (McElwain 2012). The victory of the Democratic Party of Japan (DPJ) in 2009 seemed to herald a new era of (mostly) two-party competition, marked by greater electoral accountability and government turnover. The DPJ’s loss in 2012 was consistent with this trend, but the manner of its loss, as well as the refragmentation of opposition parties, belied simple expectations of a two-party equilibrium. New smaller parties continued to crop up only to fade away quickly—a still ongoing process. The DPJ itself no longer exists, and many of its former members have splintered into other parties. The fragmentation of opposition parties is important to understanding contemporary Japanese politics. However, it is not clear if the fluidity of the party system is a short- or long-term feature, raising the possibility that any in-depth, contemporaneous account of individual parties or events will quickly become obsolete. More fundamentally, the Japanese political system has 69

Kenneth Mori McElwain become less distinctive from a comparative perspective. The LDP’s first loss in 1993 was due to the perfect storm of political scandals, economic collapse and intra-party defections, but it still remained the largest party in parliament. Its second loss in 2009 was more conventional: it was beaten by the more popular DPJ. The resurgence of the LDP since 2012, as well as the inability of opposition parties to settle on a unified banner, may have systemic roots in the institutional architecture of government, but this may be better studied when (if) the dust settles, rather than following events election-by-election. One possible avenue of comparative research is through the lens of party emergence and survival. There is a robust range of European literature on ‘new parties’, many of which have made significant gains in recent years, buoyed by growing anti-establishment or anti-immigration sentiments. There is evidence that the ideological basis of party politics in Japan has been changing over the last twenty years too. Christian G. Winkler (2017) argues that the LDP has been placing greater emphasis on post-materialist policies, rather than neo-liberal issues, in their election manifestos. The salience of constitutional revision, both in party platforms and in voter decision-making, has increased dramatically in the last decade (McElwain 2018; Sakaiya 2017). These may reflect generational shifts in ideological priorities and perceptions of the left–right dimension, as documented by Willy Jou and Masahisa Endo (2016).

What’s new is what’s old: Public opinion and economic performance One topic in which Japan researchers have renewed their interest is the political causes and consequences of economic performance. Between the 1970s and 1990s, the focus of Social Science research on Japan was the political economy: the determinants of the ‘miracle’ growth of the 1960s–70s, the asset bubble of the 1980s, and policy paralysis in the 1990s. However, as the economy remained mired in the doldrums of the Lost Decade, comparative researchers shifted their emphasis to better-performing countries. This trend has begun to shift in recent years, in large part because of the global financial crisis. What was once seen as a ‘Japanese problem’—prolonged deflation, stagnant growth—started to afflict other advanced economies. While political economy research has historically focused on elite decision-making, there has been a proliferation of work that centres on voters, drawing on the literature on ‘economic voting’. Public support for the government is correlated with macro-economic performance, but governments often need to enact long-term policies that entail short-term pain. This makes it crucial to understand how voters perceive the economy, learn about public policies and attribute changes in their own lives to government actions. In a series of public opinion projects, I have been examining these linkages. I show that people’s macro-economic evaluations are highly sensitive to stock market fluctuations, in large part because these are the most reported economic items in the news (McElwain 2015). Greg Noble and I (2015) find that Japanese voters are more likely to support increases in the consumption tax if these are explicitly tied to funding social insurance programmes. Finally, Junko Kato, Tomoko Matsumoto and I (2018) explore attitudes towards government budgets and find that older men with higher educational attainment are more likely to worry about government debt, suggesting important sub-population differences in economic beliefs. There are three reasons why Japan is an ideal case to test the relationship between the economy and public opinion. First, Japan has been a front-runner in many (pessimistic) trends, such 70

Chapter 2 How to ask as prolonged deflation and ballooning government debt. These link to other issues that indirectly affect the economy, such as the labour market consequences of an ageing population and reticence towards increasing immigration. Second, there is a wealth of public opinion data in Japan that allows us to test existing theories and develop new frameworks. For example, Jiji Press, a newswire service, has been running monthly surveys on attitudes towards the economy and politics since the late 1960s without ever changing the question wording. Third, Japan has an excellent infrastructure for running original surveys. While costs vary by vendor, survey length and sampling strategy, it is possible to run an original 2000-person survey for JPY 60,000–100,000. For many overseas researchers, this will be possible to budget in research grants, and is probably cheaper than a round-trip flight to Tokyo (see Hommerich/Kottmann, Ch. 10).

Some concluding thoughts When conducting research on Japan, there are two broad approaches: using new knowledge from Japan to inform theories of comparative politics, and applying comparative theories to explain Japanese phenomena. Most scholars try to do both, but as I argued, Japan is no longer an obvious outlier among developed democracies on conventional political topics. Exceptions exist, such as the consequences of and policy responses to demographic ageing, but it is incumbent upon the researcher to justify one’s case selection. One way to do so is to identify ways in which Japan continues to be an empirical rarity that challenges conventional theory. Another is to use original or rare data from Japan that allows us to test theories in a more refined way. My interest in Japanese public opinion, particularly on perceptions of the government’s macroeconomic competence, comes from my belief that both approaches can be married successfully. Let me end by offering some practical suggestions about conducting new research on Japanese politics from my own experience. First, the best way to learn whether you have a good topic is to present your work in front of different audiences. In my case, learning to give different types of talks to political scientists, economists and Japan specialists has pushed me to better explain the Japanese case and also broadened my comparative scope. Each group has different types of expertise and interests (e.g. quantitative versus qualitative analysis, economic policy versus public opinion), which can inform avenues for fruitful research, gaps between theory and empirics, and ways to integrate Japanese and comparative research. Second, collaboration can be crucial for innovative research. Survey work is extremely difficult to do on one’s own, and I co-author with two graduate students at the University of Tokyo for both survey design and empirical analysis. Dissertations obviously require original, individual work, but for scholars who have an opportunity to do fieldwork in Japan, attending conferences and workshops can be valuable for identifying other like-minded scholars with whom you may eventually want to collaborate.

71

2.3 Capturing social change in Japan David Chiavacci

The research question is the alpha and omega of research and academic writing. It is, in my view, even more important than theories or methods. Throughout my academic career, I most often ended up with meaningful research results when I had a good research question to start with. In these cases, I let the research question flourish and develop, and tried not to lose sight of or to forget it along the way. It is, foremost, the research question that will allow you to generate findings that are new and innovative contributions to your field. Ideally, the research question defines which theoretical approaches and methods should be used and not the other way around. However, how do you find a good research question? Actually, I have to admit that, during over two decades of doing Social Science research on Japan, I never gave this question too much thought. I could even claim that I do not know if I found my research questions or if they came to me like uninvited but highly welcome guests. Still, the door has to be open for an uninvited guest to materialise. My main advice for ‘opening the door’ is to follow your interests: read up on those research topics that excite you and about which you would like to know more. And if you happen to come across a research gap or even a puzzle, then seize your opportunity and start to frame and reframe a research question, but don’t forget to think about which data or material is needed to answer this—ideally—open question.

Research question first: From the question to the project When I review my research projects and questions, social change can easily be identified as the common and general issue that played an important role in most of them. In my experience, focusing on social change is a quite productive way to identify a research question, because a first step is to ask why change occurs. However, this does not mean that everything will go smoothly if you try to capture the meaning of social change. In fact, over the years, I encountered different challenges in developing research questions on issues of social change depending on the project. In my PhD project, I analysed the sudden surge in popularity of Western companies as employers of graduates from prestigious universities in Japan. In the final year of my master’s studies, together with a colleague I was able to secure financial support from the Swiss Asia Foundation for an empirical survey on the perception of Swiss companies as employers in Japan and the job satisfaction of their Japanese employees (Chiavacci/Lottanti 1999). As part of this research project, we found several quantitative surveys whose data showed a surprisingly sudden and strong increase in popularity of Western companies among graduates of top-ranked Japanese universities. Up to the mid-1990s, difficulties and often the complete failure in recruiting university graduates was, together with the high costs of doing business, one of the two by far most important barriers to Western companies successfully making direct invest-

72

Chapter 2 How to ask ments and entering the market in Japan. However, in the late 1990s, almost from one day to the next, Western companies became attractive employers and many secured top positions in the rankings of the most sought-after employers of university graduates, even including those from the most prestigious institutions. In my PhD project, I discussed this new popularity as an indicator of macro-sociological change from the perspective of the continuing economic stagnation and increasing discourse of crisis under the buzzword ‘Lost Decade’ (ushinawareta jūnen) in Japan. In this case, the research question was already clearly defined at the very beginning of the research project. In fact, the new popularity of foreign companies as employers of elite university graduates was one of the main findings of the earlier research project on Swiss companies, for which we had, however, no explanation. The research question in my PhD project asked how this sudden increase in the attractiveness of foreign companies could be explained in the context of a new discourse of crisis and national stagnation, a changing transition process from university to labour market and the new life course ideals of the generation of university graduates in Japan in the late 1990s. As the research design was clearly defined from the very beginning of the project, the research question did not change at all during the research. Moreover, I was fortunate that I had already been able to gain ample experience in qualitative interview methods in the preceding project about Swiss companies in Japan and was able to successfully secure access to interview partners. This resulted in the smooth and fast realisation of the whole PhD project in about two years and its publication the following year (Chiavacci 2002).

Struggling with research questions Developing good research questions in the two later projects was much more complex and time-consuming. When I came to Japan as a postdoc in the early 21st century, I soon noted an increasing number of academic publications about rising inequality and a public debate on this issue gaining momentum. My interest in social change and in social structures led me to embark on this topic immediately. However, empirical data and research were puzzling for me in this case. In public and academic debates, a new model of Japan as an increasingly unequal society became dominant up to 2005. It became condensed into a new discursive frame for Japan as a gap society (kakusa shakai) and displaced the former frame of Japan as a general middle-class society (sōchūryū shakai). Still, from a comparative perspective, the discursive juncture was not at all supported by empirical data. Quantitative indicators and research results showed an increase in social inequality, but widening social and economic gaps were very moderate in international comparison and by no means justified the complete transformation of Japan’s self-perception from a general middle-class society to a gap society. This contradiction between strong structural continuities and complete frame reversal inspired me to move away from my original research questions, which focused on explaining the context and main factors for increasing inequality. My new research question was how the gap between the dominance of a completely new frame on a discursive level versus a very moderate change according to structural data could be explained. This shift led me to explore the Sociology of Knowledge as a completely new and unplanned theoretical field. My focus was no longer on questions of structural or discursive change, but on the interrelationship between so-

73

David Chiavacci cial structures and discursive frames as central pillars of social reality. My research in this area resulted in publications on the question of why and how certain macro-sociological models reflect everyday experiences and become dominant frames in academia and society. The main finding was that these processes and the sequence of dominant social frames were not a question of scientific truth of models based on evidence, but a question of their persuasiveness in the micro-sociological lifeworld of the population and especially among those groups shaping public debates and discourses (Chiavacci 2008). While the research took years, I can also say that these unsolved puzzles helped me to turn a rather boring research question into a much more interesting research question. The excitement of this topic has not diminished for me, and I have continued to do research on it (Chiavacci/Hommerich 2017).

Asking questions about non-change My main postdoctoral project was even more complex. After finishing my PhD, I urgently needed a new research project to be able to apply for research funding. Immigration and the need for comprehensive immigration policy reform in view of Japan’s demographic trajectory started to become an increasingly important issue in the mass media and on the political agenda. The emergence of a more active and less restrictive immigration policy seemed to be only a question of time. Hence, I had the idea of studying the forthcoming immigration policy reform and writing my second book about it. However, the more I worked on immigration policies, the more I noted a huge gap between an intensive and heated debate in public discourse and among policymakers versus a standstill and reform bottleneck in immigration policy. As the years passed by without any significant reform despite heated ongoing discussions, my frustration grew. Obviously, I had completely underestimated the risks of real-time research. Its degree of timeliness is excellent, and opportunities to collect good qualitative data on the policy process are very good, but the outcome is completely unknown and insecure. At some point, I even considered writing my second book about the research question of why no comprehensive immigration policy reform was happening in Japan. Hans Geser, one of my former professors in Sociology at the University of Zurich, had formulated a theory about non-action and refraining (Geser 1986). However, Geser pointed out that writing a whole book about a non-occurrence is very unusual in the Social Sciences. Finally, I realised that I had to expand my research question to solve my problem of political standstill by covering not only Japan’s immigration reform in the 2000s, but its immigration policy since 1945 in the context of developments in the East Asian region. While I had originally planned to analyse the large, upcoming immigration policy reform and its social consequences, my new research question focused on how immigration and immigration policy in Japan had developed and changed in a larger East Asian context and against the backdrop of the national institutional setting in immigration policy. In this book, I discussed the transformation of Japan from a non-immigration country to an immigration country in the late 1980s. I also compared the immigration policymaking processes that led to significant reforms around 1990, but only very limited reforms in the 2000s. This research design enabled me to comparatively analyse the question of why some reforms stall, while other are formulated and implemented very fast (Chiavacci 2011). This whole project took me eight years, much longer

74

Chapter 2 How to ask than originally envisaged, but made me realise the importance of developing an alternative plan if the original research question reaches a dead end.

Concluding remarks Overall, my experience with research questions is that sometimes you are very lucky—like in the case of my PhD project—but normally it is a cumbersome process with many twists and turns, as in the two other projects discussed above, until you finally have your research question. As I said above, you are desperately looking for good research questions, but normally they seem to suddenly (and often after a long time) present themselves to you. This long process can be very stressful, especially for young researchers as they work under time constraints and know only too well that their academic career depends on their speed and ability to produce output in the form of publications. Still, this quest is not only a part, but a fundamental piece of the whole research process. The most important thing along the ride is not to lose track of your research question and to constantly try to improve it. Never forget to be and remain the master of your research question instead of letting it pull you in a one-way direction!

75

Further reading Bryman, Alan (2007): The research question in social research: What is its role? In: International Journal of Social Research Methodology 10, No. 1, pp. 5–20. Coleman, James S. (1990): The foundations of social theory. Cambridge, MA: The Belknap Press of Harvard University Press. Kelly, Moira (2018): Research questions and proposals. In: Seale, Clive (ed.): Researching society and culture. Los Angeles, CA: Sage, pp. 79–99. Leung, Doris/Lapum, Jennifer (2005): A poetical journey: The evolution of a research question. In: International Journal of Qualitative Methods 4, No. 3, pp. 64–82. Mattick, Karen/Johnston, Jenny/de la Croix, Anne (2018): How to … write a good research question. In: The Clinical Teacher 15, pp. 104–108.

References Bodemer, Nicolai/Ruggeri, Azzurra (2012): Finding a good research question, in theory. In: Science 335, p. 1439. Browne, Jacinta E. (2013): Getting started with research ‘Beginning: Defining a research question and preparing a research plan’. In: Ultrasound 21, pp. 102–104. Bryman, Alan (2007): The research question in social research: What is its role? In: International Journal of Social Research Methodology 10, No. 1, pp. 5–20. Calder, Kent E. (1988): Crisis and compensation: Public policy and political stability in Japan. Princeton, NJ: Princeton University Press. Catalinac, Amy (2016): Electoral reform and national security in Japan: From pork to policy. New York, NY: Cambridge University Press. Chiavacci, David (2002): Der Boom der ausländischen Unternehmen als Arbeitgeber: Paradigmawechsel in Japan? München: Iudicium. Chiavacci, David (2008): From class struggle to general middle-class society to divided society: Societal models of inequality in postwar Japan. In: Social Science Japan Journal 11, No. 1, pp. 5–27. Chiavacci, David (2011): Japans neue Immigrationspolitik: Ostasiatisches Umfeld, ideelle Diversität und institutionelle Fragmentierung. Wiesbaden: VS Verlag für Sozialwissenschaften. Chiavacci, David/Hommerich, Carola (eds.) (2017): Social inequality in post-growth Japan: Transformation during economic and demographic stagnation. London: Routledge. Chiavacci, David/Lottanti, Stefania (1999): The Japanese employees of Swiss corporations in Japan: Image and reality. SAF Research Report. Lausanne: SAF (Swiss Asia Foundation). Christensen, Ray (1994): Electoral reform in Japan: How it was enacted and changes it may bring. In: Asian Survey 34, No. 7, pp. 589–605. Coleman, James S. (1990): The foundations of social theory. Cambridge, MA: The Belknap Press of Harvard University Press. Curtis, Gerald L. (1999): The logic of Japanese politics: Leaders, institutions, and the limits of change. New York, NY: Columbia University Press. Gerth, H. H./Mills, C. Wright (1948): From Max Weber: Essays in Sociology. New York, NY: Routledge. Geser, Hans (1986): Elemente zu einer soziologischen Theorie des Unterlassens. In: Kölner Zeitschrift für Soziologie und Sozialpsychologie 38, No. 4, pp. 643–669. Hollifield, James (2000): The politics of international migration: How can we bring the state back in? In: Brettell, Caroline B./Hollifield, James (eds.): Migration theory: Talking across disciplines. New York, NY: Routledge, pp. 137–185. Institute for Analytical Sociology (2016): The Coleman Boat explained [film]. https://www.youtube.com/w atch?v=dGaz0xKG060, [Accessed 15 April 2020]. Jou, Willy/Endo, Masahisa (2016): Generational gap in Japanese politics: A longitudinal study of political attitudes and behaviour. London: Palgrave Macmillan. Kato, Junko (1998): When the party breaks up: Exit and voice among Japanese legislators. In: American Political Science Review 92, No. 4, pp. 857–870. Kelly, Moira (2018): Research questions and proposals. In: Seale, Clive (ed.): Researching society and culture. Los Angeles, CA: Sage, pp. 79–99.

76

Chapter 2 How to ask Leung, Doris/Lapum, Jennifer (2005): A poetical journey: The evolution of a research question. In: International Journal of Qualitative Methods 4, No. 3, pp. 64–82. Matsumoto, Tomoko/McElwain, Kenneth Mori/Kato, Junko (2018): Why do government deficits prevail? A survey experiment on budget making. Presented at the 2018 American Political Science Association Annual Meeting, Boston, MA. Mattick, Karen/Johnston, Jenny/de la Croix, Anne (2018): How to … write a good research question. In: The Clinical Teacher 15, pp. 104–108. McElwain, Kenneth Mori (2008): Manipulating electoral rules to manufacture single party dominance. In: American Journal of Political Science 52, No. 1, pp. 32–47. McElwain, Kenneth Mori (2012): The nationalization of Japanese elections. In: Journal of East Asian Studies 12, No. 3, pp. 323–350. McElwain, Kenneth Mori (2015): Kabuka ka kakusa ka: Naikaku shijiritsu no kyakkanteki/shukanteki keizai yōin. In: Leviathan 57, pp. 72–95. McElwain, Kenneth Mori (2018): Constitutional revision in the 2017 election. In: Pekkanen, Robert J./ Reed, Steven R./Scheiner, Ethan/Smith, Daniel M. (eds.): Japan decides 2017: The Japanese General Election. London: Palgrave Macmillan, pp. 297–312. McElwain, Kenneth Mori/Noble, Gregory W. (2015): Who tolerates tax increases? Age and gender in the raising of Japanese consumption taxes. In: Shakai Kagaku Kenkyū 67, No. 2, pp. 75–96. Miyake, Ichiro/Nishizawa, Yoshitaka/Kōno, Masaru (2001): 55-nen taisei-ka no seiji to keizai: Jiji yoron chōsa dēta no bunseki. Tōkyō: Bokutakusha. Noble, Gregory W. (2010): The decline of particularism in Japanese politics. In: Journal of East Asian Studies 10, pp. 239–273. Onwuegbuzie, Anthony J./Frels, Rebecca (2016): Seven steps to a comprehensive literature review: A multimodal & cultural approach. Los Angeles, CA: Sage. Pempel, T.J. (1990): Uncommon democracies: The one-party dominant regimes. Ithaca, NY: Cornell University Press. Ramseyer, J. Mark/McCall Rosenbluth, Frances (1993): Japan’s political marketplace. Cambridge, MA: Harvard University Press. Reed, Steven R./Scheiner, Ethan (2003): Electoral incentives and policy preferences: Mixed motives behind party defections in Japan. In: British Journal of Political Science 33, pp. 469–490. Sakaiya, Shiro (2017): Kenpō to yoron: Sengo nihonjin wa kenpō to dō mukiatte kita no ka. Tōkyō: Chikuma Shobō. Scheiner, Ethan (2006): Democracy without competition in Japan: Opposition failure in a one-party dominant state. Cambridge: Cambridge University Press. Sebe-Opfermann, Andreas (2016): Die Frage der Fragen: Was ist eine gute Forschungsfrage? In: Dunker, Nina/Joyce-Finnern, Nina-Kathrin/Koppel, Ilka (eds.): Wege durch den Forschungsdschungel. Wiesbaden: Springer Fachmedien, pp. 21–36. Sternsdorff-Cisterna, Nicolas (2019): Food safety after Fukushima: Scientific citizenship and the politics of risk. Honolulu, HI: University of Hawai‘i Press. Vogt, Gabriele (2003): Die Renaissance der Friedensbewegung in Okinawa: Innen- und außenpolitische Dimensionen 1995–2000. München: Iudicium. Vogt, Gabriele (2018): Population aging and international health-caregiver migration to Japan. Cham: Springer. Winkler, Christian G. (2017): Right on? The LDP’s drift to the right and the persistence of particularism. In: Social Science Japan Journal 20, No. 2, pp. 203–224.

77

Chapter 3 How to organise research: Research designs Kaori Okano

1. Introduction Having decided on a research topic, researchers begin considering what research design best addresses their topic. By research design, I mean the whole process of investigation from refining the research questions based on a review of the literature, to writing up the findings in the form of a paper or a thesis. We design research so that it can most effectively address our research questions. A good research design functions like a map that provides researchers with orientation when they are aiming to collect the evidence that is needed to answer their research question or to test a theory in a convincing way (De Vaus 2001, p. 9). In this chapter, I will focus on case study research design, as it is most broadly conceived. While different disciplines (e.g. Anthropology, Sociology, Political Science, etc.) and paradigms (e.g. positivist, interpretive and constructionist) may have slightly different ideas about case study research, there are generally agreed features. All research projects on Japan, which are intended to be published in a language other than Japanese, are ‘case studies’ in the widest sense of the term, in that they aim not only to understand Japanese society for its own sake, but also to advance our understanding of the social world at a more abstract level. Any study of Japan is inherently comparative in that researchers bring their own perception lens created through past experiences, when observing and interpreting the phenomena under study. Japan could be studied, for example, as an example of a non-Western liberal democracy with particular institutional features and cultural norms. This view reflects the reality of the Anglophone or Eurocentric dominance in global knowledge production, which views non-Western societies as peripheral societies with specific conditions for case studies, where researchers collect raw data or test a theory developed in the West (Okano 2018). In view of this, Yoshio Sugimoto proposes a cosmopolitan methodology which presumes that all Social Sciences are ‘Area Studies’ (including studies of Anglophone societies) without privileging studies based on societies at the centre (Sugimoto 2018). This chapter aims to address first-time or emerging researchers with little experience who are about to embark on a project. You may be an undergraduate honours student or a postgraduate student with Japanese language proficiency adequate for studying Japan (of course, your Japanese will improve as your project progresses). I write this chapter as a senpai with a particular positionality, which may help you understand what follows. Born in Hiroshima, my entire schooling was in mainstream schools in Japan until I completed an undergraduate degree. While I was an undergraduate student, I spent a year as an exchange student in Auckland, New Zealand. I subsequently studied comparative education in Australia for a master’s degree,

78

Chapter 3 How to organise research and then worked as a full-time teacher (and a participant observer) at secondary schools in Sydney and in New Zealand. Inspired by micro-level experience of schooling as a teacher in a predominantly anglo-white environment, I subsequently researched working-class high schoolers in urban Kobe for a PhD from a New Zealand university. Writing in a foreign language was a challenge. Since then I have been a university-based academic in Melbourne, Australia. In this chapter, I will first provide an overview of different approaches in case study research designs and their features, advantages and disadvantages; I will then discuss ways to select a unit of analysis (single case or multiple cases both synchronic and diachronic). There are five types of multiple case study designs: 1. multiple cases within Japan, 2. cross-national multiple cases, 3. a time-series (or diachronic) design, 4. time-series analysis of multiple cases within Japan (which provide both diachronic and synchronic analysis) and 5. time-series cross-national multiple cases. Time-series design can take the form of a wave study or a panel study. Finally, I will discuss theory building and testing, and inductive and deductive processes.

2. Case studies A case study is an in-depth empirical investigation of a case or multiple cases bounded by time and activity. These cases can be institutions, phenomena, events, processes, individuals, groups of people, organisations, activities and programmes (Hancock/Algozzine 2006, pp. 16–29). Case studies are useful when studying a phenomenon as a whole in its natural setting. Most case studies use multiple sources of data, including any combination of documents (both public and private), interviews, observations (direct and participant), archival records, physical items (e.g. CDs, pictures) (Yin 2018, p. 111), as well as surveys (see Hommerich/Kottmann, Ch. 10). Surveys can be officially designed and conducted by government agencies or companies, or by the researchers themselves. Most case studies include both quantitative and qualitative data components for different parts of a project, even when they claim to be qualitative studies (Gerring 2007, p. 11). In Japanese Studies, it is rare to see studies which comprise only narratives. For example, a study of non-regular workers in a company would first locate the place of non-regular workers in the national labour market (e.g. income levels, gender) and the company in the national context (e.g. size of companies), using quantitative data available in published statistics. The researcher may deliver a short questionnaire to employees of the company as a starting point to assist in refining research questions and formulating interview questions. Marie Roesgaard’s study of cram schools (juku) (2006), for example, included extensive quantitative data from surveys and national statistics about these institutions and their students, as well as qualitative data from interviews and observations. Key issues A case study is an in-depth investigation of phenomena in their natural setting. Case studies often use multiple sources of data, both quantitative and qualitative, at different stages of the investigation.

79

Kaori Okano

3. Different approaches of case studies While John W. Creswell and Cheryl Poth (2017, pp. 65–109) name case studies as one of five research approaches to inquiry, along with narrative (biographical) research, phenomenological research, grounded theory research and ethnographic research, I see case studies as a generic term inclusive of all these approaches. Let me explain these one by one. Studies adopting a narrative and biographical approach investigate a person or persons (or other forms of units of analysis) over a relatively long period of time, drawing on oral life histories and other forms of narrative. For example, Ruth A. Keyso (2000) examined the life histories of eight women residing in Okinawa in order to understand their life trajectories and the nature of changes in their community. David Plath’s study of maturity (1980) examined the life paths of several individuals in the Kansai region over the postwar period, drawing on oral histories, archives and other documents, as well as novels. A biographical study of a prominent individual would also take this approach. Phenomenology is often adopted by studies that have a psychological or micro-sociological orientation. This approach illustrates the shared meaning of a phenomenon that all individuals of a group commonly experience in interaction, and attempts to grasp the meaning and essential nature of that interaction (Creswell/Poth 2017, p. 75), for example, between pupils and teachers, elderly and carer workers, doctors and patients, and so on. The study aims to identify and explain the essential nature of that interaction, drawing on interviews with these individuals. For example, when Nobuo Shimahara and Akira Sakai (1995) investigated fledgling teachers in a case study of primary schools, they found that all teachers considered teacher– pupil bonding, mutual trust and empathetic relationships as the most crucial elements in learning to teach at Japanese primary schools. Shimahara and Sakai called this phenomenon kizuna. The grounded theory approach, originally developed by micro-sociologists who studied the interaction between dying patients and their relatives (Glaser/Straus 1965), is most effective when little has been studied about the process in question. It aims to seek an explanation for the process and generate a theory that is ‘grounded’ in data. Researchers collect vast amounts of data through observation and interviews, devise categories and sub-categories in that data, identify patterns and try to explain them (Creswell/Poth 2017, pp. 82–90), which eventually leads to the formulation of a tentative hypothesis. Advantages of this approach are that it is likely to lead to original understandings (since little has been known) and grant a larger scope of ‘freedom’ to the researcher. But this can be a challenge to inexperienced researchers in that they are not guided by existing theories that inspire hypotheses they can build on to the same extent as the other approaches (see Meagher, Ch. 12). An ethnographic approach captures the unit of analysis as a whole in the most natural setting. It also encompasses a natural process of knowing, like someone entering a new organisation and getting to know the people and the place by observing, talking to people and becoming familiar with the immediate surroundings (see Tagsold/Ullmann, Ch. 8). It often involves longterm immersion in, and observation of, the organisation or people, and events and programmes as they occur in a natural setting. There are many advantages. A researcher can gather data from a wide range of sources and, since it involves a relatively long period of time, can revise questions as the study progresses. It also allows the researcher to test emerging interpre80

Chapter 3 How to organise research tations (tentative hypothesis) by asking relevant questions, seeking rival theories and/or other sources of evidence while in the field. The main challenge is the length of time required, not only for conducting the fieldwork but also for negotiating access to the fieldwork site. If the study includes institutions as a unit of study, it requires the institution’s approval (see Reiher/ Wagner, Ch. 16). If a study includes minors (under 18 years old), the ethics approval process will pose more challenges, since consent needs to be obtained from their parents or guardians. This is particularly true for ethnographies that examine schooling processes: preschools (Lewis 1995; Peak 1991), primary schools (Bjork 2016; Cave 2007; Sato 2003; Tsuneyoshi 2001), middle schools (Bjork 2016; Cave 2016; LeTendre 2000), senior high schools (Aspinall 2012; Okano 1993) and minority students in schools (Bondy 2015; Gordon 2008; Roth 2002; Ryang 1997). Instead of using an ethnographic approach that includes several methods, researchers can choose only one of these individual methods like interviews, observations, document analysis (including content analysis), policy analysis or discourse analysis (see Kottmann/Reiher, Ch. 7; Tagsold/Ullmann, Ch. 8; Arrington, Ch. 13; Eder-Ramsauer/Reiher, Ch. 14), separately, or in any combination, when the research question is more specific. These methods of data collection require significantly less time commitment, since they are more or less ‘one off’, and the time spent on data collection is shorter. The data collected will usually be more focused and specific than for an ethnographic study. Tasks for researchers • • •

Consider why the case study method would be more effective for your project than other methods. Consider what approach best suits your research question (narrative/biographical, phenomenology, grounded theory, ethnographic). What aspects of your project may be a challenge for each of the above approaches if you select one?

4. Selecting cases: Analysis of a single case or multiple cases A research project can involve a single case or multiple cases. Researchers make this decision based on their research questions and practical considerations of time and resources. Some researchers select a case (or cases) because of typical and representative features that it presents (e.g. salarymen, married women with irregular employment), so that they can claim that the findings shed light on wider Japanese society. Others choose a case because of its unique and distinct features (e.g. motorbike gangs). Some choose cases with similar features: middle-income schools in suburbs (Cave 2016) and working-class urban schools with minority populations (Okano 1993). Others select cases with distinctive features like Thomas Rohlen’s study of five schools in Kobe (an elite private school, a high-ranking government academic school, a medium-ranking government academic school, a vocational high school and an evening high

81

Kaori Okano school).1 It is also possible to examine ‘all cases’ in order to cover the ‘universe of cases’, like Kay Shimizu’s study (see this chapter, Ch. 3.3). It examines all the local chapters of the Japan Agricultural Cooperative (JA) in order to understand the institutional changes that occurred in the nationwide institution. In a single case study, researchers examine one organisation, individual, phenomenon or event. Emma Dalton (2017) chose one political party, the LDP, in order to study Japanese political parties’ positions (both overt and covert) regarding women in politics. This is a relatively contained process, which may be a good starting point for a novice researcher. In multiple case studies, researchers select two or more typical cases (e.g. two urban primary schools) in order to study representative and mainstream features; or two or more schools with distinct but similar features (e.g. urban disadvantaged schools) in order to study working-class reproduction through schooling. On the other hand, a researcher may want to select two cases with contrastive features, for example, one working-class school and an elite private school, in order to illuminate differences in class-specific socialisation and reproduction. Another researcher may choose one girls’ working-class school and one boys’ working-class school in order to include gender in its analysis, with the social class factor constant. Existing studies of ‘working women’ in Japan illuminate how researchers have bounded their units of study. Studies have examined a group of clerical workers and a group of factory workers to illuminate differences (Lo 1990), factory workers (Roberson 1998; Roberts 1994) and tertiary educated white-collar ‘office ladies’ (Ogasawara 1998). The choice depends on the research questions. Analysis of multiple cases takes five different forms: 1. multiple cases of the same institution (or any unit of analysis) within Japan, 2. cross-national multiple cases (Japan and other societies), 3. a time-series design which examines a single case diachronically at different points in time, 4. a set of multiple cases within Japan across different points in time and 5. a set of cross-national multiple cases at different points in time. A project becomes more complicated both in design and implementation as it goes from 1 to 5. I will provide some examples in the following. We see the first type of multiple case design in Rohlen’s (1983; see above) study, which examined five different types of schools in Kobe, Japan. On the other hand, I chose two municipal vocational high schools (technical and commerce) with some minority students, since my interest focused on how final year high schoolers decided on and obtained postschool employment under the school-based job referral system, and the impact of class and minority backgrounds in this process (Okano 1993). An example of the second type (crossnational analysis of multiple cases in Japan and in other societies) is Lynne Nakano’s study (see this chapter, Ch. 3.1), which compares single women in Tokyo, Hong Kong and Shanghai. There are several cross-national analyses of this kind about schooling. Gerald LeTendre (2000) presents an ethnographic study of two middle schools in Japan and two in the U.S., in examining how middle schools interpret adolescence and assist their adolescent students. Shimahara and Sakai (1995; see above) compared how fledgling teachers learn to teach at primary schools in Tokyo and the East Coast of the U.S. The third type investigates the same case (or unit of analysis) at different points in time, which is often called a time-series design (Tight 2017, pp. 109–110). The time-series design considers each time period as a case, and illuminates the chronological sequence of events, continuities and changes over that period. A di-

1 For comparative case studies and most similar or most different case study designs, see also Alexander George and George L. Bennett (2005).

82

Chapter 3 How to organise research achronic approach is useful in a study of changes resulting from a new policy or intervention. When the time span covers years, then we can call it a longitudinal study. The fourth type of multiple case design is diachronic analysis of multiple cases within Japan at different points in time, presenting both synchronic and diachronic analysis of cases. This could take the form of a panel study which examines the same people over time, or a wave study which examines similar cases over time. For example, my own ongoing research, the Kobe women’s longitudinal panel study, takes the former approach. It examines a group of the same working-class girls (multiple cases) from 1989 (when they were in their final year of high school) to the present, examining their identities, and the accumulation of advantages and disadvantages over their life courses in light of structural changes during the Heisei period (Okano 2009). If the study had taken a wave study approach, the project would have studied the final year students of the same high schools (or similar high schools) in Kobe at different points of time, rather than following the same individuals over 30 years. The project’s aim would have been to identify continuities and changes in how final year high schoolers make decisions about their post-school destinations, and how the schools guide this process and with what outcomes. On the other hand, Cave’s study (2016; see above) can be seen as a wave study, in that it studied the same two middle schools 10 years apart. The fifth type is diachronic cross-national analysis of multiple cases, again presenting both synchronic and diachronic comparative analysis. Like the fourth type above, this can take the form of a panel study or a wave study. Preschools in three cultures revisited (Tobin et al. 2009) adopts a wave study design in examining changes and continuities in observed schooling practice in the same preschools in China, Japan and the U.S., 17 years apart in 1984 (Tobin et al. 1991) and 2002. Nakano’s project on single women in Tokyo, Hong Kong and Shanghai (see this chapter, Ch. 3.1) has the potential to take on a panel study design, if it were to follow the same women in the future.

5. The Kobe women’s panel study: An evolving project I must confess that I did not originally design the Kobe women’s project as a diachronic analysis of multiple cases when it began. The project began as an ethnographic study of two working-class schools in the 1989–1990 academic year (the first year of Heisei), examined how final year high schoolers decided on their post-school destinations under the institution of school-based job referral, and identified the roles of class, minority status and gender in that process. I interviewed 100 students during that time. When I interviewed some of the same students two years later (when they were employed) in order to uncover the extent of the school-based job referral system’s effectiveness, I was intrigued by their descriptions of their emerging adult lives (including relationships and childbirths). It occurred to me that an interesting project would be to follow their process of attaining adulthood as a life course study. It has since then evolved beyond this into a 30-year project with the 22 participants now in their late 40s. There has been further development in the Kobe women’s project into another discipline. I vaguely felt that these women’s speech patterns and styles had changed over the years, and casually mentioned this to my colleagues working in sociolinguistics. They then became 83

Kaori Okano interested in studying language variations (both synchronic and diachronic) in the discourse of my interviews with these women over 30 years. The sociolinguistic part of the Kobe women’s project now involves a team of five linguists and me, and is funded by the Australian Research Council (Maree/Okano 2018). A novice researcher’s postgraduate project has unexpectedly evolved into something larger, and has involved serendipity and redrawing the boundaries of the unit of analysis (see this Chapter, Ch. 3.2). To my surprise, my own speech in the interviews was also closely analysed by my co-researchers! This (for me unexpected) widening of the scope of my research leads to another issue of research design—bounding the unit of analysis or a case. Tasks for researchers • • •

Consider if your research question is best addressed by studying a single case or multiple cases in light of the advantages and disadvantages. In view of your research questions, consider what specific features your case must have, and what features you can compromise in selecting cases. If you adopt multiple cases, is it better to have cases with similar features or contrastive features?

6. Bounding cases and units of analysis Researchers need to limit the boundaries of a unit of analysis in order to make the project manageable. While we as researchers are curious about many things, we cannot study all that interests us and need to limit the focus of our research and the boundaries of a chosen case in order to complete the project within the required time and resources. I would advise novice researchers to follow Pamela Baxter and Susan Jack’s approach (2008, p. 546) and confine their cases by considering these categories: time and place, time and activity, definition (of the case) and the context. For example, the case can be limited by time and place (e.g. the third year of a particular newly introduced type of six-year secondary school), time and activity (e.g. volunteer activities at the time of the Hanshin-Awaji earthquake), and by definition (of the case) and context (e.g. the elderly’s use of day-care services, in the context of other available care options, in an urban centre). The boundaries of a case can alter in the course of the research process, when researchers encounter relevant aspects that they had never considered. While it is wise to define clear boundaries for a case (or unit of analysis) at the outset when deliberating on data collection methods, I would also suggest being sufficiently open-minded to adjust the boundary at later stages of the project, a point that Jamie Coates (see this chapter, Ch. 3.2) would concur with.

84

Chapter 3 How to organise research

7. Knowledge production: The theory building and theory testing continuum How does case study research contribute to theory building and theory testing, and ultimately knowledge production? In theory building, the research begins with an empirical observation of a phenomenon, and develops a theory by making sense of the empirical study through inductive reasoning. In contrast, in theory testing, the research begins with a theory, which will guide empirical observation in the research process; and eventually assesses the value of the theory through deductive reasoning (De Vaus 2001, pp. 5–6). Critics of the case study method have long commented that case studies cannot contribute much to theory building and testing (Maoz 2002) because of their limited scope of generalisation, non-replicability, ‘subjective’ conclusions or too many variables examined in too few cases (Gerring 2007, pp. 7–8). These debates, however, depend on what you mean by a theory. I see a theory as a tentative explanation. A theory may include identification of causal relationships, a sequence of events or simply a particular interpretation informed by analysis of evidence. The scope of tentative explanation (here in my understanding as theories) differs significantly: at what I would call a lower and immediate level (also: micro-level), a theory can be a tentative explanation of a specific event in a specific context, that is, an informed interpretation based on evidence of that particular case (e.g. class size has little impact on student learning in a Japanese primary school since learning centres on interaction amongst students there). On the other hand, at a higher abstract level (also: macro-level), a theory can be well established and deemed widely or universally applicable, such as the theory of gravity in physics. Somewhere between the two on the continuum are medium level theories (also: middle-range theories), that is, theories about phenomena which are often based on multiple units of study (e.g. school class size generally promotes self-directed learning). Therefore, theoretical discussion can occur at different levels. There are variations in how (far) studies want to contribute to theory building beyond an immediate specific level. Case studies undertaken to understand a specific society are called ‘intrinsic’, while those with the aim of contributing to general theory building are ‘instrumental’ (Stake, 1995, pp. 3–5). However, I do not see being either intrinsic or instrumental as mutually exclusive but complementary. Researchers can build theories both from single case and multiple case studies. A single case study (within-case analysis) collects data, and identifies patterns, sequences and themes in an event (the unit of analysis). In a multi-case study, researchers identify cross-case patterns and sequences, and develop a thematic analysis from within their own case study or from other published case studies. In so doing, they develop a tentative explanation, again look at another case and modify a tentative explanation to refine their theory, and the process continues. Novice researchers are likely to aim to advance theories at specific and lower levels, and possibly medium levels, rather than aiming to build grand theories. Theory building and theory testing are closely related, although it is often considered that qualitative studies contribute to the former and quantitative studies to the latter. However, case studies usually involve both inductive and deductive processes. A simple theory testing case study can investigate the extent to which an existing medium level theory is applicable in a specific local and institutional context, and where the theory might fail to provide an explanation. In so doing, the study can modify or refine that medium level theory. A more ambi85

Kaori Okano tious theory testing case study may examine a more established higher level theory (usually) developed in the West to see to what extent it is applicable in Japan, with its specific institutional and cultural features. Japan can be a useful setting for testing what is considered an established theory that has been developed elsewhere (Árnason 2002; Sugimoto 2014, pp. 24– 28). We can understand a case study’s relationship with theories as a continuum between deductive and inductive processes, drawing on and modifying classifications (Dooley 2002; Welch et al. 2011). One end of the continuum is the most deductive process, whereby a study attempts to test an existing theory for its applicability in the specific context of the chosen case. At the other end is the most inductive approach, whereby research begins with a large amount of empirical data (rather than an existing theory), which is coded and organised into categories and themes to identify patterns and sequences if any (see Meagher, Ch. 12). From these, researchers develop a tentative explanation concerning the phenomenon being studied in the particular case. A grounded theory approach especially strives to advance a theory of an unstudied theme from information from participants. Most qualitative studies are found somewhere in between the two poles, often involving both deductive and inductive processes at different phases of the study (Creswell/Creswell 2018, pp.56 – 58; pp.63 – 64). A study may begin with a tentative hypothesis that the researcher has developed from existing research literature. Even when researching an unstudied area using a grounded theory approach, a study is (at least to a certain degree) likely to be guided by some prior, if limited, understanding of the area and/or a theoretical perspective. The research process is thus both inductive in that case studies generate and build theories, and deductive in that existing theories are applied to a unit of analysis in the specific case. Let’s finally look at these processes in Nakano’s study of single women in three countries as an example (see this Chapter, Ch. 3.1). The study presents tentative explanations about single women in Tokyo, Hong Kong and Shanghai (at the immediate and specific level). By comparing the findings of these three cases, the study identifies patterns across three cases (cross-case patterns) of single women, and builds a tentative hypothesis as to how certain features of local laws, institutional practices and cultural norms may have influenced their lives in different ways (at the medium level). In the future, Nakano may wish to discuss these findings in the context of published case studies of single women in other societies, identify any cross-case patterns and develop theories at a more abstract level. At the same time, other researchers interested in single women, for example in Germany, may be interested in examining Nakano’s cases as published case studies in shaping their own theories. The value of case studies is that they provide empirical examination of a phenomenon, event and people in the natural setting of a specific local context with particular sets of institutions and cultural norms. Studies thus produced and published can be used as published cases where other researchers attempt to build or refine tentative hypotheses at a higher level. Therefore, I suggest that every case study contributes to tentative theory building in the long term, and eventually to knowledge production. Key task Think about the potential contribution of your study to theory testing or/and theory building.

86

Chapter 3 How to organise research

8. Summary The major value of case studies is their capacity to study diverse social phenomena empirically in the natural setting of their specific local context. A single case study (within-case analysis) identifies patterns, sequences and themes within the unit of analysis. In a multi-case study, researchers identify cross-case patterns and sequences. I consider all research on Japan to be case studies in the widest sense of the term, because Japan research never only aims at understanding Japanese society for its own sake, but also to advance our understanding of the social world at a more abstract level. Researchers studying Japan can advance, build or test theories based on their own empirical studies. This process involves studying other published case studies while other researchers may also be examining your case studies while modifying and refining their tentative hypotheses and theories. Therefore, every case study (not only) about Japan contributes to theory building in the long term, and ultimately to knowledge production. Knowing this gives us a sense of fulfilment in doing research. I encourage you to explore the opportunities offered by case study design for your project. In my research career, case study research has been both insightful and enjoyable. You never know where it might lead!

87

3.1 Developing a comparative study: Single women in Hong Kong, Tokyo and Shanghai Lynne Y. Nakano

Soon after arriving in Hong Kong in the mid-1990s, I enrolled in Cantonese language courses at the Chinese language school operated by the university where I worked. I took night courses with other faculty members, and learned that the daytime programme enrolled many women from Japan who were in their late twenties to early thirties and unmarried. They had quit their jobs in Japan to study the Cantonese language full-time in Hong Kong. Their actions contradicted the literature on Japan at that time, which stated that most women were very concerned about ‘marriage deadlines’ and basically married on time, that is, by their early to mid-thirties. What were these women doing in Hong Kong? Weren’t they worried about missing so-called ‘marriage deadlines’? If not, were social values changing or was something happening in Japanese social institutions such as workplaces or the employment market? These early questions led me to study the rise of singlehood in Japan. As I explored the topic, I discovered that women in Hong Kong married even later than women in Japan. In the late 1990s, women in Hong Kong were marrying at the age of about 29 and women in Japan were marrying around 28. Given the similar statistics, it seemed reasonable to construct a study that compared the two societies. I wanted to find out whether parallel changes in attitudes or social structures were occurring in the two societies or whether there were differences, and, if so, what was the nature of these differences. In my reading, I learned that the age of marriage is rising across Asia and the world, and the reasons for later marriages are complex, involving women’s increased levels of education and employment opportunity, and wealthier and smaller families. Rather than investigating statistical factors, I wanted to understand women’s views on marriage, how women’s choices reflected changing social values, and how women’s experiences were shaped by family, workplaces and friendships. To address these questions, I decided to interview women who had never married in the two cities on a broad range of topics, including their views and experiences of work, family, marriage, friendships and views of the future. When possible, I also conducted participant observation by joining single women in their everyday activities such as eating out, shopping, going to concerts, trips and other leisure activities. I met them wherever was convenient for them: in their homes, offices, coffee shops and restaurants. In structuring the study, I decided to compare Hong Kong and Tokyo rather than Japan as a whole. It made sense to me to focus on cities because singlehood is rising fastest in cities, cities attract single women and cities are home to more single women than rural areas. Tokyo has the largest percentages of single women in the country and the latest ages of first marriages. The two cities, Hong Kong and Tokyo, host consumer cultures that cater to single women such as restaurants, bars and leisure activities. The two cities are global cities (Sassen 2001)

88

Chapter 3 How to organise research that share similar structural features such as a high concentration of educational institutions, service industries and the availability of employment opportunities for women. After I started interviewing women, I began giving talks on my findings. At these talks, I was often told by members of the audience that my findings described the situation for single women in their home city as well. I was encouraged by audience members to add a third or even fourth city to my study. I received suggestions that I should conduct my study in Singapore, Taipei, Seoul, Bangkok and New Delhi. I agreed that the study would be more interesting with a third or even fourth city included for comparison. After considering my language abilities and time limitations, I decided to include Shanghai as the third city. I thought that a city in China would pose an interesting contrast to Hong Kong. People living in cities in mainland China would have very different understandings of marriage and a different recent history of marriage and family relations. I chose Shanghai because women marry later in Shanghai than in any other city in mainland China. I found it difficult to locate my study in a theoretical framework because there were many theories that might explain the data. I was not sure which body of literature I should address. One body of literature explains global shifts in marriage and family relationships. For example, the ‘second demographic transition’ argument developed by Ron Lesthaeghe (2014) identifies global patterns that describe some of the changes occurring in the societies in my study. Anthony Giddens’s (1991; 1992) theories of changing relationships in the modern period predicted greater freedom and equality in romantic relationships, but not all of his observations applied to the societies I was studying. A second body of literature explored singlehood from feminist perspectives, examining how singlehood is an inferior category that is necessary to create the superior category of married people (Borneman 1996; Lahad 2017). A third body of literature examines the rise of singlehood in East Asian societies. Eventually, all three kinds of studies helped me to shape my argument. The literature on global family changes helped me to identify commonalities in singlehood in the societies in my study, and see how they differed from Western societies. One key difference, for example, is that in Western societies singlehood is understood as a product of the decline of extended families. In the societies in my study, however, singlehood occurs in the context of strong family relationships. For a fascinating study on changes in family relationships over time in China, I recommend Yunxiang Yan’s study (2003) that explores this topic over a fifty-year period in a village in northeastern China. Feminist studies, such as those by Angela McRobbie (2004), helped me to see that single women are supposedly free to choose their own paths in life, but these paths are actually highly circumscribed. Studies of singlehood in the societies under study helped me to understand how the organisation of family resources in the three societies created different experiences of singlehood for women in the three cities. When I sat down to write about the data, I looked to studies that had taken a comparative approach for ideas about how to develop my analysis. I drew on the work of Ching Kwan Lee (1998), who compared women workers in a factory in Hong Kong and another across the border in Shenzhen, China. Building on what Michael Burawoy called ‘an extended case method’ (1991a), she asked why these societies with a similar cultural history produced very different work regimes. Don Kulick and Jens Rydstrom (2015) adopted a similar approach when they asked why two countries with similar cultural histories and welfare systems, Denmark and Sweden, had two very different ways of dealing with the sexuality and the desire for intimacy

89

Lynne Y. Nakano of individuals with severe disabilities. Similarly, I tried to explain why three cities with similar cultural backgrounds would produce three different marriage regimes. In the end, I was glad that I had the opportunity to conduct a comparative study (Nakano 2016a; 2016b). It allowed me to take a macro-perspective, which is often difficult for anthropological studies. The comparative approach prevents myopic conclusions. Scholars and the mass media, for example, frequently comment on the nature of single women in their society, but their arguments do not hold up when we compare singlehood across the region and the world. Comparative studies, particularly within Asia, prevent simple conclusions about Japan or China versus Western societies. But there were serious drawbacks to the comparative approach. The biggest drawback was that it took an enormous amount of time. To make the most out of the data, I wanted to understand the three societies well. As a Japan specialist, I was familiar with the literature on Japan but less familiar with the literature on Hong Kong and China. It took me a year to read the literature on China. A third major challenge was my low level of spoken Mandarin. I already spoke Cantonese and Japanese at the start of the study, but my Mandarin was rudimentary and required an investment of time to make it serviceable. As I proceeded with the project, another issue emerged that I could not fully address. That is, I struggled with how I could discuss three societies without flattening the great diversity within each society. The handling of diversity within a society can be partially addressed by research design; one can narrow the focus of one’s study. In my case, I focused on never-married women who were near or beyond marriage ‘deadlines’, that is, women from their late twenties through to their early forties. But I did not narrow my study to a certain socio-economic class, although this may have been a better strategy in retrospect because it would have made the project more manageable. Excellent studies that explore women’s experiences in reference to socio-economic class include Kaori Okano’s study (2009) on working class women’s transition to adulthood in Japan and Jesook Song’s study (2014) on working class women’s struggles to find appropriate housing in South Korea. As I only interviewed around 35 women in each city, I could not come to far-reaching conclusions about different kinds of class-based experiences. But I could convey some of the diversity of the three societies when I explained my findings, and I suggested to the reader how I thought different walks of life and perspectives shaped my findings. For researchers who would like to try a comparative approach, I suggest starting with societies that one knows well, including the relevant research literature. I would also recommend narrowing the focus of the study so that the comparison is manageable in terms of the time it will take to conduct the study. I would then set out to consider the ways in which the societies under comparison are similar as well as different, and look for surprises (Burawoy 1991b). The strategy of explaining surprises may lead to interesting research questions and directions.

90

3.2 Contained serendipity as fieldwork in Japan: Studying Chinese people in Japan Jamie Coates

My research has focused on how new and emergent communities in China and Japan identify with one another. In particular, I have focused on the ways young Chinese people living, studying, working and travelling in Japan develop a sense of ‘being-in-common’ among themselves and with the various contexts they found themselves in in Japan. This research is part of a wider, loosely defined conceptual interest in how people imagine their place in the world, and more specifically, how young Chinese encounters with places like Japan afford new kinds of cosmopolitan imagination. In stating that these issues are the focus of my research today, I should note that it has taken a long time, a lot of mistakes and several poorly designed projects to come to my current understanding of my work. Each step of my research has relied heavily on what we might call ‘contained serendipity’, where I set certain parameters around a research project, while still allowing myself considerable freedom to discover ‘unsought findings’ (Van Andel 1994, p. 631). I was classically trained as an anthropologist. For readers who might not be familiar with the discipline of Anthropology, it has both scientific and humanities-oriented sub-disciplines, but most sociocultural anthropologists, such as myself, rely heavily on long-term fieldwork to develop new case studies and concepts about what it means to be a person. The research I would like to focus on in this short essay is comprised of the conceptual project I mention above, and two periods of ethnographic fieldwork I conducted to try and address this broader conceptual direction. The first project was for my PhD, which was about how Chinese student-workers form new communities in Japan. While studying in China in the early 2000s, I was struck by the antiJapanese sentiment I came across in daily life. Yet, when I went to visit a friend studying in Japan in 2004, I soon discovered a large number of Chinese students in Japan whose perceptions of their place in the world had been transformed by their experience overseas. In the early stages of my PhD in 2008, I proposed to conduct fieldwork among a cohort of Chinese students in Japan, hoping to find further examples of the people I had met previously. I had originally designed this project as an investigation into how experiences of mobility, study and work affected young Chinese people’s sense of belonging in Japan. Trained in Chinese and Anthropology, I did not speak Japanese very well at the time, so I proposed to firstly enrol in a Japanese language class, where I could improve my Japanese and potentially meet more Chinese students as my sampling method. My goal was to eventually live with a group of students to better understand the daily pressures they faced, while also collecting interviews from my classmates and peers. This approach is an urban modification of the standard form of ethnographic fieldwork design in Anthropology, where you find a group of people to live with and observe how they go about their lives for extended periods of time. 91

Jamie Coates Typically, you supplement this approach with interviews, social mapping and surveys depending on your research goal. I was also interested in how experiences of transnational mobility (e.g. being a migrant or international student) contrasted with experiences of mobility in the city (e.g. walking around in Tokyo), so I also proposed to use a variety of methods by which I would follow participants across the city. My approach took inspiration from the growing emphasis on ‘multi-sited ethnography’ as the best way to understand migration and globalisation (Marcus 1995) as well as the ‘mobility paradigm’ at the time by which movement was seen as foundational to social and cultural processes (Sheller/Urry 2006). When I started my fieldwork using this approach the first few months found me following a few generous individuals through their daily routines in Tokyo. Yet, I soon came to realise that, while I was learning a lot about this small number of people, I was struggling to develop a wider framework for how their different experiences related to each other. Moreover, the sample of people I was working with was very small and not growing. Through these first few months of following a few individuals however, I came to learn that in 2008 a group of business owners near a major station in Tokyo (Ikebukuro) had proposed to have the northwest corner of the station renamed ‘Tokyo Chinatown’. So, I started focusing on this area, setting clearer boundaries around my proposed field site. I became friends with the owner of a small hairdressing salon and decided to try observing daily life among young Chinese people from this one strategic location. My methodological ‘idleness’ (Coates 2017) allowed me to see the comings and goings of a larger number of young Chinese people, and eventually I was invited to other small spaces in the area where people lived, worked, studied and played, including a small privately owned dormitory where I lived with a group of young Chinese people. By taking up the opportunity to focus on this one area, my cohort shifted from focusing solely on students to a mixture of students and entrepreneurs in the area, although most of the people I conducted research with had previously been students. Making strategic decisions about when to move or not within a smaller fixed location allowed me to cultivate a form of serendipity (Hendry 2017), learning a lot about the unexpected aspects of people’s lives, and making close observations of the details of how people fostered friendships, maintained ties with their families and developed a nascent but somewhat conflicted sense of community. Altogether I conducted two years of fieldwork for this project. The material I collected and the people I met not only served as data for my final PhD dissertation on how young Chinese people foster a sense of ‘being-in-common’, but it also led to side projects on their favourite celebrities and a growing interest in the media they consumed. The second project built on these initial approaches and experiences, while also aiming to address some of the frustrations I faced in trying to write about everything I had observed. While writing about my first period of fieldwork research, I felt that much of what I had observed was difficult to capture in field notes and ethnographic prose. In particular, I became increasingly interested in how I might capture the intense point at which the media, emotions and play intersect in the tiny spaces of conviviality that I had observed in my previous fieldwork. Feeling that my words were not enough, I took some intensive filmmaking training over the summer before starting a postdoctoral project at Waseda University in 2014. In this research project, I proposed to look at how young Chinese people’s everyday interactions with various kinds of media might reflect their changing identities in the Sino–Japanese context. My approach focused on conducting filmed interviews with people in their living spaces, asking them to speak about the various items in their homes. I would then follow this up with an 92

Chapter 3 How to organise research interview about their media habits, filming over their shoulder as they showed me various items and content. My intention was to use their living spaces as a way to define the parameters of my research, while also allowing for a very broad definition of what constituted ‘the media’ in their lives. I took inspiration from Daniel Miller’s The comfort of things (2008), where he used a similar approach among people in London to develop an argument about the importance of material culture in everyday life. My intention was to generate themes about their media consumption, rather than using pre-defined categories typical in Media Studies, such as forms of technology or platforms. My goal was to be more rigorous than in my previous study, using more structured interviews as the people guided me through whichever social media and streaming services they felt comfortable sharing with me. While filming these interviews, I also asked if I could participate in and document the online spaces they mentioned in their interviews. These spaces ended up mostly being group-based interactions on platforms such as Weibo, WeChat and to a lesser extent Instagram and Facebook. While doing this research, I was invited to film and document a range of other spaces, social events and interactions. These filmed observations blurred the lines between many of the categories I had developed from my interviews, but they also allowed me to develop contextualising insights. Further, through these invitations I met a network of young musicians and artists who were eager to share their story with me and help me produce a small ethnographic film. This opportunity saw me combine some of my PhD project themes about play, friendship and community with my interest in the media, which resulted in a 50-minute ethnographic film titled Tokyo pengyou (Coates 2018). The process of making the film provided feedback on the themes I was developing in my other work too. As I filmed and edited the work, I would show it to my participants and discuss the direction and themes of my broader research project. Fieldwork across disciplines relies on encounters with people, things and environments to produce different kinds of knowledge. As Isabelle Rivoal and Noel B. Salazar (2013) note, serendipity alongside reflexivity and openness are often quoted as the defining characteristics of anthropological fieldwork. Many classical stories of excellent anthropological findings depended on chance encounters, unsought observations and submitting oneself to the rhythms of the lives of those you hope to understand. This approach embodies Anthropology’s tendency to be the discipline of being undisciplined, which although romantic, can also prove difficult to justify in institutional settings or across disciplines. Yet, this approach provided me with a wealth of data that fed into observations about young Chinese and Japanese lives which might have been otherwise inaccessible. In the classical field sites of Anthropology, such as the Trobriand Islands in Papua New Guinea (Malinowski 2013), the excesses of this approach were often curtailed through the limits of scale. The populations they worked with were smaller and typically covered shorter geographic distances. In urban Japan, it is difficult to find contexts that naturally lend themselves to this kind of research. As Matei Candea (2007) has argued though, the importance of setting limits to one’s field site, while recognising that you yourself set those limits as an ‘arbitrary location’, are incredibly important. I would characterise this kind of research design as a way of fostering ‘contained serendipity’. So how might we build contained serendipity into fieldwork research design in urban contexts such as Tokyo? My advice is to start from a clearly defined, but relatively small field site. Recognising the somewhat ‘arbitrary’ nature of this location, I would then combine practices of both mobile and immobile participant observation to find differing sites of ‘contained 93

Jamie Coates serendipity’. This approach to research corresponds well with urban Japanese contexts such as Tokyo, Osaka or Kyoto, where life takes place in a series of interconnected spaces and sites that each have their own unique qualities. It also complements other modes of data collection, from visual ethnography, to structured interviews and perhaps event quantitative methods.

94

3.3 The universe of cases: Agricultural cooperatives in Japan as a case study Kay Shimizu

This essay draws on research concerning institutional changes to agriculture cooperatives in Japan to address the issue of research design in Political Science in general and in Japan in particular. The examples come from the book Cultivating institutional change: Economic liberalization, demographic decline, and the reform of Japan agricultural cooperatives, which I coauthored with Patricia Maclachlan and is forthcoming from Cornell University Press. I will discuss our research design, its development, characteristics and levels of analysis, and address its drawbacks and (possible) solutions to them.

Research project: Studying institutional change in Japan through the lens of agricultural cooperatives While some research tests existing theories in new settings to better understand their explanatory power or discover their limits, our research aims to build on existing theory—in this case, theories of institutional change—by examining institutional change in one country, Japan. Unlike previous studies which have focused on the national level (i.e. studies of democracy) or those that have examined individuals within an organisation, this study focuses on a relatively homogenous set of organisations, that is, Japanese agricultural cooperatives known collectively as JA. JA has a national organisation, prefectural level offices and local (municipal level) branches. Our study mainly focuses on the latter. In the Japanese setting, agricultural cooperatives (co-ops) are organisations that strive to serve their members through a broad range of services including the provision of agricultural inputs and the marketing of agricultural outputs, along with a myriad of peripheral services such as banking and insurance. Historically, these co-ops have also served as vote-gathering machines for politicians who support policies favourable to agriculture (George Mulgan 2000). The study examines how these more than 600 organisations, which are spread nationwide, have adapted to the changing environment surrounding the Japanese agricultural industry and pushed for institutional change. We find that while some co-ops have embraced market competition and the pursuit of greater profits for their member farmers, other co-ops have failed to adapt to changing consumer demands or are struggling with ageing and depopulation. This comparative research design within a single country allows us to hold many factors constant, especially national level features that may be of concern in cross-border comparative studies, such as regime type or trade openness. Beyond its structural characteristics where the co-ops are of equal rank in the nationally hierarchical organisation of JA, the research subject of agriculture in Japan proved attractive for

95

Kay Shimizu several reasons. First, the politics of agriculture is a growing concern in many parts of the world, both in developed and developing countries. How agriculture is practised and regulated directly influences many of the United Nations’ Sustainable Development Goals including the protection of life on land and the goal of zero hunger. Second, agriculture in Japan has been in crisis mode for over two decades. Japan’s ageing and decreasing population and its evolving dietary habits have dramatically changed food consumption patterns, greatly decreasing overall demand, especially for the country’s staple food, rice. Consequently, agricultural production has had to adjust to these changing demands, but has seen limited success. Third, the global market for agricultural goods has become more open and fluid, exposing Japan to increasing imports and the pressure to export. The evolution of Japanese agricultural cooperatives, and how they are reacting to the challenges facing agriculture in Japan, makes a compelling area for research. One of the most intriguing aspects of Japan’s agricultural system and the norms and institutions that govern its practices is that they were last overhauled during the postwar years in the shadow of the U.S. occupation and still assume the family farm (nōka) to be the only legitimate producer of agricultural goods. As such, corporate participation in agricultural production remains extremely limited. At the same time, both the national and local governments play an oversized role in the governance of agriculture and its institutions. JA lies at the centre of this web of institutions that govern agriculture in Japan and its evolution reflects many of the ongoing changes in agriculture and the context in which it operates. Thus, our research project concerns JA itself, how it has evolved and what it might say about institutional change more broadly.

Research design One of the first things to consider in research design is the unit of analysis—at what level should we study the issue at hand? JA as an organisation lends itself to a multilevel design due to its hierarchical nature. As mentioned above, JA has a national organisation, prefectural level offices and local (municipal level) chapters. As such, we are able to study its organisational change at all three levels. Furthermore, nearly every rural and peri-urban area with agricultural activity falls under JA rule. These local chapters each preside over several municipalities. To the extent that quantitative data on JA activity, for example the yen amount of agricultural products sent to market, is available, we wanted to examine all the local chapters in our analysis. Thus, we tried to cover the universe of cases, and thereby eliminated the need for random sampling or attempting its approximation. One caveat to this approach is that complete data for the universe of cases may not always be available. Nearly all of the data necessary for this study can be found on the websites of individual JA chapters or in the statistical yearbooks of the Ministry of Agriculture, Forestry and Fisheries (MAFF). However, some JA chapters do not make their annual data available online. In this case, a simple phone call can sometimes give you access to the missing data via postal mail. Occasionally, smaller organisations, in this case very small JAs, will not have collected the necessary data or will not want to share their data in order to hide perceived deficiencies. These smaller co-ops may need to be dropped even though this means acknowledging the bias that is introduced into the analysis when this occurs.

96

Chapter 3 How to organise research Another important consideration is the comparability of individual local cases. In our research, the very nature of agriculture and agricultural markets in Japan had to be carefully considered. First, differences in climate and topography make each region suitable for a wide range of agricultural products, and no two are the same. In Japan, the biggest difference lies in those areas dedicated to rice farming using rice paddies and those areas that are more focused on fruit and vegetables, some of which use greenhouses. Rice is in many respects the most important agricultural product in Japan, but it is also losing its influence both in volume and price. Second, recent mergers of JAs have decreased the number of JA chapters and increased the size of individual JA chapters, but not in a uniform manner. As such, the variation in geographic size of JA chapters has grown. In the most extreme case, four prefectures have just one JA chapter for the entire prefecture. In the end, the study retained the analysis of the universe of cases, supplemented by paired qualitative case studies based on more rigorous criteria for comparability. In short, studying the universe of cases has its share of drawbacks and is not always necessary or desirable. In our study, we complemented quantitative data from all local JA chapters with an indepth qualitative study of a handful of cases (see Hommerich/Kottmann, Ch. 10 on mixed methods). Understanding how to measure the extent to which JA chapters have embraced pro-farmer activities required extensive background research and interviews. We interviewed a wide range of people involved in multiple aspects of farming including farmers, JA officials, wholesalers, local government officials, scholars, national level politicians and MAFF bureaucrats (see Yamaguchi, Ch. 7.2). Retirees provided some of the best sources of information. These interviews also provided the detailed information used in our in-depth case studies. We selected our cases based on several criteria that were important in building our hypotheses but were not easily measured equally across the entire population of JAs. For example, one case study was based in a rice growing area, while another was based in a vegetable and fruit growing area. In each instance, we sought our ideal locations, but also planned for backup locations. Incidents (like sudden bankruptcies in our case) do occur, so it is always good to have a contingency plan!

Problems and potential solutions Two issues emerged from the research design of this project. First, we encountered the vexing problem of missing data. This research was designed to examine the entire universe of cases so missing data, such as the lack of financial data from some of the smaller agricultural co-ops, was bound to emerge, but several characteristics of our research topic made locating the missing data particularly challenging. In this case, the missing data tended to be from the more remote and smaller co-ops with few dedicated staff members. As such, they did not have some of the data we sought, even though their national organisations required them. Some co-ops were also reluctant to share their data, as JAs have come under increasing scrutiny in recent years. In such cases, an affiliation with a local organisation (such as a school or library) or an introduction from the local government was able to open some doors. Finally, in more rural areas of Japan, where elderly farmers kept the books, using the telephone worked much better than email. Of course, there are multiple ways to compute missing data using statistical methods, but we went to the original source.

97

Kay Shimizu Second, the selection of case studies was, as is often the case, limited by our access to our chosen subjects. Ideally, cases are selected according to strict criteria of where the cases lie along a spectrum. However, when working in rural, more conservative areas, establishing a relationship with an agricultural community was not always straightforward. One way to overcome such problems is to ask for introductions early in the research and to cultivate relationships over time. We visited some of our cases multiple times over several years. Diet members can also provide valuable connections and introductions to their home prefectures. Bureaucrats (in this case from MAFF) in local offices can also be helpful.

General advice As scholars studying Japan, it is easy to lose sight of the rest of the world. Our work focuses on telling stories about Japan, rather than a more generalisable phenomenon. This greatly limits our audience and prevents us from participating in broader discussions about political or social behaviour. When designing our research, we can move beyond Japan by placing our questions into broader categories that are not country or region-specific. The research design does not necessarily need to be comparative across countries, but it can be helpful to keep potential comparisons in mind. In our case, comparisons included co-ops in other industries such as fishing as well as agricultural co-ops from other countries. Additionally, Japan occupies an ambiguous space in Political Science scholarship that can shed a different light on studies which are often conducted in more ‘typical’ regions such as Western Europe or the Asian continent. For example, it is a democracy long dominated by a single party (see McElwain, Ch. 2.2). It is also a market economy in a centralised, regulatory state. These features of Japan can be used to test the limits and explanatory power of existing theories that have been established elsewhere. Lastly, deep familiarity with a country or region allows for more creative and precise research designs. For political scientists, a highly centralised system like Japan generates rich sources of data at multiple levels of governance. Digitisation of data is lacking, especially in areas beyond Tokyo, but new technologies such as optical character recognition (OCR) and machine learning can help to overcome this hurdle by quickly turning printed documents into digital data that can be machine analysed. Official affiliation with a local institution, such as a school or research centre, can also facilitate greater access to information. Sitting at the frontlines of many important social phenomena such as ageing and post-industrial development, Japan is rich in sources for Social Science research. Findings from Japan can serve as a strong foundation for building theories which can then be tested beyond its borders.

98

Further reading Baxter, Pamela/Jack, Susan (2008): Qualitative case study methodology: Study design and implementation for novice researchers. In: The Qualitative Report 13, No. 4, pp. 544–559. Creswell, John W./Poth, Cheryl (2017): Qualitative inquiry and research design: Choosing among five approaches. London: Sage. De Vaus, David A. (2001): Research design in social research. London: Sage. George, Alexander/Bennett, George L. (2005): Case studies and theory development in the Social Sciences. Cambridge, MA: MIT Press. Gerring, John (2007): Case study research: Principles and practices. New York, NY: Cambridge University Press. Stake, Robert E. (1995): The art of case study research. Thousand Oaks, CA: Sage. Tight, Malcolm (2017): Understanding case study research: Small-scale research with meaning. Los Angeles, CA: Sage. Yin, Robert K. (2018): Case study research and applications: Design and methods. London: Sage.

References Árnason, Jóhann (2002): The peripheral centre: Essays on Japanese history and civilization. Melbourne: Trans Pacific Press. Aspinall, Robert W. (2012): International education policy in Japan in an age of globalisation and risk. Leiden: Brill. Baxter, Pamela/Jack, Susan (2008): Qualitative case study methodology: Study design and implementation for novice researchers. In: The Qualitative Report 13, No. 4, pp. 544–559. Bjork, Christopher (2016): High-stakes schooling: What we can learn from Japan’s experiences with testing, accountability and educational reform. Chicago, IL: Chicago University Press. Bondy, Christopher (2015): Voice, silence, and self: Negotiations of buraku identity in contemporary Japan. Cambridge, MA: Harvard University Asia Centre. Borneman, John (1996): Until death do us part: Marriage/death in anthropological discourse. In: American Ethnologist 23, No. 2, pp. 215–238. Burawoy, Michael (1991a): The extended case method. In: Burawoy, Michael (ed.): Ethnography unbound: Power and resistance in the modern metropolis. Berkeley, CA: University of California Press, pp. 271–287. Burawoy, Michael (1991b): Reconstructing social theories. In: Burawoy, Michael (ed.): Ethnography unbound: Power and resistance in the modern metropolis. Berkeley, CA: University of California Press, pp. 8–27. Candea, Matei (2007): Arbitrary locations: In defence of the bounded field-site. In: Journal of the Royal Anthropological Institute 13, No. 1, pp. 167–184. Cave, Peter (2007): Primary school in Japan: Self, individuality and learning in elementary education. London: Routledge. Cave, Peter (2016): Schooling selves: Autonomy, interdependence, and reform in Japanese junior high education. Chicago, IL: Chicago University Press. Coates, Jamie (2017): Idleness as method. In: Elliott, Alice/Norum, Roger/Salazar, Noel B. (eds.): Methodologies of mobility: Ethnography and experiment. New York, NY: Berghahn, pp. 109–128. Coates, Jamie (2018): Tokyo pengyou. In: Journal of Anthropological Films 2, No. 1 e1538. DOI: https:// doi.org/10.15845/jaf.v2i1.1538. Creswell, John W./Creswell, J. David (2018): Research design: Qualitative, quantitative, and mixed methods approaches. New York, NY: Sage. Creswell, John W./Poth, Cheryl (2017): Qualitative inquiry and research design: Choosing among five approaches. London: Sage. Dalton, Emma (2017): Women and politics in contemporary Japan. London: Routledge. De Vaus, David A. (2001): Research design in social research. London: Sage. Dooley, Larry (2002): Case study research and theory building. In: Advances in Developing Human Resources 4, No. 3, pp. 335–354. George, Alexander/Bennett, George L. (2005): Case studies and theory development in the Social Sciences. Cambridge, MA: MIT Press.

99

References George Mulgan, Aurelia (2000): The politics of agriculture in Japan. London: Routledge. Gerring, John (2007): Case study research: Principles and practices. New York, NY: Cambridge University Press. Giddens, Anthony (1991): Modernity and self-identity: Self and society in the late modern age. Cambridge: Polity Press. Giddens, Anthony (1992): Transformation of intimacy: Sexuality, love and eroticism in modern societies. Cambridge: Polity Press. Glaser, Barney/Strauss, Anselm (1965): Awareness of dying. London: Weidenfeld and Nicolson. Gordon, June A. (2008): Japan’s outcaste Youth: Education for liberation. Boulder, CO: Paradigm Publishers. Hancock, Dawson/Algozzine, Bob (2006): Doing case study research: A practical guide for beginning researchers. New York, NY: Teachers College Press. Hendry, Joy (2017): From scrambled messages to an impromptu dip: Serendipity in finding a field location. In: Hendry, Joy (ed.): An anthropological lifetime in Japan. Leiden: Brill, pp. 17–33. Keyso, Ruth A. (2000): Women of Okinawa: Nine voices from a garrison Island. Ithaca, NY: Cornell University Press. Kulick, Don/ Rydstrom, Jens (2015): Loneliness and its opposite: Sex, disability and the ethics of engagement. Durham, NC: Duke University Press. Lahad, Kinneret (2017): A table for one: A critical reading of singlehood, gender and time. Manchester: Manchester University Press. Lee, Ching Kwan (1998): Gender and the South China miracle: Two worlds of factory women. Berkeley, CA: University of California Press. Lesthaeghe, Ron (2014): The second demographic transition: A concise overview of its development. In: Proceedings of the National Academy of Sciences of the United States of America 111, No. 51, pp. 18112–18115. LeTendre, Gerald (2000): Learning to be adolescent: Growing up in U.S. and Japanese middle schools. New Haven, CT: Yale University Press. Lewis, Catherine C. (1995): Educating hearts and minds: Reflections on Japanese preschool and elementary education. Cambridge: Cambridge University Press. Lo, Jeannie (1990): Office ladies/factory women: Life and work at a Japanese company. New York, NY: M. E. Sharpe. Malinowski, Bronisław (2013): Argonauts of the Western Pacific: An account of native enterprise and adventure in the archipelagoes of Melanesian New Guinea. Long Grove, IL: Waveland Press. Maoz, Zeev (2002): Case study methodology in international studies: From storytelling to hypothesis testing. In: Harvey, Frank P./Brecher, Michael (eds.): Evaluating methodology in international studies. Ann Arbor, MI: University of Michigan Press, pp. 455–475. Marcus, George A. (1995): Ethnography in/of the world system: The emergence of multi-sited ethnography. In: Annual Review of Anthropology 24, No. 1, pp. 95–117. Maree, Claire/Okano, Kaori (eds.) (2018): Discourse, gender and shifting identities in Japan: The longitudinal study of Kobe women’s ethnographic interviews 1989–2019, phase one. London: Routledge. McRobbie, Angela (2004): Post feminism and popular culture. In: Feminist Media Studies 3, No. 3, pp. 255–264. Miller, Daniel (2008): The comfort of things. Cambridge: Polity. Nakano, Lynne (2016a): Single women and the transition to marriage in Hong Kong, Shanghai and Tokyo. In: Asian Journal of Social Science 44, pp. 363–390. Nakano, Lynne (2016b): Marriage and reproduction in East Asian cities: Views from single women in Shanghai, Tokyo, and Hong Kong. In: Zheng, Tiantian (ed.): Cultural politics of gender and sexuality in contemporary Asia. Honolulu, HI: University of Hawai‘i Press, pp. 156–170. Ogasawara, Yuko (1998): Office ladies and salaried men: Power, gender, and work in Japanese companies. Berkeley, CA: University of California Press. Okano, Kaori (1993): School to work transition in Japan: An ethnographic study. Clevedon: Multilingual Matters. Okano, Kaori (2009): Young women in Japan: Transitions to adulthood. London: Routledge. Okano, Kaori (2018): Rethinking ‘eurocentrism’ and Area Studies: Japanese Studies in the Asia-Pacific. In: Okano, Kaori/Sugimoto, Yoshio (eds.): Rethinking Japanese Studies: Eurocentrism and the Asia-Pacific region. London: Routledge, pp. 1–18. Peak, Lois (1991): Learning to go to school in Japan. Berkeley, CA: University of California Press. Plath, David (1980): Long engagements: Maturity in modern Japan. Stanford, CA: Stanford University Press.

100

Chapter 3 How to organise research Rivoal, Isabelle/Salazar, Noel B. (2013): Contemporary ethnographic practice and the value of serendipity. In: Social Anthropology 21, No. 2, pp. 178–185. Roberson, James E. (1998): Japanese working class lives: An ethnographic study of factory workers. London: Routledge. Roberts, Glenda (1994): Staying on the line: Blue-collar women in contemporary Japan. Honolulu, HI: University of Hawai‘i Press. Roesgaard, Marie (2006): Japanese education and the cram school business: Functions, challenges and perspectives of the juku. Copenhagen: Nordic Institute of Asian Studies Press. Rohlen, Thomas (1983): Japan’s high schools. Berkeley, CA: University of California Press. Roth, Joshua H. (2002): Brokered homeland: Japanese Brazilian migrants in Japan. Ithaca, NY: Cornell University Press. Ryang, Sonia (1997): North Koreans in Japan. Boulder, CO: Westview Press. Sassen, Saskia (2001): The global city: New York, London, Tokyo. Princeton, NJ: Princeton University Press. Sato, Nancy (2003): Inside Japanese classrooms: The heart of education. New York, NY: RoutledgeFalmer. Sheller, Mimi/Urry, John (2006): The new mobilities paradigm. In: Environment and Planning 38, No. 2, pp. 207–226. Shimahara, Nobuo/Sakai, Akira (1995): Learning to teach in two cultures: Japan and the United States. New York, NY: Garland. Song, Jesook (2014): Living on your own: Single women, rental housing, and post-revolutionary affect in contemporary South Korea. Albany, NY: State University of New York Press. Stake, Robert E. (1995): The art of case study research. London: Sage. Sugimoto, Yoshio (2014): An introduction to Japanese society. Cambridge: Cambridge University Press. Sugimoto, Yoshio (2018): Turning towards a cosmopolitan Japanese Studies. In: Okano, Kaori/Sugimoto, Yoshio (eds.): Rethinking Japanese Studies: Eurocentrism and the Asia-Pacific region. London: Routledge, pp. 167–183. Tight, Malcolm (2017): Understanding case study research: Small-scale research with meaning. Los Angeles, CA: Sage. Tobin, Joseph/Wu, David/Davidson, Dana (1991): Preschool in three cultures: Japan, China and the United States. New Haven, CT: Yale University Press. Tobin, Joseph/Yeh, Hsueh/Karasawa, Mayumi (2009): Preschool in three cultures revisited: China, Japan, and the United States. Chicago, IL: University of Chicago Press. Tsuneyoshi, Ryoko (2001): The Japanese model of schooling: Comparisons with the United States. New York, NY: RoutledgeFalmer. Van Andel, Pek (1994): Anatomy of the unsought finding. Serendipity: Origin, history, domains, traditions, appearances, patterns and programmability. In: British Journal for the Philosophy of Science 45, No. 2, pp. 631–648. Welch, Catherine/Piekkari, Rebecca/Plakoyiannaki, Emmanuella/Paavilainen- Mäntymäki, Eriikka (2011): Theorising from case studies: Towards a pluralist future for international business research. In: Journal of International Business Studies 42, No. 5, pp. 740–762. Yan, Yunxiang (2003): Private life under socialism: Love, intimacy and family change in a Chinese village: 1949–1999. Stanford, CA: Stanford University Press. Yin, Robert K. (2018): Case study research and applications: Design and methods. London: Sage.

101

Chapter 4 How to identify relevant scholarly debates: Reviewing the literature Urs Matthias Zachmann

1. Introduction Why should I bother with what other people have to say on my subject? Can’t I just start with the raw data and work my way from there through analysis towards the finished dissertation? Conversely, others may ask: there is so much stuff already written. How can I stand on the shoulders of giants and produce something new at all? Should I not limit myself to just taking stock of what knowledge there is already and venture, at best, a modest interpretation thereof? These are doubts that trouble many scholars at varying stages of their career. As a historian of modern Japan, I wrote my doctoral dissertation on Sino-Japanese relations in the late Meiji period. At first glance, there seemed to be so much literature out there that it was hard to see how I could produce something new and relevant on top of this (after a while, it turned out there was still lots to find out). My second book on Japan’s engagement with international law was exactly the opposite: although there was an abundance of primary sources, little research had been done on this subject. This was not an ideal situation either, because in unmarked territory, one easily wanders astray. It turned out, however, that many Japanese legal scholars took a strong interest in the subject and were able to point me in the right direction. This dilemma of too much or too little scholarship on one’s research subject is, of course, a fundamental one. In his Analects, Confucius had already observed that studying without thinking makes one stupid and thinking without studying is dangerous. To paraphrase this in the context of the above questions and the subject of this chapter: Relying solely on previous scholarship without any research and analysis of your own brings no progress at all, but research and analysis without consulting previous scholarship is even more dangerous (also for your career). Professional science is a rational process of incremental knowledge production based on a division of labour. Ideally, you climb the communal edifice of knowledge and add a well-designed and functional building block of your own. At times, you might tear something down that does not fit in this edifice and replace it with something better (see Gerteis, Ch. 16.3). What you must not do is ignore previous scholarship and build your own little tower, as this is neither rational nor practical. There is a life and a career waiting for you beyond the PhD. But equally, you do not just want to paint the tiles yellow instead of brown because you found the edifice too intimidating to add something new to. This chapter and the following essays intend to help navigate your course through the sea of literature towards a completed literature review and towards writing up the dissertation itself. Here, I take the PhD research and writing process as the default situation, but what follows is 102

Chapter 4 How to identify relevant scholarly debates applicable to a master’s thesis or a second book as well. I will first introduce the differences between reviewing literature as a process and a literature review as part of a thesis or other publications. I will then discuss different kinds of sources and elaborate on how to find secondary literature in general, and finally give some guidance on writing a literature review.

2. Reviewing literature: Two kinds of review When we speak about reviewing literature, we often mean two different things, which should be kept apart because they vary significantly in their respective processes, scopes and purposes. Two kinds of review 1.

2.

Reviewing literature as a continuous process throughout your research, which is finished only when your thesis is submitted. • Process: a continuous, iterative and circular (see below) search and evaluation of material • Scope: comprehensive, includes all primary sources and secondary literature, even search tools (bibliographies, indices, etc.) • Purpose: to find all material that you need for further research and writing your thesis The literature review as a critical narrative account of ‘what has been done and what needs to be done’ in your research field with a focus on the main debates and recent trends (Department of Sociology 2012, p. 18). • Process: a written partial result of reviewing literature, done and redone at different stages during your PhD and part of the final dissertation (see below) • Scope: selective, uses only such literature that answers the above questions and serves the following • Purpose: to explain to readers what motivates your research, the ‘puzzle’ that has remained unsolved by previous scholarship; to situate your own work within the recent debates and trends and thereby explain its relevance

Reviewing literature is a comprehensive process that will accompany you from conceiving an idea until submitting your thesis. It is not a straightforward process, but takes a circular, iterative course (see this chapter, Liu-Farrer, Ch. 4.3). As you progress with your research, you will tweak your research question, pivot into a somewhat different direction or modify your theoretical framework or methods along the way. You will then start reviewing the literature again, bearing these changes in mind, find more and more items to add on your subject and ideally develop a better understanding of how your subject relates to cognate subjects in and outside your field as well. This, in turn, will inform your research question, theory and method in a circular motion. A literature review, on the other hand, is only a partial, selective result of the process of reviewing literature, which becomes part of the finished PhD thesis. In practice, you are often

103

Urs Matthias Zachmann required to write an increasingly refined literature review at different stages of your PhD process. When preparing a proposal to apply for a particular PhD programme, you already have to make a case for what you do and how this relates to what has been done in your field and to current debates and trends. This is also true for applications to scholarships and other kinds of funding. Most often, writing a literature review is part of the first-year review that decides whether you can progress with your project or not. So you will have a fair number of chances to improve your writing skills in this genre.

3. Getting started: The scope of searching for secondary literature When starting to review literature, you should be comprehensive. This means not just looking for literature on the topic itself and the field it is situated in, but also that on methods and theories and on related studies in different fields. The rationale of this is 1. that you should not narrow your search down too quickly and miss the context, and 2. that you want to situate your research in a broader field and speak to audiences beyond the confines of your direct field of research. Gracia Liu-Farrer (see this chapter, Ch. 4.3) has described this oscillating movement of the scope of your search as ‘zoning in’ and ‘zoning out’. It is only after you have done a comprehensive search that you whittle down the results in terms of relevance, quality, etc. as described below. Throughout your research, I advise you to keep two records—an annotated bibliography and a research journal (see text box below)—and constantly add to them. Keep the following records throughout your research

104

1.

An annotated bibliography (required): Keep a structured list of the literature you find, divided into primary sources and secondary literature (see below) and suitable thematic subsections (the most intuitive structure would mirror the structure of your PhD, with variations and additions as necessary). For this, use reference management software like EndNote or Citavi, or just a simple Word file. In any case, each entry should contain: • the full bibliographic data, • the main arguments, findings and method of the item and • your personal evaluation as to the relevance and usefulness of the source. This annotated bibliography will be a steady companion during your research and will provide the backbone for your literature review at the beginning of your PhD as well as at the end when you finalise your reference list.

2.

A research journal (recommended): Keeping a research journal (preferable in analogue form) is also most useful in order to jot down notes and ideas on your dissertation in general. Apart from the memory function, keeping a journal gives a sense of continuity and progress, which is sometimes badly needed in those three to four years. Plus, it is excellent writing practice (Kolmer/Rob-Santer 2006, pp. 93–94).

Chapter 4 How to identify relevant scholarly debates

4. Basic types of sources and their general ambivalence In reviewing literature, you have to make a clear distinction between ‘primary sources’ and ‘secondary literature’. These two categories should be kept apart in the bibliography and sometimes in the references in your dissertation (depending on your discipline). Primary sources are any stable body of information that is the original (primary) object of your analysis, e.g. written texts, archaeological or contemporary objects, material or digital images, statistical data or recorded and transcribed interviews. Secondary literature, in contrast, is everything that is written about this body of information and your research topic in general. It typically contains explicit analysis or interpretation of it. It is secondary in the sense that it is layered on top of the primary source as the interpretive icing on the cake.

4.1 Primary sources Traditionally, the division was not between primary sources and secondary literature but between primary and secondary literature, assuming that the main object of your study would be found in written sources. This certainly applied to heavily text-oriented, traditional philological and humanistic fields of Japanese Studies such as literature or philosophy that dealt (and deal) with texts, such as the Man’yōshū or a Buddhist treatise like Dōgen’s Shōbōgenzō. With the general diversification of the field of Japanese Studies (see Goodmann, Ch. 1), the emergence of Social Science research on primary sources on Japan today particularly has diversified. In general, any material or immaterial element of information can constitute a primary source. This includes, but is not limited to, written texts, objects like archaeological artefacts or contemporary material culture or even such intangible information as observed rituals and practices in certain social settings or interviews (Kühmstedt 2013, pp. 19–21). The precondition for being a source is, of course, that it is manifested as a stable body of information that can to some degree be stored, reproduced and referenced in research. In some cases, the process of turning something into a workable source already leads to a certain subtle gap between ‘the real thing’ and the source itself, of which one has to be acutely aware. If we study gagaku, Japanese court music, our source is not the music and the performance itself, but recordings of it. We have to be aware of the mediality of the sources that, as mere representation, makes them one step removed from reality and subject to (inadvertent) distortion and we also need to maintain a healthy critical distance towards them. This is also the reason why, with written texts, one should use the critical, authoritative edition. This is mostly the collected works of an author (zenshū).1 As the collection and analysis of primary sources is discussed in chapters 5–14, in the following sections, I will focus on secondary literature.

1 However, vigilance is useful here as well, because not all editions are as critical and thorough as they pretend to be.

105

Urs Matthias Zachmann

4.2 Secondary literature In contrast to primary sources, secondary literature is invariably just that: written texts. This body of sources comprises the previous research on your topic and constitutes the main bulk of your literature review. However, the boundary between primary and secondary sources/ literature is porous (see this chapter, Ganseforth, Ch. 4.2). Thus, at times, secondary literature too can and should become the object of analysis and interpretation, and therefore those works should become primary sources in their own right. This is particularly true with secondary literature informed by a certain political and ideological position or literature which reflects certain power asymmetries and thus becomes part of a specific discourse. A classic example of such fluidity is Ruth Benedict’s book The chrysanthemum and the sword (1946), which, at the time of its writing during wartime, was an anthropological study of the Japanese people. Today, this book is considered a seminal primary source for the genre of the so-called nihonjinron, the essentialist discourse on ‘the nature of the Japanese people’.2 These examples show that one has to be aware of the fluidity of the division of works into primary and secondary sources/literature.

5. Finding secondary literature But what is the best way to find secondary literature? There exists a multi-pronged approach of different search strategies, which you should pursue routinely and simultaneously throughout your research. In the following, I provide a checklist of common and useful strategies. Finding secondary sources 1.

When doing general bibliographic research • Search the OPAC of the university library you are affiliated with; • search several other OPACs of big research universities and institutions (e.g. Bibliothèque Nationale, Harvard University or Library of Congress) and regional aggregated or global catalogues (WorldCat); • use bibliographies such as the Bibliography of Asian Studies (BAS) and other more specialised bibliographies in your field (online and print); • ask librarians at your institution, particularly Asian languages librarians, for advice; • browse review sections of journals in your field and reviews in dedicated mailing lists (e.g. EAJS; H-Japan; SSJJ); • go into the stacks, if you can, and have a look at what is physically there (you will find more and unexpected things).

2 The same is true for Nakane Chie’s Tate-shakai no ningen kankei (1967) or Takeo Doi’s Anatomy of dependence (1973).

106

Chapter 4 How to identify relevant scholarly debates 2.

When searching Japanese sources in particular • Again, search several OPACs of big Japanese research libraries and institutions (National Diet Library, Keio University, Tokyo University, Waseda University); • use bibliographic databases like CiNii or, if you have access, CrossAsia; • use online repositories of publications and documents, such as the Japan Center for Asian Historical Records (JACAR) or the National Diet Library Digital Collections; • ask your mentors or peers at Japanese universities for bibliographic advice; • browse review sections of Japanese language journals in your field; • if you are in Japan, again, go into the stacks, and see what you can discover there and visit well-stocked bookshops, either new or antiquarian (see below).

Some of the above strategies need some explanation: when using OPACs, make liberal use of advanced search tools, as otherwise you will be hit with a deluge of results that, despite what it says, often do not come with the most relevant on top. Librarians, particularly those working in your regional field, are mostly very happy to advise you on your research; and so are your supervisors, mentors and peers. But do your homework first and consult them only when you have done some substantial digging yourself. Ask your librarian also whether there are courses on bibliographic search tools available at your university. More and more institutions put their content online. I have only given two examples (NDL and JACAR), but there are of course many, many more. For access to Japanese university libraries, it is of course best to have an affiliation, e.g. as a guest researcher (kyakuin kenkyūin), at a university with an extensive research library. I, for example, spent blissful weeks and months in the stacks of Waseda, Keio and Tokyo University libraries. But if you are there only for a short time so that an affiliation seems impractical, the easiest way is to use the library of a public university or institution, as you can show up on the day and do not need a special letter of introduction. Tokyo University Central Library (Sōgō Toshokan), for example, will issue you a day pass if you can demonstrate an affiliation with your home university; take business cards (meishi) with you (see Tagsold/Ullmann, Ch. 8) and name a source in their holdings you want to consult. And then there is, of course, always the National Diet Library as a last resort: the NDL has fantastic holdings, but it takes time to make use of them (see Peucker/Schmidtpott/Wagner, Ch. 9). To search for and buy old or used books, use Nihon no Kosho-ya, the centralised online search tool for all used and antiquarian bookstores in Japan. Some also deliver overseas, but most require an address in Japan (but you can also just check online and then go there personally, of course). The same network also organises book fairs (furuhon matsuri), frequently held at the Tokyo Kosho Kaikan in Kanda. This area is also famous for its many specialised antiquarian and second-hand bookstores (see this chapter, Maclachlan, Ch. 4.1), where you will most certainly find many books in your field that are out of print. I would like to make one final remark: in the same way as there are Western-language manuals for searching for literature, there are of course similar manuals and bibliographies in Japanese for the same purpose (for example Fujita 2020; Hamada 2020; Hiraoka et al. 2013; Satō

107

Urs Matthias Zachmann 2000). Be sure to consult these, too, if you need more inspiration and guidance on finding Japanese literature.

6. Cultivating a sense for the ‘best source’ As already indicated in the previous section, it is important to be discriminating in your selection of secondary literature from among the deluge of possible results. This is particularly so since we increasingly use electronic search tools that yield a huge number of hits if we do not teach them to narrow down the search with a number of criteria. These criteria and the cultivation of a sense and intuition help us to find the best source for our research. With ‘best source’ I generally mean that you should prefer a source that is 1. specifically targeted at your topic in scope, 2. the original source of information and analysis, and 3. likely to be the most reliable source in terms of academic quality. Let me explain these criteria in the following. Specifically targeted at your topic in scope is literature which specifically addresses the particular subject that you are writing about at any given time, not in a general way, but already by title (mostly; see below) and, of course, content. To give you just one example: suppose your dissertation is about late Meiji Japanese–Chinese relations. In your historical background chapter, you want to write about their premodern trajectories and, at this particular point in writing, about their relations during the Tokugawa period. You have a choice between Marius B. Jansen’s China in the Tokugawa world (1992) and Mikiso Hane and Louis Perez’ Premodern Japan: A historical survey (2015). In this case, Jansen’s work is the ‘best source’, because it is a specialised study that, in title and content, addresses the specific topic you are writing about. Hane and Perez’ general history textbook may have excellent passages on Japan’s foreign relations at the time and even contain information or analysis that may not be found in Jansen’s study. But in this case, Hane and Perez took it most likely from a third source, such as Ronald P. Toby’s State and diplomacy in early modern Japan (1984), which in turn is also the ‘best source’ (just not as obvious in the title, so the title’s recommendation has to be taken with a pinch of salt). In any case, the more specific literature always trumps the more general overview. The above example also brings us to the next criterion: use the original source of information and analysis. You should avoid all ‘second-hand’ literature that, in turn, only references back to another, the original locus of information. Check the original source as the ‘best source’, and reference this in your footnotes and bibliography. The original source-rule also often excludes, most importantly, Internet sources that take their information and analysis from published sources (whether explicitly or without references). Here, again, the published source is the one that should be used. Of course, this does not exclude Internet sources per se as long as these are, in turn, the original site where the particular information or analysis can be obtained (i.e. published studies may reference statistics from the Ministry of Health, Labour and Welfare website; in this case, the website is the best because it is the original source). As a rule of thumb, the use of Internet sources as the best source increases the more contemporary the subject is.

108

Chapter 4 How to identify relevant scholarly debates The final criterion, the most reliable source in terms of academic quality, is somewhat delicate because, as one should not ‘judge a book by its cover’, one cannot gauge the academic quality of a work without actually having read it. But with the deluge mentioned above, we cannot read everything (even if we try skimming) and we have to develop some strategies and an intuition that allows us, maybe not to exclude, but to order the literature in terms of what to read first and later. Here are some recommendations. How to select reliable scholarly literature Generally, as a rule-of-thumb, the best scholarly source 1. is published in a professional academic publication venue (either print or online) as an article or (E-)book; 2. has an identified author (i.e. is not anonymous); 3. if it is a monograph, it will be published with a good academic or commercial publisher: • a university-associated press (e.g. Cambridge, Columbia, Cornell, Edinburgh, Harvard, Oxford, Tokyo University); • some international commercial presses (e.g. Brill, Palgrave, Routledge, Sage); • established national academic and commercial publishers (e.g. Beck and Nomos in Germany, Presses universitaires de France, CNRS Éditions and Flammarion in France or Einaudi and Skira in Italy); • Japanese publishing houses like Chikuma shobō, Iwanami, or Kōdansha. 4.

If it is an article, it is published in • a peer-reviewed journal in Area Studies (ASIEN, Contemporary Japan, Japan Forum, Journal of Asian Studies, Journal of Japanese Studies, Monumenta Nipponica, Social Science Japan Journal); • or the discipline(s) you are working in (e.g. American Anthropologist, American Historical Review, Gaikō Forum, Shigaku Zasshi, Shirin, Sociology, etc.).

The main rationale in favour of the publication venues listed above actually follows from the first stipulation, namely that these are professional publication venues which have, as such, inbuilt quality assurance mechanisms, ideally a strict peer review process (with commercial publishers, quality assurance can be a bit patchy, though). Which means that other people, experts in your field, will have already pre-screened these publications for you and rejected some or required substantial improvements to them before publication. As cautioned above, this does not mean that one should ignore publications in ‘esoteric’ or unlikely publication venues (e.g. in journals that otherwise do not have any track record with Japan or the discipline you are working in). But often enough there is a reason why someone chose to publish there and not in an established academic journal that is central to the field. This is not snobbery but, considering the flood of sources one has to deal with, pure practicality. However, there is one exception to be made for Japanese language articles: Japanese colleagues often publish excellent articles in university-associated smaller venues, particularly bulletins (kiyō) or faculty-related periodicals. These should be taken into full consideration, despite all rules of thumb.

109

Urs Matthias Zachmann

7. An inclusive approach to literature: Literature in Japanese and other languages Now that we have discussed means of whittling down the bulk of available literature by scope and quality, this section follows the other direction and encourages you to be as inclusive as possible when it comes to languages, and particularly regarding Japanese language literature. Although it is often said that 90% of literature in Japanese Studies is produced in English, this is actually not true. The vast majority of literature on Japan and on any Japan-related subject is written in Japanese. This should be a self-evident fact just from statistics: the number of authors as well as the number of readers who write and read about Japan-related topics are, of course, biggest in Japan itself. This simple calculation applies to Japan-related topics even more so than to other national academic communities, as Japanese colleagues still tend to write more exclusively in Japanese than other academics. It is therefore paramount to consider all the Japanese language literature that is relevant to your topic (see this chapter, Liu-Farrer, Ch. 4.3; Maclachlan, Ch. 4.1). At least in Japanese Studies, you as a scholar and your publications will not be taken seriously if you have not included the relevant Japanese language literature on your subject in your work. As a non-native speaker, this requires more linguistic effort on your part. Unfortunately, this does not mean that you may read less literature just because it is written in Japanese. Nor should it be a fig leaf exercise, i.e. including Japanese language literature in a superficial way so that you have ticked this box, too. You should engage with Japanese language material as seriously as with literature in English or other languages. Consequently, previous scholarship on your topic in Japanese also sets the height of the bar over which you have to jump in order to arrive at new findings. At times this bar seems awfully high. But this does not mean that you should mystify Japanese sources. As Liu-Farrer (see this chapter, Ch. 4.3) argues, Japanese colleagues have a more intimate relationship with their subject, enjoy more sustained access to Japanese sources and profit from linguistic and cultural advantages. However, good scholarship is the product of hard work in all languages. For example, Japanese students of History have to learn to read premodern or even Meiji sources in the same hard way as Western students do. Also, an intimate relationship and cultural proximity can be a double-edged sword for a scholar. Similarly, unlimited access to sources is not always a blessing (a hard-earned personal insight after one year in the Waseda University stacks). And although it is often argued that Japanese scholarship can offer different insights to Western scholarship, this goes both ways. Thus, as a non-Japanese scholar, one should not feel intimidated by Japanese sources either, or accord them with a special status of being naturally more authoritative. But, of course, the sheer number of Japanese scholars doing research in a particular Japan-related field is much higher than elsewhere, and therefore it is likely that there exist more very good studies in Japanese than in other languages. When dealing with Japanese secondary sources, beware of another Western-centric bias that values Japanese language sources predominantly as a ‘data mine’. Some scholars appreciate it for its richness in factual detail and information, but otherwise see Japanese scholarship as lacking in analysis, theoretical contextualisation and methodological rigour. Ironically, this attitude among many Western scholars in Japanese Studies mirrors the attitude with which they, in turn, are treated by their colleagues in ‘hard disciplines’ (Cheah 2001). They use the materi110

Chapter 4 How to identify relevant scholarly debates al diligently accumulated by Japanese scholars/Japanese Studies scholars for the ‘real’ rigorous analysis done in Japanese Studies or/and Social Science or the Humanities. But often enough, this seeming lack is a matter of representation and of differences in academic cultures. In Japanese scholarship, analysis and methodology are often more implicit, but still present. However, with regard to languages other than Japanese and English, one could argue that all relevant literature should be included in your study. Depending on the topic of your research, a multilingual approach is even ingrained into your research design (see this chapter, Ganseforth, Ch. 4.2). This universal approach may seem impractical for languages you do not speak (although you could ask someone or use rough translation software for a start). But at least you should include the literature in languages you do speak, particularly when you submit your thesis in a non-English speaking country (colleagues get very disappointed when their research is overlooked just because it is not written in English). This has the practical consequence that you should always search OPAC with keywords in various languages (the language button on the side often does not work properly).

8. Reading secondary literature: Some practical advice An important part of reviewing literature is, of course, reading it. However, do not try to read ‘everything’ before starting to write. I suggest the following: During your orientation phase (usually the first year of your PhD), do several rounds of literature reviewing. After the first round, skim lightly what you have found (see this chapter, Maclachlan, Ch. 4.1) and note in your bibliography the main arguments, findings and methodological approaches of the work in question as well as your own valuation of it. Then pick out a handful of publications that you deem the most relevant and important and read them thoroughly, before starting the next round of research. Reading literature thoroughly means that you read it with absolute concentration, while taking copious notes. Without concentration, nothing good can be accomplished in research. So, develop a set routine of when and where to work, create an environment as disturbance-free as possible and, most importantly, lock away your mobile phone, shut down the email browser and cut off the Internet. Do not try to multitask and take up one source at a time. An important routine while reading is taking notes. For each piece of literature, create a separate file in which to store your notes (you can do this with Citavi, EndNote or just a simple Word file; file them in a structure that mirrors your bibliography). In these notes, jot down a summary of the book, notable quotes you want to use and also your own observations on how the arguments, findings, theory and method relate to your topic. This has several purposes: three to four years is a long time, and just scribbling in the margins will not help to find things when you need them most—when writing. Also, taking notes in an organised way will help you develop your own thoughts in conversation with what you read and, thus, is good preparation for easing into writing. Finally, and also importantly, carefully taking notes helps you to avoid inadvertent plagiarism (Massengill 2012, p. 18; see also Reiher/Wagner, Ch. 16). When you read so much, the lines between your own thoughts and other people’s get blurred

111

Urs Matthias Zachmann if you do not take notes. Consequently, when taking notes, distinguish between direct quotes, paraphrases and your own observations! A final note on the general mindset while reading secondary literature: Academics often say that one should read critically. But, to paraphrase Bertrand Russell (1946/2004, p. 47), it would be better to start with an attitude of hypothetical or detached sympathy that first and foremost seeks to really understand what the author is driving at (even if it is only imperfectly presented or evidenced) and only then start criticising it. This way, your reading will be much more fruitful and balanced.

9. Identifying relevant debates and situating one’s own research The two main purposes of reviewing literature are 1. to identify the relevant debates in your field, both in Area Studies and the discipline you work in, and 2. to situate your own work in relation to it. Also, it is quite dangerous (and not very motivating either) to ‘go it all alone’ and pursue your research in the isolation of your library carrel, your office or your desk at home. The best way to get on top of recent debates is—while doing your literature review and continuing to work at it—to embed yourself in and engage with different scholarly communities. In the following, I offer some advice on how to do this. Efficient ways to get an overview of recent trends in academic debates 1. Reading • Consult up-to-date handbooks (e.g. Brill, Routledge, SAGE) (both in Area Studies and your discipline(s)); • look for recent edited volumes and particularly screen the introductions which are supposed to outline the state of research (both in Area Studies and your discipline(s)); • look for special issues of relevant journals; • study book reviews as they usually contextualise findings in the current trends and debates. 2. Communicating with peers and senior scholars in your field • Ask senior scholars in your field for advice and introductions; • sign up to relevant mailing lists and academic fora (e.g. EAJS-L, H-Net, J-Studien); • apply to and attend as many PhD workshops as possible (e.g. EAJS PhD workshops); • apply to and attend postgraduate and regular conferences in Area Studies (e.g. Association for Asian Studies, British Association of Japanese Studies, European Association of Japanese Studies, Japanologentag) and in your discipline(s) (e.g. American Anthropological Association, American Historical Association, International Sociological Association, etc.). Note that there are often travel grants for PhD students.

112

Chapter 4 How to identify relevant scholarly debates 3. When staying in Japan • Ask senior scholars in your field for advice and introductions. • If possible, attend a zemi (a regular ‘seminar’ or colloquium of postgraduate students under the same supervisor) run by a senior scholar working in your field; participate actively and present your research in Japanese. It is perfectly fine to read from a manuscript and distribute it among participants for them to read along (Hiraoka et al. 2013). • If possible, attend one or several kenkyūkai (regular meetings of national or local research networks) in your field; if asked (do not venture to do so by yourself but wait until asked), present your research in Japanese; also make sure you attend the konshinkai, the social gathering, after the workshop. When you present at conferences, workshops or kenkyūkai, people are often eager to point out relevant literature and to contextualise your research in the relevant debates. This is also very motivating and makes you realise that research is not a solitary fight with the material, but a living debate about ideas. Even negative or muted feedback, as much as it stings at first, is extremely helpful in the long run. Just do not let it get under your skin or demotivate you (see Farrer/Liu-Farrer Ch. 17). When presenting at disciplinary (non-Japan-specific) conferences, be prepared to get other or fewer questions than you are used to from Japanese Studies conferences, or even harsher and methodologically more fundamental questions than you might be used to from Japanese Studies conferences. Finally, let me make a remark on time management: As interesting and aspiring as it is to attend conferences and workshops to discuss issues and meet people, it also takes a lot of (often underestimated) time to prepare and can be disruptive to your research routine. Keep in mind that a conference paper is not your dissertation. So, when presenting, go for quality, not for quantity, and find the right balance between presenting papers and writing your dissertation.

10. Writing the literature review The literature review is a critical narrative account in a section at the beginning of your PhD thesis (and proposal(s) leading towards it). I use the word ‘narrative’ to stress the purposedriven nature of the literature review, which is to explain what you do by way of what has been done, or not. It is therefore not the bibliography you have compiled, nor a report on the state of the field in general. It is a much more selective and purposeful narrative that always has the same ending to its tale: ‘And that is why what has been done is not enough and why what I am doing is important.’ Its general content, purpose and structure are always as follows.

113

Urs Matthias Zachmann Content, purpose and structure of the literature review Every literature review contains the same narrative elements that answer the following questions • •

•

What has been done by previous research in the field regarding your topic? What are the main findings, arguments and debates in previous scholarship? Why is previous scholarship not enough to solve the ‘puzzle’ from which your research question arises? What contradictions or gaps does this research contain that create your puzzle in the first place? How does your own research contribute to debates and recent trends in your field in terms of topic, theory and method?

The scientific purpose is, of course, to situate your project in the edifice of professional science (see introduction). Invariably, however, a literature review also serves • • •

to demonstrate your grasp of the most important arguments, debates and trends in research regarding your topic; to clear space for your own contribution and to showcase its importance; to relate your research to peers and predecessors in Area Studies and disciplines in which you intend to pursue an academic career.

The general structure of the narrative follows from these elements and purposes • •

• •

•

Its narrative clusters around larger debates and trends rather than single contributions of scholars. The order in which you present the narrative can be 1. chronological (How has the field evolved in terms of theoretical or methodological trends? How has the debate evolved?) or 2. thematic (What are the main arguments relating to the topic? How can they be broken down?). In any case, the presentation usually follows a dialectical sequence (while A argued x, B contended -x, upon which C suggested y…). Participants in the debates are introduced with their main findings, arguments and theoretical or methodological contributions, not a description of book or article contents. The relevance of all sub-fields and approaches to your project must be addressed (and vice versa).

In terms of the flow of presentation, it is useful to imagine a metaphorical dinner party (Kamler/Thomson 2006, pp. 37–38; Lloyd 2017/18) or, if you like, to visualise Raffael’s fresco The School of Athens, in which groups and factions of people lead lively debates. Imagine them to be your peers and predecessors who are discussing the research in your field regarding your topic. You are right in the middle, holding forth why this is not good enough and why something has to be done about it (see McMorran, Ch. 15)! But, of course, keep your tone civil in your counterarguments. Just like reading (see above), write your literature review with detached sympathy. However, there is no simple fit-for-all template for a literature review, as every discipline has its own style. Each sub-field of Japanese Studies takes its rules from the respective discipline (see this chapter, Maclachlan, Ch. 4.1). To get a better idea of what a literature review can

114

Chapter 4 How to identify relevant scholarly debates look like in your field, my suggestion is that you dig up the original PhD theses of the books most relevant to your topic (not the published book itself, because that contains only a muchabridged version) and take a look at the literature review. This might also help to answer the following two questions: When do I have enough literature? When do I start writing? Reviewing literature can be overwhelming. You might think that there is so much material out there that you cannot cope. Keep calm and do not let the seeming immensity of the task overwhelm you. Take one quick step back and reconsider it in the greater scheme of your research. Generally speaking, a ‘perfect’ or ‘comprehensive’ literature review is the enemy of a very good one. As you should not aspire to write a masterpiece with your dissertation (although you might certainly produce one), you should also not aim for perfection in reviewing literature. These things serve a purpose and are not ends in themselves. So, what you should aim for is a thorough literature review at first that is good enough for you to continue with the next steps in your research.

11. When to start writing Do not try to read ‘everything’ before you start writing. This takes too much time and can result in massive writer’s block. Of course, during the first year of your PhD, you mainly read and search without producing much (except your notes, and your entries in the bibliography and journal). But during the second year, you will see the fog clear and get a feeling for the lay of the land. You can start drawing up a final marching route for the dissertation and for the particular chapters with a sense of where the whole thesis and each of the chapters are heading (a rough narrative). When you have reached this point, I suggest that you start writing while continuing to read. Writing, as much as researching, is a circular process, so when you reach the end of your thesis/of each chapter, you will have to go through it/them from the beginning and add to where there is something missing. Once you have acquired a general sense of direction, follow the guide below.

115

Urs Matthias Zachmann When you roughly know your direction, start writing (and continue to read) 1.

2.

Do not try to read ‘everything’ before starting to write, but read as much as you need in order to orient yourself in a certain direction (a general ‘marching route’) in order to get going and then read along the way, filling in the gaps. When starting on a particular chapter or section • First, review the necessary literature (primary and secondary) and skim through what you find. • Pick the most important and promising works (say five or eight) and work through them thoroughly. • Draw up a general outline of the chapter with the particular points and aspects you want to address and the sequence in which to do so (a ‘storyboard’ or ‘marching route’, as it were). • Then start writing in line with this, working the rest of the literature into the various aspects while writing. • Naturally, this being a circular process, you might have to start from the beginning and readjust the chapters, etc. somewhat until the end.

12. Summary Naturally, people have different writing strategies and habits. However, I think the general principle is that you have to find a balance between reading and writing. In figurative terms, imagine research is like growing a tree. Nature does it incrementally and iteratively. The trunk, roots and some main branches come first, then the whole is strengthened and more detail (branches, leaves, etc.) is added. Nature does not make jumps and, similarly, you should grow your literature review as an important part of your dissertation with patience and tenacity. Safe journey and good luck!

116

4.1 Looking for sources in all the right places Patricia L. Maclachlan

Searching for good primary, secondary and tertiary sources in Japan Studies can be a daunting task. While Japan offers a cornucopia of riches for scholars in search of the printed and electronic word, it can be hard for the foreign researcher to fully access that information. In this chapter, I draw from my own—often trial-and-error—experiences to offer some tips on how to navigate these challenges and opportunities.

Definitions Before progressing, I will provide some definitions. Primary sources are accounts of some event or phenomenon written by someone who experienced it first-hand. They include autobiographies, interviews, photographs and newspaper accounts that are contemporaneous with the event in question. Also included in this category are government policy statements and survey data, and in Cultural Studies and other Humanities fields, original literary, theatrical and other artistic works. Secondary sources, by contrast, are works that interpret those events or artistic works; books and edited volumes, journal articles, newspaper editorials, television documentaries, literary criticisms and the like. Finally, tertiary sources are compilations of primary and/or secondary sources, and can include bibliographies, online databases and indexes, and so on. Depending on the topic, researchers should expect to work extensively with all three types of sources.

Secondary sources When launching a new research project, start by perusing the available secondary literature. This essential step in the research process will give you general background information, position you to write a literature review and help you identify your own distinctive contributions to your topic. Fortunately, the Japan Studies field has grown so much over the last half-century that there should be no shortage of materials on your chosen topic in your native language. When searching for secondary sources, pay close attention to recently published articles in peer-reviewed disciplinary and Area Studies journals and books from reputable university and commercial publishers, as these will alert you to the latest scholarly contributions to your topic. The rapid digitisation of scholarly work over the last decade or so will allow you to access most secondary sources from your desk. You can locate most major journals via databases provided by your university library. Be sure to try out the Bibliography of Asian Studies, a comprehensive catalogue of not only books and journal articles in Japan Studies and related

117

Patricia L. Maclachlan fields but also chapters in edited volumes. When searching for journal articles, it is important to consult several databases, since many journals will contract with only one of them. As for books, many—but by no means all—are now available electronically. Figure out which titles you need by consulting the citations and bibliographies of other key works and searching your university library’s holdings. If you are in the U.S. and need a print version of a book that your university does not own, you can borrow a copy via your library’s interlibrary loan (ILL) service. Locate the volume on WorldCat—a bibliographic database covering tens of thousands of libraries around the world—and take note of the volume’s bibliographic information, including the all-important ‘OCLC number’, and record that information on your library’s ILL request form. While you’re on WorldCat, remember to search for other sources as well. ILL services can also be used to access journal articles located beyond your library’s immediate reach; these are normally delivered to the user electronically. All of us are understandably most comfortable working with sources in our native language, but it is essential that you master the relevant Japanese secondary literature as well. And this is where the research process can get tricky. Many university libraries in North America and Europe have substantial holdings in Japanese, but many more do not; once again, ILL will help you fill in any gaps. Some ILL services borrow hard-to-find Japanese-language sources directly from Japan via the National Diet Library (NDL), although this service is still a work-inprogress. The NDL has been gradually digitising its collections, which is promising news for foreign users, but as of the time of my writing this paper, those digitised versions are only accessible in Japan at either the NDL’s headquarters in the Chiyoda ward of Tokyo or one of its branches. You can, however, search the library’s excellent online databases while still abroad. First-time visitors to the NDL must present their passports in order to receive a registered user card. The card is good for three years and can be renewed online. Needless to say, students affiliated with a Japanese university have access to that university’s library. If you wish to search a particular library’s specialised holdings without an affiliation, be prepared to provide a letter of introduction from a librarian at your home institution that includes a list of the sources you wish to consult. Letters of introduction are also helpful when requesting access to corporate, museum and government archives (see Peucker/Schmidtpott/ Wagner, Ch. 9). For a list of Japanese, American and European libraries and archives of specific interest to Japan Studies scholars, you can consult the website of the North America Coordinating Council of Japanese Library Resources (NCC). Once you have located your secondary sources, the next challenge is reading them. Many Japanese scholarly books lack indexes, which means you will have to skim them to find what you need. This is easier said than done when you are researching a new topic with specialised vocabulary. But with a little time and patience, you will soon master that terminology and realise that ‘skimming’ in Japanese has its advantages, since topics can often be gleaned just by glancing at a few key characters.

Primary sources Primary sources are essential to the research process, and many of those materials can now be obtained online. Government White Papers (hakusho), press releases, and advisory committee (shingikai) minutes and reports can be retrieved directly from the websites of the government 118

Chapter 4 How to identify relevant scholarly debates organs that produce the information, and some of it is in English as well as Japanese (see Aizawa/Watanabe, Ch. 9.3). The NDL website includes links to the minutes and other information relating to Diet deliberations since the early Meiji era. Newspaper articles can be accessed directly from newspaper websites (e.g., The Japan Times, Mainichi Shimbun), although a small subscription fee may be required for extensive searching. Also helpful are subscription-supported newspaper databases (e.g. Kikuzō bijuaru: Asahi shimbun kiji dētabēsu, Nikkei Telecom), each of which offers access to a variety of different news sources. And then there is the wealth of hard data that is now available online—not all of it easy to navigate. Sometimes the problem is conceptual or related to measurement. Many social scientists, for example, find out the hard way that the data provided by some government ministries are inconsistent over time, which thus complicates longitudinal analysis. Other statistical sources are simply out of reach. Every seasoned researcher tells stories about learning of the existence of a dataset that seems integral to their work, only to discover that the notion of ‘open access’ has yet to take firm root in Japan (see Reiher/Wagner, Ch. 16). If you find yourself in this position, be persistent. Visit the organisation that owns the data. If possible, arrive armed with a letter of introduction from someone who has a close connection to that organisation. Provide the powers-that-be with a detailed explanation of exactly how you plan to use and disseminate that data. And make follow-up visits, if necessary. Locating qualitative primary sources comes with a different set of challenges, particularly if you are researching a topic that few scholars know anything about. Brace yourself for false starts and roadblocks, but know that once you locate one or two good sources, other discoveries will soon follow. As a PhD student working on the impact of Japanese consumer organisations on the legislative process (Maclachlan 2002), I spent the first few weeks of a lengthy research stint in Japan spinning my wheels. But I soon snared a great interview, which led to several more. One of my interviewees invited me to attend a symposium for consumer activists, which led to invitations to attend more symposia, consumer rallies and even a few demonstrations. Another interviewee introduced me to an archive on consumer-related topics (Kokumin Seikatsu Sentā) close to Tokyo’s Shinagawa Station, where I discovered a treasure trove of back issues of consumer-related newspapers, the in-house histories and newsletters of consumer organisations, memoirs of consumer activists and the like. And repeated interviews with the wonderful activists at the Japan Housewives Association (Shufuren) eventually led to an open invitation to explore the organisation’s in-house archive. Years later, as a student of the history and politics of the Japanese postal system (Maclachlan 2011), I spent dozens of hours in a small archive at the Communications Museum (Tei Park) in the Ōtemachi district of Tokyo. Like many other specialised archives in Japan, this one had yet to electronically catalogue its holdings. Locating relevant material meant searching through printed catalogues and filling out request forms by hand. The work was tedious but well worth the effort. I found invaluable sources that were unavailable through the NDL. And much to my surprise, I got to know many of the other ‘regulars’ at the archive, some of whom turned out to be scholars and retired postmasters who generously agreed to be interviewed. One last point about primary sources: the value of memoirs. While researching the politics of the postal system, I struggled to piece together the storyline of how local postmasters participated in the electoral process; since these activities were illegal, no postmaster would divulge the details to me during interviews. I then happened upon two ‘kiss-and-tell’ memoirs written by disgruntled former postmasters that included extraordinary accounts of the political inner 119

Patricia L. Maclachlan workings of the profession. Based on those memoirs, which corroborated both one another and newspaper sources, I was able to construct a detailed chronology and analysis of an otherwise obscure topic. More recently, during a hunt for materials on Japanese agricultural cooperatives, a Japanese expert on the topic alerted me to the out-of-print autobiography of a co-op leader published during the early 1990s. The memoir filled gaping holes in my knowledge and offered a revealing glimpse into how local co-ops operated a generation ago.

Bookstores in Japan: Some concluding thoughts Finally, every Japan Studies scholar should spend some quality time in Japanese bookstores. Virtually every town and urban shopping street (shōtengai) has a small bookshop that sells the latest issues of news magazines and recent bestsellers on a variety of subjects. Large cities boast of multi-story book meccas the size of small department stores that can take hours to navigate. And there is no shortage of used bookstores, most notably in the fabled Jimbōchō neighbourhood of Tokyo, where history buffs can find all manner of out-of-print multi-volume sets and other rare gems on topics both mainstream and obscure. Visit these bookstores and do it regularly. Find the weekly or monthly news magazines that are relevant to your research interests and read as much as you want while standing in the aisles. Locate the sections that stock the sorts of primary and secondary sources that are relevant to you, and be on the lookout for new acquisitions. If your research budget precludes a purchase, worry not; you can take note of the bibliographic information and locate the volumes at a library. If you can make a purchase, take your acquisition to the bookstore café, dive into it over a cup of tea or coffee and revel in your life as a scholar!

120

4.2 Ambiguity and blurred boundaries: Contextualising and evaluating heterogeneous sources Sonja Ganseforth

In my work on Japanese development policies in Palestine (Ganseforth 2016), I conducted a discourse analysis to examine how different actors tried to frame development policies and to leverage different aspects of an industrial park project according to their respective interests. Looking for sources and information for my research, I mainly encountered challenges in three areas: first, situating my own research in scholarly discussions; second, finding relevant sources in different languages to inform my analysis; and third, identifying relevant discourses and dealing with ambiguous and politicised sources.

Beyond disciplinarity: Positioning one’s own research Having studied two Area Studies, namely Japanese Studies and Arab Studies, but no ‘core discipline’, I had received a comprehensive education encompassing the histories, politics, societies, economies, literatures and languages of my regions of specialisation. My own interests as well as academic courses and advisor choices had led me to some degree of specialisation in Human Geography, discourse analysis and qualitative fieldwork methods. I had also participated in some optional methodological courses and a longer field research expedition, but I was far from the security—or constraints—a traditional disciplinary determination might have given me. There was no obvious choice of theoretical model or methodological approach for me, and all such decisions were in need of prior deliberation, explanation and legitimisation— possibly more than in narrower disciplinary contexts. On location in Palestine, I had found the research topic for my dissertation rather quickly after reading and hearing about a contentious large-scale Japanese development project in the West Bank. However, I needed to identify the theoretical and methodological framework to guide my research as well as the scholarly discourses I wanted to contribute to with my work. After drawing on some introductory works on development theories and consulting with my doctoral advisor, I found the ideas of so-called post-developmentalist thinkers to be the most interesting and promising. Extending Michel Foucault’s discourse analysis (1977), one of the most prominent exponents, Arturo Escobar (1995), criticises development discourses as political constructs, and emulates Edward Said’s (2003) criticism of Western Orientalism in proposing that ‘development’ creates its own ‘other’: the ‘underdeveloped’. He contends that ‘development’ has become a hegemonic discourse in the 20th century and the dominating interpretative framework that allows control over countries and peoples conceived as ‘underdeveloped’ and in need of salvation by Western technology and expertise. This idea of discourses legitimising far-reaching policy interventions, while at the same time depoliticising social and political disparities through technical and economist language, is particularly relevant in the Palestinian 121

Sonja Ganseforth context. Here, the international community has started massive development programmes since the 1990s, and many different truths and narratives justify very diverse political positionalities and programmes. A second major conceptual influence came from my doctoral programme in a structured research training group, which brought together research on issues of globalisation from different disciplinary backgrounds, albeit focusing on perspectives integrating ideas of the so-called spatial turn in the Social Sciences since the late 1980s. Conceptualising space as a contingent product of social, political and cultural construction and constant renegotiation, rather than as a fixed, homogeneous container or arena of events and relations (Massey 2005), offered me a fruitful perspective from which to analyse how spaces are being appropriated by different actors during and by means of a Japanese development project in Palestine. Even if you think that an Area Studies audience is the most likely readership for many of your publications, it is nevertheless important to contribute to scholarly and theoretical discussions that are relevant beyond the regional frame. In order to reach an audience beyond Area Studies, it is therefore advisable to aim for publications in disciplinary journals as well. Treating ‘Japan’ as a case in the study of universal political, social, cultural, economic or historical questions can avoid the pitfalls of solipsistic specialisation without relevance beyond a very specific case—a case that might appear quite exotic, such as a Japanese development project in the Palestinian West Bank.

Literature research: Hunting for relevant multilingual sources Even if I would argue strongly against this apparent exoticism, and even if Japanese development cooperation has constituted a major influence in the Middle East in recent decades, finding and choosing the appropriate sources on Japanese policies in Palestine sometimes proved challenging. While there is certainly no scarcity of literature on development cooperation in general, the sources on Japanese official development policies in the Middle East were very limited. Local case studies mostly focused on Southeast Asia, and there was virtually no research on Japanese aid in Palestine. Of course, this lack of specialised secondary sources was not only a challenge, but also an opportunity for me to fill a considerable research gap with my own work. The first rounds of literature research unearthed mainly English language texts and were limited to the range of possibilities offered by access to a library at a European university as well as various online repositories. I did find an array of studies on more general fields, such as Japanese development cooperation on the one hand, and the problematic aspects of international development and aid in Palestine on the other. However, it was important for me to go beyond the English literature and include Arabic and Japanese language sources. Otherwise, I was bound to ignore crucial parts of the scholarship and some of the discourses probably most relevant to those involved in the process I was studying, even if I also conducted quite a few interviews with Japanese- and Arabic-speaking actors. Obviously, there were language hurdles and long office hours spent cursing two non-European writing systems, but accessibility of sources also proved difficult on a more practical level. Sometimes I asked travelling colleagues and friends to bring back home certain publications that were not available in Germany. Japanese sources—even grey literature and primary sources—were often well accessible via the usual means of libraries, online catalogues, Internet 122

Chapter 4 How to identify relevant scholarly debates repositories, organisations’ websites or media outlets. However, it was not so easy for Arabic language material. In some cases, it proved most useful to personally visit some of the research and activist organisations in Palestine and Israel. They often publish their own research and have additional collections on regional issues for sale in their information centres. Pertinent bookshops in the region were another good resource for some of the relevant publications. I also received a lot of material as well as recommendations directly from people I contacted for interviews or information and advice, which was extremely helpful. In this way, the search for literature became a form of field research as well.

The politics of research: Blurred boundaries and ambiguous sources The overlap of field and literature research correlated with blurred boundaries between primary and secondary sources at times. Occasionally, I had to rely on primary sources such as activists’ reports or official policy statements for information, while at the same time some of the scholarly literature, the secondary sources, had a rather politically charged character. While it would be a fallacy to assume there is an objective, neutral and detached author behind academic publications, I felt that particular care was necessary in evaluating many of the writings I had collected. Analyses by leftist activist organisations from Palestine and Israel, sometimes bordering on the polemic and highly critical of Israel’s occupation and international development policies in the West Bank, painted a very different picture of the industrial park project than studies by semi-independent research institutes from Japan or Israel. Such publications were, of course, important primary sources for my discourse analysis of the power struggle to interpret and frame development policies. At the same time, I sometimes depended on them for information and contextualisation, so it became crucial to verify their depictions as much as possible and never to take them at face value. The problem, however, was not limited to primary sources. The political character of secondary sources became most obvious in analyses of the—still ongoing—Palestinian-Israeli conflict. Studies of international development cooperation in Palestine also tended to be highly critical of either international donors or the Palestinian leadership. At the same time, polemic language and idiosyncratic Palestinian expressions, such as speaking of Israeli settlements in the West Bank only as ‘colonies’ or of Israel as ‘the occupier’ as well as usage of the term ‘apartheid’, occasionally disguised solid academic analyses by Palestinian authors and contributed to their disqualification in international academia—an interesting aspect of discourse politics in itself. It was therefore indispensable to critically scrutinise my sources for their scientific integrity and to consider their respective backgrounds when trying to evaluate the significance of their findings to inform my own research. There were also a number of rather judgemental and biased accounts of Japanese development politics, especially those by some North American scholars since the 1970s. The literature was often limited in its sources, e.g. hardly drawing on any Japanese sources, and some texts seemed to be highly influenced by the growing economic prowess of Japan and the intensifying trade frictions with the United States of America since the 1980s. They focused on critiques of policies that mainly seemed to aid the promotion of Japan’s own export industries, whereas Japanese positions since that time often seemed to take a defensive stance towards these criti-

123

Sonja Ganseforth cal tendencies. It is, of course, imperative to always be aware of the contextuality and historical contingency of academic literature and examine its validity as thoroughly as possible. As it happens, detecting these nuances and overtones proved quite interesting and fruitful for my actual study. I found that many of these secondary sources contributed to discourses that became the object of my analysis, for example the discourse on the self-interest of Japanese aid, or the discourse on apolitical economic cooperation between Israelis and Palestinians as a prerequisite to peace—as well as the fierce rejection of this idea. This way, the literature assumed an ambiguous role: it was still an important source of information, contextualisation and analytical framework, but at the same time some of the secondary literature became a form of primary source for discourse analysis. In the beginning, I worried about the lack of a neat division into primary and secondary sources as I had learned in basic academic education. However, I came to the conclusion that this ambiguity was inevitable and tolerable as long as I was aware of it and dealt with the sources in a transparent and reproducible manner. In putting the results of my research on paper, I therefore felt it was important to make explicit at every point whether I was drawing on a study in order to gain analytical insights and critically engage with their findings on the one hand, or whether I was treating a source as material for discourse analysis on the other.

Engaged scholarship The ambiguous character of some secondary sources had me pondering the positionality of my own writing, especially as my research dealt with highly contentious and sometimes unsettling issues. While assuming ‘objectivity’ or ‘neutrality’ in one’s own work might be as problematic as in other authors’ works, it is of course crucial to adhere to academic standards in terms of rigour, reflexivity, transparency, reproducibility and verifiability and not to sacrifice ethical research for political aims. Nevertheless, I would strongly argue the case for engaged scholarship (see Slater et al., Ch. 16.2) and the responsibility of academics to highlight problematic findings such as social grievances or political injustice, as science is in a potentially powerful position of knowledge production. In fact, many salient researchers and theorists—like the geographer Doreen Massey and the anthropologist Escobar—do not limit themselves to academic audiences, but become vocal advocates of social and political causes. The blurring of roles, activist engagement and collaboration with activist groups does not automatically blur the clarity of academic analysis as long as it is transparent. Good academic research can be interested and engaged.

124

4.3 Doing migration research in Japan: The roles of scholarly literature Gracia Liu-Farrer

This essay shares my experiences of using different types of literature in the process of my research about immigration into Japan. I broadly categorise the relationship between literature review and research as a three-step process of ‘zoning in’, ‘reorienting’ and ‘zoning out’. In addition, I highlight the importance of drawing on scholarly literature in Japanese. Qualitative research is often used for studying emergent social phenomena, for understanding the subjective meanings of particular practices and for investigating the processes of how some events are unravelled. The underlying assumption of qualitative research is that not all concepts pertaining to a given phenomenon have been identified, are fully developed or are sufficiently understood, and further exploration on a topic is therefore necessary to increase understanding (Corbin/Strauss 2008). Our statements about why and how such situations exist or events take place contribute to both the empirical understanding of the phenomenon and theoretical development in the related fields. In other words, qualitative research is often employed to investigate phenomena that we know very little or do not yet have a reliable theoretical statement about. In such a process, literature review recurs at different stages of the research. It provides the initial direction of our research, is used to inform the next stage of data gathering, and offers tools for interpreting our findings and debates to situate our findings in.

Zoning in: Localising the research Since our research interest is often kindled by curiosity or concerns about an emergent phenomenon, an ongoing crisis or intriguing situations, a starting point of one’s research is to find out how much has already been done on the specific subjects and phenomenon. Without surveying the field, it is difficult to know where and how to start, let alone identify gaps in research. The literature one needs to be familiar with needs to cover several areas: the theoretical and conceptual discussions in this field and the research that has been done on a specific topic. In other words, a study needs both theoretical and empirical literature to provide a road map. Some researchers might start from the specific issues and subjects, and work outward to more general conceptual and theoretical discussions. I usually do these different types of literature review simultaneously because empirical case studies allude to different theoretical discussions. Each field has its key questions as well as concepts and theories that are aimed at addressing such questions. That is why one needs to educate oneself about the broader field of one’s study. In my field of international migration studies, some key questions have been: why and how people move; why and how they move to a specific place; how people are related to both the new environment and the places they depart from; and what migration does or means to migrants, those who receive them and those who are left behind. Different disciplines tend to 125

Gracia Liu-Farrer have different issue focuses. After numerous empirical studies in different contexts, a large number of concepts and theories have been developed to answer these questions. This theoretical background might not directly enter into one’s literature review section in the end, but it informs the initial questions of one’s research. Issue and context specific literature, i.e. empirical literature, is an essential part of a literature review. The specific literature for my research is migration into Japan. Since the mid-1980s, increasing contemporary migration into Japan has stimulated interest in this social trend. When I started my research in the early 2000s, there was already a substantial amount of literature in English on immigration into Japan. For example, several anthropological studies had researched ethnic Japanese who were brokered from Latin American countries to be manual labour in the name of ethnic return (Linger 2001; Roth 2002; Tsuda 2003). Studies about Filipino entertainers and marriage migration emerged in the 1990s (Ballescas 1992; Suzuki 2003). There was also a group of Japanese sociologists that, working in the urban sociological tradition, studied immigrant communities and the neighbourhoods they dwelled in. While there was practically no English literature on Chinese people in Japan when I started my research project, Japanese scholars had already produced a number of scholarly publications in Japanese, delineating their migratory trends and describing their life in Japan (Tajima 1998). This body of literature brought me closer to the field. I could see that some of the key issues in immigration in Japan have to do with the dilemma created by a need to import labour and the conservative stance the government took on immigration, which created many informal channels for migration. The nature of Japan as an ethno-nationalist receiving context and the particular migration channels it created necessarily produced different sets of research questions. With the key questions provided by general migration theories and a knowledge of the specific migration conditions in Japan, as a sociologist, I entered my field with a broad interest in understanding how networks play a role in Chinese migration into Japan, and whether and how immigrant communities are formed in Japan, and how the migratory trajectories into a nonimmigrant country unfold. These questions remained the guiding questions throughout my later research, but the specific fieldwork findings produced more specific research outputs.

Reorienting: Continued literature review in the field Qualitative research is such that as soon as we enter the field, we are immediately inundated with new information and unexpected findings. Where we do our research and whom we approach influence what we find out. Although I did not plan to focus on international students initially, I found that among my interviewees the majority of them came as students. Their narratives presented discernibly patterned mobility trajectories. Most noticeable was their double identity as both students and migrant labour. In order to understand such an interesting trend of migration and make sense of my findings, I had to look for literature that specifically examined international students. A quick survey of the existing literature revealed to me that most of the studies about international students before the early 2000s fell within the purview of education research, and were limited mostly to their life on university campus and their education outcomes. At the same time, international migration studies then were primarily about labour migrants. Skilled labour migration had emerged as a trend and attracted concerns about the issues of brain drain. Policymakers and researchers in countries such as Australia

126

Chapter 4 How to identify relevant scholarly debates had begun to treat international students as potential skilled labour migrants. No literature had yet discussed this group as simple migrant labour. However, doing lowskilled labour was an integral part of being international students in Japan. This revelation was exciting because it showed the importance of merging the fields of international education and labour migration. Meanwhile, looking at Chinese student migrants’ career trajectories, I noticed their particular locations in the transnational businesses between China and Japan, and that their activities often involved transnational movements. Another round of focused reviewing of immigrant transnational practices and theoretical discussions about transnationalism necessitated me putting more effort into observing and identifying various patterns of Chinese student migrants’ transnational behaviour. With these newly gained conceptual insights and such empirical evidence, my research focuses therefore shifted to examining patterns of Chinese student migrants’ transnational practices, asking why such phenomena emerged in the context of immigration into Japan (Liu-Farrer 2011). Because of its role in reorienting one’s research and shaping one’s approaches and ultimately findings, the literature one draws on in the process of actual research is the most pertinent to one’s study, and therefore often ends up in the literature review section of one’s academic output.

Zoning out: Situating your study After one has immersed oneself in the field and accumulated many empirical findings, one’s next step is to think about how to reconnect the case study to the literature and situate one’s study in a broader academic field. Aside from offering more knowledge on the specific group and topic in this particular context, one’s research findings should also speak to the larger concerns of the field. In other words, who do you have in mind as your readers and interlocutors beyond the researchers who might be doing the same study? My study obviously added to the research about Chinese student migration into Japan and more broadly about migration in Japan, so anybody who wanted to carry out such studies, I hoped, would read it. At the same time, I saw my study as having broader relatability in the field of Migration Studies, both the topical area of student migration and migrant transnationalism—the two sub-fields of Migration Studies that I had to read up on while doing research in the field. They therefore became the broader literature that I situated my project in. Though I emphasise the need of ‘zoning in’ from and ‘zoning out’ to the larger theoretical conversation, it does not mean that our study is merely an empirical case to substantiate existing theories. For example, transnationalism as a framework with which to examine migrant practices has fundamentally changed how we approach migrant subjects and what we will eventually find out. But our study does not aim to prove that yes, migrants in Japan are indeed transnational and this is the proof. Instead, we want to identify the specific characteristics and patterns of transnational practices migrants in Japan are engaged in, and ask whether they are conditioned by the specific institutional, social and cultural contexts Japan presents; how our case study modifies the theory; and whether and how our findings can be generalised in other contexts.

127

Gracia Liu-Farrer One pitfall of social scientific research, particularly in Migration Studies on Japan, is to impose a concept or theory derived from very different social contexts onto the observations in Japan. Social theories are created out of concrete social contexts. More than other social scientific topics, Migration Studies were borne out of Western, especially traditional settler societies. Some concepts and theories, for example those positing and describing community formations, do not necessarily apply in a Japanese context. One important task for research, and the reason to engage with literature, is not to find a case that fits a particular concept, but to critically engage with the concepts and see in what way our own research, out of particular contexts, might help shape and modify the existing conceptualisation in the field.

Drawing on publications in Japanese Researching events taking place in Japan, one cannot ignore contributions by scholars in Japan and publications in the Japanese language. This is because scholars who are based in Japanese institutions often have more intimate relationships with the subjects they investigate. They have more access to them or better knowledge of the social contexts into which our subjects enter. Overall, Japanese scholars also tend to have more linguistic and cultural advantages in doing such research, and can offer different perspectives. In the field of Migration Studies, Japanese researchers, or researchers who write in Japanese, have accumulated a large body of literature and are particularly prolific in the following subjects: • • • • • • •

immigration and integration-related policy analysis; case studies on international education/international students, including students’ career aspirations and post-graduate mobilities; in-depth anthropological and sociological studies on specific immigrant communities and family formations; neighbourhood research, especially done by urban sociologists, geographers and anthropologists; immigrant children’s educational mobilities; immigrants’ economic practices, including entrepreneurship; Japanese people’s reaction to immigration and immigrants.

Japanese scholarship is a rich source of information and empirical insights. Aside from indepth and sometimes longitudinal qualitative research, Japanese scholars often conduct social surveys of different scopes, something non-Japanese scholars and scholars who are not based in Japanese institutions have difficulty doing. For example, my recent work (Liu-Farrer 2020) draws heavily on Japanese literature on migrant children’s educational situations in Japan (Kaji 2007; Sakuma 2006; Takaya et al. 2015). In addition, social scientists from outside Japan sometimes bring with them concepts and theories that might not apply in Japanese contexts. By looking into Japanese scholarship, one can understand how the real situations are and what concerns the people within Japan. Although it is understood that not all researchers can read Japanese proficiently, ignoring this body of literature is not only a missed opportunity for mining important research outputs, but also impairs the credibility of a ‘Japan Studies’ project. Finally, researchers who write in other lan128

Chapter 4 How to identify relevant scholarly debates guages bear the responsibility of introducing such scholarship. By presenting Japanese research outputs through our non-Japanese publications, we are helping to build a more substantial scholarly foundation for research on Japan.

Concluding remarks This essay shares some of my own experiences with academic literature in the process of doing qualitative research on migration in Japan. One needs to be equipped with an understanding of the key questions, main concepts and theoretical debates in this field as well as specific disciplinary approaches. This will be the foundation one continuously builds on throughout one’s research career. However, specific projects tend to draw on more specific empirical literature. The relationship between research and literature review is a cyclical process, continuing throughout the course of a project, broadly analogised as ‘zoning in’—localising the research, ‘reorienting’—finding theoretical constructs that help you fine-tune your ongoing research and provide interpretive frameworks, and ‘zoning out’—situating your study in a broader range of literature. This essay also emphasises the importance of drawing on the research output of scholarship in Japanese, because there is often more research done in Japan and published in Japanese on the topic we research. Works in Japanese are an indispensable part of scholarly output which we should reference and utilise. However, although other scholars’ research serves as signposts for the terrain one is treading on, one needs to contour one’s own research trajectories.

129

Further reading Callahan, Jamie L. (2014): Writing literature reviews: A reprise and update. In: Human Resource Development Review 13, No. 3, pp. 271–275. Kamler, Barbara/Thomson, Pat (2006): Helping doctoral students write: Pedagogies for supervision. London: Routledge. Oliver, Paul (2012): Succeeding with your literature review: A handbook for students. Maidenhead: McGraw-Hill Education. Onwuegbuzie, Anthony J./Frels, Rebecca (2016): Seven steps to a comprehensive literature review: A multimodal & cultural approach. Los Angeles, CA: Sage. Turabian, Kate L. (2018): A manual for writers of research papers, theses, and dissertations: Chicago style for students and researchers. Chicago, IL: The University of Chicago Press.

References Ballescas, Maria Rosario Piquero (1992): Filipino entertainers in Japan: An introduction. Quezon City: Foundation for Nationalist Studies. Benedict, Ruth (1946): The chrysanthemum and the sword: Patterns of Japanese culture. Boston, MA: Houghton Mifflin. Cheah, Pheng (2001): Universal areas: Asian Studies in a world of motion. In: Traces 1, No. 1, pp. 37–70. Corbin, Juliet/Strauss, Anselm (2008): Strategies for qualitative data analysis. In: Corbin, Juliet/Strauss, Anselm: Basics of qualitative research: Techniques and procedures for developing grounded theory. Los Angeles, CA: Sage, pp. 65–86. Department of Sociology (2012): A guide to writing a senior thesis in Sociology. https://sociology.fas.harva rd.edu/files/sociology/files/thesis_guide_sept_2012.pdf, [Accessed 13 May 2020]. Doi, Takeo (1973): The anatomy of dependence: The key analysis of Japanese behavior. Tokyo: Kodansha International. Escobar, Arturo (1995): Encountering development: The making and unmaking of the Third World. Princeton, NJ: Princeton University Press. Foucault, Michel (1977): Die Ordnung des Diskurses: Inauguralvorlesung am Collège de France— 2. Dezember 1970. Frankfurt: Ullstein. Fujita, Setsuko (2020): Toshokan katsuyō-jutsu: Kensaku no kihon wa toshokan ni. Tōkyō: Nichigai Associates. Ganseforth, Sonja (2016): Besetzungen: Japanische Entwicklungsräume in Palästina. Bielefeld: Transcript. Hane, Mikiso/Perez, Louis G. (2015): Premodern Japan: A historical survey. Boulder, CO: Westview Press. Hamada, Kumiko (2020): Nihon-shi o manabu tame no toshokan katsuyō-jutsu: Jiten, shiryō, dētabēsu. Tōkyō: Yoshikawa Kōbunkan. Hiraoka, Kōichi/Takegawa, Shōgo/Yamada, Masahiro/Kuroda, Kōichirō/Suda, Yūko/Shizume, Masato/ Nishino, Michiko/Kashida, Yoshio (eds.) (2013): Kenkyūdō: Gakuteki tankyū no michi annai. Tōkyō: Tōshindō. Jansen, Marius B. (1992): China in the Tokugawa world. Cambridge, MA: Harvard University Press. Kaji, Itaru (2007): Chūgoku shusshin seito no shinro kitei yōin: Ōsaka no chūgoku kikoku seito o chūshin ni. In: The Journal of Educational Sociology 80, pp. 331–349. Kamler, Barbara/Thomson, Pat (2006): Helping doctoral students write: Pedagogies for supervision. London: Routledge. Kolmer, Lothar/Rob-Santer, Carmen (2006): Geschichte SCHREIBEN: Von der Seminar- zur Doktorarbeit. Paderborn: Ferdinand Schöningh. Kühmstedt, Estella (2013): Klug recherchiert: Für Historiker. Göttingen: Vandenhoeck & Ruprecht. Linger, Daniel Touro (2001): No one home: Brazilian selves remade in Japan. Stanford, CA: Stanford University Press. Liu-Farrer, Gracia (2011): Labor migration from China to Japan: International students, transnational migrants. London: Routledge. Liu-Farrer, Gracia (2020): Immigrant Japan: Mobility and belonging in an ethno-national society. Ithaca, NY: Cornell University Press. Lloyd, Charlotte (2017/18): Literature reviews for Sociology senior theses. https://socthesis.fas.harvard.ed u/files/socseniorthesis/files/pres-litreview.pdf, [Accessed 13 May 2020].

130

Chapter 4 How to identify relevant scholarly debates Maclachlan, Patricia L. (2002): Consumer politics in postwar Japan: The institutional boundaries of citizen activism. New York, NY: Columbia University Press. Maclachlan, Patricia L. (2011): The people’s post office: The history and politics of the Japanese postal system, 1871–2010. Cambridge, MA: Harvard University Asia Center. Massengill, Rebekah (2012): Writing Sociology: A guide for junior papers and senior theses. https://sociol ogy.princeton.edu/sites/sociology/files/soc_ug_writing_guide.pdf, [Accessed 13 May 2020]. Massey, Doreen (2005): For space. London: Sage. Nakane, Chie (1967): Tate shakai no ningen kankei: Tan’itsu shakai no riron. Tōkyō: Kodansha. Roth, Joshua Hotaka (2002): Brokered homeland: Japanese Brazilian migrants in Japan. Ithaca, NY: Cornell University Press. Russell, Bertrand (2004): History of Western philosophy. London: Routledge. Said, Edward (2003): Orientalism. London: Penguin. Sakuma, Kosei (2006): Gaikokujin no kodomo no fushūgaku. Tōkyō: Keisōshobō. Satō, Yoshimaru (2000): Nihon kingendai-shi: Bunken. Tōkyō: Fuyō Shobō. Suzuki, Nobue (2003): Transgressing ‘victims’: Reading narratives of ‘Filipina brides’ in Japan. In: Critical Asian Studies 35, No. 3, pp. 399–420. Tajima, Junko (1998): Shakai toshi: Tōkyō no ajiakei ijūsha. Tōkyō: Gakubunsha. Takaya, Sachi/Omagari, Yukiko/Higuchi, Naoto/Kaji, Itaru/Inaba, Nanako (2015): 2010-nen kokusei chōsa ni miru gaikokujin no kyōiku: Gaikokujin seishōnen no kateihaikei, shingaku, kekkon. In: Journal of Humanities and Social Sciences—Graduate School of Humanities and Social Science, Okayama University 39, pp. 37–56. Toby, Ronald P. (1984): State and diplomacy in early modern Japan: Asia in the development of the Tokugawa bakufu. Princeton, NJ: Princeton University Press. Tsuda, Takeyuki (2003): Strangers in the ethnic homeland: Japanese Brazilian return migration in transnational perspective. New York, NY: Columbia University Press.

131

Chapter 5 How to collect data: An introduction to qualitative Social Science methods Akiko Yoshida

1. Introduction This chapter provides an overview of qualitative data collection methods used in Social Science research primarily for undergraduate students with a limited knowledge of research methods. I first explain different types of methods and discuss what each method is useful for, with a focus on fieldwork, qualitative interviews and observational research. I then present what researchers should consider in selecting data collection methods for their research. I close this chapter by discussing anecdotally what other things should be considered in preparing for qualitative research, drawing on the accounts shared by experienced researchers in the three essays included in this chapter as well as my own experience as a sociologist who conducted interviews for my doctoral dissertation research in Japan.

2. What qualitative data collection methods are there? There are different types of data collection methods within qualitative Social Science research. The most commonly employed methods, which are the focus of this book, are fieldwork, qualitative interviews, observational research and archival research (see Ch. 6–9). These methods are not necessarily mutually exclusive, meaning that there are significant overlaps among them. Researchers may employ just one or multiple data collection methods in their study; they may also combine qualitative methods with quantitative methods or use some other form of mixed methods research or triangulation (see Hommerich/Kottmann, Ch. 10). Fieldwork (see McLaughlin, Ch. 6), or field research, is generally understood as research in which researchers immerse themselves in their research sites, making observations in natural settings, interacting with research participants and interviewing or just talking with them. Despite the fact that fieldwork has been commonly employed by qualitative researchers for over a century, meanings and applications of this concept are not shared among all social scientists (Berg 2007). In this chapter, I keep the conceptual definition of fieldwork rather loose and general and treat it as a combination of qualitative interviews (see Kottmann/Reiher, Ch. 7) and observational research (see Tagsold/Ullmann, Ch. 8).

132

Chapter 5 How to collect data Qualitative interviewing, which may also be called in-depth interviewing, involves asking open-ended questions to research participants in which participants choose their own words in answering the questions or, for example, elaborate on their answers. Researchers ask followup questions to participants to clarify, elaborate or explain their answers. Researchers often make observations on participants’ facial expressions, body language, speed of speech, use of pauses, tone of voice and other non-verbal expressions. This contrasts with survey questionnaires, which typically ask sets of prepared closed-ended questions with prepared answer categories, from which research participants choose answers. Though questionnaires can include open-ended questions, what differentiates qualitative interviewing from questionnaire research is that it aims to obtain in-depth accounts instead of direct answers to particular questions. The driving force behind qualitative interviewing is therefore ‘an interest in understanding the lived experience of other people and the meaning they make of that experience’ (Seidman 2006, p. 9). Another difference from questionnaire research is that researchers may accidentally encounter new themes that were not previously anticipated. As discussed in more detail in Chapter 7, there are different types of qualitative interviews based on the nature of questions (e.g. structured, semi-structured, unstructured and life story/ history interviews), the method of interviews (e.g. face-to-face, via phone, via the Internet) and the number of interviewees (i.e. one-on-one, couple/small group or focus group interviews). Semi-structured interviewing, which utilises a prepared set of open-ended questions called an interview guide, is probably the most common method. Life story/history interviewing, in which researchers ask questions regarding participants’ life experiences or prompt them to tell stories about their lives in the past, is also common (see Kinoshita, Ch. 13.2). These can be semi-structured or unstructured interviews, the latter of which does not involve any prepared questions. Tomoko Hidaka (2010), for example, conducted life history interviews with Japanese elite white-collar salarymen of a large age range—those born between 1925 and 1984—and found interesting differences and commonalities in childhood experiences according to age group. One-on-one interviews, in which a researcher interviews one participant at a time, are the most common forms (see Kottmann/Reiher, Ch. 7). Although couple/small group interviews and focus group interviews are less common, they are nonetheless suited to certain research questions and so worth mentioning here. To study the patterns of and reasoning for the division of household labour among Japanese married couples, Scott North (2009) interviewed married couples together. Couple interview research could inhibit individuals from sharing honest accounts in the presence of the other. In his research, however, most couples ‘used the interview to proclaim, and at times, negotiate, feelings about family, the division of family work, their marriages and self-identity’ (North 2009, p. 31), which generated valuable and rich data for his research. North was also able to observe how married couples interacted with each other, which added more nuance to his data. The ‘couples’ in couple interviews need not be pairs of intimate partners. In her study of care work for elderly parents in Japan, Kristen Schultz Lee (2010) interviewed an 88-year-old mother and her 66-year-old daughter together. The interview of this pair helped Schultz Lee conceptualise the ambiguous sentiments shared among adult daughters: many harboured resentments towards their mothers and had conflictual relationships with them, yet they held a desire to provide care for their parents (Schultz Lee 2010).

133

Akiko Yoshida Focus group interviewing (see Yamaguchi, Ch. 7.2) was originally developed by marketing researchers (e.g. to obtain consumer feedback on products) and adopted by social scientists. Time efficiency is a definite appeal as researchers can interview a fairly large number of participants at once. This method, however, poses challenges such as keeping track of who said what, needing additional staff and possibly video cameras (which can conflict with confidentiality promises) or group dynamics to influence people’s accounts, to list a few. Group settings, however, may facilitate conversations among otherwise reserved individuals. Yoshie Moriki (2017) successfully obtained interesting accounts on sexless marriages and family cosleeping through two focus group interviews with 83 Japanese married women and men. Observational research is a method in which researchers make observations on social processes in natural settings, as little disturbed by researchers as possible. Chapter 8 provides details on what might be called overt participation, in which researchers participate in group activities, communicate openly about their research and build a rapport with the people they observe. Researchers, however, may assume a covert observer role, which means that they do not participate in any activities or reveal to participants that they are observing for a research purpose. You may do research, for instance, by systematically observing public spaces such as anime conventions, restaurants, train stations, protests and baseball games without explicitly interacting with any participants. You are unlikely to learn what these participants feel or think about what they do, but you might gain some understanding of norms, rituals, roles or processes by observing their normal activities as undisturbed by the presence of the researcher as possible. In some cases, researchers may assume a covert participatory role, but participate in activities without disclosing that they are doing research. The researcher can gain richer data in this case, but this approach raises ethical concerns. Researchers should be honest and respectful. Alternatively, researchers may tell the truth, or debrief, during the research or after completing the fieldwork. Whether this role can be carried out ethically is for researchers to consider and for the review board of researchers’ institutions to determine prior to data collection (see Reiher/Wagner, Ch. 16). Yuko Ogasawara (1998), in her study of a Japanese bank, did not disclose her research purpose. She entered the site as a temporary clerical worker instead of a sociology student doing research because Japanese corporations would not have hired her if she had told them her purpose. During her research, she faced a psychological dilemma. Some of the female clerical workers of Japanese companies, or so-called ‘OLs’ or Office Ladies, began to share their dreams and problems with her, but she was unable to reciprocate by sharing her own. In the end, however, she was happy with her decision, as her role as a female employee tremendously helped her gain insights into power and gender dynamics within the Japanese white-collar corporate world (Ogasawara 1989). She used pseudonyms for the bank and individuals to protect their privacy and reputation. Her book aimed for women’s empowerment; thus the OLs would probably have appreciated her work. Whether her choice was ethical, however, is debatable. Other qualitative data collection methods include, but are not limited to, archival research, content analysis, action research, visual ethnography and autoethnography. Archival research (see Peucker/Schmidtpott/Wagner, Ch. 9) entails examining archival records to assess human traces. This is one of a few examples of an unobtrusive strategy—a method that does not intrude into people’s lives—in qualitative data collection methods. Another unobtrusive method —which at the same time is a tool for data analysis (see Chiavacci, Ch. 11)—is content analy134

Chapter 5 How to collect data sis (see Arrington, Ch. 13), which involves a systematic examination of recorded communication (Babbie 2014; Berg 2007), such as written and/or visual materials in newspapers, books, magazines, advertisements, websites, songs, letters, e-mails, speeches, TV shows and more. Ayami Nakatani (2006) used magazines, newspapers and government reports to analyse the emerging concept of ‘nurturing fathers’ around the turn of century in Japan. Justin Charlebois (2013) examined discourses on ‘herbivore masculinity’ in literature on herbivores published in the late 2000s and early 2010s (see Schad-Seifert, Ch. 14.1). Action research is a unique way of conducting research in that research participants actively collaborate in the process of doing research (Berg 2007). Visual ethnography or visual sociology utilises visual images such as photography, video and other media. Photographs or videos can be the data to analyse (Dowdall/Golden 1989; Pink 2007; see also Slater et al., Ch. 16.2). Researchers may also use visual images (e.g. showing old photographs) to solicit responses from participants (Schwartz 1989; Walker/Moulton 1989). In autoethnography, researchers produce ‘accounts that draw upon the experience of the author/researcher for the purposes of extending sociological understanding’ (Sparkes 2000, p.21). This postmodern method challenges traditional scientific methods (Wall 2008) and has been questioned due to its lack of objectivity, for example, but I think it would be interesting if non-Japanese scholars recorded their perceptions, experiences and so forth of living in Japan. I encourage readers interested in this subject to consult the literature on these innovative methods. Thus far, I have used the term data without specifying what it means. Most people think of ‘data’ in relation to numbers, and in quantitative research data collected are indeed numeric or converted into numbers (such as 1 = strongly agree) and analysed statistically (see Hommerich/ Kottmann, Ch. 10). In qualitative research, on the other hand, data is anything but numbers (although, theoretically, numbers can also be part of qualitative research to some degree). Words, stories told in interviews, descriptions of scenes written by researchers, visual images, news articles or diaries, among others, are the data. Qualitative researchers transcribe interviews and write up field notes based on their observations. These are the data to analyse (see Ch. 11–14). Qualitative data collection methods include fieldwork (ethnography, field research), qualitative (in-depth) interviewing, observational research, archival research, content analysis, action research, visual ethnography (visual sociology) and autoethnography.

3. What is each method useful for? Qualitative data collection methods are generally suited to, and excellent for, gaining a deeper, fuller understanding of the social phenomena under study (Babbie 2014). By directly observing natural behaviour/settings and/or asking questions in depth, researchers can, among other things, capture social life as experienced by participants, witness and record how certain events unfold, examine individual behaviour in relation to social contexts, gain insights into reasons for certain actions and behaviour, identify subtle nuances of attitudes, and better un-

135

Akiko Yoshida derstand human subjectivity and meanings attached to words, acts or events (Babbie 2014; Berg 2007; Schutt 2012). The units of social settings appropriate for fieldwork are: 1. practices, 2. episodes, 3. encounters, 4. roles and social types, 5. social and personal relationships, 6. groups and cliques, 7. organisations, 8. settlements and habitats, and 9. subcultures and lifestyles (Lofland et al. 2006). Because qualitative research typically takes an inductive approach (i.e. not to collect data for hypothesis testing), researchers can use this method for exploratory research or dive into areas that are understudied (Schutt 2012; see also Hommerich/Kottmann, Ch. 10). Anne Allison (1994) and Akiko Takeyama (2016) wanted to understand the unknown worlds of, respectively, hostess and host clubs in Japan. In her ethnographic research, Allison (1994) worked as a hostess at a high-class hostess club in Tokyo for four months. Takeyama (2016) visited numerous host clubs and conducted interviews with hosts, managers, owners and clients of clubs over the course of ten years. They obtained rich data on routine practices among hostesses/hosts, patterns of host(ess)–client encounters, roles played on the front stage and their lives behind it, and more. Of course, not everything in human behaviour is physically observable. Researchers may wish to understand what people think, feel, perceive, remember, reason, justify, experienced in the past and so forth. Qualitative interviewing is suited to such purposes. By gaining access to subjective understanding/reasoning, perceptions, sentiments, lived experiences or life stories/histories as told/expressed by research participants, researchers can identify, for instance, the reasons behind people’s actions and life choices (or lack thereof). Ekaterina Hertog (2009) conducted qualitative interviews with unwed mothers in Japan. Her study provides a great insight into why and how these women made the ‘tough choices’—the title of her book—of bearing and rearing children out of wedlock in the Japanese cultural context, which stigmatises unwed mothers. While qualitative interviewing appears to focus on micro-level human interaction/ behaviour, data collected could help identify structural and cultural problems as well. My interview research on singlehood aimed to identify structural constraints that kept many women in Japan from marrying. Never-married women cannot pinpoint the structural causes of their single status, of course. So instead of asking why they remained unmarried, I conducted life history interviews. In other words, by learning about these women’s subjective lived experiences, especially during the prime marrying age, I reconstructed the social and cultural structure they had lived through. Their life stories indicated that, among other things, many employees were spatially segregated by gender and age (which inhibited romantic encounters at workplaces), and that the 1980s economic boom and the 1990s recession had significant impacts on their life paths (Yoshida 2017). When conducting qualitative interviews, researchers do not have to predefine concepts (whereas operationalisation of concepts is crucial prior to data collection in quantitative research). For instance, Barbara Holthus and Wolfram Manzenreiter (see this chapter, Ch. 5.3) questioned the operationalisation of ‘happiness’ in widely used international surveys that rank Japan very low. They conducted interview research in rural Japan using a creative way to assess the sense of well-being. Similarly, concepts such as ‘social class’ are not easy to operationalise and quantify. Years/level of education and household income, which are commonly used as measures of social class in quantitative research, are far from accurate measures. Social class encompasses cultural practices and identity, for example (Bourdieu 1987). To gain a better understanding of the association between social class and reproductive practices in Japan, 136

Chapter 5 How to collect data Aya Ezawa (2010) conducted qualitative interviews with divorced single mothers of various social and class backgrounds. By doing so, she identified different strategies for and perceptions of mothering depending on women’s class origin. Limitation in quantitative data—availability, inadequacy in measurements (see Hommerich/Kottmann, Ch. 10)—is another good reason for conducting qualitative interviews. One example is given by Karen Shire (see this chapter, Ch. 5.2). Her research question on Japanese corporations’ practices of recruitment and employment of transnational workers could not be answered by government statistics, as they were severely limited. Observational research is an excellent method for researchers interested in studying social processes as they happen. Ayumi Sasagawa (2006) observed, as a participant, various community and self-organised mother–preschooler groups. Her data explain how young mothers, in the era of low birth rates, were shaped into ‘good’ mothers through the practices of these groups, some of which were funded by local governments. Ogasawara’s aforementioned study (1998) of a Japanese bank provides detailed (and even humorous) accounts of how OLs expressed their resentment towards undesirable male supervisors through the use of Valentine’s Day chocolate-giving. Her observations challenged the notion of passivity among female clerical workers in Japan. Yet their resistance was certainly not to challenge the power structure. In other words, power dynamics were much more nuanced than male dominance–female subordinance, which was captured thanks to her observational research. One should, however, be aware of the limitations of qualitative methods. The flip side of the advantage of rich data is the limitation in sample size, sample representativeness and, hence, generalisability. The reliability and validity of interviewee accounts and researcher observations can also be questioned, as interviewees may lie, rationalise or remember things inaccurately. Similarly, researchers’ biases or moods (which change) may affect their observations. (There are, however, ways to manage reliability and validity, as subsequent chapters discuss.) Furthermore, ethics guidelines may inhibit data collection and the dissemination of research findings (see Reiher/Wagner, Ch. 16). This being said, rich data collected from qualitative research can outperform these limitations, and its significance in enhancing our understanding of the human social world cannot be overstated. Additionally, it can truly be a rewarding experience to enter a community—the lifeworlds of people—and become connected through the means of conducting research. We just need to keep it in mind, as good social scientists, that our research is never perfect. Key issues Qualitative research allows researchers to explore understudied areas, gain access to people’s thoughts, feelings, life experiences, etc., and observe social processes so that researchers gain a deeper and fuller understanding of the human social world.

4. Which data collection method should you select for your research? The first thing to consider when selecting your data collection method is which methods best answer your research question. The above section should already have given you some tips in 137

Akiko Yoshida this respect. There are other things to consider, such as time and budget, disciplinary norms or ‘personality and personal comfort’ (see this chapter, Cook, Ch. 5.1). Time and budget are determining factors for most researchers. Few researchers can afford to allocate time and other resources, or even have such opportunities, to conduct extensive participant observation research like the aforementioned Allison (1994) and Ogasawara (1989) did. Indeed, these studies are typically doctoral research or research generously funded by fellowships and grants. Your personality and preference also matter. Are you a good listener? Are you observant? Are you patient? These are important factors to take into consideration when selecting your data collection method. Even when researchers attain funding, other factors such as employment or family status can constrain the length and location of research (see McMorran, Ch. 15). This is especially the case for researchers who wish to conduct research in another country. As a Japanese national living and working in the U.S., I do not have the option of doing fieldwork in Japan most of the time. For my singlehood study, which I conducted when I was a doctoral student, I was fortunate because I was able to stay at my parents’ house in Tokyo (with a mother who cooked meals for me!). But I could not stay longer than two months, leaving my young child behind in the U.S. My parents’ residence in Tokyo also meant my research site was limited to the Greater Tokyo Area. In an ideal world, I would have conducted interviews in other urban centres and rural regions, and also with men. With the advent of social media and video calling services, however, qualitative interviews can be conducted from a long distance (see Kottmann/Reiher, Ch. 7). Table 5.1: What to consider in choosing data collection methods Research question

Which method best answers your research question?

Time

How much time can you spend on data collection?

Budget

How much money can you allocate to data collection? Can you obtain funding for your research?

Discipline

Which methods are considered appropriate in your discipline?

Personality

Are you observant, a good listener/communicator?

Access

Do you have access to research sites, interviewees, etc.?

5. How to prepare your research? Once you select your data collection methods, you need to consider and make plans for other things, such as how to choose your interviewees and/or research sites, when and where to meet your interviewees or what to ask or observe. These are discussed in detail in the subsequent chapters for each method. Here, I share some anecdotal episodes on this matter, addressed in the chapter’s three essays and also drawing from my own research experience. Your choice of samples (i.e. your interviewees, sites for fieldwork/observation, etc.) can be theoretical—e.g. should you interview both married and never-married women for your single138

Chapter 5 How to collect data hood research? How about divorced singles? But it can also be made for practical reasons (e.g. my choice to interview women living in the Greater Tokyo Area) or even dictated by factors outside a researcher’s control. Emma Cook (see this chapter, Ch. 5.1) illustrates the latter point well: whereas the selection in her first research (on young men in irregular employment) was relatively easy, her more recent project (on food allergies) was first inhibited by a lack of cooperation from relevant support groups. She eventually turned to one of her colleagues for help, which ended up expanding her research opportunities. Her essay demonstrates not only how research is an evolving process and can be affected by luck, but also how important it is for researchers to build and maintain good social networks. I have already discussed how the time and resources available to researchers may constrain their research projects. Researchers also need to consider such constraints imposed on research participants as well. For instance, it was a big challenge for me to interview women in fulltime career occupations in Japan because they worked incredibly long hours every day! They were often unable to keep promises regarding when they could be available on weekday nights. I am forever grateful to them for sparing their precious weekend hours for my research. I also had to cram in two or more interviews per weekend day despite the undesirability of being unable to take extensive fieldnotes right after each interview. The importance of expressing gratitude for study participation cannot be understated in the cultural context of Japan. Generally speaking, you should prepare a token gift—souvenirs from your country, gift certificates, etc.—and plan to pick up the bills for meals, coffee, etc. consumed at interviews (see McLaughlin, Ch. 6). But depending on the research sites, research participants may feel it is they who should be hospitable to the researcher. Nora Kottmann and Cornelia Reiher (see Kottmann/Reiher, Ch. 7) discuss that research participants may invite the researcher to stay for dinner after having interviews at their home. Kottmann and Reiher recommend accepting such invitations. But as a Japanese native, I have mixed feelings about this. I think acceptance of such invitations is probably appropriate for non-Japanese researchers as they are perceived as ‘guests’ (in Japan). But for Japanese nationals like me, having received such an invitation when I interviewed a married woman at her home, I saw it as a polite and obligated offer—the Japanese feel they should not kick out their guests just before mealtime. I turned it down, and she appeared relieved. This was, however, a different story when an unmarried interviewee suggested we could go out for a drink after the interview. The difference here was that I did not want to inconvenience the first woman. This is quite tricky to navigate, even for a Japanese native like me who is well-versed in Japanese culture. My rule of thumb is to apply the old-fashioned norm: you turn down the offer three times. If they insist after you politely say ‘No’ three times, they must really mean they want you to stay. When interviewing in Japan or with Japanese nationals, which language to use is an important consideration for researchers who are not native Japanese speakers (see Kottmann/Reiher, Ch. 7). Additionally, researchers should be aware that within the Japanese language, there are variations in language according to region (i.e. dialects), gender, ethnicity and social class. If possible and appropriate, it is best to use the language of the people researched, although it is not always appropriate to do so. For instance, it is simply strange if a male researcher uses the feminine form of Japanese when interviewing a woman! If it is not possible to use the same language or dialect, I strongly urge that researchers carefully consider the implications of this. Power structures are embedded in language, and researchers can easily offend or alienate the

139

Akiko Yoshida people researched by using language of a higher status than that used by their research subjects. One of the challenges in conducting qualitative interviews in Japan may be to open up the ‘closed lips’ of the Japanese, who are culturally expected to be humble and suppress their honest feelings. Holthus and Manzenreiter (see this chapter, Ch. 5.3) discuss the effective use of props. I had a similar experience in my research, in which I used cards for different life courses (e.g. married and work full-time, lifetime singlehood, unwed motherhood, etc.). The use of props set up a game-like atmosphere and relaxed the interviewees, which facilitated conversations on a very light note. Easy-to-answer warm-up questions also help. At the beginning of interviews, many of my interviewees said they were just ordinary (futsū) and had little to tell me. After answering the many questions that I posed, they were pleasantly surprised to learn they had plenty to tell about their lives. Key issues When preparing your research, consider the choice of samples, time and resources available, expressing gratitude, language issues and strategies like the use of props or warm-up questions to make people talk.

6. How to position yourself when collecting data? As mentioned earlier and discussed more in subsequent chapters, researchers need to be cognisant of their positions in relation to their study participants, or reflexibility/positionality. Western scholars may take the importance of age hierarchy too lightly (though foreigners may be excused for it). As a Japanese native, this was extremely important to me, especially because I was older than most of my interviewees. I did not want younger interviewees to speak to me in the polite form of Japanese (and establish too formal an atmosphere) and feel pressured to answer all my questions because of my seniority. I managed the situation by deliberately dressing younger and by joking (Yoshida 2017). Inequalities based on other attributes (and their intersection) of course matter as well. Of particular relevance to Japanese (and Area) Studies is the impact of nationality, gender, race and ethnicity. In Japan, Westerners, especially of white race, are regarded as ‘higher’, while other races and ethnicities, including (non-Japanese) Asians, are often placed ‘lower’ in social strata (Yamashiro 2013). In her study of Filipina women in rural Japan, Lieba Faier (2009) discusses her struggle. Filipina women were discriminated against in rural Japan, and there were tensions between the two groups of her interviewees: Filipina women and Japanese nationals. Her whiteness and U.S. nationality (along with her affiliations to prestigious universities) were received with respect by the Japanese participants, but this positioned her in a different status from Filipina women. Some of her Filipina participants questioned her openness towards Filipina women (Faier 2009). Furthermore, researchers’ outsider status (i.e. non-Japanese) could inhibit them from gaining entry to research sites. Yet some researchers may discover their outsider status to be rather advantageous, because Japanese participants may feel more relaxed

140

Chapter 5 How to collect data about sharing their honest feelings with researchers who are not part of their community (Yoshida 2017; see also this chapter, Holthus/Manzenreiter, Ch. 5.3). All of these issues discussed above lead to one important conclusion: qualitative researchers cannot anticipate everything necessary to design research perfectly in advance. In fact, one of the most important guidelines in qualitative research is that researchers remain flexible and go with the flow. Research agendas and designs may change as time passes and researchers gain trust, as Cook points out (see this chapter, Ch. 5.1). Interview questions may be added, modified or dropped based on interviewees’ accounts (Yoshida 2017). Shire (see this chapter, Ch. 5.2) discusses how her research project carried out by a team of researchers with various backgrounds evolved against the backdrop of legal constraints. They had to be flexible and to reframe their research question. This flexibility, however, is an exciting thing about qualitative data collection. Key issues Researchers always have to be aware of their reflexibility and positionality. This includes being mindful of inequalities based on age, gender, nationality, race, ethnicity, etc. Most importantly, researchers have to remain flexible and go with the flow.

7. Summary This chapter has provided an overview of various qualitative data collection methods, focusing on fieldwork, qualitative interviewing and observational research. These methods allow researchers to directly observe natural settings and/or learn about people’s views. Thus, they are suited to gaining a deeper understanding of the subjects under study. Which method best answers the research question should be the most important criterion for choosing the data collection method, but availability of time, resources, access to sites/interviewees and other things such as one’s own personality should be taken into consideration. When collecting data, researchers should remain flexible and be aware of the impact of their own position on the people they are researching.

141

5.1 Participant observation and interviews: Going with the flow and dipping in and out Emma E. Cook

[R]igorous anthropological inquiry [involves] … long-term and open-ended commitment, generous attentiveness, relational depth, and sensitivity to context. (Ingold 2014, p. 384)

Fieldwork is a technique of gathering research material by subjecting the self—body, belief, personality, emotions, cognitions—to a set of contingencies […] such that over time —usually a long time—one can more or less see, hear, feel, and come to understand the kinds of responses others display (and withhold) in particular social situations. (Van Maanen 2011, p. 151)

[E]thnography […] does not claim to produce an objective or truthful account of reality, but should aim to offer versions of ethnographers’ experiences of reality that are as loyal as possible to the context, negotiations and intersubjectivities through which the knowledge was produced. (Pink 2007, p. 22)

Take a look at any qualitative methods research book and you might be overwhelmed at the possibilities on display. Perusing just one book’s table of contents, we are provided with chapters on: interviews, oral history, biographical research, focus groups, narrative research, conversation analysis, participant observation and action research, to name just a few (Seale et al. 2007). This chapter is not an overview of the different methodologies that you could potentially use in your project; it’s not a how-to guide. Instead, I aim to provide some insight into the actual choice and use of methods within my two main research projects to date. The methods we choose are, of course, fundamentally designed to help us answer particular research questions, yet they are also more than that. They are chosen for disciplinary reasons, time and money constraints, as well as personality and personal comfort, among others. As a social anthropologist my go-to methods always involve extensive interaction with people through participant observation and semi-structured and unstructured conversations. As Tim Ingold (2014, p. 386) argues, ‘in the conduct of our research, we meet people. We talk with them, we ask them questions, we listen to their stories and we watch what they do. In so far as we are deemed competent and capable, we join in.’ As you can imagine, the ways in which we do this are contingent: on the locations we choose, the people we meet, the reception we get, and on our gender, age, ethnicity and class, to name just a few (see Goodman, Ch. 1; Kottmann/ Reiher, Ch. 7). These are also circumscribed by the project itself, the research questions and 142

Chapter 5 How to collect data also the amount of time we can spend in any one place at any one time. In this chapter, I will briefly explain the methods I’ve used in two research projects, the reasoning underlying my choices and some of the limitations that have arisen.

Project methods In my PhD research, I set out to explore masculinities and part-time labour in Japan. In particular, I was interested in how men who are not working in a normative way—as salaried fulltime workers—understood, lived and constructed their masculinities. To do this, I had to first ask, what is masculinity? How is it produced, experienced, negotiated and lived? How are labour and masculinities linked in the Japanese context? Why were young men working in the irregular labour market (‘freeters’) regarded so critically in wider society? These are just some of the questions that directed my PhD research and informed the methods I chose. It is quite difficult to ask people directly about their masculinities and the links with their labour practices because such questions are individual and inherently social (Cook 2019). I therefore decided that I needed to come at it from a variety of different angles by engaging primarily in participant observation and semi-structured interviews. Before starting the fieldwork, I made a list of all the possible places that I thought would give me a good chance of meeting and working with a number of freeters: coffee shops, bars (izakaya) or the multi-screen cinema, for example. Having arrived in the city, I also began to look for any support organisations that existed to help young people into work. Ultimately, I did participant observation at two places: a cinema where I worked three to four days a week and a non-profit organisation which helped young people build confidence and find work. After building relationships, I also conducted semi-structured interviews with male irregular workers, asking about their personal histories, experiences, lives and aspirations. Semi-structured interviews were done with people I worked with, and through some limited snowballing (for an in-depth discussion of the methods used Cook 2016, p. 16–22). As the fieldwork progressed, I realised that to understand the moral panic about young male labour practices it was also important to talk more deeply with people who were not working in the irregular labour sector. I therefore began speaking to a range of other people of different ages, genders and occupations that I met through friends and acquaintances. My most recent project on socio-cultural aspects of food allergies in Japan and the U.K. is more methodologically challenging than my PhD research on a number of fronts. It is a crosscultural project that explores experiences of food allergies in Japan and the U.K. Moreover, it explores the experiences of two different groups: specifically, parents of children with food allergies as well as adult individuals who have food allergies. As I am working in a full-time position in Japan, there are constraints on the amount of time and uninterrupted participant observation I can do. Instead, I have to dip in and out of the field. Finding my fields took time and is ongoing. It takes patience and perseverance. I had originally planned to do the Japanese side of the research primarily in Hokkaido where I’m based, and so I reached out to a support group that had an informative website as a starting point. In late 2013, I emailed the representative of the group, explaining my interest. In response to my email, the representative wanted to talk on the telephone, so I gave her my number. When she called she asked about my research, my allergies to fish and nuts, and if I had children (I 143

Emma E. Cook don’t). It was plain throughout the conversation that she didn’t want me to join, and she wasn’t interested in meeting in person. Her last comment was ‘well, if you birth a child you can join (maa, kodomo undara sanka dekimasu)’. After casting around looking for other groups—and struggling to find any—I decided to contact a fellow anthropologist who had done work in Japan on an allergic disease. She very kindly mentioned my research to the director and CEO of the non-profit organisation that she had worked with, and they invited me to attend their summer camp for children with atopic eczema, asthma and food allergies. There began a working relationship with the NPO that lasts to this day. Finding places willing to accept an anthropologist entails a certain amount of luck and the willingness to reach out to others. Having someone to speak on your behalf is also important (Bestor et al. 2003; see Hendry, Ch. 1.3). My colleague was someone who had volunteered and worked with the NPO for years for her research, and she was liked and trusted. Moreover, the director of the NPO has a degree in Sociology and is interested in—and sees the value of—qualitative research. Working with them has also opened up the project in many different directions. For example, they are members of—and represent Japan—at the International Food Allergy and Anaphylaxis Alliance (IFAAA), which brings together the heads of organisations from 21 countries each year. Since 2015, I have been able to attend the meetings with them, meeting the heads of organisations from around the world, and hearing about the various issues and solutions that different groups in different countries have been working on. This has helped me develop relationships with the two charities that work on allergy and anaphylaxis in the U.K., who have allowed me to post information on their social media sites and put out a call for interviewees for each summer that I’m in the U.K. However, it also allows me to begin to trace the ways in which scientific and received knowledge is transmitted in different cultural contexts. In addition to such kinds of participant observation, I have done a range of informal and semi-structured interviews with people who have food allergies and those who are parents of children with food allergies in Japan and the U.K.

Problems and ongoing problem-solving Unlike in my doctoral research—where the bulk of my time was spent on research—working in a full-time position puts constraints on methodologies because of time, research budgets and location. Whilst I originally intended the food allergy research to be a cross-cultural comparative project, the comparison aspect has, to an extent, taken a back seat for the present moment. With time in the U.K. limited to a month of research in the summer, it has not been possible to undertake extended participant observation in addition to travelling around conducting semi-structured interviews. I have therefore focused my attention so far on meeting with, and interviewing, individuals and families. Time issues are solvable by understanding that it’s not possible to do everything at once and by planning methodologies accordingly. There are also logistical elements to consider: the locations of allergy charities in the U.K. are not in major cities, and public transport is expensive and not always particularly reliable. This is solvable by renting an Airbnb close by or a car, but this also depends on finances. In contrast, in Japan I have done more participant observation than interviewing. With advance notice of events and meetings, I can secure lower priced tickets using low-cost air carriers to travel or I can travel by train.

144

Chapter 5 How to collect data The methods I currently use—whilst ostensibly similar on the surface to what I did in my PhD research: participant observation and semi-structured interviews—are different from when I was doing my doctoral research. In that case, I worked alongside irregular labourers three or four days a week in one location, and I interviewed people primarily in the local area. Now I dip in and out throughout the year, attending events when and where I can. Work circumstances, research funding, time and the ability to travel or live in the areas where the research is conducted are thus just as much a factor in the kinds of methods we choose as the research questions we are trying to answer. Moreover, the methodologies themselves are carried out quite differently depending on these factors. It is necessary to remain aware and reflexive of these issues (see Coates, Ch. 3.2), and how this affects the kinds of research and data being collected. Moreover, as time progresses, as trust increases and as the research moves on, methods change. Staying open, flexible and reflexive is therefore important.

General advice Think about what is doable given the time—and research budget, if relevant—that you have. Be patient, reach out to people you know and those you don’t yet know, and don’t give up when the door is closed on you: there will be other ways to do the research. And don’t be afraid for your project, research questions and methodologies to change. Research is a dynamic process that we don’t do alone, so be confident to go with the flow whilst remaining alert and reflexive to what you’re doing and how you’re doing it.

145

5.2 Transnational research in Japan Studies—an oxymoron? Studying cross-border labour mobility in globalising Japanese production organisations Karen Shire

Social surveys and official statistics define their units of analysis as located within specific national state territories. If statistical methods generate data on internationally mobile populations, it is usually in relation to their presence within rather than their mobility between national contexts. Data gathered on an international scale by organisations like the OECD or the UN are mainly compilations of nationally generated statistics and serve a comparative purpose at best. Even in the European Union, where the institutionalisation of a supra-national polity would seem to meet a fundamental condition for the generation of world regional data, Eurostat relies almost exclusively on compiling country-level reports. ‘Methodological nationalism’ continues to infiltrate the world of social indicators, despite almost twenty years of research about transnational mobility. Within this broader scientific situation, a transnational form of Japan Studies would seem to be an oxymoron. The tendency of Area and Regional Studies to focus on a specific country, almost by definition, favours research designs bounded by the national container of a single society. John Urry in his research programme on horizontal mobilities defined the task of transnational research as one of investigating ‘the respective and uneven reach of diverse networks and flows as they move within and across societal borders’ (Urry 2000, p. 18). In this essay, I recount the experience of studying employment practices which extend ‘above and between’ the scale of nation states (Djelic/Quack 2003, p. 305). The focus is on how national Japanese labour market institutions are increasingly reconstituted across Japanese borders, thus linking labour markets in Japan, including those for migrant labour, with supplies of labour and organisations of production in East and Southeast Asia. The project Cross-Border Temporary Agency Work: The Construction of Markets and Transnational Regulation in International Comparison funded by the German Research Foundation (DFG, SH/82/5–1, 2013–2016) aimed to understand how supplies and demands for foreign labour were linked through the employment of migrants in Japan and the mobility of Japanese transnational enterprises abroad in search of labour supplies, especially technical and skilled labour. Developing a multi-site, actor-centred and qualitative research design, the project focused on the role of temporary staffing firms as agents driving the making of crossborder labour markets, and their interactions with local states and regulatory requirements, Japanese client firms and local workers to develop new patterns of transnational labour mobility. In this way, the research questioned whether the Japanese labour market exists as a nationally contained exchange of labour or, at least in part, was developing as a transnational institution.

146

Chapter 5 How to collect data

Transnational research design: Following the staffing agencies and tracing mobility patterns In developing the research design, we found that two sets of national quantitative indicators were important in sketching out the spatial scope and longer-term changes—Japanese immigration statistics and data on Japanese foreign direct investments (FDI). Immigration statistics lent some insight into the increasing dependence of the Japanese domestic labour market on migrant labour (a fact that the secondary literature on migration in Japan has now well established). FDI statistics, however, hardly paid attention to foreign labour forces, local labour recruitment and employment practices, or to how Japanese transnational enterprises structured careers for local foreign labour. The research strategy was to investigate the links between foreign labour in Japan and in their home countries in relation to the globalisation of Japanese production organisations. Our previous research on Japanese temporary staffing agencies suggested that agency recruitment and employment practices were at least one important mechanism of cross-border labour market linkage. The consequence of this research strategy for the research design was the adoption of a multisite approach focusing on the activity of market making. We conceptualised staffing firms as cross-border market makers, and defined the basic research concern as following Japanese temporary staffing firms and documenting their recruitment and placement practices within and outside Japan. Interviews with staffing industry representatives in headquarters in Tokyo revealed a dense network of Japanese temporary staffing firms in all of the major urban centres of countries where Japanese FDI was located in East and Southeast Asia. Vietnam, though not among the countries named in the research proposal, was named early on by staffing firms in Tokyo as an especially important case, leading us to add Vietnamese labour in Japanese enterprises at home and abroad to our research activities. An important prerequisite for implementing research about the transnational practices of private enterprises is gaining research access to the same enterprises in multiple countries. Gaining access was also a challenge in relation to foreign actors in multi-site designs, including experts in foreign governments and the labour movement. The author had long studied Japanese staffing agency employment and could enter into a research collaboration with two scholars based in Japan (Steffen Heinrich and Jun Imai), both of whom had excellent contacts in the industry. A third researcher from Taiwan with quantitative research skills (Chih-Chieh Wang) allowed us to draw on Chinese language sources, to better analyse available official statistics in all the destination countries, and to conduct additional interviews in Taiwan. Finally, a doctoral researcher following up on the study (Aimi Muranaka) extended the research design to study the motivations of Vietnamese workers moving between Tokyo, Hanoi and Ho Chi Minh City. These contours of the research design and the constitution of the research team underline how transnational research requires an international network of researchers and a combination of linguistic and methodological competencies. Following the staffing firms abroad, we soon learned that a number of legal constraints on the industry meant that it could not operate in the same way as in Japan. Only China has a legal framework for temporary agency work, requiring all foreign agencies to enter into joint ventures with Chinese agencies. To work around legal constraints, Japanese staffing agencies abroad changed their business model to what they called shōkai, or introducing labour to clients, rather than dispatching to clients, as they do in Japan. This discovery led the research 147

Karen Shire team to readjust the research approach again, to focus on how staffing firms invented new practices for organising mobility in the context of foreign national legal constraints.

Qualitative interviewing methods The main method for the field study was semi-structured interviews (see Kottmann/Reiher, Ch. 7) with operational managers at branches of temporary staffing firms, with a focus on two specific leading firms with branches in Beijing, Shanghai, Hanoi, Ho Chi Minh City and Bangkok. Starting with corporate headquarters in Tokyo, we obtained contact to the branches in each of these cities. Japanese enterprises abroad are notorious for continuing to do most of their business in the Japanese language. For Japan Studies scholars, this was of course an advantage. Nearly all interviews in China, Vietnam and Thailand were conducted in Japanese. The research design also involved understanding the different sets of national regulatory constraints and how these affected mobility and employment practices. In many cases, the staffing firms could supply us with information, but in Vietnam and Taiwan, it became necessary to gain access to experts in the Labour Ministries. The Vietnamese Labour Ministry staff over the past decade have received policy advice and support from a number of international (for example, the ILO) and foreign government aid organisations, so that an interview could be conducted with some effort in English. Transcripts were completed in the original language by native speakers. Since the research was designed as an inter-regional comparison, with the same research conducted in the European Union, transcripts were eventually summarised as memos in English to support broader comparative analyses of the data. In designing the interviews with staffing agency managers in Japan and abroad, we structured our questions around three main types of horizontal labour mobility—staffing firm recruitment of foreign labour to Japan, their cross-border recruitment from Japan to foreign-based clients, and movements back and forth, including evidence of intra-company transfers. A short project description was sent with each request to interview, where we also named the contact person in Tokyo (or in some cases, other Asian locations). Making our networks of contacts transparent to potential interview partners was a key factor in gaining access. In most cases, it became evident that the contacts checked up on us before accepting our requests for interviews. In one case, where we contacted a trade union organisation in Hong Kong, the contact checked back with organisations close to trade unions in Germany to assess the neutrality of the principle investigator. In Taiwan, the government official contacted for an interview knew of the prior research activities of the Taiwanese team member, and told him when they met that she agreed to the interview because she had heard he was a serious researcher. Research networks and reputations have a very long reach. The interviews always began with a brief description of our research activities and with openended questions about the staffing business in the specific context (What is your business here? How does it compare to what you do in Japan?). Some questions were always repeated, for example, what the interviewees would like to see changed in the regulatory environments in which they operated (in this way, we could gather information on dealing with legal constraints). Towards the end of each interview we asked about the interviewees’ own work biography and, in most cases, these questions generated individual cases of cross-border labour mobility—one of which is presented in the following.

148

Chapter 5 How to collect data

Patterns of cross-border labour mobility: The case of JiaIi Kobayashi We received the contact to Jiali Kobayashi, the operating manager of a major Japanese staffing firm in Beijing, from her colleagues in Shanghai (whose contact we had received from headquarters in Tokyo; here and in all our publications, persons and organisations are anonymised). Jiali is a Chinese citizen, who studied in Japan and is married to a Japanese businessman. She relocated to Beijing with her two children when her husband’s employer, a Japanese automaker, transferred him to China. Unlike many Japanese women, Jiali continued to work full-time after the birth of her son. When her husband’s company announced he would be transferred to Beijing, they also expressed their expectation that Jiali would quit her job and go with him. She agreed, expecting to find a job in Beijing. The company, however, refused to allow this. Instead, they enrolled her in a training programme for wives, to prepare her and the other wives for supporting their husbands’ careers. For the first year, Jiali describes having enjoyed the break from work, but soon began to worry about the effects of a long interruption on her future employability back in Tokyo. Though a Chinese citizen, being from a rural area, she had no idea about how to find a job in Beijing. For that reason, she visited the Japanese staffing company for advice and ended up being employed by them. Within two years, she was promoted to manage the branch. She expected to continue to work for the staffing firm when she and her husband eventually return to Tokyo. Nearly all of Jiali’s clients were Japanese manufacturing firms and trading companies, who wished to employ Chinese citizens who could speak Japanese. Jiali’s biography represented exactly this pattern of mobility. Her work consisted of organising job fairs in cities back in Japan where Chinese students were known to enrol in Japanese universities. Many Chinese students in Japan, like Jiali, are originally from rural regions in China. For the students, finding a job before moving back home to China allows them to exchange their rural hukou (registration) for an urban one, and in this way re-enter China as part of the urban workforce. The problem, however, for Japanese staffing firms was that their Japanese clients were notorious for paying lower wages than Chinese competitors. Jiali recounted how she and her Japanese clients were losing her Chinese recruits to Chinese enterprises. She and others in the industry, however, noticed that Chinese enterprises were beginning to establish branches in Japan (even citing lower labour costs of young college-educated labour in Japan). For the Japanese staffing firm in China, this opened a new opportunity to recruit Chinese with Japanese language skills in China and return them back to Tokyo to work for Chinese rather than Japanese clients. Japanese client enterprises in China, facing high labour turnover (as well as geopolitical uncertainties) were, as a result, beginning to move out of China and into Vietnam. What began as recruitment for foreign invested Japanese enterprises in China evolved, in our observations, into a cross-border labour market.

149

Karen Shire

Conclusion The research experience reported above demonstrates that transnational Japanese Studies is not an oxymoron, if research is designed to move beyond the comfort zone of the ‘national container’ we call Japan, and to follow activities and processes of Japanese people, organisations and institutions across borders. Moreover, without available statistical data at present, transnational research is strongly dependent on qualitative research designs and international collaborations, but also on language-based research skills in Japan Studies.

150

5.3 ‘Bullseye view on happiness’: A qualitative interview survey method* Barbara Holthus and Wolfram Manzenreiter

Since 2014, we have been trying to decipher the social DNA of happiness in Japan. The conference of the same title in 2014 was the starting point for a joint research project focusing on the current state of well-being in Japan, in time developing into a focus on regional differences in well-being, particularly in rural Japan. Even though happiness research has been burgeoning, a comprehensive understanding of the underlying impact factors of subjective happiness is still wanting. The same is true for the definitions and conceptualisations of the terms happiness, life satisfaction and well-being. There also exists a fuzziness in the use of the terms, both in everyday life as well as academic works (Holthus/Manzenreiter 2017; Izquierdo/Mathews 2009; Manzenreiter/Holthus 2017). The relevant literature in this respect is often inconclusive and contradictory. Different disciplinary approaches do not simply highlight different aspects of happiness; they generate different data and draw diverging conclusions. This is all the more the case for happiness research that has been dominated by Psychology and Economics. And quantitative surveys all too often gloss over the cultural particularities of respondents. Large-scale international surveys on happiness that present Japan as unhappy in comparison to other highly developed societies often rely on a single-item measurement of overall happiness or overall wellbeing. The meaning of happiness and its importance in human life as a universal standard is hardly ever questioned. Qualitative studies, which by contrast do indeed take context into consideration, focus on specific sub-groups or singular factors, such as ageing, the workplace, parenthood or political participation, to name but a few. A comprehensive study of Japan that investigates the multidimensionality of happiness and pays tribute to a culture-sensitive understanding of happiness is still lacking. Our interest in rural happiness derives from conflicting views on rural life in contemporary Japan, oscillating between the dystopian vision of the countryside in irreversible economic decline and the nostalgic romanticisation of rural Japan as the repository of traditional values and institutions or an alternative space for individual self-realisation. We have good reason to believe that well-being and life satisfaction are not lower in rural Japan than in urban Japan, but instead that rural life offers a distinctive set of factors that influence residents’ happiness. Our research objective is to find out how these factors account for the subjective sense of wellbeing among types of residents in rural Japan. Because we can access happiness only in terms of conscious reflection and verbal expressions, we were looking for appropriate interview * This chapter is a shortened version of Holthus, Barbara/Manzenreiter, Wolfram (2019): Bullseye view: Developing a sociological method for studying happiness. Tokyo: DIJ Working Paper Series 19/3 and Holthus, Barbara/Manzenreiter, Wolfram (2020): A gameboard approach to studying the multidimensionality of life satisfaction. In: Cieslik, Mark/Hyman, Laura (eds.): Researching happiness: Qualitative, biographical and critical perspectives, Bristol: Bristol University Press.

151

Barbara Holthus and Wolfram Manzenreiter methods in order to address the diversity of factors that rural residents deem to be of significance.

Access to the field and the interviewees A common hurdle in social research is to get access to the field, which is potentially even more pronounced in the case of rural communities and in the case of happiness research, by dealing with the highly private aspects of individual emotions and subjective experiences. What considerably helped us to gain access to people in rural Japan was the extensive ethnographic research history of former members of the Japanese Studies section at the University of Vienna, where we both worked. This research history in the Aso area in central Kyushu dates back half a century. Since 2015, we have carefully rebuilt relations at different levels (with locals, officials, governments and scholars) in the area; numerous visits considerably eased our access to individuals in the region and lay the foundations of trust relationships. Trust and respect are crucial prerequisites of any research method that relies on self-reporting. In that regard, we believe the local population’s familiarity with us through our recurrent presence in the field was advantageous. In addition, our outsider position as visible foreigners was surprisingly helpful. No matter which culture, most people naturally feel reluctant to reveal their feelings about happiness and life satisfaction to strangers in a face-to-face interview. Yet the threshold at which one feels willing to reveal personal sentiments to complete outsiders is actually lower, and we as foreigners were allowed to act naively and address issues that native researchers seem more hesitant to ask. At the time of writing, we have interviewed 30 people from a small rural settlement of about 60 households and the nearby town, with a population below 10,000. In fairly equal measure, we have spoken with newcomers and lifelong residents and with young, middle-aged and elderly men and women. The interviews usually took place at the homes or shops of our interviewees.

Experimental design: Coming to terms with happiness We began our one to 1.5-hour interviews with a short word association part as a warm-up, naming seven terms: happiness (shiawase), sadness (kanashimi), worries (nayami), hope (kibō), success (seikō), anxiety (fuan) and failure (shippai). Our interviewees were asked to talk about anything that came to their minds associated with these terms or how they would define them. A few mentioned at the outset that they had never really thought about happiness, before trying to formulate definitions or providing us with stories of their experiences of happiness. These concrete examples of moments in which they feel or felt happy, sad or worried range from large issues to small incidents. ‘Large’ happiness often appears to be tied to the interrelatedness of the self, in many instances to the well-being of the family. One example of ‘small’ happiness is the appreciation of ‘the sensation in May when rice seedlings are just protruding from the water surface of paddy fields under a clear early morning sky.’ Such an attitude was occasionally summarised as futsū no shiawase, which either signified ‘the usual happiness’ or ‘happiness due to everything being normal.’ This opening to the interview demonstrated that

152

Chapter 5 How to collect data happiness is very much an interpretative process, embedded in social networks and across personal biographies. During the main part of the interviews, we worked with a two-coloured bullseye-shaped chart and tokens to be placed on the board, which to our knowledge is a unique contribution to happiness research (see Figure 5.1). The idea of designing an interactive tool came out of our desire 1. to make the interviews less abstract for the interviewees, 2. to have a hands-on approach as a starting point for detailed questions to follow thereafter, 3. to be able to cover the multidimensionality of happiness and life satisfaction in a comparably short period of interview time, 4. to understand the importance of some elements in people’s lives in relation to other elements in their lives and 5. to make visible and understand clusters of elements, namely how some elements are clustered together on the chart by the interviewees, whereas others were placed on the board at a distance from each other. The chart is divided into a blue and red part, the blue representing the things one is dissatisfied with, red the things one is satisfied with. Circles radiating from the centre weight the significance of the factors, here in the form of round tokens that we provided. The interviewees were instructed to first place a board game figure representing one’s self on the bullseye, and then to arrange the tokens on the board. By selecting and placing tokens on the chart, the interviewees revealed which aspects they think to be important for living a good life, how important these are—also in relation to other factors—and if they are currently satisfied or not with them. Tokens identified as irrelevant were omitted from the board, those for which interviewees felt both satisfaction and dissatisfaction were placed on the dividing line between the red and blue halves. We chose the content of tokens in accordance with our understanding of the state of happiness research. The 30+ terms cover a wide range of aspects between politics, nature, social relations, media usage, work and cultural life. In addition, we offered the interviewees the chance to label blank tokens with other terms. Once all the tokens had been placed, we embarked on an in-depth conversation about the tokens to learn more about the meaning attached to them, the reasons for their placement and their specific realisation in the lives of our interviewees. In wrapping up the interviews, we handed the interviewees a piece of paper, featuring three quantitative survey questions on happiness. All three questions offer the same answer option, a Likert-scale from 0 to 10. We used smileys instead of words to identify the extremes of the scales. Question 1 on the general state of happiness is posed in the same wording as most large-scale surveys in Japan, asking ‘These days, all things considered, how happy do you feel?’ (see Hommerich/Kottmann, Ch. 10). The single-item measurement enables us to tie in with quantitative research, as well as to understand how interviewees rate their state of happiness in comparison to others. It helps us put the overall view of life satisfaction, as extrapolated from the chart and the conversation in the main part of the interview, into perspective. Question 2 ‘In your opinion, what is the ideal level of happiness?’ is important in light of our understanding that Japanese do not universally subscribe to the Western concept of ‘the happier, the better’, but rather to an idea of happiness as fluctuating in a cyclical, sine-wave fashion, in which happiness is a transitory experience and rather based on interpersonal connectedness and balance between the self and others (Uchida et al. 2015). Question 3 ‘How important is it to be happy?’ is an additional attempt to make a cultural argument. While many studies tacitly presuppose the desire for happiness to be universal, and happiness to seemingly be the ultimate goal in life, cultural anthropological as well as cultural psychological research has 153

Barbara Holthus and Wolfram Manzenreiter Figure 5.1: Bullseye chart in practice, three examples

Copyright: authors

demonstrated that happiness is far from being an unquestionable good in every cultural context and, under certain conditions, is even seen as socially undesirable. We found for our respondents levels of happiness throughout the scale of 0 to 10. The majority chose either 5 or 8 or a number between these values. Some chose a 10, and only one person, who was outspokenly unhappy, chose a 0. As much as we see great variability in happiness levels, we also see a wider range of what people consider ideal levels.

Reflections Evaluating the benefits of the methodological approach, we find the quality of our findings as well as the ease of conducting the interviews exceeded our expectations by far. The bullseye chart made our interviewees alive and talkative. Interviewees accepted the outcome as a visualisation of a current slice of their life, and in more than one case they thanked us, saying that they experienced the interviews as enjoyable and a kind of a psychoanalytic session. The visual tool allowed us to get into the complexities of the different aspects and their interrelatedness, which would have otherwise been extraordinarily difficult to extrapolate from our respondents in such comparably short interview times. The visual aid of the chart provides stimulation for high-quality conversation. It displays the different factors in relation to each other and evaluates the strength of these indicators, also in relation to each other. This complexity could not be grasped without any visual tool. The instructions for the exercise are extraordinarily simple, which makes it appealing to many different types of interviewees. We believe this method is applicable and easily adaptable to different social groups, cultural contexts and different research topics. Therefore, we hope that our method will be tried out in many different ways and with different types of interviewees, beyond rural areas or even Japan.

154

Further reading Alasuutari, Pertti/Bickmann, Leonard/Brannen, Julia (eds.) (2008): The SAGE handbook of social research methods. Thousand Oaks, CA: Sage. Amelina, Anna/Nergiz, Derrimsel D./Faist, Thomas/Glick Schiller, Nina (eds.) (2012): Beyond methodological nationalism: Research methodologies for cross-border studies. New York, NY: Routledge. Denzin Norman K./Lincoln, Yvonna S. (eds.) (2005): The SAGE handbook of qualitative research. Thousand Oaks, CA: Sage. Leavy, Patricia (ed.) (2014): The Oxford handbook of qualitative research. Oxford: Oxford University Press. Savin-Baden, Maggi/Major, Claire Howell (2013): Qualitative research: The essential guide to theory and practice. London: Routledge. Silverman, David (2000): Doing qualitative research: A practical handbook. London: Sage. Sprague, Joey/Zimmerman, Mary K. (1989): Quality and quantity: Reconstructing feminist methodology. In: The American Sociologist 20, No. 1, pp. 71–86. DOI: 10.1007/BF02697788.

References Allison, Anne (1994): Nightwork: Sexuality, pleasure, and corporate masculinity in a Tokyo hostess club. Chicago, IL: University of Chicago Press. Babbie, Earl (2014): The basics of social research. Belmont, CA: Wadsworth/Cengage Learning. Berg, Bruce L. (2007): Qualitative research methods for the Social Sciences. Boston, MA: Pearson. Bestor, Theodore C./Steinhoff, Patricia G./Lyon-Bestor, Victoria (2003): Doing fieldwork in Japan. Honolulu, HI: University of Hawai‘i Press. Bourdieu, Pierre (1987): What makes a social class? On the theoretical and practical existence of groups. In: Berkeley Journal of Sociology 32, pp. 1–17. Charlebois, Justin (2013): Herbivore masculinity as an oppositional form of masculinity. In: Culture, Society & Masculinities 5, No. 1, pp. 89–104. Cook, Emma E. (2016): Reconstructing adult masculinities: Part-time work in contemporary Japan. London: Routledge. Cook, Emma E. (2019): Exploring masculinities and labour through intimacy. In: Alexy, Allison/Cook, Emma E. (eds.): Intimate Japan: Ethnographies of closeness and conflict. Honolulu, HI: University of Hawai‘i Press, pp. 246–247. Djelic, Marie-Laure/Quack, Sigrid (eds.) (2003): Globalization and institutions: Redefining the rules of the economic game. Cheltenham: Edward Elgar. Dowdall, George W./Golden, Janet (1989): Photographs as data: An analysis of images from a mental hotel. In: Qualitative Sociology 12, No. 2, pp. 183–213. Ezawa, Aya (2010): Motherhood and class: Gender, class, and reproductive practices among Japanese single mothers. In: Ishida, Hiroshi/Slater, David H. (eds.): Social class in contemporary Japan: Structures, sorting and strategies. New York, NY: Routledge, pp.197 – 220. Faier, Lieba (2009): Intimate encounters: Filipina women and the remaking of rural Japan. Berkeley, CA: University of California Press. Hertog, Ekaterina (2009): Tough choices: Bearing an illegitimate child in contemporary Japan. Stanford, CA: Stanford University Press. Hidaka, Tomoko (2010): Salaryman masculinity: Continuity and change in hegemonic masculinity in Japan. Leiden: Brill. Holthus, Barbara/Manzenreiter, Wolfram (eds.) (2017): Life course, happiness and well-being in Japan. London: Routledge. Ingold, Tim (2014): That’s enough about ethnography! In: Journal of Ethnographic Theory 4, No. 1, pp. 383–394. Izquierdo, Carolina/Mathews, Gordon (eds.) (2009): Pursuits of happiness: Well-being in anthropological perspective. Oxford: Berghahn. Lofland, John/Snow, David/Anderson, Leon/Lofland, Lyn H. (2006): Analyzing social settings: A guide to qualitative observation and analysis. Belmont, CA: Wadsworth/Thompson Learning. Manzenreiter, Wolfram/Holthus, Barbara (eds.) (2017): Happiness and the good life in Japan. London: Routledge.

155

References Moriki, Yoshie (2017): Physical intimacy and happiness in Japan: Sexless marriages and parent-child cosleeping. In: Manzenreiter, Wolfram/Holthus, Barbara (eds.): Happiness and the good life in Japan. London: Routledge, pp. 41–52. Nakatani, Ayami (2006): The emergence of ‘nurturing fathers’: Discourses and practices of fatherhood in contemporary Japan. In: Rebick, Marcus/Takenaka, Ayumi (eds.): The changing Japanese family. London: Routledge, pp. 94–108. North, Scott (2009): Negotiating what’s ‘natural’: Persistent domestic gender role inequality in Japan. In: Social Science Japan Journal 12, No. 1, pp. 23–44. Ogasawara, Yuko (1998): Office ladies and salaried men: Power, gender, and work in Japanese companies. Berkeley, CA: University of California Press. Pink, Sarah (2007): Doing visual ethnography. London: Sage. Sasagawa, Ayumi (2006): Mother-rearing: The social world of mothers in a Japanese suburb. In: Rebick, Marcus/Takenaka, Ayumi (eds.): The changing Japanese family. London: Routledge, pp. 129–146. Schutt, Russell (2012): Investigating the social world: The process and practice of research. Los Angeles, CA: Sage. Schultz Lee, Kristen (2010): Gender, care work, and the complexity of family membership in Japan. In: Gender & Society 24, No. 5, pp. 647–671. Schwartz, Dona (1989): Visual ethnography: Using photography in qualitative research. In: Qualitative Sociology 12, No. 2, pp. 119–154. Seale, Clive/Gobo, Giampietro/Gubrium, Jaber F./Silverman, David (2007): Qualitative research practice. London: Sage. Seidman, Irving (2006): Interviewing as qualitative research: A guide for researchers in education and the Social Sciences. New York, NY: Teachers College Press. Sparkes, Andrew C. (2000): Autoethnography and narratives of self: Reflections on criteria in action. In: Sociology of Sport Journal 17, No. 1, pp. 21–43. Takeyama, Akiko (2016): Staged seduction: Selling dreams in a Tokyo host club. Stanford, CA: Stanford University Press. Uchida, Yukiko/Ogihara, Yuji/Fukushima, Shintaro (2015): Cultural construal of wellbeing: Theories and empirical evidence. In: Camfield, Laura/Glatzer, Wolfgang/Møller, Valerie/Rojas, Mariano (eds.): Global handbook of quality of life: Exploration of well-being of nations and continents. Heidelberg: Springer, pp. 823–837. Urry, John (2000): Sociology beyond societies: Mobilities for the twenty-first century. New York, NY: Routledge. Van Maanen, John (2011): Tales of the field: On writing ethnography. Chicago, IL: University of Chicago Press. Walker, Andrew L./Moulton, Rosalind Kimball (1989): Photo albums: Images of time and reflections of self. In: Qualitative Sociology 12, No. 2, pp. 155–182. Wall, Sarah (2008): Easier said than done: Writing an autoethnography. In: International Journal of Qualitative Methods 7, No. 1, pp. 35–53. DOI: 10.1177/160940690800700103. Yamashiro, Jane H. (2013): The social construction of race and minorities in Japan. In: Sociology Compass 7, No. 2, pp. 147–161. Yoshida, Akiko (2017): Unmarried women in Japan: The drift into singlehood. London: Routledge.

156

Chapter 6 How to do fieldwork: Studying Japan in and outside of Japan Levi McLaughlin

1. Introduction All research necessitates fieldwork, to some degree. Even the most archive-dependent scholar must forge interpersonal connections within hierarchical academic communities in order to gain access to resources (see Schmidtpott/Schölz, Ch. 9.1). The same certainly applies to researchers who are intent on producing original data through participation, observation and interviews. Major fellowships frequently require an introductory letter or statement of support from a Japanese professor, and researchers who spend an extended period of time in Japan may seek a visiting position within a Japanese university, a government agency or a private institution. Ethnographers must gain a place in a Japanese community and build relationships based on trust that they can rely on, potentially for their entire careers. All introductions, no matter the researcher, require careful attention to Japanese conventions. This chapter provides guidelines on how to begin and sustain fieldwork. Ambiguity necessarily surrounds the topic, not least because ‘fieldwork’ and ‘ethnography’ are often treated as synonyms. This is partly due to the fact that, thanks to our social media-saturated world, the classic division between going ‘into the field’ and returning ‘home’ to write up results tends to no longer apply, to the extent that it ever did. The ‘field’ persists as an active presence in the researcher’s life, no matter her location. It remains essential to carry out ethnography in person in Japan to most fully learn about people’s lives, but the researcher will also construct a digital persona to perpetuate ethnography while she is not physically present. Ongoing fieldwork relationships generate exciting possibilities. They also accrue heavy responsibilities. In this chapter, I lay out strategies for initiating new fieldwork projects, ways to keep fieldwork going when you are travelling back and forth to Japan, how the researcher’s identity and disposition shape projects, and ethical concerns. Throughout the chapter, I make general points that apply at all stages. The chapter serves primarily as a how-to guide; please read the essays by Nana Okura Gagné, James Farrer and Hanno Jentzsch in this section for examples of fieldwork in action (see this chapter, Ch. 6.1; 6.2; 6.3).1

1 In a 2010 essay for the Asia Pacific Journal: Japan Focus, I made detailed suggestions on how to begin fieldwork in Japan (McLaughlin 2010), which I update here. For further Japan-specific fieldwork advice, I recommend essays in a 2007 special issue of Critical Asian Studies in which Japan researchers reflect on ethical dimensions of their ethnography (Robertson 2007), and the appendix of the 2019 book Intimate Japan, edited by Allison Alexy and Emma E. Cook, in which the volume’s authors break down difficulties they encountered in the field (Alexy/ Cook 2019; see also Cook, Ch. 5.1; Alexy, Ch. 7.3). Theodore C. Bestor, Patricia Steinhoff and Victoria LyonBestor’s Doing Fieldwork in Japan remains useful (Bestor et al. 2003b).

157

Levi McLaughlin

2. Getting started: Connecting with a Japanese university For novice researchers, the most fruitful introductions to universities are forged via personal relations between their advisors and faculty in Japan. Politely request that your academic advisor write on your behalf to top researchers in Japan who are pursuing projects related to your own. If these connections do not exist, identify leading researchers in your speciality and contact them yourself. You should be familiar with who is publishing the most useful material in your field, but do not be shy about reaching out to knowledgeable scholars for advice on who to approach. Even if your potential Japanese advisor understands English, or another language, it is best to initiate your connection in Japanese. You will most likely need to operate in Japanese in order to carry out your research, and the advisor will need to know that you can function comfortably in her institution.2 Write a message with appropriately formal openings and closings and phrase your requests in honorific language. It is best to ask a native speaker to edit your initial messages to a potential advisor. From your initial introductory email to a prospective academic advisor to your most recent correspondence with a fieldwork interlocutor, ensure that the other person is the focus. By making other people’s priorities your priority, you make yourself their priority. In his pitch for transforming Asian Studies into Asian Humanities, Donald R. Davis (2015) places ‘care first’ as the starting point for scholarly enquiry. Following Emmanuel Levinas (1969), he stresses that because the knowable forever exceeds the known, humility should drive personal relations. Put aside an instrumental approach and instead focus on what you owe as you interact with the people from whom you wish to learn. Before you ask for support, familiarise yourself with the other person’s work by reading at least one of her recent articles and/or books. Ask about this work in your introductory message (see below for written self-introductions and the ‘elevator pitch’). The best possible place to land is in a vibrant seminar (zemi) in which enthusiastic graduate students learn from an engaged professor as they publish and present papers on topics connected to your own. In your correspondence with the professor, do not be afraid to ask about her zemi and what her students research. In the future, these students will be your colleagues, to whom you will send your own grad students and to whom you will owe obligations. Do not alienate them. Turn to them for help. The professor herself will most likely be too busy to spend much time with you, so your fellow students will be key. Again, make the other person your focus. In exchange for their aid, help fellow students by offering to edit English-language abstracts for their articles. Help them identify resources outside Japan by introducing them to conferences and study abroad opportunities, and by making them aware of publications out-

2 There are numerous, ever-updating blogs and social media-based guides for writing formal messages; search for ‘writing letters in Japanese’ for the latest versions. Writing Letters in Japanese, which was published for The Inter-University Center for Japanese Language Studies and is now out of print, remains an invaluable guide (Tatematsu 1993).

158

Chapter 6 How to do fieldwork side Japan. Involve them in your own publications and panel presentations. Be a consistently relevant presence for them. General point #1: Your primary focus should always be the other person.

3. Forging new ties in Japan You learn from people by creating sincere human connections. This requires building relationships based on trust and care. I call the people I spend time with ‘friends’, and I avoid the term ‘informant’. It projects a disingenuous impression of impartial objectivity, and it creates the sense that you are using people simply as data dispensers. Recent decades have seen the American Anthropological Association and other professional organisations adopt the term ‘research participants’.3 In all fieldwork, there is a mutuality to interactions, so ‘participant’, ’interlocutor’, or preferably ‘friend’ serve as honest descriptors. An interlocutor who is initially separated from the researcher by one degree provides an ideal balance of trust and distance. This person will not suffer severe social costs from dealing with you and will therefore be more liable to share information. At the same time, because your relationship was facilitated by a mutual friend, you are sufficiently trustworthy and obligation-laden to deserve the interlocutor’s attention. General point #2: The best interlocutors are often a friend of a friend. How do you meet people who are part of the social group you want to study? You can approach people yourself, but the best way in is to be introduced by someone else. Seek introductions from people you know, or from friends of friends. Introductions are relatively straightforward: contact the person by phone, email or social media. Mention your friends in common, if you have them. Optimally, you will be introduced in person or via message by your mutual acquaintance. Sometimes, even these introductions do not work. Make this your mantra: ‘the worst thing that happens is nothing.’ If you get a negative response, or no response, move on to the next person. It is imperative to remember that every introduction to an individual is necessarily an introduction to a social group. The individual you meet will report on your conduct to other potential research participants. Be very careful: behave well as you learn where this individual is situated within her group, to whom the individual is beholden and how the person’s network operates. Some researchers seek to spend time with politicians and others who are accustomed to scholarly attention. Much of the time, you will be introducing yourself to someone who is not a public person. You may be the first academic they have met. You may even be the first nonJapanese person they have ever spoken with. Meeting you will be a big deal for them, just as

3 ‘Informant’ remains common parlance among fieldworkers, but ‘research participant’ has been official convention for over two decades. See, for example, the Association of Social Anthropologists of the UK and the Commonwealth’s Ethical guidelines for good research practice (1999) and the American Anthropological Association’s Statement on ethnography and institutional review boards (2004).

159

Levi McLaughlin meeting them might be a big deal for you. You must respect this experience by making your research relevant to them. General point #3: An introduction to one person is always an introduction to a social network. Starting out by connecting with someone in a leadership position may yield important insights, but leaders are liable to issue demands downwards to underlings, who may subsequently regard interactions with you as following orders rather than fostering friendly interpersonal relations. Be enterprising: seek to become acquainted with people who carry out the day-to-day operation of what you intend to study and ask them to introduce you to others in their circles. General point #4: Your best contacts are at the bottom or middle of a hierarchy.

4. Pitches You never know when you may meet someone who could be vital to your research. Always be prepared with 1. memorised pitches and 2. written descriptions. Imagine that you are standing next to someone in an elevator going from the twentieth floor to the lobby. This person is vital to your research. You have approximately one minute to introduce yourself and explain what you research in advance of asking for this person’s contact information. Memorise two versions of a Japanese-language ‘elevator pitch’ about your work that you are ready to deliver to anyone who will listen. Each pitch should be three sentences long. •

•

Version 1 should address an academic audience. Sentence 1: Who you are and the topic of your research. Sentence 2: Why the topic is relevant. Sentence 3: Which resources you require to carry out this work. This introduction can segue smoothly into requests for help accessing the resources you named. Version 2 should be pitched at a non-academic audience. Sentence 1: Who you are and where you are from. Sentence 2: What you study, explained in accessible terms. Sentence 3: Why this study is relevant to the person to whom you are speaking. Remember the first general point: the other person comes first. This is particularly important in your relations with non-academics. A fellow researcher will understand that you need resources to carry out your work. A non-academic needs to know why what you do is important, and why it will be beneficial for her to help you.4 General point #5: Always be ready to discuss your research.

4 See also Booth et al. (2016) for advice on crafting elevator pitches.

160

Chapter 6 How to do fieldwork

5. Written self-introductions Prepare two versions of an approximately one-page-long Japanese-language written self-introduction (jiko shōkaisho). Have them ready for dispatch into messages to potential interlocutors. These will be slightly extended versions of your elevator pitches. The first version should suit an academic audience by addressing three main points: 1. Your driving research questions and the contexts in which they emerge. 2. Anticipation of the ever-present ‘so what?’ question by discussing how your work is relevant to your sub-field and to broader scholarly enquiry. 3. Which resources you require to carry out your work. Use this document as a template for messages to potential academic advisors or collaborators. Rework this document as your research develops. The second jiko shōkaisho should be suited to potential interviewees or other parties who may grant you fieldwork opportunities. Non-academic recipients will need to know about you in personal terms. Three points you must address: 1. Who you are, where you came from and why you study Japan. 2. The specific topic you want to learn about and why you became interested in this. 3. Dilemmas you face as you seek to pursue your research and how the person to whom you are writing can help you. Most importantly, you must make your study relevant to the recipient. It is only through making your work personally relevant that you can ask for help. General point #6: Always be ready to introduce yourself in a way that suits your interlocutor.

6. Who you are matters Given the deep social conservatism that prevails in Japan, initiation costs are not evenly distributed. Who the researcher is exerts an inevitable impact on how she makes inroads. I am an able, married, white cisgender male. Because of this, I have not suffered as much as many of my colleagues have from Japan’s innate prejudices—notably its misogyny. If you are young, female, not white, not married, not cisgender, disabled or otherwise do not satisfy a stereotypical foreign professor image, you will most likely confront significant roadblocks. Japan is certainly not unique in this regard, but it is measurably worse than many other places when it comes to prejudicial attitudes towards ethnic, sexual and other minorities.5 However, you will always be able to gain insights others cannot access by virtue of who you are. When it comes to fieldwork, what can you observe that others ignore? Who can you spend time with that the majority of researchers cannot access? What, or who, can you advocate for that others have overlooked or purposefully silenced? Listing entry points you can access thanks to who you are creates a productive to-do list for your research.

5 Numerous measures could be cited, but that Japan ranked 164th in the world in 2019 for numbers of women elected to the national parliament (Inter-Parliamentary Union 2020) and that women occupy fewer than 10% of management positions in Japanese corporations (Kajimoto 2018) indicates a dispiriting imbalance.

161

Levi McLaughlin

7. Go for the ask People are busy. Japanese academics spend absurd amounts of time in boring committee meetings when they are not teaching too many courses or striving to meet daunting writing deadlines. Anyone you seek to interview or spend time with, academic or not, will almost certainly have a packed schedule. As you introduce yourself, you need to move quickly to specific requests. Do your homework. If you are looking for a place in an academic setting, learn as much as you can about the research in the institutions you are contacting. The same goes for non-academic settings: find out what you can about what the people you seek to learn from do, along with their backgrounds and present-day contexts. Preparing this way allows you to put forward a specific request. Most importantly, be clear about time. Specify when you wish to meet and how much time you will require. Be clear about what you want to ask and why your enquiry is relevant—to them and to a reading public. Specify names of documents you wish to read, people you wish to meet, the topics you wish to discuss, activities you wish to observe or join, and other clear-cut information. The more relevant detail you can supply, the clearer your objectives will be. If you are vague about what you want, you will make a new interlocutor nervous; she will have difficulty vouching for you within her hierarchy, and this may compromise her position. Put her at ease by giving your request sharp, easy-to-comprehend contours. Overly ambitious requests may still be refused, but you are more likely to get yes as an answer if your ‘ask’ is unambiguous and well-informed. General point #7: The more specific your request is, the bigger it can be.

8. Fieldwork stuff: Practical considerations Once you have made initial contact and have arranged to meet, prepare the following: •

•

162

Meishi: People in Japan are increasingly likely to share social media contact information, but exchanging meishi, the ubiquitous ‘name cards’, is still standard practice. Have cards printed on high-quality paper stock with English on one side and Japanese on the other. Have a native speaker check your translations and katakana renderings. Include your social media contact information. Kinko’s print centres in Japan have many pleasing meishi templates. Notebooks: You will collect data through sound and video recordings, paper and electronic documents, and other means, but handwritten notes are often your most vivid record. When you are starting to write up, and whenever you are unsure of where to go in your research, return to the notes you took by hand. Always have an extra notebook with you, and never lose your notes!

Chapter 6 How to do fieldwork •

•

•

•

Phone/data: You will need to stay in touch with your interlocutors via LINE, Facebook and other messaging systems they prefer. You will exchange contact information when you meet new people, and you will need to navigate to unfamiliar locations for interviews and participant observation. Purchase a phone, either before you leave for Japan or from a Japanese provider when you arrive. Purchase an external battery pack. If you are not going to be in Japan for long, rent a pocket Wi-Fi device. If you are in Japan for a year or longer, purchase a contract as soon as you receive your alien registration card. Recorder: Some fieldworkers use their smartphones to record, sometimes with an external microphone. I prefer to keep my phone free to exchange contact information, show or take photos, or look things up. Purchase a separate digital high-quality recorder, and always check the battery. Camera: Carry a small digital camera that is separate from your phone. They are preferable for taking shots of documents—you may need to record an entire book on the fly—or for capturing details that phone cameras tend to miss. Photos that make it into publications need to be very high resolution, so non-phone camera shots are often preferable. The omiyage: Do not show up empty-handed to a first meeting. Even if you are visiting a large institution, it can be appropriate to bring an omiyage, or a small gift (see Yoshida, Ch. 5; Kottmann/Reiher, Ch. 7). It is essential that you do so if you are visiting someone’s home. Do not overthink the omiyage. People will tend to prefer food and drink as gifts; avoid burdening them with non-perishable stuff. There may be sweets or other foods from your home country that will work well as omiyage, but it is generally best to err on the side of the familiar. Purchase a pre-wrapped gift in the basement of a well-appointed department or grocery store. Avoid giving alcohol unless you are certain it is appropriate.

9. Fieldwork tips Once you have forged introductions and are ensconced in a fieldwork setting, there is no onesize-fits-all way to carry out your work. However, there are dispositions you should foster, no matter what you study. Even if you are just listening, you should cultivate the understanding that you are there to be useful to the people from whom you are learning. If your role is limited, by choice or necessity, to observing without participating, you must still consider how retaining information about what is happening is useful to your interlocutors. You will learn the most by making yourself a student. Be proactive about this by seeking an apprenticeship role. Volunteer to take on mundane tasks that provide chances for you to spend time with people. Take advantage of skills you possess to join ongoing activities, and treat fieldwork as an opportunity to learn new skills. If you are a student, the people in the society you are learning about will (hopefully) offer straightforward explanations. Do not be afraid to

163

Levi McLaughlin ask obvious or even stupid-sounding questions. If you take on an apprenticeship-type role, these questions are appropriate. General point #8: Always be useful. Asking questions is how we learn. However, the veteran Japan ethnographer Jennifer Robertson put it best: when you are doing fieldwork, try to be Hello Kitty—big ears, no mouth.6 Remember that when you are talking, you are losing a chance to hear what your interlocutors have to say. Ask brief questions when you are lost, but otherwise shut your mouth and listen. Anything you think to ask will be less informative than what the people you spend time with bring up themselves. Be silent, let pauses build and let others fill them (for details on interviewing, see Kottmann/Reiher, Ch. 7). General point #9: Let others do the talking. Because the people you meet are most likely not familiar with scholarly scrutiny, they are not accustomed to being interviewed. This means that if you ask someone if you can interview them, they are liable to refuse outright or offer inhibited responses. Instead, simply ask if you can speak with them. If you suggest that ukagaitai koto ga arimasu ga (‘there are things I would like to enquire about’) and ask for permission to record the conversation, ethical guidelines on interviews still apply (see Reiher/Wagner, Ch. 16), but the conversation will flow more naturally. Seek to record the interview. Explain that you are not a native speaker, that you have difficulty retaining detailed information, or other plausible reasons why you need a recording. Explain that the recording is for your exclusive use and that no one else will hear it, barring a research assistant who may transcribe or translate the interview. Never lie about this. General point #10: Be honest about your research, but do not use the word ‘interview’.

10. Strategies for notetaking and storing data If no one else is writing in a notebook, it may be inappropriate for you to take notes while events are unfolding around you. Employ the principle of kūki o yomeru—‘reading the air’, or going with the flow—to determine if it is all right for you to pull out a notebook. Whether or not you can take notes, turn yourself into an all-senses recording device. Strive to remember every sight, sound, smell and other sensation. As soon as you can, sit down at the computer or with your notebook and record as much as you possibly can of what you remember. If you have time restraints, first take down ‘head notes’, or keywords and phrases—what Robert Emerson, Rachel Fretz and Linda Shaw (2011) call ‘jottings’—that serve as mnemonic triggers for detailed recollections.7 Do your best to write out complete fieldnotes before another bout of fieldwork overwhelms your memory. You will sacrifice sleep to make this possible. Notetaking can serve as a strategy to learn from research participants. Offer your notebook to them to ask them to write the kanji for unfamiliar words, or to sketch maps, institutional hier6 Personal communication, December 6, 2019. 7 See also Van Maanen (2011) and Hammersley/Atkinson (2019) for practical fieldnote-taking advice.

164

Chapter 6 How to do fieldwork archies or other information. Make your notes interactive, and they will become relevant and familiar to the people you spend time with (see Kottmann/Reiher, Ch. 7; Tagsold/Ullmann, Ch. 8). At regular intervals, scan your notebooks at convenience stores, or anywhere you can, and save the scans in at least three places: your computer, on a separate hard drive and in a cloud account. Subscribe to an online backup service that keeps your data safe even if your equipment is lost or destroyed. Compulsively save all of your recorded interviews, documents and every scrap of other digital data in the same three places. Save photos of all the meishi you receive, along with other ephemera. General point #11: Cultivate paranoia about your data.

11. The all-important thank you Invest in the long-term health of your fieldwork by always thanking people for their time (see this chapter, Gagné, Ch. 6.1). At bare minimum, after an interview or meeting people in a fieldwork setting, send them a thank-you message via email or social media. This is one important reason to keep meishi. Consult the aforementioned Writing Letters in Japanese and search online for up-to-date online examples of thank-you messages. Better than electronic messages are printed letters, especially on university or other institutional letterhead. The best thank you is a handwritten letter in Japanese. However, you may not have every person’s mailing address, and time and energy are always limited. No matter what, always make the effort to write a message or phone people to thank them. The arigatō denwa, the ‘thank-you phone call’, will be reported positively. Be confident that the sincerity of your thanks is always more important than how well it is executed in Japanese.

12. Ethnography in and outside Japan via social media As a fieldworker, you create the primary sources you analyse in your academic writing. To enable maximal access for primary source creation, you must create and maintain an online presence. People in Japan are busy, so face-to-face meetings are often difficult to organise, while online discussions remain robust. You will need to join social media interactions. These connections will be all the more crucial when you leave Japan. Social media platforms shift, but at the time of writing the three essential platforms for a Japan researcher are LINE, Facebook and Twitter. These are not simply data caches you can pull from; they are extensions of the people you research. As online ethnographer Kaitlyn Ugoretz (2017) notes, considerable re-evaluation of fieldwork ethics is now underway to determine how researchers should take part in online communities. For some researchers, even lurking in chat groups or otherwise observing online conversations demands privacy consider-

165

Levi McLaughlin ations (King 1996; Roberts 2015). Some advocate for proactive co-creation of online content with interlocutors (Kozinets 2015). In all cases, let the best interests of your interlocutors guide how you interact online.8 LINE groups are topic-specific, subgroup-specific, family-specific—you name it. Request to join them and contribute judiciously, if at all, to ongoing interlocutor conversations. Overall, the stakes for communicating via LINE can remain comparatively low. It essentially functions as a message service akin to WhatsApp, which your fieldwork communities may also use. Your community members will also create Facebook groups in which vigorous discussion is likely to play out. Facebook will become a sensitive platform for many researchers. Friends in Japan will pay close attention to what you post and how people comment on your posts. For your own sake, and to help anonymise people who appear in your writing, adjust your privacy settings so that no one but you can see who your Facebook friends are. Depending on the sensitivity of your research topics, you may choose to prevent people from posting on your wall. Understand that everything you do on Facebook will be observed and that you will be held accountable by interlocutors for what you post. Topics of interest to your research community will flow through Twitter feeds, so keeping up with them will keep you up to speed in your online and in-person conversations. Twitter and Facebook commentary on the topics you research will guide what you read and how you write; this will simply happen, consciously or unconsciously, so you should gird yourself to accept it. Just as you necessarily develop a persona in the field, you will cultivate an online persona. Your online and physically in-person personas will shape one another and navigating between them will become part of your long-term fieldwork. You are a different person online, and it is essential to remember that this is also true of your research participants. Keeping up with digital and in-person versions of the people you spend time with is hugely informative. In terms of your online conduct, a good principle to maintain is less is more. Just as you will do your best work in person shutting your mouth and letting others speak, so too should you think twice before offering opinions online. To retain long-term online connections, offer to take up conversations on more developed or controversial ideas through private messages rather than public posts; your conversation partners will feel more comfortable responding, and you will learn more. Anonymity, already a fraught issue before the advent of social media, has been seriously compromised by digital platforms. There is no reasonable way to keep one’s fieldwork engagements separate from one’s online presence. When it comes to concerns for your research community, you will be the one to make adjustments that best protect them. Keep discussions about your interlocutors within private posts, to the extent that you can. Do not be afraid to erase posts on your pages that may compromise the people to whom you are responsible. If things get too heated, limit your connections to personal messages. This will require more work at your end to keep conversations going but may be worth it when it comes to protecting the people you research, and your own mental health. When it comes time to write up, keep in mind how you have been corresponding on social media platforms. Adjust how you characterise your interlocutors in print in light of how readers may trawl through your posts looking for identity clues. In person and online, as you spend extended periods of time with your interlocutors, you will run up against opinions and activities you find objectionable. People you care about may speak and 8 See also Przybylski (2020) for tips on ‘hybrid ethnography’.

166

Chapter 6 How to do fieldwork act in ways that clash radically with your own values. You will need to devise personal coping mechanisms in order to learn from people on their own terms. Enormous research benefits await those of us who keep our egos in check. A mantra to repeat is: This research is not about me. It is about the people I’m learning from (see general point #1). Remaining task-oriented will prove to be a useful strategy in assessing the costs and benefits of personally challenging fieldwork situations. Ask yourself: What will I gain from continuing to spend time with people whose views challenge mine? What are the personal costs I will accrue in persisting with this work? Are they worth it? You will discover your own threshold. It is likely that this threshold will move during the time you spend with a community. General point #12: Develop compartmentalising strategies.

13. Ethical concerns If the field comes with you, no matter your physical location, so do its ethical obligations. Many institutions require its affiliated researchers to uphold mandated ethical practices, such as the Institutional Review Board (IRB) in the United States. Even if your institution does not maintain formal requirements, you must always keep a final general point in mind: the person you interview is not a document for you to cite. People’s lives will be profoundly affected by your field research. Your presence in their lives makes them highly vulnerable, particularly when it becomes time for you to write up, and you have no way of knowing just how much even a seemingly token interaction may affect them. This means that you must maintain rigorous ethical principles at all times. Here are three principles that should never leave your mind: •

•

•

Always anonymise. Unless the person you are interviewing or learning from has published under her own name or is otherwise a prominent public person, you must change her name in your publications. In particularly sensitive cases, mask her location. Even if your interlocutor declares that she is proud to be known by her own name, anonymise her for her own protection. Never misrepresent yourself. There should never be a situation in which you misrepresent who you are or why you are carrying out your work. In cases where you are confronted with opinions or actions that run contrary to your own principles, you may respond by emphasising the fundamental questions that drive your research. These may include why you wish to understand people who are not well understood, or the importance of a group or practice that has not received sufficient academic attention. Always act with extreme kindness and caution. You cannot know the consequences of your actions. Even if you do everything right, you may be unprepared for the outcome. General point #13: People are not your data.

167

Levi McLaughlin

14. Conclusion To end, I supply an illustrative fieldwork vignette from my research in Japan in 2019 to emphasise that people are not your data. Friday, January 25, 2019: I had an important lesson reinforced for me in Fukushima. I was fortunate to meet up with an interlocutor I will call Mister Arimoto, someone I had met once before. He is a resident of Iwaki, a city on Fukushima’s coast that was devastated by the tsunami on March 11, 2011. Arimoto is a hero. His house was steps from the water, and it was completely destroyed when the waves hit. Knowing that some of his neighbours were elderly and physically disabled, Arimoto waded back into the wreckage repeatedly to carry out those he knew could not make it to safety on their own. He went back and forth for hours and carried out five people. He is justly celebrated in Fukushima. When we met, Arimoto brought with him a large file folder; it is not rare for me to meet people who are eager to show me documentary records of their lives. As he spoke, he pulled out paper after cutting after flyer after paper, creating a mound on the low table before us that flowed onto the floor. What he wanted to show me most was a letter I had sent him in the summer of 2013. After I had returned to the U.S. following my visit to this region, I sent letters on North Carolina State University letterhead to people who were kind enough to meet with me, as a means of expressing my thanks. Arimoto told me that my letter was an important reason he was still alive. He had lost his job as an electrician when the company he worked for collapsed in 2014. He spoke hesitantly and elliptically, but it appears as if he had contemplated suicide after this. He stressed to me repeatedly that, in his lowest moments, he would reread my letter, which was a fairly short but sincere expression of my thanks to him for sharing his experiences and of my commitment to relaying his account beyond Japan. He found a new job in 2016, working on the lighting rig for a hospital helicopter pad, among other meaningful projects, and things had improved for him over the last couple of years. Shortly after the New Year in 2019, Arimoto had a heart attack. He received surgery to clear a ventricle and was urged by doctors to remain in hospital, but he pushed against this, driven by a commitment to meet with me again, knowing I was coming back through Iwaki. He was cleared to leave hospital this past Tuesday, and we met today. He seemed fit, energetic. Driven. The hospital stay had led him to quit smoking, which he seemed particularly grateful for. He was very happy to see me. We spoke for hours. You have to be very, very careful about how you treat people. A casual word or action by you can be treated as a life-or-death matter by the person you intersect with. This is true for all of us. But it is really, really important for those of us who do fieldwork. Some who received a thank-you letter from me may have simply thought it was a nice note and tossed it aside. Arimoto clung to it. Online connections attenuate and enhance these responsibilities, but they are not the only determining factors, or even the most important ones, necessarily, as Arimoto demonstrates. You may have published your book, and your grant funding may have elapsed, but you remain responsible to the people with whom you have forged a relationship. You must live with that commitment.

168

6.1 The cosmology of fieldwork: Relationship building, theoretical engagement and knowledge production in Japan Anthropology Nana Okura Gagné

What is ‘fieldwork’? What is ‘successful’ fieldwork? What does it mean to do fieldwork in Japanese society? Although all fieldworkers encounter unexpected challenges as well as gratification in the field, there is little discussion about what fieldwork actually entails. This chapter aims to demystify this experience by introducing the concept of fieldwork and analysing the question: Why is fieldwork particularly important for those who study Japanese society? Fieldwork is, by definition, ‘working in the field’. Fieldwork in the anthropological sense consists of ‘participant observation’. The core of fieldwork is its function of ‘contextualisation’ and ‘binding’. It contextualises structures/systems and daily routines, as well as your informants and the delicate human relationships and dynamics of power in the field. Fieldwork can also bind you to your informants and the space itself. This can be done through ‘real’ participation by ‘working on the ground,’ immersing oneself fully in participating just as your informants do (Roberson 1998; Roberts 1994; Roth 2002), or being in situations where ‘real’ participation is not possible (e.g. Bestor 2004; Raz 1999; Robertson 1998). This can be conducted through what Theodore D. Bestor variously calls ‘inquisitive observation’, ‘self-consciously work[ing] on a technique for gaining access to people’, ‘parachuting’, ‘dropping into the midst of things from multiple entry points’ (2003, p. 319) and ‘engaging in […] unstructured interviews’ (2003, p. 320). Crucially, such active participation is invaluable for anthropologists to offer grounded knowledge for contextualising and cross-checking other research methodologies such as surveys, formal interviews and archival research, as well as to reveal new avenues of research (Bestor 2003, p. 333).

The practice of fieldwork First, identifying what kind of field site is relevant for examining the particular research questions that you aim to answer is crucial to conducting good fieldwork. Scholars have expressed the importance of ‘serendipity’ as well as well-thought out plans; thus, flexibility in the field is very important. In the course of fieldwork, your research questions may guide you to do research across multiple field sites. Just as your informants are not confined to one space or one institution, researchers now conduct comparative research across different groups or spaces (comparative fieldwork), or follow informants as their lives criss-cross through multiple sites (multi-sited fieldwork). In my own research, to analyse the changing dominant ideologies and the impact of corporate restructuring on Japanese workers, I chose to conduct my fieldwork in three different spaces: 169

Nana Okura Gagné corporations, after-work leisure spaces and the weekend space of hobby activities. I chose these venues in order to fill the gaps in the previous works that had examined either corporate or leisure spheres exclusively. This way, while it was time-consuming and labour-intensive, I was able to understand individual employees more holistically as they moved through different contexts and crafted themselves by navigating through varying spheres and ideologies (Gagné 2010; 2015; 2019, 2020). Moreover, it is hard to understand sensitive corporate tensions and office politics as well as personal desires and life experiences only through corporate contexts (e.g. by working together in the office). This was especially true toward the end of the first decade of the new century, when many corporations had implemented various forms of restructuring. Resonating with Glenda Roberts’ (1994) research on female factory workers and Joshua Hotaka Roth’s (2002) research on Brazilian nikkei migrant workers, in which they pursued intense participant observation as workers themselves, I was also challenged by the actual limited time of interacting with informants when at work. Instead, lunch breaks as well as after work or non-work contexts became indispensable to understanding office politics. At the same time, if I focused only on the leisure spaces of after-work or weekend activities, I would have missed the larger contexts behind individuals’ leisure participation and desires, as work and leisure co-construct each other for many of my informants. As many scholars have pointed out, informants use different contexts to express the various sides of their selves (Kondo 1990), and the power of shifting social contexts can influence the manners in which they present themselves to others (Lebra 2004). Therefore, the venues that one chooses and the ways in which one conducts fieldwork greatly influence the kind of data one may collect. In my own research, getting into corporate sites was the most difficult part. No companies were interested in having a researcher inside their walls, especially during this period of corporate restructuring. Thus, I ended up using a bottom-up approach. I became a volunteer assistant for Company A, while I was introduced to Company B and C through former informants I had gotten to know. For the leisure spaces, I was introduced to different types of hostess clubs by my corporate informants. While the mama-san, the heads of these clubs, were sceptical about my request to do fieldwork, after I had submitted my research proposal and explained my reasons, they accepted having me there. At the same time, getting accepted by fellow hostesses was a completely different matter. At hostess clubs, where I was introduced as a researcher, many hostesses did not think of me favourably, and developing good relationships was challenging. However, it was long-term participant observation at the same site that enabled me to be accepted. We shared late nights together, helping each other through entertaining customers when they felt sick and overly drunk; other times we were harshly scolded by the mama-san. Altogether, this gradual bonding helped to reduce the distance between us. For the communal hobby and volunteer spaces, I was introduced by corporate informants who were participants in those activities and so access was relatively easier. At the same time, in such spaces, participants deliberately avoid talking about private matters, such as corporate affiliation and their occupations, because they were seen as differentiating participants and thus as taboo topics related to the ‘opposing’ corporate spaces. This once again makes it clear how each space has benefits and limits as a field site. Finally, another strength of fieldwork is familiarising oneself with the field and with informants, which can refine your knowledge and understanding about what is most important to your informants. Some of the information can be elicited from interviews, but seeing them in 170

Chapter 6 How to do fieldwork action in particular spaces or networks offers deeper insights and holistic understanding. Also, ‘good’ fieldwork can lead to fruitful interviews in later stages, as individuals become familiar with you as a person in the same field and networks. In my field study of hostess clubs, for the first four months I wrestled with the question of male to female tensions and gendered consumption. However, after my long-term involvement, I came to realise what issues were really important for participants, especially for hostesses as this was fundamentally a workplace for them. In this way, informants in the field will often teach you what is at stake through their actions (as opposed to in interviews). In this sense, fieldwork is not just about knowledge production, but it can also redirect and rewrite our research questions and agenda in the field.

Maintaining relationships Bestor (2003) highlights the importance of cultivating and expanding ‘networks’. Therefore, it is important to immerse oneself in networks and human relations consciously and continuously, which can also speak to post-fieldwork relationships. This leads to an important challenge: How can we maintain relationships with informants? And why might this be particularly important for those who do research on Japanese society? Many fieldworkers have emphasised the importance of visiting sites after long-term fieldwork to correct their early misconceptions, deepening understanding and identifying continuities and changes over time (Bestor et al. 2003a, p. 16). For example, Roger Goodman’s (2003, p. 184) long-term research reveals how the once problematised concept of kikoku shijo (returnee children) underwent a dramatic transformation between the 1960s and the 1980s, and these individuals even became appreciated as an ‘international elite’ due to the larger socio-economic changes of globalisation in Japan. I also found it important to visit the sites in the years after, as I was able to see first-hand the long-term effects of corporate restructuring. In addition, some of my informants’ lives and worldviews greatly changed after being directly affected by corporate restructuring and family medical problems, as well as due to the triple disasters of March 11, 2011. As many research phenomena are constantly subject to change, it is important for researchers to be aware of this and follow up on their field sites and informants in order to avoid being trapped in synchronic and essentialist theorisations. While essentialism is certainly a caveat, there is something particular about ‘doing fieldwork in Japan’. Several scholars have demonstrated the importance of ‘maintaining good relations in the long term’ and ‘the practice of gift-giving’. While this is true for any fieldwork elsewhere, Hardacre (2003, p. 85) calls this ‘one of the obligations’ of doing fieldwork in Japan. Gift-giving is also a marked feature of doing fieldwork, not only for anthropologists but also for historians and Religious Studies scholars. Moreover, having some third person introduce you to your field site can help open doors for your research. However, it is important to note that this involves gratitude as well as obligation, as ‘such introductions involve the standard Japanese cultural practice of borrowing trust from other people’ (Bestor et al. 2003a, p. 14, emphasis added). Thus, researchers should be aware of how their behaviour in the field affects both the researcher’s relationships with their informants as well as impacts on the person who made the initial introduction. To understand the importance of reciprocal and reflexive relations as well as the complexity and ambivalence toward these practices among Japanese people themselves, Katherine Rupp’s 171

Nana Okura Gagné (2003) fieldwork provides valuable insights into how relations within Japanese society are developed and maintained. What undergirds gift-giving practices in Japan is the strength of relationships, gratitude and hierarchy (2003, p. 50). Thus gift-giving is not a practice conducted on a purely ad hoc and case-by-case basis, but rather it is underwritten by ‘symbolic systems’ that integrate it within social life. For instance, through gift-giving, a receiver can discern the depth and commitment of one’s relationship with a giver by understanding the time, effort and monetary value that one puts it into exchanges and the giver’s perception of the strength of the relationship with the other person. Thus, Rupp pushes us to acknowledge that interpersonal exchanges create and perpetuate social relationships, just as rituals work to create and perpetuate social worlds. Moreover, researchers who have conducted fieldwork in Japan can relate how many informants are ‘responsive’: individuals will come to meetings at agreed times, or once institutions grant you access they will prepare well for your visit. However, as ‘reciprocity goes hand in hand with the process of getting along’ (Roberts 2003, p. 311), this also entails reciprocal expectations and responsibilities for the fieldworker. While this can be constraining to a researcher who deals with multiple responsibilities across various groups (Roth 2003, p. 349), it is important for the fieldworker to be aware that they are also becoming part of such cosmological relationships, as well as to recognise the meaning of such involvement in terms of trust and responsibility.

The cosmology of fieldwork The opportunity to do fieldwork is a very rare and special chance for researchers. To take fieldwork seriously, you must place value ‘on a holistic approach to the entities that are the subject of this form of knowledge’, which form ‘the constitution of anthropological theory’ (Stocking 1992, p. 284). Moreover, the time and effort you put into your fieldwork will directly impact the subsequent stages of your research, analysis and writing. As fieldwork in Japan entails becoming part of the cosmology of your informants, complete with its own challenges, responsibilities and gratification, it is a long-term endeavour that can lead to long-lasting and life-changing engagements for you and your informants, as well as to anthropological knowledge and theory production that may shape your personal and professional life for years to come.

172

6.2 A mobilities approach to ‘Japan’ fieldwork James Farrer

I once remarked to a famous ethnographer of Japanese foodways that my project, in contrast to his own, applied a ‘Sinocentric perspective’ to Japanese cuisine, a tongue-in-cheek comment, but also one meant as a challenge to traditional Japanese Food Studies. Historically, not only has China impacted many aspects of Japanese society, including foodways (Cwiertka 2006; Solt 2014), but so have other Asian countries. Beyond considering historical influences, however, my comment about a Sinocentric perspective points to a broader issue in the ethnography of Japan, which is how we capture the cross-border mobility that is inherent in modern society in our fieldwork practice, without giving up on the advantages of traditional placebased or site-based fieldwork. First of all, as a migrant in Japan myself, I see it as important to incorporate the viewpoints of migrants in Japan, especially if we remember that Asian migrants are the majority, with the Chinese the most numerous (Liu-Farrer 2020). However, the mobility intrinsic to modern Japanese society goes far beyond traditional inbound migration stories, and includes the multiple forms of mobilities of Japanese themselves as well as the mobilities of Japanese cultural artefacts and ideas around the world, many of which are not created by Japanese people. A mobilities perspective (Urry 2016), which takes movement as the norm in human societies, therefore represents exciting opportunities, as well as challenges, for conceptualising fieldwork on Japan, including my own studies of Japanese foodways (Farrer et al. 2017). A Sinocentric perspective comes naturally enough to me. I began working in Japan after years of fieldwork in Shanghai, a very cosmopolitan city in China, and this experience has shaped my take on mobilities. First, my study of Shanghai sensitised me to the outsize role a small number of migrants can play in forming a city’s larger identity. Secondly, there was my direct observation of how deeply Chinese people were involved in the spread of Japanese culture. This is evident in Japanese gastronomy around the world. Where would people in Europe, Africa and the Americas get their sushi rolls without all those Chinese and other Asian migrant restaurateurs? A lesser-known story is the boom in Japanese food in China, including Shanghai (Farrer et al. 2017). Here, I use my experiences to discuss some challenges of incorporating a mobilities perspective —especially an Asian centred one—into two fieldwork projects on Japan. The first is an ethnography of a Tokyo culinary community (Farrer n.d.). It documents foodways in a single neighbourhood with a focus on the people who make food and their relationships to customers. On the surface, this is an example of old-fashioned, place-based ethnography of Tokyo, one emphasising the rootedness and particularities of local foodways. Moreover, there is nothing fancy about the methodology, just going from door to door, entering as many places as I can, eating, drinking, observing, interviewing and writing up. For young social scientists, I still recommend this kind of ‘pedestrian’ fieldwork, which though time-consuming allows ethnographers to orient themselves to a scene, to become themselves a local expert. Place173

James Farrer based research is the best way to ground oneself empirically and to learn basic skills of observation and conversation. My traditional approach to gathering data, however, does make use of modern technology. I post most of the interviews I conduct as articles on my website (see above), a practice aimed at reaching a broader audience beyond academia. It also forces me to continually analyse and reflect upon the data I am gathering, generating feedback from informants and readers, but also from the process of writing itself (see Farrer/Liu-Farrer, Ch. 17). My point here is that a mobilities perspective can still inform traditional place-based fieldwork practice. In my door-to-door fieldwork in Tokyo, I learned that many businesses, including izakaya, bars or convenience stores, are run by migrants, even if they are not owned by them. Many are Chinese, though I would not have known this from looking at the shops’ marketing. A more selective sampling of ‘migrant restaurants’ might have missed these assimilated migrant workers in nondescript locales. Every Tokyo neighbourhood is already a migrant neighbourhood (Coates 2015; Liu-Farrer 2020), and we should include these migrant voices in the design of ethnographic fieldwork, through consideration of systematic inclusion of different voices and even interviewing in multiple languages, not only Japanese. Mobility has many faces beyond inbound migration, however. While I was learning about migrants making local Tokyo spaces, I also was learning about Japanese creating Chinese culinary spaces in Japan. One of the most popular types of restaurants in urban Japan is the ‘neighbourhood Chinese’ restaurant. My study revealed a parallel between rural Japanese who migrated to Tokyo to make a living by running Chinese restaurants in the immediate postwar era and Chinese who migrated to Japan as students and opened quite similar establishments serving the daily needs of local residents. Sometimes these migration trends intertwined, as when the son of a Japanese restaurant owner married a Chinese migrant, who continues to work with the son in the Chinese restaurant business. These are migrant businesses, but the more common Japanese-run places would likely be ignored in a study of ‘migrant’ or ‘ethnic’ restaurants. The Chinese-run places conversely might not be considered local Japanese spaces. So the ‘neighbourhood Chinese restaurant’ is an anomalous space that represents two migrant trends, one internal and one international, yet both are also very much part of the local community (Farrer 2018). They are also declining because both types of migrant flows are declining. The new culinary trend in urban Japan may be Nepali-run curry restaurants, and it is a phenomenon driven by the migration industry, not necessarily by changing consumer demands (Kharel 2016). As a fieldworker in Japan, it is not necessary to speak other Asian languages to study such spaces, but it certainly helps. It is advisable to work jointly with people who do, since one fieldworker cannot be expected to do all things well. Language difficulties are endemic to mobility, in fieldwork as in life, and no one should be shy about getting help. I work with my former Japanese teacher, who transcribes and edits the Japanese interviews we do. But I have also had Chinese students work on the project to transcribe Chinese interviews. I do the English translations (and participate in nearly all the interviews). As a graduate student, I did almost all the transcription myself (a good practice), but a benefit of research funding is affording transcription and editorial help. For students without funding, other forms of collaboration are advisable, including co-authoring with other students with compatible skills. In short, there is often a Chinese—or Vietnamese or Nepali—story to tell about Tokyo neighbourhood scenes. Mobility, however, not only concerns the Chinese, Nepalese and Vietnamese, but also the Japanese working in Japan. Indeed, even now, most of the ‘ethnic restaurants’ in 174

Chapter 6 How to do fieldwork my neighbourhood are run by Japanese who make ‘Asian’ cuisine. Many are multiple migrants, some having grown up in other parts of Japan, others having lived abroad for years, either to study cooking or for other reasons. We share their mobility stories on our webpage. In short, a mobilities perspective on Japanese society will be sensitive to the role migrants play in producing Japanese cuisine, and also to the role of migration, travel and overseas study among Japanese producers. A singular focus on ‘migrant cuisine’ in the traditional sense would miss out on these complex stories of mobility and food, which are now part of everyday Japanese life. As in other areas of social life, we need to capture how Japan is a multiply mobile society. A mobilities perspective is even more central to my second ongoing project, a study of the global spread of Japanese cuisine. It is designed as a Global Studies project based largely on ethnographic fieldwork, involving six main members and several research assistants, conducting fieldwork on Japanese cuisine in cities on every inhabited continent. We are looking at the movement of people, foods, recipes, design elements and the media images that accompany modern culinary culture. Our main focus, however, is on the people who make the food, including their mobility paths (Farrer at al. 2017). One thing that became quite apparent to us when we began is that the traditional Japanese Studies approach to Japanese culinary culture had radically undervalued, indeed largely dismissed or ignored, the contributions of non-Japanese actors in producing and recreating Japanese cuisine around the world. This bias is apparent in both the Japanese and Western language writings on the globalisation of Japanese cuisine. There are undoubtedly many reasons for this, but one is that Japanese Studies is simply too focused on Japanese actors. In our research project, we want to focus on the Asians, Europeans, North and South Americans, and Australians who are also producing, transforming and consuming Japanese cuisine. This is their story too. The point is a simple one, but has radical implications for redefining the practice of fieldwork about Japan. Even when studying the spread of Japanese cultural forms around the world, we need to equally examine these ethnically diverse producers and purveyors of Japanese culture and not simply expose them or criticise them for not selling the ‘genuine article’. This approach to fieldwork requires radical rethinking, not only of the basics of demographic sampling in our fieldwork, but also of what ‘Japan’ really is. For us, Japanese cuisine—and by extension ‘Japaneseness’—is now a global cultural product, and not simply the product of Japanese people. We let our informants and their advertising tell us what ‘Japanese food’ is. Ethnographers should not police the boundaries of authentic Japanese culture, though the attempts to do by others are part of what we study. Japan is now itself a global product, and fieldwork about Japan can now happen anywhere. The challenges of doing this type of multi-site fieldwork are large, and I don’t generally recommend multi-city fieldwork to lone young scholars, because one can very easily end up with too little information about too many places. Sometimes, however, such an approach may be fruitful. For me, a very important practical element has been to find expert informants in each location who can not only provide data about a single site—e.g. their own restaurant—but also about the larger context that one is trying to understand but will not have enough time to understand on one’s own. Collaboration with local scholars can also be helpful, though strong relationships with informants in the field are far more important than those with academics. Overall, the mobilities approach to fieldwork lends itself to teamwork, and effective teamwork requires much more than a conference leading to an edited volume. Frequent meetings within 175

James Farrer the group and joint fieldwork with multiple members are essential to make teamwork function. Fieldwork and interviewing together with other team members produces insights that weave together the multi-sited ethnography (see Tagsold/Ullmann, Ch. 8). Writing as a group takes time, but the journey is also the destination.

176

6.3 Building arguments on national policies from everyday observations Hanno Jentzsch

A political scientist by training, I have been interested in the rules, norms and practices that shape social, political and economic life in Japan, and how these rules change over time. But I have always felt particularly vulnerable to the criticism that qualitative researchers are often exposed to in the field of Political Science: How can you build an entire argument on the personal observations and accounts you gathered in a small number of field sites? While I have almost always collected my empirical data locally, i.e. on the municipal level or below, the arguments I have used this data for are mostly aimed at understanding changes to and the stability of broader institutional arrangements operating at the level of the nation state, or even addressed the logic of institutional change in general. The distance between the field site and the analytical objective, however, is a methodological challenge. But as a non-Japanese political scientist working on Japan, another challenge comes up almost inevitably: How can researchers collect reliable qualitative data despite their socio-cultural, linguistic and—given that research stays are probably limited—spatial distance? I want to address both of these potential sources of insecurity with some examples from my own experience.

Studying national policies on the local level First, I want to emphasise the value of frequent, everyday observations in a particular locality and over extended periods of time as opposed to the use of brief field trips and interviews alone. When I collected data for my dissertation on institutional change in the Japanese agricultural sector, I relied on interviews (both semi-structured and unstructured), informal conversations and participatory observation in two main field sites for three months and three and a half months respectively. While this may appear like two brief visits from an anthropologist’s viewpoint, it was a long and ethnographic engagement with the field sites from a Political Science perspective. I went into the field to understand the local manifestations of the agricultural reform process in Japan and—in a more abstract sense—the trajectories of Japan’s extensive agricultural support and protection regime, a multi-faceted arrangement of formal and informal institutions. The opportunity to collect data during two research periods helped to offset the rookie mistakes I made during interviews, particularly in the first research period. For example, the way I asked farmers to explain local rules and practices and their relations with their hamlet or their local cooperative were far too direct and too abstract to yield useful information. Moreover, my questions were based on the academic literature I had read on rural social organisation in Japan, some of which were written decades ago and typically derived from research in ricegrowing localities (Dore 1978; Fukutake 1980). In my first field site, however, the Kōfu Basin 177

Hanno Jentzsch in Yamanashi Prefecture, rice cultivation played virtually no role. During my second research stay, this time in a rice-growing locality in Shimane Prefecture, I learned to ask less directly, and I did actually encounter elements of traditional social organisation. Moreover, I realised (in hindsight) an issue that linked both sites despite their differences, namely the role of social ties and certain unwritten rules and practices for exchanging farmland. The finding that farmland exchange was organised very differently in both field sites later became a cornerstone of my thesis’s first argument. I argued that the pace and the direction of structural change in the Japanese agricultural sector are crucially shaped by the way in which local social ties, norms and practices are integrated into distinct local manifestations of farmland governance. Brief research trips for interviews with local stakeholders, the collection of (official) documents and expert interviews with policymakers at the national level could not have yielded the same results. Far more important than the respective administrative techniques that were employed to govern the exchange of land in each field site or than statistics on how much land was obtained by professional farms were factors less visible at first sight. They included personal ties between certain farms and the administration, actual practices underlying the exchange of land (sometimes quite different from what the formal contract would stipulate) or the spatial boundaries of local social networks. In my second research period, for example, I lived in one part of the city and commuted to another part, which was my actual field site. Although both localities are only separated by a river and have belonged to the same municipality since 2011, the organisation of farmland exchange was (and still is) very different. Over the three and a half months of my stay, I realised that every morning I would cross the bridge into a distinct social world. In this world, every interviewee knew everybody I had interviewed so far, and everybody I was going to interview in the future. I did not find this density of social ties on the other side of the river. This experience became the foundation of my second argument that the Japanese agricultural support and protection regimes consist of a multitude of distinct local agricultural regimes, in which national policies intersect with local social ties and norms in ways that produce very different local interpretations of the same national reform process. This does not mean that interviews with local officials and the collection of statistics, documents or presentations were not useful—they were, and so was studying the details of the national policy process. It means, however, that political scientists can profit from immersion to build arguments on abstract social phenomena such as institutional change: getting to know faces and family ties, oral local history, attending drinking parties and regular visits to the local bar; this is all fieldwork. The very mundane act of ‘being around’ has practical benefits: I found out which festivals to attend, what stakeholders I could meet there or where I could get myself a car over drinks and snacks in the local bar. But more importantly, these practical benefits can quickly turn into crucial building blocks for actual research. Informal conversations are immensely useful for finding out what the critical issues are in a certain place: which topics come up routinely, and which topics tend to be avoided? Casual conversations can also be helpful in order to double-check issues that came up in more formal interview situations. The information local officials present during formal interviews on a certain local policy will likely become far more colourful when discussed with random strangers over a beer. Even if the information sourced at the bar might be flawed, it provides the raw material for better, more grounded interview questions.

178

Chapter 6 How to do fieldwork Thus, although political scientists have often been sceptical towards immersion and ethnographic methods and rather adhere to the positivist notion of keeping maximum distance from the ‘object’ under study (Schatz 2009), I have found ethnographic methods to be immensely useful for understanding how local actors translate abstract policies into practice. The flourishing discourse on practice theory also highlights the value of studying local practices as a means to understand large social phenomena, such as organisations or institutional arrangements (Nicolini 2017). The empirical focus on the emergence and reproduction of certain practices does not imply that the arguments derived from these micro-level observations have to get stuck in ‘localism’—i.e. the descriptive study of local social phenomena with no intention to analytically connect them to other phenomena across time and space (Nicolini 2016). Much to the contrary, the ‘interest for the mundane and often unsung details of organisational life’ can—when carefully contextualised—very much become the foundations from which to ‘explain social matters’, and how they change over time through practice (Nicolini 2017, pp. 22–23).

Open-ended field research When a researcher is applying qualitative methods of data collection, fieldwork never ends, and certainly not after the interview is over. Yet, fieldwork in Japan often ends quite abruptly, when the research funding runs out or the semester starts. Leaving the field site, however, is only the beginning of other equally important tasks: contextualising the data and keeping in touch with research participants. In fact, I did substantial parts of my work on local agricultural regimes on my desk. After I had established farmland exchange as a crucial process, national statistical data helped me to put the diverging outcomes of farmland exchange in my field research sites in context. Moreover, the perspective on land exchange (as opposed to the broader topic of agricultural reform) unlocked a new batch of policies that I had only touched on superficially before. Some of the most important details on my field sites—for example, changes regarding their administrative and cooperative boundaries—were easily accessible online, but only once I knew what to look for. Such contextualisation helped to reduce the distance between the local field sites and the analytical objective of studying the trajectory of the agricultural regime as a whole. Moreover, it shaped the profile of so-called ‘small cases’ to strengthen the empirical data and helped me to identify these additional cases, including the right interview partners, and to formulate questions. The foundation of previous ethnographic field research has allowed me to conduct short field trips to additional locations in a much more productive way, since I could rely on a concrete set of questions focused on certain topics and stakeholders. While there will rarely be the time again for longer periods of field research once a dissertation is finished, such additional research trips and revisiting the original research sites whenever possible (including, of course, the bar) help to keep the data alive. Beyond occasional visits, little gaps and statistical details can be asked about via email, Facebook or text messages. I found that all of these channels were important ways to stay in touch with the field site—whether to ask if everybody is OK after the area was hit by an earthquake, or to quickly follow up on a particular detail you need to know right now to write up an article. Not least, keeping in touch via various channels helps to build stronger arguments on change. Every abstract policy change at the national level becomes easier to understand when

179

Hanno Jentzsch traced down to the local context of a familiar field site, and an email to the right person among your local contacts can be sufficient to unlock this specific knowledge. In that sense, to paraphrase Levi McLaughlin (2010), not only is all research in Japan fieldwork—keeping in touch with your field site is also the foundation for continuing research on Japan from abroad.

180

Further reading Alexy, Allison/Cook, Emma (2019): Reflections on fieldwork: Exploring intimacy. In: Alexy, Allison/Cook, Emma (eds.): Intimate Japan: Ethnographies of closeness and conflict. Honolulu, HI: University of Hawai‘i Press, pp. 236–260. Bestor, Theodore C./Steinhoff, Patricia G./Lyon-Bestor, Victoria (eds.) (2003): Doing fieldwork in Japan. Honolulu, HI: University of Hawai‘i Press. McLaughlin, Levi (2010): All research is fieldwork: A practical introduction to studying in Japan as a foreign researcher. In: The Asia-Pacific Journal: Japan Focus 8, No. 30. https://apjjf.org/-Levi-McLaughlin/ 3388/article.html, [Accessed 14 August 2020]. Robben, Antonius/Sluka, Jeffry A. (eds.) (2012): Ethnographic fieldwork: An anthropological reader. Hoboken, NJ: Wiley-Blackwell. Robertson, Jennifer et al. (eds.) (2007): Politics and pitfalls of Japan ethnography: Reflexivity, responsibility, and anthropological ethics (Special Issue). In: Critical Asian Studies 39, No. 4. Reiher, Cornelia (ed.) (2018): Fieldwork in Japan: New trends and challenges (Special Issue). In: ASIEN— The German Journal on Contemporary Asia 149.

References Alexy, Allison/Cook, Emma E. (eds.) (2019): Intimate Japan: Ethnographies of closeness and conflict. Honolulu, HI: University of Hawai‘i Press. American Anthropological Association (2004): Statement on ethnography and institutional review boards. www.americananthro.org/ParticipateAnd Advocate/Content.aspx?ItemNumber=1652, [Accessed 20 April 2020]. Association of Social Anthropologists of the UK and the Commonwealth (1999): Ethical guidelines for good research practice. www.theasa.org/ downloads/ethics/Ethical_guidelines.pdf, [Accessed 20 April 2020]. Bestor, Theodore C. (2003): Inquisitive observation: Following networks in urban fieldwork. In: Bestor, Theodore C./Steinhoff, Patricia G./Lyon-Bestor, Victoria (eds.): Doing fieldwork in Japan. Honolulu, HI: University of Hawai‘i Press, pp. 315–334. Bestor, Theodore C. (2004): Tsukiji: The fish market at the center of the world. Berkeley, CA: University of California Press. Bestor, Theodore C./Steinhoff, Patricia G./Lyon-Bestor, Victoria (2003a): Introduction: Doing fieldwork in Japan. In: Bestor, Theodore C./Steinhoff, Patricia G./Lyon-Bestor, Victoria (eds.): Doing fieldwork in Japan. Honolulu, HI: University of Hawai‘i Press, pp. 1–17. Bestor, Theodore C./Steinhoff, Patricia G./Lyon-Bestor, Victoria (eds.) (2003b): Doing fieldwork in Japan. Honolulu, HI: University of Hawai‘i Press. Booth, Wayne/Colomb, Gregory G./Williams, Joseph M./Bizup, Joseph/Fitzgerald, William T. (eds.) (2016): The craft of research. Chicago, IL: University of Chicago Press. Coates, Jamie (2015): ‘Unseeing’ Chinese students in Japan: Understanding educationally channelled migrant experiences. In: Journal of Current Chinese Affairs 44, No. 3, pp. 125–154. Cwiertka, Katarzyna J. (2006): Modern Japanese cuisine: Food, power and national identity. London: Reaktion Books. Davis, Donald R. (2015): Three principles for an Asian Humanities: Care first … learn from … connect histories. In: The Journal of Asian Studies 74, pp. 43–67. Dore, Ronald (1978): Shinohata. A portrait of a Japanese village. New York, NY: Pantheon Books. Emerson, Robert M./Fretz, Rachel I./Shaw, Linda L. (eds.) (2011): Writing ethnographic fieldnotes. Chicago, IL: University of Chicago Press. Farrer, James (n.d.): Nishiogiology. www.nishiogiology.org, [Accessed 16 October 2019]. Farrer, James (2018): The decline of the neighborhood Chinese restaurant in urban Japan. In: Jahrbuch für Kulinaristik—The German Journal of Food Studies and Hospitality, No. 2, pp. 197–222. Farrer, James/Wang, Chuanfei/Wank, David/de Carvalho, Mônica R./Vyletalova, Lenka/Hess, Christian (2017): Japanese culinary mobilities research: The globalization of the Japanese restaurant. In: Foods & Food Ingredients Journal Japan 222, No. 3, pp. 257–266. Fukutake, Tadashi (1980): Rural society in Japan. Tokyo: Tokyo University Press.

181

References Gagné, Nana Okura (2010): The business of leisure, the leisure of business: Rethinking hegemonic masculinity through gendered service in Tokyo hostess clubs. In: Asian Anthropology, No. 9, pp. 29–55. Gagné, Nana Okura (2015): Romance and sexuality in Japanese Latin dance clubs. In: Ethnography 15, No. 4, pp. 446–468. Gagné, Nana Okura (2019): Neoliberalism at work: Corporate reforms, subjectivity, and post-Toyotist affect in Japan. In: Anthropological Theory. DOI: 10.1177/1463499618807294. Gagné, Nana Okura (2020): Reworking Japan: Changing men at work and play under neoliberalism. Ithaca, NY: Cornell University Press. Goodman, Roger (2003): The changing perception and status of Japan’s returnee children (kikokushijo). In: Goodman, Roger/Peach, Ceri/Takenaka, Ayumi/ White, Paul (eds.): Global Japan: The experience of Japan’s new immigrant and overseas communities. New York, NY: Routledge, pp. 177–195. Hammersley, Martyn/Atkinson, Paul (2019): Ethnography: Principles in practice. Abingdon: Routledge. Hardacre, Helen (2003): Fieldwork with Japanese religious groups. In: Bestor, Theodore C./Steinhoff, Patricia G./Lyon-Bestor, Victoria (eds.): Doing fieldwork in Japan. Honolulu, HI: University of Hawai‘i Press, pp. 71–88. Inter-Parliamentary Union (2020): Global data on national parliaments. data.ipu.org/, [Accessed 20 April 2020]. Kajimoto, Tetsuji (14.09.2018): Women in management at Japan firms still a rarity: Reuters poll. www.re uters.com/article/us-japan-companies-women/women-in-management-at-japan-firms-still-a-rarity-reuter s-poll-idUSKCN1LT3GF, [Accessed 20 April 2020]. Kharel, Dipesh (2016): From Lahures to global cooks: Network migration from the western hills of Nepal to Japan. In: Social Science Japan Journal 19, No. 2, pp. 173–192. King, Storm A. (1996): Researching Internet communities: Proposed guidelines for the reporting of results. In: The Information Society 12, No. 2, pp. 119–128. Kondo, Dorinne K. (1990): Crafting selves: Power, gender, and discourses of identity in a Japanese workplace. Chicago, IL: University of Chicago Press. Kozinets, Robert V. (2015): Netnography: Redefined. London: Sage. Lebra, Takie Sugiyama (2004): The Japanese self in cultural logic. Honolulu, HI: University of Hawai‘i Press. Levinas, Emmanuel (1969): Totality and infinity: An essay on exteriority. Pittsburg, PA: Duquesne University Press. Liu-Farrer, Gracia (2020): Immigrant Japan: Mobility and belonging in an ethnonationalist society. New York, NY: Cornell University Press. McLaughlin, Levi (2010): All research is fieldwork: A practical introduction to studying in Japan as a foreign researcher. In: The Asia-Pacific Journal: Japan Focus 8, No. 30. https://apjjf.org/-Levi-McLaughlin/ 3388/article.html, [Accessed 14 August 2020]. Nicolini, Davide (2016): Is small the only beautiful? Making sense of ‘large phenomena’ from a practicebased perspective. In: Hui, Alison/Schatzki, Theodore A./Shove, Elizabeth (eds.): The nexus of practices: Connections, constellations and practitioners. London: Routledge, pp. 98–113. Nicolini, Davide (2017): Practice theory as a package of theory, method and vocabulary: Affordances and limitations. In: Jonas, Michael/Littig, Beate/Wroblewski, Angela (eds.): Methodological reflections on practice oriented theories. Cham, CH: Springer International Publishing, pp. 19–34. Przybylski, Liz (2020): Hybrid ethnography: Online, offline, and in between. London: Sage. Raz, Aviad (1999): Riding the black ship: Japan and Tokyo Disneyland. Cambridge, MA: Harvard University Press. Roberson, James (1998): Japanese working class lives: An ethnographic study of factory workers. London: Routledge. Roberts, Glenda S. (1994): Staying on the line: Blue-collar women in contemporary Japan. Honolulu, HI: University of Hawai‘i Press. Roberts, Glenda S. (2003): Bottom up, top down, and sideways: Studying corporations, government programs, and NPOs. In: Bestor, Theodore C./Steinhoff, Patricia G./Lyon-Bestor, Victoria (eds.): Doing fieldwork in Japan. Honolulu, HI: University of Hawai‘i Press, pp. 294–314. Roberts, Lynne D. (2015): Ethical issues in conducting qualitative research in online communities. In: Qualitative Research in Psychology 12, No. 3, pp. 314–325. Robertson, Jennifer (ed.) (2007): Politics and pitfalls of Japan ethnography: Reflexivity, responsibility, and anthropological ethics (Special Issue). In: Critical Asian Studies 39, No. 4. Robertson, Jennifer (1998): Takarazuka: Sexual politics and popular culture in modern Japan. Berkeley, CA: University of California Press. Roth, Joshua Hotaka (2002): Brokered homeland: Japanese Brazilian migrants in Japan. Ithaca, NY: Cornell University Press.

182

Chapter 6 How to do fieldwork Roth, Joshua Hotaka (2003): Responsibility and the limits of identification: Fieldwork among Japanese and Japanese Brazilian workers in Japan. In: Bestor, Theodore C./Steinhoff, Patricia G./Lyon-Bestor, Victoria (eds.): Doing fieldwork in Japan. Honolulu, HI: University of Hawai‘i Press, pp. 335–351. Rupp, Katherine (2003): Gift-giving in Japan: Cash, connections, cosmologies. Stanford, CA: Stanford University Press. Schatz, Edward (ed.) (2009): Political ethnography: What immersion contributes to the study of power. Chicago, IL: University of Chicago Press. Solt, George (2014): The untold history of ramen: How political crisis in Japan spawned a global food craze. Oakland, CA: University of California Press. Stocking, George W. (1992): The ethnographer’s magic and other essays in the history of Anthropology. Madison, WI: University of Wisconsin Press. Tatematsu, Kikuko (1993): Writing letters in Japanese. Tokyo: The Japan Times. Ugoretz, Kaitlyn (2017): A guide to unobtrusive methods in digital ethnography. Unpublished paper, https: //www.researchgate.net/publication/331287677_A_Guide_to_Unobtrusive_Methods_in_Digital_Ethnog raphy, [Accessed 13 May 2020]. Urry, John (2016): Mobilities: New perspectives on transport and society. London: Routledge. Van Maanen, John (2011): Tales of the field: On writing ethnography. Chicago, IL: University of Chicago Press.

183

Chapter 7 How to interview people: Qualitative interviews Nora Kottmann and Cornelia Reiher

1. Introduction Scholars researching Japan from across the Social Science spectrum rely on qualitative interviews with a variety of actors to learn about their lifeworlds, experiences, practices and perspectives on a range of issues. These include, for example, residents in a specific neighbourhood, local government officials, parents, married and divorced couples and farmers (see this chapter, Alexy, Ch. 7.3; Brumann, Ch. 7.1; Yamaguchi, Ch. 7.2). In this chapter, we examine what qualitative interviews are and what insights researchers can gain from interviews in general, as well as from interviews both in and outside of Japan. We detail decisions a researcher should make before conducting interviews to ensure the data produced addresses the research question. We then introduce different types of interviews, elicit the process of selecting and contacting interviewees and give practical advice on preparing for the actual interviews. This is followed by in-depth discussion on the practice of interviewing: specifically language issues, the process of interviewing, dealing with (non-)verbal expressions, negotiating interviewer–interviewee relations and recording. The chapter concludes with reflections on interview followups, particularly in regard to the transcription of interviews, and how and why to stay in touch with informants.

2. Qualitative interviews Qualitative interviews range from the ‘traditional’ question and answer type to more openended and spontaneous ‘conversations’. Interviews are probably the most frequently used tool in qualitative research (Keddi/Stich 2008, p. 2) and are utilised in the Humanities and Social Sciences to enable the researcher to ‘elicit views and opinions from the participants’ (Creswell 2014, p. 190), to learn about the interviewees’ experiences (Rubin/Rubin 2005, p. 2) and to obtain descriptions of their lifeworlds in order to understand how an interviewee interprets the meaning of phenomena they describe (Brinkmann 2014, pp. 277, 286–289). In contrast to surveying through questionnaires, interviewing people directly enables researchers to ask questions, listen to stories and respond to unexpected issues and opinions immediately. In addition, when meeting informants in their homes or coffee shops, or when following them through their paddy fields, researchers can catch a glimpse of the environments the informants live in.

184

Chapter 7 How to interview people Although differences in the definition of the qualitative interview types exist across disciplines, most scholars agree on three basic strategies to produce verbal data: asking individuals more or less fixed questions, listening to stories and talking with couples or groups (Flick 2006, p. 149). Depending on a researcher’s discipline, the boundaries between interviews and ‘normal’ conversations can be blurred. In Anthropology, for example, ‘questioning and/or just listening take place within everyday conversation’ (O’Reilly 2005, p. 116; see this chapter, Alexy, Ch. 7.3). This approach stands in contrast to more formalised interview settings where researchers and informants might meet for the first (and sometimes only) time in order to ask and answer questions for a fixed period of time. Qualitative interviews, regardless of their specific definition, always involve explicit rules concerning their content, duration and setting. These rules and practices may differ according to disciplinary conventions, but foremost according to the research topic, context and interviewees. This also applies to ethical rules ‘concerning consent for the interview, for recording and for preserving the subject’s anonymity and the confidentiality of the respondent’ (Hammond/ Wellington 2013, p. 91; see Slater et al., Ch. 16.2). Because of the plurality of approaches to interviews, it is important to familiarise oneself with how interviews are conceptualised within one’s discipline, as well as best practices for how they are to be conducted, documented and analysed. Key issues Qualitative interviews are a method to find out how people characterise and view their lives, to identify what they find important and to understand their perceptions and interpretations of the topic a researcher is interested in. Therefore, it is people’s perspectives and not ‘objective facts’ researchers can find out.

3. Choosing the ‘right’ type of interview and questions Once a researcher decides on interviews as a suitable tool to answer her research question(s), two intertwined decisions must be made: how to interview, and how and what to ask. There exist different interview types that can be distinguished between based on 1. their degree of structure, 2. the number of interviewees (individuals, couples or groups) and 3. the interview method (face-to-face, via phone or online). Deciding which interview type to use is primarily dependent on whether the data ‘made’ through interviews will be used as evidence to answer the research question (Hammond/Wellington 2013, p. 92). Other factors that influence the choice of interview type include the researchers’ abilities with regard to the local language, her personal as well as professional experience, available funding, time and staff, structural circumstances such as access to informants, geographical distance, and the abilities and needs of the interviewees in terms of health or privacy. The most obvious distinction between interview types is in their structure (see this chapter, Yamaguchi, Ch. 7.2). Interviews are commonly divided into structured, unstructured and semistructured types. In reality, however, this distinction is more of ‘a continuum ranging from relatively structured to relatively unstructured formats’ (Brinkmann 2014, p. 285). Although in 185

Nora Kottmann and Cornelia Reiher the literature on research methods, the various interview types appear as neatly separated, in practice their use is more intertwined (Helfferich 2004, p. 11). Generally speaking, unstructured interviews, such as narrative (often biographical), episodic or ethnographic interviews, strongly focus on subjective meanings or the interviewee’s perspective and system of relevance. They often evolve around a ‘generative narrative question’ as in the case of narrative or episodic interviews (Flick 2006, pp. 173–188) or often develop spontaneously out of a ‘normal’ conversation in the field. In contrast, more structured interviews mainly aim to refute or confirm a hypothesis and centre around questions the researcher has decided on before the interview(s) (Brinkmann 2014, p. 286). Most researchers in qualitative Social Science research use semi-structured interviews (ibid.). Their intent is to identify individual views, but they are based on a (loose) interview guide a researcher can use for orientation. Sticking to the guide is not, however, obligatory, and the researcher can ask various additional questions or include a new topic whenever she feels it is appropriate to do so. Mixing different degrees of openness in interviews can be necessary for collecting different kinds of information (demographic information and individual perceptions) or when researchers need to flexibly adjust their (theoretical) plans to the actual interview situation or the ‘flow’ of the conversation (Rubin/Rubin 2006, p. 35). The level of structure of an interview can change anytime during the interview process. While unstructured interviews are particularly helpful in exploratory phases of a research project, a more structured interview format is often better suited to later phases of a project, to follow up on specific issues or to test a hypothesis that derived from earlier exploratory and unstructured interviews (Froschauer/Lueger 2003, p. 35). Different interview forms: focused interview, semi-standardised interview, problemcentred interview, expert interview, ethnographic interview, narrative interview, episodic interview, couple interview, group interview, group discussion, focus group1 Questions asked during interviews may differ significantly based on the interview approach. However, regardless of the interview type, researchers should produce ‘a set of questions that are meaningful for the interviewee’ (Hammond/Wellington 2013, p. 92) and ask the ‘right’ questions in order to get answers that help her to answer the research question. In semi- and unstructured interviews, the questions should ‘invite interviewees to give descriptions’ (Brinkmann 2014, p. 287), as interviewees are expected to answer ‘as freely and as extensively as they wish’ (Flick 2011, p. 112). Hammond/Wellington (2013, p. 92) stress careful use of language, such as avoiding jargon, striving for clarity in phrasing and asking open-ended questions that relate to the expertise of the informants. Flick (2011, p. 113) suggests that questions should ‘allow room for the specific, personal views of the interviewees’ and should not influence them. Questions that refer to abstract theoretical concepts like ‘globalisation’ or ‘hypergamy’ or questions that imply a certain understanding of a term or concept like, for example, ‘love’ should also be avoided.

1 For differently structured overviews on interview forms, corresponding questions and possible research goals, see Cresswell 2014; Flick 2006; Rubin/Rubin 2006. For detailed information on couple interviews, see Yoshida, Ch. 5.

186

Chapter 7 How to interview people Key issues Consider whether the data produced through a specific interview type is best suited to your research topic and will help you answer your research question. Take available resources and practical issues like access to the field or the needs of informants into account. Mixing different interview formats in one interview or across interviews might be helpful or even necessary.

4. Selecting and finding interviewees The research question and topic are key factors in deciding on appropriate interviewees, just as they are in determining the optimal interview format. When selecting interview partners, a researcher should keep in mind the practical considerations discussed in section 3, but also consciously reflect on sampling strategies and determining what expertise interviewees need in order to produce meaningful data. Sampling refers to the process of selecting whom you will interview (case sampling) (Flick 2006, p. 122). In order to produce reliable and valid research, it is important to reflect on the criteria according to which you have selected the interviewees and make this process transparent to the audience. For some research projects, it might be useful to choose the interviewees before going to the field (a priori sampling) based on certain criteria: for example, age, occupation or gender (statistical sampling, Flick 2006, p. 123). In other projects, sampling can occur during the process of collecting and interpreting interview (and other) data (theoretical sampling, Flick 2006, pp. 125–131). Theoretical sampling is derived from grounded theory (Glaser/Strauss 2005; see Meagher, Ch. 12), but has become widely accepted in qualitative research. The gradual selection of cases and material is based on criteria concerning content and relevance rather than representativeness (Flick 2006, p. 128). The ideal informant is ‘experienced and knowledgeable’ and contributes ‘a variety of perspectives’ (Rubin/Rubin 2006, p. 64) on the researcher’s topic of interest. When you are selecting interviewees, it is best to think of all informants as experts in their respective lifeworlds, experiences and opinions. Different types of expertise exist, however. These experts derive their status from being part of a certain group (intra-systematic knowledge), from possessing knowledge of several systems in a larger field (intra-field expertise) or from external expertise (theoretical knowledge). Examples of such experts are scholars or journalists, who are often targeted for ‘expert interviews’ (Froschauer/Lueger 2003, pp. 37–38). In Japan, social networks, introductions and trust relations are crucial in order to identify and connect with informants who are willing to be interviewed (Bestor et al. 2003a, p. 14). Foreign researchers in Japan often obtain access by affiliating with a Japanese host institution (McLaughlin 2010). Although in some cases researchers prefer not to mention their affiliation (Steinhoff 2003), Japanese supervisors and colleagues can introduce researchers to potential informants and thus become important gatekeepers. Some scholars rely on snowball sampling to access ‘informants through contact information that is provided by other informants’ (Noy 2008, p. 330). But being introduced to informants by only one gatekeeper can result in a biased perspective on the research topic. Thus, to find interviewees, creative and plural modes of access via different gatekeepers are crucial (McLaughlin 2010, p. 8–9), as are reflecting on the 187

Nora Kottmann and Cornelia Reiher influence of gatekeepers and sampling strategies on the findings of research projects (see Gagné, Ch. 6.1). Contacting people to make an appointment for interviews with or without the help of gatekeepers might demand the use of different media. While in some cases an email with an introductory letter from a scholar’s home or host institution and an outline of their research is necessary to set up a connection, in other cases all that is required to gain access is a phone call or a visit to the person’s shop or farm. Sometimes social media can be the most efficient tool with which to arrange an interview. In her research on local culture in post-disaster Tōhoku, Julia Gerster (2018) arranged almost all her interviews via Facebook and Line. How to best contact interviewees after you identify them as knowledgeable depends on the people one would like to interview and demands creativity, patience and often involves a lot of trial and error. Key issues In order to decide whom to interview, you should consider different kinds of sampling methods and reflect on the kind of knowledge the ‘ideal interviewee’ should possess to answer your research question. Try plural modes of access to interviewees and don’t rely on only one gatekeeper. Different media can be helpful when contacting informants to arrange interviews.

5. Preparing interviews: Location, timing and things to bring Once a researcher has (roughly) decided how and whom to interview, the appropriate preparation is crucial. The first step is to find a location to conduct the interview. Interviews can take place at people’s homes, favourite restaurants, cheap fast-food chains, parks, coffee shops or workplaces. When possible, it is preferable to leave the decision about the location to the interviewees so they can choose a place that is convenient and comfortable for them. This has the added benefits of both giving researchers an insight into the life of the respondents, especially if they are invited into their home or one of their favourite local hangouts, as well as freeing the researcher from having to search for an appropriate location, which can prove rather difficult in a city like Tokyo. Sometimes researchers have to select locations themselves. Public spaces like coffee shops can be a good choice, even for interviews on sensitive topics. If the interviews are going to be recorded, it is crucial to consider all the sources of environmental noise and seek ways to limit it as much as possible. Nothing is more disheartening than discovering that hours of audio are ruined due to background noise (as we both know from experience). Once the meeting place for the interview is set, it is best to visit the location and confirm its opening hours. When a face-to-face interview is not possible, there are other good options, such as sending questionnaires or conducting phone or online interviews (Flick 2006, pp. 254–260). These techniques of producing verbal, written, audio and/or video data enable the researcher to carry out ‘both asynchronous and synchronous online interviews’ (Hammond/Wellington 2013, p. 92) and to widen the scope of her research. (Written) online interviewing can be less intrusive, can produce more reflective answers, and can help people feel more comfortable (ibid.). How188

Chapter 7 How to interview people ever, the responses might be less spontaneous and the relationship between the researcher and the interlocutor less familiar (Flick 2006, pp. 256–260). It is also important to take sufficient time to conduct interviews and, if possible, to have a coffee or dinner with the interlocutor afterwards (see this chapter, Alexy, Ch. 7.3).2 The length of an interview depends on the interviewee’s available time, the topic, the number of people involved, the type of interview and the overall atmosphere. We have conducted interviews ranging from thirty minutes up to around four hours and experienced interviews to last longer when invited to individuals’ homes, as interviewees showed us personal belongings and even invited us for dinner. Nora usually writes in a first email to possible interviewees that the interview will take ‘approximately one hour’ because a longer time span seems to be a deterrent, especially to working people. Before the interview starts, she also asks the interviewee again how much time she has in order to adjust the interview guide (if there is one), to plan the interview and to frame the situation. Finally, there are several things the researcher should bring to the interview. Apart from items like a recording device that help the researcher with organisational matters, we recommend bringing name cards (meishi) and a gift (McLaughlin 2010). Gifts create a pleasant atmosphere and show gratitude for the interviewees’ help and efforts. They don’t need to be expensive and can be something from the researchers’ home country or some sweets bought in Japan. We recommend always bringing extra gifts in case one is met by more people than expected or gifts are presented to the researcher. Key issues Ask the interviewee where she wants to meet for the interview. In case they leave the decision up to you, suggest convenient locations and bear topics like privacy, noise or opening hours in mind, and always check the location beforehand. Take enough time, but also think about the interviewee’s needs and constraints. Always bring name cards (meishi) and a small gift. Bring more than you think you will need.

6. Deciding on the language When you are conducting interviews in Japan or with Japanese interlocutors outside of Japan, language is of paramount importance. Usually researchers contact their interviewees in advance via email or phone in the language the interview will most likely be conducted in. In cases where contacts were established through a third person, the question of language has to be explicitly discussed. Ideally, interviews should be conducted in the language informants feel most comfortable with, so they can express their opinions and feelings. In most cases, this is Japanese. Therefore, ‘advanced reading and comprehension skills are a prerequisite to doing research in Japan’ (Smith 2003, p. 161). However, researchers should not fear conducting interviews in Japanese due to anxiety about their language skills as ‘doing fieldwork is in itself a powerful language-learning opportunity’ (Bestor et al. 2003a, p. 9). 2 For a different opinion on this topic, see Yoshida, Ch. 5.

189

Nora Kottmann and Cornelia Reiher Even with fluency in Japanese, there will likely be many new specialised terms to master (Smith 2003, p. 162). Therefore, it is important to study the specialised vocabulary necessary to conduct interviews on your research topic. In Cornelia’s (Reiher 2014) project on porcelain production in Kyushu, she learned the special terms for different types of kilns and glazes in advance. In addition, interviews in multicultural communities may also demand fluency in several languages. In Chaline Timmerarens’ (2018) project on Brazilians in Hamamatsu, for example, she conducted interviews in Japanese and Portuguese. Using multiple languages when researching Japan increases the ‘burden of checking and double-checking that we have interpreted meaning correctly’ (Smith 2003, p. 161). Thus, the interviewer must not be afraid to ask for explanations of unfamiliar Japanese terms while conducting the interview. When in doubt, one should always consult Japanese colleagues and friends for help or contact research participants to clarify what they meant by a certain term. Although most interlocutors will prefer to speak Japanese, some Japanese interviewees will insist on ‘speaking “impoverished” English’ to foreign researchers (Johnson 2003, p. 146). In transnational settings where interviewees are bilingual or multilingual, researchers are likely to encounter a language-mix in interviews, as Nora encountered in some interviews during her research on marriage decisions and relationship worlds (Kottmann 2016; 2019). Key issues Conduct the interview in the language the interviewee is most comfortable with. Study specialised terms related to your research topic. Don’t be afraid to ask your interviewee if you don’t understand what she is saying during the interview or to get help from native speakers. Reflect on the language(-mix) used in the interviews.

7. The process of interviewing: Listening, contradictions and (non-)verbal expressions The interview process can be roughly divided into an introductory, main and final part (Hopf 2007, p. 356). Before the interview starts, the researcher should briefly explain the interview’s goal and procedure, and talk about ethical norms, such as anonymity and how the data will be used. After some small talk, she should ask if she can turn on the recording device and start with more general introductory questions (see McLaughlin, Ch. 6). In case the interviewee has asked for the questions to be sent in advance of the interview, going over the questionnaire is a good place to start. Be transparent and create a pleasant atmosphere to get interviewees to talk more freely. We usually start an interview by asking informants for a brief self-introduction (jiko shōkai) or to talk about an item related to the topic. Cornelia, for example, brought fairtrade chocolate from Germany to interviews about food labelling. In all interview approaches, it is important to begin and proceed with an interview in a way ‘that makes the conversational partner feel comfortable, [the researcher] obtains [the] needed information, and [that] is compatible with the researcher’s [and the interviewees] personality’ (Rubin/Rubin 2006, p. 31). After the introduction, researchers then ask main questions, follow-up questions and probes (Rubin/Rubin 2006, pp. 134–139). In the main part of the interview, it can be helpful to con190

Chapter 7 How to interview people tribute one’s own thoughts and experiences on the subject matter to establish ‘conversational partnerships’ (Rubin/Rubin 2006, p. 79). These contributions should be limited so as to not influence the interlocutor or direct the narrative. Regardless of the interview type, it is important to listen to the interview partner and not interrupt her. Similarly, a researcher should also ‘listen’ for the stories that have not been told. This may require accepting moments of silence, ambiguities and contradictions. Both researcher and interlocutor might feel uncomfortable with silences, but not interrupting the ‘flow’ of thinking and talking can also offer insights. The same is true for evasive replies, when there is no response and when the interviewee does not address salient topics—one can gain insights from what is not said. In addition, it is important to pay attention to body language, tone of voice, speech rhythm and the overall atmosphere, as well as one’s own immediate feelings as they are meaningful and can be understood as commentaries on the verbal data (Kaufmann 1999, p. 117). Possible ways to handle contradictions in interviews that can be seen as ‘normal’ ‘internal conflicts in narratives and descriptions’ (Brinkmann 2014, p. 288) are to either ask follow-up questions or to simply continue conducting the interview as planned. This also applies to ‘untrue’ narrations. One of Nora’s informants told a life story that significantly differed from information about the interviewee that was obtained from other informants (including her marital status). Despite the interviewee deflecting questions, Nora still decided to continue with the interview, which later provided enlightening material with regard to the research question. It also demonstrated that instead of asking if stories are ‘true’ or not, it is important to reflect on how is what (not) being told and ‘what truth did [the interviewee] tell’ (North 2009, p. 31). The final part of an interview is often devoted to enquiries on specific topics that came up during the interview. To be able to ask these follow-up questions, Cornelia recommends taking notes during the interview. Before ending an interview, we usually include some open questions that allow further room for the interviewees to reflect on their interpretation of the topic (‘Is there anything you would like to add?’) (see Sternsdorff-Cisterna, Ch. 2.1). After you have obtained all the necessary information or if the interviewee starts checking the time, thanking them and switching off the recording device is a practical way to end the interview (see below). For further valuable insights and out of politeness, we recommend accepting invitations for a joint meal after the interview (see this chapter, Alexy, Ch. 7.3). Akiko Yoshida (see Ch. 5), on the other hand, is reluctant to accept invitations due to concerns about bothering the interviewees. In any case, a researcher should always carefully consider the situation before accepting an invitation. Key issues Creating a pleasant atmosphere and asking introductory questions are a good way to break the ice at the start of an interview. Don’t interrupt your interviewees and listen to the stories that are (and are not) told. Endure silence and contradictions, and also take non-verbal expressions into account. End the interview with open questions that invite the interviewee to bring up their own questions on the topic and possibly add issues.

191

Nora Kottmann and Cornelia Reiher

8. Negotiating interviewer–interviewee relations and reflexivity Just as it is important to create a comfortable atmosphere during an interview, it is also crucial to create a comfortable relationship between the interviewer and interviewee. This is greatly influenced by the interview approach. In ethnographic interviews—in contrast to more structured interviews—the relationship tends to be more ‘equal’ and the roles are more blurred. However, regardless of the interview form, the relationship has to constantly be negotiated— as do mutual expectations, aims and interests. The researcher, therefore, has to face the difficulty of balancing proximity and distance, strangeness and familiarity, being an insider and an outsider, being empathetic or not, as well as being active and passive (Flick 2006, p. 118, 120; Rubin/Rubin 2006, pp. 81, 86–89). Negotiating these balances can be particularly difficult in interviews on sensitive issues, such as illness and death (Spoden 2015). The expectations of the interviewee(s) may also differ significantly: they might expect the interviewer to adopt their perspective (and support them) or they might regard the interviewer as an ‘expert’ who can give ‘objective’ advice on various matters. These topics are closely connected to a researcher’s positionality and reflexivity: that is, being aware of one’s implicit assumptions, questioning them and being conscious of the impact or influence of the researcher on the research in general, and the interview specifically (Brayda/ Boyce 2014). The importance of reflexivity is widely acknowledged. Yet, researchers refer to this process with different terms, such as ‘being self-aware’ (Rubin/Rubin 2006, p. 31f.) or doing ‘backyard research’ (Creswell 2014, p. 87f.). Reflexivity also refers to a researcher’s sociodemographic factors, like age, gender, nationality and marital, familial or occupational status. Certainly, the interview and stories the interviewees tell depend significantly on the researcher (Takeda 2013). Genaro Castro-Vázquez, for example, experienced in his research on intimacy and reproduction in Japan that his interlocutors—all of them married women between the ages of 29 and 45—‘found conversation on intimacy […] challenging’; especially in front of a single, childless, middle-aged, non-Japanese male scholar (Castro-Vázquez 2017, p. 66, 186). Being a non-native, obviously not Japanese researcher also has significant (often positive) impacts on the interviews. Nora, for example, experienced the situation of her interlocutors explaining many things in great detail that they assumed she didn’t know as a non-Japanese. Considering reflexivity is also important when working with research assistants, as James Farrer (2013) elaborates on in his research on sexual stories of Chinese women in Shanghai. Key issues When negotiating interviewee–interviewer relationships, reflexivity is important. It is essential to be aware of one’s own implicit assumptions, to question them and to be conscious of the impact or influence of the researcher on the research in general, and the interview specifically.

192

Chapter 7 How to interview people

9. Recording, taking notes and (not) transcribing During interviews, many researchers record their interviews as audio files via a voice recorder or smartphone. Others create videos or take pictures. It is important to always ask for permission to record before the interview begins. Some researchers and ethics boards demand written consent for you to record an interview on audio and video files (see Slater et al., Ch. 16.2), and always bring a form to sign to the interview. Others ask for permission by email prior to the interview. Some scholars recommend taking notes, even if the interview is recorded, as there are cases when audio files are incomprehensible. If one does take notes during an interview, it also helps to pick up on topics your interview partner mentions and to grasp non-verbal expressions. As soon as one leaves the interview site, it is helpful to find a place to sit down and write up everything one remembers from the interview: describe the situation, number of people, the vibe, maybe even draw a map of the room and where everyone was seated during group interviews (see this chapter, Yamaguchi, Ch. 7.2). This is important even when recording, but especially when taking notes during the interview is not possible, as Patricia Steinhoff (2003) experienced when conducting interviews in prisons. After such interviews, we recommend jotting down quick notes before writing a longer report. Last but not least, make sure you store your audio files and notes. McLaughlin (2010, pp. 17– 18) advises researchers to ‘save everything in at least three places’ such as one’s computer, a flash memory, an external hard drive or online. In the likely case that a researcher records her interviews, she has to decide after the interview whether to transcribe the audio material and, if so, which method to use. While some researchers see transcribing as a ‘necessary step on the way to […] interpretation’ (Flick 2006, p. 288), others are resistant to making a full transcription as they argue that the nature of the raw material changes in the process and loses its depths and complexities, which are inherent in speech rhythm, tone of voice and breaks of silence. Jean-Claude Kaufmann (1999, p. 117), for example, recommends only transcribing important parts of the interviews, while constantly relistening to the whole audio data. If the researcher decides to transcribe (some parts of) the interview, she also has to decide on the level of detail of the transcription: from notes to written versions of the interviews up to extremely precise transcriptions that include stalling words, as well as their pronunciation. As transcription can be extremely time-consuming, it might be helpful to get (paid) assistance when a detailed full transcript is required and/or use a transcribing device operated by a food pedal and/or transcription programs.3 For most projects, however, transcripts do not have to be perfect. Decisions on the level of detail should be based on the level of analysis and ‘include any information that might influence the interpretation, such as laughter or gestures of emphasis, [breaks of silence] or puzzlement’ (Rubin/Rubin 2005, p. 204). Regardless of the level of detail, it is important to create a table in which important abbreviations and characters are introduced and explained. It is especially necessary (and vital for analysis and writing) to clearly mark direct citations and parts which are summaries of what was heard (Kottmann 2016, pp. 117–119). 3 See, for example, FOLKER (free of charge).

193

Nora Kottmann and Cornelia Reiher Key issues Always take notes, even when you record the interview. Sometimes the most interesting things are said after you have turned the voice recorder off. Document all information about the interview, the interview situation and your informants immediately after the interview and store them in different places. We recommend transcribing at least part of the interview. Transcriptions do not have to be perfect but should include non-verbal expressions. A table with abbreviations and a clear mark-up of direct citations is helpful.

10. Following up and keeping in touch Scholars writing about fieldwork in Japan advise researchers to ‘follow up every interview with a thank you, either in writing or by phone’ and suggest writing ‘thank you letters on high-quality paper’ (McLaughlin 2010, p. 13). This is not only because of the importance of trust, social networks and reciprocity for fieldwork in Japan (see Gagné, Ch. 6.1), but also because one might want to contact sources again for questions that arise during data analysis or to ask them for follow-up interviews. Recently, social media has become an equally important means for expressing gratitude and maintaining long-term relations with informants. In addition to writing thank you notes, Gerster (2018) also likes the informant’s personal social media profile and shares pictures via Facebook and Line. Through social media, Skype or emails, a researcher can stay in touch easily with informants to ask follow-up questions or to ask for help in gaining access to informants for a new research project. When Cornelia (Reiher 2012; 2017) started a new project on food safety in Japan, she first contacted farmers she knew from her previous fieldwork in rural Japan. Many researchers even maintain life-long relations with their informants, particularly when they have conducted fieldwork over a longer period of time (see Gagné, Ch. 6.1). Interviewees and gatekeepers might ask the researcher for favours as well. Although ideally informants will collaborate in research projects voluntarily (Sluka und Robben 2012), reciprocity in the research process means that researchers should return favours to informants and gatekeepers to thank them for their support and efforts (McLaughlin 2010, p. 7). Cornelia, for example, regularly meets with consumer advocates and bureaucrats from Japan who come to Berlin to study the German food system. Sometimes informants also expect researchers to take sides, represent their views and advocate for them. After Japan’s 3.11 triple disaster, for example, some called for enhanced political engagement and scholar activism and raised questions about researchers’ responsibility towards those being studied (Yamashita 2012). Particularly in Anthropology, reciprocity also refers to ‘reciprocal processes with tangible benefit for local communities, if ethnographic work is to continue’ (lewallen 2007, p. 509). This might also include presenting one’s findings to informants before and after the research is finished (Gerster 2018). In some cases, however, it is impossible to maintain relations with informants. Sometimes this is due to sensitive research topics, but in other cases research participants can no longer be located (Steinhoff 2003, p. 39, 40), are simply too busy or are not interested in remaining in touch.

194

Chapter 7 How to interview people Key issues Thank your interlocutors for the interview and try to maintain relations with them. You might have follow-up questions or want to conduct follow-up interviews. Reciprocity means that interviewees may also ask you for favours, which should be reasonable.

11. Summary In this chapter, we have provided an overview of qualitative interviews used in research on Japan. Qualitative interviews should be conducted in a reliable and ethical way. This requires researchers to actively reflect on the entire interviewing process and make the process transparent to their audience. Reading the literature on interviews from specific disciplines is a good first step to prepare for conducting interviews in Japan. But it is important to bear in mind that conducting interviews in a different cultural setting presents specific challenges, especially in regard to language and getting access to informants. Therefore, reading about interviewing in Japan, reviewing literature on methodology in Japanese and asking your Japanese colleagues for advice are all helpful before getting started. We found interviews to be a wonderful way to study people’s lifeworlds, opinions, experiences, values and beliefs in and outside of Japan. Meeting informants in person at a variety of places, listening to their always interesting stories and experiencing their kindness has been a great experience for us. We encourage you to give it a try in order to gain one more perspective on Japan.

195

7.1 The empire of interviews: Asking my way through Japan Christoph Brumann

Quite a few Japanese have told me that their culture does not encourage self-expression, but in my ethnographic research in the country, I have never managed to confirm this stereotype. On the contrary, and in a play on Roland Barthes (1970), I consider Japan the empire of interviews: many Japanese are familiar with this kind of interaction, even if only from the mass media, and are sufficiently polite, patient and curious to try and subject themselves to the exercise even when it is for the first time in their lives. Interviews can therefore be a valuable tool for social research in Japan. In the following, I will reflect on my experiences with this method, hoping that this will provide ideas and encouragement for the reader’s own practice. As a social anthropologist, I have used interviews in all my ethnographic research projects, be it on Japanese utopian communes and their histories and current conditions (Brumann 1996), the gift-giving practices of my former landlord in Tokyo (Brumann 2000), the conflicts about heritage conservation and urban development in Kyoto (Brumann 2012) and the decision processes of the World Heritage Committee (Brumann 2021), where my broad set of interlocutors included Japanese nationals. In 2016, I also used interviews for a yet unpublished study of Buddhist temples in Kyoto. The first Kyoto study in particular included over a hundred formal interviews, and I must have spent months of my life asking Japanese individuals questions and listening to their answers. Most of the time, I enjoyed myself—when interesting people open up about a topic one is interested in, this can be a stimulating experience. I am a dilettante in the sense that beyond some reading in methods handbooks, I have never been formally taught how to do interviews, proceeding in a more or less intuitive way most of the time. This is certainly easier with the ‘open-ended’, narrative type of interviews I conducted most often: in these cases, I had no rigid questionnaire to be ticked off but only a laundry list of items to be covered, often on just one or two pages. These could include fairly precise questions but also more vaguely delineated topics. Such an open methodology has the advantage of allowing for discoveries: often enough, my interlocutors revealed facts, views and connections I had not expected. Beyond this, open-ended interviews are good for exploring worlds of thought and feelings and the chains of reasoning and association in people’s minds. Interviews are better suited to retrieving stories than to checking isolated facts. Quite often, the people I met were involved in some sort of structured activity I was interested in, representing a specific cause or the organisation set up to pursue it (such as, for example, an NGO engaged in saving traditional houses). This meant that upon meeting them for the first time, I was regularly presented with a stack of brochures, leaflets and press clippings that would often explain basic facts and positions at least as well as the most structured conversation. Where possible, this might speak for a brief previous meeting in order to obtain and process such information, so that precious interview time can be reserved for what cannot be gathered otherwise. Interviews are particularly good for uncovering the informal side of formal facts: a can-

196

Chapter 7 How to interview people did conversation can reveal quite a bit about what official self-presentation really means or what it actively tries to hide, which is important in a society where appearances count.

Interview strategies Interviews are contingent social interactions, and I think that it is prudent not to see them as opening a tap out of which pre-formed pieces of information then flow freely. Rather, an interview is a not very closely controlled experiment in which a researcher and those researched coproduce something that, at least to some degree, is unique to the encounter and might not have arisen under different circumstances. Obviously, social scientific research pursues general insights and therefore aims to control this effect, rather than embellish it (different from, say, a late-night show host questioning a celebrity guest). Yet still, it is advisable to face the fact of co-creation, rather than to ignore or conceal it (see Gagné, Ch. 16.2; Klien, Ch. 8.1). This means that the personality of the interviewer and their way of conveying empathy and understanding have a bearing. The ability to think (and feel) along is important to keep an interviewee interested and engaged. There are a lot of verbal and non-verbal ways to signal that one is following what an interviewee says. Beyond this, I talk quite a bit myself in my interviews, such as by summarising an important or surprising point and making comments, just as I would in an everyday conversation. This no doubt increases my impact on what happens, but I tend to think that it helps to maintain the intellectual and emotional flow, and since not influencing what the interlocutor utters is impossible (see above), I prefer such a dialogic approach to reading out a question and then falling silent, come what may. What I still have to learn, however, is to resist the temptation to finish sentences in the interviewees’ stead, such as when they are groping for words or prefer not to express the obvious—I am sure that this has cost me a number of interesting quotes, and tolerating a moment of silence until the interviewees finish a sentence in their own words can have its advantages. When asked for an interview, quite a few people will offer one hour even when they often end up talking longer than initially intended. After one and a half hours or two, concentration often starts to drop on one or both sides, but I have also had rewarding interviews of double that length. It may then be preferable to meet more than once, however, and in my Kyoto research, I interviewed some individuals half a dozen times or more. Staggering the encounters in this way also allows the prior ones to be digested and new questions to be built on the insights gained. Interviews do not prescribe a one-on-one format, and it was sometimes the interlocutors themselves who preferred to meet me together with their married partner, their colleagues or another researcher intent on questioning them. I myself brought the person who had provided the contact or my own colleague in a number of cases. In Kyoto, I did a whole series of interviews with a circle of elderly female friends who regaled me with their reminiscences of the social mores of their youth. Here, the control I exerted over the encounter diminished greatly—the animated conversation took leaps and bounds, making systematic coverage of any given topic difficult, but the joint excavation of memories brought up things that each lady might not have thought of on her own. Group interviews can also ease the stress of simultaneously listening, taking notes and thinking about the next question, given that the flow of conversation depends less on one’s own input. 197

Christoph Brumann For some projects, systematic comparison across a number of interviews is a possibility, and I did interview series with all the members of a citizens’ group I observed in Kyoto or most households in the neighbourhood where I had followed the Gion festival (Brumann 2012, pp. 15–45, 156–208). This might require a different approach to interview questions. In my case, many of my questions were identical and a comparison was intended. For strict comparability, however, a written questionnaire is often the better choice, as it exposes all individuals to the questions in the same way. Integrating a not too arduous survey in an interview can lighten the mood, and I used one in a number of my interviews (see Holthus/Manzenreiter, Ch. 5.3; Yoshida, Ch. 5). For example, I included a test where I asked Kyotoites to sort photos of buildings according to their personal likes and dislikes (Brumann 2012, pp. 211–219). This yielded both a numerical data set and the live comments the interviewees made when fulfilling the task. In the course of long-term fieldwork, interviews often grow out of prior social encounters and are followed by further ones. This usually makes for more relevant and suitable questions and more trust. Interviews can become intimate occasions—encouraged by the flattering interest in their personality and views, quite a few interlocutors end up revealing more than they initially intended, and mutual understanding makes people feel closer. This can increase trust and support in subsequent encounters, and the choice of interview partners might reflect this hope, not only what they might have to say in the specific interview.

Recording and ethics Strategising should not involve deception, however. Research ethics require disclosing one’s own identity and purpose at the outset, and in line with common anthropological practice, I assure my interlocutors that they will remain anonymous in my writings. Sharing the contents or recordings of an interview with others requires explicit consent. Depending on the topic, individuals may not be worried about being named or even insist on it. Such demands should be treated with circumspection but, where possible, with the interviewee’s wishes as a guideline. If interviewees agree—and a number of times, I was overly cautious in asking—audio recording an interview is an option. Some interlocutors refused to be recorded and others became perceptibly nervous, but most Japanese I met did not mind and seemed not to adjust their candidness greatly. I share Ellis S. Krauss’s observation (2003, pp. 182–183) that the device is often completely forgotten—it sometimes came back to mind only when I stopped the recording in the end, prompting the occasional sceptical remark and my reconfirmation of confidentiality. I usually offer to send my interlocutors the sound file of the interview, which some of them accept. A recording is, of course, a more faithful rendition than taking notes and allows for repeated listening, including also of passages one failed to grasp initially. When the precise wording of longer passages is not an issue, however—and in much research, it is not—one can also jot down notes (in my own case, often in a mix of Japanese and my native language German). To avoid long pauses, it is advisable to be brief here, writing just enough to jog one’s memory when typing or dictating a full version of what was being said. Time should be set aside for

198

Chapter 7 How to interview people this purpose immediately after the interview, and I was often surprised how much detail I could still remember then. Sound quality is important: most current smartphones make good recordings but particularly in noisy environments such as cafes, an external microphone can be advisable, and stereo recording makes it easier to filter out the interlocutor’s voice from background chatter. Scratching and knocking on tables can produce disturbing sounds when a device is placed on them without cushioning. For outdoor situations, the device or microphone must have wind protection (I lost a good part of one interview by not paying attention to this). It is best to test the equipment beforehand or use earphones to assess conditions. Recordings are a blessing and a curse, given that in order to be useful, they must be processed. I generally discourage full transcriptions. It is better to type summaries of what was being said, including the occasional verbatim quote, but even that takes many times longer than the interview itself and has kept me busy for months on end. So unless one has help, it might be best to devise an indexing system with keywords under which hyperlinks lead directly to the respective interview passages. Text analysis software such as MAXQDA allows for this option, and I am sure there are other possibilities. In this way, analysis can precede the selection of telling passages that are worth writing down and translating. To analysis the interviews must lead: gratifying as they might be, they are not an end in itself but just a step towards a result. Excessive quotations—down to ethnographies consisting almost entirely of raw interview passages (Dwyer 1982)—can be taxing for the reader, whose main interest is often the researcher’s voice. In closing, I stress that much of what I do in interviews I learnt on the job and I have perhaps not reflected too deeply on it. The reader should take this as encouragement not to shy away from trying interviews themselves, starting perhaps with those interlocutors who are least likely to be estranged and most open to a second attempt if things do not work out as planned.

199

7.2 The art of interviewing: A Japanese perspective Tomiko Yamaguchi

In recent academic discussions in Japan, the value of qualitative research methods has been increasingly appreciated across a wide range of disciplines, from Sociology and Anthropology to Public Policy, Nursing Science and more. While ‘qualitative research’ encompasses a variety of techniques, including participant observation and text analysis, the focus of this chapter is qualitative interviews: in particular, guiding principles and practical aspects of carrying out qualitative interviews in Japan. I will address here what qualitative interviews are and what we can learn from them, as well as the art of interviewing in Japan.

What are qualitative interviews? A qualitative interview is a research method whereby a researcher seeks answers to an established research question by eliciting people’s thoughts, opinions and attitudes. Put simply, qualitative interviewing is ‘a conversation with a purpose’ (Berg/Howard 2007, p. 105). There are various ways that qualitative interviews are carried out, such as in-depth interviews, which are usually done one-on-one, and focus group discussions. While in-depth interviews are used to gather data on individuals’ perspectives and experiences, focus group discussions use group dynamics to elicit data on the cultural norms and practices of the group and to gain overviews of issues. My own rule of thumb, when a project is relatively new and I am not yet familiar with the topic, is to use the group discussion method to learn who is who in the field and their relationships. My own project that examined nanotechnology in food was new to me in the sense that I did not know key people in the field, so I chose the focus group discussion method (Yamaguchi 2011). The method helped me greatly to learn who is who and identify important issues on the topic. Similar types of information gathering can also take place by attending programmes such as public forums, conferences and seminars; that information can later be used to make an appointment for in-depth interviews. I attempt to do ‘homework’ thoroughly when carrying out a project in Japan, because the fact that I am Japanese and a professional researcher means that there are expectations that I have a decent knowledge about the subject and have maturity of judgement regarding the situation; maturity means knowing what to ask and what not to ask. When dealing with a topic some of whose elements are familiar (e.g. issues problematised within that community, technical words used by informants, or acquaintanceship with at least one or two key informants) and when I feel confident, I choose to use in-depth interviews right away. As the idea of ‘the active interview’ (Holstein/Gubrium 1995) suggests, an interview not only functions as a form of dynamic social interaction between the researcher and the respondent, but also involves the active participation of the respondent in shaping the meaning of the

200

Chapter 7 How to interview people social phenomena studied. Therefore, I try to pay close attention to the timing of carrying out interviews so as to carefully cultivate field relations.

What can we learn from qualitative interviews? First, interviews are a good method to help the voices of marginalised people to be heard, provided that we listen carefully to what they say in order to learn from the perspectives of the people interviewed (see Ganseforth, Ch. 4.2; Slater et al., Ch. 16.2). For example, using a life history approach, Yukiko Araragi (2017) carried out a study of people with leprosy, explicating the experience of people who were forced to live in a sanatorium. Through careful conversation and listening, Araragi built trustful relations over a period of years with people suffering from social discrimination, and thus gathered and reported on stories which otherwise would not have been known by the majority of people. Similarly, by listening and observing carefully, Christopher Bondy (2015) explored the question of what it is like to grow up as burakumin (Japan’s largest minority group and an outcast group) in contemporary Japan. Again, the perspectives of burakumin are rarely heard. My own project that dealt with Latino organic farmers in California was an attempt to understand the livelihood of immigrant organic farmers who are minorities in the organic sector in the U.S. (Yamaguchi 2016a). What these studies demonstrate is that qualitative interviewing can be a powerful tool with which to uncover personal experiences that individuals have hidden, either willingly or unwillingly, often because they are at odds with the dominant norms, values and beliefs of wider society. Second, qualitative interviewing can also enable researchers to understand complex and contradictory social phenomena. Complex problems are often multifaceted; thus it would be useful to carefully unpack the experiences of varying stakeholders situated in differing sociocultural, economic or political contexts. For instance, by examining radiation contamination of food and farms in Japan, some researchers have shed light on the experience of organic farmers after the Fukushima accident (Kimura/Katano 2014; Yamaguchi 2016b), while another looked at the experience of mothers in Fukushima (Sternsdorff-Cisterna 2015), and yet another carried out interviews with public policymakers, scientific experts and consumers (Reiher 2017). We learn from these studies that, for farmers, the major issue has been whether and how to continue farming; on the other hand, for mothers, or more broadly for consumers, the issue has been about the safety of food for their children. Differing takes on the problem suggest multiple and contradictory social agendas, which raises difficult issues for public policymakers who are responsible for finding a way that is agreeable to most people. An interview method which stands on the premise that none of these perspectives is privileged is useful for unpacking complex social reality. Third, qualitative interviewing allows one to go beyond current incidents and to gain insights into past events. Studying a political riot involving Korean Japanese in Osaka in the 1950s, for instance, Sara Pak (2010) interviewed people who were involved in the riot and asked them to reconstruct incidents from their point of view. In short, qualitative interviews enable the researcher to explicate the complexity of the real world by shedding light on matters that are unnoticed or hidden, and by unpacking things that are complexly entangled.

201

Tomiko Yamaguchi

The art of interviewing When carrying out interviews in countries outside one’s own, one may encounter new challenges. Some are practical, such as conducting interviews in a foreign language or with an interpreter, and others have to do with cultural logic and how to navigate interview dynamics in order to overcome cultural differences. I will delve into the latter. To gain access to interview subjects, introductions from someone who knows the people are key (see this chapter, Alexy, Ch. 7.3; McLaughlin, Ch. 6; Yoshida, Ch. 5). This is the case with any fieldwork situation but is especially so in Japan, as noted by a number of U.S. scholars with experience (Bestor et al. 2003b). Many overseas researchers who have carried out studies in Japan indicate that people know ‘how to avoid a substantive response’ if an interviewee is uncomfortable with a researcher (Bestor et al. 2003a, p. 14). One may thus spend hours asking questions, but get practically no answers, or interviewees may give socially acceptable but insincere replies. I almost always look for someone who can introduce me to an interviewee; through that person an interviewee has a chance to find out about me and my intentions beyond what it says in a formal letter of request. Providing this background information before I am physically present helps to set the tone for our eventual conversation. This will be seen as my attentiveness or thoughtfulness (kiokikasu) or even my respect for the interviewee. The other reason that I go through someone is that I deal with socially contentious issues such as the use of genetically modified food and social conflicts over food safety; society is often divided on these issues (Yamaguchi 2014; Yamaguchi/Suda 2010). Therefore, people generally tend to be on guard. Going through someone may ease tensions, and also makes it easier and less embarrassing for an interviewee to decline the interview request if they are not comfortable being interviewed. After all, qualitative interviews are carried out in a conversational style; inevitably they are entangled with cultural norms, and thus paying close attention to what is expected culturally is important. Once an appointment is made, the question is how to conduct the interview. Unfortunately, there is not one perfect way of dealing with interviews because each interview situation is unique and grounded in a particular context. Upon exchanging name cards with you, interviewees will obviously have some sense of your social position, expect that you have done some homework (studying the subject), and expect you to know and respect appropriate social norms, though often this sense of ‘appropriateness’ varies from individual to individual. So you will need to figure things out each step of the way by carefully observing interactions with interviewees. What is required resembles an art—a practice and a habit that reflects connected ideas. There are four pieces of advice I would like to give based on my experience. First, even if you might not be fully confident with your Japanese language proficiency, do not be afraid. Foreign researchers may have an easier time than native Japanese in getting people to open up to them (see Holthus/Manzenreiter, Ch. 5.3). Why is this? In Japan (and in Asia more broadly), if one is ‘a visitor’ or ‘an outsider’, people tend to be more generous and courteous, which frequently translates into willingness to share stories and information with a visitor. When I carried out fieldwork in India during my graduate school days, even without an introduction, I was given opportunities to meet and interview key informants such as farmer leaders and high-profile scientists, and was even invited to their homes to carry on conversations, etc. but my graduate colleagues who were Indian natives were not granted the same opportunities. I have heard similar stories in Japan. 202

Chapter 7 How to interview people Second, it is typical for a researcher to encounter reluctance on the part of interviewees when it comes to openly sharing their thoughts with people they have just met for the first time. Do not be discouraged by such an encounter. As you make progress with your project, it almost always gets easier and faster to establish rapport with an interviewee. Third, when asking questions, start by asking for facts, then gradually move to the questions that seek people’s opinions and ideas. Factual questions are easier to answer than requests for opinions. When a person is not directly responding to your question, it could be an indication that you have infringed on the logic of the individual, or you might have asked something culturally inappropriate. Be thoughtful and respectful, but do not be afraid; even if you make a blunder, native Japanese are likely to assume that, as a foreigner, you were not well-versed in their cultural expectations. As long as you apologise and give the interviewee an opportunity to avoid answering an uncomfortable question, you should be able to part on good terms. You can always try out that question with other interviewees if the question is important, because different people will have different reactions to the same question. Fourth, striking a balance between talking and not talking on the interviewer’s side is important. While it may be useful if you tell people about yourself, your ideas on the subject or about your country throughout an interview, too much talking on the part of the researcher will generate insufficient and sometimes poor data, so remind yourself that you are there to listen. In my experience, when responses are long and story-like, it is an indication that people are comfortable with you. When I hear a comment such as, ‘I’m not sure if I’ve given you the information you need, but it was fun talking to you,’ at the end of an interview, it is a sign that the interview has gone well. Qualitative interviewing is more than meeting people and collecting information about Japan. It is a medium through which to build relations, gain insights and learn from the experience and perspectives of others. What you learn through this method will not only change the way you see and experience the world, but it may also change the people you have interviewed. An interview is an encounter where your ideas and their ideas come together.

203

7.3 Talking through difficult topics Allison Alexy

Although my research includes many methodologies, ethnographic interviews play a key role in how I understand and represent contemporary Japan. I am a sociocultural anthropologist and have conducted two major research projects in Japan. My first project explores divorce in contemporary Japan, or how people decide what makes a marriage worth ending (Alexy 2020). My second project examines ‘parental abductions’, when one parent takes their own child and prevents the other parent from having access to it. Broadly, both of these projects investigate how intimate relationships are shaped by, expressed through and are resistant to legal categories and changing social norms in contemporary Japan. I analyse the ideologies and methodologies that people use to imagine, build, critique and dissolve romantic and family relationships at a moment when both the heteronormative marriage system, and the postwar cultural practices that have sustained it, are increasingly under stress. In my research, I often talk with people about deeply traumatic, difficult or personal topics. In this brief reflection, I describe why and how I use ethnographic interviews as a fundamental element of my analysis. In any humanistic research, the researcher’s own positionality shapes the project, basic assumptions underlying it and people’s willingness to participate in the research (see this chapter, Brumann, Ch. 7.1; see also Klien, Ch. 8.1; Yoshida, Ch. 5). In this way, it matters that I am a white American woman who started learning Japanese in high school. I am not a native speaker of Japanese, nor do I look phenotypically Japanese. I believe that my visible foreignness is helpful in my research because it provides a bit of distance for those worried about sharing personal or potentially stigmatising stories. Multiple people have told me that they imagine a Japanese researcher might judge them more harshly (see Yamaguchi, Ch. 7.2). While I do not believe that to be true, their perception shapes my work and what I am able to learn. In addition to the identifying attributes over which I have no control, during fieldwork I maintain a strict perspective of non-judgement. As people tell me about terrible family conflicts or custody fights, I focus on understanding how they understand the world, responding to what they think is important. Moreover, all my research is structured by my conviction that everyone is constantly interpreting and analysing the world. Throughout my participatory research and interviews, I know the people with whom I am conducting research are smart and thoughtful about their own lives. They do just as much analysis as I do, if through slightly different lenses and with different goals in mind. In my writing, I work hard to represent people’s own interpretations in parallel with mine. In the research moment, this means I am always interested in whatever people want to share with me and I am grateful if they want to correct or redirect my interests. Ethnographic interviews occur, in my work, within a broader context of participant observation. I spent most of my research time hanging out with people and participating in their lives or special events. For any anthropologist, everyday life is just as interesting and important as 204

Chapter 7 How to interview people particular activities. If I really want to understand how divorce changes families, for instance, I need to have a good understanding of people’s regular lives. I know this volume includes other chapters more directly focused on participant observation, but it is impossible for me to think about research interviews outside of that broader context. I can count on one hand the number of people with whom I only recorded an interview and did not spend any additional time. Like other ethnographers, my general rule is to accept any invitation; following this logic, my research has included going to movies, sharing meals, discussing politics and gossiping while watching TV, among many other mundane activities (see Yoshida, Ch. 5). So when I ask someone to participate in an interview, rather than being a one-off interaction, it is often a recorded conversation interposed within our longer relationship. These interviews regularly include reference to shared experiences or mutual acquaintances, or are bookended with their direct requests for my assistance or questions about my life. These relationships enable me to meet potential interlocutors through networks of connection. When I conduct a research interview, I use a format that is intentionally designed to be extremely flexible. I type and print out the questions I would like to ask during the interview, clustering them into broad categories. I bring this printed copy of the questions to the interview and physically put them on the table, facing the other person, so interlocutors can see what I plan to ask. After all, being interviewed by an anthropologist is not a common experience and people feel more comfortable when they know what is coming. Because anyone who agrees to be interviewed is likely interested in helping a research project, I have found that many interview subjects are very concerned about providing me with what I ‘want’ or ‘need’. While I appreciate their enthusiasm, I have developed techniques to convince them that I have no particular goals in mind and am genuinely interested in whatever they want to share. Throughout the interview, I follow their narrative and respond positively to any digressions. For instance, in my divorce research, the first cluster of questions was gathered under the heading ‘marriage’ and included questions like ‘When did you meet your (former) spouse?’ and ‘Why did you decide to get married?’ In many cases, when I asked either of these initial questions, my interlocutor’s response could last more than an hour. These are broad questions that get people talking about the categories of experience my research explores. When they say something surprising or confusing, I probe gently on that question to make sure I understand what they intend to convey. Sometimes my inquiries reflect genuine questions I have—things I am trying to figure out—and sometimes they instead reflect my attempts to confirm something I have already been wondering. For instance, see the exchange between Fujita-san and myself (Alexy 2019, p. 104), an example published in a volume available through Open Access. In that conversation, he uses the term ‘love like air’ and I respond by asking him what that means. At that point in my research, I had heard this phrase a lot and literally knew what it meant, but I needed to confirm my increasing suspicion that this was a key measure some people used to judge their intimate relationships. Mr. Fujita’s answer helped me concretise why intimacy metaphorically described as ‘like air’ was suddenly a flashpoint of tension in the mid-2000s. Stated another way: sometimes, it is important to play a little dumb in interviews, both to make sure you do not push your views on your interlocutor and to get independent confirmation that your ideas hold water. After asking permission, I record all my interviews using a relatively unobtrusive audio recorder that I place clearly on the table, pointing towards the other person. I simultaneously take notes and jot down the time when they say anything of particular interest. This makes it 205

Allison Alexy easier to quickly find choice quotes later in the process. I never, ever look at my own watch because that makes people immediately stop talking, thinking you are bored or out of time. Instead, I always schedule interviews in a very long block of time, so they can go as long as needed. I have also learned that I cannot be hungry going into an interview; even if I am conducting the interview in a restaurant or cafe, I make sure to arrive with a full stomach so I can focus entirely on the conversation, rather than shovelling food into my mouth. I bring a small gift to every interview and offer it after the conversation is over. In Japan, I most typically buy gift boxes of crackers (senbei), figuring that people either will enjoy it themselves or can easily re-gift it. In a nice department store, these typically cost about JPY 2,000, and seem like a small but appropriate gesture of my gratitude. Towards the end of each interview, I always ask a series of questions that have substantially helped me expand my research topics and gather richer data. First, I ask if my interlocutor has any questions for me, and they almost always do. People are legitimately curious about my background and motivations, and I answer honestly. Second, I ask my interlocutor what they would focus on if they were doing this research. This question often prompts excellent ideas that did not come up earlier because I had not thought to ask about them (those pesky unknown unknowns). Finally, I make a point of waiting as long as possible to turn off my recorder because the most interesting comments can come at the very end of a conversation, when everyone is packing up to leave. I always recommend letting the recorder run as long as possible. After the interview, I pay native speakers to transcribe the recordings. When I am analysing and writing from these interviews, I work with the Japanese versions and selectively translate any parts I want to publish in English. When first mentioning your research to people, it can be helpful to use different phrases to describe your project, trying out different options to see how people react. For instance, in my first project, I started by telling people that I was studying ‘divorce’ (rikon) and asking for their thoughts or reactions. While this provoked interesting reflections, after some months, I realised that it was too on the nose and therefore had the social effect of alienating people. If, say, a woman was in an abusive marriage and starting to think seriously about how to address it, she might not immediately identify herself with the word ‘divorce’, even if that might eventually be in her future. Moreover, by introducing my project with that relatively specific term, I unintentionally excluded all sorts of unhappy but ongoing marriages, and all the people who might want to divorce but had not yet done so. In the course of my research, I began to use more inclusive terminology to introduce my project, shifting from framing it as exploring ‘divorce’ to ‘family issues’ (kazoku mondai, which also has the connotation of ‘family problems’ in Japanese). This small shift had the effect of including many more people who were willing to share their experiences with me. I learned a lot from happily married spouses who explained what made their relationships ‘good’ and single people who were willing to share their worries about what was not successful with their parents’ relationship, among others. I am not sure these people would have been as willing to talk with me if I had more narrowly framed my project as only about ‘divorce’. Simply broadening my initial introduction to my project shifted how people responded to it, who was willing to talk with me and the data I was able to gather. In English and Japanese, I am a person who thinks through conversation. My research projects, and indeed my professional identity as an anthropologist, build on this personality trait. Of course, there are many ways to be a successful researcher, but my final recommenda206

Chapter 7 How to interview people tion is to find methods that correspond best to your own tendencies. No matter the topics you hope to explore, and the discipline(s) in which you are working, there are ways to gather interesting data. As this volume makes clear, no singularly ‘right’ or ‘wrong’ method exists for research and instead should reflect the researcher’s own positionality, personality and interests.

207

Further reading Brinkmann, Svend (2013): Qualitative Interviewing. Oxford University Press. Flick, Uwe (2006): An introduction to qualitative research. London: Sage. Fontana, Andrea/Anastasia H. Prokos (2007): The interview: From formal to postmodern. Walnut Creek, CA: Left Coast Press. Gubrium, Jaber F./Holstein, James A./Marvasti, Amir B./McKinney, Karyn D. (2012): The SAGE handbook of interview research. Thousand Oaks, CA: Sage. Kvale, Steinar/Brinkmann, Svend (2008): InterViews: Learning the craft of qualitative research interviewing. Thousand Oaks, CA: Sage. Rubin, Herbert J./Rubin, Irene S. (2012): Qualitative interviewing: The art of hearing data. Thousand Oaks, CA: Sage.

References Alexy, Allison (2019): What can be said? Communicating intimacy in millennial Japan. In: Alexy, Allison/ Cook, Emma (eds.): Intimate Japan: Ethnographies of closeness and conflict. Honolulu, HI: University of Hawai‘i Press, pp. 91–111. Alexy, Allison. 2020: Intimate disconnections: Divorce and the romance of independence in contemporary Japan, Chicago: University of Chicago Press. Araragi, Yukiko (2017): Yamai no keiken o kikitoru: Hansenbyosha no life history. Tōkyō: Seikatsushoin. Barthes, Roland (1970): L’empire des signes. Paris: Seuil. Berg, Bruce L./Howard, Lune (2007): Qualitative research methods for the Social Sciences. Boston, MA: Pearson/Allyn&Bacon. Bestor, Theodore C./Steinhoff, Patricia G./Lyon-Bestor, Victoria (2003a): Introduction: Doing fieldwork in Japan. In: Bestor, Theodore C./Steinhoff, Patricia G./Lyon-Bestor, Victoria (eds.): Doing fieldwork in Japan. Honolulu, HI: University of Hawai‘i Press, pp. 1–20. Bestor, Theodore C./Steinhoff, Patricia G./Lyon-Bestor, Victoria (eds.) (2003b): Doing fieldwork in Japan, Honolulu, HI: University of Hawai‘i Press. Bondy, Christopher (2015): Voice, silence, and self: Negotiations of buraku identity in contemporary Japan. Cambridge, MA: Harvard University Asia Center. Brayda, Winsome C./Boyce, Travis D. (2014): So you really want to interview me? Navigating ‘sensitive’ qualitative research interviewing. In: International Journal of Qualitative Methods 13, pp. 318–334. Brinkmann, Svend (2014): Unstructured and semi-structured interviewing. In: Leavy, Patricia (ed.): The Oxford handbook of qualitative research. Oxford: Oxford University Press, pp. 277–299. Brumann, Christoph (1996): Strong leaders: The charismatic founders of Japanese utopian communities. In: Neary, Ian (ed.): Leaders and leadership in Japan. Richmond: Japan Library/Curzon Press, pp. 175– 189. Brumann, Christoph (2000): Materialistic culture: The uses of money in Tokyo gift exchanges. In: Clammer, John/Ashkenazi, Michael (eds.): Consumption and material culture in contemporary Japan. London: Kegan Paul International, pp. 224–248. Brumann, Christoph (2012): Tradition, democracy and the townscape of Kyoto: Claiming a right to the past. London: Routledge. Brumann, Christoph (2021): The best we share: Nation, culture and world-making in the UNESCO World Heritage arena. New York: Berghahn Books. Castro-Vázquez, Genaro (2017): Intimacy and reproduction in contemporary Japan. London: Routledge. Creswell, John W. (2014): Research design: Qualitative, quantitative, and mixed methods approaches. London: Sage. Dwyer, Kevin (1982): Moroccan dialogues: Anthropology in question. Baltimore, MD: Johns Hopkins University Press. Farrer, James (2013): Good stories: Chinese women’s international love stories as cosmopolitan sexual politics. In: Sexualities 16, No. 1–2, pp. 12–29. Flick, Uwe (2006): An introduction to qualitative research. London: Sage. Flick, Uwe (2011): Introducing research methodology: A beginner’s guide to doing a research project. London: Sage.

208

Chapter 7 How to interview people Froschauer, Ulrike/Lueger, Manfred (2003): Das qualitative Interview: Zur Praxis interpretativer Analyse sozialer Systeme. Stuttgart: UTB. Gerster, Julia (2018): The online-offline nexus: Social media and ethnographic fieldwork in post 3.11 Northeast Japan. In: ASIEN—The German Journal of Contemporary Asian Studies 149, pp. 14–32. Glaser, Barney G./Strauss, Anselm L. (2005): The discovery of grounded theory: Strategies for qualitative research. Chicago, IL: Aldine-Atherton. Hammond, Michael/Wellington, Jerry (2013): Research methods: The key concepts. London: Routledge. Helfferich, Cornelia (2004): Die Qualität qualitativer Daten: Manual für die Durchführung qualitativer Interviews. Wiesbaden: Springer VS. Holstein, James A./Gubrium, Jaber F. (1995): The active interview. Thousand Oaks, CA: Sage. Hopf, Christel (2007): Qualitative Interviews: Ein Überblick. In: Flick, Uwe/von Kardorff, Ernst/Steinke, Ines (eds.): Qualitative Forschung: Ein Handbuch. Hamburg: Rowohlt Taschenbuch, pp. 349–359. Johnson, David T. (2003): Getting in and getting along in the prosecutors office. In: Bestor, Theodore C./ Steinhoff, Patricia G./Lyon-Bestor, Victoria (eds.): Doing fieldwork in Japan. Honolulu, HI: University of Hawai‘i Press, pp. 139–155. Kaufmann, Jean-Claude (1999): Das verstehende Interview: Theorie und Praxis. Konstanz: UVK. Keddi, Barbara/Stich, Jutta (2008): Qualitative Sozialforschung: Erschließung von Wirklichkeit—Auswahl der Fälle—Instrumente der Erhebung. In: DJI Bulletin, No. 82, pp. 1–4. https://www.dji.de/fileadmin/us er_upload/bulletin/d_bull_d/bull82_d/DJIB_82.pdf, [Accessed 23 May 2019]. Kimura, Aya H./Katano, Yohei (2014): Farming after the Fukushima accident: A feminist political ecology analysis of organic agriculture. In: Journal of Rural Studies 34, pp. 108–116. Kottmann, Nora (2016): Heirat in Japan: Romantische und solidarische Beziehungswelten im Wandel. Wiesbaden: Springer VS. Kottmann, Nora (2019): Verliebt, verlobt, allein. Romantische Beziehungswelten junger Erwachsener in Japan. In: Schad-Seifert, Annette/Kottmann, Nora (eds.): Japan in der Krise. Wiesbaden: Springer VS, pp. 99–120. Krauss, Ellis S. (2003): Doing media research in Japan. In: Bestor, Theodore C./Steinhoff, Patricia G./ Lyon-Bestor, Victoria (eds.): Doing fieldwork in Japan. Honolulu: University of Hawai‘i Press, pp. 176– 192. lewallen, ann-elise (2007): Bones of contention: Negotiating anthropological ethics within fields of Ainu refusal. In: Critical Asian Studies 39, No. 4, pp. 509–540. McLaughlin, Levi (2010): All research is fieldwork: A practical introduction to studying in Japan as a foreign researcher. In: The Asia-Pacific Journal: Japan Focus 8, No. 30. apjjf.org/-Levi-McLaughlin/3388/ article.html, [Accessed 12 March 2019]. North, Scott (2009): Negotiating what’s ‘natural’: Persistent domestic gender role inequality in Japan. In: Social Science Japan Journal 12, No. 1, pp. 23–44. Noy, Chaim (2008): Sampling knowledge: The hermeneutics of snowball sampling in qualitative research. In: International Journal of Social Research Methodology 11, No. 4, pp. 327–344. O’Reilly, Karen (2005): Ethnographic methods. Abingdon: Routledge. Pak, Sara (2010): ‘Jijitsu’ o tsukuru Suita-jiken to gensetsu no seiji (Creating ‘facts’: The Suita incident and the politics of discourse). In: Sociology 54, No. 3, pp. 89–104, 173. Reiher, Cornelia (2012): Food pedagogies in Japan: From the implementation of the Basic Law on Food Education to Fukushima. In: Australian Journal of Adult Learning 52, No. 3, pp. 507–531. Reiher, Cornelia (2014): Lokale Identität und ländliche Revitalisierung: Die japanische Keramikstadt Arita und die Grenzen der Globalisierung. Bielefeld: Transcript. Reiher, Cornelia (2017): Food safety and consumer trust in post-Fukushima Japan. In: Japan Forum 29, No. 1, pp. 53–76. Rubin, Herbert J./Rubin, Irene S. (2005): Qualitative interviewing: The art of hearing data. London: Sage. Słuka, Jeffrey/Robben, Antonius (2012): Fieldwork in Cultural Anthropology: An introduction. In: Robben, Antonius/Słuka, Jeffrey (eds.): Ethnographic fieldwork: An anthropological reader. Oxford: Wiley-Blackwell, pp. 1–48. Smith, Sheila A. (2003): In search of the Japanese state. In: Bestor, Theodore C./Steinhoff, Patricia G./ Lyon-Bestor, Victoria (eds.): Doing fieldwork in Japan. Honolulu, HI: University of Hawai‘i Press, pp. 156–175. Spoden, Celia (2015): Über den Tod verfügen: Individuelle Bedeutungen und gesellschaftliche Wirklichkeiten von Patientenverfügungen in Japan. Bielefeld: Transcript. Steinhoff, Patricia (2003): New notes from the underground: Doing fieldwork without a site. In: Bestor, Theodore C./Steinhoff, Patricia G./Lyon-Bestor, Victoria (eds.): Doing fieldwork in Japan. Honolulu, HI: University of Hawai‘i Press, pp. 36–54.

209

References Sternsdorff-Cisterna, Nicolas (2015): Food after Fukushima: Risk and scientific citizenship in Japan. In: American Anthropologist 117, No. 3, pp. 455–467. Takeda, Atsushi (2013): Reflexivity: Unmarried Japanese male interviewing married Japanese women about international marriage. In: Qualitative Research 13, No. 3, pp. 285–298. Timmerarens, Chaline (2018): Brasilianische Migranten in Hamamatsu: Zugang zu einem multikulturellen Feld. In: ASIEN—The German Journal of Contemporary Asian Studies 149, pp. 47–64. Yamaguchi, Tomiko (2011): Discussing nascent technologies: Citizens confront nanotechnology in food. In: East Asian Science, Technology and Society: An International Journal 4, pp. 483–501. Yamaguchi, Tomiko (2014): Social imaginary and policy practice: The food safety arena in Japan. In: Food Policy 45, pp. 67–173. Yamaguchi, Tomiko (2016a): A look at California organic farms. In: Research Institute for Humanity and Nature’s Small-Scale Economies Project Newsletter 3, January, pp. 3–7. Yamaguchi, Tomiko (2016b): Scientification and social control: Defining radiation contamination in food and farms. In: Science, Technology and Society 21, No. 1, pp. 66–87. Yamaguchi, Tomiko/Suda, Fumiaki (2010): Changing social order and the quest for justification: GMO controversies in Japan. In: Science Technology and Human Values 25, No. 3, pp. 382–407. Yamashita, Shinji (2012): The public Anthropology of disaster: An introductory note. In: Asian Anthropology 11, No. 1, pp. 21–25.

210

Chapter 8 How to observe people and their environment: Participant observation Christian Tagsold and Katrin Ullmann

1. Introduction Participant observation is one of the main methods of conducting fieldwork. It helps us generate new topics, ideas and even theories in a unique way. But being in the field is often hard work. It takes time, patience and courage to become part of surroundings that the researcher is probably unfamiliar with, not to mention a great deal of effort to understand their specifics, rules and challenges. Nonetheless, we want to emphasise that fieldwork is still a lot of fun and you can get a lot out of it. Fieldwork gets us ‘irritated and stimulated’ (Bergmann 2006, p. 16) at the same time by giving new insights into daily routines and procedures that we simply could not examine without being ‘out there’ (see this chapter, Ho, Ch. 8.3; Klien, Ch. 8.1; Takeyama, Ch. 8.2). By walking through a specific area, speaking to its inhabitants and dwelling in one place for a prolonged period of time, fieldworkers gain new perspectives, which would not have been possible if they had not left their comfort zones and shed their convictions. Participant observation is about becoming a part of an unknown—often at first glance ‘strange’—culture, milieu or scene. At the same time, we are researchers and have to stick to some important rules when conducting participant observation. Our chapter will give a short introduction to conducting fieldwork and participant observation by discussing its history with a focus on Japan, its theoretical assumptions and its practical requirements. We will discuss how to select a field—which at times is facilitated by coincidences and surprise—and how to gain or maintain access to it. In addition, we will highlight why it is so important to think about positioning ourselves as researchers in the field, and how our decisions during fieldwork affect our research. We are both Westerners and non-native speakers, a fact that influences our approach. Researchers with a Japanese or even an Asian background might find that some of our advice does not apply to them. This certainly is a first indication that fieldwork is not only about ‘others’ but also about ourselves situated in the field. We will give practical advice on how to conduct participant observation and how to document it by taking notes and pictures, using smartphones or recording sounds. We also address ethical considerations when participating, observing and interacting in and with the field.

211

Christian Tagsold and Katrin Ullmann

2. What is participant observation all about? Participant observation requires a specific frame of mind and lots of practice. While this is true for every method introduced in this volume, participant observation most likely stands out as a case with special needs. Even though the term participant observation seems to be self-explanatory—being there, observing and taking part—it can raise tricky questions. To what extent should we participate? Does our involvement not run counter to science as we ourselves get involved in the field and, as a consequence, alter events? As we mentioned before, participant observation is about dwelling, about making contact and establishing a rapport. We aim to gain access to a field in a way that enables us to participate, indeed without disturbing the setting inappropriately. Participant observation is thus related to daily life experiences in various ways. However, without us being visible, people in the field would not react to us and we would have a hard time finding out what they are thinking by just looking at their actions. In contrast, being visible will help us to establish rapport, which in turn will also give us access to the underlying logic of actions. But how can one fit into unknown surroundings while being a curious, yet informed outsider at the same time? H. Russell Bernard (2011) has stressed two very important skills which participant observers have to train: explicit awareness and memory. Explicit awareness, a term he has borrowed from Bernhard Spradley, refers to the ability to focus on the ‘little details of life’. We usually tend to ignore such nuances, since they do not seem important for everyday life. As anthropologists, however, we have to notice these small details. As Clifford Geertz (1987, p. 9) puts it, our research is about understanding culture as ‘webs of significance’ spun by people, in which social interaction takes place. To that end, we have to create ‘thick descriptions’. For example, when someone moves their eyelids very quickly, we cannot be sure whether the reason for this is because something is irritating their eye or whether it is a cultural way to signal or parody something (ibid., pp. 10–15). As a researcher doing fieldwork in a novel setting, it is important to note these details and reflect on them: What is happening? Which questions do we have to pursue? How are these actions significant in a wider context and can thus be part of a thick description? And at what point do our own cultural codes and ideas influence our interpretations? Therefore, participant observation is also a hermeneutic technique about sense-making, recognising, interpreting and understanding signs (Illius 2012, p. 76). Yet, the ability to notice subtleties does not suffice. We have to memorise what we discern in order to document it. This is the first step towards making sense of our observations and fitting them into the bigger picture of culture. A good researcher should be able to memorise a short sentence verbatim and write it down ten minutes later, when the time and situation finally allow for it, without distorting an informant’s words. In formal interviews, recording devices perform this task, but in many instances of fieldwork we have to rely on our faculty of memory. Explicit awareness and memory are best trained by actually doing fieldwork. Empathy, curiosity, a sense for formal frames, etc. are all necessary. Once again, practice and self-reflection are the path to better fieldwork. In other words, a good participant observer not only observes others, but also herself! All these points and requirements may sound intimidating for a neophyte, but it all comes down to practice. Going into the field will teach you what locals expect from you in this very specific situation. This cannot be generalised or gleaned from even the best manuals. Learning about these expectations and catering to them in the field is the first 212

Chapter 8 How to observe people and their environment step in collecting valuable data. We recommend training your professional awareness in daily life situations and observing, for example, routines, habits and communications while waiting for an appointment, travelling by train or other moments when you can watch and learn to be a precise observer. Try to memorise as much as possible without taking notes and without guessing (see below). Becoming a good fieldworker means accepting errors as part of the game. Not every conversation or participant observation will immediately yield great results and lead to new theories. Wasting time, getting ill, forgetting about important questions, losing contact to a key informant—all this can and will happen from time to time. However, such mishaps are quite normal and part of the process. Entering the field is not laboratory work and can sometimes be chaotic (Ehn/Löfgren 2012, p. 273); therefore, one should distinguish between the often-idealised descriptions from handbooks and the ‘messy’ practice of actual research in the field. Key issues Participant observation requires a specific frame of mind and lots of practice. Explicit awareness, i.e. the ability to focus on small details and changes, is important for giving good interpretations and recognising structures. Empathy, patience and learning from mistakes in the field are important. Participant observation is essential for good fieldwork because it can add information to your research that your informants may not be aware of themselves and that you can only gain by observing, reflecting and being out there.

3. Participant observations in Japan: From the 16th to the 21st century Writing on Japan based on participant observation actually has a long history. When Europeans first reached Kyushu in the mid-16 th century, Jesuit missionaries started to send back letters to their order, in which they assessed the chances of converting the Japanese. Jesuits were keen observers who tried to understand others, albeit with the strategic goal of conversion (Rubiès 2017). In the case of Japan, Francisco de Xavier’s letters are an excellent example of this. Late 19th century travel reports often published as books also used participant observation (see for example Schliemann 1995). As many introductions to Anthropology have noted, participant observation first started off as an acknowledged method of Cultural and Social Anthropology with Bronisław Malinowski’s book Argonauts of the Western Pacific (1922) Because of World War I, Malinowski got stuck in the Papua region in 1914. The region was controlled by the British, who did not allow him to return to Europe. Malinowski made a virtue out of necessity and studied the islanders of the region in-depth. The ideal of staying in the field for at least a whole year –four seasons!– to understand the full cycle of rural activities partly originates from his work. Nonetheless, much of current anthropological research is no longer tied to agriculture. Even though it still makes sense to stay in the field for prolonged periods in order to fully understand foreign lifestyles, values and beliefs, the ideal of four seasons has lost its meaning. For a long time, fieldwork based on participant observation in Japan was often tied to rural regions 213

Christian Tagsold and Katrin Ullmann (Hannerz 1980, p. 1). Yet Akiko Takeyama and Swee-Lin Ho’s accounts prove that the division of cities for sociologists and rural regions for anthropologists is severely outdated. Research today does not have to focus on ‘foreign’ cultures; research just ‘around the corner’ might also yield new insights. Participant observation is no longer bound to a specific spatial setting. While Ho, Takeyama and Susanne Klien have done their research in rather well-defined places and settings, such as among urban employees, northern Japan or host clubs, fields are often threaded together by specific questions. For example, Christian’s research (Tagsold 2017) on Japanese gardens in the West spanned three continents and included fieldwork in more than eighty gardens. It even extended to unlikely places such as a factory, where a temporary stone garden was employed by a laminate company to impress customers with the endless possibilities of designing flooring patterns (Tagsold 2016, pp. 293, 294). In his seminal paper Ethnography in/of the world system: The emergence of multi-sited ehnography (1995), George Marcus proposed that anthropologists follow certain concepts rather than stick to a specific place. He gives a couple of threads to take up when suggesting that one could follow ‘the people’, ‘the thing’, ‘the metaphor’, ‘the plot, story, allegory’, ‘life or biography’, or ‘the conflict’ (Marcus 1995, pp. 90–94). A focus on researching concepts, specific mobile groups and challenges instead of places allows the researcher to deal with the complex conditions of the intertwined globalised world we live in today. Global flows of people, ideas, goods, capital, technologies and media, and their unequal distribution, demand fluid, flexible and mobile research methods (Inda/Rosaldo 2008, p. 4; Welz 1998). The development of digital communications technologies made it necessary to open up to a form of research that can cross borders and act in different, but connected places. As a consequence, ethnography becomes multi-sited and fit for the 21st century. Multi-sited, mobile and even virtual ethnography needs to reflect on its methods and fields like every other kind of ethnography. Sometimes it can also be useful to use autoethnography, that is, the observation of oneself and one’s own culture. Key issues Participant observation in Japan has a long history that began in the 16th century. However, it only turned into a well-established method in the 20th century. Today, fields of participant observation are no longer defined solely in terms of place, but rather by the questions that thread them together. New mobile research practices have arisen. They are able to connect different places, follow mobilities or/and add to information that is raised in virtual spaces.

4. Selecting field sites Before we can commence engaging in participant observation, we need to select a field site as a starting point. ‘Field’ has become the term used for the site we want to observe. A field was once a well-defined place like a village. Nowadays, we need to think much more about the boundaries of our fields, as they can actually extend over the entire globe. Yet field sites do have meaningful boundaries, and it is up to us to define them cleverly—otherwise we would lose ourselves. Choosing a geographical field for research can be based on a very pragmatic 214

Chapter 8 How to observe people and their environment decision, for example when a junior researcher has a scholarship for a year abroad in one specific area. But settling for a field that relates to a specific research question can be a very demanding task, while following a research topic through space is often a complex challenge in terms of logistics. This is because fieldworkers do not just study books, but real-life situations that tend to be messy. Sophisticated plans and theories may turn out to be inappropriate when the reality of the field and the practices explored differ from presumptions or there is a lack of access to the field. However, coincidences may offer surprising new paths to follow, and a skilled observer will exploit such situations to find out much more than she could have imagined beforehand. This is why the ideas of lingering, patience and serendipity are so crucial for fieldwork (Ehn/Löfgren 2012, pp. 273–285; Illius 2012, p. 85; see also this chapter, Klien, Ch. 8.1). As Paul Rabinow (1977, pp. 125–126) puts it, significant processes of understanding will sometimes occur when the researcher is waiting and feeling bored or stuck: ‘Slowly and sporadically, I was moving in the kind of understanding I was seeking.’ In fact, just staying somewhere and being alert to unfolding events can open up fresh perspectives. In this way, surprise, chance and attentiveness often lead to new fields and consequently to new findings. Klien (see this chapter, Ch. 8.1) provided an example of the importance of serendipity from her research in Japan. When she as a female was not allowed access to sacred events while researching the Shinto mountain festival, this was not a setback in the end as it turned into a chance to generate fresh research opportunities. In his attempt to do research about the social strategies of young people in a neighbourhood of Chicago's South Side in 1988, French anthropologist Loïc Wacquant (2010, p. 9) joined a boxing gym and became an accepted member of the local boxing community. After 16 months of practice, he changed his original research topic into a ‘sociology of boxing’ (ibid., p. 11). With the approval of the boxing community, he turned into a fully fledged boxer fighting in the ring. Wacquant then made the best use of his own bodily experience, the thrill of being a boxer and all the insider information he was able to gain. He turned his data gleaned through ‘observant participation’ (ibid., p. 12) into an impressive book and thus proved that coincidental access to fields can lead to deepened theoretical understanding if a fieldworker stays alert and open to any kind of new information. These examples prove that being open, flexible and most of all attentive are core skills in successful fieldwork. Of course, being open and attentive does not mean going into a field entirely unprepared; you still have to do your homework. Fieldworkers should certainly acquire language skills and knowledge about political, social and economic conditions, while also keeping in mind social and cultural rules, habits and taboos. In addition, pre-existing research on the region/field will yield valuable insights and help to avoid conflicts. However, being open to surprises and questioning pre-existing knowledge concerning the field are indispensable. Ho’s research on Tokyo’s working life and nightlife and its influence on gender roles as well as Takeyama’s fieldwork in Tokyo’s host clubs and her findings about affective sensations are excellent examples of such an attitude (see this chapter, Ch. 8.2; Ch. 8.3). Openness and (self-)reflexivity in specifying the field as something that does not exist ‘beyond the imagination of the ethnographer’ (Madden 2017, p. 38) have become even more crucial since the postcolonial and postmodern turns in Anthropology led to a heightened awareness of often stereotyped ‘localization of “natives”’ (Clifford 1997, p. 78). Reproducing cultural stereotypes and established power relations must be avoided (ibid.; Robben/Sluka 2012, pp. 215

Christian Tagsold and Katrin Ullmann 18–20). If we follow this train of thought, participant observation is not concerned with dreams of investigating as yet undiscovered and isolated fields. Rather, our task is to generate a useful construct that helps us hone the research as a defined ‘synthesis of concrete space and investigative space’ (Madden 2017, p. 39). Furthermore, we must document the processes of construction in the most transparent way possible (see Reiher/Wagner, Ch. 16). Key issues The selection of a field is about finding a good point to start from in terms of place and topic. Although there is a lot of preparation to be done in advance, it is important to note that even the best plans will have to be adapted at some point. Being open to coincidences and serendipities helps to refine research topics and reshape the field. (Self-)reflection is important because choosing and defining an appropriate field might otherwise reproduce cultural stereotypes and established power relations.

5. Gaining access The next step after identifying a field where you can begin to search for answers to your research questions is gaining access. If, for example, you want to research how cosplayers organise their conventions in urban Tokyo, simply going to one of these events is the most obvious way to start. In many fields, being non-Japanese will be enough to attract some basic curiosity from the people involved and give you a head start in establishing a deeper rapport. Rapport has long been a keyword for good access to the field. The term describes the ability to connect to locals, show empathy and gain trust. Only then will a researcher be able to elicit the information and feelings that lead to new insights. However, some fields are not as accessible as a public convention and require more developed skills to be accessed. People you encounter in the field may not be interested in you at all or may even distrust you. Luckily, from the various experiences of Christian’s fieldwork, Japanese people’s interest in foreigners is usually substantial enough to overcome initial thresholds. But bureaucratic fields can be quite different. Often enough, you will need to identify yourself and persuade gatekeepers, as we call people who control access to the field, to grant you access. In these cases—and generally in Japan—having well-printed business cards (meishi) with a thorough Japanese translation on the reverse side is an important precondition for building up professional credibility. Furthermore, a letter of recommendation in Japanese can be very helpful in overcoming obstacles. Knowing people with good networks is important everywhere around the world. But this is especially relevant in Japan, since being introduced by a trusted business partner, advisor or friend usually opens up almost any field. Thus, kindly ask professors at home and in Japan to help with networking. Obviously, in these cases preparing some sort of interesting omiyage is a good way to show gratitude and will facilitate conversation in the beginning (see Alexy, Ch. 7.3; Kottmann/Reiher, Ch. 7; McLaughlin, Ch. 6). Finally, the obvious inroad is having skills and knowledge connected to the field. When one of us, Christian, was visiting a residential care home for elderly people suffering from dementia, he first gained access through a colleague. However, his ability to bake a German cake togeth216

Chapter 8 How to observe people and their environment er with the elderly patients enhanced his contact to the caregivers considerably. Still, fields can ask for skills you do not have! Studying in rural areas often entails talking to people with a strong dialect, which will limit the degree of your communication skills considerably, even for those of you who speak Japanese very well. Here, the best solution is to spend time and learn the dialect as much as possible—and this, by the way, is what the classic anthropologists did, who often did not even speak the language of the field at all before going there. Often though, teaming up with an informant who is well versed in both the dialect and standard Japanese can help as she can translate for you. In general, it is a good idea to look for one privileged contact, someone whom you get along with well and who understands your research goals. This is also a typical strategy of experienced anthropologists in the field. Key issues Rapport describes the necessary ability to connect to locals, show empathy and gain trust. It is very important to bring in your personality and skills to be able to connect to people and encourage them to spend time with you. Think about who you are and what you can offer to the field. Whom do you know who could help you to gain access or have valuable contacts to the area, the topic, etc.? Recommendations, mediation and credits are valuable. How does the field you want to enter work? What can help you gain credibility?

6. Ethical implications In the wake of postmodern and postcolonial critiques of fieldwork, important debates on ‘writing culture’ and a new awareness of power relations, the notions of ethics, fieldwork and participant observation have shifted considerably in recent decades (Clifford/Marcus 1986; Robben/Sluka 2012, pp. 18–23). As a consequence, professional standards for conducting fieldwork were established: from making prior decisions about field investigation itself to writing and communicating the findings later on. The American Anthropological Association (AAA) regularly publishes a Code of Ethics, which states the core ideas about the professional responsibility of anthropologists (AAA 2012). Obviously, the AAA standards are valuable for everyone who engages in fieldwork and employs participant observation as a method (see Reiher/Wagner, Ch. 16). The points regarding a researcher’s responsibility towards informants and the fields of research are particularly relevant. The AAA (2012) reminds us to be ‘sensitive to the power differentials, constraints, interests and expectations characteristic of all relationships’. This requires reflection on the potential consequences and impacts research may have for and on the individuals, communities, identities, institutions and environments involved. Furthermore, we are reminded to be aware of ‘possible ways that the research might cause harm’ (ibid.). Therefore, it is important to disclose the general aims of one’s research and protect those who participate in it. Participants should be informed about ‘the purpose, methods, outcomes and potential sponsors’ of the investigation and be able to consent with clarity to their participation (ibid.). Rules of anonymity and credits should be negotiated in advance and renegotiated during the research process if necessary. The informants’ privacy should be protected through the use of 217

Christian Tagsold and Katrin Ullmann pseudonyms (ibid.). In her chapter, Ho states that ‘[a]ccess is both a privilege and a responsibility.’ To sum up, participant observation and fieldwork can be considered interrelated processes that involve social relations to others and therefore need to be conducted in an ethical and responsible way. Japan is probably not the most complex case for applying the standards of the AAA—for example, PR China poses much more pressing questions, especially in regions with minorities such as Uygur. Yet we can easily think of fields in Japan which need extra care too. Doing research with people like the homeless or sex workers can quickly lead to conflict with official institutions and cause harm to research participants if you do not take care (see McLaughlin, Ch. 6). This responsibility towards our research partners is closely intertwined with the idea of reciprocity. Reciprocity means that researchers have to give something back to their research partners for the data they acquire. This can include drawing public attention to a specific problem as well as (financial) compensation for informants (AAA 2012, p. 22ff.). Fieldwork may also entail some sort of collaboration and partnership with informants (see Ganseforth, Ch. 4.2; Slater et al., Ch. 16.2). In many cases, the lives of potential informants are not strictly separated from the researcher’s own life. The informants will often be able to access our publications, for example through the Internet. While the latter can be a great form of reciprocity, it also demands careful anonymisation in order to protect research participants’ privacy. For the protection of their privacy, responsible storage of research data is also essential (see Reiher/Wagner, Ch. 16). Key issues Assumptions and attitudes towards fieldwork have changed over recent decades. In this process, ethical questions have become central. We should reflect carefully on our responsibilities to informants as well as the field during the whole research process. This includes transparency about our research as well as considering compensation for the informants’ time and expenses. The protection of data and anonymity, and the possible impact of our fieldwork have to be taken into account.

7. Positioning oneself in the field When entering a field site, researchers learn a lot about their new environment. But in order to do so, they have to become visible as people themselves (Bourdieu 2005, p. 404; Schlehe 2008, p. 139). The idea of the neutral observer who invisibly studies ‘the other’ while adding nothing to the field is a futile ideal. Such an approach will not invite others to open up and feel comfortable (ibid.). Apart from that, hiding oneself is simply impossible over an extended period of time—especially as a Westerner in a country like Japan. The Swedish movie Kitchen Stories (Hamer 2003) convincingly and comically narrates the story of a completely failed non-participant observation project and is worthwhile for anyone who wants to understand why this seemingly neutral approach does not work. Fieldwork does not take place under laboratory conditions. As a consequence, fieldworkers have to bring in their own personality. Furthermore, and even more importantly, they are at218

Chapter 8 How to observe people and their environment tributed to and perceived by different, intersectional, intertwined settings that influence the way they are addressed and accepted by potential informants. This can entail freedoms or restrictions in terms of the way they are able to explore or even access the field they wish to enter and examine. At that point, we have to reflect on our own status and personality in relation to the fieldwork at hand. For example, doing research on marriage in Japan as a married woman puts the researcher in a specific relation to the (un)married informants (Kottmann 2016). Likewise, conducting research on the love lives of employees in Tokyo will yield different results depending on the researcher. Ho’s fieldwork (see this chapter, Ho, Ch. 8.3) vividly proves this point. Thus, our gender, race, power and educational background, profession, nationality, age, etc. all influence how others in the field will deal with us and, consequently, impact our findings. A focus on power relations during a study, for example, can help disclose our status and position to the field and our informants. Possible forms of relations are: 1. Studying down denotes research relations where informants are typically less educated or empowered than the researcher. An example is Stephanie Osawa’s (2018) research on deviant students in a school in Tokyo. Osawa’s position as a researcher employed at a German university and as an adult obviously entailed more power than the position of her informants. This was underlined by the teachers introducing Osawa to her informants. 2. Studying up works the other way around. Here, the researcher has less power and influence, for example in research on political or economic elites (Gusterson 1997; Nader 1972). Studying up is not that rare in Japan, though it has been rather untypical in conventional Anthropology, which can usually be considered as studying down. In the Japanese case, studying up also raises problems of language. Usually, we would have to use very polite forms to address our informants, not only because we are outsiders, but also because of status differences. However, seeming too clever and native might have the adverse affect, because Japanese usually do not expect foreigners to be fluent in grammatical politeness. 3. Studying sideways means that the researcher and the informant in the field have a comparable status (Hannerz 1998, p. 109; Marcus 2006, p. 21; Rao 2006, p. 27). One of this chapter’s authors, Katrin, conducted participant observation studying sideways. She wanted to study members of the mostly well-educated, widely travelled, transnational backpacker scene. Therefore, it was necessary for her to represent similar experiences in travel to gain accepted status among her target group (Ullmann 2017).1 As a consequence, it is essential for us to reflect on our major traits, which people will perceive and maybe react to when facing us as a fieldworker. Reflecting on our position is not only valuable for collecting data. It is also part of our analysis later on as it helps us to evaluate social relations in the field (Robben/Sluka 2012, p. 2). Former fears of ‘going native’, meaning that the researcher loses her professional distance and gets taken in by the field, are often contested today. Without getting temporarily lost, deeper insights will be hard to gain. In order to write about ‘commodified romance’ in host clubs in Tokyo, Takeyama, for example (see this chapter, Ch. 8.2), let herself be seduced emotionally and sensorially while developing ‘affective ethnography’. Even though this is not a research practice that we would recommend for beginners, it shows what is possible without losing

1 There exist a number of other terms and concepts to describe the way research is done, e.g. ‘studying through’ (Hannerz 2010).

219

Christian Tagsold and Katrin Ullmann focus when you are an experienced and self-aware researcher. In most situations, less involvement is sufficient and appropriate. ‘Selective authenticity’, a term from counselling (Amendt-Lyon 2000, p. 55), might be helpful here. For fieldwork, it means that you should be fully yourself while doing fieldwork, but like in every human relationship you can choose which aspect of your personality you actively want to involve in a relationship and the field. How much you get involved of course depends on you, your subject, and your ability to stay focused as a researcher and to protect yourself and the people involved. Key Issues Getting involved is essential to get any useful information. Fieldwork does not take place under laboratory conditions. People will reacting to you anyway; when you hide as a person they will not open up. Think about the (power) relations involved and how to deal with them. Reflecting on your position in the field will help you to collect better data and to do better analysis.

8. Cell phones, writing pads and field notes Traditionally, participant observation has been documented in field notes collected in a field diary, and this practice still holds up today. Of course, we now use cell phones to make notes or record voice memos and laptops to write our diaries. Nonetheless, field notes should be written immediately after finishing the observation process. Memories and other notes can be backed up by jottings from the field. Carrying a small notepad to instantly jot down an impression, a quote or a sketch is highly beneficial and can be expanded into more extensive field notes afterwards. In some fields – if, for example, we wanted to know how Japanese behave in public libraries – it is completely natural to use a laptop from the start. In other settings, even a traditional pen and notepad might be disturbing, and we have to rely on our memory (see Kottmann/Reiher, Ch. 7; McLaughlin, Ch. 6). As a rule of thumb, taking down field notes after observation takes as long as observing itself. We should be prepared to roughly spend another hour on writing field notes based on observing an NGO meeting for one hour. Sometimes observing takes the whole day, because our role in the field does not allow us to leave and start documenting our observations right away. In these cases, it is wise to focus on specific moments and events because it is simply impossible to remember everything after a full day of work. In any case, our field notes should be rich in detail and add a sense of how we came to conclusions on the spot. Noting that ‘the old man looked puzzled’ is unlikely to be very helpful halfa-year later when you start writing your thesis. Instead, recording more details about what made you think that this man was puzzled – raised eyebrows or a comment to his wife – will provide much welcomed information when carving out the significance of this moment at a later stage. A better description of how you came to the conclusion that this was an old man – roughly how old? – will also be very valuable in the end. Going into detail can be quite tiring at times, but without the discipline to write down our observations accurately, participant observation will not yield good data. 220

Chapter 8 How to observe people and their environment Personal feelings are an important part of field notes. We should make the best use of our own subjective views and emotions, as illustrated by Takeyama’s account (see this chapter, Ch. 8.2) of ‘affective ethnography’ (Robben/Sluka 2012, pp. 20–23). Your own relation to a situation is not entirely objective, but it may be useless, or even disturbing, if you do not learn to observe yourself. It is important to take down what, for example, made you ‘irritated’, and how this feeling came about and affected you. Otherwise, this bit of information is likely to be of no worth in the future. When you are documenting spatial settings, photos are an important supplement to traditional sketches. Indeed, not using a cell phone to take photos or short movies will often enough mark you as an outsider today, especially in high-tech countries like Japan! Owing to modern media we are able to take the field with us in various ways and, for example, learn about changes in a community via social media or call an important informant when we realise that we forgot to ask a crucial question. In contrast to the early days of ethnography, these extended connections demand researchers to think carefully about how they want to go on with the relations they have formed during their fieldwork (see McLaughlin, Ch. 6). Key issues Fields and fieldwork change along with technology. This can be helpful when jotting down notes, as notepads and laptops are nowadays used nearly everywhere and thus can be employed without causing confusion in the field. Social media is helpful for staying in contact after the end of fieldwork, though we have to take care to what degree we want to let the field intrude into our daily life as researchers at home and how much time we can devote to staying in touch.

9. Summary Participant observation can be a lot of fun—and sometimes one even has to be careful about becoming too immersed in it! Documenting the process and taking field notes certainly feels more demanding and tedious, but they are extremely important steps in transforming observations into useful data. And participant observation quite often yields data which could not have been attained through other methods, thus producing rich and lively examples of ethnography. Students often assume that participant observation is too subjective and introduces too much bias into research, since we get directly involved with our informants and even affect their lives while in the field. But it is precisely this kind of immersion and reciprocity that helps us to see and understand things which may otherwise remain concealed. The great potential of participant observation and fieldwork is that there will be surprises which will inevitably lead to new results and findings. Furthermore, personal experience and its detailed codification in field notes make excellent material for presentations, papers and theses. Anthropologists usually quote their own field notes in well polished papers and books in order to add lively impressions. Field notes can also 221

Christian Tagsold and Katrin Ullmann give a vivid account of the researcher’s own problems in grasping the meaning of situations and coming to terms with local sense-making. Examples of this writing strategy abound. Sometimes writers are more concerned with establishing their authority in relation to their readership and sometimes they just aim at conveying data. But when employed mindfully, field notes can be extremely convincing. Wacquant’s above-mentioned book Body & soul: Notebooks of an apprentice boxer (2006) is one such powerful account that uses field notes to a rich and intoxicating effect. Wacquant succeeded in demonstrating his own, occasionally painful conversion into a boxer in Chicago through many excerpts from his field diary. His textual strategy certainly is debatable, but that is precisely why his book offers a good starting point for reflection on field notes as part of a finished text.

222

8.1 Of serendipities, success and failure and insider/outsider status in participant observation Susanne Klien

‘You are female, aren’t you?’ I will never forget that question asked by the local government representative I was talking to on the phone back in 2009 in order to inquire about permission to conduct fieldwork during a renowned local Shinto mountain festival in Okinoshima, Shimane Prefecture. It turned out that I could not actively participate in the festivities as it was a sacred Shinto event, with women only taking part in backstage preparatory roles. I remember my initial frustration, thinking that this was a major setback for my research. In hindsight, however, I found out that I was fortunate. My partner (who hardly spoke any Japanese but actively participated) ended up with black and blue knees and aching muscles all over as he had joined the team that did the two-day preparatory work for the festivities. I, on the other hand, accompanied the team throughout, but was not actively involved and focused on observing, gaining important insights in the process. This episode shows that what may be perceived as a setback may eventually turn out to be a merit. Similarly, Thomas Hylland Eriksen has observed that ‘…Even unsuccessful is rarely entirely unsuccessful, and it is often stressed that the notorious ability of anthropologists to make fools of themselves in the field […] can actually be a methodological advantage’ (2009, p. 54). The above episode also highlights the broad range of roles that participant observation may entail. Being actively involved is usually perceived as a key feature of fieldwork; yet, external circumstances may limit the scope of participation. Evidently, being a member of the team preparing the festivities is bound to give you insights that cannot be obtained by being a mere observer. Spending extensive periods of time with the group one is researching; sweating and downing sake together with locals creates a level of visceral rapport that is beyond reach for armchair researchers. However, ‘simply hanging out’, lingering at the site may in fact give you a larger analytical sense of the site and its people. It may in fact yield valuable insights into how people relate to one another, communication patterns, power relations and other key features of social life. Thus, the next section is concerned with the concrete process of observation.

How to observe Few ethnographers would contest Eriksen’s description of fieldwork as a ‘time-intensive enterprise’ (2009, p. 49). After all, researchers spend months, if not years, at their sites. Compared to other research methods, fieldwork involves large amounts of individual time spent researching a relatively small sample of interviewees. Whereas sociologists and other researchers relying on quantitative methods regularly criticise this small sample, this constellation has afforded ethnographers the leeway of implementing in-depth observation of their chosen sites (see 223

Susanne Klien Jentzsch, Ch. 6.1). Ethnography is an effective research tool that facilitates systemic observation, which does not rely on first (or second) impressions, but combines the researchers’ extensive observations with interviewees’ narratives, often blending these multiple layers of meaning into an incisive analysis. Once you enter the field, many things that you observe as a researcher may strike you as odd. At this stage, I would say that the key to turning your fieldwork into a success is to sharpen your sense of your own view and its limitations—‘seeing one’s seeing’ (jibun no mie o miru), as Fumitoshi Kato (2009, p. 51) incisively puts it. Certain assumptions that we take for granted may not be valid in the field; expanding your self-reflective skills is essential. Bronislaw Malinowski (1922, p. 2) famously pointed out that ‘foreshadowed problems’ should be a key feature of fieldwork, yet ethnographers are bound to encounter unanticipated occurrences, phenomena and events. From my own experience, it is these serendipities that often, literally, open doors to new networks, interviewees and insights. I vividly remember standing on a square in Ishinomaki, Miyagi Prefecture, during lunchtime, wondering where I could get a bite to eat. Just then, three elderly local women appeared, opening a door to a place that looked like a private house. I seized this opportunity to ask them whether that place was a restaurant; they nodded and asked me to join them. This moment eventually resulted in me eating a large part of the ladies’ lunch and befriending the restaurant owner, who became a key interlocutor and, incidentally, my home stay host in later phases of my fieldwork. This episode indicates that during fieldwork, acting on the spot, behaving a tad beyond what is considered socially appropriate is often productive. It does not need to be emphasised that respecting the local community (or the group of your interlocutors) should be accorded the highest priority at all times during fieldwork. In order to seize such moments effectively, however, we need to sharpen our sense of visceral instinct, something we have often unlearned in academic contexts. I generally recommend lingering around wider rather than smaller groups of people, especially in the initial stages of your fieldwork, but starting with a clearly bounded group may be helpful. After some time spent in the field, more specific sets of questions and issues will emerge that strike your research interest so that you can narrow down your themes. For example, during my ethnographic research into disaster volunteers after the Great East Japan Earthquake in March 2011, I started by signing up as a disaster volunteer myself. First, I mostly spent time with other disaster volunteers, but later expanded my fieldwork to disaster volunteer coordinators, local government representatives, local residents supporting disaster volunteers and local residents who had no personal relations with volunteers. Engaging with diverse groups of interlocutors gave me a more comprehensive view of the phenomenon I was researching, i.e. the motives of disaster volunteers to embark on altruistic work. In a similar vein, Roth observes about his fieldwork among Japanese and Japanese Brazilian workers in Japan that ‘[b]y talking to as many of these workers, bureaucrats, and other intermediary cultural brokers as I could early on in my research, I was able to get my bearings in the field site more quickly than I could have on my own’ (Roth 2003, pp. 343–344; see Goodman, Ch. 1).

224

Chapter 8 How to observe people and their environment

Challenges during preparation and implementation As emphasised by numerous other researchers, the way ethnographers enter the field considerably shapes the research to come. The position of the person introducing the researcher to others in the field, the gatekeeper, must not be underestimated. For example, during my recent fieldwork with hip-hop practitioners in Hokkaido, northern Japan, one of my key interlocutors and gatekeeper was a local male who had previously been a DJ in the club that was my key field site. On the one hand, the fact that he was a veteran practitioner gave me access to events and parties that were only open to members of the scene, as such events were disseminated by word-of-mouth (Klien 2020). However, at the same time, the fact that I attended many events together with him placed me in a special segment of the scene. In other words, I recommend making informed choices when choosing gatekeepers and keeping in mind that there are always downsides, regardless of what choice you make. I believe that we can minimise such downsides by behaving in an accessible (i.e. open and amicable) manner when hanging out and encountering members of the site that we are researching, although we have to accept downsides as facts of life. Related to this, from my own ethnographic experience, I would like to argue that the most insightful fieldwork results from a balance between the researcher’s insider and outsider status. As Ian Reader (2003, p. 103) has previously argued: ‘By contrast, my outsider position worked to my advantage, as my Japanese academic colleague recognized, I had been able to make a suggestion and get away with what could have been an indecorous request’. Laura Dales and Beverly Anne Yamamoto similarly note that some of their interviewees ‘would not have felt comfortable sharing certain stories with us had we been Japanese’ (2018, p. 242). Evidently, being an outsider does entail numerous advantages that we need to work with in order to obtain incisive results as fieldworkers. During my fieldwork into the hip-hop scene, I aimed for a balance between inconspicuousness and out-of-the-box questions. My gender turned out to be more of a merit than a disadvantage as male practitioners seemed eager to share their thoughts with a female outsider unrelated to the scene—something I had not anticipated at the beginning of my fieldwork.

General recommendations To sum up, I will provide four pieces of advice for a smooth fieldwork experience. First, as previously indicated by Joy Hendry (2003, p. 69), seizing chance encounters is key to successful ethnographic research. I encourage all fieldworkers, especially neophytes, to overcome inhibitions to talk to strangers. As outlined in my episode with the senior ladies in Ishinomaki, I would not have met my long-term host and gatekeeper there had I not followed them on the spur of the moment (Klien 2016a, p. 44). Second, verbal statements by interlocutors need to be taken with a grain of salt. Carla Freeman emphasises the importance of ‘the ways in which they spoke, the timbre and lilt of their voices, the intensity of their expression, and the look in their eyes’ (2014, p. 135). Make sure that you do not focus too much on oral narratives, but also observe people’s facial expressions and body language (see Kottmann/Reiher, Ch. 7). I remember a group interview in northeast Japan after the Great East Japan Earthquake in March 2011 that I was conducting with two

225

Susanne Klien colleagues, one of whom did not speak a word of Japanese. Sharing our impressions after the interview was a real eye-opener since the colleague with no Japanese language skills was forced to focus on facial expressions and body language. Thirdly and related to that, I highly recommend talking to key interviewees several times and arranging meetings with them outside the field, if possible (Klien 2016b, p. 362). The field shapes interlocutors’ statements to a larger extent than we assume, especially in rural communities. During my fieldwork with lifestyle migrants in Tokushima Prefecture in 2017, a female settler described her life in her newly chosen community in overly positive terms, although she did refer to some challenges. Six months later, we met again in Hokkaido. It turned out that she had decided to leave the community soon after our meeting as she generally disliked rural life. In a small rural town, she felt that she could not talk freely about her real thoughts. In other words, talking to interlocutors outside the field may provide more nuanced insights. Last but not least, listening well is a skill that is usually taken for granted, but is the key to gaining in-depth insights during fieldwork.

226

8.2 Doing and writing affective ethnography Akiko Takeyama

A splendid wonderland awaits at the bottom of a long staircase as I descend from the midnight darkness. On the dance floor, well-dressed men and women hold each other closely as a romantic melody plays. Under dim light, mildly intoxicated couples flirt. At one particularly lively table, a handful of young men surround a woman to entertain her. A man kneels down with a lighter on his palm waiting for a woman to put a cigarette in her mouth. Another man with a steaming towel in hand waits next to the powder room for a lady’s return. This is a scene in a Tokyo host club. The men serving the women are called hosts, or hosuto in Japanese. Their job is to sell love, romance, companionship and sometimes sex to women for exorbitant sums of money. Deploying stylised masculinity—feathered fringes, polished nails and fine European suits—they serve women attentively and seek their fortunes assiduously. Women of a wide range of ages visit the club to escape from their everyday lives and indulge themselves in fantasy. In and outside the club, these men and women mutually seduce one another to foster a commodified form of romance. I closely studied the art of ‘love business’ for my book Staged Seduction: Selling Dreams in a Tokyo Host Club (Takeyama 2016). I asked: What does it mean for these actors to engage in commodified romance? What are the hosts and their female clients getting out of the seemingly feminised labour and apparently fake love, respectively? What does it say about the changing gender dynamics, lifestyle consumption and market economy in contemporary Japan? To answer these questions, I conducted ethnographic fieldwork in Tokyo for 36 months. My research methods included participant observation (observing and analysing social events and activities), in-depth interviews and extensive ‘hanging out’ with my informants in order to see things from their perspective. Through my fieldwork, I have found that romance, which evokes anticipation, is intertwined with the future-oriented aspiration of not only individuals but also of the hosting business and, by extension, Japanese society at large. The aspiration process itself is capitalised on in post-industrial society: seducing people out of the present and into a future where hopes and dreams are imaginable. Some might ask: How can you study something as vague and undefined as the art of seduction in future-oriented aspiration? What does it look like? What kind of fieldwork does it entail? How do you write about the findings? My answer is to do and write what I call ‘affective ethnography’. Affective ethnography is a tool to invoke one’s own feelings to sense and study the often invisible dimensions of human experience. It is about vicariously experiencing what it is like to be seduced by, and seducing, another person into acting out for one’s own, as well as the other’s, ends.

227

Akiko Takeyama Let me share how I was seduced into a host club in the field. One late Saturday night in September 2004, I walked alone into Tokyo’s Kabuki-chō red-light district, the bustling centre of Japan’s sex and entertainment industry. When I arrived, I was hit by the lively sounds of hundreds of people on the streets, solicitors shouting, computer games and the colourful arrays of billboards and neon signs. It was getting late and I was feeling a bit fatigued by it all, yet a strange excitement kept me awake. ‘Hey lady, interested in a host club?’ a young man in a black designer suit addressed me off guard. Across the district, men in such suits attempt to lure salarymen and other passers-by into hostess clubs and pornographic peep shows. This man, however, was different. He was a host who advertised both himself and his club to passing women. I became nervous. I did not intend on visiting a host club that night. I planned to observe the street scene only as an entry point to my future study. Why did he approach me? I need to be cautious around men like him. But then again, this would be a great opportunity to learn more about the hosting business. Should I just walk away or ask where he works at least? Up close, he looked different from other hosts, who wore gaudy accessories and had bleached hair. He was very polite. ‘Well, I am researching host clubs. I came here tonight…’ ‘Why don’t you come over and see my host club? It’s only 5,000 yen for the first visit.’ ‘Where do you work?’ I asked. ‘I work for club Orion.’ ‘I have heard of it! It’s a famous one, isn’t it?’ ‘Yes, it is a long-standing club and known for its fair business practices. You can trust us.’ I decided to follow him. His congeniality and openness about his life put me at ease immediately. His host name was Shin. Passing several clubs and bars on the way to Orion, Shin and I got to know each other. We learned that we both grew up outside of Tokyo and sought alternatives to conventional sex roles—salaryman and housewife—in Japanese society. With our growing rapport, the district’s night scene no longer felt alien to me. Opening Orion’s heavy door, I felt as if a theatre curtain had been lifted. As I walked in, a couple of hosts welcomed me with deep bows. Once I was seated, Shin swiftly sat down next to me. His three ‘helper hosts’ followed to assemble beside us. One carefully laid a lace napkin on my lap. Another started to make drinks. The other handed me a steaming hand towel and asked, ‘What is your name?’ ‘Akiko.’ ‘No wonder you are so pretty! There are statistics showing that women named Akiko tend to be beautiful.’ I doubted there were such statistics. The helper who was making drinks jumped in, ‘Haven’t you ever been told that you look like a TV announcer?’ ‘Um… no, not really,’ I said. ‘You know, the kind of announcer at a local station who is lovable … but not quite sophisticated!’ I could not help bursting into laughter. So did everyone at the table. ‘Cheers!’ They gave me a toast. At this point, I explained my intention to conduct participant observation and obtained informed consent for my research (see Slater et al, Ch. 16.2). As both a researcher and a client, I tried to observe the club scene and examine the hosts’ art of seduction as much as possible. I devoted my initial attention to the space’s theatrical effect and the minutiae of people’s flirtatious interactions. While making mental notes, I noticed my knee was slightly touching Shin’s. I straightened myself and slid my leg away. As I was drawn back into the conversation at our table, my trouser leg once again rubbed against his. Shortly after, Shin inched slightly closer and leaned over to me. ‘Are you having a good time?’ he murmured into my ear. His whisper left a ticklish sensation and the sweet fragrance of his cologne that entranced me. The subtle 228

Chapter 8 How to observe people and their environment interaction dramatically transformed the club’s open space into an intimate fantasy world, wherein Shin’s move, accidental or not, seemed deliberate. This experience allowed me to see questions I would ask of women who engaged in this service, such as, did it matter whether every aspect of their experience was motivated by money or not? Thus, my own affective experience, and my later reflections on it, fed into the interviews and analysis in later stages of the research. Intimacy aroused because of—not despite—the presence of others who noted our secrecy but left us alone. In the imaginary world, my sensual experience and cognitive interpretation felt all-encompassing. While I tried to make a distinction between what is performed and what is not, that line became blurry. In this circumstance, nothing seemed clear-cut. There was no way to solicit the hosts’ real intentions behind their bodily and speech acts and test their truthfulness. Affective ethnography, unlike the more traditional approach that prizes objective knowledge, ponders what is unknowable—the territory that is somewhat like the experience of an eclipse blocking out the light of clear sight. The unknowable territory of ‘eclipse’, however, did not keep me from seeking further. Instead, it fuelled my pondering about the possibilities of my future research-and-adventure. This research experience is, by no means, universal. It is not even a typical one in my own career. The experience, especially a sensory one, is contingent on who you interact with, when, where and how. I would not have visited any club if Shin had not approached me that night. And even if I had ended up visiting one, I would have had a totally different experience with another host. These contingencies play an important role in the shaping of our sensory experience and the determining of its significance. There was a time, however, when there was little room for ethnographers to reflexively take into account their sensory experience. The premise in the discipline of Anthropology and Social Sciences at large was the production of objective and impartial knowledge. The positivist approaches not only suppress anything subjective, including sensory experience, but also erase it to safeguard objective truth. But, how could I really have elicited anything meaningful about commodified intimacy from clinical-style interviews and observation at a distance? Doing and writing affective ethnography was thus central. The sensory experience I shared above, after all, allowed me to understand commodified romance and theorise the commercialisation of hopes, dreams and future-oriented aspiration in Japan’s service-centred economy. Some might see the importance of affective ethnography but wonder how it is done. The answer is that there is no one right way. There are, however, some principles that affective ethnographers can draw from. Being open-minded and non-judgmental is first and foremost vital. Allow yourself to believe in your own sensory experiences as part of the fieldwork experience. You also need to be aware that you do not always know where you are headed before your actual interactions in the field, even if you have developed solid research questions. There are multiple possibilities that constantly appear and disappear while you decide which lead to follow. Affective ethnographers also need to be self-reflexively responsible for their projects. We recognise that ethnographers are human beings who are situated in a society and whose access to information is never entirely objective (Pink 2009). Instead of conveniently removing the researcher-self from what is written, affective ethnographers should be honest and self-reflexively critical about their positionality in the field. In other words, the affective ethnographer’s job 229

Akiko Takeyama is not just to study others but also to question their own positionality and potential bias (see Ho, Ch. 8.2). They should also be aware of the privilege of producing ethnographic knowledge. To sum up, traditional ethnographers rarely discuss in print how they might have been, for example, attracted to some subjects of study or seduced into building a rapport with particular individuals (Kulick/Wilson 1995). However, it is time to rethink the research premise that affects are separable from human experience in the doing and writing of ethnography. Feminist scholars have long argued that the body, emotion and care have been largely associated with feminine traits and systematically excluded from scientific research (Haraway 1988; Jagger 1989). In this sense, affective ethnography is a feminist approach to challenge what has been traditionally dismissed in the masculinist enterprise of Social Sciences. It is not so much about seeking the truth. It is rather an alternative mode of knowing which affective ethnographers use to prise doors open and to shed light on things behind, beneath and beside observable realities.

230

8.3 Reflections on fieldwork in post-bubble Japan: Gender, work and urban space Swee-Lin Ho

Framing the fields I first lived in Japan in 2002 when my then employer, a multinational corporation, sent me to supervise the restructuring of an ailing subsidiary in Tokyo for a year. I was out most evenings drinking with colleagues to understand the local work culture, and to get to know one another better in informal settings when conversations were candid and spontaneous. The drinking often extended beyond the first round to a second, third or fourth, even on weeknights. I had initially thought the nocturnal merriment would be short-lived, since my stay coincided with the FIFA World Cup co-hosted by Japan and South Korea, but I gradually realised that social drinking after hours was very much a part of the daily routine for working adults in Japan. Everywhere I went—from Shinjuku, Shibuya and Akasaka, to Roppongi, Ginza and Shinagawa—bars and restaurants were thronging with office workers, both male and female. As the evening progressed, the group’s size shrank, venues changed, conversation topics shifted away from work and some late-night entertainment activities became more playful. When a night out ended long after public transport services had ended, some of my co-workers would grab several hours of sleep in a love hotel, which I realised were plentiful across Japan, and were hardly sleazy establishments for illicit sexual activities. Most certainly, I was quite flummoxed by my encounters. ‘Are drinking and late-night entertainment no longer socially unacceptable for women in Japan?’ I wondered. Many establishments had menus listing alcoholic beverages exclusively for ‘ladies,’ while advertisements pasted in drinking venues, commuter trains and public billboards portrayed women drinking traditionally masculine drinks such as beer, sake and whisky. There were also small bars, strip clubs and host clubs serving only a female clientele. What has effected these commercial changes? Having met many female corporate managers, lawyers and bankers, I also wondered if better work opportunities had enabled women to negotiate themselves out of their gender roles. I thus began to frame my research interest to focus on the lives of women, work and the urban night-time economy in post-bubble Japan. When I returned to Tokyo in 2003 as a graduate student, large numbers of housewives and female office workers of all ages swarming airports, stadiums and various venues to attend events and catch a glimpse of actors, male singers and popular music bands from South Korea confounded me. The Korean Wave reinforced my interest in understanding women’s lives and the public space, as well as the role played by commercial processes in enabling women to actively affirm themselves as subjects of their own desire, and not as mere objects of the male gaze and men’s pleasure.

231

Swee-Lin Ho

Selecting sites and sampling subjects I have so far interviewed more than 500 Japanese men and women for various related studies on processes of change that affect women’s work patterns, family formations, consumption practices, leisure activities and various social relations, in the hope of broadening the scope of existing studies by dispelling some myths about the uniqueness or exoticness of Japanese society that many have depicted. My field sites are mostly places and spaces pertaining to the activities of my subjects of study, which encompassed the private sphere of the home, and the public spheres of work and play. For my first project, I had gathered a considerable amount of data from one year of employment in a Japanese company when I started studying women managers (Ho 2018). The daily notes I took from observing and interacting with co-workers helped me understand the corporate environment in which women managers work: corporate policies on hiring, promotion and remuneration; assignment of duties; lines of reporting; allocation of authority and power of decision-making. Other field sites included public places where my subjects spent their time eating, drinking and engaging in playful activities. To understand the various institutional structures that affect how women work, I also interviewed male workers, and gathered information about companies—small and large—before I requested interviews with corporate executives. I also conducted interviews with government employees to understand the implementation and management of state policies on labour, family and health, as well as with labour union representatives and legal advisors, non-governmental organisations, and several support groups dedicated to helping workers seek redress for unfair dismissal and various work-related problems. Studying female fans of the Korean Wave took me to concerts, fan meetings, product endorsement events, filming sites, airports and birthday parties in Japan and South Korea (Ho 2011). To examine the production, distribution and marketing processes driving the popularity of Korean popular music, television dramas and movies—my second project—I spent several years in South Korea interviewing actors, actresses, singers, instructors, directors, scriptwriters, promoters and distributors. I also made observations at training academies, auditions, filming locations, recording studios and broadcasting stations. To understand how Japanese women’s fan activities might affect their daily life, I followed my informants to neighbourhood and community events, and observed some women in their homes. Using the methodology of snowball sampling offers a useful chain of referrals by people who share or know of others with similar interests and experiences, but it could potentially narrow the research scope, since each chain of referrals tends to offer specific perspectives on a given practice. When studying women’s extramarital experiences—a third project (Ho 2012)—for example, one chain of referrals led me to ten women who were once full-time housewives, and explained their extramarital activities as provoked by their husband’s infidelity. As I searched for informants with different marital and extramarital experiences, I found some who agreed to be interviewed after months of private messaging through Internet chatrooms and blogs hosting discussions on the extramarital experiences of women. Several others came from a speed dating agency that arranged ‘extramarital adventures’ for men and women, who had responded to my interview request, which the agency had circulated. I met others through the owner of a cable television company that broadcast many channels of adult programmes, who also disseminated my request to its subscribers and helped me obtain several responses. 232

Chapter 8 How to observe people and their environment Similarly, for another research project on love hotels (Ho 2008), I spoke to people using these establishments as my main subjects, and others who could help me understand commercial and legal change. These comprised cleaners, receptionists and managers of love hotels; employees of Japan’s Ministry of Land, Infrastructure, Transport and Tourism; and project consultants at advertising companies which had contributed to marketing assignments to change the image of love hotels.

Ethics and positioning in the field Selecting sites and subjects for diversity is not easy, and that is only the start. Managing fieldwork over a long duration of time is far more challenging. Over the years, I have come to appreciate the importance of positioning in the field with an ethical code of practice by which my subjects would define me as an ethnographer. I had once thought about ethics in the field in terms of do’s and don’ts. Nearly two decades of fieldwork in Japan taught me that it is about the constant management and negotiation of the perceptions and positions of both the ethnographer and research subjects. I would like to think of this as the process of subject–object reorientation, which involves the delicate balancing of proximity and distance, to allow for more egalitarian, reciprocal and dynamic interactions between ethnographers and subjects. I might have the privilege of knowing a few women in each of the two friendship networks of women managers (Ho 2018), but having initial access by positioning myself as ‘one of them’— I was once a corporate executive myself—only gave me opportunities to meet with the women. Careful about protecting their respective circles and meticulous in selecting individuals to include, the women gradually shared more about their lives after several years of careful negotiation of my identity and character, as I made continual efforts to understand their needs, wants and concerns. Access is both a privilege and a responsibility. It is a form of trust given to an ethnographer to respect the behaviour and thoughts of research subjects (see McLaughlin, Ch. 6; Reiher/Wagner, Ch. 16). Far from being an entitlement, access is an obligation to protect the confidentiality of the information gathered, and a responsibility to not adversely alter the lives and activities of informants. Researchers and informants often asked me about my studies of women’s late-night drinking, voyeuristic play in host clubs and various entertainment venues, sexual encounters, marital issues, intimate feelings and workplace difficulties, all of which are sensitive, morally contentious issues with serious implications on individuals’ lives. Breaching an informant’s trust could have serious repercussions for many parties. As the ethnographer’s subjectivity is an inherent part of our research (Madison 2005, p. 9), we risk discrediting our cultural and academic communities. It is also important to tread carefully between the polar ends of sympathy and apathy in interacting with informants. While documenting the changes to Japan’s urban night-time economy that promote drinking among women (Ho 2015), I often checked the interview questions I posed to informants working in drinking and entertainment venues, regardless of whether they worked as bartenders, waiters, waitresses, supervisors, managers, public relations corporate executives, hosts, hostesses, strippers or government officials responsible for urban zoning laws and licencing. For my research on love hotels too, I tried to understand the perspectives of cleaning workers, receptionists, janitors, and suppliers of drinks and products to vending 233

Swee-Lin Ho machines in love hotels. These individuals may represent the institutional structures whose practices are often portrayed negatively, but they are also workers with real lives, whose jobs could be equally susceptible to broader controlling processes that affect the lives of my main subjects. In conclusion: Going native is extremely difficult. The best that ethnographers can do is to minimise the tendency of othering our subjects so that they are not treated as mere objects of scrutiny. Negotiating shifts in position by becoming objects of inquiry for field subjects at times helps build trust and respect in our interactions with research subjects, who may at times be equally curious about and interested in our own experiences. Doing fieldwork in Japan—as cross-cultural studies elsewhere are doing—can be particularly trying due to the plethora of stereotypes many have formed about the people and their social practices. But one’s field experience will be meaningful, enriching and rewarding—as mine has been—if due care and respect are exercised in our engagement with and representation of our research subjects.

234

Further reading Bernard, H. Russell (2011): Research methods in Anthropology. Lanham: Rowman & Littlefield. Clifford, James/Marcus, George E. (1986): Writing culture: The poetics and politics of ethnography. Berkeley, CA: University of California Press. Hannerz, Ulf (ed.) (2010): Anthropology’s world: Life in a twenty-first-century discipline. London: Pluto Press. Madden, Raymond (2017): Being ethnographic: A guide to the theory and practice of ethnography. London: Sage. Madison, D. Soyini (2005): Critical ethnography: Method, ethics, and performance. Thousand Oaks, CA: Sage. Pink, Sarah (2009): Doing sensory ethnography. Thousand Oaks, CA: Sage. Satō, Tomohisa (2013): Fieldwork 2.0: Gendai sekai wo firudowāku. Tōkyō: Fukyōsha. Smartt Gullian, Jessica (2016): Writing ethnography. Rotterdam: Sense Publishers.

References Amendt-Lyon, Nancy (2000): Authentizität, selektive. In: Stumm, Gerhard/Pritz, Alfred (eds.): Wörterbuch der Psychotherapie. Wien: Springer, pp. 55–56. American Anthropological Association (2012): Principles of professional responsibility (Code of Ethics). https://www.americananthro.org/LearnAndTeach/Content.aspx?ItemNumber=22869&navItemNumber =652, [Accessed 24 June 2019]. Bergmann, Jörg R. (2006): Qualitative Methoden der Medienforschung: Einleitung und Rahmung. In: Ayaß, Ruth/Bergman, Jörg R. (eds.): Qualitative Methoden der Medienforschung. Reinbek bei Hamburg: Rowohlt, pp. 13–41. Bernard, H. Russell (2011): Research methods in Anthropology. Lanham: Rowman & Littlefield. Bourdieu, Pierre (2005): Verstehen. In: Bourdieu, Pierre (ed.) (2005): Das Elend der Welt: Studienausgabe. Zeugnisse und Diagnosen alltäglichen Leidens in der Gesellschaft. Konstanz: UVK, pp. 393–426. Clifford, James (1997): Routes: Travel and translation in the late twentieth century. Cambridge, MA: Harvard University Press. Clifford, James/Marcus, George E. (1986): Writing culture: The poetics and politics of ethnography. Berkeley, CA: University of California Press. Dales, Laura/Yamamoto, Beverly Anne (2018): Relating to ‘unconventional’ women. In: Alexy, Allison/ Cook, Emma (eds.): Intimate Japan: Ethnographies of closeness and conflict. Honolulu, HI: University of Hawai‘i Press, pp. 241–242. Ehn, Billy/Löfgren, Orvar (2012): Nichtstun: Eine Kulturanalyse des Ereignislosen und Flüchtigen. Hamburg: Hamburger Ed. Eriksen, Thomas Hylland (2009): What is Anthropology? London: Pluto Press. Freeman, Carla (2014): Entrepreneurial selves: Neoliberal respectability and the making of a Caribbean middle class. Durham, NC: Duke University Press. Geertz, Clifford (1987): Dichte Beschreibung: Beiträge zum Verstehen kultureller Systeme. Frankfurt a.M.: Suhrkamp. Gusterson, Hugh (1997): Studying up revisited. In: Polar 20, No. 1, pp. 114–119. Hamer, Bent (dir.) (2003): Kitchen stories [film]. 96 min., IFCFilms. Hannerz, Ulf (1980): Exploring the city: Inquiries toward an urban Anthropology. New York, NY: Columbia University Press. Hannerz, Ulf (1998): Other transnationals: Perspectives gained from studying sideways. In: Paideuma: Mitteilungen zur Kulturkunde 44, pp. 109–123. Hannerz, Ulf (2010): Field worries: Studying down, up, sideways, through, backward, forward, early or later, away and at home. In: Hannerz, Ulf (ed.): Anthropology’s world: Life in a twenty-first-century discipline. London: Pluto Press, pp. 59–86. Haraway, Donna J. (1988): Situated knowledges: The science question in feminism and the privilege of partial perspective. In: Feminist Studies 14, No. 3, pp. 575–599. Hendry, Joy (2003): From scrambled messages to an impromptu dip: Serendipity in finding a field location. In: Bestor, Theodore C./Steinhoff, Patricia G./Lyon-Bestor, Victoria (eds.): Doing fieldwork in Japan. Honolulu, HI: University of Hawai‘i Press, pp. 55–70.

235

References Ho, Swee-Lin (2008): Private love and public space: Love hotels and the transformation of intimacy in contemporary Japan. In: Asian Studies Review 32, No. 1, pp. 31–56. Ho, Swee-Lin (2011): Old texts, new desires: How Korean television drama Daejanggeum evokes reflexivity, renewal and resistance among Japanese women. In: The Review of Korean Studies 14, No. 2, pp. 91–113. Ho, Swee-Lin (2012): ‘Playing like men’: The extramarital experiences of women in contemporary Japan. In: Ethnos 77, No. 3, pp. 321–343. Ho, Swee-Lin (2015): ‘License to drink:’ White-collar female workers and Japan's urban night space. In: Ethnography 16, No. 1, pp. 25–50. Ho, Swee-Lin (2018): Friendship and work culture of women managers in Japan: Tokyo after ten. New York, NY: Routledge. Illius, Bruno (2012): Feldforschung. In: Beer, Bettina/Fischer, Hans (eds.): Ethnologie: Einführung und Überblick. Berlin: Reimer, pp. 75–100. Inda, Jonathan Xavier/Rosaldo, Renato (eds.) (2008): The Anthropology of globalization: A reader. Malden, MA: Blackwell. Jagger, Alison M. (1989): Love and knowledge: Emotion in feminist epistemology. In: Inquiry 32, No. 2, pp. 151–176. Kato, Fumitoshi (2009): Kyanpuron. Atarashii fīrudowāku. Tōkyō: Keio University Press. Klien, Susanne (2016a): Reinventing Ishinomaki, reinventing Japan? Evolving creative networks, alternative lifestyles and the search for quality in life in post-growth Japan. In: Japanese Studies 36, No. 1, pp. 39–60. Klien, Susanne (2016b): Shinto ritual practice in Miyagi prefecture after the Great East Japan Earthquake: The case of the Ogatsu Hoin Kagura. In: Asian Ethnology 75, No. 2, pp. 359–376. Klien, Susanne (2020): Accommodation and resistance in Hokkaido hiphop practitioners: An ethnographic analysis of generation resignation in post-growth Japan. In: Ethnography. DOI: https://doi.org/10.117 7/1466138120907339. Kottmann, Nora (2016): Heirat in Japan: Romantische und solidarische Beziehungswelten im Wandel. Wiesbaden: Springer VS. Kulick, Don/Wilson, Margaret (eds.) (1995): Taboo: Sex, identity and erotic subjectivity in anthropological fieldwork. London: Routledge. Madden, Raymond (2017): Being ethnographic: A guide to the theory and practice of ethnography. London: Sage. Madison, D. Soyini (2005): Critical ethnography: Method, ethics, and performance. Thousand Oaks, CA: Sage. Malinowski, Bronisław (1922): Argonauts of the Western Pacific. London: Routledge & Kegan. Marcus, George E. (1995): Ethnography in/of the world system: The emergence of multi-sited ethnography. In: Annual Review of Anthropology 24, pp. 95–117. Marcus, George E. (2006): Reflexivity unbound: Shifting styles of critical self-awareness from the Malinowskian scene of fieldwork and writing to the emergence of multi-sited ethnography. In: Rao, Ursula/ Hutnyk, John (eds.): Celebrating transgression: Method & politics in anthropological studies of culture. New York, NY: Berghahn, pp. 13–22. Nader, Laura (1972): Up the anthropologist: Perspectives gained from studying up. In: Hymes, Dell (ed.): Reinventing Anthropology. New York, NY: Random House, pp. 284–311. Osawa, Stephanie (2018): Devianz aus der Sicht von ‘Tätern’: Normabweichendes Handeln in den Selbstdeutungen devianter Jugendlicher in Japan. Wiesbaden: Springer VS. Pink, Sarah (2009): Doing sensory ethnography. Thousand Oaks, CA: Sage. Rabinow, Paul (1977): Reflections on fieldwork in Morocco. Berkeley, CA: University of California Press. Rao, Ursula (2006): News from the field: The experience of transgression and the transformation of knowledge during research in an expert-site. In: Rao, Ursula/Hutnyk, John (eds.): Celebrating transgression: Method & politics in anthropological studies of culture. New York, NY: Berghahn, pp. 23–37. Reader, Ian (2003): Chance, fate and undisciplined meanderings: A pilgrimage through the fieldwork maze. In: Bestor, Theodore C./Steinhoff, Patricia G./Lyon-Bestor, Victoria (eds.): Doing fieldwork in Japan. Honolulu, HI: University of Hawai‘i Press, pp. 89–105. Robben, Antonius/Sluka, Jeffry A. (2012): Fieldwork in Cultural Anthropology: An introduction. In: Robben, Antonius/Sluka, Jeffry A. (eds.): Ethnographic fieldwork: An anthropological reader. Hoboken, NJ: Wiley-Blackwell, pp. 1–47. Roth, Joshua Hotaka (2003): Responsibility and the limits of identification: Fieldwork among Japanese and Japanese Brazilian workers in Japan. In: Bestor, Theodore C./Steinhoff, Patricia G./Lyon-Bestor, Victoria (eds.): Doing fieldwork in Japan. Honolulu: University of Hawai‘i Press, pp. 335–351.

236

Chapter 8 How to observe people and their environment Rubiès, Joan Pau (2017): Ethnography and cultural translation in the early modern missions. In: Studies in Church History 53, pp. 272–310. Schlehe, Judith (2008): Formen qualitativer ethnographischer Interviews. In: Beer, Bettina (ed.): Methoden ethnologischer Feldforschung. Berlin: Reimer, pp. 119–142. Schliemann, Heinrich (1995): Reise durch China und Japan im Jahre 1865. Leipzig: Merve. Tagsold, Christian (2016): Japanese gardens unleashed. In: Die Gartenkunst 28, No. 1, pp. 293–300. Tagsold, Christian (2017): Spaces in translation: Japanese gardens and the West. Philadelphia, PA: University of Pennsylvania Press. Takeyama, Akiko (2016): Staged seduction: Selling dreams in a Tokyo host club. Palo Alto, CA: Stanford University Press. Ullmann, Katrin (2017): Generationscapes: Empirie und Theorie einer globalen Generation. Bielefeld: Transcript. Wacquant, Loïc (2006): Body & soul: Notebooks of an apprentice boxer. New York, NY: Oxford University Press. Wacquant, Loïc (2010): Leben für den Ring. Boxen im amerikanischen Ghetto. Konstanz: UvK. Welz, Gisela (1998): Moving targets: Feldforschung unter Mobilitätsdruck. In: Zeitschrift für Volkskunde 94, pp. 177–194.

237

Chapter 9 How to access written and visual sources: Archives, libraries and databases Theresia Berenike Peucker, Katja Schmidtpott and Cosima Wagner

1. Introduction As a researcher, the hunt for primary and secondary sources will soon lead you to libraries and archives, both at your local institution as well as in Japan. In this chapter, the authors have joined forces to provide you with advice on gathering both physical and digital Japan-related sources in the digital age. This includes hints on developing a general and Japan-specific form of digital literacy when using libraries and archives. In addition, a selection of important reference tools for primary and secondary sources will help you to find appropriate sources for your research project. While all parts of this chapter will be of interest for first-time researchers studying Japan, the sections on digital literacy and archives will also be of relevance to more experienced researchers. The chapter is divided into three sections and begins with the first step in the search process: how to approach and operate research infrastructures, especially search engine-based catalogues. This is followed by a section on where to look for Japanrelated library reference tools and sources. The final section of the chapter is dedicated to Japanese public archives, many of which hold sources that are yet to be discovered, interpreted and published.

2. General hints on how to approach and operate library catalogues The digital transformation of everyday life has profoundly changed the infrastructure framework within which research is conducted, especially in libraries and archives. Easy access to online catalogues and other sources may lead us to believe that we can find any information we need at the mere click of a button. However, despite the sheer number of resource discovery tools available to researchers, digital access to overseas Area Studies collections can still prove difficult (Asato 2013, p. xx; Pitman 2015, pp. 69–80). This is partly because the holdings of some collections are not visible in online catalogue records (‘hidden collections’), partly because of the way overseas publications in non-Latin scripts are handled in Latin-script-centric catalogues or discovery tools and partly because libraries license rather than own electronic resources, making them accessible to members of the libraries’ institution only (Pitman

238

Chapter 9 How to access written and visual sources 2015, pp. 69–70). The latter also applies to access to resources in Japan, which on top of that are often heavily protected by copyright laws or not available in a digital format yet. Therefore, information literacy and digital literacy are the key to finding and accessing resources for a research project. Both types of literacy involve finding, understanding, critically evaluating and using information. While information literacy refers to reflectively discovering information and understanding how it is produced and valued (ACRL 2016, p. 3), digital literacy involves a good command of digital tools and information technologies through which information is discovered, accessed, analysed, (re)created and communicated (ALA 2013; Becker 2018). Here it is important for students and faculty alike to be aware of the marketplace concentration in digital services and the increasing reliance of libraries and other knowledge infrastructure providers on ‘closed system’ products, e.g. search algorithms for library catalogues and their transformation into discovery systems, which implies an increasing lack of transparency of how catalogue data is processed (Reimer 2020, p. 5). Therefore, developing digital literacy enables a deeper and critical understanding of these tools/information technologies as biased filters through which a search result is produced.1 How to deal with these conditions is the subject of the following sections, where we will look in detail at how bibliographic information is presented in catalogues in the digital age and at the diverse technologies available when searching for Japan-related sources (including how to conduct search queries in Japanese script).

2.1 Where to start your search? Always begin your search for primary or secondary sources at your local academic library. Almost every institution will provide access to an online catalogue. Many library catalogues are interconnected and provide access to their combined bibliographical data through meta catalogues like the Karlsruher Virtueller Katalog (KIT n.d.).2 These can also search for bibliographic data worldwide, including from WorldCat (OCLC n.d.),3 and the Japan-centred ‘Scholarly and Academic Information Navigator’ CiNii catalogues (NII n.d.).4 If you are at the start of your research in a certain library, we strongly recommend you to contact the Japanese Studies librarian in charge of the collection that you are interested in. She or he might give you source-based bibliographic advice that could change your research design (Lyon-Bestor 2003, p. 370), especially regarding the barriers to accessing hidden collections. Also, check whether your institution has a subject guide for Japanese Studies.5 This is a collection of thematically 1 Investigating the bias of knowledge infrastructure discovery systems has only just begun. However, a thoughtprovoking paper by Sharon Block (2020) on how ‘seemingly racist or sexist topical labeling [...] impedes knowledge discovery’ in the well-known academic data base JSTOR highlights the importance of critical algorithm studies and encourages academic communities and digital providers to jointly create systems that reflect scholarship better (Block 2020, p. 56). 2 The KVK is provided by the Karlsruher Institut für Technologie (KIT) in Germany at https://kvk.bibliothek.edu. 3 The WorldCat is the world’s largest library catalogue and is provided by the OCLC Online Computer Library Center, Inc. in the U.S. at https://www.worldcat.org. 4 CiNii is a database aggregating bibliographical data of Japanese academic libraries provided by the National Institute of Informatics (NII) in Japan. You can search for academic articles at https://ci.nii.ac.jp, books at https://ci. nii.ac.jp/books/en/ or dissertations at https://ci.nii.ac.jp/d/?l=en. 5 Examples of Japan-related subject guides are the ‘Subject Guides Portal’ from the North American Coordinating Council on Japanese Libraries Resources (NCC) at https://guides.nccjapan.org/subjectguidesportal or the ‘Japan Studies Subject Guide’ of Leiden University Asian Library at https://www.library.universiteitleiden.nl/subject-guid es/japanese-studies.

239

Theresia Berenike Peucker, Katja Schmidtpott and Cosima Wagner sorted resources (databases, catalogues, websites, etc.) carefully curated by the subject librarian.

2.2 How to search the library catalogue Today, most library catalogues look like search engines and operate under a ‘discovery system’. Discovery system catalogues are intuitive, can process simple queries and the results are ranked by relevance algorithms (Böhner 2013, p. 49). The basic search mode tends to resemble a Google search box and is your first port of call when you are broadly exploring a topic. Before you start browsing, write down a list of key words that relate to your planned research. What is your overall research field, what are the subfields and what are the aspects of your topic (c.f. ULBM 2019)? These words should also include synonyms of your core concepts and superordinate and subordinate terms (ibid.). If you are planning to read literature in different languages, include the corresponding translations too. To ensure that you are using the correct terminology in a foreign language, you can consult Wikipedia, an online thesaurus,6 digital encyclopaedias7 or specialised dictionaries. Note that glossaries at the end of articles or books might help you enlarge your browsing terms. In doing so, remember that terms related to academic concepts can change depending on their historic, social and geographical contexts. When browsing catalogues by topic, note that libraries use classification systems for labelling the subject of a book. However, classification schemes outside Japan have often been criticised for their Anglo-American and European bias, paying less attention to coverage of other geographies (Pitman 2015, p. 74). In Japan, Japanese bibliographic data tends to be classified by the Nippon Decimal Classification (NDC) (Nihon jisshin bunruihō), and most libraries will use the NDC to systematically arrange their books on the shelves. Ask for a tailored handout of the classification scheme at the respective counter. This will help you to orientate yourself quickly when visiting a library. If you already have information about the specific primary or secondary source you are looking for, you should switch to the advanced search mode. To use it efficiently, you have to be familiar with the concept of bibliographical data, which includes categories like author/creator, title, year/place of publication, publisher, the series a publication might be part of and the edition. You can search for the majority of these categories (e.g. family name of an author) and combine them with other categories (e.g. title).

2.3 How to search for Japanese language content in catalogues Libraries give instructions on their website about bibliographical searching and topical browsing. However, outside Japan you might face difficulties when searching for and retrieving bibliographical data in Japanese, Chinese and other non-Latin scripts as well as in transcriptions. 6 Examples of multilingual online thesauri for the academic community are Sowiport hosted by the German Social Science Infrastructure Services/Leibniz Institute for the Social Sciences (GESIS) at http://sowiport.gesis.org/thesaur us and Skosmos developed by the National Library of Finland at http://skosmos.dev.finto.fi/en/. The so-called Research Navi (Risāchi nabi) by the National Diet Library provides an online tool for creating a Japanese word field at https://rnavi.ndl.go.jp/ln-search. 7 For a short overview of digital Japan-related encyclopedias, Weber/Krickel 2018, pp. D.73–14–D.73–15.

240

Chapter 9 How to access written and visual sources One reason for this is that the search algorithms in library catalogue software outside Japan are often not well adapted for non-Latin scripts. Furthermore, library catalogue search results increasingly come from a wide range of sources, including library-owned resources and licensed databases. With the net cast so wide, users can access more resources, but ‘might well not realise that they are not seeing all the potential hits that they might, because of the wide range of different standards used in these systems’ (Pitman 2015, p. 76). Hence, searching for foreign language material with varying transcription standards requires a skilled researcher, who can find all the resources a library holds (Pitman 2015, p. 77). It is advisable, in addition to searches in Japanese script, to always run searches in rōmaji (Hepburn/modified Hepburn system and kunrei romanisation) as well. In library catalogues based on an algorithmic discovery (search engine) technology, the results of your search will automatically be ranked by ‘relevance’. ‘Relevance’ is generally determined by the commercial producer of the software, often a third party, without making the rationale behind the programming of the ranking algorithms transparent to library staff and users. The ranking of hits will draw your attention to the data prominently displayed at the top of the list. This pre-selection of visible bibliographic data takes place unless you actively turn to other sorting categories like ‘newest/oldest first’. In addition, note that bibliographical data about book chapters and titles of journal articles might not be indexed in a library catalogue. In these cases, you have to check the availability of the superordinated entity like an edited volume or a journal. There are specialised Japan-centred bibliographic databases that focus on incorporating these data like the aforementioned CiNii books or CiNii articles. The Bibliography of Asian Studies (BAS)8 contains journal articles and book chapters in English. JSTOR9 and Project Muse10 cover journal articles in English. However, be aware that full access to these databases requires a subscription from your institution. Use the following criteria to further evaluate your search results: 1. 2. 3.

Does the journal match the academic discipline(s), e.g. Japanese Studies or Social Science you are referring to in your research? Is the quality of the articles ensured by a peer review process? Some library catalogues identify these journals. In regard to other publication formats, you might also take into account these points: • Is the author known in her field of research? • Do you know which institution she or he belongs to? • Is the publisher specialised in the topic of the book? • Look for reviews of the book and check the journals these reviews have been published in. • Check if the criteria for good academic practice (Balzert et al. 2008, pp. 9–47; see also Reiher/Wagner, Ch. 16) have been fulfilled. This also means taking a look at the accuracy of the citations and at the references.

8 BAS is hosted by EBSCO at https://www.asianstudies.org/publications/bibliography-of-asian-studies. 9 JSTOR is part of the not-for-profit organisation ITHAKA at https://www.jstor.org. 10 Project MUSE is produced by Johns Hopkins University Press at https://muse.jhu.edu.

241

Theresia Berenike Peucker, Katja Schmidtpott and Cosima Wagner

2.4 Japan-related library reference tools If we were asked to recommend just one print and one digital meta reference tool for starting your Japan-related search for sources from outside Japan, we would refer you to the Handbook for Asian Studies specialists: A guide to research materials and collection building tools, edited by Noriko Asato (2013) as the print meta reference tool. This book not only lists the most common tools for addressing search queries, but also provides a short abstract about each reference tool. For a digital ‘community hub for the field of Japanese Studies’ (NCC 2020), we would refer you to the Guide to library and information resources (NCC 2017), a website published by the North American Coordinating Council on Japanese Library Resources.11 Both of these resources will lead you to appropriate starting points for accessing print or digital sources for your research project and give you ideas about where to look further afield.12 But note when discovering special collections or sources not available through your university library, you do not (always) have to travel to Japan in order to get hold of these items. Many libraries take part in national and international inter-library loan systems (see Maclachlan, Ch. 4.1). So even if going abroad is out of reach, it is possible to get hold of the literature you need. Furthermore, many libraries also welcome acquisition requests or provide ‘patron driven acquisition’ (PDA) schemes in their catalogues. If you study Japanese Studies at a German university, you are eligible to use the CrossAsia portal (crossasia.org) managed by the East Asia department at Staatsbibliothek zu Berlin (SBPK). The CrossAsia search provides access to original language bibliographic records in library catalogues from Asia and around the world and full-texts from licensed databases through its Integrated Text Repository (see Heckel, Ch. 11.3).13 In order to get access to all licensed content in CrossAsia you have to fill in an application form, get approval from your institute (stamp) and mail it to the CrossAsia librarian team (SBPK 2020). However, there are some resources that can only be accessed in Japan. In the following two sections, we will introduce you to the general systemic structure of libraries (section 3) as well as archives (section 4) in Japan and to ways to approach and make use of them for your research project.

3. Libraries in Japan When one is in Japan, it is important to think about the structures and purposes of libraries, before searching their catalogues and visiting different libraries. According to the National Diet Library Law (Kokuritsu Kokkai Toshokanhō), the National Diet Library (NDL, Kokuritsu Kokkai Toshokan) is in charge of collecting and providing all Japanese publications and administering the national bibliography (Ōwa 2018, p. 11). Here the chances are high that any Japanese books and journals you are searching for will be found. Always prepare for your stay in Japan by searching the online bibliographic records of the NDL (https://ndlonline.ndl.go.j 11 Check the comprehensive institutional index of international collections on Japan provided by the North American Coordinating Council on Japanese Library Resources at https://guides.nccjapan.org/researchaccess. 12 For an overview of access to digital sources for the field of Japanese history, Weber/Krickel (2018). 13 For more information, see CrossAsia in a nutshell at https://blog.crossasia.org/about/?lang=en.

242

Chapter 9 How to access written and visual sources p). As described in detail in the essay by Shinichi Aizawa and Daisuke Watanabe (see this chapter, Ch. 9.3), the NDL is also a hub for all government-related material (e.g. White Papers) and statistics. Its ‘Research Navi’ guide14 is a comprehensive starting point for an overview of the wealth of information on Japan available through the NDL (NDL 2009). NDL’s WARP (Web Archiving Project) database is an important resource when conducting research on institutions and their representations in the digital space in Japan. Here, the ‘websites of the national, prefectural, and municipal governments, including those of prefectures, designated cities, cities, and towns as well as committees for municipal mergers, independent administrative corporations, semi-governmental corporations or agencies, universities, events, online periodicals, and similar sites’ (NDL 2013) are archived regularly.15 Besides the NDL, the bibliographic databases of prefectural libraries (kenritsu toshokan) may also be worth consulting. They are part of the public library (kōkyō toshokan or kōritsu toshokan) sector (NTK 2020) and are charged with collecting resources that are closely related to the history of their prefecture. Catalogues of public libraries on a city, town and village level can likewise provide an insight into the collection building of the regional body they are operated by. In these libraries, expert staff can help you access sources concerning the region, its history and socio-political structures. Public libraries are social spaces and are actively promoted as a hub for actors from local civil society and for citizen scientists, so they might also be a good starting point for field research. If you are doing research on corporate bodies (e.g. firms, museums, NPO/NGO, labour unions) and their history, keep in mind that they too might have libraries or archives of varying size, possibly without providing accessible bibliographic data online (see this chapter, Schmidtpott/Schölz, Ch. 9.1). In contrast to public libraries, they are mainly for internal use and may hold rare primary and secondary sources. In fact, the collection of resources provided by a certain institution itself might be an interesting object of research (e.g. which types of literature are provided for inmates of a certain prison?). Do not forget to search the catalogues of those university libraries that are home to the experts you refer to in your research (see Zachmann, Ch. 4). Usually, their interests will be mirrored in the catalogue or special data archives (see this chapter, Aizawa/Watanabe, Ch. 9.3) of their home institution. Unfortunately, physical access to Japanese university libraries is also very likely to be restricted. Many institutions will require a letter of introduction.16 At the same time, open access publications are gaining momentum in Japan, as they are elsewhere, and you might look for sources in open access repositories of all universities via the directory of the Japan Consortium for Open Access Repository (JPCOAR).17 If you are uncertain which university library catalogue to head to, turn to the already mentioned meta catalogue and bibliographic database CiNii18 of the NII.

14 Available in English (https://rnavi.ndl.go.jp/rnavi/english.php) or Japanese (https://rnavi.ndl.go.jp/rnavi). 15 WARP website: see http://warp.da.ndl.go.jp. 16 Please refer to the draft for a ‘letter of introduction’ as well as a ‘library materials request form’ incl. an English translation provided by the NCC (2017). 17 For a list of member institutions, see https://jpcoar.repo.nii.ac.jp/?page_id=40. 18 Note that a new ‘CiNii Research’ next generation discovery platform is under development at the NII Research Center for Open Science and Data Platform (RCOS), which will not only provide access to academic publications but also to research data (RCOS 2017).

243

Theresia Berenike Peucker, Katja Schmidtpott and Cosima Wagner

4. Archives in Japan Although libraries provide you with access to vast amounts of primary and secondary sources, there are still numerous ‘raw’ primary materials waiting to be uncovered in archives in Japan. Drawing on the example of transnational history, Sheldon Garon (see this chapter, Ch. 9.2) has described the joy of ‘serendipitous discoveries in archives’ in his essay and the importance of archival research not only for the purpose of deepening but also widening one’s research scope. In addition, Katja Schmidtpott and Tino Schölz (see this chapter, Ch. 9.1) recommend not only considering written artefacts that are preserved in private (and often secluded) archives for your research, but also openly accessible objects such as public memorials. Although you might be intimidated by the necessary language proficiency and research experience required for working in archives and with written/objective archival sources, we would strongly encourage you to consider archival material for your research projects on Japan. The following sections will give you a short overview of working with and in archives in Japan with special consideration of regional public archives and archives in museums, libraries and universities. Public archives in Japan can be subdivided into two groups: a small group of archives run by the central government19 and a larger group of regional public archives, which are run by regional bodies, namely the prefectures and municipalities. National archives are, in general, easily accessible and parts of their holdings have even been digitalised and are accessible online free of charge.20 Japanese archival researchers and historians generally complain that in Japan the public interest in the preservation of public records is comparatively weak. They base these complaints, for instance, on the fact that the first public archives in Japan, the Yamaguchi Prefectural Archives, were founded only in 1959 and that the National Archives of Japan (Kokuritsu Kōbunshokan), which were established in Tokyo in 1971, were the result of years of lobbying by Japanese historians and followed a recommendation by UNESCO (Takayama 2008, p. 51). The legal framework is also frequently found to be lacking. An extremely short Public Archives Law (kōbunshokanhō)21 was passed in 1987, but it comprises little more than the duty of the central government and the regional bodies to enact appropriate measures for preserving administrative documents and other materials of historical importance and making them available for use (Kokuritsu Kōbunshokan 2001, p. 96). In this respect, Japan was trailing in last place behind all other OECD countries (Matsuoka 2011, p. 16). Precise rules for the proper management of public records were finally laid out in the Public Records and Archives Management Act (kōbunsho kanrihō) of 2009. However, this legislation only applies to the

19 These include the National Archives of Japan (Kokuritsu Kōbunshokan), the Diplomatic Archives of the Ministry of Foreign Affairs of Japan (Gaikō Shiryōkan), the National Institute for Defense Studies of the Ministry of Defense (Bōei Kenkyūsho Shiryō Etsuranshitsu) and the Archives and Mausolea Department of the Imperial Household Agency (Kunaichō Shoryōbu). 20 The databank of the Japan Center for Asian Historical Records (JACAR), which encompasses more than 30 million document pages from the National Archives of Japan, the Diplomatic Archives and the National Institute for Defense Studies, is likely to be already known to those familiar with the field. What is, however, probably less well-known is that JACAR is progressively starting to include holdings from other archives. Information about this can be found in the JACAR newsletter (https://www.jacar.go.jp/newsletter). 21 See http://elaws.e-gov.go.jp/search/elawsSearch/elaws_search/lsg0500/detail?lawId=362AC1000000115.

244

Chapter 9 How to access written and visual sources National Archives of Japan, whereas the regional public archives are merely required to pay heed to its core intention. While the National Archives of Japan are well organised and easily accessible, quite often one may encounter gaps in their holdings. As a matter of fact, the extent of the holdings is rather modest when compared internationally: in 2008 they amounted to 48 km as opposed to the 300 km held by the German national archives, the 175 km held in Great Britain and the 930 km in the U.S. Substantial losses were caused by fires, by the Great Kantō Earthquake of 1923, by bombings during World War II as well as the deliberate burning of files in the face of the impending defeat in 1945. In addition, however, there has been a tendency among public and private organisations to primarily keep filed materials that presented them favourably, whilst other materials were often destroyed (Takayama 2008, pp. 49–50). To those trying to fill the gaps that exist in the National Archives in Tokyo, Japanese colleagues recommend visiting the regional public archives. With a little luck, this is where one might find documents from the correspondence between the prefectural government and the central government bureaucracy, which may serve to reconstruct, to a large extent, some of what was lost.

4.1 Regional public archives (chihō kōbunshokan) Most of the 86 regional public archives currently in existence were created after the Public Archives Law came into force in 1988. So far, 39 of Japan‘s 47 prefectures and 47 of its 1,700 municipalities have public archives.22 Public archives are primarily intended for the preservation of public records but, in reality, they often also gather a wide range of material from private sources, such as private individuals or civil society organisations (Matsuoka 2011, p. 109). Consequently, the regional public archives are of interest for a wide variety of different research areas. The Amagasaki Municipal Archives (Amagasaki Shiritsu Chiiki Kenkyū Shiryōkan),23 for instance, house extensive materials about the history of the city’s industrialisation, which contains, among others, many private documents and objects from workingclass families. After being expertly advised, visitors will be permitted to access these sources and to photograph them without any bureaucratic difficulties. Although public archives are open to the public at large, it might be wise to visit the archives in the company of Japanese researchers who are already familiar with the archive in question. They will be able to arrange meetings with staff in advance, which might be very helpful in terms of getting relevant information about the specific archive’s holdings. In general, you are recommended to contact archives prior to visiting them in order to make an appointment. In some archives, it might happen that you are only allowed to see anonymised material due to staff being anxious not to violate the provisions of the Personal Data Protection Act (kojin jōhō hogohō), which was introduced in Japan in 2003. Indeed, the law does not require public archives to hire qualified specialist staff. Article 4 of the Public Archives Law does make it obligatory to employ specialist staff, but the additional article No. 2 allows them to dispense with this requirement initially. The reason for this loos22 See https://ja.wikipedia.org/wiki/公文書館#日本国内の公文書館一覧. Another list, which at 74 entries is less extensive but nonetheless just as up-to-date, having been compiled in 2018, can be found on the website of the National Archives of Japan at http://www.archives.go.jp/english/links/index.html. 23 See http://www.archives.city.amagasaki.hyogo.jp.

245

Theresia Berenike Peucker, Katja Schmidtpott and Cosima Wagner ening of the law may be that training for archivists has lagged behind that in Europe, the U.S. or China (Kokuritsu Kōbunshokan 2001, p. 22) and that, therefore, qualified specialists are comparatively rare. Since 2013, the Japan Society for Archival Science (Nihon Ākaibuzu Gakkai), whose aim is to pursue a more academic approach to training archivists, has been issuing accreditation for academically trained archivists. The professionalisation of archive staff can therefore be expected for the future.

4.2 Archives and collections of other public institutions: Libraries, museums, universities For those of you who are researching the history of a city or region where there are no public archives, the recommendation is not to give up too quickly but instead to check out other local institutions. One thing that is important to know about Japanese archives is that many of them are not called archives since they are part of libraries, museums or universities, and it is not always visible from the outside that they exist at all. As there is no nationwide directory of all Japanese archives,24 one will have to rely upon hints and tips from Japanese colleagues to hunt down these archives. A first port of call should be the approximately 3,000 public libraries (Takayama 2008, p. 46). Many of these were founded before 1945 and even back then already functioned as places for the preservation of historical materials (Aoyama 2003, p. 16), which means that, in some cases, their tradition as archives goes back much further than that of the public archives. Museums, too, can house archives or collections and, like libraries, they are easily accessible to everyone. The Nagasaki Museum of History and Culture (Nagasaki Rekishi Bunka Hakubutsukan), for example, holds around 81,000 objects and documents dealing with the history of the foreign trade that was conducted via Nagasaki port.25 Its documents can be viewed there without any problems and can be photographed upon payment of a small fee. Other initiatives like the Knowledgebase of Historical Resources in Institutes (khirin)26 of the National Museum of Ethnology in Osaka promote open science and aim at forming a crossinstitutional hub for Japanese history sources (written, visual and objective). Public archives, libraries and museums are often well connected to the local community. Thus, when chapter co-author Katja Schmidtpott was researching the history of Osaka’s armaments industry at the Peace Osaka museum (Pīsu Ōsaka), she received valuable pointers towards archive material held by private individuals and civil society organisations. In addition, you should not overlook local educational establishments. The Central Library of Kyushu University, for instance, has a huge collection of historical materials related to the history of coal mining in the region, but also many other sorts of manuscripts and records, some of which were collected by some of the university’s former history professors in other regions of Japan.

24 An incomplete list is constituted by the member list of the professional body Japan Society of Archives Institutions (Zenkoku Rekishi Shiryō Hozon Riyō Kikan Renraku Kyōgikai), which currently lists 139 institutions as members, amongst which can be found public archives, historical museums, libraries, university archives and company archives at http://www.jsai.jp/kikan/index.html (as of May 2018). An overview of the holdings and particularities of 30 selected public and private archives as well as data material (until 2007) regarding public archives and libraries can be found in Fujiwara (2008, pp. 223–283). 25 See http://www.nmhc.jp/collection.html. 26 See the catalogue portal at https://khirin-ld.rekihaku.ac.jp.

246

Chapter 9 How to access written and visual sources The collection is described online,27 as are other collections in other major universities, but unfortunately it is usually not possible to access them without being affiliated to the university in question, and sometimes even to the faculty of which the library is a part. Key issues Away from the central national archives in Tokyo, Japan offers a fertile landscape of public archives and collections where there is still much left for (foreign) historians to discover. Despite growing digitisation and the increasing availability of information online, it seems that in Japan there is still the need for a relatively strong degree of personal interaction in order to successfully conduct archival research. Good contacts with Japanese colleagues are important for gathering information about what institutions house what collections and, in many cases, also for gaining access to their holdings. Once an archive has been discovered and accessed, archivists or librarians often turn out to be extremely helpful and are happy to share their knowledge about where to look for further material. Last but not least, one must not rely on the assumption that all materials have already been gathered in archives or other institutions. This is one of the reasons why Japanese historians like to take their students on exploratory trips to historically interesting sites in order to search for, order and catalogue materials at family homes, businesses, temples and other local institutions.

5. Final comment In times of the ubiquity of information available digitally, it might be regarded as untimely to give advice on how to access Japan-related sources in a written ‘static’ publication like this handbook. However, with our focus on how to and not only where to and our encouragement for you to discover sources that are not yet visible digitally, we hope to have contributed to your understanding of Japanese Studies as well as to your digital literacy and to have enabled you to track down changed URLs as well as new reference tools by yourself. In the future, we aim to enhance this book by developing a subject guide website. Your cooperation is highly appreciated: just send us a note with further recommendations from your search discoveries. Meanwhile, good luck with hunting sources in and outside Japan!

27 See https://www.lib.kyushu-u.ac.jp/ja/collections?field_display_tid_i18n=All.

247

9.1 Clever approaches to tricky sources: How to extract information from business archives and war memorials Katja Schmidtpott and Tino Schölz

Just as fieldwork serves as a tool for gathering data in Anthropology or Sociology, archival research is the equivalent tool for historians. However, archival research differs from fieldwork because it does not involve observation or questioning. But although historians deal with ‘lifeless’ materials in archives, it would be wrong to assume that the cultural context and a researcher’s cultural competence, which matter greatly in fieldwork, can be neglected when it comes to successful archival research in Japan. The degree of cultural competence one needs, however, depends on the subject area as different areas involve varying levels of social interaction. In the field of Social and Economic History as well as in Urban or Local History for instance, historians often face special challenges ‘regarding access and discovery of archival material’ (Gordon 2003, p. 262), which frequently can only be mastered by means of ‘personal connections and introductions’ (ibid., p. 265). While this is generally true for all sorts of archives apart from those maintained by the central government, it especially applies to business archives. What is more, historians of modern Japan—mostly political or intellectual historians—currently mainly deal with written or, to a lesser extent, (audio)visual sources; however, historical sources are not only to be found in archives. Depending on the nature of the project, public space can also serve as an archive. It contains myriads of openly accessible sources ready to be analysed and interpreted as witnesses of the past at any given time. Towns and villages, streets, squares and their respective names, castles and palaces, office buildings, schools and universities, dwellings and mansions, monuments, shrines and temples, etc. shape our image of the present and the past. This essay introduces the reader to different types of historical sources and archives. In the first part, Katja Schmidtpott discusses business archives and shows that they tend to be carefully secluded from the outside world and often difficult to access. In contrast, in his account on public space as an archive, Tino Schölz shows that the material public space provides is very often in plain sight. He encourages historians to use objects and other non-textual materials as sources. Based on his research on war memorials, he shows that this can lead to a multitude of new insights, although it might be exhausting in comparison to usual archival research. The authors argue that working with these distinct types of archives requires quite different skills.

248

Chapter 9 How to access written and visual sources

Private archives: Business archives For many research projects in Economic and Social History, business archives can provide useful source material (Schmidtpott 2012). At first sight, Japanese businesses appear to have an extremely strong sense of tradition as almost every large and medium-sized company will have processed their history in the form of a company history (shashi). There are thousands of these shashi. Nonetheless, it would be misleading to conclude that there must be a similarly large number of company archives. As Andrew Gordon (2003, p. 264) pointed out, the ‘paradox of obsessive organizational history writing combined with poor archiving’ exists in Japan. As Schmidtpott has been told repeatedly by fellow historians, it is not unusual for a company to dispose of historical documents after a shashi has been published, mainly because its storage space is limited. She encountered this problem herself when visiting the library of a public housing corporation in 1999, where most of their historical material had been discarded to clear the shelves for newer publications. Digitisation may solve this problem, but it is doubtful whether companies in general consider this worthwhile. This is because one of the major functions of shashi is to serve as in-house reference material for employees. So once the shashi has been written, companies feel no need to keep the raw materials any longer. Therefore, until the turn of the millennium, only relatively few Japanese companies maintained an archive and usually did not employ professional archivists. Since the 2000s, the Shibusawa Eiichi Memorial Foundation (Shibusawa Eiichi Kinen Zaidan) and the Business Archives Association of Japan (Kigyō Shiryō Kyōgikai) have been working towards raising awareness among Japanese companies about the importance of professionally run business archives for corporate accountability. Since then there has been a small but noticeable rise in the number of business archives and in the levels of specialised, trained archivist staff employed there (Matsuzaki 2017). The first task for historians who want to work with business archives is to check which companies own an archive. This is not the easiest of undertakings as there is no nationwide directory of business archives. As the result of an initial stock take, the Shibusawa Eiichi Memorial Foundation published a list of business archives in 2008 (Shibusawa Eiichi Memorial Foundation). Likewise, the members’ list of the Business Archives Association can be used as a pointer towards which companies may have archives. Also, if there is an industry museum, the business it relates to will possibly have an archive. Special tourist guidebooks contain lists of such industry museums (Nichigai Asoshētsu Henshūbu 2003; Takeda 2008). Their advantage is that they are open to the public and that, ideally, one might strike up a conversation with the curator, who would be able to provide information about its holdings. This is what happened to Schmidtpott in the Seiko Museum in Tokyo, for example. There are, however, also companies that do possess all three: shashi, a museum and an archive, but do not allow anyone access to their records so as to keep full control over the interpretation of their history. Thus, Schmidtpott failed utterly in her attempts to gain access to one of the largest, internationally active Japanese industrial companies, which had already issued a number of company histories, had its own archive that was renowned as exemplary amongst experts, and also had a state-of-the-art museum that attracts large numbers of visitors. When asked, Japanese colleagues confirmed that the business archive rejected any attempts at contact. Even later on, when Schmidtpott was introduced personally to a staff member of that archive purely by chance, her subsequent enquiries remained unanswered. 249

Katja Schmidtpott and Tino Schölz To sum up, accessibility probably poses the largest problem with regard to business archives in Japan. Researchers need an introduction from a Japanese historian who is already in contact with the archive in question (Matsuzaki 2007, p. 8). It’s important to note that companies can freely create their own rules for their archives. During her research, Schmidtpott asked the employees of a big consumer products manufacturer’s archive how they decide which material to preserve. They answered frankly that they keep materials that testified to the successful development of new products, while they throw away materials on accidents or failures. Consequently, the range of possible research topics seems to be quite limited.

Public space as an archive: War memorials In contrast to the rather closed-off business archives, war memorials are located in public spaces. Various memorial sites have been erected and used in Japan from the 1860s to the present to mourn and pay tribute to the war dead. They include shrines such as the famous Yasukuni Shrine in Tokyo, the national defence shrines (gokoku jinja) in the provinces and cities, several thousand monuments in almost every community, military cemeteries, individual and collective graves in regular cemeteries, or museums that exhibit personal possessions of the victims. There were also thousands of trees planted and even small lakes created. For a project on the history of public commemoration of the war dead in modern Japan, Schölz (2016) analysed these memorial sites. In selecting his field, he considered regional diversity as well as differences between the victors and losers of the civil wars of the Bakumatsu (1853– 1867) and early Meiji periods (1868–1877). When focusing on monuments, it is important to actually visit and ‘document’ them. Schölz tried to systematically answer the following questions, which address both practical considerations and theoretical and methodological problems. First of all, the route to locations itself is sometimes very revealing: Is the site placed in the middle of a town or isolated in a forest or on a mountain that is difficult to reach? Is it visible from a distance? Is the path signposted? Do employees in the local tourist office or taxi drivers know the site? When arriving at the destination, the next task is to carefully ‘document’ it: How big is the memorial site? Where is it located within the public space? What does it look like? Who built it, when and why? Who paid for it? Is it located in a battlefield where people actually died? What is the visual message of the object? Are there inscriptions; if so, by whom? Who is remembered; who is not? Are war victims mentioned by name or not? What additional information about the lives and deaths of the fallen is given? Is there a hierarchy of the dead expressed? Which symbols are (not) used? Do gravestones have a Shintoist or Buddhist shape? Which trees or flowers are part of the site, for instance, cherry trees as symbols of the ‘Japanese spirit’ and the death of the warrior? Two things are important with regard to the documentation: firstly, especially at the beginning of a project, researchers do not know what will be important in the end. Therefore, it is wise to include as much information as possible and sort it out later. It is advisable to visit memorial sites as early as possible in the research process and to take enough time, as historians can never be sure how much work lies ahead. A good camera with several batteries and memory cards is indispensable, and a GPS function is very helpful in order to locate photos quickly on maps. And one should also calculate enough time in order to systematically catalogue notes,

250

Chapter 9 How to access written and visual sources photos and sketches as soon as possible. Especially during short research stays, researchers usually tend to visit as many places as possible, but without good organisation quickly drown in their material—less is more. Secondly, one problem Japan researchers often encounter is that they may not perceive certain cultural codes, like iconographic motifs, symbols or references, and may place them in their own cultural context and therefore misinterpret them. Historians should beware of hasty interpretations and they should not be afraid to ask Japanese colleagues specialising in the field under study even the most trivial questions (see Liu-Farrer, Ch. 4.3; McLaughlin, Ch. 6). The same is true of talking to and asking people at the memorial sites for information. Schölz experienced that many Japanese are happy to help—even if they just tell foreign researchers whom to contact for further information. After the documentation of the memorial sites’ current state, the next step historians have to take is an analysis of the context in which the object was built and the historical change in both its appearance and its use. Thus, it is essential to consult further sources. Usually, the researcher can find them in archives: There may be references to the intention and the financing of the monument in the private archives of donors, for example. Materials produced by the artist can provide numerous references to the context in which a monument was designed, possibly also to alternative plans that were not realised. Public archives, mostly prefectural archives (see this chapter, Peucker/Schmidtpott/Wagner, Ch. 9), often hold sources about the approval of a monument’s construction and possibly required changes to its appearance. Laws and ordinances provide an indication of the legal dimension of building monuments in public. Newspaper archives, image databases or albums can help researchers to understand an object’s historical form and its use, including inauguration ceremonies or rituals performed. Edited collections of primary sources and local histories as secondary sources may provide further insights in this respect. Historical maps may help researchers to locate former sites if monuments have been relocated or to understand the situation of a public space that has changed over time. And finally, in some cases, court records, although they are rather difficult to access, can also provide information about conflicts that took place about sites of remembrance.

Summary Depending on the research project, it is sometimes necessary to explore different kinds of sources located at various places. In doing so, one will not always be successful. Many visits to an archive or a local library end in disappointment, because the researcher cannot find anything, or existing sources do not reveal anything of value. But it is always worth a try, because one can also come across information that changes and expands our knowledge of the object under study. The historian should never forget that sources are of immense importance but, at the same time, only the basis of a good study; even more important are smart questions, meaningful approaches and hypotheses, and above all clever interpretations and answers.

251

9.2 Writing transnational history through archival sources Sheldon Garon

One of the most exciting fields today is transnational or global history. Historians increasingly explain developments within Japan by going beyond national history to explore the country’s important connections with the rest of the world. Transnational history spotlights the movement of ideas, institutions, peoples and practices across borders and oceans. The challenge for scholars of Japan is how to research topics that require the use of not only Japanese archives, but also those of other countries (Garon 2017).

Discovering the transnational in the sources I offer my own experiences as a researcher, having devoted the past two decades to writing transnational history centred on Japan. My book Beyond our means: Why America spends while the world saves (Garon 2012) is a transnational history of saving and consumption. Despite the subtitle, one-third of the book deals with Japan. This study became a global history— including America and several European and Asian countries—primarily because of serendipitous discoveries in the archives. I had originally intended to write a national history of the Japanese state’s efforts to promote popular saving. However, as I read reports by Japanese officials and reformers, I realised that they were fundamentally involved in the global exchange of knowledge. Like their Western counterparts, the Japanese vigorously investigated best practices in other nation-states that similarly sought to cultivate hardworking, thrifty populaces towards the end of the nineteenth century. The Home Ministry’s journal Shimin (The Subject, 1906–44) frequently presented models of local savings associations in Europe while popularising Victorian ideas of thrift. One official’s enthusiastic report of Britain’s Post Office Savings Bank led, for example, to the establishment of Japan’s famous postal savings system in 1875. The archives also reveal the impact of transnational models on Japan’s savings-promotion programmes in the course of the twentieth century. The Ministry of Finance’s archives of pre-1945 and postwar financial history (Shōwa zaiseishi shiryō and Sengo zaiseishi shiryō) were particularly useful. During the First World War, the Ministry of Finance and Home Ministry dispatched teams of bureaucrats to Europe and America to survey war savings campaigns and other home-front programmes. They were especially impressed by the British government’s National War Savings Committee, which mobilised local savings associations and women’s groups. Based on a report by its resident official in London in 1924, the Ministry of Finance explicitly emulated Britain’s National Savings campaign structure in its own Campaign to Encourage Diligence and Thrift. After Japan embarked on war with China in 1937, officials again drew on the British model to establish a war savings campaign that continued through the Pacific War.

252

Chapter 9 How to access written and visual sources Following Japan’s defeat in 1945, according to Ministry of Finance documents, the state once again investigated European national savings campaigns. Officials observed that Japan was not alone in mounting savings and austerity drives to recover from wartime devastation. The Soviet, French, Belgian and Dutch governments were all haranguing their people to save more and spend less. The Japanese people, asserted the ministry’s leadership, should emulate the postwar British, who accepted lives of austerity to save for their country’s economic recovery. In 1952, the government created the Central Council for Savings Promotion. Housed within the Bank of Japan (BOJ), the Central Council coordinated savings campaigns with local councils and civic groups for the next several decades. A former vice-minister’s introduction permitted me to access the Central Council’s working archive. Materials in the Bank of Japan library supplemented its newsletters and documents. The BOJ journal Chochiku jihō (Savings Times) regularly reported on innovative savings promotion practices in Europe during the 1950s and 1960s. The working archives also told the little-known story of Japanese efforts to promote savings-led development in Asia in the 1980s and 1990s, as illustrated by the BOJ’s frequent seminars with the region’s central bankers. Reading Japanese sources for their transnational connections places Japan in a more global framework. But one could go further—and I did—to use Japanese archives as a springboard to explore the global circulation of knowledge among other societies. Since the Meiji era, Japanese have been among the world’s most energetic transnational actors and learners (Konishi 2013). Their detailed investigations of other nations’ institutions permit the historian of Japan to see global currents more readily than scholars of Western Europe and the United States. Midway through my research, I decided to follow my Japanese sources to their sources and write a truly global history of saving since 1800. I first followed Japanese savings-promotion officials to Britain, where I mined the National Archives to reconstruct the story of the statesponsored National Savings Movement from the First World War to the 1970s. I similarly uncovered the history of the U.S. Treasury Department’s savings campaigns during the two world wars. The history of saving had not been studied comprehensively in either country. I was, moreover, able to connect Anglo-American developments with those occurring in Japan and elsewhere. My Japanese actors next led me to the obscure, but immensely useful archive of the World Savings Banks Institute (WSBI), a Brussels-based consortium of savings banks and postal savings banks. The collection contains historical materials from many nations around the world from the 1920s to the present. Most are written in English, German or French. I stumbled across the WSBI’s predecessor, the International Thrift Institute, after reading of the dispatch of a Japanese official to the organisation’s First International Thrift Congress in Milan in 1924. The WSBI archivist in turn arranged my research visits to affiliated savings banks associations in France, Italy, Sweden and Germany. Increasingly, my visits to the Bank of Japan have made me aware that Japan has been a maker, as well as a taker, of transnational knowledge. The Central Council for Savings Promotion inspired the establishment of a similar council in the Bank of Korea in 1969. Japan’s enormous postal savings system also served as a model for postal savings banks in Singapore and Malaysia. In 2001, I visited the archives of the Korean and Malaysian central banks, plus Singapore’s National Archives. The Bank Negara Malaysia’s staff were particularly helpful. They 253

Sheldon Garon were in the midst of emulating the Bank of Japan’s savings campaigns, and were quite interested in my knowledge.

Designing a transnational study I am currently writing another transnational history, this time about home fronts in Japan, Germany and Britain during the Second World War (Garon 2020). Whereas my book on saving became a global history only after I had discovered connections in the Japanese archives, my current project rests on a transnational-historical design from the start. Home fronts did not develop in isolation. Rather the very concept and practices of the home front were transnationally constructed from the First World War through the Second World War. Officials and experts in each nation investigated others’ mobilisation of civilians in the areas of food security, civilian defence against air raids and the maintenance of morale. At the same time, strategies of how to destroy home fronts through bombing and blockades circulated throughout the world. In each country, I selected archives that would best reveal these connections. I began with the user-friendly U.K. National Archives. Commencing in 1918, Air Ministry files detail the Royal Air Force’s (RAF) growing commitment to strategic bombing that targeted cities and strove to break civilian morale. The interwar RAF refined its strategy based in part on surveys of air strategists in Italy, France, Germany and the United States. Moreover, in 1938–39, the British Committee of Imperial Defence issued several secret reports on the ‘air lessons’ derived from the Japanese bombing of Chinese cities and air raids in the Spanish Civil War. Transnational knowledge similarly informed Britain’s efforts to defend its own cities from air attacks. By the early 1930s, the government’s Air Raid Precautions Committee recognised that Britain lagged far behind the Soviets, Germans and other Europeans in preparations for civilian defence. Learning from others, Britain entered the Second World War with a comparable nationwide system of neighbourhood air wardens, first-aid workers and ‘fire watchers’. The archives also spotlight the widely accepted transnational lesson of the First World War— that the British blockade of Germany’s food supply had broken civilian morale and forced the German government to surrender in 1918. In the Second World War, as key documents make clear, the high command attempted, unsuccessfully, to bring about another ‘1918’ against Germany through bombing and blockades. German archives were more of a challenge. I had only recently learned to read German. Rather than replicate the histories of National Socialism written by German scholars, I focused on the under-researched issue of transnational connections between the Nazi home front and those of other nations. For the exchange of knowledge with Japan, I read the records and clippings of the Deutsch-Japanische Gesellschaft in the Bundesarchiv Berlin. Another rich source for transnational learning was Die Sirene (1933–44), the magazine of the multi-millionmember German Air Defence League (Reichsluftschutzbund). As Nazi officials devised their air defence structures, articles featured civilian defence organisations and air raid drills in Japan, Poland, Sweden and elsewhere. Die Sirene in turn inspired Japanese authorities to publish their own illustrated magazine, Kokumin bōkū (Civilian Defence, 1939–44). I compared the magazines in their depictions of everyday practices in both countries, notably the mobilisation of neighbourhood women to extinguish incendiary bombs. The best collection turned out to be the political archive (Politisches Archiv) of the German Foreign Office. From 1927 to

254

Chapter 9 How to access written and visual sources 1941, the Foreign Office instructed its embassies to gather information on others’ civilian defence while arranging visits by Japanese and other foreign delegations to Germany’s air defence facilities. The several folders on air defence (Luftschutz) also contain the German Air Ministry’s bimonthly reports on civilian defence in more than twenty countries, including Japan. For the American bombing of Japan, I relied on the Air Force Historical Research Agency archive in Alabama. We assume that racism played a huge role in the U.S. firebombing campaign against Japanese cities. Yet the planning documents show that the incendiary attacks were instead modelled on what the Americans learned from British ‘area bombing’ of German cities. In Japan, as in Germany, some of the best transnational evidence comes from the Ministry of Foreign Affairs archive. It holds numerous reports of the German and British home fronts from Japanese diplomats and military attachés in Europe. They are supplemented by the microfilmed MAGIC Documents, the daily translations of U.S.-intercepted Japanese diplomatic cables. The National Institute for Defense Studies (Bōei Kenkyūjo) possesses the influential reports of Japanese military missions that went to Germany to study air defence. The most comprehensive collection of Japanese home-front magazines and books is found in the National Shōwa Memorial Museum (Shōwakan). These materials showcase the impact of foreign models on Japan’s food rationing system and civilian defence practices.

Some advice To researchers who wish to conduct a transnational study, I encourage you to be ambitious. Transnational archival research may appear intimidating, but there are steps you can take to realise its potential. First, leverage your languages. Most scholars of Japan know at least one European (or Asian) language in addition to English. A little French or German can lead to many discoveries. Second, be prepared to spot the transnational connections in your documents that others have overlooked. Finally, think about how your transnational perspectives could contribute both to debunking myths of Japanese ‘uniqueness’ and enriching the field of Global History itself.

255

9.3 Accessing quantitative data for qualitative research: White Papers, official statistics and micro datasets Shinichi Aizawa and Daisuke Watanabe

This handbook mainly discusses qualitative research methods. However, researchers use quantitative information in qualitative research, too. Both authors of this essay applied qualitative research methods in their doctoral research. Shinichi Aizawa analysed social transformation and discourses about school education in postwar Japan, and Daisuke Watanabe studied participation and activity processes in club activities for people after retirement. By contextualising qualitative research results with quantitative information, our findings became more persuasive. We consider both qualitative and quantitative data to be valuable, depending on the research question you would like to answer. Because comparing qualitative results with quantitative data can increase the relevance of one’s findings, this essay introduces ways to access White Papers (hakusho) and quantitative data on Japan, mainly official government statistics on e-Stat, and micro datasets. We begin this introduction with White Papers, because they are a good starting point for researchers who want to understand Japanese society, both concerning numbers and qualitative contexts.

White Papers in Japan: Characteristics and access In Japan, White Papers are officially defined as ‘government publication materials edited by central government agencies, which are prepared to inform the nation of the status of politics, economics, societies and government measures’ (NDL 2019) by the administrative vice-ministers’ conference in 1963. Generally, White Papers are annual government reports. Nowadays, the Japanese government publishes three types of White Papers: 1. statutory White Papers which are submitted by the Cabinet to the Diet upon Cabinet decisions based on laws, 2. White Papers distributed at Cabinet meetings and 3. other White Papers. Notably, White Papers of the first and second type are decided at Cabinet meetings and introduce a series of current policies and policy planning. White Papers are official documents and often written in a formal style that is as dry as dust. In historical or Political Science research, Japanese scholars analyse documents such as the minutes of various councils’ meetings, resources of parliaments or interview data. Just like the latter, White Papers are very useful documents in understanding how policymakers recognise social issues, laws and social institutions. White Papers contain longitudinal data and are structured in a similar manner. Most White Papers are published every year over a longer period of time. Therefore, long-term comparison of White Papers is relatively easy. All White Papers are openly available online, but most White Papers are available in Japanese only. Often, only a summary of the respective White Paper is translated into English (see Table

256

Chapter 9 How to access written and visual sources 9.1). The e-Gov website run by the Government of Japan and the National Diet Library collects and presents information on White Papers. Older White Papers that were only published in Japanese have recently been digitalised and made available by the National Diet Library (see Schmidtpott/Schölz, Ch. 9.1; Zachmann, Ch. 4). Table 9.1: List of main White Papers in Japan Governing agency

Title [since] (Japanese / English)

Cabinet Office (CAO)

Annual Report on the Japanese Economy and Public Finance [2001–/2001–] (**) White Paper on Nuclear Energy [1956–/2016–] (**) White Paper on Disaster Management [1974–/2016–] (*) White Paper on Children and Young People [1956-/2012–] (**) Declining Birth Rate White Paper [2004–/2008–] (2008–2015*; 2016–**) Annual Report on the Ageing Society [1996–/2002–] (**) Annual Report on Government Measures for Persons with Disabilities [1996–/2003] (**) White Paper on Traffic Safety in Japan [1971–/1997–] (*) White Paper on Gender Equality [1996–/1996–] (**)

Ministry of Justice (MOJ)

White Paper on Crime [1960–/2000–] (*) Immigration Control [2003–/2005–] (*)

Ministry of Education, Culture, Sports, Science and Technology (MEXT)

White Paper on Science and Technology [1958–/1963–] (*) White Paper on Education, Culture, Sports, Science and Technology [1959–/ 2003–] (**)

Ministry of Health, Labour and Welfare (MHLW)

Annual Health, Labour and Welfare Report [1956–/2007–] (*)

Ministry of Land, Infrastructure, Transport and Tourism (MLIT)

White Paper on Land, Infrastructure, Transport and Tourism in Japan [2001–/2001–] (*)

National Police Agency (NPA)

White Paper on Police [1973–/2016–] (2016–2017*; 2018–**)

Japan Tourism Agency (JTA)

White Paper on Tourism [1997–/2007–] (2007–2017*; 2018–**)

White Paper on the Labour Economy [1948–/2003–2005, 2012–] (**)

White Paper on Land [1990–/NA] Measures for Crime Victims [2006–/2008–] (**)

* English version is available (as of Oct. 2020). ** English summary version is available (as of Oct. 2020).

Each ministry and agency assigns a team, whose head is the director general of the respective agency, to collect data and to write a White Paper about topics that fall under its jurisdiction every year. All White Papers utilise various official social survey data, such as 53 fundamental statistics, which include the population census, surveys on employment status and data of administrative records on public health insurance, number of crimes, immigration and other social surveys. Most White Papers visualise statistical data in graphs and tables in order to make

257

Shinichi Aizawa and Daisuke Watanabe them more comprehensive. However, most White Papers in Japan do not reference academic articles. Writing White Papers is teamwork. All teams who write White Papers are strictly managed in a similar manner as in lawmaking processes. Although strictly managed, White Paper writing teams have some freedom to address middle- and long-term policy directions. Therefore, researchers should pay attention to changing terms, vocabulary and expressions. White Papers which are adopted in Cabinet meetings have a considerable influence on policymaking and political discourses. For example, the White Paper on gender equality (Danjo byōdō sankaku hakusho) referred positively to separate surnames for married couples in 2011, when the DPJ was the ruling party. After 2013, this was changed into the expression ‘give careful consideration’ to the separation of surnames for married couples under the Abe Cabinet, which is known as a very conservative government. Thus, we can grasp social change by analysing expressions or vocabulary in White Papers. Particularly, we should try to analyse not only what is written but what is not written in White Papers. We also should take into account photographs and other figures in order to not focus only on texts. Most statutory White Papers and White Papers distributed by the Cabinet Office have two parts. One part is a selected yearly theme, and another part features categorised information on laws, policies and empirical data. The part on the selected yearly theme is very useful when comparing past and current White Papers. By tracing yearly themes, we can reveal dynamics in Japanese society and politics. For example, the themes in the Annual health, labour and welfare report (Kōsei rōdō hakusho), which has been published from 1956 onwards, have changed over time. Before 1973, the themes were related to the establishment of a universal health insurance and pension system and the development of social welfare institutions. From the 1973 oil shock to the beginning of the new millennium, the Ministry focused on policies related to an ageing society and pursued realising a ‘Japanese-style welfare system’ (nihongata fukushi shakai). Finally, after 2000, the Ministry of Health, Labour and Welfare (MHLW) struggled to introduce measures to address the declining birth rate and the so-called super-aged society. White Papers from this period onwards emphasized new roles and activities for older people and local communities in a super-aged society (Watanabe 2009) as well as on work-life balance recommendations for the working population. To sum up, analysing the changing topics and terms in White Papers can become useful resources in understanding discourses on Japanese society.

Official macro statistics via e-Stat After reading White Papers, you may become interested in official statistics. Of course, you can access World Bank data, OECD data and UN data, too. It is often insightful to compare Japan with other countries. However, we think that it is even better to check each national dataset in addition, because these international comparative datasets may sometimes use nonreliable sources. Japanese official statistics are all collected on the above-mentioned website eStat (see Hommerich/Kottmann, Ch.10). You can access official statistics from various fields on this website. However, we advise you to use the Japanese rather than the English website, because there are great differences with regard to the quantity and diversity of the available data. Currently, you can access 1,534,675 datasets through the Japanese website, but only

258

Chapter 9 How to access written and visual sources 395,464 datasets in English. In other words, the English website provides less than 30% of the datasets available on the Japanese website. Furthermore, the English website lacks many statistics; for instance, there is no English access to the Basic survey on Japanese schools (Gakkō kihon chōsa), which provides the most reliable official statistics about Japanese school education (as of the end of October 2020). Therefore, we strongly recommend using this website in Japanese. If you can break through a wall in Japanese, you can collect official statistics easily. I will present one example of how to access, calculate and make a table in the field of education from this website. First, you can choose a year, a school level and a certain kind of statistic. Together with my colleague Kagawa Mei, I created a table based on these School basic statistics from e-Stat (Aizawa 2018). It shows not only the numbers of high school enrolments in general, but also the differences between public and private schools. The website of e-Stat makes various tabulations possible. Although e-Stat is convenient, we recommend you to check printed materials at the beginning of your research in order to understand the numbers in the Japanese social context and to select the most suitable tables for your research. Materials related to official macro statistics are not only available in the National Diet Library (NDL; Kokuritsu Kokkai Toshokan) but also in the library of the Statistics Bureau (Tōkeikyoku) in Tokyo. Naturally, you can access some materials in the university libraries and public libraries in Japan. You can also purchase them from the official gazette cooperation of Japan. The reason why we recommend you to check paper materials at the initial stage of your research is that you should first familiarise yourself with the topic by grasping a sense of relevant and realistic numbers. For instance, do you know how many students advanced from upper secondary schools to colleges in 2019 in Japan? You do not need to know the accurate number (the answer is 578,382) and may not know the approximate number, but it is helpful to make yourself familiar with some rough numbers about this topic. For example, the birth rate of Japanese babies is around one million in the twenty-first century, but it is decreasing and has not exceeded one million since 2016. Japan’s advancement rate to upper secondary education is over 95%, and 55% of the graduates from the upper secondary level advance to tertiary education. The last number, 578,382, is connected to this social, institutional and demographic situation in Japan. You will be able to imagine Japanese society with approximate numbers if you become familiar with numbers relevant in your field of research.

Accessing micro datasets from data archives If you are interested in statistical analysis in Japan, you can also access datasets from data archives and execute secondary analysis using these data sets. Emma Smith’s book about secondary analysis is a valuable resource that refers to access to quantitative data in Japan (Smith 2008, p. 199). The Inter-University Consortium for Political and Social Research (ICPSR) also features datasets about Japan, for example from the Japan General Social Survey (JGSS). The Social Science Japan Data Archive (SSJ Data Archive) is the most famous data archive among Japanese researchers in Japan, although some universities and research institutes like Rikkyo University and Keio University have their own data archives. As of the end of October 2020, SSJ Data Archive has provided 1,424 datasets, and there is no difference between Japanese

259

Shinichi Aizawa and Daisuke Watanabe and English websites when it comes to access to datasets on the matter of quantity of datasets, although some datasets include instructive information only in Japanese. You can get access to this data archive’s datasets online after registration (see Hommerich/Kottmann, Ch. 10). Aizawa has worked with data from data archives and has also made his own data available through the SSJ Data Archive. In his research, Aizawa (2016) analysed the effects of social origins and schooling on advancement to schools comparing Japan and Taiwan. The dataset that Aizawa (2016) used for the Japanese case from the SSM 2005 (Social stratification and social mobility survey, provided by the University of Tokyo) is accessible to all researchers now via the SSJ Data Archive. This data archive collects not only new survey data sets but also collects and digitally restores older datasets. Aizawa and Yutaka Koyama (2016) carried out data restoration and secondary analysis. The two authors of this essay continue to restore historical datasets. We also published a book in Japanese about time use in collective housing in 1965 (Watanabe et al. 2019). These data on time use have also been digitally restored and will be accessible to researchers via SSJ Data Archive. Therefore, secondary analysis via data archives expands frontiers not only in contemporary statistical fields but also in the historical analysis of Japan.

Final remarks Not only should researchers in Japanese Studies combine various methodologies across qualitative, quantitative, historical and comparative methods, but so should those in all Area Studies. The division between quantitative and qualitative work remains all over the academic world. However, quantitative information on societies has qualitative contexts, although some highly mathematical Social Sciences ignore this context. Therefore, researchers in Japanese Studies should contextualise both quantitative and qualitative information on a cross-national or global level.

260

Further reading Armstrong, Catherine (2015): Using non-textual sources: A historian’s guide. London: Bloomsbury Academic. Asato, Noriko (ed.) (2013): Handbook for Asian Studies specialists: A guide to research materials and collection building tools. Santa Barbara, CA: Libraries Unlimited. Becker, Bernd W. (2018): Information literacy in the digital age: Myths and principles of digital literacy. In: School of Information Student Research Journal 7, No. 2, pp. 1–8. Pitman, Lesley (2015): Supporting research in Area Studies: A guide for academic libraries. Amsterdam: Chandos Publishing. Reidsma, Matthew (2019): Masked by trust: Bias in library discovery. Sacramento, CA: Library Juice Press. Smith, Emma (2008): Using secondary data in educational and social research. Berkshire: Open University Press.

References ACRL (Association of College & Research Libraries) (2016): Framework for information literacy for higher education. http://www.ala.org/acrl/sites/ala.org.acrl/files/content/issues/infolit/Framework_ILHE. pdf, [Accessed 17 May 2020]. Aizawa, Shinichi (2016): A comparative sociological study of Japanese and Taiwanese upper secondary education. In: Educational Studies in Japan 10, pp. 33–48. Aizawa, Shinichi (2018): Universal participation in school education as a historical process in modern Japan. In: Yonezawa, Akiyoshi/Kitamura, Yuto/Yamamoto, Beverly/Tokunaga, Tomoko (eds.): Japanese education in a global age: Sociological reflections and future directions. Singapore: Springer, pp. 35–52. Aizawa, Shinichi/Koyama, Yutaka (2016): Japanese historical challenge of social inclusion: From the digitally restored household survey data on receiving social security in 1952. In: International Journal of Pedagogy, Innovation and New Technologies (IJPINT) 3, No. 2, pp. 74–84. ALA (American Library Association) (2013): Digital literacy, libraries, and public policy: Report of the Office for Information Technology Policy’s Digital Literacy Task Force. http://www.districtdispatch.org/ wpcontent/uploads/2013/01/2012_OITP_digilitreport_1_22_13.pdf, [Accessed 17 May 2020]. Aoyama, Hideyuki (2003): Joshō: Nihon ni okeru ākaibaru saiensu no keisei to kadai. In: Zenkoku Rekishi Shiryō Hozon Riyō Kikan Renraku Kyōgikai (ed.): Nihon no ākaibuzu ron. Tōkyō: Iwata Shoin, pp. 15–46. Asato, Noriko (ed.) (2013): Handbook for Asian Studies specialists: A guide to research materials and collection building tools. Santa Barbara, CA: Libraries Unlimited. Balzert, Helmut/Schäfer, Christian/Schröder, Marion/Kern, Uwe (2008): Wissenschaftliches Arbeiten: Wissenschaft, Quellen, Artefakte, Organisation, Präsentation. Herdecke: W3L-Verlag. Becker, Bernd W. (2018): Information literacy in the digital age: Myths and principles of digital literacy. In: School of Information Student Research Journal 7, No. 2, pp. 1–8. Block, Sharon (2020): Erasure, misrepresentation and confusion: Shortcomings of JSTOR topics on women’s and race histories. In: Digital Humanities Quarterly 14, No. 1. http://www.digitalhumanities.org/dh q/vol/14/1/000448/000448.html, [Accessed 29 May 2020]. Böhner, Dörte (2013): Verbessern Discovery Systeme die Informationskompetenz? In: 027.7 Zeitschrift für Bibliothekskultur/Journal for Library Culture 1, No. 2. DOI: 10.12685/027.7 – 1–2–26. Garon, Sheldon (2012): Beyond our means: Why America spends while the world saves. New Jersey, NJ: Princeton University Press. Garon, Sheldon (2017): Transnational history and Japan’s ‘comparative advantage’. In: Journal of Japanese Studies 43, pp. 65–92. Garon, Sheldon (2020): On the transnational destruction of cities: What Japan and the United States learned from the bombing of Britain and Germany in the Second World War. In: Past & Present 247, pp. 235–271. Gordon, Andrew (2003): Studying the social history of contemporary Japan. In: Bestor, Theodore C./ Steinhoff, Patricia G./Lyon-Bestor, Victoria (eds.): Doing fieldwork in Japan. Honolulu, HI: University of Hawai‘i Press, pp. 261–273.

261

References Fujiwara, Yoshio (ed.) (2008): Toshokan: Ākaibuzu no genba kara. In: KAN Special Issue No. 15: Toshokan: Ākaibuzu to wa nani ka, pp. 223–295. Kokuritsu Kōbunshokan (2001): Kokuritsu Kōbunshokan 30shūnen o mukaete. Tōkyō: Kokuritsu Kōbunshokan. Konishi, Sho (2013): Anarchist modernity: Cooperation and Japanese-Russian intellectual relations in modern Japan. Cambridge, MA: Harvard University Asia Center. Lyon-Bestor, Victoria (2003): Appendix: Digital resources and fieldwork. In: Bestor, Theodore C./Steinhoff, Patricia G./Lyon-Bestor, Victoria (eds.): Doing fieldwork in Japan. Honolulu, HI: University of Hawai‘i Press, pp. 367–373. Matsuoka, Tadaaki (2011): Ākaibuzu ga shakai o kaeru: Kōbunsho kanrihō to jōhō kakumei. Tōkyō: Heibonsha. Matsuzaki, Yuko (2007): Business archives in Japan: An overview and access issues. Paper presented at the Japan-U.S. Archives Seminar on ‘Access to Archives: The Japanese and American Practices’, Tokyo, 9– 11 May 2007. http://www.archivists.org/publications/proceedings/accesstoarchives/06_Yuko_MATSUZ AKI.pdf, [Accessed 11 March 2019]. Matsuzaki, Yuko (2017): Communicating the value of business archives to business archives: The Shibusawa BA project and corporate archives in Japan. Presentation at the 2017 ICA SBA conference ‘The Future Roles of Business Archives’, Stockholm. https://www.shibusawa.or.jp/center/ba/bunken/doc012_ icasba_stockholm.html, [Accessed 11 March 2019]. NCC (The North American Coordinating Council on Japanese Library Resources) (2017): NCC’s Online Guide to Research Access for Japanese Museums, Libraries, and Archives (MLA): General affiliation procedure for Japanese institutions. https://guides.nccjapan.org/researchaccess/procedure, [Accessed 14 February 2020]. NCC (The North American Coordinating Council on Japanese Library Resources) (2020): About the NCC. https://guides.nccjapan.org/homepage/about/ncc, [Accessed 22 August 2020]. NDL (National Diet Library) (2009): Research navi: English guide. https://rnavi.ndl.go.jp/rnavi/english.ph p, [Accessed 15 May 2020]. NDL (National Diet Library) (2013): Web archiving project (WARP): FAQ. https://warp.ndl.go.jp/info/W ARP_qanda_en.html#01_02, [Accessed 15 May 2020]. NDL (National Diet Library) (2019): White Papers. https://rnavi.ndl.go.jp/politics/entry/e-white-papers.ph p, [Accessed 21 August 2020]. Nichigai Asoshētsu Henshūbu (ed.) (2003): Shintei: Kigyō hakubutsukan jiten. Tōkyō: Nichigai Asoshētsu. NTK (Nihon Toshokan Kyōkai) (2020): Toshokan rinku shū. https://www.jla.or.jp/link/tabid/95/Default.a spx, [Accessed 17 February 2020]. Ōwa, Kunio (2018): Ronbun sakusei gaido shakai kagaku o manabu gakusei no tame ni. Tōkyō: Tōkyō Nōgyō Daigaku Shuppankai. Pitman, Lesley (2015): Supporting research in Area Studies: A guide for academic libraries. Amsterdam: Chandos Publishing. RCOS (Research Center for Open Science and Data Platform) (ed.) (2017): CiNii research (Discovery Platform). https://rcos.nii.ac.jp/en/service/research, [Accessed 14 February 2020]. Reimer, Torsten (2020): A manifesto for the digital shift in research libraries: Report from RLUK’s digital shift working group. https://www.slideshare.net/TorstenReimer/a-manifesto-for-the-digital-shift-in-resea rch-libraries, [Accessed 20 May 2020]. SBPK (Staatsbibliothek zu Berlin Preussischer Kulturbesitz) (2020): Über CrossAsia. https://blog.crossasia. org/about/, [Accessed 20 May 2020]. Schmidtpott, Katja (2012): Neue Perspektiven der historischen Industriestadtforschung in Japan. In: IMS —Informationen zur Modernen Stadtgeschichte 1, No. 12, pp. 87–103. Schölz, Tino (2016): Die Gefallenen besänftigen und ihre Taten rühmen. Gefallenengedenken und politische Verfasstheit in Japan seit der Mitte des 19. Jahrhunderts. Berlin: De Gruyter. Shibusawa Eiichi Memorial Foundation (n.y.): Sekai/nihon no bijinesu ākaibuzu. https://www.shibusawa.o r.jp/center/ba/bunken/index.html, [Accessed 29 February 2020]. Smith, Emma (2008): Using secondary data in educational and social research. Berkshire: Open University Press. Takayama, Masaya (2008): Nihon ni okeru bunsho no hozon to kanri. In: KAN, Special Issue, No. 15: Toshokan: Ākaibuzu to wa nani ka, pp. 42–58. Takeda, Tatsuya (2008): Nihon zenkoku sangyō hakubutsukan meguri. Tōkyō: PHP Kenkyūsho. ULBM (Universitäts- und Landesbibliothek Münster) (2019): 4. Wie suchen Sie? (= LOTSE: Kompass zum wissenschaftlichen Arbeiten). https://www.ulb.uni-muenster.de/lotse/literatursuche/suchstrategien/thema tisch_suchen/wie.html, [Accessed 27 November 2019].

262

Chapter 9 How to access written and visual sources Watanabe, Daisuke (2009): Creating a community: Ageing in urban Japan. In: Umegaki, Michio/Thiesmeyer, Lynn/Watabe, Atsushi (eds.): Human insecurity in East Asia. Tokyo: United Nations University Press, pp. 211–231. Watanabe, Daisuke/Mori, Naoto/Aizawa, Shinichi (eds.) (2019): Sōchūryū no tanjō: Danchi to seikatsu jikan no sengoshi. Tōkyō: Seikyūsha. Weber, Torsten/Krickel, Nina C. (2018): Japan. In: Busse, Laura/Enderle, Wilfried/Hohls, Rüdiger/Meyer, Thomas/Prellwitz, Jens/Schuhmann, Annette (eds.): Clio guide: Ein Handbuch zu digitalen Ressourcen für die Geschichtswissenschaften. Berlin: Clio-online und Humboldt-Universität zu Berlin, pp. D.73–1– D.73–24. DOI: 10.18452/19244.

263

Chapter 10 How to combine methods: Mixed methods designs Carola Hommerich and Nora Kottmann

1. Introduction ‘In quantitative research, you do not see people’s faces; in qualitative research you tend to see nothing but their faces’ (Kobayashi/Hommerich 2018, p. 321). To overcome such drawbacks of purely quantitative or qualitative research designs, a methodological approach that integrates both has received increasing attention over the past few years. By combining qualitative and quantitative data, mixed methods research2 enables researchers to use the strengths of both approaches while minimising—and compensating for—their respective limitations. Integrating different methods can, therefore, be a strategy for supplementing or complementing the findings of one approach with the other, thereby gaining a more comprehensive, multifaceted understanding of social problems or phenomena (Creswell/Creswell 2018, p. 316; Venkatesh et al. 2016, p. 442). This means that ‘mixed methods research can […] answer research questions that the other methodologies cannot’ (Tashakkori/Teddlie 2003b, p. 15). Since the late 1980s, this approach has gained significant attention in the Social Sciences as well as in Area Studies. While mixing methods might seem simple, there are complex rules, models and hurdles involved, which researchers need to consider. We wrote this chapter with Japan researchers in mind (mostly those who are in the early stages of their careers), who have a qualitative background and are unfamiliar with mixed methods research but interested in combining qualitative and quantitative data. While we also mention some topics that might be more relevant for experienced scholars (like conducting your own survey), the chapter mainly aims to provide an overview of basic ideas and key terms in mixed methods research as well as basic guidelines on why and how to use mixed methods designs. After a brief introduction, we outline three core models that researchers can adapt to their individual needs and offer concrete advice on how to select a research design, how to collect and analyse data, and how to present results in a written report. Finally, we address possible obstacles that early career researchers might face and share practical advice on how to avoid them. Given that this handbook focuses primarily on qualitative methods, we have included a few additional paragraphs on quantitative methods and share information on further reading and relevant online sources. Throughout the whole chapter, we refer to and give advice on topics that are specific when doing research on (and possibly in) Japan.

1 Translated from Japanese by the authors. 2 In line with John Creswell, Abbas Tashakkori and Charles Teddlie we here use the term ‘mixed methods research’. Other terms that are often used interchangeably are ‘multi-method research’, ‘mixed research’ or ‘mixed methodology’ (see this chapter, Pekkanen/Pekkanen, Ch. 10.1).

264

Chapter 10 How to combine methods

2. What are mixed methods designs? This relatively new methodological approach is defined as follows: A mixed methods design requires the collection of both qualitative and quantitative data in response to an underlying research question or to test theory-driven hypotheses (Johnson et al. 2007, p. 119). The combination of quantitative and qualitative methods—two traditionally strongly separated approaches—was initially received with enthusiasm as well as scepticism (for an overview of early discussions Brannen 1992). Recalling its early stages, Michael Fetters (2016, p. 3) acknowledges that previously researchers had also combined the two methods in the fields of Anthropology, Psychology, Sociology and Natural and Health Science. However, it was in the late 1980s that a ‘new intellectual and practice environment’ led to the ‘birth of modern mixed methods research’ (ibid., p. 4). Since then this approach has significantly increased in popularity to become the ‘third methodological movement in social science research’ (Tashakkori/ Teddlie 2010, p. ix). This can be seen in the vast body of literature published across various academic disciplines (Hesse-Biber/Johnson 2015; Johnson/Christensen 2016; Tashakkori/ Teddlie, 2003a), in the emergence of influential journals (see text box below), and recent private as well as public funding opportunities (Dahlberg et al. 2010). Journals promoting mixed methods research This is a selection of journals featuring and promoting mixed methods research (also Creswell/Creswell 2018, p. 314): • • •

International Journal of Multiple Research Approaches: www.ijmra.org Journal of Mixed Methods Research: https://journals.sagepub.com/home/mmr Quality and Quantity: www.springer.com/journal/11135

There are various reasons for using mixed methods designs, from theoretically induced decisions to pragmatism and opportunity structures. The most commonly cited reason for using a mixed methods approach is methodological triangulation (see text box below; see also this chapter, Imai, Ch. 10.3).3 Denzin (1978) identifies methodological triangulation as one way of achieving a more comprehensive understanding of a research object. Differentiating between within-method and between-method triangulations, he recommends the latter, as the ‘bias inherent in any particular data source, investigators, and particular method will be cancelled out when used in conjunction with other data sources, investigators, and methods’ (ibid., p. 14). The use of different methods can help a researcher acquire a more comprehensive awareness of their research problem; for example, using qualitative data to interpret quantitative results (i.e. complementarity) or using results from one method to develop the other method (i.e. development) (Hesse-Biber 2010, p. 5). However, the appropriateness or necessity of a mixed methods design depends on the research question and previous research on the topic (see this chapter, Pekkanen/Pekkanen, Ch. 10.1).

3 Other forms of triangulation include data, investigator and theory triangulation (Denzin 1978).

265

Carola Hommerich and Nora Kottmann Methodological triangulation Methodological triangulation usually refers to the ‘use of more than one method for gathering data’ (Hammond/Wellington 2013, p. 145). In mixed methods research, data collection abides by the established rigorous procedures for each approach, which implies profound knowledge of both qualitative and quantitative research methods. Throughout the research process, the two forms of data must be integrated (Tashakkori/Creswell 2007, p. 3). This can be done in several ways: by merging the data, by building on one data point when collecting the other or by embedding qualitative and quantitative data within an overarching framework in the analysis process. When writing up a coherent research plan, it is important to decide right at the outset which data-integration method will be used. Research related to Japanese society has also seen several applications of mixed methods designs, as for example in ‘Mixed-method analysis of Japanese depression’ (Arnault/Fetters 2011), an investigation of the ‘Fukushima effect’ in Germany (Hartwig/Tkach-Kawasaki 2019), or in research on friendships and intimacy (see this chapter, Dales, Ch. 10.2; Kobayashi/Kawabata 2019), labour markets and employment regulations (see this chapter, Imai, Ch. 10.3), or political parties and governmental industrial policies (see this chapter, Pekkanen/Pekkanen, Ch. 10.1). Before guiding you through the steps necessary for a successful mixed methods project, we will briefly outline the respective rationales of qualitative and quantitative research methods and then introduce three core models of mixed methods research.

3. Overcoming the qualitative–quantitative divide: A pragmatic approach The large divide between qualitative and quantitative research approaches has its roots in the underlying paradigms on which research questions and designs are based. Qualitative research is in most cases grounded in a constructivist worldview (Creswell/Creswell 2018, p. 7), which means that great emphasis is placed on individuals’ subjective comprehension of their world. Here, theory is not necessarily the starting point, but often rather a result of the research process (theory building), generated based on individual constructions of realities that the researcher has collected from the participants. Quantitative research, on the other hand, usually conforms to a post-positivist worldview, assuming that societies operate based on general laws (often also called theories), which need to be empirically verified. To this end, the quantitative researcher collects data on the ‘objective reality that exists “out there” in the world’ (Creswell/ Creswell 2018, p. 7), which either supports or rejects the initially assumed theory (see Goodman, Ch. 1). The stark distinction that exists between the comprehension and application of a theory is also found in the case of research methods: qualitative researchers tend to collect in-depth information on a limited number of cases (i.e. in ethnography), while quantitative researchers gather a limited amount of information on a large number of cases (i.e. in a population survey, also

266

Chapter 10 How to combine methods Ragin/Amoroso 2011, p. 28). In this way, qualitative research emphasises the commonalities found among a small—but intensely studied—number of cases, or looks for patterns or typologies to differentiate between when comparing a moderate number of cases (ibid., p. 36). Quantitative researchers, however, usually examine differences across large numbers of cases. Most often the intention is to explain these differences as a consequence of the mutual impact of certain impactor variables and to extrapolate the results from a sample to a population. Mixed methods research—often called a ‘pragmatic approach’—combines these two research traditions to procure the largest possible amount of information to answer a specific research question. Considering the still strong animosities between the two fields, this is a challenging but rewarding task, as we will lay out in more detail below. As this handbook does not focus on quantitative research, we insert several text boxes below, labelled Basics in quantitative methods, in which we focus on a selection of core questions often asked by newcomers and offer basic definitions. For more detailed information, we recommend Charles Ragin and Lisa Amoroso (2011, pp. 163–187) or John Creswell and J. David Creswell (2018, pp. 147–177) as two accessible introductions to this vast methodological field.

4. Combining qualitative and quantitative data: Three core designs In mixed methods research various classification systems and typologies exist (Hesse-Biber 2010, pp. 68–72; Tashakkori/Teddlie 2003b, pp. 25–33). According to Creswell and Creswell (2018, p. 217), three basic models can be identified that differ considerably in 1. terms of emphasis on quantitative and qualitative data sets; 2. the timing of the overall research process (simultaneous/ sequential collection, combination, analysis, etc.) and 3. the relationship between theory and empiricism (developing or testing theory). Here, we introduce these three core designs—convergent, explanatory sequential and exploratory mixed methods design—with their basic features, respective processes of data collection, data analysis, integration and interpretation. Please note that in practice the designs might not be as neatly separated as presented here. The convergent mixed methods design is the most common and least complex design (for examples, Hatta et al. 2018; see also this chapter, Imai, Ch. 10.3). In a single-phase approach, the collection as well as the primary analysis of both quantitative and qualitative data takes place separately, followed by an analytical comparison aimed at identifying the convergence or divergence of findings. The key assumption underlying this approach is that both types of data —meaningful statistical results in the case of the former and extensive in-depth data in the case of the latter—provide different types of information which complement each other. Before the collection of data, the researcher needs to consider four points: 1. (ideally the same or parallel) variables (see Basics in quantitative methods I), constructs, and concepts in both forms of data; 2. sample sizes (usually less so for qualitative than quantitative data) (see Basics in quantitative methods II); 3. the handling of potential inequality in sample size and 4. whether or not to include participants in the qualitative survey in the quantitative sample. Data collection is followed by data analysis and interpretation in three phases. After coding (see Meagher, Ch. 12) the qualitative data (i.e. phase I), the analysis of the quantitative data takes place (i.e. phase II). Phase III consists of integrating the two data sets by merging the results in the form

267

Carola Hommerich and Nora Kottmann of either a side-by-side comparison—through data transformation of the qualitative data into quantitative data—or as a joint display in a table or a graph. Finally, the discussion and interpretation of the comparison aim to understand whether there is convergence or divergence between the two sources of information (Creswell/Creswell 2018, pp. 217–221). The explanatory sequential mixed methods design focuses on quantitative data and involves a distinctively separated two-phase data collection process. Here, the main idea is that qualitative data collected after quantitative data helps the researcher to understand quantitative results in greater detail and explains possible contradictions or surprising survey responses. The data-collection process occurs in two distinct phases: after quantitative sampling in the first phase, the purposeful sampling of qualitative data builds on the quantitative results (see Basics in quantitative methods II). Challenges here include the identification of results to follow up on and the selection of sources of qualitative data, for example, whether informants for the qualitative part should be recruited from the quantitative sample or not. Data analysis takes place separately because quantitative results are used to plan the qualitative data collection. Here, the integration of both kinds of data occurs in the form of basing one approach on the results of the other. The qualitative data are then used to further analyse the quantitative part (Creswell/Creswell 2018, pp. 221–223). In contrast, in the three-phase design of the exploratory mixed methods model the researcher first collects and analyses qualitative data (the ‘exploring phase’), then, based on this, develops hypotheses to be tested, and finally creates a measurement instrument to empirically test these hypotheses: the social anthropologist Laura Dales (see this chapter, Ch. 10.2), for example, initially only planned and conducted a qualitative study, but was then encouraged to add a quantitative perspective in order to contextualise her qualitative findings. Similarly to the previous model, data is collected during the first and third phases, but interrupted by integration of the data in phase II to inform the design of the—in this case—quantitative survey. Challenges here resemble those faced in the other designs and include the question of whom to choose for the qualitative study and whether to include these participants in the quantitative sample. After collecting the quantitative data, both data sets are analysed separately. This means that the final interpretation of the data can include a report on the qualitative findings and the hypotheses drawn from them, the development of the measurement instrument and the results of the quantitative test. For this strategy, there is no inherent need to compare the findings, as the basic intent is to ‘determine if the qualitative themes in the first phase can be generalised to a larger sample’ (Creswell/Creswell 2018, pp. 224–226). In addition to the above-mentioned core designs, variations and (individual) adaptations of these core designs exist, as do several more complex designs that involve more steps and/or procedures (also Creswell et al. 2003; Creswell/Creswell 2018, pp. 226–236). Without going into details here, we recommend the researcher to familiarise themselves with the core designs and adapt them wherever appropriate or necessary (see this chapter, Pekkanen/Pekkanen, Ch. 10.1). In any case, it is necessary for the researcher to explain the choice of her design, the underlying theoretical and practical assumptions and the implications regarding the subsequent data collection, data analysis and interpretation. Especially when writing a proposal, we recommend including a chart or table that outlines the basic features of the chosen design (for useful abbreviations, Teddlie/Tashakkori 2009, p. 27).

268

Designs based on Creswell/Creswell 2018, p. 218

Figure 10.1: Three core mixed methods designs

Chapter 10 How to combine methods

269

Carola Hommerich and Nora Kottmann

5. Practical advice In the following chapter, we offer concrete advice on how to get started, how to collect your data, how to analyse your data, how to present your findings and how to overcome common hurdles.

5.1 Getting started with your research Your starting point should always be a thorough review of the available literature on your chosen topic (see Zachmann, Ch. 4). Only then can you decide whether or not you wish to employ a mixed methods research design—this choice should be grounded in your research question. When the application of either method alone is unlikely to help you thoroughly understand your research problem, a mixed methods approach can be helpful. For example, you might want to discover how widely distributed a certain phenomenon is among a certain population (i.e. how many people feel they belong to the Japanese ‘middle class’), but at the same time you want to understand what this phenomenon means for or how it is understood by individuals in more detail (i.e. which criteria individuals associate with being ‘middle class’). A mixed methods approach permits you to perform both and, thus, allows you to obtain results that can be extrapolated to a broader population (e.g. more than 75% of the Japanese population think they belong to the ‘middle class’), while acquiring detailed information on how the respondents interpret this phenomenon (i.e. the criteria for ‘middle class’ membership vary widely among individuals). In this way, you will probably obtain better results than when applying only one approach. If you have concluded that you can best answer your research question by applying a mix of the two methods, you have to specify the questions to be answered in the qualitative or quantitative part of your study. Moreover, you need to decide the order in which the two parts should be executed, as described in the three core models outlined above. At this point, you should contemplate the role that theory will play in each part of your study. For the qualitative part, a theory is likely to be more akin to an underlying guideline, an ‘orienting lens’ or perspective (Creswell/Creswell 2018, p. 62) that influences the selection of your topic and your questions (see Okano, Ch. 3). Most probably, especially in grounded theory approaches (see Meagher, Ch. 12), developing a theory inductively from the collected data will be the overall goal rather than the starting point. In contrast, in quantitative research, theories are applied more rigidly, providing a clear structure as to what is to be investigated (see this chapter, Imai, Ch. 10.3). Theories are understood as underlying rules that specify the strength and direction of relationships among different variables (see Basics in quantitative methods I), thereby explaining social phenomena. Explicit hypotheses are deduced from these theoretical assumptions, which are to be tested with empirical data. When presenting specific hypotheses to be tested in the quantitative part of your project, you should, therefore, include the theoretical rationale of how and why you expect a certain variable to impact another (Creswell/Creswell 2018, p. 53; for a practical example, Hommerich/Tiefenbach 2018, p. 1098). The application of a mixed methods approach enables you to use theory deductively as well as inductively, to empirically test its assumptions, and at the same time, further develop it. We 270

Chapter 10 How to combine methods recommend using an overarching theoretical framework for both parts of your study (for a practical example, Inglehart’s theory of value change, Hommerich 2009), and to point out how each part utilises the theoretical perspective in its own way. When doing so, lay out your methodological strategy: Which relationships will be tested with what kind of data? What are the specific contributions of the quantitative and qualitative parts?

Basics in quantitative methods I: Variables and quantitative data A variable, plainly described, is something that can take on different values. In the Social Sciences, the term ‘variable’ refers to attributes or characteristics that can be measured, and which vary among individuals or institutions (Creswell/Creswell 2018, p. 50). Quantitative data is any type of data that can be counted or expressed numerically. In the Social Sciences, this means that variables such as social characteristics (i.e. age, gender, education, occupation, income, etc.), forms of behaviour (i.e. time spent on housework per day, participation in volunteer activities, etc.) or social attitudes (i.e. view of governmental redistribution, tolerance for minorities, etc.) are transformed into numbers when they are measured. Quantitative data can have different measurement levels: nominal, where a certain value is assigned to a certain characteristic, but there is no order to the values (i.e. 1 = married, 2 = not married); ordinal, where an order is assigned to different features, but the distance between the numbers has no meaning (i.e. 1 = very much agree, 2 = somewhat agree, 3 = neither agree nor disagree, 4 = somewhat disagree, 5 = very much disagree); interval, where features are equally spaced along a scale, but no zero exists (i.e. Celsius temperature); or ratio, where the differences between different categories are equally spaced and an absolute zero exists (i.e. height, weight or annual household income). The most common method of acquiring quantitative data in the Social Sciences is via survey research.4 Surveys can be cross-sectional, carried out at one point in time only, or longitudinal, which means that a survey is repeated several times (see Okano, Ch. 3). Quantitative surveys can help find answers to descriptive questions (i.e. what percentage of the population works overtime?), questions about the relationship among different variables (i.e. is there a positive relationship/correlation between family background and educational achievement?), or—in the case of a longitudinal study—to test causal relationships5 (i.e. does an increase in female labour force participation cause decreased fertility rates?). In the latter case, the variable that causes the impact is the independent variable (e.g. x = female labour force), while the variable that is impacted is called the dependent variable (y = fertility rate). When you are investigating causal claims (i.e. the increase in female labour force participation has decreased the fertility rate), it is important not to overlook an unmeasured phenomenon that affects both the dependent and the independent variables, being the main cause of the phenomenon investigated. This third variable is called the confounding variable (e.g. z = family-friendly/unfriendly policies).

4 Experimental designs, where certain variables are manipulated to test how this impacts outcome

are more common in psychological studies. Despite being less common in the Social Sciences, they do exist, i.e. to test different survey methods against each other. 5 Many researchers also make assumptions about causality when working with cross-sectional data. In these cases, their assumptions are grounded in theory, not in the empirical data.

271

Carola Hommerich and Nora Kottmann In the next step, you need to consider whether you possess the resources required to accomplish your project (see checklist below). This includes financial resources, human resources, and time. It is important to consider if you have the necessary skills to execute both the quantitative and the qualitative parts of your intended study. If not, is there anyone who you can cooperate with? How much money will you need to complete your study in the intended format? If the expenditure exceeds the budget, are there other sources of funding for which you could apply? What is the expected time frame of your study? This might depend on a funding scheme and/or on your career stage. Is it realistic to finish the survey within this time frame? Last but not least, as with any other research proposal, you need to consider ethical issues (see Reiher/Wagner, Ch. 16). Are there any possible conflicts of interest for you to consider? Is it possible that by studying a certain population and writing about them, you might harm your respondents? Are there questions you intend to ask that might cause respondents to experience psychological stress? How will you ensure that the data you are collecting (both in a quantitative and qualitative format) is stored safely and cannot be accessed by third parties? It has become common practice at most universities to have research proposals that involve human participants reviewed by the universities’ Research Ethics Committee.6 However, this is only the first of many ethical considerations you will have to deal with throughout your project. For example, you will need to be aware of possible problems regarding the anonymity and privacy of your respondents and the safety of the collected data. Checklist for conducting mixed methods research Before starting to collect data, ensure you have considered the following questions: • • • • • • • •

What is the overarching research question? What is the rationale behind choosing a mixed methods approach? What specific research questions will be investigated in the quantitative and qualitative parts? How is your theory applied in the two methodological parts? What information surplus do you expect from the combination of quantitative and qualitative methods? What kind of data do you intend to use? Do you have the resources (skills, budget, time, etc.) to execute this project? Are there ethical issues to be considered?

If you find problems with any of these questions, reconsider and, if necessary, adjust your research design. Once you have started collecting data, it will be too late for bigger changes. Therefore, it is important not to rush into anything. Instead of being overly ambitious, it is advisable to be realistic and pragmatic.

6 Guidelines about this differ widely across universities, also depending on the career stage of the researcher. Therefore, check whether you are expected to submit your proposal for review or not.

272

Chapter 10 How to combine methods

5.2 How to collect data The research design that you choose (convergent, explanatory sequential or exploratory sequential) determines the time and sequence of data collection. Again, the manner of data collection will differ between the qualitative and the quantitative parts of your study. As data collection methods for qualitative research are described in detail in chapters 5 to 8 of this book, we focus mainly on core questions to think about for the quantitative part of your study. Carrying out a quantitative survey is time-intensive and expensive. Therefore, time and funds at your disposal determine if you will be able to conduct your own survey. Free online survey tools, which are widely available nowadays, may tempt you to quickly write down some questions, upload them and circulate your online survey via social networking sites (SNS), but we strongly advise against this. When considering a quantitative investigation of your research question, you should think carefully about the following points: 1. 2. 3. 4.

Who is my target population? How can I reach my target population? Which questions do I need to pose in order to get answers to my research question? Have these or similar questions been asked before? If yes: 5. When and where was this survey conducted? 6. Does this include a sample of my target population? 7. Is this data available for secondary analysis?

If you find a survey that is representative of your target population (see Basics of quantitative methods II), includes most of your questions and was carried out not too long ago in order to give relevant results for the phenomenon you are looking at, we recommend that you try to access this data for secondary analysis (see text box below; see also Aizawa/Watanabe, Ch. 9.3). This will not only save you time and money but will also be highly likely to yield more reliable results than you would obtain from a convenience sample recruited through friends or SNS. The Social Science Data Archive (SSDA) at the University of Tokyo is the richest repository of existing surveys on Japan, open to use for secondary analysis. Surveys can be accessed online, applications for data usage can be submitted online and data can also be downloaded from outside Japan. Microdata from the Japanese government can be accessed through so-called ‘on sites’—you need to physically visit the office to register—or online through the miripo-portal of the Statistics Bureau of Japan (SBJ). A more detailed overview, which includes additional data sources for Japan, can be found on a website directed at students by the Japanese Association for Social Research (JASR). Cross-national surveys, such as the World Values Survey (WVS) or the International Social Survey Program (ISSP), also include data on Japan, which can be compared with data on other countries or used on its own. Data is available for download through their websites. The Organisation for Economic Co-operation and Development (OECD) also provides a rich pool of data on Japan in a cross-national context. While data is not easily available for secondary analysis, the OECD provides population statistics you will need to outline your research population or the societal context of your study.

273

Carola Hommerich and Nora Kottmann Data archives and resources (selection) • •

Social Science Data Archive, University of Tokyo: https://csrda.iss.u-tokyo.ac.jp Statistics Bureau of Japan (SBJ): www.e-stat.go.jp/microdata/

The Japanese Association for Social Research (JASR) provides a detailed list of data archives and other data sources (in Japanese only): www.jasr.or.jp/students/links.html Cross-national surveys: • •

World Values Survey (WVS): www.worldvaluessurvey.org/wvs.jsp International Social Survey Program (ISSP): www.issp.org/menu-top/home/

Nevertheless, there will be occasions when the variables you need will not be included in previous surveys—or at least not in the combination that you need for your analysis. Or your target population will be so specific and/or difficult to access that it will be challenging to find existing survey data that include this population. In these cases, you should consider conducting an original survey. Your choice of a specific method of data collection (i.e. face-to-face interviews, computer-assisted personal interviews, drop-off surveys, postal surveys, online surveys; De Leeuw/Berzelak 2016) is highly likely to be influenced by the size of your budget. Nevertheless, you should carefully consider whether a certain method will give you access to the population you want to survey. While an online survey is usually the cheapest option, it is also a risky endeavour in terms of representativeness and reliability (Bryman 2016, p. 191). If your target population is likely to use the Internet and large parts of the population can be accessed, for example, through a mailing list or an online platform, using an online survey tool and distributing the survey yourself can be meaningful. Still, this procedure will not deliver data that will be representative of your population, as you have no information on the original target population, or because of the bias caused by the route of questionnaire distribution. If the population is less specific, but likely to use the Internet, a more expensive alternative would be to use registered monitors of a research company. Research companies in Japan (selection) There are numerous research companies in Japan, and the market is expanding. Here, we list some companies often used by social researchers. We recommend comparing several research companies in terms of weighing up the procedures and prices they offer, before making a final decision. • • • • •

Shin Jōhō Center: www.sjc.or.jp/english/ Nippon Research Center: www.nrc.co.jp/english/index.html Chūō Chōsa: www.crs.or.jp/english/ RJC Research: www.rjc.co.jp/ Cross Marketing: www.cm-group.co.jp/

Most companies offer quota sampling procedures, which allow control of the distribution of certain attributes (e.g. gender, age, education, income, etc.) in the sample. However, if your target population is unlikely to use the Internet—for example, if you want information on elderly people in rural Japan—then this is not a recommendable approach. If your budget allows, you should opt for a random sample of your population, as this will give you the best

274

Chapter 10 How to combine methods possible representation of your target population (see Basics in quantitative methods II; for an example, Hommerich/Tiefenbach 2018, p. 1100). Basics in quantitative methods II: Sampling in quantitative research ‘How big should a sample be to be representative?’, is a common question asked by newcomers in the field of quantitative research methods. Several misconceptions exist regarding the relationship between the sample size and representativeness of a study. We introduce the three most common misconceptions, based on Fowler (2014, pp. 37– 41), who we highly recommend for more details. The first common mistake is 1. to make assumptions about the ‘necessary’ sample size based on the ‘usual’ sample size of other studies (i.e. samples of around 1,500 respondents in national surveys, or the like). Whether or not a sample is adequate depends on the population to be researched and the questions to be answered. Other inappropriate but common approaches are 2. that a certain fraction of the target population should be included in the sample, and 3. that a researcher should decide the margin of error she can tolerate for specific estimates. As most quantitative studies make use of not just one but numerous estimates, it is rather unusual for the researcher to be able to specify an acceptable margin of error for each estimate in advance. Instead, the sample size should be decided depending on the analysis plan. Which subgroups within your general population are you interested in (i.e. comparing different levels of education, for males and females separately, etc.)? To be able to perform the intended statistical analyses, you have to make sure you achieve the minimum sample size that can be tolerated for the smallest sub-group included in your analysis plan (for more details, Fowler 2014, p. 39). When drawing a sample, you should first set a sample frame that defines the target population (e.g. residents of Japan between 20 and 64 years of age). Next, you need to select a sample design, which is the actual procedure to sample the individuals. In order to make inferences about the population from the sample, a probability sample is required. This kind of sampling is only possible when you have a list of all members of your target population. In Japan, the population registry is often used for this purpose. The most basic form of a probability sample is a simple random sample, in which each individual in the population has an equal chance of being selected. A variation of this is the systematic sampling procedure in which a random starting point is set and respondents are selected according to a certain strategy (i.e. select every Xth number from the list; ring the doorbell of every Xth house and ask to interview the person with the most recent birthday, etc.). To ensure that your sample is representative of certain characteristics of the target population, stratified random sampling procedures can be applied. For example, when you have a large target population and want to ensure regional representativeness, you can use a multistage (or stratified) sampling procedure that identifies clusters (e.g. by region) within which individuals are sampled randomly (Fowler 2014, p. 14)

275

Carola Hommerich and Nora Kottmann Non-probability sampling procedures such as convenience sampling are less desirable because respondents are chosen based on their accessibility, which means inclusion in the sample is biased. A similar procedure is snowball sampling, where respondents introduce the researcher to additional respondents, or survey questionnaires are circulated via SNS among a group of friends. A quota sample—often claimed to come close to random sampling procedures in terms of predictive power—is another form of nonprobability sampling. The goal of quota sampling is to produce a sample that closely reflects the original population in terms of the distribution of different characteristics (i.e. gender, age, education, income, region, etc.). For a more detailed introduction to sampling procedures, see Bryman (2016, p. 170). Be aware that regardless of the sampling procedure, a sample can only be representative of the sample frame, i.e. the people who had an actual chance of being included in the sample. This means that if you work with a random sample of registered residents of Japan who were between 20 and 64 years of age at the time of the survey, your results will only be representative of that population. You cannot draw any conclusions for residents of Japan who are below 20 or above 64 years of age. Before you conduct a quantitative survey, it is most important to take ample time to think about what questions need to be included in the questionnaire, in order to find answers to your research question. These questions need to be specific enough for you to be able to make assumptions about what respondents imagined when answering them. Where standard questions about issues you want to research exist which have been used and checked for reliability in other surveys, we recommend using them, rather than inventing new questions for which you have no information on whether respondents will understand them in the way you intend. When using questions which have been used in another language before, refer to Dorothée Behr and Kuniaki Shishido (2016) for more information on what points to consider when translating research questions (see also Basics in quantitative methods III). Keep your hypotheses in mind when writing up a questionnaire. Will you receive all the information you need by posing these questions? This should include a clear outline of the analytical strategy and consideration of the level of measurement variables needed for the type of statistical procedures you want to execute (Creswell/Creswell 2018, p. 155; p. 159 for some introductory examples).

276

Chapter 10 How to combine methods Basics in quantitative methods III: Translation issues When you are conducting research in Japan (or any foreign context) or in a comparative setting, it is necessary to translate questionnaires, interview questions, forms of consent, or—at a later stage of the research—interview transcripts or questionnaire responses.7 As the translation of a questionnaire and/or an interview guide is an essential prerequisite for valid and reliable findings—as well as their comparability—it is crucial to put sufficient effort into it, including time and money. A meaningful translation takes into account not only language issues, but also the cultural context, ways of thinking and communication styles (for concrete examples in the East Asian context, Behr/Shishido 2016, pp. 278–284). This is not a one-person job, but a team effort. Different experts should contribute to and review the translation at various stages; for example, in the form of an ‘expert review’ (which should be supplemented by a questionnaire pre-test; Willis 2016). Experts can be professional translators/linguists, methodological experts and/or topical experts (Behr/Shishido 2016, pp. 270–272). When you are translating a source questionnaire into Japanese or designing an original questionnaire in Japanese as a non-native researcher, the help of Japanese natives is indispensable (Huber 2018, pp. 23, 25, 31; Schrauf 2016, pp. 94–96). When you are planning to carry out your own survey, it is just as important to review existing surveys that are thematically related and to stick to the established terminology. This requires thorough knowledge of the relevant surveys in your field of interest. In order to create such a questionnaire, you either need to be trained in quantitative data analysis (see text box above) or need a research partner with the necessary training. Once a survey has been conducted, it is too late to make adjustments. Therefore, carry out a pilot survey with a small number of respondents similar to those who will eventually take the survey (Bryman 2016, p. 260). Pre-testing in this way will enable you to simplify the complicated questions before carrying out the actual survey.

5.3 How to analyse your data In qualitative research, data analysis tends to happen concurrently with data collection because both are intertwined and influence each other in a processual way (for an overview of qualitative data analysis and individual methods, see Ch. 11–14). The analysis of quantitative data, however, starts after data collection. You should always begin with becoming familiar with your data set. This is true for the data you collected yourself as well as for the data you have acquired for secondary analysis. While we emphasise the importance of the sampling strategy, it must be remembered that sampling design and sample size are not the only possible sources of error. It is, for example, just as important to check for patterns of non-response or response bias after data has been collected. This is needed to identify your sample population and determine whether it is representative of your target population. After this process of data cleaning, you will start the actual analysis. Before moving on to more complex multivariate analyses, remember to report the descriptive statistics of your sample. Writing up the results of 7 In this context, a distinction should be made between the translation of an already existing questionnaire, which has already been tested in an empirical setting, and the creation of a new questionnaire in a foreign language.

277

Carola Hommerich and Nora Kottmann your quantitative analysis should begin with a brief outline of which statistical procedures you applied and for what purpose. Always indicate which inferential statistical tests are used to test your hypotheses. Basics in quantitative methods IV: Analysing quantitative data For researchers who did not undergo training in quantitative research methods as part of their university education, several institutions offer intensive courses in quantitative methods and data analysis. The GESIS Spring and Summer Seminars, for example, have a long tradition of offering courses to newcomers as well as advanced learners. Similarly, the Inter-university Consortium for Political and Social Research (ICPSR) organises an annual summer school with a wide selection of courses. Data is analysed with software designed for this purpose. Some of the commonly used software packages such as IBM SPSS Statistics or Stata are relatively expensive, unless the licence is shared, as is the case at some universities. A free open-source alternative is the programming language R, which is increasingly also taught at universities. GESIS Spring and Summer Seminars: https://training.gesis.org/?site=pOverview&cat=all ICPSR Annual Summer School: www.icpsr.umich.edu/icpsrweb/content/sumprog/about.html An important step in the mixed methods approach is the integration of the results of the two parts of the project. This integration happens according to the order in which the two parts were executed, as described above in the outline of the three core approaches. In sequential designs, the second part of the study builds on the first part; this is one way of connecting the two analytical parts. When you are reporting the final results, however, the contributions of each part need to be emphasised. The extent of integration depends on the closeness of the two data sets. In an explanatory sequential research design in which in-depth interviews (phase II) were conducted with a selection of respondents from a larger quantitative survey (phase I), the results of the two parts can be discussed in close reference to each other. In other designs, the data sets might be more difficult to connect, but you might be able to draw conclusions at a theoretical level based on both parts of the project. Checklist The final results should emphasise the specific contribution of each part of your study. A crucial point, however, is the integration of both parts. How did each part inform the other? What is the information surplus achieved by applying a mixed methods approach?

5.4 How to present and report your findings While writing will be a part of the entire research process—from formulating an initial proposal (Creswell/Creswell 2018, pp. 75–80; see also above) to writing an abstract, to constantly keeping track of the research process (see McMorran, Ch. 15)—presenting and reporting the

278

Chapter 10 How to combine methods final findings will most likely occur at the end of the research process. Obviously, general conventions apply when presenting and reporting the findings, depending on the audience and the dissemination form (see Farrer/Liu-Farrer, Ch. 17). However, there are some specific points regarding mixed methods research. Generally speaking, the crucial point is to outline the reasons for choosing a mixed methods research design and to ‘address the integration challenge in order to reap the rewards of the integration equation of 1 + 1 = 3’ (Fetters/Freshwater 2015, p. 204); in short: ‘to persuade readers of its merits’ (Sandelowski 2009, p. 321). There are several ways of doing it, depending on the intended audience. Therefore, we recommend a careful selection of the journal and its audience: Are readers likely to be familiar with mixed methods research? If so, are they specifically interested in methodological questions? If not, are they used to either quantitative or qualitative research or are they sceptical towards quantitative or qualitative narratives (which do differ significantly)? Especially with regard to an audience not familiar with mixed methods research, we recommend providing a rationale for choosing a mixed methods design, to sufficiently outline underlying assumptions and to avoid technical terms and jargon. Instead, it might be best to use succinct and clear language, to adapt to the scientific conventions of your specific audience, and to present your data in such a way that your audience can easily relate to it (more narratively in the case of a qualitatively trained audience, more formally in the case of a quantitatively trained audience; Sandelowski 2010, pp. 329–331).8 The structure of an article (or a book), in turn, is independent of the readership and should follow a common structure of academic publications (Fetters/Freshwater 2015, pp. 205–207; see Farrer/Liu-Farrer, Ch. 17). However, the presentation, analysis and discussion of the empirical findings depend on and follow the research design. After outlining the design, that is, convergent, explanatory sequential or exploratory mixed methods design (see above), you should describe the methods—and their specifics with regard to sampling, collection and analysis—in the same order in which they were executed (Fetters/Freshwater 2015, p. 208). That is to say, depending on the design, the presentation of the data will either occur in different sections for qualitative and quantitative data or through integration (known as ‘weaving’). Alternative approaches are data transformation and the use of a joint display (which we recommend using in any case). By doing so, you can create a coherent narrative your audience can relate to (Sandelowski 2010). When finally presenting your results, you again need to stick to the order of the design and present the results consistently with the research process. As mixed methods designs can be extremely complex, it is helpful to visualise the research process and your findings using diagrams, graphs, lists, charts or tables. This not only helps you as a researcher/author to organise and present your research/writing, but also helps your audience to follow your argument (Sandelowski 2010, pp. 335–338). In this context, the use of common abbreviations can also be helpful. It might also be necessary to add an appendix that contains additional information on the empirical data, a survey questionnaire, interview transcripts, statistics or selected calculations (e.g. Hommerich 2009, pp. 241–271). Finally, at the level of content, we recommend openly and transparently discussing setbacks and problems encountered, any unexpected (supposedly ‘wrong’ or ‘bad’) findings and the limitations of the 8 The American Psychological Association, for example, recommends in its Mixed Methods Design Reporting Standards (JARS-Mixed) to ‘refrain from using words that are either qualitative (e.g. ‘explore’, ‘understand’) or quantitative (e.g. ‘determinants’, ‘correlates’)’ (American Psychological Association 2020, p. 1).

279

Carola Hommerich and Nora Kottmann findings—as Dales shares in her essay (see this chapter, Ch. 10.2). This kind of transparency augments the validity and reliability of your findings and promotes further research. Checklist While writing starts at the very beginning of the research process, the presentation and publication of the final results take place at the end of the process. When publishing your findings, follow the common structure of an academic publication, but 1. specifically consider your audience and ‘speak’ to it (Are they familiar with mixed methods research? Are they used to quantitative/qualitative narratives/logics? What is the focus of the journal?); 2. follow the overall structure and logic of your design and create a coherent narrative to which your audience can relate; 3. visualise your research by way of diagrams, charts, etc. (findings); 4. use generally accepted abbreviations and add an appendix with necessary additional information; and 5. openly discuss setbacks and limitations.

5.5 Stumbling blocks and how to avoid them Due to the characteristics of mixed methods research, there are specific challenges at different levels that researchers should be aware of. Here, we summarise some common obstacles, although this list is not exhaustive. The focus, however, should always be on overcoming these obstacles in order to enjoy the advantages of this research approach. Complexity, time and financial resources: As outlined above, mixed methods research designs can be extremely complex due to the collection, analysis and combination of two completely different data sets. Visualising the different steps in the research design and a detailed (realistic) time schedule can help manage this complexity and save you from getting lost in the details. Since research with mixed methods is extremely time-consuming, it is essential to be aware of possible time constraints regardless of whether they are professional (teaching responsibilities, administrative work, etc.) or personal (family, friends, care work, hobbies, etc.) (see McMorran, Ch. 15). Financial resources, e.g. for conducting an original survey or securing professional assistance for the analysis of quantitative data or the transcription of qualitative interviews, also need to be considered: you should consider respective funding options or collaborating with experts. Translation issues: Serious challenges include ‘translation issues’, which might appear ‘within as well as between qualitative and quantitative phases’ (Schrauf 2016, p. 98). While we discuss the challenges of language translation above (see Basics of quantitative methods III), a different type of translation is likely to become necessary when moving back and forth between the two methodological fields of qualitative and quantitative research. We recommend being aware of the respective conventions and traditions of each scientific community and adapting to them. Skills, infrastructure and other resources: Finally, on a pragmatic level, we emphasise the need for you to honestly and realistically evaluate your own (and possibly your research team’s) knowledge and abilities with regard to mixed methods research in a Japanese context: Are you (and/or your collaborators) familiar with both qualitative and quantitative data collection and analysis, and will you be you able to perform it in Japanese? Dales (see this chapter, Ch. 10.2), 280

Chapter 10 How to combine methods for example, was first ‘ambivalent’ about ‘adding a quantitative lens’ to her primarily qualitative work and then faced serious problems when trying to analyse her data since she did not undergo the necessary training.9 With regard to infrastructure, it is important to first identify everything that you (and/or your team) will need to collect, combine and analyse the data sets. Next, you need to realistically evaluate whether you have all the resources you need in terms of manpower, software and hardware: How can you offset the abilities, knowledge and infrastructure you lack? Where can you find help and—most importantly—with whom can (and will) you collaborate at which stage of the research? Overall, we recommend discussing the research design and all questions and ideas with experienced peers and/or colleagues; using professional services, if necessary and affordable (e.g. for data collection, transcription, quality control, etc.) and undergoing training in necessary research methods. Checklist While mixing methods might sound simple, there are some serious challenges researchers should be aware of. Do not underestimate 1. the complexity of the research process; 2. time constraints; 3. financial burden; 4. translation issues (between languages and methodologies); and 5. conventions of specific scientific communities. Realistically evaluate your own abilities and methodological knowledge, the available infrastructure, and the need for/possibility of collaboration. Never hesitate to reach out for help and never lose your enthusiasm for your research question!

6. Summary: Prospects and challenges As discussed above, mixed methods research is a complex field of growing importance. Various developments in terms of sampling, data collection and analytical procedures as well as philosophical foundations have evolved and developed, and specific funding opportunities have been established. ‘These developments signal optimism’ and—according to some proponents of the mixed methods approach—‘the strengths and power of mixed methods research […] has only just begun to emerge’ (Fetters 2016, p. 8). This seems to be especially true in the context of increasing transnationalisation and digitalisation, which expand the possibilities of (and necessity for) data collection and interdisciplinary collaboration. However, despite this enthusiasm, advocates of quantitative and qualitative research remain sceptical towards this research approach. Researchers formally trained and experienced in one of the methodological traditions often find it difficult to learn and apply the other approach. In addition, when collaborating, researchers might find it challenging to integrate the viewpoints of qualitative and quantitative scholars. Therefore, when deciding on using a mixed methods design, you need to

9 These might be common hurdles among (young) researchers in Japanese Studies around the globe, as training in quantitative methodology is generally not a part of curricula in Japanese Studies. Robert and Sadia Pekkanen (see this chapter, Ch. 10.1), on the other hand, work in the realm of Political Science, and observe ‘that training in qualitative methods […] has been demoted in the Social Sciences’ and thus call for more training at universities. Against the background of these inequalities, Creswell et al. (2010, pp. 619–637) provide some practical tips on how to teach mixed methods research.

281

Carola Hommerich and Nora Kottmann be aware of your abilities and experience(s) and your discipline’s conventions. It might be useful to be well-prepared for criticism from both sides. In summary, as outlined above, mixed methods research is complex, time-consuming, moneyintensive and challenging on multiple levels—especially for junior researchers with little experience, funding and/or institutional backing. Nevertheless, mixed methods research provides a great opportunity to achieve a deeper and more comprehensive understanding of a research objective, fostering (interdisciplinary) collaborations and broadening the researcher’s horizons through the application of (in most cases) new or unfamiliar research approaches (qualitative or quantitative, respectively).

282

10.1 Reflections on multi-method research Robert J. Pekkanen and Saadia M. Pekkanen

The common wisdom today is that we need to combine quantitative and qualitative methods in order to shed more light on causal research questions in the Social Sciences. This thinking is increasingly pervasive in Political Science and has made substantial inroads into Area Studies, including the field of Japan. In this essay, we draw on our scholarly and pedagogical experiences in approaching the three questions set out for us in this volume. We discuss how multimethod research (MMR) has influenced our own work, what we see as some of the problems it poses and what we see as takeaways for others. First, what do we understand by multi-method research? Typically, in Political Science, with causal inference in mind, this means combining statistical or formal models (the quantitative part, with a big number of observations or Large-N) that give an aggregate overview of findings with other methods such as historically attuned case studies (the qualitative part, with a smaller number of observations or Small-N) that draw attention to causal processes, mechanisms and sequences (Gerring 2012, pp. 362–366). MMR is widely thought of as disciplinary best practice for the purposes of correcting the weaknesses of any one method and providing external validity to the findings (Ahmed/Sil 2012). Selecting MMR represents a methodological choice. We have also deployed MMR in our own Japan-focused research, finding the ‘triangulation’ among different methods and data sources a useful way of answering the questions we set ourselves. In past projects, one of our authors has chosen qualitative methods (Pekkanen 2006), quantitative methods (Pekkanen/Nyblade/ Krauss 2006) and MMR (Krauss/Pekkanen 2011). The other has chosen both a strict MMR approach (Pekkanen 2003) and a purely qualitative one in a collaborative work (Pekkanen/ Kallender-Umezu 2010). Below we discuss our methodological choices in three books: two in which MMR was pursued, and one in which qualitative methods were chosen and not MMR. The research for The rise and fall of Japan’s LDP involved a study of what happened to central institutions of the LDP after electoral reform. These were: personal support organisations for politicians (kōenkai), factions and the party’s policymaking body, the Policy Affairs Research Council (PARC). The authors used MMR to investigate all of these, but LDP party factions provide perhaps the readiest example. The central research question of the book was how electoral reform affected institutions, and the argument revolved around historical institutionalist concepts of sequencing, institutional complementarity and path dependence. For factions in particular, the question was how factions had changed after the electoral reform of 1994 from a single non-transferable vote in multi-member districts (SNTV MMD) to the current mixed-member majoritarian system. One aspect that particularly interested the authors was how factional membership aligned with personnel decisions by the LDP. The most widely known example of this is that for decades cabinet positions were allocated roughly in proportion to factional size, in a kind of intra-LDP version of Gamson’s Law. So, one question the book was particularly keen to investigate was the role factions played in personnel decisions— 283

Robert J. Pekkanen and Saadia M. Pekkanen meaning assigning LDP Diet Members to positions in PARC (party), Diet committees (legislature) or government (cabinet and sub-cabinet). The authors adopted a multi-methods strategy to answer this question. The reasons for choosing a multi-method analysis were that the authors wanted to both understand the process and to have an overview of the outcomes. Either element by itself would have been a piece of the whole but insufficient for the research objectives of the authors. In other words, multi-methods were deliberately chosen as they represented the only viable path to achieving the research objectives of the authors. The qualitative methods research tools employed were documentary analysis and interviews. Documentary analysis meant a thorough review of the secondary literature on factions in Japanese and English, as well as a review of some primary sources, primarily LDP party documents. Documentary analysis was particularly valuable in the analysis of the development of factions over time. As the research question was framed as change over time, these were essential. Equally important were the elite interviews. One of us has argued elsewhere (Bleich/ Pekkanen 2013) for the importance of rigorous reporting standards being applied to interviews. In this case, the interviews were essential for an understanding of the process; the authors wanted to understand how factional affiliation could matter in deciding which Diet Member would be assigned to which committee. This type of information is not available from looking at the final committee assignments, but adheres to the deliberation process. To gather information on the nature of the decision-making process, interviews were a required method. In interviews with Diet Members, the authors asked about the decision-making process, both the how and the why. The authors wanted to know the process of how decisions were made, specifically who met with whom and how often. They learned that factional representatives negotiated the distribution of Diet and PARC posts. The authors also wanted to know what kinds of arguments were advanced in these meetings, and what kinds of arguments were considered convincing in advantaging one potential Diet Member appointee over another. As a result of employing these qualitative methods, the authors were able to create a model of how post allocation works within the LDP. All of this information could not be ascertained from public documents, nor could the process be inferred from looking at statistical analyses of committee assignments. Rather, interviews were required to obtain this information. To complement the authors’ qualitative methods employed for an understanding of the process, the authors also used quantitative methods to analyse the importance of factions in committee assignments. The key question here was whether factional affiliation or the lack thereof had any impact on the likelihood that a Diet Member would be given particular posts in the party, legislature or government. The authors found evidence that factional affiliation increased the chances of legislators receiving posts. This kind of analysis—weighing up how much a particular factor mattered compared to other factors—is particularly amenable to quantitative methods. We also used measures of proportionality to assess how factionally balanced Cabinets were before and after electoral reform, finding that proportionality did decrease markedly but remained relatively high. The authors were able to construct their arguments about the continuing importance of factions despite electoral reform by combining their qualitative and quantitative analysis. Interviews and documentary analysis showed that factions still mattered in the process, and quanti284

Chapter 10 How to combine methods tative analysis provided evidence that factional affiliation mattered in the results. The result of the combination of these two different research methods was a much stronger evidentiary basis for the arguments in The rise and fall of Japan’s LDP. Our other author relied heavily on MMR in one project but not another. This is because not every research question is amenable to the MMR approach. A lot depends on existing sets of theoretical literature, which can orient one’s thinking, the clarity of the question that can be posed and from which neat rival hypotheses can be derived, the availability of data that often comes one’s way fortuitously in the field, the pressures of time and resources and vagaries that invariably go with fieldwork, the methodological competences one possesses, and the epistemological orientation that resonates with the researcher in the final analysis. In Picking winners? From technology catch-up to the space race in Japan, the author combined a statistical analysis with structured focused comparisons and case studies from dissertation work. The book provided a look at the underpinnings of the developmental state model, drawing on the frameworks of the literature on strategic trade policy and endogenous trade policy. The general question it asked was how do governments choose industries to favour? If winners were to be picked, by what selection criteria could governments choose some industries over others? What did the case of postwar Japan teach us about these larger roiling theoretical issues of concern? MMR was the right choice for this book project because the author was concerned with doing justice to Japanese institutional peculiarities but also with extracting generalisations that could go beyond Japan. There was an element of luck that came out of the author’s professional networks. One thing that expedited the research was the availability of a time-series cross-sectional dataset to which the author could add, and that was key to providing an aggregate overview of the entire manufacturing sector fairly early; it facilitated the setting up of a test of rival economic and political logic to industrial selection, carried out through a variety of hard trade and industrial policies. This was the first glimmer that economic logic had held sway, and that bureaucrats had been at the helm of Japan’s industrial strategising for a good bit of time; the choice-theoretic logic that put politicians at the centre of industrial selection mattered of course, but not as consistently. That macro overview was necessary but not sufficient to establish more credible evidence on this finding, and was followed by structured data analysis that helped confirm roughly the same pattern of industrial selection. These aggregate windows helped set up the case studies, in which the author drew on interviews, and primary and secondary sources to extract the criteria for selection that were consistently dominant. These helped to illuminate the set of factors that actually mattered to policymakers on the ground and it grounded assessments of the theoretical criteria that were assumed to drive things. One thing that came out of this study for the author (Pekkanen 2003, p. 203) was that even a simple question can become very complicated very fast in the real world. The layers of methods helped tremendously in clarifying the answer from different perspectives; moreover, triangulating between them kept the author aware of the complexities of approximating the ‘truth’ (encouraging reflexivity) and cautioned against any extreme depictions of Japanese industrial strategy. In setting up the next book project, however, it was clear at the outset that the MMR approach would be wholly unsuited to the enterprise. The book sought to focus on Japanese space policy in a thematic and chronological manner, seeking to illuminate the changes in the legal and institutional context, the technology trajectories, and the motivations and manoeu285

Robert J. Pekkanen and Saadia M. Pekkanen vrings of corporations that were central to both. There were few other works available to build on. The author was able to join forces with a leading space journalist, leading to a book that drew on academic and real-world backgrounds. Our primary objective when we set out (Pekkanen/Kallender-Umezu 2010, p. 223) was to document specific developments in Japanese space assets across public and private actors, and to analyse what we saw as the market-tomilitary trajectories in Japanese space technology and policy. To that end, decade by decade and across all sets of space technologies, we process-traced the space activities of the Japanese public and especially corporate actors. Until our work, these were all thought of as scattered dots—the corporations, rockets, satellites and emerging technologies—and we focused on showing through detailed case study-based chapters how they were connected and what that implied about Japan’s broader militarisation controversies. We concluded then that Japan was on track to become a bigger military space power; and almost ten years later our predictions have been borne out. In retrospect, it is difficult to conceive of this project being carried out in any other way than through qualitative approaches. One was a practical constraint, as there was little to no previous work on the topic. Another was that the research goal was not to test but to illuminate some aspect of the unfolding realities on the ground. In the space book, given its centrality to real-world developments and policy, we recognised that there was also a heavy premium on getting things right. We therefore relied on piecing together and corroborating findings through a variety of qualitative methods and techniques, such as case studies based on primary and secondary sources, process tracing, interviews and participant observation. Our observations were drawn out across different cases and time, and it took some time to see how the dots were connected and what they suggested about Japan. MMR is certainly useful but it is affected by a deepening inequality in the training of social scientists. One significant problem in advocating this approach is that training in qualitative methods, noted for their usefulness to societies and policymakers, has been demoted in the Social Sciences (Desch 2019). If we are not formally training students and younger scholars in qualitative methods, how can we credibly talk about MMR? These realities also have the unfortunate effect of making qualitative methods, which can stand on their own, seem to be mere appendages to big data analysis. One way to address this gap is to strengthen training with respect to qualitative methodology in the Social Sciences, and one of us has taken steps in this respect as founding director of the Qualitative Multi-Methods Research Initiative (QUAL) at the University of Washington (2019) in Seattle. As takeaways for researchers, we suggest that methods should be chosen to best address the research question; there is no reason to always choose MMR as, in some cases, either qualitative or quantitative methods alone will be appropriate. We have found MMR particularly valuable when we wish to analyse both processes and outcomes. But when the nature of the observations across time and cases is unclear to the researchers, and when the data often (as in the field) have to be generated from first principles, qualitative methods are best suited to illuminating the processes at play and keeping the research accountable to the realities and people on the ground. Quantitative methods are best in other cases, for example when weighing up the differential contributions of multiple factors.

286

10.2 Texts, voices and numbers: Using mixed methods to sketch social phenomena Laura Dales

Researching friendship As a graduate student, I was trained in the ‘non-discipline’ of Asian Studies, under the good guidance of an anthropologist. As a result of this background, the social research I have conducted in Japan has tended towards the anthropological, with attention to Gender Studies and Cultural Studies along the way. My research has involved two main methods: the collection of subjective, reflective perceptions and experience elicited through semi-structured interviews and ad hoc discussions, and the observation and analysis of contextualised behaviour in particular field sites over extended periods. Using these methods, I have explored individuals’ perceptions, experiences and expectations of (inter alia) feminism, agency, singlehood, marriage, friendship, extramarital relationships, career, loneliness, belonging and happiness. When embarking on my most recent research project, I decided to build on this practice, with time and energy allowed by funding from the Australia Research Council for a project entitled Beyond the family: Relationships of intimacy in contemporary Japan (ARC DECRA (DE120101702)). This project primarily aimed to examine what role friendships and intimate relationships play in Japanese society, as singlehood becomes more common, fertility declines and the population ages. More specifically, I sought to use case studies of individual experiences with an eye to: 1. Mapping the roles and ramifications of non-family networks of support, in the context of broad demographic shifts (ageing, delay of marriage, low fertility). 2. Investigating gendered and generational differences in experiences and expectations of friendship and extra-familial intimacy. 3. Clarifying the effects of marriage on friendships and the effects of friendships on marriage (that is, marital plans, prospects and relations). 4. Analysing the discourse of intimate extra-familial relationships in recent popular media, notably magazines and popular non-fiction literature. In this project, as in my earlier work, I primarily used interviews and participant observation: spending several months over the course of four years in a café that presented itself as a hub of queer community and interpersonal relationships. I interviewed 68 people, either singly or in selfselected ‘friend’ groups of up to six people. Through both interviews and participant observation, I endeavoured to elicit and convey the ways that individuals see and move through their world, with attention to details that offer specific explanation or contextualisation of these perceptions and modes of behaviour. Of course, there are limitations to this approach in terms of the picture that can be captured. For example, when an unmarried woman observes in dialogue that a friendship has become complicated since her counterpart married or had children, I take this observation as both a reflection of her lived experience, and a response to the particular discussion on friendships and marriage which I am eliciting. That is to say, it features centrally in my interview/fieldwork notes—because 287

Laura Dales this is a central thematic interest of my research—but it may not feature as a prominent daily concern in my interviewee’s life. Thus, it may only be raised by interviewees incidentally, or in specific response to the questions I posit. Furthermore, the details of the friendship—for example, the nature of its complication or how the interviewee has dealt with this complication—may be specific to the particular instance, and in that sense cannot be extrapolated to a larger population.

Triangulating with contemporary literature To determine whether the individual might be understood as representative of a greater group, I compare the details of individual reflections with those elicited in discussions with others, both like (in the case above, unmarried women) and unlike (e.g. married women/men). This is not triangulation in a narrow sense, because the data is similar: interviews and discussions. Rather it is contextualisation that provides a richer sense of the ways that individuals fit into groups. For example, what are the key themes among the responses of married women interviewees, and are these similar or different to responses given by unmarried women? Is there meaning in that similarity/difference? But where there are only few interviewees in a category, this might yet remain insufficient as an indicator of significance or prevalence. If one recognises this, triangulation can be helpful. Triangulation, in my understanding, involves the comparison of different kinds of data, to check or verify the accuracy or uniformity of data, but also to provide deeper insights into the context of the study (Taylor et al. 2015, p. 94). I use contemporary literature, typically essays, non-fiction social commentary, academic texts and popular media, to examine the salience and spread of particular themes emerging in interviews. How have Japanese authors written about friendships? Is conflict between friendship and marriage experienced broadly, or chronically, among particular demographic groups in society, and if so how is this understood in popular discourse? How are extra-familial intimate relationships viewed by sociologists and social critics? Thus, in exploring friendships and singlehood, I have looked at work by the sociologists Masahiro Yamada and Chizuko Ueno, the feminist writers Yōko Haruka and Minori Kitahara, the research think tank leader Kazuhisa Arakawa and the journalists Momoko Shirakawa and Megumi Ushikubo. The genres of these texts can be defined variously as scholarly, social commentary, social critique and self-help, with differing levels of popularity and circulation. The value of using these texts is twofold: firstly, to assist in triangulating the experiences (Gray et al. 2007, p. 75). Secondly to indicate the currency of the themes that arise in discussions and interviews with individuals. Particularly for researchers based outside Japan, it can be easy to rely on English-language material as the secondary sources scaffolding our own research findings, without engaging with the work of local scholars, writers and social observers. The dangers in this are self-evident; using Japanese texts can help to avoid the potential insularity (or at worst, Orientalism) of this approach (see Liu-Farrer, Ch. 4.3; Zachmann, Ch. 4). Incorporating popular as well as academic literature reflects the ways that social phenomena are interpreted more broadly: how the media and popular culture can shape the discourse around social issues such as singlehood, and drawing attention to gaps between ideal and practical realities (Dales 2014).

288

Chapter 10 How to combine methods

Adding a quantitative lens Following the collection of qualitative data over four years of the project, I was encouraged by a colleague to consider developing and applying a research tool with which to gather a bigger, and more statistically significant, set of data on friendships. I should note that at the outset, as a scholar who is not trained in quantitative methodologies, I was particularly ambivalent about designing a survey for large-scale implementation: both because of uncertainty regarding the tools of quantitative analysis, and because I was not convinced that it would provide insight beyond the data already gathered in my qualitative fieldwork and interviews. But the potential insights offered by a broad-scale (n=2,500) national survey were enticing and ultimately productive to the project. I began by looking at others’ large-scale surveys, introduced in sociological articles, with attention to how they had been facilitated. It is beyond the scope of most single researchers to conduct large surveys without significant support, and the key consideration for many in this position is likely to be costs: the larger and more detailed the survey—and the more efficient and professional the service—the more expensive it is likely to be. Because I was fortunate to have a budget for this study, I was able to focus on the services provided by particular market research companies: the capacity to conduct a national survey with specified demographic delineators, i.e. a general balance across age groups, sex, employment categories and marital status. Ultimately, I chose to use a particular market research company because a senior Japanese sociologist colleague, who had used their services for his own research in the past, recommended it. Given that many social researchers in Japan have experience in quantitative studies of this kind, I would recommend looking to these colleagues for advice, particularly for those embarking on quantitative research for the first time. In designing the survey, I decided my first step was to develop a list of the areas of which I sought further (or broader) examination, and to clarify the target audience of respondents for the proposed survey. I aimed to attract respondents from across Japan to ensure my findings were less geographically specific than my qualitative work (which had been focused in urban and semi-urban areas), and as equally as possible across age groups (20–50 years) and economic strata. The latter was a particularly important consideration as I sought to counter biases among my qualitative research population: 44 women (relative to 24 men), skewed towards a more highly educated middle class. I also hoped to include a sizeable sample of individuals identifying themselves as same-sex attracted, but this proved difficult: only 2.6% of both the females and males surveyed indicated that they were in a same-sex relationship. I also considered the kinds of information I had found revealing and/or recurring in interviews: the lengths and resilience of particular kinds of friendships, the effects of marriage upon friendship, the impact of gender on experiences of friendship, and the extent to which friendships supported or challenged singlehood. These formed the basis for questions developed for the survey. Further, because I am interested in perceptions of well-being and connection, I decided to include a psychological tool: the Subjective Vitality Scale, a 5-term model adapted from the Basic Psychological Need Satisfaction Scale. This tool requires participants to respond to questions about (inter alia) their sense of vitality, freedom, satisfaction, connection and competence, by indicating the degree to which they agree with set statements. Following design completion, I submitted the survey questions, instructions and related documentation to my university’s Human Research Ethics Committee to obtain permission to conduct the survey (see Reiher/Wagner, Ch. 16). De-

289

Laura Dales pending on the institution and its requirements, this can be a lengthy process, so it is important to factor this in to the time frame for your research. Analysis of the survey results is ongoing. I enlisted the help of a research assistant experienced in SPSS analysis to provide an initial overview of first research findings, while also exploring the data in its basic Excel format. The next steps will involve comparing and contrasting key themes and findings from the survey and those from the qualitative research, and then situating this analysis within the overall study. So far, the survey has produced some interesting findings. Although I did not ask explicitly about happiness and life satisfaction in my qualitative interviews, I was curious to see the impact of marriage and children on surveyed individuals’ sense of satisfaction; however, the overall findings showed little difference between never-married and married. Further, having children slightly increased satisfaction rates (compared to those without children), even for single parents. Interestingly, the highest scores for satisfaction were seen in the category ‘widowed’! I can already conclude that the survey produced valuable contributions to my overall study: both in the breadth of its scope and findings, and in its support for the micro-level observations made through individual case studies. The granularity of interviews is obscured in the broad brushstrokes of the survey, but the latter reveals national trends, group patterns of satisfaction and practices of friendship that could not easily be generalised from the individual interviews. Political and social concerns around precarity and disconnection in contemporary Japan, reflected in the recent proliferation in literature on living alone, are not obvious from the survey findings on relatedness and connection. But these are nonetheless themes that emerge from discussions and observations conducted in the qualitative research, and are therefore worthy of scholarly attention.

Conclusion Plainly, there are many ways to answer questions about contemporary Japanese society. Insofar as each researcher has only so much time, energy and resources, before one begins a project, it is important to consider its scope and possibilities: how much time and resources you have to devote to the study, from which to ascertain both the methods and the size of the sample. In a shorter-term research period, such as a master’s programme, it may be difficult to design, plan, conduct and analyse a large-scale survey, even assuming that permission has been obtained from the appropriate research ethics bodies. In this context, consider ways to conduct smaller studies with similar approaches. For example, you may limit the target audience to a particular demographic, with fewer responses needed to obtain a sense of the particular perspective. Similarly, when considering interviews and qualitative fieldwork, bear in mind that organisation of interviews (not to mention transcription and analysis) can be time-consuming and difficult to organise, especially if you are not based in Japan. A large-scale survey, when conducted by market research companies, can be expensive; in-depth ethnographic exploration of individuals or a specific environment requires time and energy. When deciding on an approach, one should obviously also consider the aims of the project: Is it imperative to map broad trends and produce generalisable conclusions? Or is it more important to produce a ‘narrow but deep’ study of a particular site or individuals? There are benefits to both

290

Chapter 10 How to combine methods approaches; it simply depends on the aims of the research and the interests and material capacities of the researcher.

291

10.3 Examining facts from different angles: The case of the deregulation of employment relations in Japan Jun Imai

Questions as a starting point The term ‘mixed methods approach’ does not simply mean encouraging researchers to use multiple methods. It means encouraging a strategic use of different methods under the principle of triangulation. In my view, triangulation facilitates productive ‘communication’ between analytical frameworks and data. Research textbooks often illustrate research as either a deductive or an inductive process. However, research does not proceed in such a linear fashion; rather it is a process of constant (re)adjustment of interpretation and observation (retroductive) (Ragin/Amoroso 2011). The idea of triangulation facilitates this process, and a mixed methods approach helps to produce well-informed, nuanced and convincing interpretations of data and causal sequences and enables researchers to articulate their findings in explanatory narratives. The examples I introduce here are from my research on the deregulation of the employment relations and labour market in Japan, a trend that happened mainly during the period between 1995 and 2005, but is still ongoing. The deregulatory reforms at the turn of the century are known to have had significant impacts on Japanese society today. The reforms included the expansion of non-regular employment forms such as temporary dispatched work (rōdōsha haken, hereafter temporary work), which was first established in the late 1980s and significantly expanded in and after the decade of deregulatory reforms. It also included the case of the Discretionary Work System (sairyō rōdōsei, hereafter DWS), a working time regulation that almost exempts a segment of white-collar workers from the regular working time regulations. This was also first established in the late 1980s, and significantly expanded due to deregulations in this period. Its further expansion was discussed during the recent work-style reform (hataraki-kata kaikaku, implementation April 1, 2019). I had questions such as why, how and for whose benefit these deregulatory reforms were implemented. What are the consequences of these reforms? By asking these questions, I wanted to understand the impact of the deregulatory reforms on employment relations and labour markets in Japan, focusing on policy debates, the changes to labour management practices, the changes to the structure of inequality, the patterns of social mobility and the ways workers make effort at the workplaces. I planned to contribute to the research literature that examines the changes to and continuity of the employment institutions in industrial democracies (Yamamura/Streeck 2003), and that attempts to identify new characteristics of inequality emerging in these societies (Emmenegger et al. 2012). The results of this research project were published as books and articles in the early 2010s (Imai 2011; Sato/Imai 2011). In order to produce finely tuned explanations about social reality, I recommend always keeping the concept of triangulation in mind. In my case, using multiple methods was a natural and appropriate choice to understand the policy process and its social consequences. The most common methods I used in my research were archival work/document analysis, expert inter292

Chapter 10 How to combine methods views and quantitative data analysis. Putting all these into historical perspective was indispensable, as having knowledge about the development of the employment relations and labour market structure in Japan greatly helped me to interpret my observations. By combining the different data and methods, I constantly performed triangulation on the issue in question.

Tuning the interpretation via triangulation Usually, the main research questions guide the research, but the steps described below often take place in parallel and/or even in an intertwined manner. In the case of my research, the first thing I needed to do was to collect documents related to the regulative reforms, such as policy proposals published by related parties and transcriptions of the important meetings, in addition to the laws in question. It was necessary to read all these materials to roughly grasp what was going on in the field related to the issue in question. However, understanding legislative texts, related documents and policy discussions is often difficult for those who are not policy or legal experts. The legal explanation provided by legal experts often did not help us to sociologically understand the impact of these reforms since those explanations were for legal scholars and practitioners. Legal scholars need to put reforms into the context of legal developments and theory, and practitioners need to understand the legal details to practice law properly. Since legal abstraction is different from sociological abstraction, we needed to put the laws, their revisions and the discussions on them into a sociological context. The guiding principle of a sociological interpretation should be the theoretical argument. In my case, the argument was that the policy process and the articulations of the legislative texts (and the guidelines provided by the ministry) reflect the balance of power between the related actors, such as the state, employers and employees, in the case of employment/labour market regulations. Since the actor who has the upper hand in this matter benefits most from legal revisions, measuring to what extent the reform favours specific stakeholders is a crucial analytical point. Expert interviews greatly helped us to read between the lines of the documents and the discussions. For instance, the minutes of the ministry’s advisory councils’ (shingikai) meetings were the main source of information about the major points of the ongoing discussions and the stances taken by the most important actors. However, only relying on these sources frustrated us since they appear to lack some important issues. When the expansion of the list of occupations for temporary work was considered at the advisory council, I wondered why non-expansion was never considered and why the discussion was so hasty. It was a series of expert interviews with the members of the advisory councils and Deregulation Committee, including representatives from employers, labour and scholars, who taught me what the ‘shifting balance of power between related actors’ means in reality. During the deregulatory reforms, the Deregulation Committee, which includes entrepreneurs and deregulation advocates, became an important actor that tilted the balance of power in traditional tripartite corporatism. By that time, I already knew that the alliance between employers and the Deregulation Committee had the upper hand in the process of deregulation. In an interview, a public representative, who was supposed to be neutral towards employers and workers, told me that due to the tilted balance of power, the points I raised were not the issue or an option. Whether to include working at manufacturing lines in the list of non-applicable jobs in the category of temporary work or not was the real issue from the beginning. This in-

293

Jun Imai terview clearly taught me that the shifting balance of power could push potential issues and options away, and that this was exactly the reason I felt the discussion was so hasty. The imbalance of power facilitated discussions in favour of one party. A brief stint of participant observation was also helpful in me understanding the growing influence of business and deregulation advocates in the policy process, which confirmed which I was told in the expert interviews. I attended one session of the above-mentioned advisory council, where three actors were traditionally on an equal footing about agenda setting. However, in the session I attended as an observer, I witnessed that the agenda setting power was clearly in the hands of employers due to the rising status of the Deregulation Committee that backed them up. It was my moment of ‘seeing is believing’. When the employees’ side refused to discuss the issue, which was a normal way of doing things at advisory councils, the employers had no hesitation in pushing forward their agenda, citing that the agenda had already been set by the Deregulation Committee. While the labour representatives were up against the ropes, there was no support for them, even from the public representatives. In this research project, I mainly used survey data collected by government ministries, such as the Employment Status Survey (shūgyō kōzō kihon chōsa), the Labour Force Survey (rōdō ryoku chōsa) and the General Survey on Working Conditions (shūrō jōken sōgō chōsa) to check the quantitative significance of the deregulation and the shift of power relations. My examination of that data, using cross-tab tables including mobility tables (that show the patterns of job changes) and graphs of chronological development, helped me to understand the changes in the sections of the population affected by the policy changes. I examined the situations of every employment form, considering important variables for the study of inequalities such as gender, company size and mobilities. These examinations did not provide an in-depth understanding of the phenomena. However, they did reveal an accurate profile of the variables and their impact on the issue. It is important to connect these findings with qualitative observations. Since the reforms were driven by employers, I expected an expansion of non-regular employment that further helps employers to achieve flexibility in labour management, and, in fact, the number of temporary workers increased greatly after the revisions in 1997 and 1999, especially in clerical jobs that were mainly occupied by women. This was followed by an even greater increase in male-dominated manufacturing jobs after the revision in 2004. My analysis confirmed this ‘expansion’ proposition, but the combination of my findings from the survey data and the knowledge drawn from the qualitative investigation informed by the existing literature helped me to reach an even deeper understanding of the issue especially concerning the relationship between the expansion and gender. For example, men’s increase in temporary work was eye-catching. This increase, I concluded, prepared the ground for the turmoil after the Lehman crisis in 2008. However, I recognized that it could mean more to the social change because it looked like it violated the conventional relationship between employment and gender. Following the crisis, so-called ‘tent cities’ became a visual symbol of those temporary workers, who had lost their jobs due to a termination of contract. Since temporary work was usually considered as a possible form of work for women, the media coverage of men, who are supposed to be breadwinner, thrown onto the streets as homeless people was so shocking that it seemed to trigger the re-regulation of temporary work in the following years. The observation of such a tight relationship between employment form and gender even generated new questions. For instance, these clearly gendered developments posed the question of whether the initial establishment of temporary work in 294

Chapter 10 How to combine methods the mid 1980s and the establishment of the Equal Employment Opportunity Law (EEOL) in 1986 was coincidence; was the disproportionate allocation of women into the expanded temporary work intentional? Although I could not follow up on this question, it is an important historical issue as it may reveal that gendered inequality was deepened or deliberately preserved when it was supposed to be being dismantled. There was a partial answer to the continuity/change question. The expansion of the new forms of employment such as temporary work and contract work (keiyaku) grew more significantly at large firms than in small and medium-sized firms, where traditional non-regular employment such as part-time work (pāto) seems to be enough to achieve flexibility. This proves that the shift in employment practices by large Japanese companies was enabled by the deregulatory reforms.

Importance of the historical context Putting any observation into a historical context is necessary to measure and clarify the significance and the exact meaning of the findings. It is an indispensable part of the efforts to interpret policy discussions, legislative texts, power dynamics and changes in numbers. In my research, for instance, I found that employers tended to demand ‘autonomous decision-making at an organisational level’ (rōshi jichi). The discussion on the above-mentioned DWS was no exception. In order to decide to whom the DWS should be applied, employers constantly argued that these decisions should be made in each respective organisation and not through government regulations. It is no surprise that employers opted for deregulation and preferred decentralised, organisational-level decision-making. However, to properly understand the abovementioned discussions, specific knowledge about the history of Japan’s labour market and working time regulations is necessary. The Labour Standards Act (LSA) stated that the upper limit of overtime work is 45 hours per month in Japan. However, there was a big loophole in this law. Article 36 of the LSA includes a clause on exceptional circumstances that allows companies to negotiate the upper limit of overtime work depending on the situation of individual companies. This agreement—negotiated at the organisational level—made extreme overwork possible and completely undermined the social regulations. The media reported the shocking examples of this agreement’s consequences: overtime at some major companies exceeded 100, or even 200 hours per month. Knowing that it has been already negotiated at the organizational level in favor of employers, employers’ assertion cannot be interpreted simply as the general preference of decentralized decision making. Japanese employers knew what they could benefit from controlling working time and were confident to control it at the organizational level. They wanted to maintain the status quo of the power relations by extending it to the governance of the DWS.

Summary In this essay, I introduced examples of triangulation that aimed to produce well-informed, nuanced and convincing interpretations of observations on deregulation processes in the Japanese labour market and presented findings and their explanations. In sum, I examined the same issues in documents, people’s narratives and behaviour, numbers and accounts of histori-

295

Jun Imai cal contexts to finalise my interpretation. Although it sounds simple, I want to conclude this essay by pointing out some difficulties of this approach. Triangulation, once it is in progress, is clearly effective in answering research questions. However, it is also effective in generating further questions. It is a productive way to conduct research, but it also raises the difficult question of ‘saturation’: When should researchers stop collecting and analysing data (see Gerster, Ch. 12.3; Rosenberger, Ch. 12.1; Spoden, Ch. 12.2)? In order to make this decision, researchers have to reconsider their data and interpretations, reflect on the breadth and depth of their research question and ask whether it has reached the point of saturation.

296

Further reading Creswell, John W. (2014): Research design: Qualitative, quantitative & mixed methods approaches. Los Angeles, CA: Sage. Creswell, John W./Creswell, J. David (2018): Research design: Qualitative, quantitative & mixed methods approaches. International student edition. Los Angeles, CA: Sage edge. Johnson, Burke R./Onwuegbuzie, Anthony J./Turner Lisa (2007): Toward a definition of mixed methods research. In: Journal of Mixed Methods Research 1, No. 2, pp. 112–133. Schrauf, Robert W. (2016): Mixed methods: Interviews, surveys, and cross-cultural comparisons. Cambridge: Cambridge University Press. Tashakkori, Abbas/Teddlie, Charles (eds.) (2010): SAGE handbook of mixed methods in social & behavioral research. Los Angeles, CA: Sage. Teddlie, Charles/Tashakkori, Abbas (2009): Foundations of mixed methods research: Integrating quantitative and qualitative approaches in the Social and Behavioral Sciences. Los Angeles, CA: Sage. Venkatesh, Viswanath/Brown, Susan A./Sullivan, Yuila W. (2016): Guidelines for conducting mixed methods research: An extension and illustration. In: Journal of the Association for Information Systems 17, No. 7, pp. 435–494.

References Ahmed, Amel/Sil, Rudra (2012): When multi-method research subverts methodological pluralism: Or, why we still need single-method research. In: Perspectives on Politics 10, No. 4, pp. 935–953. American Psychological Association (APA) (2020): APA style journal article reporting standards: JARSmixed, table 1: Mixed methods article reporting standards (MMARS): Information recommended for inclusion in manuscripts that report the collection and integration of qualitative and quantitative data. https://apastyle.apa.org/jars/mixed-table-1.pdf, [Accessed June 14, 2020]. Arnault, Denise S./Fetters, Michael D. (2011): RO1 funding for mixed methods research: Lessons learned from the ‘mixed-method analysis of Japanese depression’ project. In: Journal of Mixed Methods Research 5, No. 4, pp. 309–329. Behr, Dorothée/Shishido, Kuniaki (2016): The translation of measurement instruments for cross-cultural surveys. In: Wolf, Christoph/Joye, Dominique/Smith, Tom/Fu, Yang-Chih (eds.): The SAGE handbook of survey methodology. Los Angeles, CA: Sage, pp. 269–287. Bleich, Erik/Pekkanen, Robert J. (2013): How to report interview data. In: Mosley, Layna (ed.): Interviews in Political Science research. Ithaca, NY: Cornell University Press. Brannen, Julia (1992): Combining qualitative and quantitative approaches: An overview. In: Brannen, Julia (ed.): Mixing methods: Qualitative and quantitative research. Aldershot: Avebury, pp. 3–37. Bryman, Alan (2016): Social research methods. Oxford: Oxford University Press. Creswell, John W./Creswell, J. David (2018): Research design: Qualitative, quantitative & mixed methods approaches. Los Angeles, CA: Sage edge. Creswell, John W./Plano Clark, Vicki L./Gutmann, Michelle L./Hansons, William E. (2003): Advanced mixed methods research design. In: Tashakkori, Abbas/Teddlie, Charles (eds.): SAGE handbook of mixed methods in social & behavioral research. Los Angeles, CA: Sage, pp. 209–240. Creswell, John W./Tashakkori, Abbas/Jensen, Ken D./Shapley, Kathy L. (2010): Teaching mixed methods research: Practices, dilemmas, and challenges. In: Tashakkori, Abbas/Teddlie, Charles (eds.): SAGE handbook of mixed methods in social & behavioral research. Los Angeles, CA: Sage, pp. 619–637. Dahlberg, Britt/Wittink, Marsha N./Gallo, Joseph J. (2010): Funding and publishing integrated studies: Writing effective mixed methods manuscripts and grant proposals. In: Tashakkori, Abbas/Teddlie, Charles (eds.): SAGE handbook of mixed methods in social & behavioral research. Los Angeles, CA: Sage, pp. 775–802. Dales, Laura (2014): Ohitorisama, singlehood and agency in Japan. In: Asian Studies Review 38, No. 2, pp. 224–242. De Leeuw, Edith/Berzelak, Nejc (2016): Survey mode or survey modes. In: Wolf, Christoph/Joye, Dominique/Smith, Tom/Fu, Yang-Chih (eds.): The SAGE handbook of survey methodology. Los Angeles, CA: Sage, pp. 142–156. Denzin, Norman K. (1978): The research act: A theoretical introduction to sociological methods. New York, NY: Praeger.

297

References Desch, Michael C. (2019): How Political Science became irrelevant. https://www.chronicle.com/article/Ho w-Political-Science-Became/245777, [Accessed 10 March 2019]. Emmenegger, Patrick/Häusermann, Silja/Palier, Bruno/Seeleib-Kaiser, Martin (eds.) (2012): The age of dualization: The changing face of inequality in deindustrializing societies. Oxford: Oxford University Press. Fetters, Michael D. (2016): ‘Haven’t we always been doing mixed methods research’: Lessons learned from the development of the horseless carriage. In: Journal of Mixed Methods Research 10, No. 1, pp. 3–11. Fetters, Michael D./Freshwater, Dawn (2015): Publishing a methodological mixed methods research article. In: Journal of Mixed Methods Research 9, No. 3, pp. 203–213. Fowler, Floyd J. (2014): Survey research methods. Los Angeles, CA: Sage. Gerring, John (2012): Social Science methodology: A unified framework. New York, NY: Cambridge University Press. Gray, Paul S./Williamson, John B./Karp, David A./Dalphin, John R. (eds.) (2007): The research imagination: An introduction to qualitative and quantitative methods. Cambridge: Cambridge University Press. Hammond, Michael/Wellington, Jerry (2013): Research methods: The key concepts. London: Routledge. Hartwig, Manuela/Tkach-Kawasaki, Leslie (2019): Identifying the ‘Fukushima effect’ in Germany through policy actors’ responses: Evidence from the G-GEPON 2 survey. In: Quality and Quantity: International Journal of Methodology 53, No. 4, pp. 2081–2101. Hatta, Taichi/Narita, Keiichi/Yanagihara, Kazuhiro/Ishiguro, Hiroshi/Murayama, Toshinori/Yokode, Masayuki (2018): Crossover mixed analysis in a convergent mixed methods design used to investigate clinical dialogues about cancer treatment in the Japanese context. In: Journal of Mixed Methods Research 14, No. 1, pp. 84–104. Hesse-Biber, Sharlene N. (2010): Mixed methods research: Merging theory with practice. New York, NY: The Guilford Press. Hesse-Biber, Sharlene N./Johnson, Burke R. (2015): The Oxford handbook of multimethod and mixed methods research inquiry. Oxford: Oxford University Press. Hommerich, Carola (2009): Freeter und Generation Praktikum: Arbeitswerte im Wandel. München: Iudicium. Hommerich, Carola/Tiefenbach, Tim (2018): Analyzing the relationship between social capital and subjective well-being: The mediating role of social affiliation. In: International Journal of Happiness Studies 19, No. 4, pp. 1091–1114. Huber, Matthias (2018): Reflexive modernity in practice: The methodology of comparing German and Japanese parents. In: Bertram, Hans/Holthus, Barbara (eds.): Parental well-being: Satisfaction with work, family life, and family policy in Germany and Japan. München: Iudicium, pp. 16–42. Imai, Jun (2011): The transformation of Japanese employment relations: Reform without labor. Houndmills, Basilstoke: Palgrave Macmillan. Johnson, Burke R./Christensen, Larry (eds.) (2016): Educational research: Quantitative, qualitative and mixed approaches. Los Angeles, CA: Sage. Johnson, Burke R./Onwuegbuzie, Anthony J./Turner Lisa (2007): Toward a definition of mixed methods research. In: Journal of Mixed Methods Research 1, No. 2, pp. 112–133. Kobayashi, Jun/Hommerich, Carola (2018): What words make people happy? A mixed methods approach by text mining analyses of open-ended data in Japan (in Japanese). In: Sōsharu Uerubīingu Kenkyū Ronshū 4, pp. 31–47. Kobayashi, Jun/Kawabata, Kenji (eds.) (2019): Henbō suru ren’ai to kekkon: Dēta de yoku heisei. Tōkyō: Shinyōsha. Krauss, Ellis S./Pekkanen, Robert J. (2011): The rise and fall of Japan’s LDP: Political party organizations as historical institutions. Ithaca, NY: Cornell University Press. Pekkanen, Robert (2006): Japan’s dual civil society: Members without advocates. Stanford, CA: Stanford University Press. Pekkanen, Robert/Nyblade, Benjamin/Krauss, Ellis S. (2006): Electoral incentives in mixed-member systems: Party, posts, and zombie politicians in Japan. In: American Political Science Review 100, No. 2, pp. 183–193. Pekkanen, Saadia M. (2003): Picking winners? From technology catch-up to the space race in Japan. Stanford, CA: Stanford University Press. Pekkanen, Saadia M./Kallender-Umezu, Paul (2010): In defense of Japan: From the market to the military in space policy. Stanford, CA: Stanford University Press. Ragin, Charles C./Amoroso, Lisa M. (2011): Constructing social research: The unity and diversity of method. Los Angeles, CA: Sage.

298

Chapter 10 How to combine methods Sandelowski, Margarete (2010): The challenges of writing and reading mixed methods studies. In: Tashakkori, Abbas/Teddlie, Charles (eds.): SAGE handbook of mixed methods in social & behavioral research. Los Angeles, CA: Sage, pp. 321–350. Sato Yoshimichi/Imai, Jun (eds.) (2011): Japan’s new inequality: Intersection of employment reforms and welfare arrangements. Melbourne: Trans Pacific Press. Schrauf, Robert W. (2016): Mixed methods: Interviews, surveys, and cross-cultural comparisons. Cambridge: Cambridge University Press. Tashakkori, Abbas/Creswell, John W. (2007): The new era of mixed methods. In: Journal of Mixed Methods Research 1, No. 3, pp. 3–7. Tashakkori, Abbas/Teddlie, Charles (eds.) (2003a): SAGE handbook of mixed methods in social & behavioral research. Los Angeles, CA: Sage. Tashakkori, Abbas/Teddlie, Charles (2003b): Major issues and controversies in the use of mixed methods in the Social and Behavioral Sciences. In: Tashakkori, Abbas/Teddlie, Charles (eds.): SAGE handbook of mixed methods in social & behavioral research. Los Angeles, CA: Sage, pp. 3–50. Tashakkori, Abbas/Teddlie, Charles (eds.) (2010): SAGE handbook of mixed methods in social & behavioral research. Los Angeles, CA: Sage. Taylor, Steven J./Bogdan, Robert/DeVault, Marjorie (2015): Introduction to qualitative research methods : A guidebook and resource. Hoboken, NJ: John Wiley & Sons. Teddlie, Charles/Tashakkori, Abbas (2009): Foundations of mixed methods research: Integrating quantitative and qualitative approaches in the Social and Behavioral Sciences. Los Angeles, CA: Sage. University of Washington (2019): University of Washington qualitative multi-method research program. http://depts.washington.edu/qual/, [Accessed 04 November 2019]. Venkatesh, Viswanath/Brown, Susan A./Sullivan, Yuila W. (2016): Guidelines for conducting mixed methods research: An extension and illustration. In: Journal of the Association for Information Systems 17, No. 7, pp. 435–494. Willis, Gordon B. (2016): Questionnaire pretesting. In: Wolf, Christoph/Joye, Dominique/Smith, Tom/Fu, Yang-chih (eds.): The SAGE handbook of survey methodology. Los Angeles, CA: Sage, pp. 359–381. Yamamura Kozo/Streeck, Wolfgang (2003): The end of diversity? Prospects for German and Japanese capitalism. Ithaca, NY: Cornell University Press.

299

Chapter 11 How to analyse data: An introduction to methods of data analysis in qualitative Social Science research David Chiavacci

1. Introduction Data analysis is like baking a cake (Kornmeier 2018). The raw data or materials you have collected are just as important to your argument as the ingredients—sugar, eggs, flour—are for baking a cake. But to develop this argument from your data or material—your final goal—you have to work with the ingredients. That means that you have to think through your data and transform them into evidence for an innovative argument that should be as pleasing as a splendid cake. Not many people will be interested in your raw data—or ingredients—as such, but an innovative argument or a good cake might bring you fame. Hence, data analysis is arguably not only an important step but also the core of qualitative Social Science research. While for a long time the main methodological focus lay on data collection per se, data analysis has increasingly attracted attention in recent years. Parallel to the huge expansion of, differentiation in and the many turns in qualitative research methods and methodology of the last few decades,1 qualitative data analysis has also and perhaps even most markedly advanced and diversified: while earlier handbooks and textbooks usually cover the entire qualitative research process (Denzin/Lincoln 2005a; Flick et al. 2000; Silverman 2000), an increasing number of thick introductory publications have appeared in recent years that focus primarily or even exclusively on qualitative data analysis (Bazeley 2013; Bernard et al. 2017; Flick 2014a; Grbich 2007; Harding 2019; Miles et al. 2019). This leads to highly complex recent definitions of qualitative data analysis, which try to incorporate different traditions and worldviews of the diverse field of qualitative research. Uwe Flick (2014b, p. 5), for example, proposes the following definition, which looks rather like an abstract than a definition with a length of 114 words: Qualitative data analysis is the classification and interpretation of linguistic (or visual) material to make statements about implicit and explicit dimensions and structures of meaning-making in the material and what is represented in it. Meaning-making can refer to subjective or social meanings. Qualitative data analysis also is applied to discover and describe issues in the field or structures and processes in routines and practices. Often, qualitative data analysis combines approaches of a rough analysis of the material 1 For an overview of the history and development of qualitative research, see Norman K. Denzin and Yvonna S. Lincoln (2005b, pp. 14–20). However, this model has been criticised as representing primarily the development of qualitative research in Anglo-Saxon areas. Uwe Flick (2014b, pp. 8–10), for example, has developed a quite different account of the history of qualitative research in German-speaking countries.

300

Chapter 11 How to analyse data (overviews, condensation, summaries) with approaches of a detailed analysis (elaboration of categories, hermeneutic interpretations or identified structures). The final aim is often to arrive at generalizable statements by comparing various materials or various texts or several cases. This chapter will not be able to address all of the issues raised in the above definition in detail but will seek to provide an introduction to and overview of the main aspects of qualitative data analysis in general and research on Japan in particular. The chapter is primarily intended for PhD students and young researchers who want to learn more about qualitative research, and it discusses some of the major steps and challenges of qualitative data analysis. First, the position of qualitative data analysis in the entire research process and its relationship to the other research steps is introduced. Then a brief overview of some of the main analytical approaches is given. In the next section, the main work stages of qualitative data analysis—data transcription, data coding, identification of patterns and the development of theoretical models—is addressed. This overview is followed by a discussion of the pros and cons of the use of computerassisted (or aided) qualitative data analysis software (CAQDAS), which has undergone huge advances in recent years. The next section focuses on reflexivity as an important means to ensure the quality of data analysis. This is followed by a discussion of issues related to the Japanese language in qualitative data analysis, before the chapter closes with a brief summary and a few final remarks.

2. Qualitative data analysis and the research process The Social Science research process is often depicted as a sequence of clear-cut steps in the neat research plans of PhD students: decide on your research topic → identify a research gap → determine your theoretical approach → formulate your research question → decide on the method(s) → collect your empirical data → analyse your data → write your academic publications. However, the research process is often not a simple sequence of steps but is circular and consists of overlaps between the different steps. In qualitative Social Science research, in particular, data analysis begins during the collection process and not afterwards. Researchers often engage in a number of rounds of data collection and data analysis in qualitative research. In such cases, analysis of data from earlier and sometimes preliminary fieldwork is often intentionally used to further focus and sometimes even to completely redefine the research question (see this chapter, Castro-Vázquez, Ch. 11.2). These loops are not a problem, but actually a strength of qualitative research. They allow us to reflect on our research question and on the appropriacy of our theoretical and methodological approach. However, it is also important to remain pragmatic. Research projects can be extended to any length of time through these loops. However, the world of academic careers requires researchers to come to an end and produce output. The goal should be to make an important and innovative contribution to the understanding of the world, and not to produce a full and final explanation of it. The degree to which researchers engage in such loops and the circular research process depends on the research question and research methods. For example, if the policymaking process is to be analysed through semi-structured qualitative interviews with political actors, in301

David Chiavacci terviews will most likely only begin once the researcher has a good overview and sound understanding of the entire policymaking process and its key players. Accordingly, the data analysis of early interviews may rarely lead to a reformulation of the research question, but most likely help to identify important interview partners and to refine the questions asked in the later interviews by using new insights from earlier interviews (see this chapter, Hülsmann, Ch. 11.1). However, if you want to analyse, for example, the behaviour and emotional experience of customers in convenience stores, you will most likely start with participant observation, begin qualitative interviews or focus group research much earlier. In such a case, the early analysis of the data collected is actually needed for you to understand the field of research better and be able to formulate a meaningful research question. In any case, it is very important to plan enough time for careful data analysis during data collection to ensure the quality of your research.

3. Approaches to data analysis The rich and long history of qualitative research has led to a large number of research traditions that often include not only various data collection methods but also various data analysis methods. These research traditions are often linked to certain theoretical positions and philosophical worldviews, such as constructivism, phenomenology, hermeneutics or post-modernism (see Goodman, Ch. 1). In addition, each of these traditions consists of several subfields, and it is often not only the different traditions but also their sub-fields that stand in quite stark contrast to each other. However, aside from complex theoretical and philosophical questions, the simple typology of analysis approaches proposed by Catherine Dawson (2002, pp. 111–120) is, in my view, appropriate for a useful and hands-on overview (Harding 2019, p. 194). Thematic analysis This approach is very inductive. Thematic analysis aims to identify the main recurrent themes that arise from the data and material itself. In other words, researchers avoid imposing themes on the data but try to approach their data as open-mindedly as possible in order to allow them to emerge from the data themselves. The goal is to reach a saturation point where additional analysis does not lead to any new emerging themes (see Meagher, Ch. 12). Comparative analysis Comparative analysis is closely linked to thematic analysis. It consists of comparing and contrasting data from different sources to identify all the themes and issues in question. The goal is to reach a saturation point where no new issues arise from the material. Thematic and comparative analyses are often used jointly or alternately in qualitative research projects, as they are often mutually complementary (see Meagher, Ch. 12; Okano, Ch. 3).

302

Chapter 11 How to analyse data Content analysis The basic goal of content analysis is to code by content. Researchers use a list of categories or labels to quantitatively identify content and patterns and to interpret them qualitatively. This list of categories may have emerged from the data itself, but may also be set up in advance, depending on the theoretical approach or state of research. Generally, the content analysis is often more deductive than the two aforementioned inductive approaches to thematic and comparative analysis. However, even in highly deductive research projects, researchers should not exclude in advance the possibility that some new categories may be identified in the sample during the data analysis itself (see Arrington, Ch. 13). Discourse analysis The focus of discourse analysis is on patterns of communication. It includes a wide range of approaches. Conversational analysis, based on the conceptual works of Harold Garfinkel (1967) and Erving Goffman (1974), focuses on patterns of informal daily conversation and interaction. Studies in line with Michel Foucault (1969) search on a much more abstract macro level for institutionalised patterns of ideas and knowledge that guide society. The goal of discourse analysis is to identify the rules and regularities for communication and how meaning is conveyed through it (see Eder-Ramsauer/Reiher, Ch. 14). These approaches are located on a continuum from highly qualitative to almost quantitative approaches (Dawson 2002, p. 115). Thematic and comparative analysis are highly qualitative approaches. As a result, data analysis normally takes place in these approaches throughout the entire data collection process. Content analysis can be much more quantitatively oriented. Therefore, the entire research process is much more consistent with the clear-cut sequence template described above, with the (bulk of) data analysis often taking place after the data collection has been completed. Discourse analysis is generally located somewhere in the middle of these two pools of the continuum (Dawson 2002, p. 115). Which one of these approaches a researcher chooses for a research project depends on her research interests and research questions as well as the project’s theoretical (and philosophical) foundations. When deciding for an approach to data analysis, the researcher should keep in mind the field of research that she would like to contribute to through her project.

4. Steps in the analysis: From data to theories Concrete analytical work with data and its subsequent steps vary greatly depending on the research question or the analytical approach in qualitative data analysis. The general strategy, however, often consists of four steps.

303

David Chiavacci Four steps in data analysis 1. 2. 3. 4.

Transcribing and exploring the data, summarising and identifying themes in the data, developing codes and categories for the data, and finally identifying conceptual patterns and building theoretical models based on the data.

4.1 Transcribing and exploring the data In the initial step, the data is transcribed from the recording of an interview or field notes to an annotated written text. The extent and degree of detail of this transcription is again a matter of the research question. While it is important to bring all data into a form in order to be able to compare different recordings of interviews or notes from participatory observations, the transcription level should also be guided by pragmatism: do the transcription in as detailed and exact a way as necessary. To go back to the two examples mentioned above: recorded interviews with political actors about the policy process are rarely analysed with regard to the interviewees’ emotional feelings but are mainly used as a source to better understand their political role and their influence on the dynamics and chronology of the policymaking process. Thus, the transcription of recorded interviews is often only partially word-for-word and needs limited annotation. However, a study on the behaviour and emotional experience of customers in convenience stores will most likely require very detailed and accurate transcription with comprehensive annotation in order to provide a basis for analysing the relationship between behaviour and the emotional experiences of consumers. In this first step, the data is in its raw and transcribed form and should be explored through careful studying and reading (and restudying and rereading). In cases where, for example, an audio or video recording of the interviews or real-life interactions is not possible this exploration becomes even more important. It should be done as soon as possible after the fieldwork. Keep in mind that the ability to check, review and complement one’s field notes is much better when the fieldwork is still fresh in your memory.

4.2 Summarising and identifying the first themes The second step in working with the data is to summarise it and to identify the first themes in it. Summarising data is an important step as it enables us to become more acquainted with the material and to work out its core content and important points. This also includes the identification of recurrent and central themes in the data. Once themes have been identified, we can go back to the data to compare different interviews or observations on these themes. In semistructured, guided interviews, most themes will have already been identified through a guideline developed prior to the interviews. However, in more open formats like narrative interviews, themes will almost entirely be derived from the data, and the pool of themes will be identified in parallel with the data collection.

304

Chapter 11 How to analyse data

4.3 Coding The third step is then to develop codes through coding (see Meagher, Ch. 12). The researcher develops codes and later on (theoretical) concepts by naming, organising and refining the first content-related themes. This coding process is still tentative in the early interviews or observations of highly qualitative research projects and is undergoing a process of revision and refinement. However, in the analysis of later interviews and observations, the coding process becomes increasingly saturated when no more new codes emerge. At this point of (theoretical) saturation, the codes can be finalised into a codebook. This is especially recommended when you are working in a team. Ideally, this is also the moment in research when data collection can soon be concluded as new interviews or participatory observations will not lead to new important analytical insights. Even in research projects with semi-structured interviews or content analysis, researchers are strongly recommended to consciously develop codes and invest time in doing so. Even in a highly deductive content analysis research project, a good codebook with revised and refined codes should be developed on the basis of a pretest of first codes (or concepts) that most likely derived from a review of research literature.

4.4 Identifying patterns and structures In the fourth and final step, the analysis consists of the identification of conceptual patterns and the construction of theoretical models. This theoretical construction is a major objective of qualitative research. In this step, the researcher links codes and categories and builds hierarchies between them in order to build theoretical knowledge. The objective is to develop explanatory and coherent models for the patterns of codes identified in the data. The bestknown approach to constructing middle-range theories from qualitative data is grounded theory, as developed by Barney G. Glaser and Anselm L. Strauss (1967) in their seminal monograph The discovery of grounded theory: Strategies for qualitative research. The authors proposed that ‘qualitative inquiry could make significant theoretical and empirical contributions in its own right, rather than merely serving as a precursor to quantitative research’ (Charmaz 2007, p. 2023). In so doing, Glaser and Strauss made a strong case for theoretical ambitions in qualitative research.2

5. Computer-assisted qualitative data analysis software (CAQDAS): Pros and cons More and more researchers have recently been using software to facilitate data analysis. Software programs that assist researchers in qualitative research date back to the 1970s. Today, they are well established but still highly contested in qualitative data analysis. To some degree, 2 The discovery of grounded theory has become a foundational text with an immense influence on qualitative Social Science research. However, nowadays we have to speak of several grounded theory approaches, since Glaser and Straus parted ways and revised the original grounded theory separately. This led to the emergence of ‘grounded theories’ by Glaser, Strauss, colleagues, collaborators and students (Charmaz 2007).

305

David Chiavacci a generational gap in the perception and use of CAQDAS can be identified. Quite a number of older researchers have renounced CAQDAS as some of them even see a fundamental contradiction between qualitative research and the use of CAQDAS. Carol Grbich (2007, p. 234), for example, speaks of ‘[t]he tyranny of a system, however useful, which has the capacity to direct and simplify the construction of the views of researchers and ultimately those of readers [and] will thus always be problematic.’ However, many younger researchers have often learned about CAQDAS in their introductory courses to qualitative methods and usually emphasise the advantages of using them. When researchers are using CAQDAS, it is important for them to be aware of its main advantages and disadvantages. Certainly, CAQDAS provides valuable support in routine tasks in the entire data collection and analysis process. It helps to organise, manage and analyse data and materials. Experienced users of CAQDAS can save a lot of time in the research process by using it. Recent software programs can even assist researchers not only in identifying themes and developing categories but also in gaining conceptual insights and constructing theoretical models during data analysis. In addition, CAQDAS opens up completely new opportunities for data analysis. For example, some programs allow researchers to produce graphs of networks based on the data, which could not be done without them. Or content analysis of a large amount of data from a wide number of sources is only possible with CAQDAS (see this chapter, Heckel, Ch. 11.3). The main argument against using CAQDAS is that it entices researchers to focus more on the software and to follow its lead instead of studying and analysing their qualitative data and material in-depth and intensively. By becoming a kind of filter between researcher and data, CAQDAS poses the risk of superfluous qualitative data analysis, which would clearly contradict the fundamental principles of all qualitative analysis approaches. Moreover, using CAQDAS also means that a significant amount of time and energy must be spent on becoming an experienced user, which will not be available for fieldwork and data analysis (see Wiemann, Ch. 13.1). Overall, it is believed that the pros prevail against the cons when using software in qualitative data analysis. However, the decision to use or not to use CAQDAS should depend on its potential advantages over the efforts needed to work with it in view of your specific research question and project. Another problem is how to decide which software package is best suited for your research project. Increased qualitative research in recent years has also led to the development and refinement of a large number of software programs which support data collection and analysis. The CAQDAS entry in Wikipedia alone lists nearly 20 software packages.3 Most of these programs have some special features and functions, which makes it difficult and time-consuming to find the most appropriate software (for a recent overview, see SánchezGómez et al. 2019). A pragmatic strategy in deciding on the software you will use is to find out which packages are offered by your university or research group as well as whether introductory courses and further support are offered in order to get the most out of CAQDAS in data analysis. Finally, please also note that not all software packages are capable of supporting Japanese characters. However, there are still good programs available today that cover most writing and language systems, such as CATMA (Computer Assisted Textual Markup and Analysis).4 Finally, re3 See https://en.wikipedia.org/wiki/Computer-assisted_qualitative_data_analysis_software [as of July 2020]. 4 See https://catma.de/.

306

Chapter 11 How to analyse data searchers must be mindful that the use of CAQDAS should never be a shortcut and cannot replace intellectual engagement with their data and material, which brings us to the next section on reflexivity.

6. Ensuring the quality of data analysis: Reflexivity, validity and reliability As described above, the loops between data collection and data analysis offer a huge opportunity for qualitative research, which also allows researchers to continually reflect on their own position throughout the entire research process. It is only during data analysis that we can become fully aware of our position as researchers, especially on issues of power and bias. Data analysis involves not only analysing data but also thinking about possible traps and engaging in ethical questions in the research process (see this chapter, Hülsmann, Ch. 11.1). The metaphor of making a cake for data analysis that was used in the introduction does not really fit in this context. Social scientists do not work with death material, but with human beings, which raises all kinds of ethical questions. To mention just one example of ethical problems that may arise during the research process: non-Japanese researchers are often by nature outsiders when doing qualitative fieldwork in Japan. Depending on the research field and question, this can be helpful in opening up new opportunities for data collection, but it could also be a problem that may lead to bias. A general solution to this problem does not exist since, depending on the individual project, being an outsider may have different effects on the research process. But in any case, it is important that researchers reflect on this issue and address it in the most transparent way possible. In addition, reflexivity in qualitative data analysis is also of the utmost importance for the quality of one’s research results. The problem, however, is that researchers do not agree on the criteria for quality in qualitative research, but are divided into several camps (Steinke 2000, pp. 320–321). Validity and reliability, for example, are regarded as being important and meaningful objectives in qualitative research by some researchers. However, others denounce these concepts as being imposed on qualitative research by dominant standards of quantitative research. Some postmodern approaches even argue that quality criteria in qualitative research are impossible. Still, in line with Patricia Bazeley (2013, pp. 401–421) and David Silverman (2000, pp. 268–291), we can identify a few central strategies to ensure the quality of research findings most scholars agree on.

307

David Chiavacci

•

•

•

•

•

Strategies to ensure the quality of your research Even in a highly qualitative research project, researchers should try to develop descriptive quantitative statistics during data analysis in order to give their readers a plausible overview of their entire material. Deviant cases in their data, which contradict the conceptual explanations and theoretical models developed in qualitative data analysis, should be carefully analysed and included in the publication of research results. Deviant cases often help to strengthen and clarify the theoretical models by making their limits transparent. Researchers should, if possible, analyse data collected through different methods and from different sources. In addition, they should also check their findings with complementary data. Researchers should compare their findings with those of other studies and/or predictions made by alternative theories. This helps them to put the findings into a larger context and see the bigger picture after being trapped in their own qualitative data analysis. Finally, researchers should not be insecure. Of course, their analysis is not the only meaningful, coherent and relevant analysis of the data and material used since the analysis highly depends on the researcher. But, despite the often used and completely objective neutral language used in quantitative research, qualititative analysis of the same dataset also produces different and sometimes even completely contradictory findings.5

7. Data analysis with Japanese language material For non-native Japanese speakers, even after years of studying the Japanese language and being generally fluent speakers and readers, qualitative analysis of Japanese language material can be a very thorny issue. Particularly if the research project consists of qualitative interviews with completely open questions or participatory observations and the aim is to carry out an in-depth analysis, including emotional aspects, every non-native Japanese speaker will reach the limits of his or her interpretative capabilities. This is nothing to be ashamed of, but it has to be recognised. It is a myth that Japanese is the most difficult language in the world, but a non-native speaker is a non-native speaker. Researchers should also bear in mind that the spoken Japanese language may seem easier to understand than the apparently more complex written version, but many nuances of the spoken language are never explained in language teaching and not found in dictionaries or grammar books.6

5 See, for example, a study about racism in football, in which 29 teams of experienced quantitative researchers reached different findings despite having the same research question and analysing the same dataset (Silberzahn et al. 2018). 6 My most impressive Japanese language class was a very advanced Japanese language course at the University of Tokyo that I took when I was a young postdoctoral researcher. The instructor explained many subtle nuances in a recorded debate, which I and all my classmates had completely missed despite years of Japanese language training.

308

Chapter 11 How to analyse data The best way to address these issues in such cases is to involve native speakers in your data analysis. Discuss your interpretation, the data section you struggle with or your (preliminary) categories and interpretation of your material with Japanese collaborators (more than one if possible). Actually, collaboration in qualitative data analysis, in which two or more researchers work on the same data and continuously discuss their findings, is a strategy that is often used to improve quality in qualitative research. Therefore, by comparing your views with the interpretation of native speakers, you will not only get an insight into the nuances of the Japanese language but also increase the quality of your research (see Kottmann/Reiher, Ch. 7). Nevertheless, these language issues do not in any way imply that non-native speakers should refrain from qualitative research about Japan. Misunderstandings are common in all languages, even among native speakers. Therefore, native speakers are not immune to misinterpretations in qualitative data analysis. Non-native speakers may sometimes be even more prepared for qualitative data analysis, as they will struggle with certain passages and have to work harder to understand the material, which can prevent superfluous analysis.

8. Concluding remarks Now that I have written a few points on the most important aspects of qualitative data analysis, a few general comments may be appropriate as concluding remarks. What are the most important strategies in qualitative data analysis? In my opinion, the two most important ones are the following. First, qualitative data analysis is not something that a researcher can learn from reading a handbook or a short introduction like this chapter. You have to practise it! Thus, try to obtain as much training as possible by doing the exercises in handbooks, attending methodology courses at research institutions, but also reading methodological publications. Wolff-Michael Roth (2015), for example, published a book in which five instructors of qualitative methods code and interpret the same transcripts, allowing even experienced researchers the opportunity to look over the shoulders of other researchers and see how they do it. Second, never forget the truly wise words of Robert E. Stake (1995, p. 19): ‘Good research is not about good methods as much as it is about good thinking.’ Sound knowledge and experience of methods provide an array of useful and proven tools, but at the end of the day in qualitative data analysis like all research steps, there is only one important analytical tool— your mind.

309

11.1 Negotiating the ethics of gathering research data in a subcultural context Katharina Hülsmann

Like many scholars from my generation, I became interested in Japanese Studies through my interest in Japanese popular culture, such as manga. I was always interested in the way that cultural flows spread with little regard for national borders. The first time that I visited Tokyo Big Sight in 2012 for an event geared towards women at which artists could exchange their fan works (dōjinshi) with fellow fans; I was able to—seemingly—connect seamlessly with and chat to artists that were producing fan work for the manga and anime I was interested in back then, which made a big impression on me. Even though I looked different from everybody else, a pin on my chest suggested that I was a fellow fan and immediately we found things to discuss. On the other hand, while a lot of international fan work is shared freely on social media sites on the internet, Japanese fan work seemed strangely reclusive, confined to online platforms like Pixiv and personal homepages that could often only be accessed with passwords. It gave me pause that I was able to have this personal experience of immediately feeling welcome even though my difference was obvious, but that, as a whole, dōjinshi culture seemed a bit hidden and closed off, not least because the dominant form of the narrative fan work is still the printed zine. This discrepancy stuck with me during my master’s degree, as I voraciously consumed academic literature from the Anglophone field of Fan Studies. The number of academic works that examined the existence of homoerotic narratives in Japan (yaoi or boys’ love) also grew to the point that keeping tabs on all the articles coming out was difficult. After finishing my MA degree however, I realised that I was less interested in analysing the content of these works. The question of why ‘these women’ write the often erotic works and what that says about their ideas of gender, feminism, sexuality, etc. was, in my opinion, 1. difficult to answer and 2. had been answered too often already. I grew slightly frustrated with the exemplary analyses of fan works (knowing full well that I was guilty of having done this sort of analysis in the past) and more interested in the agency of the authors as participants in an economy of cultural goods. I wanted to find out to what extent dōjinshi culture could be conceptualised as a transcultural phenomenon. Finally, I decided that asking the authors themselves to talk about broader trends in dōjinshi culture would offer more insight than just analysing exemplary works or relying solely on quantitative data about participation in dōjinshi events. The study of fan and subcultures is appealing because it often allows the researcher to deploy expertise and skills to obtain research materials they previously accessed through non-academic means (see Vogt, Ch. 2). You can save time if you already know how to operate the search system on a fan site because you have previously used it privately. Thus, it is not surprising that some academic examinations of fan work have drawn criticism from fans themselves because they feel their works and practices have been misread and exposed without their con-

310

Chapter 11 How to analyse data sent. It is part of your job as a researcher to examine your own position and point of view. Are you dragging something out into the open that you only have knowledge of because you were —more or less—a participant yourself? You have to remind yourself that you, as a researcher, are benefitting professionally (and maybe financially) from the fact that you have insider knowledge of a cultural phenomenon (Nielsen 2016, p. 237). You have to become very aware of your stance as an academic and of your stance as a participant in the cultural phenomenon that you plan to examine. You also have to ask yourself: Are you analysing texts or are you analysing people? And if you are analysing people, then ethical guidelines apply. Examining the visibility and spreadability of Japanese fan work for my PhD research made me intimately question my own approach as a researcher and what material was even okay for me to use. Many of my senior colleagues, whose advice I sought, came from the field of Literary Studies and thus the idea of asking for authors’ consent to examine their works seemed ludicrous to them. When I started my research into dōjinshi culture in Tokyo in spring 2017, the so-called Ritsumeikan/Pixiv incident happened. Three researchers from Ritsumeikan University had presented a paper at the 31st Conference of the Society for Artificial Intelligence in Nagoya. In this paper, they used written Japanese fan works from the platform Pixiv as a text corpus to develop a text-filtering software for obscene and harmful content. The paper itself supposed, for example, that the desired purpose of the software would be to detect sexual expressions, even if they were expressed through metaphors. Thus, the software should be able to detect precisely when the word banana is used to describe a penis and censor the word banana accordingly because it is a harmful expression in this context. At no point does the paper discuss exactly why sexual metaphors are harmful, but the authors do establish the fact that ‘[t]he internet is overflowing with harmful content’ (Omi et al. 2017, p. 1) in the very first line of their paper and they used the fan works as an example of such harmful content. A big discussion ensued in the Japanese Twittersphere on whether the right to academic citation trumps the need for informed consent. The paper was taken offline but remains accessible through Internet archiving services. At the time, I had not yet been able to secure any interview partners and I felt rather alarmed at this incident and the consequences that it might have on my ability to secure interview partners in the future. While it was, of course, possible for me to obtain fan works at dōjinshi exchange events and specialised bookstores, it did not feel right for me to consider these materials fair game for my analysis. I decided that it would be of the utmost importance to obtain the informed consent of the cultural agents that I was going to be researching and that it would be safest to produce the material for my analysis myself through the interviews with the authors. I selected my interview partners by reviewing my field notes from my participant observation of the events and the quantitative data that I had drawn from event catalogues recording their (and their circles) attendance figures. Thus, I was able to see which were the most popular and fastest growing genres during my period of research. When data is ‘out there’, it is sometimes impossible to ask for consent or very improbable that you, as a researcher, will receive consent from the author, especially if you are analysing phenomena such as racism, sexism or other problematic issues. However, you have to ask yourself: Are you adequately equipped to be researching your subject? What will the consequences for the authors of the content be, especially if they are operating within a legal grey area? Are you going to protect their identities and how far will you go to ensure anonymity? There are phenomena that are important to research, but you have to ask yourself, especially as a young researcher with little institutional security, if you are able to accurately assess the risk that 311

Katharina Hülsmann your research poses to those it is conducted on, if you are able to follow the ethical guidelines of sociological research, and how you will be able to deal with the consequences in case something unexpected happens. Last but not least, I cannot stress enough that being comfortable with your position as a researcher is important as well, especially when conceptualising your PhD project. This research will be your defining feature as an academic for the coming years and will define your job opportunities. The Ritsumeikan/Pixiv incident helped cement my decision that I would use Jean-Claude Kaufmann’s method of the comprehensive interview to approach my set of questions about Japanese dōjinshi culture. It would allow me to grasp the cultural phenomenon that I was examining in all its complexity and, at the same time, allow me as a researcher to bring my knowledge and expertise into the interviews. Kaufmann encourages the researcher to bring themselves into the interview situation instead of trying to be ‘sterile’ (Kaufmann 1999, p. 25). This proved to be a very valuable approach to examining a marginalised fan practice that takes place in a legal grey area. My interview partners would test my own credentials when it came to fan culture and actively asked for participation in the interviews. Had I been prevented from providing these credentials, that is, my own involvement in the form of attending events, consuming media, etc. I am quite certain I would have received only very superficial and unpersonal replies about my interview partners’ involvement in fan culture. As Kaufmann argues, an impersonal interviewer will yield unpersonal answers. In my case, being perceived as somebody that understands fan culture and does not aim to expose the participants helped create a space of trust, so that my interview partners would paradoxically extensively speak to me about the ‘hidden’ and ‘secret’ culture they participate in. I have come to the conclusion that this secrecy is partly ostentatious but an important constitutive practice of the subculture that I examined. Beyond talks with my academic advisors, what helped me most in deciding my approach to the research questions I posed was attending international conferences and speaking to many different people from different fields and disciplines. Japanese Studies does not necessarily come with its own treasure chest of research methods and approaches, and it is up to you as a researcher to poach from any academic disciplines that you may come into contact with. It is advisable to connect early on with fellow researchers and keep an eye on what happens in the field. Any mistake others make is something you can observe and learn from. This networking is not just useful during the conceptional phase of your research project. While I was conducting a qualitative analysis of the interviews, it also helped me immensely to present the project in different contexts—to researchers concerned with Japanese popular culture, with Gender Studies, with Comic Studies, with Fan Studies, with Digital Humanities—to draw inspiration from their views on the topic and, occasionally, to confront them with excerpts of the interviews and discuss my interpretations. Kaufmann, a scholar 40 years my senior, advises the researcher to use analogue devices such as the trusted filing card to work towards a theory. My colleagues advised me to use digital tools such as Citavi to sort through my analysis notes. What helped me more, especially in my making the shift from empathetic interviewer to objective researcher, as Kaufmann describes it, was presenting my temporary findings to diverse audiences and gathering from their critique whether I was considering all facets of the meanings that were communicated in the interviews.

312

11.2 Researching sex and the sexuality of Japanese teenagers: The intricacies of condom use Genaro Castro-Vázquez

This essay draws on part of my experience of conducting a research project for my PhD dissertation. First, I discuss how I became interested in an investigation of sex and sexuality among teenagers and why I felt attracted to the Japanese case. I present why and how I moved from a functionalist quantitative perspective to a qualitative one that would help me better understand condom use as a gendered matter. Finally, I succinctly introduce the theoretical and methodological approach that I employed in this study on a group of young Japanese men. My first encounter with research related to sex and sexuality dates back to the time when I worked as a counsellor at a senior high school attached to the National University in Mexico. It was a fruitful but disheartening experience. In a country where abortion remained illegal and teenage pregnancies a rampant social and medical problem, teaching about how to efficiently and judiciously choose contraceptive methods—condoms in particular—entailed one of my main duties. With a strong emphasis on empiricism and the functionalist tradition, I firmly believed that rational thinking and objective measurement were necessary to underpin any pedagogical action. I thus endorsed the Health Belief Model—largely suggesting that perceived benefits trigger healthier and positive behaviour—to help students reasonably calculate the risks and benefits of using contraception and/or postponing intercourse (Rosenstock et al. 1988). No matter how hard I tried, however, my pedagogics appeared rather ineffective. Many of my female students ended up facing the difficult conundrum of an unplanned pregnancy or unsafe, illegal abortion. I did not know the theories and methodologies to explain it, but found it extremely strange that female students had to face unwanted pregnancies almost alone. In looking for alternative teaching methods, I was drawn to Japan as it is the country that boasts the highest condom use rates in the world. I assumed that this was due to its top-notch school-based sex education. Little did I suspect, nonetheless, that the Japanese situation was rather similar to the Mexican one. Condom use rates were largely associated with the unavailability of birth control pills, and the ineffectiveness of sex education was reflected by the number of women who might have recourse to legal abortion as a ‘contraceptive method’ (Norgren 2001). Furthermore, Japan has one of the largest pornographic and prostitution industries, erotic art expressed in a woodblock print format known as shunga—produced during the Edo period (1603–1867), festivals such as hōnen matsuri—one of the fertility festivals where copious visual displays of handmade statues of phalluses can be seen, and the supposed oriental knowledge of sensual pleasure pervading Japanese culture—referred to as ‘ars erotica’ (Foucault 1990). Nevertheless, a daily conversation about sex and sexuality entailed a great taboo (Castro-Vázquez 2015). As one of the teenagers that I interviewed as part of my doctoral dissertation suggested:

313

Genaro Castro-Vázquez [S]ex education related topics [are] difficult to deal with in school and daily conversations […] because of the meaning embedded in the Chinese character shade (kage). […] Japanese people, teachers in particular, do not feel at ease talking about sex-related issues. The character is formally used in words to refer to most sex-related vocabulary (CastroVázquez 2007, p. 115f.). Thus young people are most likely left in the dark, unable to engage in conversations about sex with teachers and parents, and mostly depending on pornography, the mass media and peers as a source of information. In trying to produce a research proposal for my PhD dissertation, I realised another similarity between my experience in Mexico and the Japanese case: functionalism was the tenet of most academic reports on sex and sexuality in Japan, which were basically surveys that could be referred to by the construct of the ‘Taylorization of sex […] or the production of rationalized means of producing pleasure [expressed in] a number of stages to be gone through before the final output: foreplay leading to coitus culminating in orgasm’ (Jackson/Scott 2010, p. 61–62). Within this framework, contraception use was considered one of these precisely calculated stages (ibid.). This viewpoint could be clearly seen in the biannual reports that the Japan Family Planning Association produces to offer an ‘accurate’ panorama of the sexual behaviour of the Japanese (Japan Family Planning Association 2017), where, for instance, details of the frequency and number of contraceptive means used are offered, but information concerning the social processes involving decision-making can hardly be obtained. The reports largely allow a description of behaviour but not really an explanation that would help understand whether negotiations underpinning the use of contraception took place and/or why young women most likely experience the consequences of unplanned pregnancies alone. I intuited a gendered view of contraception decision-making processes—wearing condoms or not was a man’s decision, but the consequences of not using them a woman’s concern—and was very keen on conducting an ethnographic study involving interviews and participant observation. Nevertheless, my university would not approve such a research plan as it was deemed largely non-scientific. I thus had to offer statistical evidence that would prove that gender influenced condom use among Japanese teenagers. Grounded in the gender schema theory (Bem 1981), through an attitude scale and a semantic differential, I managed to show statistically that a traditional male gender role was detrimental to condom use negotiation and connected to an unfavourable perception of condoms. Methodologically, the difficult part was to produce a valid instrument, but data analysis was rather easy; it was basically done through the SPSS computer programme (Cronk 2017), which in the end would help to accept or reject the hypotheses. This ‘science of contraception’ made my academic adviser satisfied as we were apparently speaking the same language. In line with Raewyn Connell (1987, p. 51), the gender schema theory, however, reduces masculinity and femininity to a simple form of dualism and sets a normative standard that hardly serves to comprehend ‘the way things usually happen’. I was unable to explain how gender might influence condom use in ‘reality’. Statistics and figures, however, helped me open the door. I was permitted to continue with an ethnographic study where I could interview Japanese teenagers and observe sex education classes. I was particularly interested in finding out how condoms were an element of teenagers’ gender relationships. Interviewing male senior high school students was insightful and absolutely fascinating. I discovered the gendered intricacies of contraception, through which the significance of language 314

Chapter 11 How to analyse data and sexual reputations became apparent. Starting the conversations with them was almost always challenging due to a lack of agreed words to refer to sex, sexuality and condoms in general. Most of the available Japanese terms were either too technical or too ‘vulgar’—dirty jokes and double entendres (shimoneta)—to have them included in a ‘normal’ daily conversation. The language of sex mostly entailed silences, omissions, similes and innuendo. An open discussion of condom use elicited sexual expertise, which largely jeopardised the sexual reputation of the person involved—a young man willing to talk about sex would likely appear as an ‘unreliable’ guy who sleeps around (karui yatsu). In dealing with these issues, I took the interview transcripts to be the interviewees’ accounts of their experiences of dealing with condoms, rather than considering their words to directly represent reality. Interview transcripts are not portraits of reality, but pieces of the participants’ life histories (Plummer 1995). When the young men felt distress and at risk of losing their reputation, my strategy was to allow them to decide how far they wanted to talk about their knowledge and experiences, and I proposed sources of information and support when it appeared appropriate. Transcripts of my interview and participant observation notes entailed the main raw data of this part of the investigation. Given my teaching experience, the statistical evidence and a framework of feminist sociological theory (Ramazanoglu/Holland 2002), I assumed that condom use decision-making was both unequal and gendered, and although my reading of the data significantly buttressed this, it did not necessarily support the young men’s viewpoints, who tended to endorse a discourse of essentialist gender difference: men are naturally protectors and women are to be protected (mamoru and mamorareru ningen). As such, my main conclusions arose from my attempts to explain contradictions, the unexpected, and absences and silences in the interviews, because in different sectors of the same conversation the young men turned to and answered to differing versions of masculinity and femininity and presented their gendered identities accordingly. I did respect the informants’ recollections as what they wanted to convey, and presented verbatim excerpts of the conversations when I reported on the study, but I did not think of the conversations as an open window to the ‘truths’ of the social world. I rather considered the outcomes of the interviews as resulting from specific circumstances strongly determined by a social interaction between a researcher and an interviewee, and was particularly interested in delving into how gender relationships were presented in the experiences reported, how the young men were able to become both subject and object, how they placed themselves in relation to the other, and how these relationships made sense to me. In contrast to how I dealt with the statistical analysis, I was slightly daunted and mostly overwhelmed when trying to systematise and analyse the huge amount of information generated. Transcribing was the first step of the interview analysis. Transcribing interviews encompassed a tedious and time-consuming process; a one-hour interview turned into an approximately 30 page document. Although I noted down every word included in the conversations, along with interruptions, pauses and hesitations, I did not count them up. This was not a quantitative analysis. The transcripts together with my observation notes served to propose a number of provisional concepts with which to analyse data, and the software Ethnograph helped me set a number of codes that I included in a map of ‘systemic networks’, which entailed a graphic representation of codes and subsidiary codes affixed to the transcripts (Bliss et al. 1983). A network, which preserves and shows the information as it was presented in the conversations, entails a practical and analytical device to sort out and categorise data efficiently. The entire process was an attempt at theorising condom use in the context of everyday life and ‘wider pat-

315

Genaro Castro-Vázquez terns of sociality’ (Jackson/Scott 2010, p. 1). The networks helped me to translate the conversations with the young men into a language of theorising, and this interactive process involved both induction and deduction. As the whole investigation was conducted by a single researcher, and every participant was interviewed twice, comparative analysis thus largely circulated around detecting inconsistencies, consistencies and correlations between ‘intra-data’—contrasting the outcomes of the two interviews with each young man—and ‘extra-data’—contrasting the outcomes of interview with each participant with the rest of the group. In the reports on my investigation, I always made explicit and justified how data had been analysed and provided evidence by including verbatim interview excerpts to help the reader comprehend my assumptions on theory and methodology as well as the origin of my conclusions. This largely served to strengthen the quality of my analysis and allowed me to respond to issues of validity. A researcher is obliged to provide ‘detailed descriptions of the path of their research and decision-making processes so that the reader can inspect’ and judge the study’s dependability, consistency and accuracy (Sparkes/Smith 2014, p. 181).

316

11.3 Studying economic discourse Markus Heckel

Introduction Economic theory has created a framework for understanding the functioning of the economy, formed a structure for economic policymaking and even shaped our reality. Usually economists are expected to offer clear statements regarding the condition and direction of the economy. However, Harry Truman, U.S. president 1945–53, is said to have stated, ‘All my economists say, “on the one hand ... on the other.” Give me a one-handed economist!’ (Fadiman/Bernard 2000, p. 542). This quote hints at the fact that there is no one truth in Economics. Economics can be viewed as a construct that changes frequently and adapts to new phenomena and new research insights. As a result, within the context of economic policy recommendations, one finds differing and sometimes totally contradictory statements in Economics, especially in the area of fiscal and monetary policy. The role of language in economic discourse is very important. Indeed, stressing the role of language, Leslie Armour (1997, p. 1062) argues that Economics is ‘largely made up of language and belief’. In fact, John Maynard Keynes, the famous British economist whose ideas played a fundamental role in macroeconomic theory, tried to find a plain language to describe the economy. Specifically, Keynes developed a pedagogic habitus and communicated in the language of monetary theory, which laid the foundations for meaningful public discourse about the economy (Holmes 2013, p. 32). Benjamin Braun (2014, p. 48) even argues that economic discourse is so powerful that it ‘contributes to the cycles of boom and bust’ in the economy. According to Warren J. Samuels (1990, pp. 1–6), economic discourse can be viewed from different perspectives: knowledge or ‘truth’ (economic theory), discourse (here in the sense of language and ‘rhetoric’) and the identification of meaning. The central idea is that Economics, just as in any field, uses language and that the words used create a certain meaning which can then be studied. Below are two examples that show how economic discourse can be analysed: one example focuses on the issue of the well-being of employees in East Asia and the other on monetary policy in Japan.

Employee well-being in China and Japan Under the umbrella of the Volkswagen Foundation’s research initiative ‘Key Issues for Research and Society’, a group of scholars from Goethe University in Frankfurt launched a project in 2014 called ‘Protecting the Weak—Entangled Processes of Framing, Mobilisation and Institutionalisation in East Asia’. Together with Stefan Hüppe-Moon (Sinology and Political Sciences) and Na Zou (Economics), we used media content to compare employee well-being measures instituted by the governments of China and Japan. One important source for media analysis was newspapers. Japan scholars (in Germany) can make use of the CrossAsia

317

Markus Heckel database (see Peucker/Schmidtpott/Wagner, Ch. 9). This rich source allows access to many daily Japanese newspapers (e.g. Asahi Shimbun, Mainichi Shimbun, Nikkei Shimbun). For this project, we conducted a media content analysis by accessing a Chinese (The People’s Daily) and a Japanese (Asahi Shimbun) daily newspaper through the CrossAsia database. By collecting data from both newspapers for the time period 1995 to 2014, we analysed how the Chinese and Japanese media report on governmental policies for protecting employee well-being, specifically focusing on migrant workers in China and regular employees in Japan. We used the headline approach, i.e. we focused on articles in which keywords such as ‘hukou/huji reform’ (household registration system) in China and ‘work-life balance’ in Japan appeared in the headline of the article. This method was employed in order to extract the most relevant articles, as a headline conveys an article’s relative importance (Dor 2003) and also helps filter out less pertinent news articles. Another possible approach would have been to conduct a quantitative analysis of the whole sample of articles, e.g. through a text-mining analysis. One challenge of this kind of analysis is how to compare two very different countries such as China and Japan. As the economic development and institutional design of both countries are quite distinct, the subject of ‘employee well-being’ requires country-specific measures. It is for this reason that we focused on migrant workers in China and on regular employees in Japan. Furthermore, a detailed knowledge of the language (e.g. with the help of native speakers) in order to understand the nuances of the articles was a necessary requirement. Our results showed that the discourse of ‘hukou/huji reform’ and ‘work-life balance’ is linked to issues that go beyond solely the subject of employee well-being. For China, increasing domestic consumption by improving the general situation for migrant workers is tethered to the larger question of how to reorient China’s economic growth model. The results for Japan showed that the government was well aware of problems for regular workers regarding work-life balance, such as overwork and burnout. As a consequence, the government implemented some family-friendly measures (e.g. the Charter for Work-Life Balance in 2007). However, the intent was not necessarily to improve conditions for individual workers, but rather to tackle problems related to Japan’s declining birth rate and ageing society (Heckel et al. 2018).

Monetary policy My project on Economic Discourses of Monetary Policy focuses on central banks, especially the Bank of Japan (BoJ), central bank communication and inflation/deflation. Given Japan’s prolonged struggle with deflation and low inflation, the BoJ is a good case study for discourses in monetary policy. Since the start of the financial crisis in 2008, financial stability has been precarious and monetary policy has been high on the political agenda. The BoJ is often the first, or one of the first, to try unprecedented monetary policy tools. A zero interest rate policy and quantitative easing were first applied in 1999 and 2001 respectively. After the financial crisis, the monetary arsenal was augmented with quantitative and qualitative monetary easing (QQE; 2013), QQE with a negative interest rate (2016) and QQE with yield curve control (2016). Next to financial analyses of the efficiency of these monetary policies, discourses of communication strategies have become increasingly important as central bank communication impacts financial markets and financial stability (Born et al. 2014).

318

Chapter 11 How to analyse data One strand of relevant papers in the academic literature on this subject reveals a steadily growing number of studies on central banks’ communication. For example, Mikael Apel and Marianna B. Grimaldi (2014) as well as Stephen Hansen and Michael McMahon (2016) analyse the communication of central banks by employing automated content analysis techniques that transform qualitative information to quantitative data. That is to say, in many cases economists analyse texts with different approaches, including computational and statistical methods in a quantitative research design. Kohei Kawamura et al. (2019) are a good example of how to apply this approach to monetary policy in Japan. They analyse the BoJ’s monetary policy by using the BoJ’s Monthly Report of Recent Economic and Financial Developments in the time period between 1998 and 2015. They employ natural language processing techniques, which involves using computer programmes to analyse, process and derive meaning from a large quantity of human language data. This method helps to classify expressions according to polarity (whether an expression is positive, negative or neutral) and modality (whether an expression is clear-cut, ambiguous or subjective). A simple game-theoretic model is applied to understand the empirical observations from the viewpoint of strategic central bank communication. It implies a principal–agent relationship between the BoJ (acting as the agent) and market participants (acting as the principal). This game-theoretic model is a persuasion game in which the sender—the BoJ—uses its information advantage (asymmetric information) towards market participants (receiver) in an attempt to influence inflation expectations in order to achieve the goal of 2% inflation. The authors show that the BoJ has an upward bias in reporting about economic conditions by increasing ambiguity when economic prospects are rather unfavourable. This finding is in line with Born et al. (2014), who show that Financial Stability Reports released by central banks have positive effects on financial markets (in terms of stock market returns) if their views are optimistic but no effects when they are pessimistic. One general problem with these kinds of analyses is the occurrence of multiple similar announcements in official documents, such as minutes of monetary policy meetings, and monthly and annual reports. Consequently, it can be argued that the level of variance is rather low, possibly casting some doubts on statistical results.

Text-mining tools One technique which has gained momentum as an analytical tool in Economics is text mining. Text mining refers to the process of extracting new and relevant information from a large collection of texts by means of automated computations. Typical data generated includes term frequencies, term patterns and term correlations. If text mining is combined with natural language processing techniques, text semantics can be examined as well, which allows for additional insights through, for instance, sentiment analysis and topic modelling such as latent Dirichlet allocation (LDA). LDA is a statistical machine-learning model for collections of data, such as text corpora, and it is used to group words and expressions to a fixed set of topics. Text mining can be performed with the help of software programmes: Python and R are open source programming languages that are commonly used for conducting data analysis. While Python is a general-purpose language with a focus on productivity and easy-to-understand syntax, R is a language (and at the same time an environment) with strengths in user-friendly statistical modelling and data visualisation. Generally, Python usage is more prominent in

319

Markus Heckel fields such as engineering (machine learning and artificial intelligence), whereas R is most commonly used in research focused on data analysis and statistics (employed not only in academia, but also by companies, journalists, etc. who can benefit from the rich visualisation capabilities of R). However, the question of what programme is the best choice depends on what problems need to be solved and what tools are most frequently used in one’s field. Additionally, it is possible to use Python in R and vice versa through the application of special packages. There are many web resources and books that help students ranging from beginners to experts to master Python and R (Bird et al. 2009; Wickham/Grolemund 2020). A tool that uses the Japanese language is the open source software MeCab. MeCab is a morphological analyser used for text segmentation (tokenisation) and it is compatible with different platforms: through a wrapper it can be used with Python, while the TopicExplorer (a webbased system for topic modelling developed by the University of Halle-Wittenberg) has integrated MeCab. Additionally, a user can install and call on the package RMeCab to benefit from the MeCab tool within an R environment.

Conclusion One major challenge for Japan scholars in Economics is the need to acquire additional skills in order to access and analyse large data sets. Especially in the field of Economics, learning statistics and economic theory can be a great burden and very time-consuming. However, once one succeeds in mastering this field, there are many opportunities for research and also job prospects. The analysis of discourse in Economics requires inter alia linguistics, software skills and (where applicable) statistics. In addition to this, detailed knowledge of the institutions and language of the country in question are required. Especially in the case of Economics, many international experts who have an excellent reputation in their field comment on Japan without any specific knowledge about the country. In many cases, standard explanations and policy recommendations are used which very often have a strong Western bias. However, it is often the case that Japan needs different explanations. This is where Japan scholars can play an important role.

320

Further reading Bazeley, Patricia (2013): Qualitative data analysis: Practical strategies. London: Sage. Bernard, H. Russell/Wutich, Amber/Ryan, Gery W. (2017): Analyzing qualitative data: Systematic approaches. Los Angeles, CA: Sage. Flick, Uwe (ed.) (2014): The SAGE handbook of qualitative data analysis. Los Angeles, CA: Sage. Grbich, Carol (2007): Qualitative data analysis: An introduction. London: Sage. Harding, Jamie (2019): Qualitative data analysis: From start to finish. Los Angeles, CA: Sage. Miles, Matthew B./Huberman, A. Michael/Saldaña, Johnny (2019): Qualitative data analysis: A methods sourcebook. Los Angeles, CA: Sage.

References Apel, Mikael/Grimaldi, Marianna B. (2014): How informative are central bank minutes? In: Review of Economics 65, No. 1, pp. 53–76. Armour, Leslie (1997): The logic of economic discourse: Beyond Adam Smith and Karl Marx. In: International Journal of Social Economics 24, No. 10, pp. 1056–1079. Bazeley, Patricia (2013): Qualitative data analysis: Practical strategies. London: Sage. Bernard, H. Russell/Wutich, Amber/Ryan, Gery W. (2017): Analyzing qualitative data: Systematic approaches. Los Angeles, CA: Sage. Bem, Sandra Lipsitz (1981): Gender schema theory: A cognitive account of sex typing. In: Psychological Review 88, No. 4, pp. 354–364. Bird, Steven/Klein, Ewan/Loper, Edward (2009): Natural language processing with Python: Analyzing text with the natural language toolkit. https://www.nltk.org/book/, [Accessed 30 October 2020] Bliss, Joan/Monk, Martin/Ogborn, Jon (eds.) (1983): Qualitative data analysis for educational research: A guide to uses of systemic networks. London: Croom Helm. Born, Benjamin/Ehrmann, Michael/Fratzscher, Marcel (2014): Central bank communication on financial stability. In: The Economic Journal 124, No. 577, pp. 701–734. Braun, Benjamin (2014): Why models matter: The making and unmaking of governability in macroeconomic discourse. In: Journal of Critical Globalisation Studies 7, pp. 48–79. Castro-Vázquez, Genaro (2007): In the shadows: Sexuality, gender and pedagogy among Japanese teenagers. Lanham, MD: Lexington Books. Castro-Vázquez, Genaro (2015): Male circumcision in Japan. New York, NY: Palgrave Macmillan. Charmaz, Kathy (2007): Grounded theory. In: George Ritzer (ed.): The Blackwell encyclopedia of Sociology. Malden, MA: Blackwell, pp. 2023–2027. Connell, Raewyn W. (1987): Gender and power: Society, the person and sexual politics. Stanford, CA: Stanford University Press. Cronk, Brain C. (2017): How to use SPSS®: A step-by-step guide to analysis and interpretation. London: Routledge. Dawson, Catherine (2002): Introduction to research methods: A practical guide for anyone undertaking a research project. Oxford: How To Books. Denzin Norman K./Lincoln, Yvonna S. (eds.) (2005a): The SAGE handbook of qualitative research. Thousand Oaks, CA: Sage. Denzin Norman K./Lincoln, Yvonna S. (2005b): Introduction: The discipline and practice of qualitative research. In: Norman K. Denzin/Yvonna S. Lincoln (eds): The SAGE handbook of qualitative research. Thousand Oaks, CA: Sage, pp. 1–32. Dor, Daniel (2003): On newspaper headlines as relevance optimizers. In: Journal of Pragmatics 35, No. 5, pp. 695–721. Fadiman, Clifton/Bernard, André (2000): Bartlett’s book of anecdotes. Boston, MA: Little, Brown and Company. Flick, Uwe (ed.) (2014a): The SAGE handbook of qualitative data analysis. Los Angeles, CA: Sage. Flick, Uwe (ed.) (2014b): Mapping the field. In: Uwe Flick (ed.): The SAGE handbook of qualitative data analysis. Los Angeles, CA: Sage, pp. 2–18. Flick, Uwe/von Kardoff, Ernst/Steinke, Ines (eds.) (2000): Qualitative Forschung: Ein Handbuch. Hamburg: Rowohlt. Foucault, Michel (1969): L’Archéologie du savoir. Paris: Gallimard.

321

References Foucault, Michael (1990): The history of sexuality: An introduction. Vol. 1. New York, NY: Vintage. Garfinkel, Harold (1967): Studies in ethnomethodology. Englewood Cliffs, NJ: Prentice Hall. Glaser, Barney G./Strauss, Anselm L. (1967): The discovery of grounded theory: Strategies for qualitative research. Chicago, IL: Aldine. Goffman, Erving (1974): Frame analysis: An essay on the organization of experience. London: Harper and Row. Grbich, Carol (2007): Qualitative data analysis: An introduction. London: Sage. Harding, Jamie (2019): Qualitative data analysis: From start to finish. Los Angeles, CA: Sage. Hansen, Stephen/McMahon, Michael (2016): Shocking language: Understanding the macroeconomic effects of central bank communication. In: Journal of International Economics 99, pp. 114–133. Heckel, Markus/Hüppe-Moon, Stefan/Zou, Na (2018): Employee-wellbeing in China and Japan: A media content analysis. In: Amelung, Iwo/Bälz, Moritz/Holbig, Heike/Schumann, Matthias/Storz, Cornelia (eds.): Protecting the weak in East Asia: Entangled processes of framing, mobilisation and institutionalisation. London: Routledge, pp. 166–195. Holmes, Douglas R. (2013): Economy of words: Communicative imperatives in central banks. Chicago, IL: University of Chicago Press. Jackson, Stevi/Scott, Sue (2010): Theorizing sexuality. Berkshire: Open University Press. Japan Family Planning Association (JFPA) (2017): Daihachi-kai danjo no seikatsu to ishiki ni kansuru chōsa hōkokusho 2016-nen: Nihonjin no sei ishiki, kōdō. Tōkyō: Ippanshadanhōjin Nihon Kazoku Keikaku Kyōkai. Kaufmann, Jean-Claude (1999): Das verstehende Interview. Konstanz: UVK. Kawamura, Kohei/Kobashi, Yohei/Shizume, Masato/Ueda, Kozo (2019): Strategic central bank communication: Discourse analysis of the bank of Japan’s Monthly Report. In: Journal of Economic Dynamics and Control 100, pp. 230–250. Kornmeier, Martin (2018): Wissenschaftlich schreiben leicht gemacht: Für Bachelor, Master und Dissertation. Stuttgart: UTB. Miles, Matthew B./Huberman, A. Michael/Saldaña, Johnny (2019): Qualitative data analysis: A methods sourcebook. Los Angeles, CA: Sage. Nielsen, Ej (2016): Dear researcher: Rethinking engagement with fan authors. In: The Journal of Fandom Studies 4, No. 3, pp. 233–249. Norgren, Tiana (2001): Abortion before birth control: The politics of reproduction in postwar Japan. Princeton, NJ: Princeton University Press. Omi, Ryuichi/Nishihara, Yoko/Yamanishi, Ryosuke (2017): Study of harmful expressions filtering focusing on word’s meaning depending on document’s domain and context. Research paper presented at the 31st annual conference of the Japanese Society for Artificial Intelligence, NYP. Plummer, Ken (1995): Telling sexual stories: Power, change, and social worlds. London: Routledge. Ramazanoglu, Caroline/Holland, Janet (2002): Feminist methodology: Challenges and choices. London: Sage. Rosenstock, Irwin M./Strecher, Victor J./Becker, Marshall H (1988): Social learning theory and the health belief model. In: Health Education & Behavior 15, No. 2, pp. 175–183. Roth, Wolff-Michael (ed.) (2015): Rigorous data analysis: Beyond ‘anything goes’. Rotterdam: Sense Publishers. Samuels, Warren J. (1990): Introduction. In: Samuels, Warren J. (Ed.): Economics as discourse: An analysis of the language of economists. Berlin: Springer Science & Business Media, pp. 1–14. Sánchez-Gómez M. C./Martín-Cilleros, Maria Victoria/Sánchez Sánchez, G. (2019): Evaluation of computer assisted qualitative data analysis software (CAQDAS) applied to research. In: Uden, Lorna/ Liberona, Dario/Sanchez, Galo/Rodríguez-González, Sara (eds.): Learning technology for education challenges. Cham: Springer, pp. 474–485. Silberzahn, Raphael/Uhlmann, Eric L. et al. (2018): Many analysts, one data set: Making transparent how variations in analytic choices affect results. In: Advances in Methods and Practices in Psychological Science 1, No. 3, pp. 337–356. Silverman, David (2000): Doing qualitative research: A practical handbook. London: Sage. Sparkes, Andrew C./Smith, Brett (2014): Qualitative research methods in sport, exercise and health: From process to product. London: Routledge. Stake, Robert E. (1995): The art of case study research. Thousand Oaks, CA: Sage. Steinke, Ines (2000): Gütekriterien qualitativer Forschung. In: Flick, Uwe/von Kardoff, Ernst/Steinke, Ines (eds.): Qualitative Forschung: Ein Handbuch. Hamburg: Rowohlt, pp. 319–331. Wickham, Hadley/Grolemund, Garrett (2020): R for data science. https://r4ds.had.co.nz, [Accessed 30 October 2020].

322

Chapter 12 How to make sense of data: Coding and theorising Caitlin Meagher

1. Introduction This chapter outlines the procedures of coding and theorising, the fundamental processes by which researchers organise their data into meaningful categories and concepts, and then into theoretical schema. The chapter is primarily intended for junior scholars of Japan Studies— undergraduates and early graduate students—as it discusses strategies to develop theory out of raw data. In the following pages, I will provide a step-by-step guide to the stages involved in coding: data collection, initial coding, focused/selective coding and finally theoretical coding; and I will discuss the methodological and epistemological significance of each step, particularly with regard to the study of Japan. Throughout, I will provide concrete examples of each type of coding, and of related concepts, from my own work—an ethnography of a mediumsized, mixed-sex sharehouse in Osaka Prefecture (Meagher 2017; 2018; 2020)—and from the essays that make up the rest of this chapter. This is to root the methodological guidance in concrete examples from a variety of projects, research topics and disciplines. The goal is to provide a blueprint for effective coding and to offer a reflection on the value of this methodology for social scientists producing scholarship in and about Japan.

2. Grounded theory as one framework for coding The process of coding and theorising, to be laid out in this chapter with reference to recent and ongoing Social Science research in Japan, is associated with the grounded theory method. Grounded theory emerged in the late 1960s—specifically, with the publication of The discovery of grounded theory: Strategies for qualitative research, co-authored by sociologists Barney Glaser and Anselm Strauss (2009), in 1967. The aim of that core text was to provide a clear road map for generating theory from qualitative data. Discovery was written in response to what the authors saw as theoretical stagnation in the Sociology of the 1960s, with researchers merely finding new quantitative data to reaffirm accepted theories, rather than producing new insights. Grounded theory’s emphasis on using data to generate new theory is relevant for Japan scholars in planning and constructing our research because ethnographic/empirical research on Japan is as likely to defy or challenge grand Social Science theories generated elsewhere as it is

323

Caitlin Meagher to affirm them. As Japan scholars, we endeavour to formulate theoretical models grounded in our data that may, however, be relevant and useful to scholars of other regions. The conclusions we reach on the basis of our data can and ought to be generalisable, to some extent at least, if we are to make meaningful contributions to our fields. Although not all coding follows the grounded theory method, I have chosen to emphasise grounded theory in this chapter for this reason, and because it offers the most comprehensive and formulaic approach and is therefore most instructive to novice coders.

3. A blueprint for the process of coding There are, in essence, two types of coding: substantive and theoretical. The first, substantive coding, subsumes all the activities or steps through which the researcher identifies, isolates, names and delimits central ideas in her data. It consists of ‘initial’ or ‘open’ coding (of which in vivo coding is one form), and ‘focused’ or ‘selective’ coding. The following table sets forth a blueprint for the process of coding, from data collection to preliminary analysis. It may be useful for you to refer back to it as you read the remainder of the chapter. Please note, however, that this is provided as a guideline and that the process has been simplified (perhaps oversimplified) for the sake of exposition and clarity. In practice, the process is inevitably more complicated, often repetitive and/or recursive, with ideas impelling the researcher to forge new conceptual paths or return to older ones. Additionally, some of these steps, particularly generating memos and diagrams, may and often do take place at every stage of this process. This table is an attempt to capture a non-linear process in linear form, so it necessarily obscures the more creative and serendipitous aspects of coding and theorising. Table 12.1: The process of coding Step

Data collection (4)

Initial (also known as open) coding (5, 6)

In vivo coding (7)

324

Purpose

Example

Related concepts

Amassing a robust body of qualitative data for analysis and generating theory

Interview transcripts, field notes, archival materials, technical and professional literature, government documents, life histories

Elicited and extant texts

The sharehouse as emancipation (jiyū) from home and the workplace

Word-by-word, lineby-line, incident-by-incident coding

Identifying central codes and concepts for development in generating theory A method of initial coding in which the original term or phrase is retained to keep the concept as faithful as possible to its use in the original text

Amaeru: the expectation that women feign dependence

Chapter 12 How to make sense of data Step

Focused (also known as selective) coding (8, 9)

Purpose

Example

Developing the central themes and concepts identified in initial coding by seeking as many incidents of the code as are necessary to establish theoretical saturation or sufficiency: when no new insights emerge from the data

The various uses of jiyū in my data: naturalised gender expectations with regard to household labour; relationships between daughters and parents in the parental home; sexbased discrimination in the labour market; travel, internationalism, and cosmopolitanism;

Related concepts

Theoretical saturation or sufficiency

sexual enfranchisement, etc.

Theoretical coding (10)

Memos and diagrams (11)

Establishing relationships among the concepts identified in initial coding and elaborated in focused coding; generating theory through these relationships; weaving the threads together into a tapestry

‘Emancipation’ in the sharehouse as ‘investing in yourself’ or ‘study abroad at home’; emancipatory rhetoric versus observed realities

Clarifying and elaborating emerging ideas; articulating nascent categories in written or visual form; giving way to new insights

4. Data collection: What counts as data? One of grounded theory’s contributions to Social Science enquiry was the expansion of the concept of ‘data’ to include various qualitative forms. Qualitative researchers collect data through various activities and strategies, including participant observation, interviewing, and the collection of archival materials, amongst others. For example, Julia Gerster’s more than one year of fieldwork researching community rebuilding efforts in Tōhoku included conducting 150 semi-guided interviews, collecting oral narratives and compiling several books with a huge amount of fieldnotes in addition to her own reflections on her participant observation experiences (see this chapter, Ch. 12.3). The two fundamental categories of textual data are elicited and extant texts. Elicited texts include interview transcripts and other informant statements rendered in textual form. Obviously, as these elicited texts emerge out of the interaction between researcher and research subject(s), it is important to recognise that they are not value-neutral; rather, they represent a situated interaction between parties with often divergent interests and emphases.

325

Caitlin Meagher In my case, for example, I had spent nearly a year living closely with young men and women, most of whom were living outside their family home for the first time. Though I had recorded my interviews with housemates as faithfully as possible, the content of these interviews was determined, of course, by what I thought necessary to ask and how/whether my informants thought fit to respond. This recognition (that my data set is both partial and determined in part by my own research agenda) is referred to as reflexivity. Another related concept is that of positionality. Positionality refers to the recognition that elements of my own identity, including my age, my gender and my outsider status, made me both: 1. more interested in some ideas, which I pursued in my interviewing and notetaking, than others; and 2. that my research subjects’ awareness of these identifiers made them more or less willing to talk to me about certain issues (see McLaughlin, Ch. 6). Extant texts, on the other hand, consist of existing materials and encompass various textual forms, from personal diaries to technical materials that offer some insight beyond what the researcher and research subjects produce. These offer a view of the research topic at a remove from the researcher’s and research subjects’ immediate concerns, though they are not to be construed as objective sources any more than elicited texts are: they, too, are artefacts of their authors’ motivations, circumstances and beliefs. Concrete examples from my own research include the numerous examples of sharehouse marketing, including sharehouse guidance literature, as well as government reports on the state of shared housing, that I collected and read during and after my fieldwork. In fact, a major theme of the book I produced out of this research (Meagher 2020) is the contrast between the ways sharehouses were marketed, on the one hand, and the realities of living in a sharehouse (my observations and what was revealed in interviews), on the other. Coding treats all data, essentially, as text, and the coding process takes an obviously textual approach to sorting data. In addition, new technologies have made it possible to code audio/ visual materials in this way (see this chapter, Gerster, Ch. 12.3). As a result, according to Kathy Charmaz (2006, p. 35), elicited and extant texts may be used as primary or supplementary data. Both are subjected to the same analytical scrutiny, the same method of analysis (i.e. coding), and both may spark insights that might otherwise escape the researcher.

326

Chapter 12 How to make sense of data Table 12.2: Elicited and extant text Definition

Example

Considerations and caveats

Elicited text

A text generated out of interactions between researcher and interlocutor(s)/research subjects: interview transcripts, fieldnotes, etc.

Transcripts of interviews with my housemates/ informants and industry professionals; field notes from participant observation

Positionality of the researcher; influence of interpersonal relationships and power structures between researcher and informant(s) on texts produced

Extant text

A text already in existence and independent of the research project: may include professional or technical literature on the topic, marketing materials, other academic literature

Sharehouse marketing literature; sharehouse guidance literature; sharehouse real estate websites; depictions of sharehouses in popular media; government white papers and other reports on shared housing

Purpose and conditions of the texts’ production; possible bias or agenda; completeness and accuracy of the text; suitability and relevance of the text

5. Identifying and isolating codes and concepts: Initial/open coding Imagine that you have returned from a productive period of fieldwork in Japan. You are at once satisfied with and overwhelmed by the volume of data you have collected: most likely, it is an amalgam of interview transcripts, observational field notes, various types of images and varying kinds of extant text. Your task is now to make some sense of it all, to construct a coherent analytical narrative out of patches and fragments. This was precisely the situation I encountered when I returned from research with boxes of field notes, some handwritten and others typed and saved in numerous separate documents; hundreds of photos; transcriptions of interviews; and various forms of literature. As Anselm Strauss and Juliet Corbin put it, ‘Science could not exist without concepts. Why are they so essential? Because by the very act of naming phenomena, we fix continuing attention on them’ (1990, p. 62). In coding, a ‘concept’ is a significant word or idea that a researcher identifies and elaborates as a unit of meaning. Concepts are, of course, not stand-alone phenomena, but exist in various relationships to each other; a theme is a recurrent and pervasive master concept that brings disparate other concepts into conversation with each other, leading to the generation of theoretical insight. A code is the term assigned to a concept and stands in relation to a concept in much the same way that a symbol stands in relation to an idea: it is simply the single, consistent term assigned to represent a concept or set of concepts. As with 327

Caitlin Meagher the concepts they represent, some codes are more pervasive and salient than others: a key code represents a master concept or theme. The first step in generating theory, then, is isolating concepts and naming phenomena through open coding or initial coding, which Charmaz defines as the process of ‘naming segments of data with a label that simultaneously categorizes, summarizes, and accounts for each piece of data’ (2006, p. 43). Identifying and naming these concepts, which will serve as the building blocks for analysis, is not merely housekeeping. Instead, it is the necessary first step to generating theory from the chaos of raw data. Note that the derivation of theoretical concepts from the qualitative data, as opposed to the use of data to verify existing concepts, is embedded in this process. The researcher does not pore through data in search of concepts borrowed from existing theory. Rather, to the extent possible, the researcher approaches the data without any preconceived itinerary, reading the signposts as they appear. This is particularly relevant to our work on Japan, which often fits imperfectly, or does not fit at all, into theoretical concepts developed elsewhere. Rather than starting from the perspective of testing the Japanese case against established models, coding in this way prioritises, in a quite literal sense, the Japanese data. Any theory generated should be grounded in the specificities of the (in this case, Japanese) sociocultural context, not an ill-fitting hand-me-down from scholarship generated elsewhere (for quantitative data and predefined hypotheses see Hommerich/Kottmann, Ch. 10). To return to my own research, for example, I found that theoretical work on alternative housing generated elsewhere had limited relevance to my study, given the rarity of and stigma against living with non-kin (tanin) in modern and contemporary Japan. Indeed, even the concept of tanin (literally: ‘others’) is not an easy idea to translate, because the ideological isometry of home and family that emerged in modern Japan imbues tanin with ideas about contagion and otherness. Had I approached the data from the perspective of verifying existing theory about the meaning and experience of shared housing which, at that time, had only been produced in other contexts, I would have missed considerable insight into my own contextually situated data. Instead, by grounding the theoretical aspects of my work in the specific data I collected, I was able to generate fresh theoretical insights into the meaning of sharing a home with others that, while based in the Japanese data, can inform broader knowledge about the meanings of ideas like ‘home’, ‘family’, ‘independence’ and ‘community’.

6. How to do initial coding? In coding, the researcher seeks out central codes and concepts that will form the basis of the generation of theory. Initial coding takes place at a number of levels, as Charmaz (2006, pp. 50–53) describes: word-by-word coding, line-by-line coding or incident-by-incident coding. In the beginning, it is advisable to stick as closely as possible to the data, and to this end it may be useful to code word by word. Simply put, this means that the researcher reviews every word of her data in search of meaningful concepts. Though it is labour-intensive and may be somewhat tedious, it minimises the risk that a salient trope, one that might serve as a key concept in the generation of theory, escapes the researcher’s attention. 328

Chapter 12 How to make sense of data A number of relevant codes will emerge through this laborious first step, and eventually the researcher will graduate to line-by-line coding, where the unit of coding is a line of text rather than the individual word. This progression occurs quite naturally as the researcher grows more familiar and more confident with her data; indeed, many researchers, especially those already familiar with the research topic and/or more experienced in grounded theory, may skip word-by-word coding entirely. As with word-by-word coding, line-by-line coding requires that the researcher remains attentive to the content of the data and the concepts emerging therefrom. Finally, initial coding may proceed to the incident level; researchers may or may not want to reach this level of abstraction during the initial coding process and, if they do, they will undertake incident-level coding once word-by-word or line-by-line coding has generated a preliminary corpus of codes that can be identified and applied to ‘incidents’ in the data. One of the challenges of initial coding is to resist the urge to make conceptual leaps or indulge in the application of a priori categories that originate in the researcher, rather than emerging from the data. Glaser (1978) suggests that, in order to keep initial codes as ‘open’ as possible, and to retain the processual nature of the data, researchers should use gerund forms rather than nominal forms whenever possible. Celia Spoden (see this chapter, Ch. 12.2) provides examples of the codes that emerged early in her own research into advanced directives. These included ‘experiencing illness/death of a loved one’, ‘suffering as a surrogate’, ‘experiencing conflicts at the deathbed’, ‘feeling insecure about the own end’, ‘worrying about being a burden’ or ‘being informed of advance directives’. As she explains, this initial labelling is ‘provisional’, and she maintained an attitude of openness to possibilities, ‘considering different interpretations and delaying their fixation’ (ibid.). This is important because it exemplifies grounded theory's emphasis on theory being driven by data, rather than the researcher's own (potentially subjective or premature) analytical conclusions. In my initial plunge into the data I had collected, I began by coding word by word, searching for recurrent and seemingly significant themes. Of course, though I did not begin coding in earnest until I returned from the field, my data-collection strategies in the field were driven by early inklings and cues that presented themselves, so it was not as if I was seeing the data for the first time. One of the first codes to emerge in my word-by-word coding was jiyū (freedom), which arose especially with regard to female sharehouse residents. It refers to the idea that the sharehouse offered emancipation from gendered behavioural conventions in other spheres of their lives, in particular: the family home, the marital home and the workplace. Once I identified jiyū as a central or key code, I could look for instances of jiyū and its variants using lineby-line and eventually incident-level coding, recalling contexts and scenarios (conversations, events and materials) where this concept arose and was elaborated. I will return to this example with reference to in vivo coding and focused coding below, but this concept and its variants arose with enough frequency and in enough permutations that I knew it warranted a central place in my analysis.

329

Caitlin Meagher

7. In vivo coding One strategy to ensure that codes remain as faithful as possible to the data is the use of in vivo codes (Glaser 1978, p. 70; Charmaz 2006, pp. 55–57), which simply refers to retaining the original language and phrasing when naming concepts. In vivo codes ‘anchor your analysis in your research participants’ worlds [and] can provide a crucial check on whether you have grasped what is significant’ (Charmaz 2006, p. 57). In her contribution to this volume, Gerster (see this chapter, Ch. 12.3) discusses her own use of in vivo codes in her research on regional revitalisation in Tohoku: as her special focus was social relations within communities affected by the natural disaster, she isolated phrases that emphasised this element rather than, for example, details about the disaster itself. Thus, the in vivo codes she identified included ideas like ‘everyone was supporting me’, ‘strongest community’ or ‘not able to express my opinion’. One value of in vivo coding is that it protects the researcher against inadvertently imposing her own interpretation on the data, in recognition of the fact that language is never value-neutral. This risk is ever-present in Social Science research, when researchers translate their informants’ statements and expressions for an academic audience. But the risk is considerably greater when the researcher and informants are working across different languages, and this ‘translation’ becomes more literal. This is a special concern in Japan Studies, as translations from Japanese to English and other languages of instruction/research can be unwieldy. Though in theory any emic (native) concept is translatable given enough space, Japan scholars are often tasked with how to work with terms and concepts that defy a neat one-to-one translation. We rely on approximations, exegeses and elaborate analogies. But these should come last, during the explication of a theory, not during its development. In the preliminary stages of coding, it is important to retain these original concepts, as expressed in the original language, lest they become confused or degraded by the researcher’s analytical agenda. To return to an example from my own research, I found frequent references to the concept of amaeru. This code emerged as one of the many sub-concepts or sub-codes under the theme of jiyū. The concept of amaeru, with a storied history in mid-20th century Japanese social psychology, means very much more than the reductive English language translation as ‘dependence’. In my interviews with young female sharehouse residents and in sharehouse marketing literature, amaeru often meant something like ‘to feign dependence or beg indulgence as a form of flattery and to downplay one’s own abilities and independence’. The sharehouse was depicted as a space of emancipation from the burdensome cultural mandate for women to amaeru. It was important for me to retain the idea of amaeru in its original Japanese to retain the full force of the concept, a full exegesis of which is impossible here due to spatial constraints. However, settling for the conventional English language translation ‘dependence’ would have foreclosed further enquiry into the nuances of this important code, which emerged as a central concept in one of the chapters of my thesis.

330

Chapter 12 How to make sense of data

8. Developing concepts: Focused/selective coding Once the researcher has achieved some analytical purchase on the mound or wave of data before her through the rather arduous and slow-going process of initial coding, she can begin to sift through the remaining data by performing focused (or ‘selective’) coding. In advancing from initial to focused coding, the researcher’s purpose shifts from discovering codes to finding instances of known codes and specifying the parameters of those codes. Thus, in this phase of coding, the researcher seeks out new incidences of designated codes, essentially to test whether these new incidences challenge or add to the meanings she has already assigned to them. Obviously, this process is not mechanical. On the contrary, focused coding is a recursive process through which the researcher tests the adequacy and relevance of previously discovered codes with an eye towards refining or revising them to more accurately capture the realities expressed in the data. This is an important step in ‘grounding’ grounded theory, as it forces the researcher to justify the analytical frames she will use in generating theory. After my initial coding, during which jiyū emerged as a key code or theme, I began to actively look for instances of jiyū in my data to elaborate the ways it is used, and the contexts in which it was used, through the process of focused coding. It emerged, for example, in discussions about naturalised gender expectations with regard to household labour; about the relationships between daughters and parents in the parental home; about sex-based discrimination in the labour market; about travel, internationalism and cosmopolitanism; and about sexual enfranchisement, among others. My goal in initial coding had been discovering and naming concepts. Having designated the idea of jiyū in its various forms as a key code or theme, I endeavoured, during the focused coding stage, to compile a more or less exhaustive list of the contexts in which it appeared in my data.

9. How much is enough? Theoretical saturation The threshold for determining that a code or concept is developed enough for analytical use is theoretical saturation (Glaser/Strauss 2009, pp. 61–62). The aim of theoretical saturation is to ensure that the codes designated correspond to the data; that they are sufficiently specific; and, finally, that they are sufficiently complete to carry theoretical weight. Theoretical saturation is achieved once ‘gathering fresh data no longer sparks new theoretical insights, nor reveals new properties of these core theoretical categories’ (Charmaz 2006, p. 113). The researcher proceeds cautiously at the beginning, venturing a provisional analysis and checking it against (grounding it in) the empirical evidence. However, especially with regard to the patterns of human behaviour, it is crucial to recall that whether an incident reaffirms or challenges the ‘pattern’ observed by the researcher depends, at least in part, on the researcher’s judgement and on the core theoretical categories established by the researcher. As a result, Ian Dey (1999, p. 257) rejects the notion of saturation in favour of sufficiency, as a reminder that ‘data suggest categories but involves conjecture’.

331

Caitlin Meagher With regard to my key code of jiyū, there finally came a point when I was confident that I had achieved this sufficiency. I was no longer in Japan, having returned to the United Kingdom to write up my thesis. The material conditions of fieldwork, which are typically dependent on funding over a set period, mean that we as researchers cannot continue our research indefinitely, however dear it is to our hearts. Fortunately, technology allows us to maintain some access, for free and in real time, with our contacts in the field (see McLaughlin, Ch. 6). In the U.K. and actively writing, I continued to collect and review new materials about sharehouses and continued to be connected to my former housemates/informants on social media, where I was privy to their discussions about sharehouse living from afar. When I realised that every new iteration of the idea of ‘emancipation’ idea was captured within the parameters I had previously set for that code, I was satisfied that it was, if not exhaustive, sufficiently elaborate for use in generating a meaningful analysis. Had I continued to encounter new permutations of the concept, however, or unanticipated applications, it would have been my cue that I had not achieved sufficiency, much less saturation, and it would have been incumbent on me to seek out new data until no new theoretical insights emerged.

10. Coding as theory: Theoretical coding So far, we have discussed initial/open coding and focused/selective coding. To reiterate, the first is to discover central concepts and pin identifying markers (codes) to them, and the second is to elaborate and define these concepts through saturation. These processes are primarily descriptive/definitive, and are both subsumed under the category of substantive coding. Theoretical coding is distinct from substantive coding in that it moves the research from establishing and defining categories to establishing the relationships among concepts and categories. Thus, while substantive coding breaks the undifferentiated mass—the mound or the wave— into its component parts, theoretical coding re-combines these parts into a coherent pattern (like shards into a mosaic, or threads into a tapestry). As Glaser (1978, p. 55) puts it: ‘Substantive codes conceptualize the empirical substance of the area of research. Theoretical codes conceptualize how the substantive codes may relate to each other as hypotheses to be integrated into the theory.’ Thus, as the name ‘theoretical coding’ suggests, coding and theory are elements of the same process, of discovering regular relationships among concepts. In my case, ideas about emancipation, as expressed by my housemate informants and in sharehouse marketing literature, centred around two themes: the idea of living in a sharehouse as ‘investing in oneself’ and one’s career; and the idea of the sharehouse as a way to ‘study abroad in Japan’ (Meagher 2018; 2020). These two somewhat interrelated ideas emerged to me through theoretical coding and became the themes of the first two substantive chapters of my study.

332

Chapter 12 How to make sense of data

11. Practical considerations: memos and diagrams Coding is the strategy used by grounded theorists to wrangle an expansive and unwieldy data set—the ‘mound’ to be climbed or the ‘wave’ to be surfed, in the imagery of Gerster and Rosenberger (see this chapter, Ch. 12.1; Ch. 12.3). With a seemingly infinite array of directions for the researcher to follow, she makes conscious decisions at every step, following some directional cues in the data and declining (or postponing) others. At the early stages of the coding process, they allow the researcher to explore hunches or possible connections without immediately committing these to a linear narrative. The reader has likely experienced the way that putting our ideas down on paper—whether in a written or visual format, and however provisionally or schematically—forces us to engage with our ideas in a different way from casual speculation. Rosenberger describes her construction of a theme map as a ‘playful’ and experimental process, writing themes in bubbles and moving them around a page. In the more advanced stages of the research process, memos and/or diagrams keep track of the researcher’s own line(s) of thinking, a theoretical trail of breadcrumbs. Should the researcher find herself too far afield of her initial signposts, she is able to ‘double back’ to any point along the route by referring back to them. Thus, the construction of memos and diagrams is useful at every stage of the research process: it allows the researcher to clarify and elaborate emerging ideas; or, in articulating these nascent categories in written or visual form, they may give way to new insights. The hunches and possible connections captured in these artefacts may end up on the cutting room floor, so to speak. Or they may find their way—in their entirety or in an excerpted form—into the published research.

12. Conclusion I would like to conclude this chapter with a brief discussion of Spoden’s anecdote (see this chapter, Ch. 12.2) of her coming to terms with the concept of songenshi (‘death with dignity’) in her own research on advance directives, which demonstrates a number of the principles and ideas presented herein. She mentions that this term caught her attention as she was coding, and emphasises that it was not a frequently occurring concept in her data; on the contrary, it only occurred twice in her interviews. Spoden, by tunnelling into the significance of this idiosyncratic expression rather than discarding it as irrelevant, discovered a central concept that might otherwise have escaped her attention. This is the advantage of staying close to one’s data in the exploratory phase of initial coding. Second, by retaining its original form as songenshi rather than translating it into English or German (in vivo coding), she also retained the peculiar history of this term and its linguistic implications. The anecdote also demonstrates the way that a researcher’s particular background and theoretical engagements influence the choices she makes in dealing with her data and the directions she will follow. Note that, far from denying the role of her existing beliefs and interests, Spoden emphasises the role of her ‘academic socialisation in Germany’, which made her sensitive 333

Caitlin Meagher to the implications of songenshi. As researchers, we bring certain ideas and assumptions with us to the data but, to the extent that we can be circumspect about them, they may contribute to, rather than undermine, our theoretical abilities (reflexivity). The process of coding, as I have set forth in this chapter, is a strategy that can be used to grapple with a substantial and varied mound of data: to tease out important strands of meaning (initial coding), discover their parameters (focused coding to theoretical sufficiency) and to weave them back together into a cohesive narrative (generating theory through theoretical coding). Coding in this way is especially useful to scholars of Japan (and Area Studies more broadly), as it ensures that theory proceeds from the data, and not vice versa. This is important because it can be a corrective to the occasional tendency in the Social Sciences to impose procrustean theories, developed elsewhere, to specific cases. Coding is an opportunity to generate theory grounded in the Japanese context for application or assessment in other spheres. It can reposition Japan, and scholarship on Japan, within disciplinary conversations and theory building.

334

12.1 Cresting the wave of data Nancy Rosenberger

It is you and the data. You are home from three weeks, three months or a whole year of fieldwork. Your machine is full of transcribed interviews, conversations and observations. The wave has arisen from the ocean and is swelling towards the shore. Your job is to surf the back of the wave—to ride it up to the point where it breaks into a story. This is ethnography—the writing of people’s experiences—being born at its creative edge. I have always found this a stressful, exciting ‘don’t-know’ point. The flow of words that have filled your ears and notebooks is going to join with your particular manner of riding the wave. The question is whether you can do it in such a way that mines the data for all it is worth, stays honest to the data, yet ultimately results in a narrative that communicates what you understand to be the most important tale to tell. Of course, you sit at your desk with some ideas already as to what is going on in the data. You have shaped the research questions, the questionnaire, the sample, and you have melded it with theoretical perspectives and previous research on the topic. But the task at hand is to put that aside and listen to what the data has to tell you. Now you are being asked to dive into the words of the people who have given you their thoughts and hearts, and let the cold water clear your brain so that you can listen to the roar of the data.

Coding Careful coding is the first step because it reduces the complexity, makes you focus on small bits of text and interrupts the overall story that you are already creating in your head. Coding is essentially classification. This stage entails going through all of your data carefully and classifying it into parts. Name each stream of water that flows through the data; some streams might get coded in several parts because they relate to two or more aspects of your research. Do not miss the spindrift that blows off the edges—felt senses of what is going on, things alluded to, half-spoken, gone in a second, but once glimpsed, important to preserve. Coding is tedious and one is often tempted to pay someone else to do it. Do not ignore the value of reading through all of your data again, reacquainting yourself with what you have forgotten or think you know, and considering how you want to categorise it. The classifications will highly influence what the data tells you. If you yourself are unable to do the actual coding, read through as much data as you can in order to suss out the codes that you want your coder to use. Make a list and refine it. If you have the luxuries of time and money, you might have both you and a coder unfamiliar with the material code the data. The familiarised eye and the unfamiliarised eye will see differently. Coding follows the technology of our brains—sort out the blue blocks, the red and the green. Put them in piles: same, different, same, different. Let us say the blue blocks represent what I 335

Nancy Rosenberger call a bucket, which simply names a general topic. Some topics are obvious, such as when I asked all of the women I interviewed about their dreams for the future: ‘dreams for the future’ becomes a general topic or what I call a bucket—something that has no specific meaning yet; it is just carrying the water. Different shades of blue quickly emerge and with them a hierarchy of buckets under the main blue bucket: large-scale dreams of geographical movement (going to New York City to work or leaving northern Japan to open a coffee shop in Okinawa); small-scale dreams (returning to flower-arranging lessons or working at a friend’s store); relationship dreams (travelling with your husband or visiting an old friend). How you divide these up makes a difference to how the tale will be told. Code your data into more buckets than fewer. The hierarchy of smaller buckets saves you because you can always pour the little buckets back into the big one, but you cannot divide the big one easily after the coding is finished. You will probably use coding software, such as NVivo or Dedoose. I started out coding on paper and this was a great boon. But never forget that your brain is called on to draw the connections and express the nuances that lie in the data. An important challenge is to be sensitive to feelings or values that are spread throughout the data without an obvious flag attached to them. Sometimes this is embedded in a word such as jibunnari (something that suits me, or is of me). If it comes up often, it becomes its own code. And, often a sentiment may not be represented in a word, but come out as a general feeling or tone. For example, if all the dreams for the future have to do with time away from the demands of other people, and if that sentiment pops up in other buckets as well, this becomes a code: ‘desiring time away’. All segments of conversation or observed action that attest to the depth of this feeling are stored under this code. As you go through your data, a hierarchy may in time develop under this code, classifying different forms of time away, different intensities of desire, the nature of desires, the way the person judges her own desires or feels other judging them, and so on. When you write up this code, you return to the hierarchy to figure out how to portray this desire and what to emphasise. Size matters in this decision, like for example, if half of the people dream of travelling overseas, then this needs to be stressed, analysed for nuances and described with rich quotations and stories. The unusual also matters. A small group of people, maybe only ten per cent, may turn to religion to get away, and another small group, another ten per cent, to drawing, ballet or piano. Do not lose this group. In Dilemmas of adulthood (Rosenberger 2013), I merged these last two groups into people getting away into the irrational or unproductive side of neoliberal life, and included quotations and case studies under this rubric, with religion and art still divided out. Looking carefully inside codes and searching across codes is essential to staying honest to the data and emerging with your own unique story to convey—a story that relates to inherited theory and studies, but that adds a new hue to the ocean. In finally writing up your material, it could be easy to classify time away from the demands of others as individuality—a major preoccupation of theory about Japan versus the West. But characterising individuality accurately for this group of people is important. Do not let theory define it for you. What does your data for these people in this time and place tell you? This will become your contribution to the field. Study the context and manner of speaking about time away. Was it said with a sense of futility, anticipation, embarrassment, anger, resistance or joy? Furthermore, it will be important to look for other codes that relate to this sense of wanting time away, and may in the long 336

Chapter 12 How to make sense of data run contribute to your characterisation of individuality. The research will heighten your awareness to utterances and contexts containing words like jibun (self), wagamama (selfish) and watashi wa (I am/do), all of which show nuances of the Japanese language and specifics of the representation of the self. These might also be considered in contrast with codes that trace feelings about relationships or identity. How do feelings about relationships connect to the feelings about time away? How does desire for time away relate to giving up individual identity in the flow of art or religion? We might assume they would be in opposition, but think it through for nuances.

Themes Themes are statements that characterise a particular study. They grow out of the topic buckets mentioned above. Themes carry meaning particular to your study with these specific people, and it is around themes that you will wrap your story. From the buckets of ‘dreams for the future’ and ‘wanting time away’ comes a theme such as: ‘Japanese women (in this study) dream of having a time or place where they can live for themselves away from the responsibilities of everyday life’. Such a theme statement can become the centre of a chapter or section of your work. As you write it, you get the luxury of delving into all the neatly coded sections mentioned above to draw out pithy quotes that represent different aspects, intensities, emotions, behaviour and judgements that will bring this theme alive for your readers. For most studies, three to five main themes will either evolve or be chosen as most important. Now you are shaping the wave as it rolls in towards the shore—the mutual product of what the data has told you and what you think tells a story that is analytically robust (supported by the data), persuasive and accurate to what you have experienced as an ethnographer. Qualitative research involves a lot of data, and while your data analysis may have resulted in ten or more themes, each of which deserve to be told, your job is to winnow these down. This process may feel overwhelming. Live with that for a while, struggle with it, freewrite about it, talk to friends about it, and in time you will see through the froth to the clear water. I suggest leaving some themes for later papers and choosing the ones that, like pieces of a jigsaw puzzle, will paint a picture that tells an engrossing story when assembled. But keep in mind that various jigsaw puzzles are possible; you have to decide which pieces to carve out and reassemble. When choosing themes, the researcher often goes back to their research questions once more. What combination of themes is going to best answer your research questions —or perhaps unsettle them? What themes relate to each other in interesting ways? In the process of choosing themes, one more analytic tool can come to your aid: theme maps. Theme maps arrange the most important themes of your study in graphic relationship to each other. Play with your themes, considering various possible relationships between them. They may complement each other. The theme above about Japanese women’s dream of time or a place away might be complemented by a theme such as: ‘Japanese women in this study attest to the importance in their lives of a period of enjoyment before and after children/career’. The ‘dream of time or a place away’ theme may also exist in apparent contradiction to another theme such as: ‘Japanese women expressed a high sense of responsibility to family members and work colleagues’. Then again, there may be an ironic relationship such as: ‘(though they want time away,) Japanese women yearn for more time with their friends, partners, husbands 337

Nancy Rosenberger and children’. Another relation could be assumed to be causal: ‘Japanese women feel overworked’. Many people can think through these themes best if they write them in bubbles and move them around a page in relation to each other. Make sure to characterise the type of relationship between the variables and draw arrows between them; the arrows may go both ways between bubbles or across the page as well as around a circle. This helps you to perceive the overall shape of the argument that you will make. Try explaining your theme map to someone else. This helps your story to unfold. You might make several theme maps before you settle on one. Themes may be included and taken out, others revived; you may develop sub-themes. Stay in touch with the quotations that support these themes, perhaps adding them to your theme map. Do not err in either direction: towards telling the story that you wanted to tell in the first place that rests mainly on your own assumptions, or towards clinging to the data so hard that the exposition becomes a list of quotes under themes. You are the expert on this data and it is the melding of your expertise with the depth and accuracy of the data that will make your paper, article or book zing. The final stage of the analysis is to choose an overall statement to characterise your paper or book. To some extent, it brings together the various themes. Contradictions are often interesting to emphasise, such as yearning for a relationship and longing for flow of the self. The title of my book Gambling with virtue (Rosenberger 2001) expresses the contrary position in which young, single Japanese women found themselves in the 1990s. In my book Dilemmas of adulthood (Rosenberger 2013), the tensions of ambivalence and uncertainty seemed to characterise the population I was studying. For organic farmers that I studied, the irony between themes is clear: organic farmers are resistant to typical farming methods, operating outside of the market; yet they are ideal citizens of neoliberal Japan—individualistic and responsible for themselves, taking advantage of government subsidies (Rosenberger 2017). Such tensions as these make for an intriguing story to sort out and, luckily for anthropologists, human life is full of them.

338

12.2 Lost in translation? Grounded theory and developing theoretical concepts Celia Spoden

In a project on advance directives in Japan (Spoden 2015; 2017; 2020), I conducted narrative interviews (Schütze 1983) with people who documented their wishes on medical treatment for end-of-life situations in advance in a so-called living will (ribingu uiru) in case they might not be able to decide for themselves any more. My primary interest was to study the motivation and subjective perspectives of my research participants to formulate such a living will. Following a grounded theory approach (Glaser/Strauss 2009), I analysed ten cases of five women and men respectively who were between 45 and 88 years of age. I developed the key concept ‘becoming critically aware of life-sustaining treatments’ and the underlying concepts ‘lifetime and timing’ and ‘conceptions of the self as a mirror of decision-making’.

Grounded theory and coding Grounded Theory (GT) is a systematic approach to constructing theories which are grounded in the data. This means generating codes inductively from the data and developing them further into theoretical concepts. The research follows a theoretical sampling approach, which means that the researcher decides which data to gather next based on the emerging categories in order to develop them further. This process of analysing and going back into the field takes place until theoretical saturation is reached, meaning that categories become refined and filled out, and no new properties emerge any more (Charmaz 2006). However, this is the ideal pattern, which is sometimes difficult to realise due to time and financial restrictions, especially when the research takes place in a different country and funding is limited. Coding is the core element and there are different approaches to coding in GT, but regardless of what approach you go for the first step is called open coding (Strauss 1987; Strauss/Corbin 1990) or initial coding (Charmaz 2006) and the intention is to break up the data. Kathy Charmaz suggests constructing short codes, preserving actions using gerunds and moving quickly through the data (Charmaz 2006, pp. 48–49). The labelling of data in this first step is provisional; the codes are not related to each other yet and can be renamed later. The basic attitude is to be open. This can mean looking at the data from an alienated perspective, considering different interpretations and delaying their fixation, asking analytical questions of the data and questioning one’s own preconceptions. In the following step, the initial codes are related to each other, bundled together and their properties and dimensions considered. This process of axial coding means sorting and bringing order from chaos. I draw mind maps and start developing the relations between codes and subcodes, or I write codes on cards and cluster them on a wall. I always write my first analyti-

339

Celia Spoden cal thoughts down in a memo whatever the sorting technique. This helps to clarify my thoughts and get them out of my head or detect gaps in the data, and it is the first step into writing. Through selective coding––the third step––the goal is to develop one or more key categories, which are grounded in the data and explain the processes or actions studied. This means constantly asking analytical questions or determining which theoretical categories are indicated by the codes. When I was coding my data on advance healthcare directives, different codes such as ‘experiencing the illness/death of a loved one’, ‘suffering as a surrogate’, ‘experiencing conflicts at the deathbed’, ‘feeling insecure about one’s own end’, ‘worrying about being a burden’ or ‘being informed of advance directives’ emerged. By asking which kind of process is indicated by these codes and how the codes are related to each other, I developed the category ‘becoming critically aware of life-sustaining treatments’, which comprises feelings of insecurity among my study participants about their final days, especially in the event that they might no longer be able to voice their own wishes regarding medical treatment. Therefore, drafting a living will is a precaution against potential problems in the future, which restores an indivdiual’s feeling of being secure in the present.

Grounded in the data or forced onto it? Besides the category ‘becoming critically aware of life-sustaining treatments’, which is grounded in the experiences of my research participants, there were other codes, indicating ‘notions of lifetime’ or pointing towards ‘self-conceptions’ in connection to social roles and responsibilities. For example, ‘having lived long enough or longer than expected’, ‘living ten years in one’, ‘existing eternally/soul (tamashii)’, ‘(not) thinking about one’s own future’ were expressions related to ‘notions of lifetime’. Codes I attributed to ‘self-conceptions’ included, for example, ‘having always decided for myself’, ‘being afraid of losing myself’, ‘searching for a decision in accordance with myself’ or ‘leading an independent life’. Here, the perception of social roles and corresponding responsibilities was important and expressed in formulations such as ‘having always cared for others/not being able to endure being cared for’, ‘not wanting to burden others’, ‘giving up social roles in order to give up living’ or ‘having no one who depends on me’. However, the theoretical categories ‘lifetime’ and ‘self’ that, in my understanding, were indicated by these codes, originated in a Western context. Could they be applied to a Japanese context or was I imposing the theoretical framework of a ‘Western researcher’ on my data? I struggled with this question and searched for answers. This took me to the literature about the period of modernisation in Japan, during which philosophical and sociological concepts were translated into Japanese with the aim of modernising society. Key concepts of modernity––such as society, individual, nation, religion and so forth (Shimada 2007)––were first translated and introduced into the academic discourse. They became key concepts of modern life and now belong to general knowledge and language, and are seldom remembered as translated concepts. This is also the case with ‘time’ (jikan) and ‘self’ (jibun). Coding involves many choices the researcher has to make, such as which route to take and labels to use. The more I reflected on the language used by myself and the study participants as well as my preconceptions, concepts used by participants and theoretical concepts developed, 340

Chapter 12 How to make sense of data these choices became conscious decisions. I decided to work with the concepts ‘lifetime’ and ‘self’ and elaborate on how they were translated into Japanese society, which structural changes they brought about and how their meaning shifted. Aside from theorising my empirical data, this allowed me to contribute to a wider discourse on key concepts of modernity. In a chapter on ‘lifetime and timing’, I outlined the historical process of translation and cultural adaptation of this concept. This made it possible to consider the evolving modern concept of time in Europe in its relation to its introduction in Japan and its consequences for modern societies. Opposed to a pre-modern understanding of time––where time and space were closely connected––an abstract concept of time and the institutionalisation of the life-course characterise the modern concept ‘lifetime’. Together with standardised scientific measurements of life expectancy, this concept forms the basis for the interpretations of life and its finiteness by the research participants. On the other hand, beliefs in spirits (tamashii) or the afterlife are related to pre-modern concepts of time, which offer us the opportunity to transcend the finiteness of lifetime. For the concept of ‘self’, I searched for existing theories which best explain the reflections and actions of the study participants. I chose the symbolic interactionist understanding of the self (Mead 2015). In a chapter on the concept ‘self-conceptions as mirroring the decision-making’, I illustrated how the symbolic interactionist understanding of the self corresponds to the expressions of self by my research participants and how their self-conception is mirrored in their decision to draft a living will.

Absent or grounded in the data? While you are coding, it is helpful to reflect on the ways a specific term is used: Is it present in the everyday, medial or academic level of discourse? Does it belong to a specialised field or a certain professional language? Sometimes meanings may differ in everyday and academic language. Reflecting on a code’s usage in different languages reveals the historical and sociocultural contingence and makes it possible to reconstruct certain developments in relation to each other with their similarities and disruptions. For example, when I was coding the first interview, a statement by a 45-year-old hospice nurse puzzled me: ‘For me dying in dignity (songenshi) holds an exceptionally high standing. But there are others who say that is not the case for them.’ It seemed obvious that she wanted to highlight the importance of ‘dying in dignity’ for her, but why should there be others who do not think that a dignified death is important? Or could this statement even be interpreted as there being others who do not want to die in dignity? As soon as I started comparing codes with data, I realised that ‘dying in dignity’ was mentioned in only one other interview. Only the participants who had access to expert knowledge and professional discourses––the hospice nurse and a former journalist––explicitly mentioned songenshi. So why consider a concept which is more absent than grounded in the data? As someone whose academic socialisation took place in Germany––where Würde (dignity) is referred to quite often in the context of advance directives––this seemed to be a difference and caught my interest. I wanted to understand how the hospice nurse uses the term ‘dying in dignity’, why the concept was almost absent in the other interviews and how and with what meaning it was used in the Japanese debate on advance directives. 341

Celia Spoden This took me back to debates from the 1970s when the concept of songenshi was invented as a translation for the English ‘dying with dignity’ in 1976. The Asahi Shimbun (1976) coined it when covering the Karen Ann Quinlan case in New Jersey––the world’s first court ruling in favour of the withdrawal of a mechanical ventilator in a continuous comatose patient. In 1983, the Japan Society for Euthanasia renamed itself Nihon Songenshi Kyōkai (Japan Society for Dying with Dignity, JSDD). Ever since, the JSDD has used songenshi to campaign for the spread of advance directives and the legalisation of (passive) euthanasia. Digging deeper into the Japanese debate on euthanasia and especially the arguments of its opponents, I soon realised that songenshi is indeed the object of rejection (Otani 2010). Not in the literal sense, but because it stands for the legalisation of euthanasia in the JSDD’s campaign. Against this background the nurse’s statement––songenshi holds an exceptionally high standing with her––acquires a new nuance and can be understood as a self-positioning act in the debate on euthanasia. This shows that although songenshi seemed a good candidate for a key concept from the viewpoint of a researcher socialised in a ‘dignity’-laden environment of German bioethics, it was more absent than grounded in the data. It had no explanatory power for the empirical part. However, for the discursive context of my study and an understanding of the Japanese debate and its relation to the international debates, songenshi was an important concept.

Co-construction, self-reflexivity and cultural translation In GT, the researcher is involved in a process of co-constructing the data. What we see in the data depends on our research question, which is connected to our research interests, theoretical knowledge in a certain field and our preconceptions. We decide which codes we use to label our data, what to bring into focus, which paths we follow in the theoretical sampling and which ones we put aside. However, it is not simply we as researchers who construct a theory, we are also affected by our data: we have experiences in the field which might alter our point of view and we constantly reflect on emerging concepts and our preconceptions. Therefore, self-reflection and an open attitude are important prerequisites for interpretation and help prevent us imposing concepts on the data. Moreover, being aware of our own concepts, cultural backgrounds and the language used by research participants––and sometimes the terms they do not use––offers us the chance to reconstruct how key concepts of modern society were constructed under specific historical, social and cultural circumstances. By tracing how certain concepts were translated into different sociocultural contexts, we can show the relatedness of the debates and transformations of meaning.

342

12.3 Coding: Mapping the mountains of ethnographic postdisaster data Julia Gerster

Fieldwork was without doubt one of the most exciting times in my life. As a PhD candidate, I spent one year at Tohoku University in Sendai beginning in December 2016. I found my research topic because I was present in Tokyo when the Great East Japan Earthquake, Tsunami and Nuclear Disaster of 2011 happened. Ever since that time, the recovery process of northeastern Japan has captivated me. Soon after the disaster, people from severely affected areas revived their local festivals and strived to hold on to their hometown traditions. Cultural patterns from the Tohoku region, like local food culture or dialect, were used in other regions in Japan as well, either to express support for northeastern Japan or, in cases of displaced communities, to preserve their identities. These observations finally led me to the research questions I focused on in my dissertation (Gerster-Damerow 2019): Why do people turn to local culture after a disaster? How does local culture affect community building among those who have been displaced? And lastly: What are the differences between communities mainly affected by the tsunami and communities additionally severely affected by the nuclear disaster in terms of the impact of local culture on community building? To compare different aspects of the disaster, I chose Natori City in Miyagi Prefecture and Namie Town in Fukushima Prefecture, before and after Namie’s partial release from the evacuation zone, as my primary field sites. There, I joined in local cultural activities, followed discussions on recovery that took place at the temporary housing, and helped to organise community events at newly built public housing. The data I gained through my participant observations became the foundation of my dissertation. Additionally, I collected more than 150 semi-structured interviews, unstructured narratives, and several books filled with almost unreadable field notes gathered during my visits. In short, I was content at first and then rather terrified by the amount of information I managed to collect. Then I reached the point that every researcher dreads: making sense of the mass of data I was so eager to accumulate. It reminded me of the deep forests and dark mountains in Northeast Japan, where it is difficult to figure out where to go and what to encounter. Although it is always exciting to enter such a forest, it is also easy to get lost. Coding was like creating a map that helped me to find a way through the mountains of data and keep track of it. The longer I used the tool, the more I was able to recognise reoccurring patterns and structures that eventually helped me to answer my research questions. The map became clearer and clearer with some big roads connected to my research questions and many sideways that seemed interesting as well. I knew that it was impossible to follow all the streets that emerged from this project. Nevertheless, with the coding map I could always go back to explore the sideways in other projects later without having to start all over again. According to Charmaz (2006, p. 43), coding refers to ‘naming segments of data with a label that simultaneously categorizes, summa-

343

Julia Gerster rizes, and accounts for each piece of data’. Since the way researchers conduct coding is influenced by various aspects like individual background, discipline, choice of methods, type of data, field site or research questions (Benaquisto 2008), Elliott (2018, p. 2850) proposes conceptualising coding as a ‘decision-making process’. Although coding is mostly done with text, i.e. transcripts of interviews, field notes or other types of textual data, it has recently also been applied to pictures, video and audio recordings. Coding takes a lot of time, but Elliott (2018, p. 2851) emphasises that ‘coding is a way of […] essentially indexing or mapping data, to provide an overview of disparate data that allows the researcher to make sense of them in relation to their research questions. Most simply, it can be a way of tagging data that are relevant to a particular point […]’. Hence, coding is a crucial tool with which to contextualise one’s arguments through the data collected. It helps to find connecting themes and patterns as evidence. The challenges of the coding process depend on how familiar the researcher already is with the data. Especially in grounded theory approaches (see Spoden, Ch. 12.2), many start with socalled in vivo coding in order to stay close to the raw data and not jump to conclusions. In vivo coding describes the practice of taking words or expressions from a piece of data and literally using them as a code (King 2008). Let me illustrate what in vivo coding looked like within my research project. Since I was investigating social dynamics, my special attention focused on references to social relations within these communities. If somebody said, for instance, ‘after the disaster everyone supported me’, the in vivo code could be ‘everyone supported me’. The statement ‘my hometown used to have the strongest community’ could be coded as ‘strongest community,’ and another person’s comment ‘after the disaster I felt not able to express my opinion anymore’ could lead to the in vivo code ‘not able to express my opinion’. This kind of in vivo coding should serve as a first step towards familiarising oneself with the data. This is recommended when working with data that was collected a long time ago or by other researchers, although researchers working with grounded theory especially would always recommend beginning with in vivo coding as early as during the data collection process. Since I started working with the data immediately after its acquisition, I only used this coding method to highlight a certain expression and keep it until similar codes emerged. This relates to one of the crucial steps for that kind of qualitative data analysis: putting similar references together and summarising them using an overarching code. The examples mentioned above refer to various aspects of social dynamics. Therefore, in the process of coding, I developed the overarching code ‘social relations’. For the various aspects of social relations, I had identified sub-codes—sub-categories of a certain topic. Examples of my sub-codes are ‘social ties before the disaster’, ‘social ties after the disaster’, ‘positive relations’, ‘suppression’ and the like. Sub-codes that contain the terms ‘positive’ or ‘negative’ might appear too simplistic, but these codes were mainly meant to structure the data or to discover common themes while further exploring the data. For instance, I eventually summarised references indicating a time before and after the disaster as ‘change’; I renamed, deleted or shifted other codes to other sub-codes. In the end, a code structure emerged that represented the most important topics within the data. In view of my research design and questions, I was able to compare different cases. This led to findings like a gender-based or age-based difference related to interviewees mentioning certain topics; ‘fear from radiation’, for example, proved to be age-related.

344

Chapter 12 How to make sense of data During my years as a student and as a PhD candidate, I applied coding within different research projects and in various ways (Gerster 2018, 2019; Gerster-Damerow 2019). Initially, I tried to code segments from interviews by printing them, using multi-coloured post-it notes, or by marking them directly. In that way, each ‘code group’ was able to get a specific colour, which gave structure to the whole and made a future search easier. For smaller projects, this method worked. Yet, it might be harder to manage with larger projects, for the overall coding structure is not always easy to grasp. Many researchers therefore use Excel files for code summaries and attach examples from the data in a table. Nevertheless, for a large project, such as my dissertation, I preferred to use MAXQDA, a type of software that allows managing qualitative data easily. One of the main advantages of this and similar software like Delve, F4analyse or NVivo is the neat display of the coding structure, which makes it easier to keep track of the topics within the data. Further advantages are fast localisation of the main topics and actors as well as an easier comparison and processing of codes for different projects. This is especially helpful when going back to the data after a long time. However, it should not be expected that digital tools do all the work for the researcher. Software may be easy to handle and lure users into playing around with all the beautiful graphic tools. Yet, these types of software are only tools which support working with the data. The real work—that is, making sense of all the colourful codes—still has to be conducted by the researcher. Of course, there are also some pitfalls regarding coding as a method. As mentioned before, the time needed to examine the data, even with the help of software, should not be underestimated. Another challenge is to know when to stop the analysis and start the writing process. Some scholars even argue that it is ‘nonsensical’ to subdivide data into useful and brute words (St. Pierre/Jackson 2014, p. 716). There is no standard procedure that is applicable to different research projects in the same way. Repeated evaluations of the data might be necessary as new topics emerge in future interviews, which relate to similar, but yet unnoticed patterns in earlier interviews. It is key not to get stuck in the analysis process but also to avoid jumping to conclusions. Especially if you have an approaching deadline, this is not an easy task! I furthermore want to point to another challenge regarding coding that is of special importance when working in teams. The instructor of a course I attended once had the participants of different research groups code the same interview on their own and present the results in class. The outcome was stunning: although all four participants in my group used the same data, each of us came up with different codes, even for the same sentence. For instance, a paragraph about the time the triple disaster of March 11 occurred was coded with ‘3.11’, ‘the day of’, ‘earthquake’ and ‘natural disaster.’ This may have severe consequences when other team members are searching for a certain aspect or code within a set of interviews. The way a segment is coded points to different aspects of the interviewees’ experiences, even if the same segment is marked. Of course, more than one code can be attached to a segment and codes may also overlap sometimes, but when you are working in groups, it is important to agree upon a coding system in which the meaning of a code is explained in a way that other members and people not involved in the project may understand and apply later on. This is often done in the form of codebooks. In sum, although coding is time-consuming, it proved to be a helpful method to structure my data and to compare my two cases. While new software allows researchers to code video or audio files as they are, I still recommend transcripts for larger projects. In that way, it is easier to compare different interviews or other types of data, and to revise files without losing sight 345

Julia Gerster of the bigger structure. Furthermore, coding requires researchers to stick to their data. This prevented me from jumping to conclusions or turning to topics that were never raised in the interviews. For beginners in Japanese Studies, coding may seem challenging, especially when having to work with sources in Japanese. Nevertheless, it is a great opportunity to thoroughly research all kinds of data while staying close to the evidence.

346

Further reading Benaquisto, Lucia (2008): Codes and coding. In: Given, Lisa M. (ed.): The SAGE encyclopedia of qualitative research methods. Thousand Oaks, CA: Sage, pp. 86–88. Charmaz, Kathy (2006): Constructing grounded theory: A practical guide through qualitative analysis. London: Sage. Elliott, Victoria (2018): Thinking about the coding process in qualitative data analysis. In: The Qualitative Report 23, No. 11, pp. 2850–2861. Gibbs, Graham R. (2013): A Discussion with Prof Kathy Charmaz on Grounded Theory. Interviewed by Graham R Gibbs at the BPS Qualitative Social Psychology Conference, University of Huddersfield, U.K., September 14-16 2013. www.youtube.com/watch?v=D5AHmHQS6WQ, [Accessed 19 October 2020]. Glaser, Barney G./Strauss, Anselm L. (2009): The discovery of grounded theory: Strategies for qualitative research. London: Aldine Transaction. Schensul, Jean J./LeCompte, Margaret D. (2013): Essential ethnographic methods: A mixed methods approach. Lanham, MD: AltaMira Press. Thomas, Gary/James, David (2006): Reinventing grounded theory: Some questions about theory, ground and discovery. In: British Educational Research Journal 32, No. 6, pp. 767–795. Timmermans, Stefan/Tavory, Iddo (2012): Theory construction in qualitative research. In: Sociological Theory 30, No. 3, pp. 167–186. Yi, Erika (2018): Adding codes to your data: Qualitative coding tools review. www.medium.com/@project ux/adding-codes-to-your-data-qualitative-data-coding-tools-review-8aa44221382f, [Accessed 10 July 2020].

References Asahi Shimbun (1976): Karen-san no songenshi saiban. Shinu kenri mitomeru: Amerika no shūsai kōsai sekai hajime no hanketsu. Asahi Shimbun, April 1. Benaquisto, Lucia (2008): Codes and coding. In: Given, Lisa M. (ed.): The SAGE encyclopedia of qualitative research methods. Thousand Oaks, CA: Sage, pp. 86–88. Charmaz, Kathy (2006): Constructing grounded theory: A practical guide through qualitative analysis. London: Sage. Dey, Ian (1999): Grounding grounded theory: Guidelines for qualitative inquiry. San Diego, CA: Academic Press. Elliott, Victoria (2018): Thinking about the coding process in qualitative data analysis. In: The Qualitative Report 23, No. 11, pp. 2850–2861. Gerster, Julia (2018): The online-offline nexus: Social media and ethnographic fieldwork in post-3.11 Northeast Japan. ASIEN 149, pp. 14–32. Gerster, Julia (2019): Beneath the invisible cloud: Kamishibai after 3.11. between disaster risk education and memorialization. In: Amfiteater, Journal of Performing Arts Theory 6, pp. 64–82. Gerster-Damerow, Julia (2019): The ambiguity of kizuna: The dynamics of social ties and the role of local culture in community-building in post-3.11 Japan [Dissertation]. Berlin: Freie Universität Berlin. Glaser, Barney (1978): Theoretical sensitivity. Mill Valley, CA: The Sociology Press. Glaser, Barney G./Strauss, Anselm L. (2009): The discovery of grounded theory: Strategies for qualitative research. London: Aldine Transaction. King, Andrew (2008): In vivo coding. In: Given, Lisa M. (ed.): The SAGE encyclopedia of qualitative research methods. Thousand Oaks, CA: Sage, pp. 473–474. Mead, George Herbert (2015): Mind, self and society: From the standpoint of a social behaviorist. Chicago, IL: University of Chicago Press. Meagher, Caitlin (2017): Constructing an interior public: Uchi and soto in the Japanese sharehouse. In: Home Cultures 15, pp. 113–136. Meagher, Caitlin (2018): Make yourself at home: Dreams and realities in a Japanese sharehouse [PhD thesis]. University of Oxford. https://ora.ox.ac.uk/objects/uuid:729fa4e2-bfda-43c5-912d-c7eff445bfd8, [Accessed 22 April 2020]. Meagher, Caitlin (2020): Inside a Japanese sharehouse: Dreams and realities. London: Routledge.

347

References Otani, Izumi (2010): ‘Good manner of dying’ as a normative concept: ‘Autocide’, ‘granny dumping’ and discussions on euthanasia/death with dignity in Japan. In: International Journal of Japanese Sociology 19, pp. 49–63. Rosenberger, Nancy (2001): Gambling with virtue: Japanese women and sense of self in a changing nation. Honolulu, HI: University of Hawai‘i Press. Rosenberger, Nancy (2013): Dilemmas of adulthood: Japanese women and the nuances of long-term resistance. Honolulu, HI: University of Hawai‘i Press. Rosenberger, Nancy (2017): Young organic farmers in Japan: Betting on lifestyle, locality, and livelihood. In: Contemporary Japan 29, No. 1, pp. 14–30. Schütze, Fritz (1983): Biographieforschung und narratives Interview. In: Neue Praxis 13, No. 3, pp. 283– 293. Shimada, Shingo (2007): Die Erfindung Japans: Kulturelle Wechselwirkung und nationale Identitätskonstruktion. Frankfurt a.M.: Campus. Spoden, Celia (2015): Über den Tod verfügen: Individuelle Bedeutungen und gesellschaftliche Wirklichkeiten von Patientenverfügungen in Japan. Bielefeld: Transcript. Spoden, Celia (2017): Well-being and decision-making towards the end of life: Living wills in Japan. In: Holthus, Barbara/Manzenreiter, Wolfram (eds.): Life course, happiness and well-being in Japan. New York, NY: Routledge, pp. 221–237. Spoden, Celia (2020): Deciding one’s own death in advance: Biopower, living wills, and resistance to a legislation of death with dignity in Japan. In: Contemporary Japan 32, No. 1, pp. 63–82. St. Pierre, Elizabeth A./Jackson, Alecia Y. (2014): Qualitative data analysis after coding. In: Qualitative Inquiry 20, No. 6, pp. 715–719. Strauss, Anselm L. (1987): Qualitative analysis for social scientists. Cambridge: Cambridge University Press. Strauss, Anselm L./Corbin, Juliet (1990): Basics of qualitative research: Grounded theory procedures and techniques. London: Sage.

348

Chapter 13 How to systematise texts: Qualitative content and frame analysis Celeste L. Arrington

1. Introduction Fieldwork yields a plethora of materials and texts, ranging from interview transcripts and notes to policy deliberations, council minutes, blog posts, photos and news stories. How does the researcher systematically analyse such diverse sources? Qualitative content analysis and frame analysis are methods through which researchers can examine such materials’ formal structure, content, meaning, latent features and context. Qualitative information about a corpus of materials can also be transformed through coding or cross-tabulation into a form amenable to quantitative analysis. Scholars from across the Social Sciences, Law, Public Health, Humanities and other disciplines use qualitative content analysis and frame analysis. They extract, summarise and interpret qualitative information and reconstruct meanings from diverse texts and from unspoken facets of communication. This chapter discusses such qualitative content analysis and frame analysis, with a focus on research about Japan. It describes how to conduct qualitative content and frame analysis and assesses the strengths and weaknesses of these analytical tools to aid students and researchers who might consider using them. These methods, which are often used in combination, can illuminate important features of texts, such as shifts in issue framing, how honorifics (keigo) are used, the social construction of meaning, what tactics groups adopt or which actors define issues. Additionally, this chapter considers how qualitative content analysis differs from quantitative content analysis and when it can be productively combined with quantitative tools. Prior chapters addressed earlier stages in the research process, including how to formulate a research question and link it to broader scholarly debates or how to balance trade-offs in sampling decisions. While this chapter focuses on the analytic and data-presentation stages, it acknowledges how interconnected they are with the processes of research design and data collection. Since categorising qualitative information involves abstracting from or interpreting complex and changing realities, analysts should be thinking about methods of analysis and how to operationalise concepts during data collection. I write this chapter as an American scholar of comparative politics, specialising in Japan and the Koreas. Drawing mainly from my home discipline of Political Science, but also from Sociology and Law, this chapter offers a few examples of how to develop, apply and refine analytical categories, share data and document the analysis. The three accompanying essays elaborate on this with further examples from other social scientists trained in Europe and Japan.

349

Celeste L. Arrington

2. What are qualitative content analysis and frame analysis? Qualitative content analysis describes the process of systematically considering and interpreting materials to make meaningful descriptive, correlational and causal inferences or generalisations (Mayring 2007). Most often, the materials are written texts and transcribed speech. Such texts may be generated by the researcher in interaction with human subjects (e.g. focus group and interview transcripts, site visit notes or emails) or collected by the researcher from other sources (e.g. newspaper articles, posts on social networking sites, politicians’ speeches, archival records, or social movement flyers). Some scholars call for integrating analysis of visual and non-verbal content, such as images on protest posters (Cooper-Cunningham 2019). Frame analysis is generally considered a subset of qualitative content analysis and discourse analysis (see Eder-Ramsauer/Reiher, Ch.14). It focuses more specifically on investigating how people ‘locate, perceive, identify, and label’ events and experiences; Erving Goffman’s influential work defined frames as ‘schemata of interpretation’ (Goffman 1974, p. 21). Frames are socially constructed and contested. Analysing them illuminates how shared understandings of the world develop (Benford/Snow 2000). Studying frames also helps us comprehend such phenomena as collective action or policy change since issue framing mobilises people by defining a problem and associating it with possible solutions. Analysis entails examining materials in a consistent, rule-based way that fits with the research question and can be validated by other scholars. The researcher takes texts and categorises or sorts their component parts, which can then be analysed and presented qualitatively or quantitatively. Such coding (see Meagher, Ch. 12) may be manual or done with computer software. Researchers consider not just words or phrases and their meanings but also the context in which the communication occurred, other properties of the text like length or formality and implied or missing factors. They may both summarise evident content of materials or illuminate latent content such as information about context or language (Mayring 2000). Description and summary help reduce the volume of content so that researchers can draw conclusions about it (Schreier 2012, p. 5). Such analysis aims to systematically discern patterns in the materials and to reconstruct people’s understanding of the meaning of their past words and actions. But it also involves interpretation, which is why it is important for the researcher to be transparent about her analytic framework and position or potential impact on the materials. For instance, my identity as a non-Japanese has led interviewees to both sugar-coat their replies at times and to reveal more than they normally would to an in-group member. Additionally, researchers must recognise that subjects may exhibit omissions or revisions in recalling the past. Triangulation (see Hommerich/Kottmann, Ch. 10) or using data from various sources, such as contemporary media accounts or secondary scholarship, can help fill in gaps and alleviate biases.

350

Chapter 13 How to systematise texts Key terms Coding: the processes of categorising, summarising or describing texts or parts of texts to highlight and/or compare important features of the texts; coding may be done manually or with computer assistance Corpus: a collection of written, visual or aural materials to be analysed Data Access and Research Transparency (DA-RT): a movement to enhance evidence-based Social Science by developing clear standards and procedures for making explicit how and on what bases researchers reached their conclusions and for sharing research materials Diagnostic evidence: information that prior knowledge and/or theory suggests is important for establishing some sequence or relationship Frame analysis: the investigation of how people perceive and interpret events and experiences; frames are socially constructed and contested shared understandings of the world Mechanisms: generalisable statements about how and why a variable contributes to some outcome Operationalisation: deciding what indicators signal the presence of some abstract concept or how to measure a theoretical concept in a reliable and valid way Process tracing: assembling and interpreting information about the unfolding of events in a temporal sequence Qualitative content analysis: methods that aim to systematically examine and interpret patterns in the materials to summarise them and reconstruct what people understood were the meanings of their past words and actions Quantitative content analysis: systematically assigning elements of communication to categories and using statistical methods to analyse relationships among those categories Texts: written or transcribed communication; visual or non-verbal forms of expression; generated by the researcher interacting with human subjects (e.g. interview transcripts, site visit notes) or collected by the researcher from extant sources (e.g. newspaper articles, social media posts) Triangulation: gathering and evaluating evidence for something from multiple sources

3. Getting started: Don’t wait! Often, the processes of conducting fieldwork and managing data receive more attention than the important task of analysing that data. Close knowledge of the case and the broader context is essential for persuasive content analysis. Hence, analysis is most effective when done in dialogue with and parallel to data collection. Analysing and collecting data iteratively results in better definition and operationalisation of concepts and more context-conscious analysis (Kapiszewski et al. 2015, pp. 335–337). Operationalising concepts refers to the process of taking an abstract idea or social phenomenon and figuring out what indicators signal its presence for observers, as Kai Schulze outlines (see this chapter, Ch. 13.3). It can also improve interviewing and other data collection. Iteration is more feasible with ethnographic or archival research methods than, for instance, with survey experiments, which are hard to adjust or repeat once fielded.

351

Celeste L. Arrington As a comparative politics specialist, I have found moving iteratively among national or cultural contexts to be analytically productive because comparison often lays bare factors that are taken for granted in one context. It can also help me discern when I have enough data and can stop interviewing or transcribing. I often adopt qualitative analysis in conjunction with the method of paired comparison, which matches relatively similar cases or contexts to reduce the range of potential confounding factors and focus on the core research question (Tarrow 2010). Most often I compare cross-nationally between Japan and Korea, across issue areas and across time. Japan and Korea have the advantage of being relatively similar socio-cultural contexts, but as an American I also bring preconceived ideas of these two societies to my research. For researchers studying only Japan, iteratively comparing across regions, issue areas, groups or time periods can similarly reveal factors taken for granted. Conducting research and analysis over iterations and remaining attuned to such differences bolsters findings. Qualitative content analysis and frame analysis also frequently rely on a comparative sensibility towards recognising patterns, such as recurring phrases or relationships among themes. And researchers constantly compare their evidence with conventional wisdom in existing scholarship. Thus, be explicit and conscious about your comparative referents. Do not wait until after fieldwork to think about and analyse your materials because you may end up duplicating work or wishing you had collected different sorts of materials if you do. Organising texts and preparing them for analysis can be a daunting task, which is a second reason why content analysis should not be put off until after data collection. Qualitative research tends to produce rich and voluminous data. Good organisation greatly facilitates analysis, and software can, too. For example, Atlas.ti or MAXQDA are software packages for qualitative research, but even labelling files in systematic ways and using Microsoft Word’s Navigation or hyperlink functions can help with organisation. Naming files clearly with the interviewee’s name/code and the interview date and method as you conduct the interviews will also expedite subsequent analysis. I recommend writing down specifics about the material (see checklist below). It is important to adhere to clear procedures when recording supplemental observations about non-verbal communication in interviews, descriptions from site visits or other latent information. Do not forget to repeatedly back up your materials and adhere to your procedures for maintaining their security (see McLaughlin, Ch. 6). Effectively managing materials makes them easier to use as data and renders them verifiable to subsequent researchers (see Reiher/Wagner, Ch. 16).

• • •

•

A checklist for documenting how you organise and manage materials1 Who is working with the materials (i.e. research assistants or translators for transcription)? What steps were taken to collect the materials (i.e. what search terms and sampling methods were used)? In what formats are the materials stored? How are you maintaining the materials, including their confidentiality (e.g. an informed consent protocol)? How are the materials organised? How are you handling all the documentation you produce in analysing the materials (i.e. codebooks, annotations, tabulations or other metadata)?

1 For guidelines, see the Qualitative Data Repository website, https://qdr.syr.edu/guidance/managing.

352

Chapter 13 How to systematise texts Once the researcher settles on a significant, clear and researchable research question and data collection procedures, there are interrelated decisions she must take with regard to the analysis of such data.

• •

•

• •

Decisions to be made before data analysis What are the study’s core goals and the ontology behind the research design and mode(s) of analysis? What units of analysis and sampling strategies should the researcher focus on, considering the challenges of qualitative content analysis, especially in complex languages like Japanese? By what means should concepts be operationalised and the categories for coding derived: inductively from the texts or deductively from theories, many of which were developed in the West? Should the researcher use manual coding methods or computer-assisted methods? How can researchers present qualitative materials and findings in credible and persuasive ways?

The next sections discuss these choices. There are rarely right or wrong choices. Instead, as detailed below, researchers must weigh disciplinary norms, feasibility, language capabilities, access to sources and other factors. The best choices are ones that are transparent and well-considered.

4. The logic behind the methods and research design To make persuasive inferences, a researcher must first consider her objectives. Research involving qualitative materials can have conceptual, descriptive, correlational or causal aims. Qualitative content analysis and frame analysis can contribute to them all. Different logics underlie each, and researchers should be clear about their aims. Yet social scientists working with qualitative materials often combine the positivist approach, which seeks generalisable explanations, with the interpretivist’s eye, such as when they consider why interviewees answered the way they did or when they convert responses into discrete categories (Mosley 2013, pp. 10– 11; see Vogt, Ch. 2). This section compares interpretivist and more positivist causal logics.

4.1 Interpretivist accounts Qualitative content analysis and frame analysis are ideal for constructionist or interpretivist approaches to studying complex socio-political phenomena, which are often difficult to ‘measure’ with quantitative methods. These methods also work for positivist approaches, which assume research can uncover truths and demonstrate correlations or test hypotheses to infer causation. Socio-constructionist accounts, however, explore how communicative processes and

353

Celeste L. Arrington narrative forms constitute common knowledge or shared understandings (e.g. Neumann et al. 1992). Interactions and communication—the texts to be analysed—produce and reproduce shared understandings of events or issues. Socio-constructionist accounts elucidate the recursive processes by which this occurs. In studying mobilisation processes, for example, Anna Wiemann (see this chapter, Ch. 13.1) used qualitative content analysis to explore how representatives of social movement organisations talked about dynamics in their organisational networks in Japan after the 3.11 triple disaster. Similarly, Schulze (see this chapter, Ch. 13.3) leverages qualitative content analysis to investigate the social construction of Japan’s foreign policy identity towards China. Reconstructing how shared meanings emerge from social interactions is not easy in one’s mother tongue because it requires careful and self-aware observation and interpretation. For those of us non-native speakers of Japanese, the nuances of communication and connotations of words and phrases in a different cultural context can be challenging to navigate. In my experience, embedding oneself in a topic and its discourses before, during and after fieldwork helps mitigate these challenges. With the growing availability of online blogs and other ‘minimedia’ from groups (i.e. newsletters or public social media pages), this process need not involve extended field research trips. As Levi McLaughlin (see Ch. 6) notes, groups often engage with each other on social media and react to events publicly in ways that can be useful to researchers because they include commentary on the language with which they or the mass media choose to characterise an issue. For example, I initially used a literal translation of the word victim (higaisha) to specify the type of movements I studied in my first book (Arrington 2016) but quickly replaced it with ‘directly-affected party’ (tōjisha) after reading in activists’ online blogs how they self-identified. From there, I uncovered an entire social discourse that used the term tōjisha to assert both innocent suffering and agency before the law (Nakanishi/ Ueno 2003). I recommend reading as much as possible in Japanese regarding one’s topic from secondary scholarship, media coverage and websites or social media pages. In particular, content gleaned through digital ethnography, which describes long-term immersion in and study of online sites, can supply linguistic insights to non-native speakers (Han 2015).

4.2 Correlational or causal accounts While qualitative content analysis and frame analysis have elective affinities with interpretivist approaches, they can also be useful for the more causal or correlative inquiry of positivist scholars. For instance, they can illuminate the mechanisms by which X causes Y and the conditions under which a cause is activated. Identifying patterns and relationships among variables is also foundational for middle-range theorising (see Okano, Ch. 3). Content analysis can contribute to causal inference by drawing attention to potential alternative explanations, including causal theories held by participants in a process. In addition, qualitative materials are critical for process tracing, which is a common method for assembling and interpreting information about the unfolding of events in a temporal sequence, as well as about context and causal processes (Collier 2011). Longitudinal qualitative content analysis reveals patterns or recurring relationships among phenomena or actors, which can supply diagnostic evidence or

354

Chapter 13 How to systematise texts evidence that prior knowledge and/or theory suggests is important for making causal inferences and lay the groundwork for qualitative network analysis. Compared to its quantitative counterpart, qualitative content analysis often produces more diverse, abstract and interconnected findings. Quantitative content analysis focuses more on numerically tabulating coded features of a corpus of materials (Schreier 2014, p. 173). Qualitative and quantitative content analyses are often productively paired, however (see Hommerich/ Kottmann, Ch. 10). For example, qualitative content analysis enhances concept development, which is critical for accurate measurement and contextualisation in quantitative methods like survey research (Gallagher 2013). Qualitative content analysis and frame analysis can also be transformed into data amenable to statistical analysis, such as with phrase frequencies, variance between categories of interest, or data on co-located words. For instance, in conjunction with qualitative content and frame analysis, I compared the pairwise correlation coefficients for coverage of four issues across Japan’s and Korea’s main newspapers to show the relatively high degree of news content homogeneity in the Japanese mediascape (Arrington 2017). Amy Catalinac (2016) also conducted quantitative text analysis of Japanese campaign manifestos alongside qualitative fieldwork to explain why Japanese ruling party politicians paid more attention to national security issues after the 1994 electoral reforms. As researchers consider the logic behind their research design, they should be wary of feasibility and issues related to limited sample size, missing data, the difficulty of coding spoken and unspoken features of texts in the same way, or conceptual stretching. For example, since manual coding in qualitative content analysis requires multiple close readings, the researcher might only be able to include several dozen speeches in her analysis. But she can enhance the credibility of her findings by being explicit about her sampling strategy to show that it did not introduce systematic biases (Mayring 2014). For one article, for instance, I conducted frequency analysis of all articles published in the past twenty years in Japan’s top three newspapers mentioning the words ‘North Korea’ and ‘abduction’, but then supplemented the analysis by closely reading any that included those words in the title to trace shifts in issue framing (Arrington 2018). No method is without drawbacks, but they can complement each other when used in a logical and transparent way. As demonstrated by the multi-method research trend in Political Science and the Social Sciences in general, using diverse methods produces more robust conclusions.

5. Units of analysis Intimately related to the process of identifying a research question and the logic behind the research design is a second issue: specifying the units of analysis (Mayring 2014). After considering what types of materials you have or can access (e.g. newspaper articles versus editorials, websites created by the social movement organisation versus by its supporters) as a quantitative researcher might do with descriptive statistics, then you can embark on qualitative content analysis and/or frame analysis. Here are five possible levels (not mutually exclusive) at which a researcher could begin to parse the materials and the frames they contain (based on Ferree et al. 2002, p. 50–52). 355

Celeste L. Arrington •

•

•

•

356

The article or speech: One might analyse newspaper articles on an issue or how a prime minister’s official speeches discuss a topic. For example, I examined all of the speeches by Japanese prime ministers that mentioned abductions over twenty years (N=77) to discern trends in how they discussed North Korea’s abductions of Japanese nationals (Arrington 2018). I found that from the late 1990s to early 2002, prime ministers referred to the abductions as ‘suspected’ but also ‘an important issue that concerns the lives and security of our country’s citizens’. In 2002, Prime Minister Junichirō Koizumi (2001–2006) dropped the term ‘suspected’ and began consistently listing the abductions first among the ‘outstanding issues to be addressed with North Korea’ (Arrington 2018, p. 489). The number of speeches and the words on the abductions issue in each speech rose under Prime Minister Shinzō Abe (2006–07, 2012–2020). Additionally, one could analyse whether a newspaper story appears on the front page or includes a photo, both of which tend to draw attention to the story. The speaker: Another form of qualitative analysis investigates differences in interpretation and emphasis among speakers quoted in news articles. For instance, the terminology used by a social movement may start appearing in speech acts by legislators after the movement has gained media coverage. To show a movement’s discursive influence, I traced the emergence of its idea of ‘uniform financial assistance’ (ichiritsukyūsai) for all victims of hepatitis C-tainted blood products, regardless of differences in their symptoms (Arrington 2017, p. 18). From within my corpus of all the articles that mentioned ‘hepatitis C’ in Japan’s main newspapers over a thirty-year period, I closely read a sub-sample of articles from 2007 with hepatitis C in their titles to reveal how news coverage hedged or qualified the victims’ claims about the state and drug makers’ liability less and less as more plaintiffs in the lawsuits revealed their real names during that year. The plaintiffs thereby signalled their empirical credibility, which is an important part of effective issue framing (Snow/Benford 1988). Utterances: This term describes speech acts by a single speaker. At a rally or on a news talk show, any given speaker might speak multiple times. Whereas formal speeches or individuals’ quotes in newspaper articles stand alone, utterances from one speaker may interact with those of others. For instance, when analysing legislative debates over a proposed bill regarding discrimination against people with disabilities in ongoing research, I compare the issue framing that opposition lawmakers use across utterances with that of politicians from the ruling Liberal Democratic Party (LDP). Opposition lawmakers often cite NGOs’ studies in legislative debates, whereas LDP lawmakers tend to cite business groups’ research more. But ruling party lawmakers may shift their terminology in response to opposition pressure. By examining utterances in their context (i.e. the parliamentary debate or a talk show), as Wiemann (see this chapter, Ch. 13.1) does, researchers can examine the recursive processes through which meaning is constructed and phenomena interpreted. The frame or idea: An article or utterance generally includes multiple ideas and sometimes multiple issue framings. The analyst could examine how the idea of second-hand smoking is discussed in a policy council meeting about increasing restrictions on indoor smoking in Japan. I found it framed as ‘smoking harassment’ (shortened to sumo hara in Japanese), violations of children’s and women’s rights to health and clean air, ‘harm to others’ (tasha kigai), and ‘smoking manners’ (kitsuen manā in katakana). These represent ways of thinking about a phenomenon and are commonly the target of frame analysis. Sometimes, the

Chapter 13 How to systematise texts

•

framing devices are visual (Van Gorp 2007). In Japanese, katakana (for foreign words) can visually and aurally emphasise a concept or frame, such as sumo hara. Time period: Detailed chronologies are central to many qualitative research methods. Analysing articles or speeches by time period can help researchers see the relationships between observed data and changes in the political environment unfolding over time.

As noted above, one tendency in qualitative content analysis and frame analysis has been to privilege textual or spoken forms of expression. Yet considering non-verbal forms of communication and signals is integral to understanding context and the social construction of shared understandings. This is especially true in the Japanese language, which arguably involves more unspoken forms of communication than many languages. But visuals can send important signals in many polities. For instance, politicians often wear lapel pins. The ubiquitous blue-ribbon pin on Prime Minister Abe’s lapel signaled his support for the families of Japanese nationals abducted by North Korea; the pins were created as a reminder that only the sky and the Sea of Japan connect abductees to their families in Japan (Arrington 2018). Hence, they can be interpreted as content that a wearer intends to convey and as expressing the activists’ issue framing. Future research should more systematically examine illustrations, graphic text and other visual effects in conjunction with texts (Cooper-Cunningham 2019). Posters and photos used in political activism or websites and social media posts are promising targets for qualitative content analysis and frame analysis that consider both textual and visual features. Interpreting communication and meanings at each or several of these levels of analysis and via non-verbal signals is something we do unconsciously daily. But qualitative content analysis aims to make such processes more systematic in order to assess more data or texts than one normally would in everyday contexts. Explicit procedures for coding or interpreting materials and clear units of analysis also facilitate knowledge building by making the findings verifiable to other researchers, even if every scholar brings their own priors to the research process.

6. Coding, categories and concepts Coding schemas make the analyst’s interpretations and the link between concepts and categories explicit. Concepts are abstract Social Science phenomena, whereas categories are ‘those aspects of the material about which the researcher would like more information’ (Schreier 2014, p. 174). Researchers must consider whether they develop the coding framework in inductive or deductive ways, or through a combination of both. Inductively developing categories involves abstracting up from the materials and identifying labels that emerge from the texts. The process may be theoretically informed but is also deeply linked to the research question and materials and is more subjective (Mayring 2014). On the other hand, deductively applying categories entails using prefabricated concepts from extant theories and can thus more readily connect to existing scholarship but may not match the Japanese context or language. There are trade-offs to either inductive or deductive category development, and many researchers combine both. Often, researchers toggle between deductively applying concepts from existing scholarship and inductively developing concepts based on close study of a given con357

Celeste L. Arrington text. For example, in my current research about tobacco control policy and the anti-smoking movement in Japan and Korea, I look for ‘rights-based’ terminology, which is predicted by theories about transnational advocacy networks and socialisation through international interactions (Finnemore 1993; Keck/Sikkink 1998). Yet I also code local issue framings, such as the right to hate or dislike smoking (ken’enken), which Japanese activists coined in the 1970s. Social Science concepts are often multifaceted. Thinking through how to operationalise them is a critical but difficult step in research that is intertwined with coding. Words have different connotations or valence in different contexts, which can complicate operationalisation and categorisation. As a non-native Japanese speaker, I have used my status as an outsider to ask interview subjects to explain their self-perceptions or terminology at length (Steinhoff 2006; see Kottmann/Reiher, Ch. 7). For instance, I study Japanese lawyers who go beyond just serving clients to advocate for policy reforms (Arrington 2014). The Western research literature calls such activity ‘cause lawyering’, which is not normal English parlance and has no Japanese translation (Sarat/Scheingold 2006). Yet I describe cause lawyering in Japan by qualitatively analysing how such lawyers conceive of their work, how they divide their day, how they interact with clients (and what clients say about them), what their backgrounds are, and what social networks they belong to. One question I asked repeatedly was about labels found in the media and academic scholarship, such as jinken bengoshi versus jinkenha bengoshi (human rights lawyer versus human rights-type lawyer). The latter apparently softens the communist overtones of the former, which emerged in the 1960s as communist and socialist party affiliated lawyers led challenges to LDP dominance and the social costs of rapid industrialisation; one of my interviewees explained that it is ‘like red versus pink.’ The technique of citing my outsider status to probe such nuances emerged through trial and error over the course of interviewing, which is why iteration between data collection and analysis is helpful. Deep familiarity with the issue and context helps with valid conceptualisation and operationalisation. Some ideas may be nested in specific contexts, which should also be coded. As Schulze and Wiemann (see this chapter, Ch. 13.1; 13.3) note, it is important to be self-reflective in coding and to redefine categories or labels if needed. Additionally, consider what interviewees or other texts omit, distort or imply, because doing so can reveal shared understandings, practices and unspoken assumptions about past interactions and developments (Fujii 2010). The ‘evidence of absence’ can be compelling (see Kottmann/Reiher, Ch. 7). Interviewees may also adjust their replies after you press them with follow-up questions. Triangulating among various sources can help you discern when such omissions or revisions are relevant and deliberate or irrelevant and accidental. In short, coding classifies texts or parts of texts into predefined categories and sub-categories. By doing so, the researcher compiles, summarises and describes important features of the texts —as defined by the research question and associated theory. The clearer the coding process is, the more other scholars can evaluate and replicate it. So, document and annotate every step.

358

Chapter 13 How to systematise texts Key terms Inductively developing categories: abstracting up from the materials and identifying labels that emerge from the texts Deductively applying categories: using prefabricated concepts from extant theories, which can thus more readily connect to existing scholarship but may not match the Japanese context or language

7. Manual versus computer-assisted analysis and feasibility Qualitative content analysis and frame analysis simplify rich data and reduce the volume of data through summary, interpretation and abstraction. Getting to see the forest for the trees requires that the researcher decide whether hand-coding is feasible or whether computer assistance or some systematic sampling strategy is needed. How coding occurs ultimately depends on research objectives, journals’, theses’ or other outlets’ space limits, the researcher’s capabilities and disciplinary norms. Increasingly, most qualitative and quantitative software work with Japanese language texts, although Japanese’s multiple scripts and writing directions and lack of spaces between words were challenges for optical character recognition (OCR) software (Catalinac/Watanabe 2019). KH Coder is one free software program for quantitative content analysis that was developed in Japan and handles Japanese.2 In general, computer programs can facilitate analysis, especially with voluminous material, but they cannot replace the researcher (Mayring 2000, p. 18). Deciding how to organise and categorise the materials still falls to the researcher. Hence, software is most profitably used after the analytic framework is well-developed. Computer-Assisted Qualitative Data Analysis Software (CAQDAS) includes programs like NVivo, Atlas.ti or MAXQDA (see Chiavacci, Ch. 11; Gerster, Ch. 12.3; Rosenberger, Ch. 12.1). Microsoft Access is another common database creation and management system that can help organise and analyse qualitative data, as Schulze (see this chapter, Ch. 13.3) describes. Wiemann’s essay (see this chapter, Ch. 13.1) offers examples of how MAXQDA helped with transcribing interviews and annotating passages for easy retrieval in analysis, but she warns about the importance of being self-reflective and revising coding schemas if needed. For instance, she combined deductive and inductive approaches to coding to avoid having overly individualised categories. Having the researcher first code a subset of the materials manually and then checking its correlation with computer-assisted coding helps to validate findings and identify errors or biases in either the researcher or the computer analysis. As Wiemann notes (see this chapter, Ch. 13.1), the upfront costs of learning how to use the software can be daunting but have long-term payoffs. Among others, the National Science Foundation has offered workshops on text analysis tools to alleviate some such challenges. However, data access and research transparency initiatives (see below) note that sharing materials can be difficult across qualitative research software.

2 See website at https://khcoder.net/en/index.html.

359

Celeste L. Arrington Even without such software, most texts will be stored on and accessed through computers. General-purpose programs like Word or Excel are also useful for searching, annotating, navigating and tabulating a corpus of texts. Ultimately, categorising and coding texts requires close reading by the researcher at some stage, but computers can facilitate these tasks.

8. Presenting qualitative content analysis and frame analysis findings Researchers present qualitative content and frame analyses in different ways, depending on the research question and other constraints such as word limits of journals. Findings can remain qualitative, presented in narrative or discursive form. They include block quotes from interviews, summaries of how a cluster of editorials framed a particular issue, or dissections of the indicators of any given issue framing. Block quotes have the advantage of enabling other scholars to assess and confirm the author’s conclusions. They let the sources speak for themselves. But, perhaps especially in my field of Political Science, much qualitative research bumps up against the reality of limited space in journal articles, books or master’s theses, and the fact that analysis should reduce the materials’ volume and complexity to arrive at conclusions. Rather than valuing thick description, the premium in Political Science is on demonstrating that one’s findings are generalisable, at least within certain scope conditions. Hence, researchers must select small portions of text in order to illustrate their findings, which can seem like cherry-picking. One solution is to explicitly explain why you think a particular quote is important and/or representative, such as that the speaker wrote an influential first draft of a particular bill or that seven other interviewees echoed the terms used by that speaker. Area Studies publications may value thick description more. Nevertheless, being explicit about the representativeness of one’s qualitative findings leads to more credible and legitimate social inquiry. Findings can also be presented quantitatively, such as by tallying the frequency of certain words and phrases, creating tables or graphs or conducting statistical analyses on categories. For example, I investigated how the media environments in Japan and Korea affect the amount of media coverage a social movement gets, as well as the message and reach of that coverage (Arrington 2017). I presented graphs tallying cross-temporal variations in the frequency of newspaper articles that mention hepatitis C as a proportion of the ‘news hole’ (i.e. the average number of articles per month devoted to health policy). To retain some of the richness of the qualitative content analysis, these quantitative analyses were then coupled with close reading and manual coding of editorials and movement newsletters to discern issue framing patterns and the message of news coverage.

360

Chapter 13 How to systematise texts

9. Reliability and validity: Annotating and documenting the analysis Social scientists are increasingly recognising that documenting every stage of data collection and analysis makes research understandable for and assessable by others. Standards and priorities differ across disciplines, so I recommend informing yourself about your discipline’s best practices for data and research transparency. The American Political Science Association has extensively deliberated the discipline’s Data Access and Research Transparency (DA-RT) initiative in recent years, with many resources available online.3 Journals and funding agencies are also increasingly requiring data management plans and online methods appendixes to enhance research transparency and data preservation and access (see Reiher/Wagner, Ch. 16). An important component of such appendixes is not only to describe the corpus of texts and how it was generated, as a researcher using quantitative data sets would, but also to detail how the author analysed those texts. As with transparency in citations, the researcher should take detailed notes along the way about how materials were collected and analysed. Annotating texts enables the researcher to keep track of what she has done and leaves a record for subsequent scholars. As Schulze (see this chapter, Ch. 13.3) notes, relevant details include the data source, background information about the source, such as who produced it or where it is archived, any sampling procedures (i.e. is the source representative of a class of similar sources or is it unique), and the unit of analysis. Annotation should also include descriptive or causal inferences and interpretations. Working with co-authors or research assistants raises the issue of inter-coder reliability. Training and close monitoring can help avoid inconsistencies. Have research assistants carefully annotate their coding decisions. As David Chiavacci (see Ch. 11) notes, involving Japanese researchers whenever possible can help overcome the challenges of interpretation in another language. Because the interpretation involved in coding qualitative materials is sometimes considered—perhaps incorrectly—more subjective than most quantitative analyses, transparency about coding and interpretation is vital. Many political scientists working with qualitative materials concur that protecting subjects or respecting copyright remains a challenge in the face of calls for transparency. But disciplines are developing best practices about how to share information about one’s sources without compromising them. For example, methods appendixes often contain tables of interviewees and their professions but not their names or a narrative explanation of the sensitive archival materials gleaned from subjects. Ultimately, data sharing entails judgement calls by the researcher. As I found in my research about stigmatised disease populations in Japan and Korea, some interview subjects may express a willingness to be quoted by name—especially in English-language scholarly publications that few Japanese will read—but anonymising all interviews avoids disparities among interviewees (Arrington 2019). With any research involving human research participants, be attuned to potential ethical difficulties or risks that you might unleash by sharing content from them. For some topics, information may already be in the

3 For reports, see https://www.qualtd.net/. Additional options in Political Science include a transparency appendix (TRAX) or an Active Citation Index (ACI), which was originally proposed by Andrew Moravcsik (2010). The idea is to provide hyperlinks to more detailed, annotated versions of the citations to support claims and inferences. If possible, the links also lead to the actual sources on which the analysis is based.

361

Celeste L. Arrington public domain, such as in newspaper editorials or court rulings, enabling researchers to protect human subjects (see McLaughlin, Ch. 6; Reiher/Wagner, Ch. 16).

10. Summary This chapter has examined qualitative content analysis and frame analysis as methods for researchers studying Japan. A key take-away is that scholars adopting both qualitative and quantitative methods must make choices about what to analyse and how. These choices include the study’s core goals, the units of analysis and sampling strategies, the materials to be analysed, categories for coding, types of analysis—manually or computer-assisted—and the style of presentation. The chapter detailed how making these choices depends on the research question, the researcher’s access to sources and Japanese language abilities, the study’s objectives and disciplinary norms. They are not easy choices. The most effective and persuasive analysis leverages transparent and self-reflective iteration, conducting data collection, analysis and writing in dialogue with each other and the existing literature. This level of discretion can be daunting but still yield valid inferences and interpretations, if done as transparently and selfconsciously as possible at every step.

362

13.1 Qualitative content analysis: A systematic way of handling qualitative data and its challenges Anna Wiemann

When researchers are at the point of applying qualitative content analysis, their research project has already gone through important stages in a research process: the designing of the research framework and gathering data in the field. They are then at a point where their study’s research question, theoretical background and chosen methodology gear into each other and are about to show first results. A qualitative content analysis (especially one that is computer-assisted) can bring order into data collected in the field, facilitates their systematic interpretation and, in so doing, leads to reliable findings. This essay gives insight into why and how I used computer-assisted qualitative content analysis in my PhD thesis Networks and mobilization processes: The case of the Japanese anti-nuclear movement after Fukushima (Wiemann 2018). After a short overview of the method itself (which adds to the main chapter to some degree), I will focus on how I applied this technique as well as the difficulties I encountered when doing so and how I dealt with them. The essay closes with some recommendations for readers who decide to use computer-assisted qualitative content analysis. Qualitative content analysis developed in the first half of the 20th century at the same time but in contrast to quantitative content analysis (Mayring 2014, p. 18ff.; Schreier 2014, p. 173). At that time—in the context of a growing media landscape—scientific and political interest in media content grew significantly and researchers had to deal with the analysis of large amounts of data material. Both qualitative and quantitative content analysis are systematic ways of interpreting (mostly but not exclusively textual) data. While quantitative content analysis is defined as an objective, numerical way of analysing the ‘manifest content of communication’ (Schreier 2014, p. 171), qualitative content analysis represents an interpretive form of analysis where text evaluation and coding or categorisation rely on the cognitive processes of the researcher (Kuckartz 2014, p. 38). Qualitative content analysis thus comprises the meaning and context of communicative data ‘by assigning successive parts of the material to the categories of a coding frame’ (Schreier 2014, p. 170). A coding frame consists of ‘at least one main category and at least two subcategories’ (Schreier 2014, p. 174). Main categories are ‘those aspects of the material about which the researcher would like more information’, while sub-categories ‘specify what is said in the material with respect to these main categories’. The main categories thus relate to the cognitive interest of the research (in other words to the research question). The sub-categories bring together and structure data passages that provide different aspects of meaning to the main category. The structure of the main categories and sub-categories reflects the coding frame; hence, coding is the act of sorting data passages into categories according to their meaning and context. In

363

Anna Wiemann qualitative research, such processes of categorisation are necessary to reach a level of abstraction that enables the researcher to make a statement related to the purpose of the research. In my eyes, the most comprehensive procedure on how to conduct qualitative content analysis has been proposed by Mayring (2014). Philipp Mayring’s suggested workflow starts with a preparatory phase, consisting of finding a research question, selecting and characterising the data material, and determining the direction of the study (including the characteristics of the data producer and the communicative context) as well as the units of analysis. This is followed by what is at the heart of the method: the coding of the data material. Coding of data material can be performed deductively or inductively. If necessary, both approaches can be combined. When categories are defined deductively, it means that they are derived from theory, that is, before the researcher works with the data. Consequently, establishing a coding guideline before starting coding helps with applying deductive categories consistently throughout the process of analysis. Defining categories inductively means categorising data passages while going through and interpreting the data at hand. Accordingly, this approach derives categories from the meaning of the collected data. Before inductive coding, however, Mayring recommends determining a guideline for category definition and the level of abstraction that needs to be reached. Deductive and inductive coding both require a double check of the categories applied and the coding guideline after 10 to 50% of the data material has been coded. Additionally, after the coding has been finished, the reliability of the categorisation should be doubly tested. Thus, a systematically applied qualitative content analysis is a method with a very high level of traceability, reliability and validity and equates to the highest social scientific standards— which is why I chose to employ this systematic interpretive research technique. In my study on network mobilisation processes of social movements after the nuclear accident in Fukushima in 2011, I analysed social movement actors’ (collective) behaviour based on the perceived quality and dynamics of network relations. I applied qualitative content analysis to analyse their perceptions of the intergroup networks in which they are embedded and the dynamics within these networks before and after the nuclear disaster. Looking at such network dynamics allowed me to assess mobilisation processes at the intergroup level of a social movement. I used qualitative content analysis to analyse interviews and other documentary data to systematically discover the meaning organisational actors attribute to their relational embeddedness in the broader movement field. Thus, I used qualitative content analysis as a tool for a qualitative network analysis. In my thesis, I triangulated qualitative network analysis with quantitative network analysis to understand the interrelation between different but overlapping networks and the network centrality of certain actors as well as to draw sensible network boundaries and to visualise the networks. Quantitative network analysis—which is not the focus of this essay—is a methodological toolkit (which is sometimes also regarded as a theoretical paradigm) to analyse relational patterns among defined units of analysis by means of mathematical (matrix algebra and graph theory) and computational models which aims at displaying graphic network imagery (for more on quantitative and qualitative social network analysis as both a method and a theory, various authors in Scott 2011). In sum, I contributed to the field of social movement research by providing an analytical framework for looking at mobilisation processes through networks at the meso-level after a disruptive event. Having a clear understanding of the analytical perspective and the phenomenon under study was important for me in approaching the field but also for the subsequent qualitative content analysis of the collected data. In my case, the richest data I gathered in the field were qualita364

Chapter 13 How to systematise texts tive semi-structured interviews with representatives of social movement organisations. When interpreting and coding these data, I had to keep in mind that I recorded the voices of individuals, but these individuals—through their individual lenses—spoke on behalf of organisations/ groups. They explained the embeddedness of their groups from their point of view. This was a bias I had to bear in mind throughout the analysis. Before I started my systematic content analysis, however, the data, especially the interviews necessitated transcription to enable a systematic computer-assisted analysis. Depending on the amount of spoken data, transcription can be very time-consuming. There are several software programmes for transcription and content analysis. My choice for transcribing the interviews in question fell on f4transkript because it is compatible to the coding software MAXQDA, which I used to support my qualitative content analysis. However, through recent programme updates, it is now (2019) also possible to transcribe directly into MAXQDA. To ease transcribing, I recommend investing in a foot pedal to be able to run and stop the audio while typing. I decided to code my data with the help of qualitative data analysis software because it facilitates looking at the material systematically and it is possible to retrieve all original data passages belonging to one code with just one click. Moreover, it is possible to connect analytical memos to codes as well as to certain text passages or groups of memos, which is very helpful in getting to different levels of abstraction in the process of coding, analysing and writing. The software also allows several researchers to work on the same data material, if one is working in a research group. Some universities provide qualitative data analysis software to their graduate students free of charge. In the case of MAXQDA, it is also possible to buy a price-reduced student licence. After transcription and the installation of coding software, and in accordance with the research design, the researcher needs to decide on a coding strategy. In my case, I had a clear research question concerning relational patterns of social movement organisations before and after Fukushima, and I wanted to discover the meaning the organisations attribute to their relations. I thus followed a deductive–inductive coding procedure. Based on my research interest, I established the two first categories deductively: ‘network antecedents’ and ‘networks’, which referred to relations pre- and post-Fukushima. The coding guideline for the category ‘network antecedents’ envisaged all passages relating to past relational patterns, past issues of contention, but also to the emergence of two coalitional meso-level networks which were founded after Fukushima and which serve as case studies. The guideline for the category ‘networks’ on the other hand referred to actual working procedures, ideas and frames, and cooperation and conflict within the operating coalitional networks of the case study. As a result, I was able to induce sub-categories inductively to uncover the meaning the actors attribute to their relations. The combination of deduction and induction while coding had the advantage of me not losing sight of the research question to be answered, while avoiding only coding according to my own expectations. Instead, the meaning derived from the data themselves. During my first round of inductive coding, it proved a challenge to find meaningful sub-categories. Every interviewee had his or her own style of expression, and after finishing the first round of coding, I realised that the established categories in this first round were far too individual and lacked the necessary level of abstraction to add points to my argumentation. I thus decided to redo the inductive coding—within the two deductive categories—from scratch. This proved to be very time-consuming, but it was also a chance for me to become more familiar with the data material. I thus recommend revising and specifying established codes regularly, 365

Anna Wiemann just as Mayring (2014) suggests. If one has to go back to the start, coding may be frustrating at times. But repetition and revision help researchers to develop a feeling for the data, the speakers and the structure of ideas. The coding process, in other words the really confronting the data with the research question, might be painful and it is difficult to estimate the time needed for it, but it is a necessary part of reliable social scientific research. In conclusion, even though it is an interpretive form of data analysis, computer-assisted qualitative content analysis enables high traceability of the qualitative data analytical procedure. Data passages and their coding may be retrieved with just one click even years after the initial research. Moreover, the systematic approach as proposed by Mayring also increases the researcher’s awareness of her or his way of looking at the data, and thereby it facilitates transparency of the methodological approach to the future reader. Qualitative content analysis may also be used for a large variety of research questions. The only disadvantage I see is that when working with qualitative content analysis software for the first time, you may find it a timeintensive and at times frustrating procedure until it runs smoothly. However, in my opinion, applying computer-assisted qualitative content analysis increases the explanatory power of qualitative research immensely, so that any hardship in the process is more than worthwhile.

366

13.2 Analysis of biographical interviews in a transcultural research process Emi Kinoshita

Area Studies often face challenges in using foreign languages, which accompany the whole research process from fieldwork to publication (Kruse et al. 2012). This is connected not only to language proficiency but also to understanding and participating in academic discourses in a foreign tongue, which always include specific perspectives. In this sense, Area Studies require transcultural translation of academic discourses, since researchers move repeatedly between home and foreign field sites as well research contexts (Bachmann-Medick 2016). This short essay reports on some of the transcultural challenges I faced in applying biographical methods.

Confusion For my dissertation at a university in Tokyo (2005–2010), I explored educational ideas during the societal changes in East Germany since the 1970s by focusing on people’s lives. I chose a biographical approach, ‘life history (seikatsushi)’, as a central method in the attempt to write a lived history of educational ideas of ‘ordinary people’ (Nakauchi 1992). I interviewed East German teachers and educators in German, which is my second foreign language after English, while my mother tongue is Japanese. In the early phase of my project, I did several interviews, as suggested in Japanese textbooks on life history (Nakano/Sakurai 1995) and life stories (Sakurai 2002), which emphasise respect for the interviewees and their words to embrace their subjective lifeworld. Questions on collecting data such as ‘how to conduct an interview, where and with whom’ were dominant issues in Japanese literature around 2000. To analyse interviews, I examined, for example, repeated episodes and discrepancies among interviews and, if applicable, written life documents, as discussed in Japanese literature. At the same time, I read relevant German academic literature, which also focused on ‘ordinary people’ using similar methods. Besides this, research concepts developed in Germany enabled me to consider educational theories such as socialisation and human development by means of empirical research. Japanese language, German language and Anglophone discourses had seemed to stand harmoniously with each other, until I began my research stay as a doctoral student in Germany. ‘How do you analyse your data?’—This was the most difficult question for me at a workshop in Germany. It sounded quite simple, but I was not able to give an adequate answer that day. I explained some of the techniques mentioned above as analysing methods, but they were understood as a concept, framework or even a moral view. Basic terms were different as well. I realised that I shared only little theoretical background with other participants from Europe. Today, I must confess that I was not conscious of the analytical procedure or its concrete steps.

367

Emi Kinoshita But neither data collection nor analysis were explained as a systematic procedure in Japanese literature at that time. Then, I was unsure of myself and wondered whether my research would contribute to any academic community. Thus, I started comparative reading of German, Japanese and Anglophone literature on qualitative and particularly biographical research to identify differences (Kinoshita 2010). Sometimes, it was a long detour, but it gave me the chance to identify the theoretical and field-related significance of my research in terms of methodology. Around the millennium in Japan, life historians introduced constructivist views and developed life story interviews as their method. Hereby, they discussed the societal implications of qualitative research, mostly reflecting critically on the researchers’ position and power asymmetry in research situations. In this context, the thoughts on oral history of the anthropologist Yanagita Kunio, the founder of Japan’s life history method in the 1920s, was often referred to. It aimed at listening to voices of the ‘ignoramus’ about their daily lives and shares the motivation of empowerment with feminist or postcolonial researchers in the U.S. Their approach to research as a form of ‘societal’ critique framed the theoretical concepts of life history in Japan. In contrast, German discourses on qualitative or biographical research pursued questions on specific procedures since the 1970s. The traceability of data analysis was intensively examined, and procedures of data collection and analysis were clearly distinguished. Until today, the latter have been discussed in different ways—based for example on the objective hermeneutics, the documentary methods or the content analysis respectively—to suit specific research questions and theoretical foundations. Anglophone research discourses, such as the grounded theory approach, were introduced to accurately analyse detailed communication in interviews or to focus on ‘social’ or interactional situations in the field. There has hardly been any exchange between German and Japanese biographical or even qualitative research. Even though a famous German handbook by Uwe Flick (2002) was translated into Japanese, it has not been discussed critically in Japan. Anglophone discourses are also perceived contrastively in both countries. The discourse referred to in the German context almost only aimed at improving given analysis procedures of social interaction, whereas in Japan, they were actively introduced to constitute a research fundament for societal critiques in Japan. Besides this, not all the terms are used identically. For example, the Japanese term life history could be replaced by Lebensgeschichte in daily use but only partly carries its meaning as a technical term. In turn, a more extensive German term Biographie does not cover the meaning of life history.

Decision The divided discourse challenged my research practice in between the German-speaking and Japanese-speaking research contexts. Comparative reading helped me to decide on a suitable method with which to analyse my data. As for the biographical approach in Educational Studies, researchers in Germany establish their own concepts on biography from a pedagogical perspective, whereas the ones in Japan directly adopt the biographical approach to Educational Studies (Kinoshita 2010, p. 160). German biographical interviews enabled me to contribute to the critical development of educational theories. They allowed me to concentrate on personal

368

Chapter 13 How to systematise texts narratives first and to add societal aspects in the Japanese context later. For me, this seemed honest towards ‘ordinary people’ and applicable to both research contexts. In consequence, I adopted the narrative interview developed by German sociologist Fritz Schütze (1983) in the 1970s as a method, which is barely known outside Germany (Fiedler/ Krüger 2016; Nohl 2010). It includes detailed procedures for interviewing and analysing. This method corresponded to my framework and made it possible for me to bridge the gap between the discourses in Germany and Japan due to its following two characteristics. Firstly, my analysis focused on the whole span of someone’s life and suited my interest in the development and socialisation process in Educational Studies. Secondly, the controlled method made it possible for me to overcome the heated debates among constructivists in Japan, as storytelling in interviews is understood as a social construct and embraces it not as a hurdle against fact-finding but as a natural condition. In my opinion, Schütze’s analytical procedure opens up the results for further discussion on societal changes. As for interviewing, Schütze’s approach strictly controls the interview situation through socalled ‘ascetic listening’, which lets an interviewee’s narrative expand as it should. Data is interpreted in a structured way in four steps, so that the ‘interpretation pattern’ (Deutungsmuster) of the narrating person is traced as a ‘process structure on the course of a life’ (Verlaufskurven). First, the different types of expressions are identified in the transcriptions of the interviews, so that narrative, i.e. storytelling passages, can be distinguished from description or argumentation. This narrative text is segmented to obtain a formal structure of the biography (formale Textanalyse). Second, the content and performance of the narrative is analysed by the researcher focusing on its structure. Thereby, he or she examines, for example, how each story connects to each other and which markers are set to make each passage relevant (e.g. time, place, the person in question, reason). On this basis, typical characteristics of biographical stages are ascertained (strukturelle inhaltliche Beschreibung). Building on this, the researcher transfers the results on each biographical stage to the other stages and then determines the whole structure of the biography (analytische Abstraktion). This is finally interpreted as biographical knowledge in terms of the specific context of the interviewee (Wissensanalyse). Schütze concentrates on deep analysis of each case, whereas today’s qualitative biography research often aims at case comparison (Nohl 2010, p. 197f.). Beyond analysing my narrative interviews based on Schütze’s method, I needed to methodologically bridge German and Japanese methodological discourses. I risked classically introducing ‘the more developed’ procedures from the West to the East and thus reproducing the asymmetry of knowledge. However, both methodological trends belong to different strands. For example, the common attempt to identify a ‘genuine’ narrative in an interview has different motivations: to deliver the voices of the ‘ignoramus’ in Japan or to reconstruct processes of someone’s life in Germany. Both could not be simply combined, but the focus on narratives accounts for the discourse of educational theories in international research contexts. The transformation of someone’s development is sought after in Germany, while in Japan the focus on turning points in someone’s life is considered to be epiphany. This insight into a specific moment in life does not suggest a ‘perfect fusion’ but invites, for example, further discussions on socialisation. Thus, Schütze’s method had to be ‘translated’ for Japanese discourses. Here is the transcultural challenge in Area Studies. Due to researchers’ transcultural migration between the field and the home and participation in foreign discourses, the original theories need ‘translation’ into ‘understandable’ conceptual language during research practice. 369

Emi Kinoshita Analysing interviews in transcultural exchange could never be realised without research workshops in a biography research community. To a workshop, one can bring one’s transcripts and exchange interpretation with other participants. A relaxed, cooperative atmosphere with tea and cookies enables researchers to widen their interpretations and to explore various, sometimes controversial perspectives with specific methods. I started to recognise characteristics in the German language which I had not learned in language schools, lectures and seminars: e.g. colloquial or regional expressions, specific uses of personal pronouns which are not really seen in Japanese, and markers of certain nuances.

Prospects Comparative reading is a kind of a detour. Nevertheless, it is still a part of my transcultural research practice, because methodological perspectives can support deeper understandings of and critical reflection on a research field and discourses. This journey goes on. I am participating in a comparative qualitative project on school lessons in Germany now and see similar challenges. Comparative inquiries in Germany and Japan have different academic and practical motivations as well institutional origins, which are accompanied by different theoretical concepts and methodological approaches. This makes our intercultural dialogue more complex but, at the same time, allows us to reflect on our perspective in terms of theory and methods in transcultural research practices.

370

13.3 Qualitative content analysis and the study of Japan’s foreign policy Kai Schulze

My main field of research is the normative and ideational basis of Japan’s foreign and security policy. I have investigated changes in Japan’s foreign policy identity, matters of securitisation and ontological security particularly in Japan’s China policy. All of these topics deal with complex social phenomena—identities, norms, ideas and perceptions—that are socially constructed. To explain why the method of qualitative content analysis was very helpful and sufficient for my research, I will briefly point out why and how I adopted and executed this method for my doctoral thesis on the changes of Japan’s foreign policy identity towards China in reaction to China’s rise. When I conducted my PhD research on identity constructions in the field of international relations and foreign policy analysis, I often received very harsh and fundamental criticism from senior scholars. This is because the study of international relations at the time was still dominated by positivist-minded scholars that base their research on ‘hard facts’, such as military and economic strength. A large part of the criticism was targeted at the complexity and immateriality of the social phenomena—such as identities—that I was studying. Critics found my research too abstract and purely based on theoretical concepts, which they claimed do not deal with ‘reality’. ‘How can you define and measure something that only exists in the minds of social actors and does not materialise?’ they asked. ‘If you cannot measure the things you research, how can you generate new insights into what is really going on in Japanese politics?’ Thus, I needed to prove that my research on the normative and ideational basis of Japan’s foreign policy met the basic academic requirements of validity. At that time, I was a rather young and inexperienced scholar and, at first, this quite fundamental criticism from established scholars in the field made me very insecure, and I did not really know how to respond. The reason for my insecurity was simple: the critics were right— at least at first sight—because it is difficult to directly measure social constructions, such as identities, norms and ideas. However, the representation of social constructions in Japan’s political discourse can be measured. In fact, they materialise in different forms, such as texts or pictures. In my case, the representations of identities usually materialised in written texts. Therefore, I needed to operationalise my research by adopting methods that allow systematic and intersubjectively comprehensible analysis of text-based sources. At the beginning of my project, I worked my way through the very rich and complex literature about discourse and discourse analysis. This literature provided me with many insights into how to define important actors, how to select primary sources and how to relate them to one another. However, it did not offer a detailed description of how to actually deal with textbased sources in a structured manner. Therefore, I searched for instructions on supplementary methods and techniques that explain how to analyse the texts and how to categorise the infor371

Kai Schulze mation. This is when I came across qualitative content analysis. This method enables researchers to identify and extract the important passages from written sources and to interpret them in a systematic and reliable manner. There are many different forms of qualitative content analysis that can be applied to different types of research. I wanted to investigate the changes in Japan’s foreign policy towards China and identify different levels of the related identity concepts. For the analysis of these identity concepts, I was looking for a data interpretation strategy that is based on the systematisation of patterns, the contextualisation of regularities in the sources and on categorisation. Philipp Mayring (2002; 2008; 2014) calls the categorisation of the defining patterns ‘content-related structuring’. This technique of interpreting different text corpora does not focus solely on special keywords in the texts, but rather searches for passages that show certain similarities regarding their content. Basing their work on these similarities, researchers can code the text and form the categories that are necessary for their research. Categories can be developed in two different ways. They can be derived deductively from existing theoretical concepts or inductively from primary sources. It is important to note that these two categorising systems are not mutually exclusive, but can be combined (Mayring 2014). For my own study, I first deduced two main categories from theoretical definitions of identities. According to social identity theory, identities are defined by the categories of a ‘self’ and an ‘other’ (Abdelal et al. 2006). Thus, in a first step, I extracted passages from my sources that described and defined China as the ‘other’ and Japan as the ‘self’. In a second step, I inductively developed sub-categories to find out how the ‘self’ and the ‘other’ were defined in detail in the documents I analysed. This process of categorising the content of my sources was the most interesting but simultaneously also the most exhausting part of my research. I literally had to go through all the texts individually, and more than once. To do this, it is advisable to find technical support for your analysis that helps you to clearly arrange the individual steps in your research. By now, there are many computer-based programmes available that can support your analysis. For the practical implementation of qualitative content analysis, I, for example, took advantage of Patricia Steinhoff’s (2009) approach based on the databank system Microsoft Access. In this system, I created a main table and first inserted basic data, such as the title of the source, the date of publication and additional information about the source, like its author, issuing authority and place of publication. For example, I distinguished between sources depending on whether they were published on behalf of the Prime Minister or the Minister of Foreign Affairs or on whether they were published online or as a printed version. Then, I also created columns and text boxes in the databank. The first text box was rather large. This enabled me to insert the whole passage from the respective source, the whole speech or the statement. I did so because it is very important to always keep the context in mind in which certain assertions and definitions are made. Otherwise the possibility of misinterpretation is very high. In addition, I inserted text boxes in which I only collected the passages that directly defined the construction of ‘self’ and ‘other’ for each individual source. I labelled them ‘Japan’ and ‘China’ respectively. Below these text boxes, I created boxes where I labelled the different categories that I developed from the data. Every single category was visible on each form and could be tagged with a check mark whenever a passage was grouped in its respective category. One big advantage of the software for qualitative content analysis is the fact that the main table can constantly be adjusted. This is helpful because I had to adjust the

372

Chapter 13 How to systematise texts databank over and over again in the process of data analysis; for example, when I identified new categories. The process of the actual analysis was quite wearing. I began with the first text, extracted the important parts and grouped them into content-related categories. It is very important that every category is precisely defined, so that every extracted part of the texts only fits into one category. When I found a pattern in one of the texts that fitted into two or more categories, I either needed to define a new category, in which it fitted or had to redefine the categories I had already formed. After I had finished with the first text, I proceeded to the second text, and the process started again. Hence, whenever a new pattern became visible, the categories needed to be adjusted accordingly. While analysing one text after another, I ordered and reordered content-related categories until I had identified certain stable patterns in the whole sample. This process was very frustrating at times, especially when I found a new pattern only after I had already analysed a huge number of sources. In this case, and in fact this happened fairly often, I had to go through all the sources again and look whether or not the pattern could be found there as well and if I needed to redefine my categories. This admittedly caused stress because I was working with a sample of texts in a rather complicated language like Japanese. But it is exactly this constant defining and redefining of the categories that brings about advancement in the analysis and the research process as a whole. Therefore, I encourage researchers to be very patient and to maintain their motivation to go through their sources over and over again until they find stable categories. To maintain my motivation in this process, I received good advice from one of my supervisors to trick myself a little bit. It was pretty simple: I just had to reward myself for finding a new category. That could mean buying a book I had wanted to buy for a long time, going out for a beer with friends, taking a day off or going to my favourite rāmen (noodle soup) eatery. Whatever makes people happy works! And then, I went back to the sources and worked on them again. Once I had finished with this process, I had a great data set and I could start the last step of the analysis. For example, I could see when a certain category first occurred in the discourse and when it vanished. This enabled me to identify various patterns of change and continuity in addition to the content-related results I had gained from the categories alone. With regard to changes in Japan’s foreign policy identity towards China, for example, I gained valuable insights into changing patterns of representation of Japan as the ‘self’ and China as the ‘other’, and how these were related to the broader discourse on the ‘rise of China’. By filtering the content until the patterns became visible, I was able to measure the significance of the respective categories in and for the discourse. In contrast to quantitative approaches to content analysis, however, this is not a matter of pure quantity, but rather of regularity and continuity. A pattern that was expressed in various sources and regularly mentioned within a certain time span could be considered to be more important for the discourse than a category that was only mentioned occasionally, even in the unlikely case that the absolute quantity of mentions might be higher. A second determinant for measuring the importance of a certain category was strongly content-related. When a category, for example, served as a precondition for another category, I considered it to be more important than a category that was built upon its foundation or a category that existed unconnected to the other categories, given that the other parameters were similar. Basing my work on this data, I was able to create a hierarchy of the different categories and definitions of ‘self’ and ‘other’ as well as the changes to this hierarchy over time.

373

Kai Schulze These were just a very few examples of the richness of data one can gain from a well-structured qualitative content analysis. In any case, it is a wonderful method to employ when dealing with social phenomena like identities, which require an interpretative approach based on empirical, text-based analysis. And although you still might not be successful in convincing positivist-minded scholars, you will at least have a strong empirical and methodological basis for an intersubjectively comprehensible analysis with which to validate and confirm your conclusions and results. The time-consuming and hard work you need to put into an analysis like this is definitely worth it.

374

Further reading Benford, Robert D./Snow, David A. (2000): Framing processes and social movements: An overview and assessment. In: Annual Review of Sociology 26, No. 1, p. 611–639. Goffman, Erving (1974): Frame analysis: An essay on the organization of experience. New York, NY: Harper & Row. Lindekilde, Lasse (2014): Discourse and frame analysis: In-depth analysis of qualitative data in social movement research. In: Della Porta, Donatella/Malthaner, Stefan (eds.): Methodological practices in social movement research. Oxford: Oxford University Press, pp. 195–227. Mayring, Philipp (2014): Qualitative content analysis: Theoretical foundation, basic procedures, and software solution. Klagenfurt: SSOAR. https://nbn-resolving.org/urn:nbn:de:0168-ssoar-395173, [Accessed 08 April 2020]. Schreier, Margrit (2012): Qualitative content analysis in practice. Los Angeles, CA: Sage Publications.

References Abdelal, Rawi/Herrera, Yoshiko M./Johnston, Alastair I./McDermott, Rose (2006): Identity as a variable. In: Perspectives on Politics 4, No. 4, pp. 695–711. Arrington, Celeste L. (2014): Leprosy, legal mobilization, and the public sphere in Japan and South Korea. In: Law & Society Review 48, No. 3, pp. 563–593. Arrington, Celeste L. (2016): Accidental activists: Victims and government accountability in Japan and South Korea. Ithaca, NY: Cornell University Press. Arrington, Celeste L. (2017): The access paradox: Media environment diversity and coverage of activist groups in Japan and Korea. In: Journal of East Asian Studies 17, No. 1, pp. 69–93. Arrington, Celeste L. (2018): The mutual constitution of the abductions and North Korean human rights issues in Japan and internationally. In: Pacific Affairs 91, No. 3, pp. 471–497. Arrington, Celeste L. (2019): Hiding in plain sight: Pseudonymity and participation in legal mobilization. In: Comparative Political Studies 52, No. 2, pp. 310–341. Bachmann-Medick, Doris (2016): The translational turn. In: Bachmann-Medick, Doris: Cultural turns. New orientations in the study of culture. Berlin: De Gruyter, pp. 175–209. Benford, Robert D./Snow, David A. (2000): Framing processes and social movements: An overview and assessment. In: Annual Review of Sociology 26, No. 1, p. 611–639. Catalinac, Amy (2016): Electoral reform and national security in Japan: From pork to foreign policy. Cambridge: Cambridge University Press. Catalinac, Amy/Watanabe, Kohei (2019): Nihongo no ryōteki tekisuto bunseki. In: Waseda Institute for Advanced Study Research Bulletin 11, p. 133–143. https://scholar.harvard.edu/files/amycatalinac/files/te xt_analysis_in_japanese.pdf, [Accessed 08 April 2020]. Collier, David (2011): Understanding process tracing. In: PS: Political Science & Politics 44, No. 4, pp. 823–830. Cooper-Cunningham, Dean (2019): Seeing (in)security, gender and silencing: Posters in and about the British women’s suffrage movement. In: International Feminist Journal of Politics 0, No. 0, pp. 1–26. Ferree, Myra Marx/Gamson, William A./Gerhards, Jürgen/Rucht, Dieter (2002): Shaping abortion discourse. Cambridge: Cambridge University Press. Fiedler, Werner/Krüger, Heinz-Hermann (eds.) (2016): Fritz Schütze. Sozialwissenschaftliche Prozessanalyse: Grundlagen der qualitativen Sozialforschung. Opladen: Barbara Budrich Publishers. Finnemore, Martha (1993): International organizations as teachers of norms. In: International Organization 47, No. 4, pp. 565–597. Flick, Uwe (2002): Qualitative Sozialforschung. Reinbek: Rowohlt. Fujii, Lee Ann (2010): Shades of truth and lies: Interpreting testimonies of war and violence. In: Journal of Peace Research 47, No. 2, pp. 231–241. Gallagher, Mary (2013): Capturing meaning and confronting measurement. In: Mosley, Layna (ed.): Interview research in Political Science. Ithaca, NY: Cornell University Press, pp. 181–195. Goffman, Erving (1974): Frame analysis: An essay on the organization of experience. New York, NY: Harper & Row. Han, Rongbin (2015): Defending the authoritarian regime online: China’s ‘voluntary fifty-cent army’. In: The China Quarterly 224, pp. 1006–1025.

375

References Kapiszewski, Diana/MacLean, Lauren M./Read, Benjamin L. (2015): Field research in Political Science: Practices and principles. Cambridge: Cambridge University Press. Keck, Margaret E./Sikkink, Kathryn (1998): Activists beyond borders: Advocacy networks in international politics. Ithaca, NY: Cornell University Press. Kinoshita, Emi (2010): Kindai kyōiku to seikatsushi: Kenkyū hōhōron no nichidoku hikaku ga shisa suru mono. In: Hikaku Kyōikugaku Kenkyū 40, pp. 158–178. Kruse, Jan/Bethmann, Stephanie/Niermann, Debora/Schmieder, Christian (eds.) (2012): Qualitative Interviewforschung in und mit fremden Sprachen: Eine Einführung in Theorie und Praxis. Weinheim: Beltz Juventa. Kuckartz, Udo (2014): Qualitative text analysis: A guide to methods, practice, and using software. London: Sage. Mayring, Philipp (2000): Qualitative content analysis. In: Forum: Qualitative Social Research 2, No. 1, pp. 1–28. Mayring, Philipp (2002): Einführung in die qualitative Sozialforschung: Eine Anleitung zu qualitativem Denken. Weinheim: Beltz Verlag. Mayring, Philipp (2007): On generalization in qualitatively oriented research. In: Forum: Qualitative Social Research 8, No. 3. DOI: 10.17169/fqs-8.3.291. Mayring, Philipp (2008): Qualitative Inhaltsanalyse: Grundlagen und Techniken. Weinheim: Beltz Verlag. Mayring, Philipp (2014): Qualitative content analysis: Theoretical foundation, basic procedures, and software solution. Klagenfurt: SSOAR. https://nbn-resolving.org/urn:nbn:de:0168-ssoar-395173, [Accessed 08 April 2020]. Moravcsik, Andrew (2010): Active citation: A precondition for replicable qualitative research. In: PS: Political Science & Politics 43, No. 1, pp. 29–35. Mosley, Layna (2013): ‘Just talk to people?’ Interviews in contemporary Political Science. In: Mosley, Layna (ed.): Interview research in Political Science. Ithaca, NY: Cornell University Press, pp. 1–29. Nakanishi, Shōji/Ueno, Chizuko (2003): Tōjisha shuken. Tōkyō: Iwanami Shoten. Nakano, Takashi/Sakurai, Atsushi (eds.) (1995): Raifu hisutorī no shakaigaku. Tōkyō: Kōbundō. Nakauchi, Toshio (1992): Zōhoban. Atarashii kyōikushi: Seidoshi kara shakaishi e no kokoromi. Tōkyō: Shinhyōron. Neumann, W. Russell/Just, Marion R./Crigler, Ann N. (1992): Common knowledge. Chicago, IL: University of Chicago Press. Nohl, Arnd-Michael (2010): Narrative interview and documentary interpretation. In: Bohnsack, Ralf/ Pfall, Nicolle/Weller, Wivian (eds.): Qualitative analysis and documentary method in international education research. Opladen: Barbara Budrich Publishers, pp. 195–217. Sakurai, Atsushi (2002): Raifu sutōrī no shakaigaku. Tōkyō: Serika Shobō. Sarat, Austin/Scheingold, Stuart A. (eds.) (2006): Cause lawyers and social movements. Stanford, CA: Stanford University Press. Schreier, Margrit (2012): Qualitative content analysis in practice. Los Angeles, CA: Sage Publications. Schreier, Margrit (2014): Qualitative content analysis. In: Flick, Uwe (ed.): The SAGE handbook of qualitative data analysis. London: Sage, pp. 170–183. Schütze, Fritz (1983): Biographieforschung und narratives Interview. In: Neue Praxis 13, No. 3, pp. 238– 293. Scott, John (ed.) (2011): The SAGE handbook of social network analysis. London: Sage. Snow, David A./Benford, Robert D. (1988): Ideology, frame resonance, and participant mobilization. In: Klandermans, Bert/Kriesi, Hanspeter/Tarrow, Sidney G. (eds.): From structure to action: Comparing social movement research across cultures. Vol. 1: International social movement research. Greenwich: JAI Press. Steinhoff, Patricia G. (2006): Radical outcasts versus three kinds of police: Constructing limits in Japanese anti-emperor protests. In: Qualitative Sociology 29, No. 3, pp. 387–408. Steinhoff, Patricia G. (2009): Seminar in qualitative content analysis spring 2009. www2.hawaii.edu/~stei nhof/soc715lassignments.htm, [Accessed 17 December 2019]. Tarrow, Sidney G. (2010): The strategy of paired comparison: Toward a theory of practice. In: Comparative Political Studies 43, No. 2, pp. 230–259. Van Gorp, Baldwin (2007): The constructionist approach to framing: Bringing culture back in. In: Journal of Communication 57, No. 1, pp. 60–78. Wiemann, Anna (2018): Networks and mobilization processes: The case of the Japanese anti-nuclear movement after Fukushima. München: Iudicium.

376

Chapter 14 How to understand discourse: Qualitative discourse analysis Andreas Eder-Ramsauer and Cornelia Reiher

1. Introduction How people make sense of the world they live in and how they act upon this particular understanding are important questions for Social Science and Area Studies researchers alike. Discourse analysis (DA) allows us to study the various meanings people attribute to specific phenomena and situations, and to analyse how meaning is mediated or constructed through language, images or routinised practices. Thus, DA is a useful tool for studying how people understand, represent and thereby shape the world. In order to analyse discourse, concise knowledge of linguistics and societal, economic, political and historical contexts is important. This is why Area Studies researchers are particularly qualified to carry out DA due to their language competence and their contextual knowledge of the area being studied (Schäfer 2011). Japan scholars, for example, have produced deeply embedded empirical studies by analysing discourse on ethnicity (Kamada 2010), the reconstruction of the Tohoku region after the 3.11 triple disaster (Samuels 2013), education policies (Rear/Jones 2013) and ‘Japaneseness’ (nihonjinron) (Befu 2001). There are various types of DA and each one is rooted in a different set of theoretical assumptions about the concept of ‘discourse’. This makes it especially tricky for researchers who want to apply DA for the first time. Analysis can also be daunting due to the rather open research design of DA. Thus, in this chapter, we provide some guidance on what discourse analysis is, introduce different types of DA and discuss how DA has been used in the study of Japan. Our goal is to help students and early career researchers to make informed decisions on whether DA is helpful for their research project and in answering their research question(s). We will also give some advice on how to conduct DA. After a brief introduction to the concept of discourse and DA, we break down the process of DA into its individual steps to explain them in more detail. For a hands-on introduction of DA, we will focus on Critical Discourse Analysis (CDA) as one specific form of DA, and particularly draw on Norman Fairclough’s work. We chose this type of CDA since it is prevalent in the analysis of political discourse in Japan. Also, because Fairclough’s CDA approach has a strong focus on linguistic methods of text analysis, we think that Area Studies specialists are particularly well-equipped to conduct this type of DA due to their language skills. Although DA is complex and diverse, this very short introduction is kept simple in order to motivate readers to engage with it. Before you start a DA project, studying different discourse theories and related concepts of discourse in more detail is necessary. Therefore, we introduce literature 377

Andreas Eder-Ramsauer and Cornelia Reiher on discourse theories and corresponding types of DA for further reading. Throughout the chapter, we discuss examples of DA in the study of Japan. The three essays that follow this introductory chapter illustrate how scholars have analysed discourse on and in Japan from a Political Science, sociological and anthropological perspective, respectively (see Heinrich, Ch. 14.3; Schad-Seifert, Ch. 14.1; White, Ch. 14.2).

2. What is discourse? Discourse is not a static concept. It is the continuous attempt of different actors to stabilise— often only for a certain time—meaning and interpretations and to establish a specific order of knowledge (Keller 2013, p. 2). Discourse is therefore always related to power. It can help to sustain and reproduce but can also transform the social status quo (Angermuller et al. 2014, p. 362). Discourse also has the power to define what can be said and what must not be said (Foucault 1970). Thus, while on the most abstract level, discourse is language that contributes to shaping the world and people’s understanding of the world, it is also shaped by its economic, political and social conditions. While there are differences in how discourse scholars evaluate the scope of language’s influence on society and vice versa, all agree that language matters in social processes (Fairclough 2001, p. 122). In other words, language and the world it shapes or represents are mutually interlinked (Fairclough 2003, pp. 3–4). Because of this emphasis on language, concepts of discourse are often based on a constructivist worldview and structuralist or post-structuralist perspectives.1 Many discourse scholars, however, consider discourse to be more than just language. From their perspective, discourse is not only an interrelated set of texts (language), but also includes the practices of its production, dissemination (sociocultural contexts) and reception (cognition) (Lindekilde 2014, p. 198). Through language, specific actors in society might try to create a certain type of discourse to promote their understanding of a given problem (language). However, depending on how other actors understand the problem (cognition), discourse and, possibly, society can change (sociocultural context) (Keller 2013, p. 13). Thus, by studying discourse, researchers can analyse power relations and/or specific conditions in a society and study how and why they persist or change. It is important to note that researchers define discourse differently, but concepts of discourse are mostly not mutually exclusive. Japan scholars, for instance, have combined quite different theories and concepts of discourse. In their CDA on education policy and work skills in Japan, David Rear and Alan Jones (2013, p. 375) defined discourse as ‘a particular way of representing certain parts of the world’ and call it an ‘important strategic resource for reproducing or challenging the social, political and economic status quo’. Although they mainly followed Fair1 According to constructivists, all knowledge about the world is constructed from human experience instead of reflecting external or ‘transcendent’ realities (Harvey 2012). Structuralists think about the world as a system of interrelated objects, concepts or ideas and aim to identify the structures that underlie human behaviour, thoughts and practices (Blackburn 2008). Post-structuralism emerged out of a critique of Marxist economic reductionism, structural determinism and methodological individualism. Post-structuralists argued that no social order is fully structured through one underlying mechanism (economic structure or social concepts) that explain everything and that there is always the potential for political intervention (Panizza/Miorelli 2013).

378

Chapter 14 How to understand discourse clough’s concept of discourse, they merged it with ‘nodal points’ and ‘hegemony’ (see Figure 14.1)—concepts developed by two other discourse theorists, Ernesto Laclau and Chantal Mouffe (2006). In her Feminist Poststructuralist Discourse Analysis (FPDA) on ethnic and gendered embodied identities of adolescent girls in Japan with Japanese and ‘white’ mixed-parentage, Laurel Kamada (2009; 2010) understands discourse as connected to and structuring social practices (Kamada 2009, p. 330). In his CDA on Japanese national identity in the Yomiuri Shimbun’s coverage of the Fukushima nuclear disaster, James W. Tollefson (2014) equates discourse with ideology (see Figure 14.1). Thus, conceptualisations of discourse differ, but concepts from different discourse theories can be combined, when necessary to answer a particular research question—if they do not contradict each other. However, researchers should always explain their understanding of discourse and related concepts in a transparent way. Figure 14.1: Three concepts related to discourse

Three concepts related to discourse Hegemony is temporarily fixed meaning (i.e. a particular set of ideas) that expands a discourse into a dominant collective understanding, social orientation and/or action (Akerstrom Andersen 2003, p. vii; Donoghue 2018, p. 396; Torfing 1999, p. 101). Ideology: There exist quite different definitions of ideology (see for example Althusser 1971; Gramsci 1971; Mannheim 2015), but in its most basic sense, the term refers to a set of beliefs shared by members of social groups (van Dijk 1995, p. 248). Nodal Points organise discourse by creating centres, like ‘people’ in populist discourse, for example. They assign meaning to elements in a discourse and partially fix it through creating relations between these discursive elements (Žižek 1989, pp. 95–97).

3. What is discourse analysis? How can social scientists analyse how people understand, represent and shape the world or society? One option is to analyse how people communicate and make sense of the world through language. There are not only various definitions of discourse, but also many variations of DA. The latter are always based on a specific understanding of discourse. In general, DA seeks to systematically understand and empirically analyse discourse. Despite its plurality, DA can be broadly defined as the study of how meaning, and thus social reality, is discursively constituted or represented by analysing the interplay between individual texts, discourse (interrelated sets of texts) and its historical, social, economic, cultural and political contexts (Phillips/Hardy 2002, pp. 3–4). One specific form of DA, Critical Discourse Analysis (CDA), aims to understand this interplay between discursive elements, discourse and society in order to critically study social problems, power and inequality (van Dijk 1997). However, not every analysis of discourse is DA, as DA does not simply summarise, paraphrase or quote texts, but follows specific rules to uncover concepts and structures (van Dijk 1997, p. 32). Depending on the concept of discourse, different dimensions of discourse (see above) can be analysed. DA can study the interplay between the ‘discursive units’ (the text), the ‘discursive 379

Andreas Eder-Ramsauer and Cornelia Reiher practices’ (production, dissemination, reception) and the ‘social practices’ (the wider order of discourses in society), for example (Fairclough 1992). It can start at the macro-level and move to the micro-level of individual talks, texts or specific contexts (top down). But DA could also begin at the micro-level and analyse discourse bottom up, beginning with sounds, words, gestures, meanings or strategies (Keller 2013, pp. 13–14). It can sometimes be difficult to distinguish between discourse theory and DA, but also between DA, content analysis and frame analysis. Simply put, discourse theories develop general theoretical perspectives on the linguistic constitution of reality. DA empirically investigates discourse (Keller 2013, p. 3). Therefore, discourse theory is the foundation of DA, but not DA itself. Content and frame analysis (see Arrington, Ch. 13) on the other hand are ‘focused subvariants of discourse analysis’ (Lindekilde 2014, p. 197). Just like DA, they cast an interpretive perspective on social interaction and can both become methods within DA in which texts are understood in a larger socio-historical context. While content analysis focuses on individual texts, frame analysis scrutinises the strategic intent of texts (Keller 2013, p. 31). Compared to DA, both methods’ analytical scope is rather narrow (Lindekilde 2014, p. 204). In other words, when researchers build larger connections between texts and study how texts relate to each other in a particular context, they are performing DA. As mentioned above, there are many different approaches to DA. We explain CDA in more detail below, because it is often used in research on Japan. But we want to at least mention a few more types of DA here: political discourse theory or post-structuralist discourse analysis builds on Laclau and Mouffe’s (2006) work, as well as on post-structuralists’ ideas, like those by Jacques Lacan (Glynos/Howarth 2007). Culturalist discourse analysis, on the other hand was developed in Sociology in the context of symbolic interactionism2 and builds on the works of Pierre Bourdieu (Keller 2013, p. 33). The Sociology of Knowledge approach to discourse (SKAD) originates from the Sociology of Knowledge theory by Peter Berger and Thomas Luckmann and is interested in the social production, circulation and transformation of knowledge, rather than in power and hegemony (Keller et al. 2018). Handbooks and compendia are an important source of learning about these different types of discourse theory and DA (see for example Angermuller et al. 2014; Gee/Handford 2014; Keller 2013; Schiffrin et al. 2005).

4. How to conduct discourse analysis? Due to the existence of diverse approaches, DA is not restricted to just one method or methodological toolkit. DA is more a research design than an individual method and, thus, theoretical assumptions, research questions and methods should not contradict each other. The assumptions about what constitutes discourse determine what kind of research questions a researcher can ask within a certain framework, what kind of DA she will use, how she defines the discursive field under study and what sources she uses. Just like in any research design (see Okano, Ch. 3), researchers conducting DA should select the tools that work best with their theory and 2 Symbolic interactionism is a framework, developed by George Herbert Mead that aims to understand the interactions between individuals that create symbolic worlds, and how these worlds in turn shape individual behaviour (West/Turner 2018).

380

Chapter 14 How to understand discourse data, and adapt DA to their own study (Gee 2011, p. ix). While scholars should not be afraid of flexibly adapting theories and methods, it is crucial to understand that depending on the assumptions about science and society (epistemology and ontology), different methods of data collection and data analysis make sense. Most importantly, they should be used in a coherent and transparent way. Despite the differences, performing DA involves some common steps (see Figure 14.2). Figure 14.2: Key steps in discourse analysis

Key steps 1. 2. 3. 4. 5. 6.

Decide on a research topic Explore the field under investigation Develop research question(s) Clarify if DA is appropriate to answer these questions Clarify what type of DA and what discourse theory you want to use Define concepts (discourse, discourse field, actors, audience, etc.) in accordance with this theory 7. Select data sources 8. Find formal (and linguistic) structures 9. Situate statements in their situational and material context 10. Interpret findings

Of course, these steps do not necessarily take place in the chronological order proposed in Figure 2. Researchers might read about discourse theory first and find their research topic later. Sometimes, researchers begin their projects with a different methodology, but realise later in the research process that DA might be a better way to approach their questions. But once a researcher has decided to use DA, it is important to clarify which discourse theory and type of DA she wants to use. Throughout the research process, it is important to check the ‘fit’ between theory and research questions (Keller 2013, p. 69). Since Japan scholars are often more interested in the empirical study of Japan and not primarily in theoretical or philosophical musings about how language in general influences society, power relations or interactions in a society, only certain parts of discourse theory are relevant. Nonetheless, some basic knowledge about underlying philosophical and theoretical considerations is necessary for any analysis. This is mainly because it empowers the scholar to argue and explain the relevance and effects of language and certain practices. In the following parts of this chapter, we will mostly draw on CDA to explain the individual steps of DA in more detail, but we also include examples from the essays and our own work.

381

Andreas Eder-Ramsauer and Cornelia Reiher

Key steps 1–3: Decide on a research topic, explore its context and find a research question First, researchers should decide on a research topic, make themselves familiar with its context and develop research questions. CDA usually focuses on social problems and critically examines politics and society. Accordingly, questions of power imbalances, exclusion from societal participation, sexism or racism are among the major topics in CDA. This is also reflected in CDA on Japan. Tollefson (2014, p. 299), for example, analysed discursive constructions of national duty and the ideology of technoscience to mitigate the risks from the Fukushima nuclear disaster, and to call Japanese national identity into the service of the nuclear industry. Rear and Jones (2013) analysed competing discourses related to education policy and identified a dominant discourse harnessing the individuality of Japanese citizens into a vision of patriotism and national solidarity (Rear/Jones 2013, pp. 375, 377). After deciding on a research topic—a social problem in the case of CDA—researchers should familiarise themselves with the topic, its context and important actors. In the process of studying the context of a social problem, CDA researchers aim to identify a ‘network of practices’, describe key actors, learn who interacts with whom and how they negotiate a certain problem. Rear and Jones (2013), for example, identified the Liberal Democratic Party, the Ministry of Education, Culture, Science, Sports and Technology (MEXT), as well as the Japan Business Federation (Keidanren) as the major policy players in education policy. In Annette SchadSeifert’s analysis on masculinity and herbivore men in Japanese print media, two particular journalists were key players (see this chapter, Ch. 14.1). Daniel White’s study on emotions in the Cool Japan campaign (see this chapter, Ch. 14.2) focused on politicians and television broadcasting. In her study on local identity and rural revitalisation in Arita, a town famous for porcelain production, Cornelia Reiher (2014) identified several relevant actors on different geographical and administrative levels: local government officials and politicians, museum curators, art historians, local potters, civil society organisations, national tourist agencies, prefectural government officials, journalists, department stores in Tokyo and national ministries like MEXT. When analysing discourse, researchers can ask a wide variety of questions (see Figure 14.3). In CDA, the question of dominance—who dominates the meaning-making process—is key in explaining why a problem persists in a society. Therefore, CDA asks about how social life is structured and organised, and how these structures relate to the problems of domination, exploitation and power imbalances (Fairclough 2001, p. 126). Most importantly, CDA (just as any DA) should not be conducted for its own sake, but in order to explain the emergence, continuity or shift of a particular social problem. Japan researchers should always think about how their research contributes to the body of knowledge about Japan or Japan’s society, politics, culture or history. A simple test for researchers is to explain their research in one sentence. If all they have to say is: ‘I am conducting DA on hikikomori’, for example, they need to rethink their research question and explain what they actually want to find out (Turabian 2007, p. 9). For example, a researcher could say: ‘I am conducting DA on hikikomori, in order to

382

Chapter 14 How to understand discourse find out how expert knowledge by medical doctors changes public discourse on hikikomori in Japan.’3 Figure 14.3: Questions in Social Science discourse research

Questions in Social Science discourse research • • • • • • • • •

When and why does a specific discourse appear or disappear again? How, where and with what practices and resources is a discourse (re)produced? What formations of objects, utterance modalities, concepts and strategies does a discourse contain? What are the decisive events in the development of a discourse, and how does it change over time? What actors occupy the positions of speakers using what resources, interests and strategies? Who are the bearers, the addressees and the audience of the discourse? What links does a discourse contain to other discourses? How can a discourse be related to more or less far-reaching temporal-spatial social contexts? What (power) effects result from a discourse, and how do they react to fields of social practice and ‘everyday representations’? (Keller 2013, p. 75)

Key steps 4–6: Clarify if and what type of DA to use and define key concepts The next steps are to think about whether the research question can be answered through DA in a meaningful way and if so, what type of DA is the best choice (steps 4 and 5). In order to make an informed decision, researchers should first learn about DA to evaluate its aim and what skills it requires. Considerations with regard to DA’s fit with a research question could include the following questions: Is your research about the analysis of the interrelations between language and power, as for example in Kamada’s (2010) work on ethnic identities of adolescent girls with mixed ethnicities and their struggle for control over marginalising discourses which disempower them as ‘others’ within Japanese society? Here, discourse analysis is probably helpful, because it can uncover how linguistic representations might reproduce unequal relationships and conditions in society. Or is the aim of the research project to find out how language is strategically used in order to achieve certain goals in a government campaign, for example on radionuclides in food in post-Fukushima Japan (Reiher 2017)? In this particular case, frame analysis was the better methodological choice, because it helped to identify ‘the strategic and deliberative side of language usage’ (Lindekilde 2014, p. 197) by particular actors and, thus, is smaller in scope than DA. Researchers should also think about whether they want to study the content of individual documents or to analyse how their content is interrelated. For the study of individual documents content analysis is a more straightforward method, while DA is better suited to analysing the interrelatedness of texts and society.

3 Please refer to Chapter 2 for more information about research questions.

383

Andreas Eder-Ramsauer and Cornelia Reiher Practical considerations include the question of language proficiency, access to data and time constraints. For DA, language particularly matters. Naturally, when studying Japan, researchers should be aware that they probably have to deal with large amounts of data or texts in Japanese. It is also possible or sometimes even necessary to conduct DA in a language other than Japanese when the discourse, for example, concerns the representation of Japan as a cultural or political actor in international relations in other countries or international organisations. Sometimes, it even takes several languages to analyse a particular discourse, particularly if transnational phenomena are the research topic (see Garon, Ch. 9.2; Shire, Ch. 5.2). Researchers should realistically evaluate whether they are able to read sources in the respective language before starting their analysis. Once a researcher has decided to use DA and selected a specific type of DA, she should make herself familiar with the specific discourse theory the selected type of DA is based on, define concepts like discourse according to this theory and explain what this means for the project at hand. CDA scholars, for instance, take a critical position towards the societal status quo and, as ‘engaged researchers’ (see Slater et al., Ch. 16.2), seek to change social practice and social relationships reproduced in the discourse under study (Keller 2013, p. 25). This is because CDA scholars consider linguistic representations to potentially reproduce unequal relationships and conditions in societies. But even in CDA there exist different approaches (see Figure 14.4). While this chapter introduces Fairclough’s dialectical-relational approach, Schad-Seifert (see this chapter, Ch. 14.1) in her essay on discourse on herbivore men in Japan, provides a step-by-step manual to a discourse analysis based on Siegfried Jäger’s approach to CDA and his concept of interdiscourse. Figure 14.4: An overview of Critical Discourse Analysis

Critical Discourse Analysis Major proponents

Teun van Dijk, Ruth Wodak, Siegfried Jäger, Theo van Leeuwen and Norman Fairclough

Approaches inside CDA

• • • • •

Dialectical-Relational Approach (Fairclough) Socio-Cognitive Approach (van Dijk) Socio-Semiotics and Visual Grammar (van Leeuwen and Kress) Dispositif Analysis or Kritische Diskursanalyse (Jäger) Discourse-Historical Approach (Wodak)

Fairclough’s dialectical-relational approach puts a focus on the interaction of social practices. He argues that social life can be observed as interconnected networks of social practices of diverse sorts (economic, political, cultural and so on). His starting point is the question of who or what stands in the way of resolving a social problem. He also analyses counter-hegemonic discourses that strive to resolve the social problem. Each of the social practices identified as relevant has a semiotic element. Semiotic elements are all elements of communication that convey meaning, either in the form of spoken or written language, or, for example, body language. The researcher’s task is to choose a network of practices that deals with a specific social problem and study the construction of meaning that occurs within this network (Fairclough 2001, p. 122).

384

Chapter 14 How to understand discourse

Key step 7: Select sources Once the CDA researcher is familiar with a specific social problem and relevant institutional structures, important socio-economic factors, actors and opportunities (or the lack thereof) to participate in discourse or status inequalities in wider society, she needs to identify important ‘texts’ to analyse. Thus, getting to know the field and context of one’s research topic and identifying the key actors of a particular discourse is an important step towards finding sources, no matter what type of DA a researcher is conducting. In general, texts are produced within certain institutional and social contexts (Keller 2013, pp. 26–27). Newspaper editorials, for example, are produced inside what constitutes the media; government White Papers originate from practices within government bureaucracies (Fairclough 2001, p. 129). Thus, a wide variety of different sources should be selected by any researcher who carries out DA. These sources should be widely distributed and influential with regard to their range and reception (see this chapter, Heinrich, Ch. 14.3). In a nutshell, DA researchers should select texts that are relevant for answering their research question and analyse them in their particular contexts of production, dissemination and reception. Japan scholars have studied texts produced by a wide variety of actors when conducting DA. In their analysis on education policies, Rear and Jones (2013) used policy reports, policy speeches and advisory reports by the aforementioned actors. Steffen Heinrich (see this chapter, Ch. 14.3) shows how only by using different types of sources for his analysis was he able to show the scope of discourse on the recent work-style reform (hataraki-kata kaikaku) in Japan and its relevance over time and across different realms of society, including: politics, the media or public opinion. Schad-Seifert (see this chapter, Ch. 14.1) analysed mainly media content to study discourse on masculinity in Japan. Both authors began their selection of sources by tracing media buzzwords like herbivore men or hataraki-kata kaikaku. White (see this chapter, Ch. 14.2) analysed the rhetoric on soft power in Japanese media in order to study emotions in Japan’s society. As pointed out above, depending on which discourse theory the DA is based on, sources can include more than written texts. This could be ethnographic material like interviews, field notes with observations on social practices, visual representations or artefacts. DA allows for a wide variety of methods of data collection (and analysis). In her research on local identity and rural revitalisation, Reiher (2014) conducted DA based on interviews with local actors, participant observations, and text and images in local, prefectural and national government policy papers, reports and plans, as well as in media coverage, memoirs, tourist information pamphlets and the advertisements of local businesses.

Key steps 8–9: Find formal (and linguistic) structures and situate statements in their situational and material context After you have identified and collected sources and data, the content of the specific discourse should be scrutinised. For linguistics-focused CDA, not only what is said, but how, is of importance. Fairclough (2001, p. 130) suggests first looking at genres, styles and (sub-) discourses that structure a specific discourse. This is because texts are typically hybrid in terms of genres, discourses and styles: Is the call for more patriotism only communicated in typical bureau-

385

Andreas Eder-Ramsauer and Cornelia Reiher cratic style and language, or do government agencies use pop-cultural motifs and genres to communicate the same message (Rear/Jones 2013)? Fairclough (2001) uses insights from linguistics, more specifically systemic functional linguistics that focuses on grammar and finds specific functions behind the concrete use of grammar (Martin 2016). To give an example, Fairclough (2001, p. 130) writes about the absence of agency in neo-liberal discourse implied through the grammar used in a text—i.e. many passive clauses—and analyses how ‘new competitors’ and ‘new opportunities’ are claimed to exist, without being located within active processes of change. Therefore, they remain abstract and thus, in Fairclough’s analysis, the social agents responsible for political processes remain an abstract ‘we’. In a similar manner, the term globalisation is used in neoliberal discourse in Japan as a force from above with no alternative that forces specific actors like ‘the government’ or ‘the economy’ to cut social services or to outsource manufacturing to Southeast Asia (Yoda 2000). Grammatical features like the passive voice can conceal the actual actors behind certain social problems that are responsible for these processes and, thus, complicate counter-movements due to the lack of a specific actor to whom to address protest (Massey 2005). Another key step in CDA is the analysis of interdiscursivity. This means the recontextualisation of discursive elements in new and different contexts (Fairclough 2001, p. 131). In her research on revitalisation strategies in Arita, Reiher (2009) found that a programme by the Japanese government to promote food education (shokuiku), and particularly the related concept of food culture (shokubunka), was recontextualised in local discourse on Arita’s identity and revitalisation. The Basic Law on Food Education (shokuiku kihonhō), its related policy plans and financial support programmes were intended to promote a healthy diet. Yet, in order to receive subsidies, Arita’s local government integrated food culture into their local revitalisation plan but referred to food culture as ‘serving food on beautiful porcelain’ (made in Arita). In their local plan, images showed junk food, like cup noodles, being served in handcrafted noodle bowls and presented as traditional Japanese food culture (Reiher 2009). In summary, not only are linguistic methods possible tools for the analysis of written texts or visual data, but so are approaches from content analysis, text analysis or frame analysis (see Arrington, Ch. 13).

Key step 10: Interpreting data After identifying the linguistic structures and patterns of grammar, researchers conducting CDA have to interpret these structures to answer their research questions. First, they should reflect on how the empirical results of the linguistic analysis, frame analysis or any other type of textual analysis relate to other dimensions of discourse, i.e. its context or cognition (see above). There are different ways to do this. Fairclough (2001, p. 126), for example, suggests interpreting linguistic structures from the perspective of ideology (see Figure 1). In such an analysis, the researcher would study whether and how a discourse contributes to sustaining specific power relations and the dominance of a certain understanding of a social problem by negating or ignoring alternatives to certain practices or interpretations. In his work about neoliberalism and language, Fairclough (2001) identified claims that markets operate based on ‘their own logic’ and cannot be socially changed as a strategy for disqualifying alternatives,

386

Chapter 14 How to understand discourse thus denying agency to those actors who are trying to challenge neoliberalism. Yoda (2000) has identified similar strategies in neoliberal discourse on globalisation in Japan. Those defending the hegemonic discourse on neoliberalism, present alternative discourses as a misrepresentation of economic matters and by doing so consolidate unequal relations of power by dominating the representation of the problem. This is what Fairclough calls ideology (Fairclough 2001, p. 132). Strategies to discredit alternative discourse have occurred in Japan, for example after the Fukushima nuclear disaster, when mothers concerned about radiation were ridiculed as irrational radiation brain moms (hoshanō mama) incapable of understanding science (Kimura 2016).Thus, a next step in relational-dialectic CDA is to describe the alternatives to the dominant arguments in the analysed texts (counter-hegemonic discourse). Issues that remain completely uncontested in society are rare. Depending on the research design, researchers analyse one or several alternatives brought forth by counter-hegemonic discourses. Alternative discourses can be studied, for example, by inquiring how they relate to the dominant discourse, how the latter responds (if at all) to counter-hegemonic discourse and vice versa. In her study on citizen scientists who monitor radionuclides in food in post-Fukushima Japan, Aya Kimura (2016) shows, for example, how citizen scientists provided alternative knowledge to the Japanese government’s assertion that food was safe. The final stage in CDA includes what some scholars who cling to a positivist idea of ‘objectivity’ might find controversial: to critically reflect on the results of the analysis and to suggest possible solutions to the social problem under study. This includes thinking about how research results can support critical engagement in society and thereby contribute to social emancipation. In addition, CDA also demands that researchers question their own role and the implications of their research for upholding dominant power relations. This might be due, for example, to a researcher’s own position in academic practices and her related networks with the market and the state (Fairclough 2001, p. 127). Through this reflexivity (see Coates, Ch. 3.2; Cook, Ch. 5.1), research results, the researcher, discourse and the social problem being studied are put into perspective and allow for thinking about possible solutions to the problem. Of course, this kind of engaged scholarship exists beyond CDA and also in the study of Japan (see Slater et al., Ch. 16.2). But to be sure, DA does not have to go as far as to suggest solutions to social problems, particularly when DA is carried out by students. Therefore, we would like to stress again that CDA is just one of many possible ways of conducting DA. For many scholars, however, research and social engagement are closely related and CDA is one approach that acknowledges this ideal.

5. Summary Discourse structures how we see, understand and represent the world. Through DA researchers can find out ‘how particular texts either reproduce or challenge established definitions and understandings of social reality’ (Lindekilde 2014, p. 197) by focusing on the interplay between discursive elements and between discourse and society. There are several approaches to DA and various methods of data collection and analysis within DA that a researcher can choose from, i.e. content and frame analysis (see Arrington, Ch. 13). Thus, before 387

Andreas Eder-Ramsauer and Cornelia Reiher beginning DA, researchers should familiarise themselves with the various approaches of DA and the specific rules it follows. Because the various approaches to DA relate to different discourse theories and, thus, assumptions about the relationship between language and the world, researchers should make sure that their research questions, the theory and their chosen approach to DA correspond to each other.

388

14.1 Media buzzwords as a source of discourse analysis: The discourse on Japan’s herbivore men Annette Schad-Seifert

Introduction Discourse analysis is a valuable device that helps us to understand processes of social communication and representation. By discourse I mean all kinds of public communication that influence how people in modern societies think and conceptualise the reality they live in. Discourse is usually produced by social institutions such as the media, academia or politics. For social scientists, analysing the functioning of discourse is essential in order to understand how people’s thoughts, ideas, beliefs, values, identities, interaction with others and behaviour are shaped. There are several useful (sometimes contradictory) instructions available in research literature and on the Internet that provide guidance on technical and analytical steps when conducting a specific research project, especially on the basis of East Asian languages; one example in this respect is Florian Schneider (2019). In the following, I will give an example of qualitative discourse analysis by referring to my own research on gender in Japan, specifically on herbivore men.

Step one: The broader theoretical picture I am very much indebted to Gender Studies as a research approach. Since gender identities have come to be viewed as both socially and culturally constructed phenomena, it is no surprise that the deconstruction of femininity and masculinity forms a crucial part of discourse analysis in Gender Studies. With regard to men and the male gender, Raewyn Connell developed the theory of hegemonic masculinity, against which subordinated or opposing masculinities are positioned (Connell 1995; Connell/Messerschmidt 2005). The concept was swiftly adapted to the Japanese context, and the social existence of the so-called salaryman as the male breadwinner and economic provider for the family was identified as the hegemonic model of Japan’s postwar middle-class society (Dasgupta 2013; Roberson/Suzuki 2003; SchadSeifert 2007).

Step two: The socio-economic context Taking the salaryman into the perspective of Gender Studies means framing the study of men within a larger socio-economic context. The economic crisis after the burst of the financial bubble in the 1990s had detrimental consequences for Japan’s regular labour market. The main pillars of the company welfare system, such as lifelong employment and in-house recruiting, started to erode. Under the new legislation of neoliberal deregulation, companies tended to avoid regular employment of new recruits, relying instead on atypical forms of labour such as part-time and casual work for men. The fact that men increasingly lost their chances to find

389

Annette Schad-Seifert regular jobs was perceived as generating socially harmful effects, and stirred a debate about the hegemonic salaryman.

Step three: Material for analysis The generation of university graduates entering the labour market between 1993 and 2005 was particularly hit by the restructuring measures and they found themselves trapped in a socalled employment ice age. The discourse around this generation, called rosujene in Japanese— a Japanese–English abbreviation of lost generation—also created terminologies for explaining modifications in Japanese masculinities. One of them is the term herbivore man (sōshoku (kei) danshi)—referring to young men who have lost interest in sex, work and consumption. I chose the discourse on Japan’s herbivore men as part of my research on men’s studies (Schad-Seifert 2016) and will show in the following how I investigated it through qualitative discourse analysis. Other examples of discourse analysis on the topic are Steven Chen’s (2012) study on herbivore men and consumption, and Constanze Noack’s (2015) work on knowledge and masculinity. After I had screened online and print sources with regard to the above-mentioned context, one of my noticeable findings was that buzzwords and neologisms such as herbivore man have drawn public attention extensively. In order to analyse the character of this attention, I applied Siegfried Jäger’s term interdiscourse, which has to be differentiated from the specialised academic discourse (Jäger 2004, p. 159). Interdiscourse refers to all kinds of statements that have not been proven empirically, but into which elements of academic discourse are interwoven. While specialised academic discourse aims to accumulate ‘objective’ and ‘true’ knowledge, interdiscourse intertwines this ‘objective’ knowledge with other, non-academic discourse. In so doing, it aims at making information more relevant and relatable for individuals and their everyday lives by providing concrete images, identity constructions and possible modes of behaviour (Link 2003, p. 23). Typically, journalists are key in producing interdiscourse. This is why I collected my main material from Japan-related media sources with a high circulation in both the Japanese and English languages. In step five, I am going to provide examples of different types of enunciations on herbivore men as threads of discourse (i.e. sequences of discourse fragments with a common theme) that generate images and models in order to explain so-called odd or non-normative behaviour among men.

Step four: Identifying the actors of discourse I found out that the Japanese term sōshoku (kei) danshi was created and has been circulated by two marketing journalists in particular, namely Fukasawa Maki and Ushikubo Megumi. With their publications, the term gained such an amount of attention that it was awarded with the Grand Prize of Buzzwords and Neologisms in Japan in 2009. As a result, it was much received and commented on by Japanese and foreign reporters. I argue, based on Michel Foucault’s ideas (1970), that during this process of production, dissemination and reification of discourse (discursive reproduction), and as an effect of repeated modes of enunciation, herbivore men became a social fact. At the same time, public statements confirmed, complemented

390

Chapter 14 How to understand discourse or rebutted the typologies of the social phenomenon. In my analysis, I tried to capture exactly this process of reproduction and interaction in discourse.

Step five: Content analysis From 2006 until 2008, as a columnist for the online magazine Nikkei bijinesu, one of Japan’s largest business newspapers, Fukasawa (2006) published a 32-part column entitled Marketing lexicon of U35 men (U35 danshi māketingu zukan). In this column, Fukasawa pictures various lifestyles and forms of consumption behaviour of young men between the ages of 18 to 34 years. Among a wide range of types, the herbivore is described as a ‘new type of person’ (atarashii jinshu) that has to be differentiated from traditional male categories. Fukasawa defines two types of traditional male characters: one is popular among women and actively seeks sexual relationships with them, while the other is unpopular and has no chance of finding a mate. According to her, recently a new type has emerged: men who are attractive and popular, but not interested in sex and love, and are therefore called ‘herbivores’ (Fukasawa 2006, p. 1). What became influential for the discourse is that although Fukusawa identified a variety of new male types, it was only the herbivore that became so popular afterwards. She quickly responded to the hype by publishing a book on the The era of the herbivore man (Sōshoku danshi sedai. Heisei danshi zukan; Fukusawa 2008). When sōshoku (kei) danshi reached the top of the list of the most important Japanese buzzwords in 2009, it seemed undeniable that a whole generation of young men in Japan had turned into herbivores. Ushikubo fuelled this impression by maintaining that these guys ‘will definitely be in your surroundings’ (2008, p. 4). She calls them ‘feminine men’ (ojōman) of the herbivore type, because of their distinct preferences for sweets, fashion and cosmetics, and their strong dislike of products usually consumed by men, such as cars, alcoholic beverages and real estate (Ushikubo 2008, p. 48). The reader is provided with an explanation of why the new generation has lost its appetite for high-priced products that were considered indispensable in demonstrating status in the bubble era. In sum, both Fukasawa and Ushikubo give an account of how the long recession must have had an impact on the tastes of the young, therefore, making their restraint comprehensible. This marketing-oriented discourse is clearly written in favour of such feminine men and as a lesson to those markets and producers that have yet to come to terms with the new tastes and trends in an era of economic decline. After analysing these two authors’ works as the first thread of discourse, I tried to come to terms with the spillover of the discourse into Japanese newspaper articles (here Asahi Shimbun published in 2009 and 2010). My reason for choosing Asahi Shimbun was that it is one of the largest newspapers in Japan. In addition, I had a pragmatic reason: I was able to access the newspaper archive from the University and State Library Düsseldorf via CrossAsia (see Peucker/Schmidtpott/Wagner, Ch. 9). I decided the time span for my investigation would be the years when the discourse started to manifest itself as discursive knowledge in Japanese and international media sources. Most of the articles give an account of how and why the phenomenon should be seen as a serious social issue. Concern is expressed about the fact that a whole generation of young men is becoming feminised and ignorant towards love and sex. Their passive character in particular is interpreted as a weakness. Alongside their poor physical strength, all kinds of mental shortcomings are attributed to herbivore men, such as fear of

391

Annette Schad-Seifert failure or lack of ambition. As a third thread of discourse, I took the international media coverage, such as that by CNN, Japan Times Online and other English-language channels, in the respective time span. One example here is a video report by CNN’s journalist Morgan Neill: It covers an original video recording that shows Fukasawa—as an expert—who spots herbivoretype men on the streets of Tokyo, or interviews senior Japanese salarymen, who express their negative views about the young generation of herbivores (Neill 2009).

Step six: Conclusive evaluation of discourse Qualitative discourse analysis helped me to show that statements about herbivore men have to be recognised as a social practice of self-assurance and as knowledge construction of social reality. The traditional salaryman, in other words Japan’s hegemonic masculinity, is defined in contrast to the men of the lost generation. It is claimed with regard to gender identity that normal men have a natural sexual need as a carnivore, while herbivore men have lost this interest in erotic intimacy. This attitude is seen as an expression of a deeper emotional state, which manifests itself as withdrawal from society and a general passivity. The question of whether this behaviour is a consequence of the long-lasting stagnation of the Japanese economy can indeed be found in the first thread of marketing analysis, but does not determine the discourse in the articles of the Asahi Shimbun. Rather, the assertion that herbivore men caused the economic recession with their consumer reluctance tends to use them as a scapegoat. At the same time, the discourse has produced empirical surveys, according to which society and people in their everyday lives have started recognising men as herbivores. Moreover, male individuals are presented as identifying themselves with this character and as calling themselves by this name. This self-identification is again empirically proven in surveys and interviews conducted by Japanese newspapers, which—as an effect—generates a high degree of authenticity. Herbivore men are actually starting to exist and to speak. It can be concluded that discourse analysis helps us to understand how the knowledge production of discourse does not only influence but also actually constructs gender images in social reality.

392

14.2 Analysing affect, emotion and feelings in fieldwork on Japan Daniel White

It is not likely coincidental that one of the most iconic early studies of Japanese culture conducted from outside Japan focused so closely on feelings. Ruth Benedict's The Chrysanthemum and the sword (2005) introduced ‘the Japanese’ to both scholars and English-reading publics through a framework of sentimental polarities. From her culturally relativist perspective, understanding who the Japanese were required seeing them through contrasting patterns of discipline and sensitivity. Today, wary of the essentialising effects of her schema, only few anthropologists of Japan would adopt Benedict’s approach to culture. But it is interesting to note that in the wake of the decline of the psychological interpretations of culture that she inspired (later called the Culture and Personality School), a renewed anthropological interest in the emotions and, more recently, in affect has emerged. This departure from overarching models for analysing emotion combined with a recognition of how an increasingly global Japan is transforming lives on intimate and emotional levels leaves many students and scholars unsure of how to proceed. The following chapter offers a general framework for attending to and analysing data focused on emotion and affect, or what we might more broadly call feelings. By drawing on recent theorisations of emotion and affect in the Humanities and Social Sciences, I argue that any successful social account of feelings in Japan needs to pay attention to at least three components: the discursive, the sensorial and their mutual interaction within material cultural environments.

Discourse and emotion Despite the variety of methodological approaches to analysing emotion from across the Humanities and Social Sciences, a feature they share is an attention to how feelings are expressed discursively. A discursive approach analyses how feelings are organised through the representational practices and codes that assign meanings and values to their expression in processes of social interaction—including those of both consensus and contest. While there are a variety of terms under which social theorists have traditionally grouped this aspect of feeling (e.g. sentiment, effervescence, emotion, affect, feeling, sensation), a growing consensus has emerged around the use of the term ‘emotion’ to refer to how experiences of feelings embed themselves, circulate and acquire affective intensity in part through mediums of representation. Representational modes of feeling operate primarily through sign systems and processes of signification. The way different representational systems order and ascribe value to certain categories of feeling determines how emotion comes to matter in social life. The importance of cultural variability to this process becomes clear when considering how emic terms like meiwaku (nuisance), omoiyari (consideration) or omotenashi (hospitality) guide conduct and interaction in

393

Daniel White Japanese public culture. These key words in Japan’s ‘emotional lexicon’ (Frevert at al. 2014) not only communicate a sense of how one is expected to express feelings in specific contexts in everyday life in Japan; they also become values by which national identity is defined, managed and leveraged towards political expedience in response to perceptions of globally structured pressures and threats. For example, consider the importance of the term ‘soft power’, which has captured the attention of politicians and bureaucrats since the early 2000s. The term describes how a nationstate cultivates prestige through attraction to its culture, values and policies rather than through traditional hard power sources such as a strong economy and the military (Nye 1990). Although originally applied to counter critiques of America’s waning power on the global stage in the wake of the Cold War, the term was taken up by Japan’s politicians, who used it as a discursive tool with which to recover geopolitical influence in the cultural field where the nation had lost it in the economic one in the two decades following the collapse of its asset bubble in 1991. Analysing the rhetoric surrounding soft power in Japan, one can trace how a hope in the potential for Japan’s popular culture industries to rescue the nation from decline was both rooted in and generated by discourses that began focusing less on Japan’s traditional arts and more on those of pop culture and anything marketable as ‘Cool Japan’. One particularly illustrative example comes in a list of programme themes for the national public broadcaster’s TV series, Cool Japan (NHK 2010): Stationery, shopping, winter, examinations, childbirth, childrearing, memorial services, Japanese men, Japanese women, mothers, fathers, anniversary parties, sweets, discipline, hot pots, sightseeing, toys, health, luck, rain/the rainy season (tsuyu), privacy, the Japanese language, and Japanese companies (parts 1 and 2)…. By way of a nation branding strategy that sought to label as many mundane aspects of Japanese life as ‘cool’, content and policymakers both signalled and produced a discursive shift that increasingly tied national identity to images of a cool, popular and alternative Japan that was attractive to foreign consumers. Whether in registers of national politics or of everyday life, cataloguing discursive shifts in the lexicon, rhetorical strategies, symbolic content and other representational strategies for ordering and reordering value constitutes a fundamental starting point for analysing how emotion operates through signifying practices.

The sensorial and affect Recognising the power of discourse’s impact on emotion is what inspired much of the recent anthropological and sociological work on feelings that emerged as an alternative to national character studies; however, facing the limits of discourse is what inspired a subsequent turn to the body, sensation and affect. Defined in contrast to emotion, in which feelings are rooted in narrative, symbols and other conscious and forms of representation, affect refers to nonconscious intensities and capacities of bodies that do not take shape in the usual signs of discourse (Massumi 2002). Indications of affect can be found in the seemingly spontaneous emergence of new political movements, an unexpected meme or slogan gone viral, collective waves of anxiety or anomie vaguely sensed but not explicitly articulated, or in the unexpected eruption of anger, animosity and violence. The task of the fieldworker interested in both emotion and

394

Chapter 14 How to understand discourse affect is to account for what is seemingly unexpected or unexplainable through discourse alone. Consider again the popularisation of soft power in Japan. What is curious about the excitement over soft power among bureaucrats in the 2000s is how it grew despite a number of complicating factors: criticism that it was ambiguously defined, a lack of metrics for tracking growth, stagnation in the culture markets cited as primary soft power resources, and a broad consensus among nation branding experts that government investment in popular culture often backfires (White 2011, p. 8–10). In short, despite a variety of criticisms over the potential for a pop culture-driven form of soft power to revitalise Japan, interest and investment in soft power grew nevertheless. Contradictions, breakdowns and even profusions of discourse can often indicate where affect is bubbling underneath. Getting in touch with this requires not only a discursive mapping of as much of the social, political and other environmental ‘arrangements’ (Slaby et al. 2017) that the researcher can identify as fuelling affect. More importantly, it also involves paying attention to the body—of both the researcher and interlocutors. In short, it requires not only an analysis but also ‘practices of feeling with the world’ (De Antoni/ Dumouchel 2017). Indeed, it was only after substantial time doing the same things that policymakers do—attending meetings, editing documents, evaluating programmes—that I began to sense the urgency of my interlocutors. Tuning in to this urgency helped explain the contradictions of logic, tensions in speech, and a variety of creative and experimental policies (White 2015) which were previously unimaginable outside the recent discursive rearrangements of soft power. Researching affect requires sensing how bodies tense or relax according to certain situations, noting how environments and events resonate or grate, tracing how moments of frustration transform into action, and simply but somatically doing what one’s interlocutors do.

Interactive approaches Although I have divided the emotional from the affective dimensions of feelings for the sake of illustration, a comprehensive approach would integrate discursive analyses with sense experiences. Margaret Wetherell (2012, pp. 7, 46–48) refers to these as the ‘looping’ effects of emotion and affect. In my own work (White 2011; 2017), I have used the phrase the ‘affect-emotion gap’ in order to draw attention to how the epistemological fissure between what we feel affectively and what we know of what we feel emotionally becomes a productive site for friction generating social and material reproduction, cultural transformation and political contest. Combining a discursive analysis of emotion with a sensorial analysis of affect allows one to analyse and ultimately argue for how they are held together in dynamic relation. For example, making sense of the affective and emotional dimensions of soft power’s rise in Japan required an understanding of how they are mutually constructed. Soft power cannot be described as only an affective anxiety over Japan’s geopolitical decline or an emotional hope for Japan’s resurgence. Rather, its discursive and affective dimensions are held together in a productive tension. Put summarily, soft power rhetoric functioned as a means to transcribe an ambiguously described but poignantly felt anxiety over Japan’s present into a communicable and manageable hope for its future.

395

Daniel White

Conclusion Although no single strategy of analysing feelings can be applied uniformly irrespective of one’s particular research aims, a comprehensive and historically sensitive approach would combine a semiotic attention to discursive ruptures in emotional discourses with a ‘somatic mode of attention’ (Csordas 1993) to affective sensations, tensions, gestures, expressions and atmospheres. Such integrative work can at times seem vague and unwieldy, but it is hardly unempirical. That the process of not only semiotically reading but also somatically sensing feelings at first involves an abstract relation to affect by no means renders affect unverifiable, as anyone who has been acknowledged in Japan for ‘reading the air’ (kūki o yomu), or ‘getting it wrong’ (kūki o yomenai), knows well enough. Even more importantly, that this process of verification is inescapably a subjective one does not then require repressing or removing one’s subjective feelings from the ethnographic encounter, as if such a thing were even possible; instead, it requires an even more reflexive accounting of them (see Coates, Ch. 3.2; Cook, Ch. 5.1; Spoden, Ch. 12.2).

396

14.3 From buzzwords to discourse to Japanese politics Steffen Heinrich

In spring 2016, the Japanese government under Prime Minister Shinzō Abe announced that the fight against excessive working hours and wage inequality was now a major policy objective for his government. In autumn of the same year, a government-appointed commission presented sharply worded guidelines for future reforms that promised, among other things, to ‘eradicate’ all unjustified differences in the working conditions of so-called regular and non-regular workers. Many observers at the time wondered whether this so-called ‘work-style reform’ (hataraki-kata kaikaku) programme meant that the male-dominated and business-friendly governing Liberal Democratic Party (Jiyūminshutō, LDP) was suddenly, and rather unexpectedly, responding to the concerns of societal groups it had never sought to appeal to previously: women and non-regular workers. In the following essay, I will show that discourse analysis can illuminate the motives behind such unexpected policy decisions and help to enhance one’s understanding of dynamic shifts in Japanese politics more generally.

Political discourse in Japan and the discourse on the work-style reform Until fairly recently, public political discourse in Japan was usually described as restrained and as offering little insight into actual policymaking processes. In particular, research on the role of the bureaucracy and the LDP’s Policy Affairs Council suggested that crucial parts of policy formulation were happening outside the public’s view (Inoguchi/Jain 2011). This perception draws mostly from three features of Japan’s postwar politics. First, under the single non-transferable votes (SNTV) electoral system for the Lower House of the Diet, politicians of the LDP, the dominant party since its formation in 1955, have had relatively few incentives to mobilise voters through public speech and detailed policy statements. Instead, they sought to foster close ties with specific organisations who would act on their behalf. Second, for politicians their ability to secure support from bureaucrats was long deemed more important for electoral success than their policy positions. Third, political reporting in Japan was often criticised as being rather close to the official line of the government, thus providing an incomplete glimpse on crucial discourses and controversies. This has prompted many scholars to conclude that ‘the discrepancy between façade and substance […] is more pronounced in Japan and may be the single most important feature to be considered by any analyst of politics’ (Feldman 2004, p. 8). However, though this distinction remains relevant, the following sections will demonstrate, analysing public political discourse in Japan is far from pointless. In fact, such analyses may never have been as insightful as today, because the link between public discourse and policy decisions has become stronger since the 1980s. Since then a major reform of the electoral system in 1994 and administrative reforms have not only strengthened the role of elected politi-

397

Steffen Heinrich cians vis-à-vis bureaucracy but also the value of public speech and communication in voter mobilisation. Takashi Inoguchi and Punendra Jain (2011) have described this as a shift from a ‘karaoke democracy’, where politicians merely represent what bureaucrats have conceived, to a more populist and engaging ‘kabuki democracy’, where politicians mobilise support also by promoting popular policies. Moreover, not all scholars agree that the representation of public discourse in the Japanese mass media is overly narrow. Tasaki Taniguchi (2018, p. 123), for example, even argues that editorials as well as the readership of newspapers vary so much with regard to their ideological orientations that they offer a range of opinions unseen in many other comparable democracies. Hence, there are not only good reasons to consider discourse analysis when studying Japanese politics but there are also several readily available sources on which such endeavours can draw. To illustrate how they can be put into practice, in the following sections, I will focus on a specific discourse that is connected to the buzzword ‘work-style reform’ (hataraki-kata kaikaku). Such buzzwords can be, under certain conditions, can be signposts of potentially significant policy changes. The term ‘work-style reform’ has been in use for some time, but since 2016 it has come to represent a range of issues, including the problem of overly long working hours and other unfair and overburdening work practices in Japan. The buzzword ‘work-style reform’ has even been used frequently outside the political realm. It has been referenced in talk shows and comedy shows on TV and even in advertisements for products ranging from beverages to a new office building in Tokyo’s Toyosu district. The vast popularity of the term is not sufficient to asses whether it indeed marks a fundamental policy change or whether this discourse is politically relevant. To answer such questions, it is necessary to first establish how politically salient a buzzword is.

Salience Political salience—that is, whether a buzzword resonates with the public and remains in use for some time—can be assessed by using sources such as parliamentary debates, government documents, manifestos of political parties, opinion polls, political newspaper reporting and other media. Most of these can be accessed online and analysed with relatively moderate investments in time. Arguably the quickest and simplest initial option for getting a sense of salience is to look at how interest in a buzzword has fluctuated in online search queries. Google trends, a commercial but free tool offered by the Google empire, allows you to survey the number of web searches for a specific term over time and provides a rough indication of the temporal changes in interest. Moreover, it indicates, again in a rather rough form, to what extent a buzzword has attracted interest online over a longer period of time and lists search queries and keywords that have been used in context. This will work, however, only for queries that have been fairly widely used as otherwise Google Trends may report that not enough data is available. In the case of the Japanese term hataraki-kata kaikaku, the number of queries increased sharply in late 2016, a few months after the announcement of a government action plan with the same name, and peaked in early 2019. Among the top five related topics listed are ‘legislative bill’ (hōan) and ‘working time’ (rōdō jikan). Hence, we can assume that the buzzword ‘work-style reform’ has attracted attention at least in part because it is related to specific legis-

398

Chapter 14 How to understand discourse lation that deals with working hours. To substantiate this further, I then used the online database of a newspaper and focused on those periods when the number of queries surged. The big five of Japan’s newspapers, Nihon Keizai Shimbun (Nikkei), Asahi Shimbun, Yomiuri Shimbun, Mainichi Shimbun and Sankei Shimbun all maintain extensive electronic databanks of their publications, in which terms and combinations of expressions over several decades can be searched for. Using Asahi Shimbun’s databank, I found that the announcement to establish a commission tasked to work out concrete measures for reforming work-styles in early 2016 was the reason for the first hike in interest. The release of a government action plan in March 2017 marks a second peak. This plan entails proposals for legislative changes, many of which were enacted in April 2019, when search queries on ‘work-style reform’ on Google peaked overall. However, this does not suffice to explain the popularity of the term hataraki-kata kaikaku beyond policy debates. To shed light on this aspect, it can be worthwhile to draw on polls and surveys. The national public opinion poll on the livelihoods of the people (Kokumin seikatsu ni kansuru seron chōsa), for example, has been conducted annually by the Cabinet Office (CAO) since the 1950s (CAO n.d.) and is easily accessible online—albeit only in aggregated form. One question that is particularly useful for assessing salience and that has been consistently asked from the beginning surveys the issues citizens want the government to address with the greatest urgency. Here, I found that the topic closest to the term hataraki-kata kaikaku, namely work and employment, peaked during the global financial crisis of 2008 and 2009. Although mentions of these terms later declined, it still ranks among the biggest concerns for citizens in Japan. From this it can be concluded that there had been no exogeneous factor or event that had prompted the government to address work-style reform in 2016, almost a decade after the crisis. Instead, the objectives of the reform are, if at all, salient because they connect to pressing and long-standing concerns of citizens.

Content and context But what explains the Japanese government’s sudden interest in reforming work-styles? To understand its motives, I first looked at the aforementioned action plan—drafted by an advisory committee headed by the Prime Minister—and found four conspicuous statements and expressions: The document claims that 1. the reform aims at changing Japanese work culture, which often entails excessive working hours for many employees. It uses 2. the term ‘work-styles’, which describes, in a rather neutral fashion, the variety of employment forms and connected working conditions instead of the more common distinction between non-regular (hiseiki koyō) and regular employment (seiki koyō), which has a somewhat more negative connotation. Although many specific problems of each employment form are acknowledged, the document does not depict regular employment as a desirable norm, nor does it portray non-regular employment per se as precarious. Instead, the main objective 3. is to ‘eradicate’ unjustified differences in the working conditions of different worker groups. To achieve this, 4. the plan is to be implemented swiftly and in broad societal consensus. To get a sense of whether the action programme represents a departure from previously held positions and provides evidence of an emerging consensus, it can be insightful to examine parliamentary debates or party manifestos. The former can be accessed and searched for online

399

Steffen Heinrich with keywords via the National Diet Library’s website (see Peucker/Schmidtpott/Wagner, Ch. 9) and the latter, for example, through the Manifesto Project (Volkens et al. 2019). In the case of work-style reform, I analysed the party manifestos of the two largest parties in Japan at the time (Heinrich 2017), the LDP and the Democratic Party of Japan (Minshutō, DPJ), for the lower house election in 2014, the last election before the work-style reform was announced. In their manifestos, both parties devote a similar amount of text to the issue of work and both acknowledge that there are discrepancies between the working conditions of regular and nonregular workers. However, they differ noticeably in their visions for reform. While the LDP’s manifesto sees higher mobility between different types of jobs as the most important objective, the DPJ calls for stricter legislation to end practices of unfair treatment and long working hours. Seen against this background, it appears that with the work-style reform the LDP has shifted noticeable towards the positions of the DPJ. In this sense, the buzzword ‘work-style reform’ could indeed signify a shift in Japanese labour politics. However, newspaper reporting on the actual policy process reveals some important caveats. For example, Japan’s largest trade union federation Nihon Rōdōkumiai Sōren Gōkai (Rengō), which had been invited to participate in the drafting of working time policy after 2016 and whose consent had been deemed essential by the cabinet, eventually withdrew its support for the proposed rules on capping excessive working hours. It gave in to members’ protests, who had claimed that the reform was in effect legitimising rather than limiting excessive overtime. A similar ambiguity characterises another centrepiece of the work-style reform, the new legislation on improving standards of equal treatment between regular and non-regular workers. Here, newspaper reporting indicates that the government’s primary goal was to prompt employers into raising wages in order to support its plan for economic growth known as ‘Abenomics’. Whether the reform would achieve its intended goal of moderating inequality was only of secondary importance. This is also mirrored in how newspapers discussed the reform proposal. While the Asahi Shimbun looked at it mostly from a social justice point of view, Nikkei and Yomiuri Shimbun focused more on its possible implications for economic policy and growth (this draws on unpublished work by Tamara Fuchs and the author). These differences hint at a strong strategic motivation for reform, which reflects less an unexpected belief shift on part of the LDP but instead the strategic opportunity to connect one short-term policy objective, wage growth, with issues that are known to resonate with many voters. Put differently, the new labour policy consensus appears to hold only as far as problem perception or description (diagnostic framing) are concerned, but to a much lesser degree regarding the substance of policy because the commitment of the government to realizing the objectives of the reforms often seems to be shallow. Although this somewhat disqualifies the assessment of the work-style reform as a major change, it is nonetheless a significant finding as it shows that concerns regarding excessive and discriminatory work practices are not only salient but acknowledged by parties and politicians across the political spectrum. This plausibly explains why the work-style reform has resonated so strongly even beyond questions of policy. Examining the content and context of the buzzword ‘work-style reform’ can therefore indeed help us to understand ‘the dynamics of change (and continuity)’ (Schmidt 2011, p. 107) in contemporary politics and policy choices.

400

Chapter 14 How to understand discourse

Concluding remarks In this essay, I showed that buzzwords can be a fruitful starting point for an investigation that aims to shed light on Japanese politics. A note of caution is due, however. Since they are highly vulnerable towards strategic issue framing, buzzwords are not per se reliable indicators of transformative change. Only when it can be established that buzzwords are linked to a salient discourse, can they offer the opportunity to uncover aspects of current political dynamics that might otherwise remain indiscernible. Discourse analysis can be the method of choice for studying highly relevant research questions, such as why the Fukushima nuclear crisis led to a major policy shift in Germany but not in Japan (Rinscheid et al. 2019) or cases of electoral success and failure of Japanese prime ministers (Sauzier-Uchida 2014). Yet tools and strategies connected to discourse analysis can also be used to enhance one’s understanding of Japanese politics or to generate meaningful research questions. Fortunately, nearly all of the sources of data and tools needed are readily available to scholars regardless of whether they are based in or outside Japan.

401

Further reading Akerstrom Andersen, Niels (2003): Discursive analytical strategies: Understanding Foucault, Koselleck, Laclau, Luhmann. Bristol: The Policy Press. Angermuller, Johannes/Maingueneau, Dominique/Wodak, Ruth (eds.) (2014): The discourse studies reader: Main currents in theory and analysis. Amsterdam: John Benjamins Publishing. Gee, James P. (2011): How to do discourse analysis: A toolkit. London: Routledge. Gee, James P./Handford, Michael (eds.) (2014): The Routledge handbook of discourse analysis. London: Routledge. Keller, Reiner (2013): Doing discourse research: An introduction for social scientists. Los Angeles, CA: Sage. Schiffrin, Deborah/Tannen, Deborah/Hamilton, Heidi E. (eds.) (2001): The handbook of discourse analysis. Malden, MA: Blackwell.

References Akerstrom Andersen, Niels (2003): Discursive analytical strategies: Understanding Foucault, Koselleck, Laclau, Luhmann. Bristol: Policy Press. Althusser, Louis (1971): Lenin and Philosophy and other essays. New York, NY: Monthly Review Press. Angermuller, Johannes/Maingueneau, Dominique/Wodak, Ruth (eds.) (2014): The Discourse Studies reader: Main currents in theory and analysis. Amsterdam: John Benjamins Publishing. Befu, Harumi (2001): Hegemony of homogeneity. Melbourne: Trans Pacific Press. Benedict, Ruth (2005): The chrysanthemum and the sword: Patterns of Japanese culture. Boston: Houghton Mifflin. Blackburn, Simon (2008): Oxford dictionary of philosophy. Oxford: Oxford University Press. CAO (Cabinet Office) (n.d.). Kokumin seikatsu ni kansuru seron chōsa. https://survey.gov-online.go.jp/ind ex-ko.html, [Accessed 05 March 2020]. Chen, Steven (2012): The rise of the 草食系男子 (soushokukei danshi): Masculinity and consumption in contemporary Japan. In: Otnes, Cele/Zayer, Linda (eds.): Gender, culture, and consumer behavior. New York, NY: Routledge, pp. 283–308. Connell, Raewyn (1995): Masculinities. Berkeley, CA: University of California Press. Connell, Raewyn/Messerschmidt, James W. (2005): Hegemonic masculinity: Rethinking the concept. In: Gender and Society 19, No. 6, pp. 829–859. Csordas, Thomas J. (1993): Somatic modes of attention. In: Cultural Anthropology 8, No. 2, pp. 135– 156. Dasgupta, Romit (2013): Re-reading the salaryman in Japan: Crafting masculinities. London: Routledge. De Antoni, Andrea/Dumouchel, Paul (2017): The practices of feeling with the world: Towards an anthropology of affect, the senses and materiality—Introduction. In: Japanese Review of Cultural Anthropology 18, No. 1, pp. 91–98. Donoghue, Matthew (2018): Beyond hegemony: Elaborating on the use of Gramscian concepts in Critical Discourse Analysis for Political Studies. In: Political Studies 66, pp. 392–408. Fairclough, Norman (1992): Discourse and social change. Cambridge: Polity Press. Fairclough, Norman (2001): Critical Discourse Analysis as a method in social scientific research. In: Wodak, Ruth/Meyer, Michael (eds.): Methods of Critical Discourse Analysis. London: Sage, pp. 121– 138. Fairclough, Norman (2003): Analysing discourse: Textual analysis for social research. London: Routledge. Feldman, Ofer (2004): Talking politics in Japan today. Brighton: Sussex Academic Press. Foucault, Michel (1970): The order of things: An Archaeology of the Human Sciences. New York, NY: Vintage Books. Frevert, Ute/Bailey, Christian/Eitler, Pascal/Gammerl, Benno/Hitzer, Bettina/Pernau, Margrit/Scheer, Monique/Schmidt, Anne/Verheyen, Nina (2014): Emotional lexicons: Continuity and change in the vocabulary of feeling 1700–2000. Oxford: Oxford University Press. Fukasawa, Maki (2006): Dai 5kai sōshoku danshi: U35 danshi māketingu zukan. In: Nikkei Bijinesu, 5 October. http://business.nikkeibp.co.jp/article/skillup/20061005/111136/, [Accessed 23 June 2019]. Fukasawa, Maki (2008): Sōshoku danshi sedai: Heisei danshi zukan. Tōkyō: Kōbunsha. Gee, James P. (2011): How to do discourse analysis: A toolkit. London: Routledge.

402

Chapter 14 How to understand discourse Gee, James P./Handford, Michael (eds.) (2014): The Routledge handbook of discourse analysis. London: Routledge. Glynos, Jason/Howarth, David (2007): Logics of critical explanation in social and political theory. London: Routledge. Gramsci, Antonio (1971): Selections from the prison notebooks. New York, NY: International Publishers. Harvey, Lee (2012): Constructivism. In: Social Research Glossary, Quality Research International. http://w ww.qualityresearchinternational.com/socialresearch/constructivism.htm, [Accessed 13 May 2020]. Heinrich, Steffen (2017): Does employment dualisation lead to political polarisation? Assessing the impact of labour market inequalities on political discourse in Japan. In: Chiavacci, David/Hommerich, Carola (eds.): Social inequality in post-growth Japan: Transformation during economic and demographic stagnation. Abingdon: Routledge, pp. 73–87. Inoguchi, Takashi/Jain, Punendra (2011): Introduction: From karaoke to kabuki democracy: Japanese politics today. In: Inoguchi, Takashi/Jain, Punendra (eds.): Japanese politics today: From karaoke to kabuki democracy. London: Palgrave Macmillan, pp. 1–9. Jäger, Siegfried (2004): Kritische Diskursanalyse. Eine Einführung. Münster: Unrast. Kamada, Laurel (2009): Mixed-ethnic girls and boys as similarly powerless and powerful: Embodiment of attractiveness and grotesqueness. In: Discourse Studies 11, No. 3, pp. 329–352. Kamada, Laurel (2010): Hybrid identities and adolescent girls: Being ‘half’ in Japan. In: Multilingual Matters 11, No. 3, pp. 329–352. Keller, Reiner (2013): Doing discourse research: An introduction for social scientists. Los Angeles, CA: Sage Publications. Keller, Reiner/Hornidge, Anna-Katharina/Schünemann, Wolf J. (eds.) (2018): The Sociology of Knowledge approach to discourse. London: Routledge. Kimura, Aya (2016): Radiation brain moms and citizen scientists: The gender politics of food contamination after Fukushima. Durham, NC: Duke University Press. Laclau, Ernesto/Mouffe, Chantal (2006): Hegemonie und radikale Demokratie: Zur Dekonstruktion des Marxismus. Wien: Passagen. Lindekilde, Lasse (2014): Discourse and frame analysis: In-depth analysis of qualitative data in social movement research. In: Della Porta, Donatella/Malthaner, Stefan (eds.): Methodological practices in social movement research. Oxford: Oxford University Press, pp. 195–227. Link, Jürgen (2003): Kulturwissenschaft, Interdiskurs, Kulturrevolution. In: KultuRRevolution. Zeitschrift für angewandte Diskurstheorie 45/46, pp. 10–23. Mannheim, Karl (2015): Ideologie und Utopie. Frankfurt a. M.: Klostermann. Martin, James R. (2016): Meaning matters: A short history of systemic functional linguistics. In: Word 62, pp. 35–58. Massey, Doreen (2005): For space. London: Sage. Massumi, Brian (2002): Parables for the virtual: Movement, affect, sensation. Durham, NC: Duke University Press. Neill, Morgan (2009): Japan’s ‘herbivore men’: Less interested in sex, money. https://edition.cnn.com/2009 /WORLD/asiapcf/06/05/japan.herbivore.men/index.html, [Accessed 26 June 2014]. NHK (2010): Cool Japan [2010 Broadcasts]. https://www4.nhk.or.jp/cooljapan/ [Accessed 27 February 2020]. Noack, Constanze (2015): Knowledge and masculinity construction of the Japanese sōshoku danshi (herbivore boy) through discourse in international media coverage. In: French Journal of Media Research 3. http://frenchjournalformediaresearch.com/ lodel-1.0/main/index.php?id=453 [Accessed 26 June 2020]. Nye, Joseph S. (1990): Bound to lead: The changing nature of American power. New York, NY: Basic Books. Panizza, Francisco/Miorelli, Romina (2013): Taking discourse seriously: Discursive institutionalism and post-structuralist discourse theory. In: Political Studies 61, pp. 301–318. Philipps, Nelson/Hardy, Cynthia (2002): Discourse analysis: Investigating processes of social construction. London: Sage. Rear, David/Jones, Alan (2013): Discursive struggle and contested signifiers in the arenas of education policy and work skills in Japan. In: Critical Policy Studies 4, pp. 375–394. Reiher, Cornelia (2009): Bestimmt der Staat, was auf den Tisch kommt? Die Umsetzung des Rahmengesetzes zur Ernährungserziehung im ländlichen Japan. In: Japan 2009, pp. 63–88. Reiher, Cornelia (2014): Lokale Identität und ländliche Revitalisierung. Die japanische Keramikstadt Arita und die Grenzen der Globalisierung. Bielefeld: Transcript. Reiher, Cornelia (2017): Food safety and consumer trust in post-Fukushima Japan. In: Japan Forum 29, pp. 53–76.

403

References Rinscheid, Adrian/Eberlein, Burkard/Emmenegger, Patrick/Schneider, Volker (2019): Why do junctures become critical? Political discourse, agency, and joint belief shifts in comparative perspective. In: Regulation & Governance. DOI: 10.1111/rego.12238. Roberson, James E./Suzuki, Nobue (eds.) (2003): Men and masculinities in contemporary Japan: Dislocating the salaryman doxa. London: RoutledgeCurzon. Samuels, Richard (2013): 3.11: Disaster and change in Japan. Ithaca, NY: Cornell University Press. Sauzier-Uchida, Emi (2014): The rise of consumer-oriented politics in Japan? Exploring the party–citizen relationship through discourse analysis. In: Japanese Journal of Political Science 15, No. 2, pp. 231– 257. Schad-Seifert, Annette (2007): Dynamics of masculinities in Japan: Comparative perspectives on Men’s Studies. In: Derichs, Claudia/Kreitz-Sandberg, Susanne (eds.): Gender dynamics and globalisation: Perspectives from Japan within Asia. Münster: LIT, pp. 33–44. Schad-Seifert, Annette (2016): Der Grasfressermann als mediales Trendwort im Diskurs um Japans Lost Generation. In: Köhn, Stephan/Unkel, Monika (eds.): Prekarisierungsgesellschaften in Ostasien. Wiesbaden: Harrassowitz, pp. 261–281. Schäfer, Saskia (2011): Expanding the toolbox: Discourse analysis and Area Studies. In: Schneider, NadjaChristina/Gräf, Bettina (eds.): Social dynamics 2.0: Researching change in times of media convergence: Case studies from the Middle East and Asia. Berlin: Frank & Timme, pp. 145–164. Schiffrin, Deborah/Tannen, Deborah/Hamilton, Heidi E. (eds.) (2005): The handbook of discourse analysis. Malden, MA: Blackwell. Schmidt, Vivien A. (2011): Speaking of change: Why discourse is key to the dynamics of policy transformation. In: Critical Policy Studies 5, No. 2, pp. 106–126. Schneider, Florian (2019): How to do a discourse analysis. www.politicseastasia.com/ studying/how-to-doa-discourse-analysis/, [Accessed 02 March 2020]. Slaby, Jan/Mühlhoff, Rainer/Wüschner, Philipp (2017): Affective arrangements. In: Emotion Review 11, No. 1, pp. 1–10. DOI: 10.1177/1754073917722214. Taniguchi, Tasaki (2018): Changing political communication in Japan. In: Darling-Wolf, Fabienne (ed.): Routledge handbook of Japanese Media. London: Routledge, pp. 121–135. Tollefson, James W. (2014): The discursive reproduction of Technoscience and Japanese national identity in The Daily Yomiuri coverage of the Fukushima nuclear disaster. In: Discourse & Communication 3, pp. 299–317. Torfing, Jacob (1999): New theories of discourse: Laclau, Mouffe and Zizek. Oxford: Blackwell. Turabian, Kate L. (2007): A manual for writers of research papers, theses, and dissertations. Chicago, IL: University of Chicago Press. Ushikubo, Megumi (2008): Sōshokukei danshi ‘ojōman‘ ga nihon wo kaeru. Tōkyō: Kōdansha. van Dijk, Teun A. (1995): Discourse semantics and ideology. In: Discourse & Society 6, pp. 243–289. van Dijk, Teun A. (1997): The study of discourse. In: Dijk, Teun A. van (ed.): Discourse as structure and process. London: Sage, pp. 1–34. Volkens, Andrea/Krause, Werner/Lehmann, Pola/Matthieß, Theres/Merz, Nicolas/Regel, Sven/ Weßels, Bernhard (2019): The manifesto data collection: Manifesto Project (MRG/CMP/MARPOR). Version 2019b. Berlin: Wissenschaftszentrum Berlin für Sozialforschung (WZB). DOI: 10.25522/manifesto.mpds.2019b. West, Richard L./Turner, Lynn H. (2018): Introducing communication theory: Analysis and application. New York, NY: McGraw-Hill Education. Wetherell, Margaret (2012): Affect and emotion: A new Social Science understanding. Los Angeles, CA: Sage. White, Daniel (2011): The affect-emotion gap: Soft power, cultural diplomacy, and nation branding in Japan [Dissertation]. Houston, TX: Rice University. White, Daniel (2015): How the center holds: Administering soft power and cute culture in Japan. In: Ertl, John/Mock, John/McCreery, John/Poole, Gregory (eds.): Reframing diversity in the Anthropology of Japan. Kanazawa: Kanazawa University Press, pp. 99–120. White, Daniel (2017): Affect: An Introduction. In: Cultural Anthropology 32, No. 2, pp. 175–180. Yoda, Tomiko (2000): A roadmap to millennial Japan. In: The South Atlantic Quarterly 4, pp. 629–668. Žižek, Slavoj (1989): The sublime object of ideology. London: Verso.

404

Chapter 15 How to finish: Writing in a stressful world Chris McMorran

1. Introduction In late 2007, my PhD research in Japan ended several months sooner than expected. I had been in Kumamoto for 17 months, studying the physical and emotional labour of hospitality in Japanese inns (ryokan), but my wife and I were not yet ready to return to the United States. As November drew to a close, our return was decided for us, as we learned that my father was dying from cancer. He was being moved to a hospice, and if we wanted to share any final moments with him, we needed to return as soon as possible. Within two days, we packed our belongings, sold our car, and hopped on a plane. In a flash, my precious time of field research ended. After the funeral and a few weeks spent trying (and failing) to comfort my mother, I took stock of my situation. I had more than a year’s worth of fieldnotes, transcripts from a few dozen interviews with ryokan employees and owners, and a mountain of ephemera, like ryokan brochures and daily work schedules. I needed to make sense of everything and write a dissertation that would meet the standards of my department, contribute something original to my discipline and do justice to everyone who had supported my project, from funding agencies and advisors, to informants and family. I had a six-month writing fellowship waiting for me at the University of Colorado, but nothing secure after that. My loving partner had made sacrifices for me for nearly a decade, following me back and forth between the U.S. and Japan. I owed it to her to complete the PhD as quickly as possible and begin my career. I share this critical juncture in my life to highlight some of the ways intersecting challenges— from disciplinary expectations to time management and family responsibilities—can interrupt or set in motion the final stages of a research project. In this chapter, I discuss how to turn one’s research into a coherent whole, while also being pragmatic about organisation, writing habits, deadlines and more. I highlight the importance of communicating your work to the appropriate audience and establishing routines. I conclude by discussing the importance of maintaining a healthy work-life balance amid the realities of a precarious job market and heightened expectations in today’s universities. I wrote this chapter with PhD students in mind, but I hope all those who must write to succeed in academia will find it valuable.

405

Chris McMorran

2. Getting started Often the greatest obstacle to completing a project is deciding how to begin. When I returned from the field as a PhD student, I felt overwhelmed by the volume of information I had collected and the expectation to contribute something valuable to my discipline of Geography. Some researchers return from the field to find so much new work has been published on a similar topic—whether conceptually or empirically—that they fear they have nothing new to say. Others struggle to switch from one phase of the project to the next; from collecting data to producing knowledge. Sometimes it helps to remember that there is no best way to communicate research to an audience. You must trust yourself, choose a way and stick with it to completion. Fortunately, scholarly outputs follow established patterns, or as Christian Tagsold (see this chapter, Ch. 15.2) puts it: ‘academic texts follow the rules of their community’. You can use these rules to your advantage. As a permanent faculty member at the National University of Singapore, when I first meet graduate or undergraduate students, I suggest they begin any project by reading several recently completed dissertations or honour’s theses. I do not expect them to read cover to cover, but to see how the chapters are structured and to understand what is expected by their department and discipline. How long are they? How are the introduction and literature review structured? Are the methods buried in an early chapter, or do they stand alone in their own chapter, or are they omitted entirely? What is the overall balance between conceptual discussion and empirical findings? What is the average number of references cited? All writing must address an audience (see Farrer/Liu-Farrer, Ch 17), and this audience— whether members of a PhD committee, readers of a particular journal or a general audience— will expect you to follow a certain structure. PhD dissertations must meet departmental and disciplinary expectations, and most journals not only have strict word limits, but also may expect contributors to follow a very narrow style and structure. It is best to understand these general expectations before you begin writing up. Key ideas There is no single best way to structure your work. Refer to dissertations or articles recently published in your field and take note of standard conventions: length, number of chapters, balance between conceptual framing and empirical findings, etc. Create a rough outline of your own work that would correspond to these conventions, then proceed with confidence. It will never be perfect, but it will be finished.

3. Engaging with theory While there are well-established ways to structure a dissertation or journal article, there is more flexibility when it comes to engaging with theory. This is one place where you can ‘be brave’ (see this chapter, Samuels, Ch. 15.3). In graduate courses, students are often exposed to 406

Chapter 15 How to finish a range of theories and theorists. In my own graduate days, I read Judith Butler (2006), Michel de Certeau (2011), Tim Cresswell (1996), Michel Foucault (1995), David Harvey (1989) and Doreen Massey (1994). My professors assigned these scholars, and we spent hours reading and debating their work. They were held up as exemplars of the ‘big ideas’ our work should engage with and push in new directions. When we developed our own theoretical frameworks and imagined the scholarly debates to which our work might contribute, we naturally mentioned these authors. At the same time, we were warned not to simply ‘apply’ their theories to new cases, especially cases outside the Euro-American theoretical context where those ideas emerged, such as Japan. Many of us felt stuck. How were we supposed to engage with those theories? Challenge them? Ignore them? There is no easy answer, but engage with them we must. There are many metaphors that can be used to conceptualise such an engagement in the practice of scholarship: a hunt for knowledge, a garden to be tended, a journey to an unexplored planet. I find the metaphor of scholarship as a conversation particularly useful. This is commonly referred to as the Burkean parlour (Burke 1941), an imaginary room filled with scholars deep in conversation about a particular field. As you enter the room and walk around, you overhear different conversations, gradually recognising the big questions in the field. The real-life equivalent is the literature review (see Zachmann, Ch. 4). Eventually, you must step into one of the small clusters and join the conversation. Perhaps you know a case that contradicts, complicates or supports what others are saying. When you write any academic piece, you must similarly choose a field and join its ongoing conversations. The venue helps determine the conversation and how you might enter.

3.1. PhD dissertations A dissertation in History or Anthropology must address a slightly different audience than one in Japanese Studies, and each work must engage with the field’s central themes. These were likely established in the research proposal, but now that you have returned from the field (literally or figuratively), you must contradict, complicate or support the theories and answer the research questions that framed your study. A dissertation tends to be a very specialised work written for a very small audience: a committee and perhaps an outside reader. Dissertations tend to be full of jargon, and most include a long literature review (see Zachmann, Ch. 4) that serves a specific purpose: demonstrating your broad grasp of the field and providing a conceptual framework for your findings. There are many ways to organise and present your empirical findings in subsequent chapters, but consider how you might arrange them in a way that complements the context of the study. For instance, one of my students did an ethnography of a Japanese high school in Singapore and arranged his empirical chapters sequentially, from opening ceremony to graduation (Toh 2019). Similarly, Lieba Faier’s (2009) study of Filipina women in rural Japan moved from their dreams of working abroad while in the Philippines, to their entrance into hostess clubs in rural Japan, to their marriage to Japanese men, and finally, for some, to their decision to run away. Empirical chapters might also be arranged spatially, moving from one research site to another. For instance, many classic ethnographies of Japan begin with a walk around a neighbourhood, before going ‘deeper’ inside the businesses and homes in later chapters (Bestor 1989; Kondo 1990). Finally, you might highlight three or four

407

Chris McMorran key words or phrases that were meaningful to the people in your study and use each one to frame a chapter.

3.2. Journal articles A journal article is a very different piece of scholarship that requires a different way of engaging with theory and presenting empirical results. If it is the flagship journal of a major academic society, the article will need to address a major trend in the discipline and make a significant contribution. If it is a specialist journal or an Area Studies journal, the breadth of theory can be narrower, but the article will need to make a significant empirical contribution (see Farrer/Liu-Farrer, Ch. 17). Read several recent issues to get a sense of the balance between depth and breadth. Some scholars write a rough draft of an article before deciding where to submit it, while others find it more efficient to first identify the journal they want to engage with. As you read the journal, ask yourself: What conversations are occurring in the discipline, and how are these being addressed in this particular journal? What are the gaps, and how does your work address these gaps, either through your empirical or conceptual findings? By writing your article with a specific journal in mind, you will not only write to suit the journal’s structure and style, but also be in a better position to join the conversations found in its pages. For example, a few years after I completed my dissertation, I became excited by the field of mobilities, especially as its scholarly conversations were being presented in the still new journal Mobilities (established 2006). ‘Mobilities’ referred to scholarship across the Social Science fields that understood the movement of people and ideas around the world not as a simple, free-flowing, judgement-free process, but as deeply meaningful and shaped by political relations on multiple scales, the creation and policing of borders, and unequal access to movement based on race, gender, class and more (Sheller/Urry 2006). However, the more I read, the more I felt something was missing. During my PhD fieldwork in Japan, I had witnessed a complex relationship between ‘mobility’ and ‘fixity’ (the terms found in the field) among ryokan owners and employees, and a deep ambivalence people felt about both. I did not write explicitly about this in my dissertation, but when I read more scholarship on mobilities later, I found a gap in the conversation I could fill. My resulting article in the journal Mobilities drew on findings from Japan to suggest a new way of thinking about mobility (and fixity) beyond Japan (McMorran 2015). As Richard Samuels (see this chapter, Ch. 15.3) and Kaori Okano (see Ch. 3) remark, we students of Japan are often told ‘Japan is unique’. However, many of the questions we ask about Japan apply beyond Japanese Studies. Whenever possible, we should try to contribute to these broader scholarly conversations, as Samuels did in his work comparing the post-disaster narratives of 3.11 and Hurricane Katrina, or as Aya H. Kimura (see this chapter, Ch. 15.1) did through her introduction of the idea of ‘food policing’. Journal articles provide a useful format for thinking beyond Japan and making theoretical contributions to wider conversations outside Japanese Studies, and not just adding case studies from Japan to ideas developed by others.

408

Chapter 15 How to finish

3.3. Books Books typically provide the most freedom for engaging with theory. One example of an innovative method of engaging with theory comes from anthropologist Nancy Rosenberger in her book Dilemmas of adulthood (2013). In this study of over fifty Japanese women carried out over two decades, Rosenberger frames her work within broad scholarly debates about ‘agency’ and ’resistance’, but she does so without the typical literature review. Instead, she reproduces a conversation she had with two Japanese graduate students one night over Indian food in Tokyo. Rosenberger tells the students about her struggle to apply these widely-used concepts to her informants, and a student replies, ‘They don’t fit my feeling of what it is like to be a woman in Japan now. It’s too hard to resist in Japan’ (2013, p. 4). During the meal, the students discuss different authors who have been at the centre of scholarly conversations about agency and resistance, including Lila Abu-Lughod (1990) and Sherry Ortner (1995), and they explain why these perspectives don’t exactly suit their own lives. Instead of providing readers a dry literature review, Rosenberger lets her conversation with the students frame both the existing scholarship and her effort to join the scholarly conversation with a concept that corresponds better to the experience of these Japanese women: long-term resistance. Through effective storytelling (see this chapter, Tagsold, Ch. 15.2) and her own version of scholarship as conversation, Rosenberger reveals both her struggle with theory and her efforts to develop theory to engage audiences beyond Japanese Studies. Key ideas Scholarship as a conversation is a useful metaphor for conceptualising the practice of scholarship. There are significant stylistic and structural differences between dissertations, articles and books, but each format requires you to choose your audience, find a gap in the existing conversation and make an original contribution that engages with theory. In Japanese Studies, we must find ways to contribute more than case studies that simply support existing theories.

4. Establishing good habits for writing Once you know how to structure your work and what scholarly conversations you want to join, you still need to make a habit of writing in order to finish in a timely manner. If you enjoy writing and always complete work on time and without stress, you may skip this section. The rest of us must find effective ways to manage our writing obstacles and establish good habits.

4.1 Avoid writing myths Writing myths are common ideas people have about writers and writing that can cripple productivity. Joseph Moxley (1992) outlines fourteen such myths. These include myths like gifted

409

Chris McMorran writers are overflowing with ideas, writing is a lonely craft best done by introverts and gifted writers rarely revise. One particularly toxic myth is that writers enjoy writing. In fact, Moxley notes that many prolific writers freely admit that they don’t enjoy writing. For them, writing can be painful, but it pays the bills. Moxley argues that if you hold the myth that writers enjoy writing, you may worry something is wrong with you because you do not enjoy it. To make his point, Moxley quotes Lafcadio Hearn (1920), an educator and author who lived and taught in Japan at the turn of the twentieth century, including at Tokyo Imperial University. Hearn wrote: ‘Nothing has been more productive of injury to young literary students than those stories, or legends, about great writers having written great books in a very short time. […] Above all things do not imagine that any good work can be done without immense pains’ (1920, p. 36; cited in Moxley 1992, p. 8). For a rare few scholars, writing is a joy. But for most of us (myself included), writing is work, and it is perfectly normal to not enjoy it. Another obstacle that interferes with writing is the belief that careful scholarship requires long blocks of uninterrupted time. Whether this means four to five hours of uninterrupted time in a day or four to five weeks of uninterrupted time during a school break, many people believe they need such long periods in order to write, and that without such blocks of time, they simply cannot write. Graduate students without families who have just returned from the field may experience that rare luxury of a semester, or even a year, to do nothing but write. But the reality is that most scholars’ lives are full of interruptions, and long blocks of uninterrupted time are increasingly rare. Those who believe they can only write during long stretches of uninterrupted time soon find themselves unable to write anything, as teaching, service, job searches, family, friends, meals, exercise, hobbies, childcare and other essential elements of a busy, rewarding life make such writing blocks impossible. It is best to develop brief, regular writing habits early and maintain them as best you can.

4.2. Writing routines So how can you establish good habits? One particularly prolific writer with something useful to say in this regard is novelist Haruki Murakami: To keep on going, you have to keep up the rhythm. This is the important thing for longterm projects. Once you set the pace, the rest will follow. The problem is getting the flywheel to spin at a set speed—and to get to that point takes as much concentration and effort as you can manage. (Murakami 2008, p. 5). Write every day, and as Kimura (see this chapter, Ch. 15.1) points out, this daily practice should even begin during the research process, ideally during fieldwork. Indeed, routine writing is one of the central ideas in Robert Boice’s (2000) insightful Advice for New Faculty Members. Boice’s long-term research with new faculty members suggests that writing for even 15-minute chunks, between meetings or before classes (instead of checking emails or posting on social media), makes the difference between scholars who are highly productive and just average. One trick is knowing when to stop each day. Again, Murakami offers some advice. ‘I stop every day right at the point where I feel I can write more. Do that, and the next day’s work goes surprisingly smoothly’ (2008, p. 5). Samuels (see this chapter, Ch 15.3) mentions a similar practice by a former professor who never left his desk ‘without writing the first sentence of the 410

Chapter 15 How to finish next paragraph of whatever manuscript he was wrestling with at the time’. These writers begin each day with momentum. As for the writing process itself, there are many ways of staying organised. Samuels (see this chapter, Ch. 15.3) uses physical notecards, which makes it possible to keep everything within grasp and to quickly make connections that you might overlook. Others organise their ideas and write in the cloud, which makes it possible to work anywhere, search for terms quickly, and copy and paste from one place to another. Some write out everything by hand, since they spend too much time revising when working on a screen. On paper, their ideas flow more smoothly, and they can cross out a word or insert an idea more easily. Others dictate long passages to themselves, which they later transfer to the computer by hand or through a transcription function. None of these tricks will make writing (and revising and rewriting) easy. Routine writing will not eliminate the need to wrestle with big ideas or the self-doubt most writers feel. But by making writing routine, you can keep the ball rolling. If you wait for those big blocks of time, you may find it increasingly difficult to return.

4.3. Write with others Another way to help make writing routine is to make writing social. Kimura (see this chapter, Ch. 15.1) mentions the value of joining a writing group that meets regularly. Both aspects of this writing group are important and speak to themes covered earlier in this chapter. First, the writing group meets regularly, which incentivises the participants to maintain momentum. Second, the writing group requires participants to consider the audience. Returning to the metaphor of scholarship as a conversation, regularly sharing your work with others (either in your field or not) forces you to clearly communicate your ideas. Having constant feedback from others in the group, or even hearing your writing read aloud, keeps the project moving forward and directed at the audience. Some writers work fine alone, setting their own goals and keeping to a self-directed schedule. But many of us need some peer pressure and support to accomplish our goals. A writing group makes you accountable to someone who is supportive and equally in need of support. Even if a writing group does not suit your style, remember that you will always be writing within a community of scholars whose ideas you must engage with. Some fledgling scholars struggle to decide what to say, largely because they have not yet decided who they want to say it to. Joining a writing group, or imagining your community of readers, may help you maintain the pace and finish the project. Key ideas Finishing scholarly projects takes routine and concerted effort. Learn to recognise the obstacles that prevent writing, establish and maintain a writing routine, and make your writing social by writing to an audience and sharing your work with others. Establish good habits as early as possible (even during fieldwork), so you can remain active even after other commitments (family obligations, teaching, service, job searching) take up more of your time.

411

Chris McMorran

5. Managing mental health Throughout this chapter, I have implied that completing a research project can be stressful. Of course, PhD students writing dissertations are not the only ones who feel under pressure. Indeed, universities have increasingly become institutions characterised by unhealthy and unsustainable levels of competition and stress. In recent years, prospective students have found it more difficult to gain entrance to them, undergraduates feel more competition to achieve excellent grades and get a good first job, and many students face rising tuition costs and the long-term impacts of rising student debt. Graduate students find the job market more competitive than ever, and recent PhD graduates are increasingly turning to precarious, short-term post-doctoral positions or adjunct teaching positions as they wait for an elusive permanent job. Plus, the competition among universities in national and international rankings of schools trickles down to faculty members fortunate enough to have permanent positions. They are expected to produce a greater quantity and quality of publications; to teach more, with better student feedback; to do more committee work; to hold more leadership positions in international scholarly societies; and to make an impact beyond the university through outreach to the community or as public intellectuals. This heightened competition has negatively impacted the well-being of students and faculties alike. One recent report prepared by the American Association of Geographers cited ‘the proliferating number of peer-reviewed studies which indicate that anxiety, depression, and other mental-health conditions have reached crisis proportions in the North American academy and beyond’ (Peake et al. 2018, p. v). While some of the stress of university is inevitable—it is work after all, and most workplaces involve some stress—you do not have to manage that stress on your own. I conclude this chapter by discussing ways to manage stress and help coproduce an academy that is a safer, healthier and more supportive environment for all. Throughout this chapter, I have argued that being pragmatic—by knowing your audience, writing to suit the format (dissertation, journal article, book), making writing social, and writing regularly—can help you manage some of the stress associated with finishing your project. The key is striving for work-life balance. This means not only writing every day, but also exercising regularly, eating healthy meals, getting enough sleep, maintaining social relationships and pursuing hobbies. Attend concerts, be active in politics and read novels. Graduate students and early faculty members in particular tend to imagine they have no time for these things, which take precious time away from their writing. But no one can write for ten hours a day every day without quickly burning out. If you push too hard, you may endanger your physical and mental health. I have colleagues who scuba dive, sing in community choruses, play brass instruments and admit (in embarrassment) to watching too much television. These things make us interesting and make our work lives sustainable. Plus, we need outside sources of inspiration and perspectives, which are unlikely to come from staring at our own work. Admittedly, some sources of stress may be outside your control. In the opening vignette, I mentioned my father’s unexpected death. It prematurely ended my PhD field research, but it also put my own life in perspective. You cannot plan for such life-altering moments, but hopefully you can build a support network of friends, family, mentors and fellow students or faculty members who can help you through. When that is not enough, most universities have counselling services that provide valuable support for anything from writer’s block to family issues 412

Chapter 15 How to finish and depression. By utilising these services and openly addressing the realities of the stressful 21st century university, we can help destigmatise the topic of mental health in academia and help make the university more supportive for all. Key ideas The 21st century university is a stressful place, characterised by increasing student debt in many countries and precarious post-doc and adjunct positions, and growing expectations on faculties everywhere. It is important to care for both your mental and physical well-being by maintaining a healthy work-life balance and to seek support from professionals when needed.

6. Conclusions In today’s increasingly stressful universities, completing the final stages of a research project requires careful planning, a healthy work-life balance and writing, writing, writing. Use the format and expectations of your venue—dissertation, journal article or book—to help you plan your writing. Know your audience—both within Japanese Studies and beyond—and join their conversations through your writing. Avoid common writing myths that may threaten to derail your progress. Use others, and support them in return, to receive feedback on your writing and maintain momentum. Plan time for writing as part of a healthy daily routine. And do not hesitate to admit your struggles and utilise professional counselling services when necessary. Life is messy and largely out of our control. One of the things you can control is your commitment to writing. Write, and you will feel the satisfaction that comes from completing projects and contributing to the world of knowledge.

413

15.1 Training your ‘writing muscle’: Writing constantly and theoretically Aya H. Kimura

As a writer, I have faced multiple issues that are perhaps familiar to many students, such as a lack of time, self-criticism and the resultant paralysis, or information glut. There is now a lot of writing on how to write, so I do not want to replicate it here. In this chapter, I will focus on three issues that might be particularly interesting to Japanese Studies scholars: when to write, how to make writing social and the audience of your writing. Using the metaphor of ‘writing muscle’, these three issues might be phrased as follows: Muscle needs to be constantly trained. Training is fun and often more effective when you do it with others. And muscles need diverse exercise; you write for different audiences and often we need to stretch our muscle a bit to speak to the larger, non-Japanese Studies audience.

Write as you go For those of us who do fieldwork in Japan but do not teach or study there, our time in the field is often very limited. So there is a temptation to just focus on collecting data (do interviews, participant observation) and wait to go home to start writing. Writing up field notes and interview notes itself takes a lot of time, so I often feel that I am too tired to write anything beyond them. However, writing is not only about communication, but also about analysis. Writing hones thinking and helps crystallise what the researcher is exploring in fieldwork. There is something about writing that forces you to explain what you are thinking in detail and in a logical manner, so it is a good idea to always write, even in the field. There is a benefit to this approach of writing during fieldwork. I can think of so many instances when new ideas about additional interviewees or interview questions emerged in the process of my preliminary writing. And I did not have to wait for another year until I went back to Japan to do these interviews. How do I write as I go? I usually create a research log document. Research log documents could contain free writing in addition to notes on research activities like interviews and meetings. I also create several folders on my computer with emergent themes or book chapters. I metaphorically ‘throw in’ bits of things as I continue my research. For instance, I might read some books I feel I could use for theoretical framing. I would write a short summary, create a Zotero (bibliographic software) entry and insert that citation with my summary into a particular chapter folder. Never wait to create a bibliographic entry—it will be a nightmare if you wait until the very end of your manuscript! I might read a newspaper article—I might take a picture of it and paste that in, or find a web version and put the link in this document, and write several sentences about it. I also throw paragraphs from my free writing into these folders. Writing can be like a sport where consistent exercise of your muscle could have a good 414

Chapter 15 How to finish pay-off. Obviously, there will be a lot of junk in free writing. The point is to use your writing muscle consistently and to use the momentum when it emerges—even when you feel like you cannot envisage the whole picture yet.

Making writing social The word ‘writing’ can invoke the image of a lone writer sitting in a den, staring at their computer. But it can be very communal if you have something like a writing group (Luker 2010). Having an audience and a sounding board makes writing fun, engaging and less isolating. When I started as an assistant professor at the University of Hawai‘i, my colleague Christine Yano invited me to a writing group. What a blessing it was! This writing group’s system is so brilliant that I believe it has contributed immensely to my productivity. Here is how it works: the group meets every week. Unlike other writing groups that I had been in, this system does not require you to produce a full-length paper and read a number of papers by others, which eats up a lot of time. But our system is to just come prepared to read your writing for seven minutes. It can be an abstract for a conference, or two pages from your manuscript. You read and others listen, with pencils and scratch papers at the ready to jot down notes. And then you read the same thing a second time. After the reading, the listeners take turns giving comments. Another rule is that when you get comments, you do not debate or argue. You just take the comments in, and the group moves on to the next person. This way, each person is done within about twenty minutes. Writing groups force you to produce something weekly at least and get feedback on it. In the system that I described, reading out loud is also effective in pinpointing what works and what does not. Because people are listening rather than reading the paper, your sentences cannot be too dense, which improves the clarity of writing. Writing groups can contain non-Japanologists. Christine Yano is a distinguished scholar in Japanese Studies and I have benefited immensely from her comments, which are based on a deep knowledge of Japanese culture and society, but other members can come from outside disciplines. In fact, having those fresh sets of ears is often very helpful in you not glossing over important issues. Another long-standing member of the writing group is ethnomusicologist, Jane Moulin. I had little clue what she was writing about (Tahitian dance and festivities, for instance), and I am sure she was the same way with my topics, such as Japanese consumer cooperatives and agriculture. But having to explain your argument without resorting to disciplinary shorthand is really helpful in refining it.

Writing to different audiences For those of us who study Japan, it is rare that we can only operate in Japanese Studies. For instance, topics like post-Fukushima citizen monitoring, food education (shokuiku) or kids’ cafeterias (kodomo shokudō) may all be interesting to Japanologists, but any writing on them needs to be able to articulate why non-Japanese Studies scholars should also care about them. It has been one of my biggest challenges to relate to the grant agencies and journal reviewers who might say, ‘Well, that only happens in Japan. Why should I care?’ I strive to write about

415

Aya H. Kimura Japan in a way that explains why Japanese case studies provide insights into broader dynamics that affect many other societies. Of course, this question of broader relevance is not limited to Japanologists or Area Studies scholars: social scientists who use qualitative methods and Small-N cases face this question, too, as they try to contribute to social analysis beyond localised and historically specific observation. If you are conducting in-depth ethnography or interviews with a limited number of people, the question of broader relevance and importance is paramount. Lofland et al. (2006, chapters 6 and 7) suggest eight questions with which to frame your analysis so that you can convey broader sociological relevance. These abstract propositions may help students situate their Japan-based cases in relation to more general social inquiries. The Japanese case study might contribute a new setting or identify a new aspect of types, structures, strategy or others: 1. 2. 3. 4. 5. 6. 7. 8.

Type: what are the topic’s types? Frequencies: what are the topic’s frequencies? Magnitudes: what are the topic’s magnitudes? Structures: what are the topic’s structures? Processes: what are the topic’s processes, such as sequences and cycles? Causes: what are the topic’s causal relations? Consequences: what are the topic’s consequences, for whom/what? Agency: where is the agency? How do people use strategies and tactics?

When reporting the research results, the use of metaphors and ironies can help abstract your observation and allow you to move away from local particulars (Lofland et al. 2006). I try not to come up with new jargon or concepts because I think they tend to unnecessarily make the writing esoteric and inaccessible. But good metaphors could crystallise the essence of the phenomenon and help to uncover similarities with other contexts. For instance, in my book, Radiation brain moms and citizen scientists (Kimura 2016), I used the term ‘food policing’ to capture how radiation concerns were seen as harmful and socially sanctioned. This concept emerged out of my observation of post-3.11 Japan, but has the potential to be applied to a non-Japanese context where citizens’ fears about food quality may be dismissed and ridiculed as irrelevant and unscientific. The challenge is how to do the abstraction, but do it in a way that still captures localised dynamics. For instance, another concept that I came up with in the book was ‘activist joshi’. This may be described as ‘irony’ because activists and girly girls (joshi) are often not thought of as compatible with each other. I was trying to capture the phenomenon of feminine women who are socially active but yet friendly and properly feminine (not angry or outraged). In the book, I tried to explain this in relation to the history of Japanese women’s mobilisation (specific and localised) but also neoliberal feminism (broader and global relevance). Another way to think about the broader relevance is to use existing literature and sociological conceptions. I have been trained in the Extended Case Method (ECM) (Burawoy 2009), so I tend to follow some variant of it. Using ECM means that I start with some ideas about theories that guide my questions in my fieldwork. ECM directs you to use a case study; not to come up with a brand new theory of something but to refine the existing theory. This can happen when you find a ‘puzzle’ or ‘anomaly’ in your case that cannot be explained with existing theories. Take the case of citizen radiation measuring organisation that I wrote about in my book and also continue to theorise on in my co-authored book on environmental citizen science (Kimu416

Chapter 15 How to finish ra/Kinchy 2019). I started my fieldwork with some expectations based on existing theories of citizen science. They tended to portray citizen scientists as mavericks and radicals, using data to challenge polluting corporations and complacent bureaucracies. The primary example were environmental justice activists in the U.S. who collected their own health and environment data to prove pollution and environmental harm in order to hold corporations and government agencies accountable. But then, once I started my fieldwork and conducted interviews, I found tremendous diversity in citizen radiation monitoring organisations (CRMOs). There were obviously strongly activist-oriented ones, but I also saw a mail-in CRMO in which there was not really much interaction between the CRMO and ‘customers’. It could well be like a private lab. I also saw many that were kind of in the middle—they did not want to seem ‘political’ and they did not want to be seen as ‘activists’. In ECM’s parlance, here was an ‘anomaly’ or ‘problem in the field’ that you could leverage to refine the pre-existing theory. The difficulty for Japanese Studies scholars or, for that matter, those who work in non-EuroAmerican contexts, could be that it would be tempting to see the ‘puzzle’ as something that is purely explainable due to the case’s non-Western context. For example, the above instance of subdued political activism on the part of CRMOs may be explained as a manifestation of the strong state, the weak civil society and the culture of conformity and harmony in Japan. Of course, these particularities and context dependency do matter. In fact, there is a bit of privilege in being ‘the norm’ if you are doing Sociology in an American context; you do not have to defend why you are studying the U.S. to journal reviewers and editors. But Area Studies scholars have to master the intricate dance of not negating the deeply rich history and a cultural understanding of the context that they work in, but retain the ability to talk to broader scholarship.

Conclusion Japanese Studies scholars may face challenges of writing in a way that has resonance and relevance to broader academic fields. But this compels researchers to answer ‘so what’ questions in a focused manner, improving the quality of the writing in the end. Despite the limits imposed by the necessity of fieldwork at a distance, writing during research is recommended. Writing is like a physical exercise: the more regularly you do it, the easier it is to maintain momentum and continue with it. You can also avoid procrastination and burnout if you do it with other people. Having non-Japanologists in your feedback circle may also improve your writing if it helps you to unpack issues that we might take for granted. Although not a perfect match, the analogy of training muscle is helpful. You need to constantly write to train your writing muscle. Make it engaging and get feedback immediately by having a social network that you can turn to. And we need to write to different audiences. Area Studies scholars often need to justify their case vis-à-vis the presumed ‘universal’ that is often embedded in Euro-American contexts. But take such pressure as a good opportunity to flex your writing muscle in new and innovative ways.

417

15.2 Writing stories Christian Tagsold

In the early 1980s, a volume changed the worldview of many anthropologists. In Writing culture, edited by James Clifford and George E. Marcus (1986), various authors discussed the impact of creating texts in Anthropology—field notes, research reports, but most of all papers and books. Hitherto, bringing observations, interviews and other anthropological data into a textual form had mostly been seen as a mere technical task. The resulting text was judged according to its ability to convey the meaning of data and theories to its readers. Reviews of such texts certainly did not neglect style and prose altogether; though they made little effort to discuss these systematically. The ensuing debate on Writing culture not only influenced anthropologists but spilled over to the Humanities in general. Ever since, our texts have been scrutinised not just for their content, but also for the language creating their very meaning. Writing culture thus brought the gist of the so-called linguistic turn into the realm of academic writing. The linguistic turn, a theoretical movement of the 1970s, posited that language is not simply an instrument that replicates reality. On the contrary, it creates this very reality of its own accord. The ramifications of the linguistic turn may sound terrifying to those who still want and have to produce texts on their research. But I would argue instead that the self-reflective stance of Writing culture enables us to think anew about the trade of writing and to develop stronger self-confidence, ultimately coupled with pleasure, concerning this task. In the following, I will introduce a playful approach to organising research results and forging them into words. This approach frees us from the heavy burden of perfectly replicating reality—a lofty aspiration that many thinkers of the linguistic turn would ultimately see as vain.

The power of stories If it is not in language’s ability to faithfully replicate reality, the persuasiveness of academic texts must lie elsewhere. Two main factors are relevant here. First of all, academic texts follow the rules of their community. These rules turn strings of words into significant descriptions which are read, shared, referenced and quoted by other texts of the same genre. Second, texts gain their power of persuasion from a good and appropriate storyline. It is not enough to gather data and dump it into a Word document. The data has to come alive before the readers’ eyes. It has to be wrapped into a convincing and compelling narrative. Usually, we learn the first set of rules at university when being introduced to the craft of writing. Examples consist of employing neutral, scientific language, which stresses ‘objectivity’ and avoids first-person pronouns or perspectives. Often enough, however, authors who overemphasise these rules write quite uninspiring texts because they are terrified of overstepping boundaries. The skill of forging a good and convincing story, however, is rarely systematically taught. Lately, ‘storytelling’ has become a fashionable buzzword and is marketed as a miracle cure for academics

418

Chapter 15 How to finish who struggle to gain the attention of a broader readership. Gurus of scientific storytelling, such as the mathematician Marcus du Sautoy (2015) and maritime biologist Randy Olson (2015), argue that by packaging our findings into narratives, our data will arouse wider interest. Furthermore, it will help readers to remember the information better, because humans have been used to stories as a tool of information for practically as long as we can remember. I am very much convinced of the power of storytelling myself, but would go one step further. For the theory of storytelling, the narrative usually comes at the end of a research process. First, we create data by means of our methods. Then, we analyse this data and draw our conclusions in line with the standards of our academic community. Finally, we translate these conclusions into a story. However, I am sceptical about this model, and in my own work stories come into play from the very beginning. I doubt that we can create good stories without material that has a good narrative core. But in order to find this core, we must interact with our data within the framework of storytelling at a much earlier stage than in the model just introduced. And we already do, if we think about it. Even quantitative social scientists rely on the power of typical narrative elements when analysing their data. On discovering a correlation between variables, they try to find a good and convincing explanation that clarifies why a link exists between two sets of numbers. Without making sense of the mere mathematical correlation, social scientists would discard the findings in their data as meaningless and not follow up on this numerical coincidence. If even quantitative research relies on small stories to stitch the elements together, qualitative research should do so even more. After all, fieldwork and interviews usually tap a rich reservoir of stories. People we meet in the field tell us stories about themselves and how they view the world in informal talks and chats. Interviewees do the same in a more formalised setting. Last but not least, written sources like newspaper articles and books employ stories. Having recorded and gathered these stories from the field, we can rearrange and refocus them to suit our needs and answer our research questions.

Storify but don’t get carried away My research process revolves around stories and storifying to a large degree. Like most researchers, I enjoy sharing my experiences from the field with colleagues, friends and family. I also try to draw the attention of like-minded people to my work and test whether my stories are convincing. Bit by bit, a larger narrative evolves out of small stories, and sharing is often an excellent test of whether it will carry over into a paper or even a book—this is probably a good moment to thank all my voluntary, but—most of all—my involuntary listeners over many years for patiently listening to the stories about Japanese gardens which formed the core of my book Spaces in translation: Japanese gardens and the West (Tagsold 2017). Thus, one of my most important pieces of advice to students who are preparing to write a thesis is to go to parties: when everybody is slightly drunk, try to find someone you like and tell them about your research. Try to get their full attention for ten minutes! This helps you gauge what interests potential readers, and it is a perfect way of finding out whether a thread connects the different pieces of the story, which might become chapters of a thesis later on. Without such a thread, you will quickly lose your audience at the party.

419

Christian Tagsold In addition to telling my stories and narratives to other people, I usually write them down at an early stage. My preferred piece of software for papers and books is Scrivener, a multifunctional programme for creative writing. Among other things, Scrivener has a corkboard where I can pin cards with summaries of different text segments. I use this corkboard to store my stories long before I actually start writing on my findings. When deadlines get closer, I set out with the introduction, which functions as a schedule and constant reminder of what the text will be about. I constantly rewrite this introduction as I move towards the finished text, but also keep in mind that it has to introduce the reader (me first of all!) to a good storyline which forms the backbone of the full text. The flexibility to change stories throughout the process of researching and writing and adapt them to new data is one of the most important skills in my version of storytelling. If I created a convincing story and stuck to it regardless of contradictory findings, I would wall myself in. However, staying flexible is very much in line with good storytelling. After all, good stories require unexpected turns and surprising twists. If the reader knew the end of the story beforehand, they would hardly indulge in reading it! As a consequence, I am usually very happy if new data conflicts with parts of the story already in my mind or written down in Scrivener. Such contradictions help to form new elements of tension that ultimately enrich the text. One of my recent research projects is a good example of this process. I wanted to find out more about gardens in Kyoto and Tokyo which were commissioned by politicians and the nouveau riche around 1900. Initially, these gardens were classified as eclectic by garden experts: not quite Western, but not Japanese either. Lately, however, many of the gardens have become listed as examples of national heritage for being important ‘Japanese gardens’. The story thus was about changing perceptions of gardens and incorporating them into the canon of national tradition. But my further research revealed that the classification of these gardens is still not fully resolved. Sometimes they are considered Japanese; sometimes they are still seen as eclectic. Yet these complex classifications are much more telling than a seamless shift from eclectic to Japanese. These blurred classifications ultimately allow for a richer story, which does not argue in black and white but exposes a whole spectrum of shades in between. The same is true for theory. Reading around the topic of one’s research is usually an ongoing process. We constantly discover books and papers related to our own projects. Some of them back up our stories, but the more exciting ones challenge them and ask us to move forward and adjust them. Often enough, I am most inspired by those texts which tell excellent stories themselves and set an example through their elegance. Bruno Latour’s Pandora’s hope (1999) is certainly one such book. His description of fieldwork in the tropical forests of Brazil and interaction with different researchers as they try to make sense of changes in the environment is a thrilling story. In the end, Latour’s theoretical explanation of this story formed the backbone of my book Spaces in translation. Latour’s pursuit of scientific studies hardly seems to have much in common with Japanese gardens. Nonetheless, his underlying theory made me understand the communication between Western and Japanese garden experts in a different light.

420

Chapter 15 How to finish

Conclusion One of the great side effects of thinking in stories and sharing them with others is that research and writing both get more enjoyable! Certainly, many stories you will encounter are not of the funny kind. At times, we run into dramas and tragedies. During my research into care homes for the elderly in Japan, not everything I saw was pleasant. However, some moments were truly heart-warming. One old woman, who had developed severe Alzheimer’s, called me a ‘star-grabber’ each time she saw me and never failed to explain that in her youth tall people were nicknamed that way—and I enjoyed her attention each time. Even the most burdensome fieldwork experiences become less depressing when shared with friends as a story. Often enough, theoretical issues only become clearer through the process of storifying. Talking to friends and colleagues will help straighten out the story. In addition, talking to different people helps you to avoid an obtuse style and accustoms you to writing for a broader public. This brings me to my final suggestion: writing about your topics in non-academic contexts can very much help to make your style more accessible. A couple of years ago I had the pleasure of authoring a popular book on Japan (Tagsold 2013) for a well-established book series on various countries. This experience did not directly translate into my more academic texts, but it helped me to reflect more on style and the expectations of my audience—and once again the power of stories as the book was largely built around anecdotes of my time in Japan, which I used as starting points for more general reflections. My next book Spaces in translation: Japanese gardens and the West gained very much from this as I started each chapter with a stroll through a garden and thus attracted my readers to my more abstract conclusions through a vivid story connected to the core topic of gardens. However, you do not have to start with a book. A blog or a short article in a popular magazine is a great start too!

421

15.3 Writing about Japan Richard J. Samuels

I have been asked to think aloud about how to write about Japan. It is hard to know where to begin. But I am reminded of the frequent (and always unsolicited) advice of the great China scholar and political scientist, Lucian Pye. When passing graduate students in the hallway, he invariably asked us: ‘Are you being brave, scholar?’ We were not sure how to respond, so we all smiled and walked on. But I have never forgotten that rhetorical question, and over the course of a long career writing and teaching about Japan, I have come to realise he was not joking. ‘Brave’ means many things, of course, but it appears as a choice in one form or another at every stage of the writing process. This starts with choosing a topic. It is easier to take on subjects that have already been validated by mentors and peers than it is to get out ahead of the discipline and outpace one’s teachers. Anyone ought to be able to review existing literature and identify a narrow niche in which to drill down for new information. But it is much more difficult—and riskier—to ask new questions and explore the previously unexplored. Great teachers should not mind. The same goes for scholars’ choices of methodology—the way they approach the chosen topic. There are plenty of ‘plug and play’ approaches to any topic, including Japanese Studies. But only the brave scholar will attempt to hack through the undergrowth of difficult problems with new tools. The risks are greater, and for some the rewards may not warrant taking them. But the brave scholar will discount the risks and tackle the problems head-on and creatively. And again, teachers should not only not mind, but respond enthusiastically. There are norms in every field, not least of all in Japanese Studies, where one is taught the importance of decorum. But the problem with excessive decorum is that one may end up writing for one’s mentors—and to ensure one’s own job security—rather than to generate new knowledge. This can constrain creativity. In my own case, this challenge was most painfully manifest when I asked my teacher (onshi) for an affiliation at his new research centre in Tokyo. He declined because my topic—one on which he had organised an impressive study group—was ‘too sensitive’. Then, when I asked for a copy of his group’s published report, he ignored the request. Since I had followed all the ‘rules’ of ‘Japanese Studies 101’, including regular greetings (aisatsu), gifting and accommodating him over many years, and since he was supposed to be an academic, I was shocked. In the event, I wrote the book I wanted to write and it won a prize. But I suffered for doing so, as I never met him again and was excluded from a range of relevant activities in Japan for at least a decade. Lucian Pye’s injunction to be brave was in my head throughout. Where do topics come from? In my experience, research projects have always had an organic quality to them. Small topics have taken me to larger, wholly unanticipated places. In one case, I was interested in trying to understand how Japanese firms could collaborate in applied re422

Chapter 15 How to finish search, a widespread practice in Japan in the 1970s–1980s that was contrary to what I was taught about market competition. U.S. textbooks on technology management described such behaviour as collusion. That short study of research collaboration became a book on technonationalism (Samuels 1994). In another case, tired of hearing second-hand (and unconvincing) explanations for why Japan’s Cabinet Legislation Bureau was so powerful, I tried to understand what it actually was and what it actually did. This grew into a book on Japanese national security and grand strategy (Samuels 2007). My current example is research I undertook in order to understand Japan’s highly controversial State Secrets Law in 2014. Before I knew it, I had stumbled into a book project on the Japanese intelligence community (Samuels 2019). But I must confess that there were times when I did not ‘listen’ to my projects, times when I was insufficiently curious or, I might say, excessively polite. This was the case in a study I did on Japanese energy markets (Samuels 1987). In the chapter on nuclear power, I neglected to ask the hard questions about Japan’s possible interest in nuclear weapons. I suspect I was too wed to the conventional wisdom about a ‘nuclear allergy’ and to the idea peddled by diplomats and politicians that a nation that had suffered a nuclear attack would never arm itself with nuclear weapons. Now that the archives are open a crack—and now that Japan is confronting the possibility (and potential consequences) of a leaky U.S. nuclear umbrella—that idea seems quaint. I finally scratched that itch 30 years later, but still wonder why so little attention has been paid to the topic. Insufficient bravery and excessive decorum, perhaps. Each of these projects—and each of the others I have tackled over the years—try to place contemporary Japan in comparative and historical perspective. After all, without a historical and comparative context, how can one establish a clear and comprehensive portrait of any country? How does one know what questions to ask? As my Japanese friends never tire of reminding me, Japan is unique. But, since every country is unique in its own way, I have always found this a tiresome and trivial claim. (For the record, I find it just as exasperating when the same claim is made about the United States.) As someone determined to make Japan comprehensible to the rest of the world without resorting to essentialist (and tautological) claims, it has always seemed more important to show how and why Japan is different. Not surprisingly, in that process one learns that Japan may be differently different than other countries are, and that in the areas I have cared most about—politics and public policy—Japan exhibits many of the same strengths and weakness as the rest of the world. One thing is certain: Japan is more comparable in generic terms than we have been taught. I first realised this in a project I conducted regarding political leadership (Samuels 2000). A historical comparison with Italy revealed that Tokyo did not have any greater ‘leadership deficit’ and did not suffer from a bureaucratism that was any more insidious than Rome’s. In fact, each had to accommodate to prewar authoritarianism and to the same postwar U.S. agenda, and they did so in parallel ways. Likewise, I learned in a project on national grand strategy that, contrary to what we are often told, Japan has never lacked grand strategists. And in a study of the 2011 triple disaster in Tohoku, it helped to compare the national discourse of failure in Japan to the one after Katrina in the United States in 2005 and to the one in Sichuan in 2008 (Samuels 2013). These histories and these comparisons have always been the most reliable guides to know what questions to ask about Japan—at least for me. As I figure it, if one does not know what questions to ask, any answer will suffice. I have long had the sense that effective scholarship on Japan depends on how the scholar engages with—and knows—what is truly sui generis and what is merely a variation on a widely shared problem. 423

Richard J. Samuels Other contributors to this volume have spoken on the issue of how to collect data, and I have addressed briefly how I try to know what data to collect. But this raises a next order—and equally critical—question: How do we make sense of the data we collect? Must we be brave here too? The short answer is ‘Yes.’ The longer answer is ‘Absolutely.’ It takes guts to ignore data that one has collected often at great personal cost in terms of time and energy. I have found that tossing materials that are original to my work is equal to the most difficult decision I have to make, and I have discovered that when I am writing I have had to make those decisions many times each day. Outlines are useful, but can only take the author just so far. The true organisation of collected data only emerges when one sits down to write. This is when data prove their worth—or their irrelevance. One must be undaunted by the fact that much of what one has collected inevitably falls, as filmmakers say, to ‘the cutting room floor’. It takes guts and is always painful to let material go. My colleagues and students are always intrigued by how old-fashioned I am regarding the organisation of what I learn. And, I am indeed ‘old school’; I use a system I learned in high school, at a time when we used slide rules, rather than computers. (I mean it when I say ‘old’; I wrote my first book on a portable typewriter.) Still today, rather than store everything on a computer, I do what I have always done: transcribe facts and interview notes on what we in the States call ‘3 x 5 note cards’. I make no claim that my system is as efficient as Niklas Luhmann’s famous ‘Zettelkasten’, but it has worked for me. Certainly one size does not fit all. No two scholars approach their analysis the same way. I urge my students to be pragmatic, to use what works and perfect it in their own way. But, for me, it has always been the note card. I would not claim that index cards produce a manuscript or enhance one’s bravery quotient, but in my experience, they do help produce a first draft by making it easier to sort and toss. There are several reasons why this is so. One benefit of the note card system is that it is nearly impossible to fill any single card with more than a single idea or set of data. This means that once it is used to transfer ideas or facts from card to a screen, it can be filed (and even forgotten) behind a divider labelled ‘used’. It can also be filed under an omnibus divider labelled ‘unused’—the author’s cutting room floor. Another benefit is that note cards can be numbered and cross-referenced. The genius of the system is not its user, of course, but its utility. In particular, I have found note cards to be amenable to migration in two ways: The first is when cards that are at first filed behind a divider that is labelled ‘Japan’ are re-sorted behind a divider labelled ‘Japan— Prewar’, and then, as the data inevitably grow, are moved to ‘Japan—Prewar Military Strategy’, before being further subdivided into ‘Prewar Strategy—Imperial Army and Imperial Navy’, and so forth. Collected and organised this way, note cards are almost amoebic. They have a way of not only forming themselves into new topics and subtopics, but into whole paragraphs as well. The second aspect of the mobility of cards is that they are portable. Once these amoebae have formed a critical mass of ideas/facts/data—and only an author can judge when that occurs— they can travel with the author far from the field or their home institution. It is awkward to say so, but there are benefits to being able to write at a distance from one’s subjects, colleagues and students. One can be as fond of all of them as I am, but still must confess that being in their midst is suboptimal for communing with one’s note cards. Everyone has their own system, and most of us can benefit from learning more about those developed and deployed by other scholars. Indeed, here is where Professor Pye comes back in424

Chapter 15 How to finish to the picture for me. I do not recall if he used note cards or not—I only remember his long yellow legal pads (notepad)—but he was astonishingly prolific and always willing to explain his productivity to graduate students. The secret, he revealed, had two steps. First, he would block out several hours every day for writing. Usually, that was at his desk at home. No telephone calls were allowed to intrude and, of course, email was not a problem in his day. And second, he made sure never to leave his desk without writing the first sentence of the next paragraph of whatever manuscript he was wrestling with at the time. That way, he pointed out, he never had to spin his wheels trying to find traction during the rare and dear moments he was free to compose text. Of course, none of this applies to an indifferent or craven scholar. At the end of the day, motivation is the equal of bravery as an important ingredient. After all, one must awaken each morning eager to continue wrestling with ideas and generating text. Often—too often, perhaps —this is not much fun. But unless it is always stimulating, it is not worth doing. Nor does any of this apply to a scholar who is content to till well-furrowed rows. To be an effective scholar in any field, one should be a bit of a ‘Contrary Mary’, my father’s favourite term for me. As in any other discipline or Area Study, in the case of Japanese Studies, this involves reflexively questioning what we think we know and what we have been told is true. After all, ‘conventional wisdom’ is conventional for any number of reasons, only one of which may be because it is true. Keep in mind that like bureaucrats, politicians, activists and everyone else, scholars have an agenda. Interrogate it, but not just for the sake of interrogation. A healthy and civil commitment to questioning what we think we know is the only sure way to improve upon what we actually do know. This applies not just to Japan, but to any other topic of study, and ought to lead us to generate better knowledge. This is, after all, the business to which scholars must commit—and serially recommit—themselves.

425

Further reading General Elbow, Peter (1998): Writing with power: Techniques for mastering the writing process. Oxford: Oxford University Press. Fernandes, Sujatha (2017): Curated stories: The uses and misuses of storytelling. New York, NY: Oxford University Press. For PhD students Bolker, Joan (1998): Writing your dissertation in fifteen minutes a day: A guide to starting, revising, and finishing your doctoral thesis. New York, NY: H. Holt. Lewin, Beverly A. (2010): Writing readable research. A guide for students of Social Science. London: Equinox. For new faculty members Boice, Robert (2000): Advice for new faculty members: Nihil nimus. Boston, MA: Allyn and Bacon. Caro, Sarah (2009): How to publish your PhD: A practical guide for the Humanities and Social Sciences. Los Angeles, CA: Sage. Germano, William P. (2013): From dissertation to book. Chicago, IL: The University of Chicago Press. Kelsky, Karen (2015): The professor is in: The essential guide to turning your Ph.D. into a job. New York, NY: Three Rivers Press.

References Abu-Lughod, Lila (1990): The romance of resistance: Tracing transformations of power through Bedouin women. In: American Ethnologist 17, No. 1, pp. 41–55. Bestor, Theodore C. (1989): Neighborhood Tokyo. Stanford, CA: Stanford University Press. Boice, Robert (2000): Advice for new faculty members: Nihil nimus. Boston, MA: Allyn and Bacon. Burawoy, Michael (2009): The extended case method: Four countries, four decades, four great transformations, and one theoretical tradition. Berkeley, CA: University of California Press. Burke, Kenneth (1941): The philosophy of literary form: Studies in symbolic action. Baton Rouge, LA: Louisiana State University Press. Butler, Judith (2006): Gender trouble: Feminism and the subversion of identity. New York, NY: Routledge. Clifford, James/Marcus, George E. (eds.) (1986): Writing culture: The poetics and politics of Ethnography. Berkeley, CA: University of California Press. Cresswell, Tim (1996): In place/out of place: Geography, ideology, and transgression. Minneapolis, MN: University of Minnesota Press. de Certeau, Michel (2011): The practice of everyday life. Berkeley, CA: University of California Press. du Sautoy, Marcus (2015): How mathematicians are storytellers and numbers are the characters. https://w ww.theguardian.com/books/2015/jan/23/mathematicians-storytellers-numbers-characters-marcus-du-sa utoy, [Accessed 19 June 2019]. Faier, Lieba (2009): Intimate encounters: Filipina women and the remaking of rural Japan. Berkeley, CA: University of California Press. Foucault, Michel (1995): Discipline and punish. The birth of the prison. New York, NY: Vintage Books. (Original work published 1975) Harvey, David (1989): The condition of postmodernity: An enquiry into the origins of cultural change. Cambridge, MA: Blackwell. Kimura, Aya Hirata (2016): Radiation brain moms and citizen scientists: The gender politics of food contamination after Fukushima. Durham, NC: Duke University Press. Kimura, Aya H./Kinchy, Abby J. (2019): Science by the people: Participation, power, and the politics of environmental knowledge. New Brunswick, NJ: Rutgers University Press. Kondo, Dorinne K. (1990): Crafting selves: Power, gender, and discourses of identity in a Japanese workplace. Chicago, IL: University of Chicago Press. Latour, Bruno (1999): Pandora’s hope: Essays on the reality of Science Studies. Cambridge, MA: Harvard University Press.

426

Chapter 15 How to finish Lofland, John/Snow, David/Anderson, Leon/Lofland, Lyn H. (2006): Analyzing social settings: A guide to qualitative observation and analysis. Belmont, CA: Wadsworth: Cengage Learning. Luker, Kristin (2010): Salsa dancing into the Social Sciences: Research in an age of info-glut. Cambridge, MA: Harvard University Press. Massey, Doreen (1994): Space, place, and gender. Cambridge: Polity Press. McMorran, Chris (2015): Mobilities amid the production of fixities: Labor in a Japanese inn. In: Mobilities 10, No. 1, pp. 83–99. Moxley, Joseph Michael (1992): Publish, don’t perish: The scholar's guide to academic writing and publishing. Westport, CT: Greenwood Press. Murakami, Haruki (2008): What I talk about when I talk about running: A memoir. New York, NY: Alfred A. Knopf. Olson, Randy (2015): Houston, we have a narrative: Why science needs story. Chicago, IL: The University of Chicago Press. Ortner, Sherry B. (1995): Resistance and the problem of ethnographic refusal. In: Comparative Studies in Society and History 37, No. 1, pp. 173–193. Peake, Linda/Mullings, Beverley/Parizeau, Kate/Thornburg, Gina K./Magee, Jon/Metzel, Deborah/ Wadhwa, Vandana/England, Kim/Worth, Nancy/Mountz, Alison/Finlay, Jessica/Hawkins, Blake/ Pulsipher, Lydia (2018): Mental health and well-being in Geography: Creating a healthy discipline: Report of the AAG task force on mental health (2015–2018). American Association of Geographers (AAG) Task Force on Mental Health. DOI: 10.13140/RG.2.2.14123.28960. Rosenberger, Nancy (2013): Dilemmas of adulthood: Japanese women and the nuances of long-term resistance. Honolulu, HI: University of Hawai‘i Press. Samuels, Richard J. (1987): The business of the Japanese state: Energy markets in comparative and historical perspective. Ithaca, NY: Cornell University Press. Samuels, Richard J. (1994): ‘Rich nation, strong army:’ National security and the technological transformation of Japan. Ithaca, NY: Cornell University Press. Samuels, Richard J. (2000): Machiavelli’s children: Leaders and their legacies in Italy and Japan. Ithaca, NY: Cornell University Press. Samuels, Richard J. (2007): Securing Japan: Tokyo’s grand strategy and the future of East Asia. Ithaca, NY: Cornell University Press. Samuels, Richard J. (2013): 3.11: Disaster and change in Japan. Ithaca, NY: Cornell University Press. Samuels, Richard J. (2019): Special duty: A history of the Japanese intelligence community. Ithaca, NY: Cornell University Press. Sheller, Mimi/Urry, John (2006): The new mobilities paradigm. In: Environment and Planning A 38, No. 2, pp. 207–226. Tagsold, Christian (2013): Japan: Ein Länderporträt. Berlin: Ch. Links. Tagsold, Christian (2017): Spaces in translation: Japanese gardens and the West. Philadelphia, PA: University of Pennsylvania Press. Toh, Jia Han (2019): National internationalization [Honours Thesis]. Singapore: National University of Singapore, Department of Japanese Studies.

427

Chapter 16 How to conduct reliable and fair research: Good research practice Cornelia Reiher and Cosima Wagner

1. Introduction Throughout the research process researchers need to employ good research practice, academic integrity and research ethics. These can include properly quoting sources, ensuring fairness and respect towards the work and findings of colleagues or protecting the privacy of interviewees. Ethical considerations related to scholarship on Japan have only been discussed infrequently within the international scholarly community (Bestor et al. 2003; Reiher 2018; Robertson 2007), but the March 2011 tsunami, earthquake and nuclear disaster in Japan have somewhat spurred discussions about ethics, the perceptions of the role of researchers and their responsibility towards those being studied (Gill 2014; Numazaki 2012; Yamashita 2012). But reliable and fair research encompasses more than just fieldwork ethics, and we must adhere to the fundamental rules of good research practice in general, no matter what kind of research we conduct. In this chapter we will provide an overview of these rules used by academic institutions and funding organisations worldwide. We think that ethical considerations are particularly important for Japan researchers because of the multiple translation processes throughout the research process, including the translation of legal, technological and social norms during fieldwork or literally translating scholarly literature written in Japanese. Therefore, this chapter aims at introducing readers to guidelines for good research practice, raising awareness about research ethics and encouraging researchers and students to critically reflect on their own research practice(s). After discussing what constitutes reliable and fair research in general, we will provide an overview of good research practice in the study of Japan throughout the research process and particularly in three different stages: 1. research design and data collection, 2. data management and 3. the writing process. The chapter closes with a discussion of open scholarship in the Japanese Studies community. In order to break down the sometimes abstract rules, we will discuss examples from research on Japan, drawing on our own experiences and the three essays accompanying this chapter (see this chapter, Gagné, Ch. 16.1; Gerteis, Ch. 16.3; Slater et al., Ch. 16.2). While advanced scholars will be familiar with some parts of the chapter, we address all scholars using sources from Japan or conducting research in/on Japan. The chapter can be used as an introduction to research integrity by lecturers teaching undergraduate classes or when planning your own PhD or research project.

428

Chapter 16 How to conduct reliable and fair research

2. What constitutes reliable and fair research? Research is a social activity and involves numerous people, such as colleagues, students, research participants or funding organisations. It is often characterised by unequal power relations in a field of complex responsibilities. Ethical challenges lurk in almost every stage of the research process, and it is necessary to make—sometimes difficult—choices and constantly assess their consequences (AAA 2012, pp. 2–3). Although most people would agree that research should rest on international and interdisciplinary principles (DFG 2013, p. 67), it is not always easy to define and to follow these principles to ensure reliable and fair research. They differ from country to country and among organisations and disciplines. Things get even more complicated when, as is the case in the study of Japan, research is conducted transnationally, possibly in international teams from several universities located in different countries. By now, most universities, funding organisations and academic associations have adopted ethical guidelines or principles. For example, All European Academies (ALLEA 2017), the Japan Society for the Promotion of Science (JSPS 2019) and the Ministry of Education, Culture, Sports, Science and Technology (MEXT) (2014) in Japan have decided on standards for research integrity, rules of good scientific practice and measures against misconduct. Most universities around the world have established their own ethics guidelines. Universities and funding organisations often have ethics committees and ethical review boards—or institutional review boards (IRB), as they are called in the U.S.—to which researchers have to submit their research proposal for approval in order to undertake fieldwork. This is common practice in the U.S. and is becoming increasingly common in European countries and in Japan. Despite the plurality of guidelines for and definitions of fair and reliable research, there are some values that are universally shared across national borders and disciplines (DFG 2013, p. 67). They rest on an understanding of research as ‘the quest for knowledge obtained through systematic study and thinking, observation and experimentation’ that aims ‘to increase our understanding of ourselves and the world in which we live’ (ALLEA 2017, p. 3). Because research is foremost about creating knowledge and not about individual researchers and their careers, all researchers have a responsibility to comply with the principles of research integrity. The European Code of Conduct for Research Integrity (ECCRI), for example, introduces the following fundamental principles of good research practice. Principles of good research practices ‘Reliability in ensuring the quality of research, reflected in the design, the methodology, the analysis and the use of resources Honesty in developing, undertaking, reviewing, reporting and communicating research in a transparent, fair, full and unbiased way Respect for colleagues, research participants, society, ecosystems, cultural heritage and the environment Accountability for the research from idea to publication, for its management and organisation, for training, supervision and mentoring, and for its wider impacts’ (ALLEA 2017, p. 4)

429

Cornelia Reiher and Cosima Wagner Ethical guidelines and principles can be helpful in making informed decisions in difficult situations, sensitising researchers to ethical issues and encouraging students and researchers to reflect and be critical. This is particularly important when using digital methods and the Internet1 for data collection on human subjects. Thus, students and researchers should familiarise themselves with their own institution’s as well as their host institution’s and their discipline’s guidelines and procedures for good research practice (O’Reilly 2012, p. 62).2 But why should students care about these ethical guidelines? Because the guidelines address not only fieldwork ethics, but also how to present other researchers’ findings in a term paper or dissertation in a fair and transparent manner. Therefore, it is also the responsibility of university teachers to convey these principles to their students and young researchers (DFG 2013, p. 67). Why is it important to follow these rules? Research misconduct can have serious consequences for research participants, researchers, a researcher’s institution and sometimes the discipline or academia as a whole. Falsification of data is prominently and frequently featured in international media. In Japan, for example, a researcher from Nobel laureate Yamanaka Shinya’s stem cell research team was found to have published falsified data in 2018 (Reuters 2018). In Germany, PhD plagiarism scandals have ended the careers of several high-ranking politicians. In times of increasing mistrust in research and decreasing budgets for Social Science in particular, incidents like this shake people’s trust in research and harm academia as a whole. The consequences for violators of good research practice differ among institutions. Students caught plagiarising may fail their exams or be expelled, researchers can lose their jobs or be stripped of their qualifications long after being awarded them. Nonetheless, the harm caused to others by misconduct is even worse. The American Anthropological Association’s Principles of professional responsibility (AAA 2012; see also Tagsold/Ullmann, Ch. 8) provides more hands-on rules that help researchers to stay honest and true to the fundamental principles of good research practice.

1. 2. 3. 4. 5. 6. 7.

American Anthropological Association’s principles of professional responsibility ‘Do no harm. Be open and honest regarding your work. Obtain informed consent and necessary permissions. Weigh competing ethical obligations due collaborators and affected parties. Make your results accessible. Protect and preserve your records. Maintain respectful and ethical professional relationships.’ (AAA 2012)

These rules apply to all steps in the research process from identifying a topic to publishing your work, and we will explain the ethical considerations involved in each step below.

1 For more information on ethics in Internet research, see for example AoIR (2019) and NESH (2019). 2 There are great differences between countries and institutions; thus readers should check the ethics guidelines of their own country, university, scholarly association or funding body.

430

Chapter 16 How to conduct reliable and fair research

3. Good practice during the research process Doing research is a process of many different steps that are often carried out in a circular rather than a linear manner. Research often begins with a research question and the designing of a research project and continues with the collection and analysis of information and data, eventually culminating in the researcher reporting the results. Acting in a fair and reliable manner throughout all these steps constitutes ‘scientific integrity’, which makes research ‘trustworthy’ (DFG 2019, p. 7). The freedom of research guaranteed by most countries’ constitutions comes with responsibilities for every researcher, who should take them into full account (ibid.). The following sections look in more detail at the challenges all researchers face when trying to adhere to the standards of good academic practice during research. We discuss examples of scholarly misconduct, such as plagiarism and abusing the trust of colleagues and partners, but also give positive examples of good research practice.

3.1 Research design and data collection Ethical guidelines are just that—guidelines, and there is no one-size-fits-all solution. Therefore, researchers have to solve the various ethical problems they encounter on a case-by-case basis (O’Reilly 2012, p. 62). This begins when defining a research topic and question and continues when collecting data, particularly when this happens via fieldwork (see McLaughlin, Ch. 6). When beginning a project, students especially wonder if they can work on a research topic others have already written about or are still working on. The answer is ‘yes’, but the first step is to review the scholarly literature on what has already been done in your respective field (see Zachmann, Ch. 4) in order to decide what you want to and actually can contribute to ongoing debates. It is the responsibility of every researcher to explain the selection of research topics, question, designs and cases honestly and in a transparent manner and to respect their colleagues’ work (see Gerteis, this chapter, Ch. 16.3). For example, asking similar questions or analysing issues from a similar perspective as other scholars is okay as long as you reference their work and your research produces new insights. This might be the case when you ask the same questions to analyse a different time period or different empirical materials. When collecting data, researchers are confronted with more ethical challenges. Social scientists collect data about people and organisations through surveys, interviews, experiments and observations. But because ‘people are not your data’ (see McLaughlin, Ch. 6) ethical standards and legal norms must be observed to protect their rights and autonomy when collecting personal data. Researchers must obtain permission to collect their data and grant them the right to anonymity (GESIS 2018, pp. 1–2). When collecting data through fieldwork in or outside Japan, researchers should reflect on ethical challenges and obtain approval from their institutions’ ethical review board, if required (Bestor et al. 2003, p. 13). Although the rules of good research practice claim universality, there are some specifics to the Japanese setting where trust is very important. Theodore Bestor et al. (2003, p. 14) explain in detail how borrowing trust from a host institution or people willing to act as ‘social guarantors’ creates sensitive social relationships affected by your behaviour in the field. This is true for your informants, yourself and the person who made the introduction, since the latter is the 431

Cornelia Reiher and Cosima Wagner one research subjects will complain to if difficulties arise. Not considering the importance of trust relations can cause harm in various ways. Imagine, for example, you introduce a colleague to several of your research participants from Japan during a workshop. The colleague later visits your informants to conduct research similar to your research in the same places without telling you but pretends to the informants she came based on your introduction. Later you receive emails from your research partners, who complain about your colleague’s behaviour, alleging she had not shown up for appointments or had asked inappropriate questions during interviews. It will take you quite some time to restore their trust and to sort things out. This example shows how the behaviour of one researcher in the field can harm the trust relations between research participants and another researcher, whom they considered to guarantee the trustworthiness of the other researcher. You should always keep this in mind when being introduced to research participants through others. However, this example also raises another question: Is it okay to conduct research at another scholar’s field site? Theoretically, yes, because scholars do not own a field site, but it is important to talk to the colleagues who are already conducting research there, especially if they are working on a similar topic. Ignoring their work or behaving in a way that discredits them is a serious violation of good research practice. You need to mention previous research. Just imagine anthropologists conducting research in Suye Mura or Onta without mentioning the work done by John Embree (1946) or Brian Moeran (1997). Collaboration is always best, but it can be difficult at times, for example, when senior researchers ask you to share your informants’ contact details. This reflects how unequal power relations can affect fieldwork ethics. In such a case, the protection of your research partners’ privacy and the hierarchies within academia can conflict with each other, and it is important to talk to supervisors, colleagues or an ombudsman to solve ethical problems that may arise. Trust is not only important for gaining access to a field site, but throughout fieldwork. This is particularly true for anthropologists who live with people and develop intimate and reciprocal relations with their research partners. Trust, vulnerability, and reciprocity are often closely linked. Researchers might try to gain their interlocutors’ trust by sharing their own experiences, but this makes researchers themselves vulnerable (Alexy/Cook 2019, p. 236). This vulnerability can be emotional, when listening to disturbing stories, or sometimes physical, when entering dangerous situations (ibid, pp. 246–247). The vulnerability of the people we conduct research with is just as important. They might be vulnerable because they are poor, discriminated against or have been traumatised by a disaster (Gerster 2018). Research on minors comes with particular ethical issues as Kathryn E. Goldfarb (2019, p. 252) explains, drawing on her research about foster care, adoption and life within Japanese child welfare institutions. Although she had received clearance to interview children and even prepared a child-friendly oral informed consent form, she realised that ‘informed consent’ in this context was ‘meaningless and misleading’, because the ‘children were already documented and traced in so many ways they could not control’ and she decided not to write about them in a language they did not know (ibid.). This example shows that even when informed consent is obtained, ethical questions remain. Yet, most guidelines for good research practice strongly recommend seeking written consent

432

Chapter 16 How to conduct reliable and fair research from research participants (see Slater et al., this chapter, Ch. 16.2).3 In Japan, written contracts cannot always be used because they ‘would call into question the researcher’s cultural understanding and trustworthiness’ (Bestor et al. 2003, p. 14). While introductions and the trust they create might still suffice in some research contexts in Japan, many universities or funding organisations require written consent. No matter how research participants’ consent is documented, it is necessary to obtain your research participants’ permission to conduct research. There are cases where gaining the full consent of the research participants is not possible and they might not know that they are being observed. This is called covert research. Ideally research is conducted openly—overtly—and all participants know about the researcher’s identity. Social Science associations and funding organisations advise against covert research because it can harm research participants through ‘deception, dishonesty, invasion of privacy and lack of consent’ (O’Reilly 2012, pp. 64–65). But some scholars call for ‘situated ethics’ in the field, because there are cases in which information can only be collected through covert research (Calvey 2008). In reality, the distinction between covert and overt research is blurred and depends on the type of research and its goal (see Yoshida, Ch. 5). Using social media during the research process, for example to gain access to the field, to learn about research topics and important actors in your field site or to keep in touch with your informants after fieldwork ends raises new ethical problems (AoIR 2019). Even if you don’t maintain an online presence to stay in touch after leaving Japan (see McLaughlin, Ch. 6), the importance of protecting your own privacy and that of your interlocutors, and critically reflecting on fieldwork ethics for online research is indisputable. Levi McLaughlin (see Ch. 6) gives valuable advice on how to protect interlocutors’ privacy online. Julia Gerster (2018) reveals how in her research on social relations in municipalities in the Miyagi and Fukushima prefectures after the March 2011 triple disaster, online and offline worlds merged and affected social relations within the communities she studied and between her and her research participants. When one is researching controversial topics, openly sharing photographs or the names of informants online can cause problems for researchers and informants alike. Through social media, it might be easier to find out who talked to whom about what, but ‘researchers are less in control of how information is shared and who will be able to access it’ (Gerster 2018, p. 30). While recommendations by ethics committees help to address some of these issues, many problems related to social media are not yet covered by such guidelines. Online research and staying in touch with informants via social media pose new challenges in terms of the protection of interlocutors’ privacy and their anonymisation, but anonymisation was not particularly easy before the emergence of the Internet. While it is important to ensure informants’ anonymity in order to encourage them to speak frankly, during her research in rural Japan, Cornelia Reiher experienced that anonymisation can be tricky when everybody knows each other (Reiher 2014; Reiher 2020). Sometimes, even pseudonyms do not ensure anonymity, and researchers have to think about additional ways to protect their research participants from personal or professional harm in their local community. This can include additionally anonymising the name of places or organisations. But such decisions have to be made carefully, because too much anonymisation limits the explanatory power of research results. 3 Examples of informed consent forms in Japanese are provided in Japanese introductions to Social Science research and fieldwork (Nishiyama et al. 2015, pp. 133–135).

433

Cornelia Reiher and Cosima Wagner

3.2 Data practices and management During the data collection process and once you are back from the field, you face the challenge of storing and managing your data following ethical principles. These are particularly important against the backdrop of the digital transformation in scholarship and with the Internet as an ‘integral tool for scholars across disciplines’ (Mai/Repnikova 2018, p. 2). Here, the responsibilities for students and researchers alike start with storing all your research files and data according to your institution’s basic IT-security guidelines to ensure their confidentiality, integrity and availability. Research data is all ‘evidence used to inform or support research conclusions’ (CESSDA Training Team 2017–2020), for example, transcribed observations, audio or video files, written text, software or photographs (University of Sheffield n.d.). Good research practice with regard to data management in the digital age means ensuring that, despite the short life span of hardware and software and research project websites, the sources your published findings are based on stay comprehensible and verifiable in the long run. For a term paper this might mean saving a digital copy of a cited website, while for a dissertation or a research project where empirical research is conducted, more profound and elaborate data management has to be applied to the questionnaire, audio files, anonymised transcripts, data coding tables or digitised manuscripts/objects. If you are applying for funding for a research project from a research funding organisation like the European Research Council (ERC), for example, you are requested to prepare a data management plan (DMP) for your project. This plan should address how you will ‘handle, organise, and structure your research data throughout the research process’ (ERC Scientific Council 2019, p. 3).

• •

•

Research data management (RDM) begins with your initial considerations regarding what will be necessary for using or collecting your particular type of data; includes measures for maintaining the integrity of the data, making sure that they are not lost due to technical mishaps, and that the right people can access the data at the appropriate time; looks toward the future, making it clear that you should provide detailed and structured documentation to be able to share your data with other colleagues and prepare the data for long-term availability (CESSDA Training Team 2017–2020).

The ‘FAIR Guiding principles for scientific data management and stewardship’ (Wilkinson et al. 2016) have become internationally acknowledged guidelines for making research data FAIR (F-indable, A-ccessible, I-nteroperable and R-eusable). This also applies to the Social Sciences, where funding agencies, journals, academic associations and institutions are now advocating data transparency and preservation through discipline-specific data management agreements and best practice (see Arrington, Ch. 13). If you are doing empirical research for your PhD or planning an empirical research project, look for research data management (RDM) policies or ongoing discussions in your discipline4 and your university. RDM consulting services might be available at the university library (Purdue University Library 2019), your local eResearch support or your university’s central research administration department. 4 For Anthropology, see Pels et al. (2018), Imeri (2018, 2019) and Huber (2019).

434

Chapter 16 How to conduct reliable and fair research In Japan, the introduction of research data management (kenkyū dēta kanri) as a means of good research practice was initiated by the Cabinet Office in 2013 and is closely connected to the open science movement (see below). Social Science research centres like the Centre for Social Research and Data Archives (CSRDA) at the Institute of Social Science at Tokyo University and its Social Science Japan Data Archive (SSJDA) have long established their best practice recommendations for the use of their data. In order to enable a joint national infrastructure for data management and data sharing, the National Institute of Informatics (NII), as the central research infrastructure provider in Japan, has recently taken the lead and developed a technical framework called ‘GakuNin’ (see text box below), which all research institutions in Japan can use to foster RDM at their institution. Furthermore, a ‘Guideline for the establishment and operation of research data repositories’ (in Japanese; CAO 2019) has been published and together with the NII’s repository (cloud) platform software ‘WECO3’ constitutes the basis for the setting up of open access repositories at the majority of universities in Japan.5 Whether the availability of a technical infrastructure will lead to a new culture of research data management and sharing in all academic disciplines remains to be seen in Japan and elsewhere. Although a central hub for accessing social sciencerelated research data on Japan is still lacking, the chances are good that it will become easier for students and scholars to access research data from Japan to inform their own empirical studies on Japan (see Hommerich/Kottmann, Ch. 10).6 Resources on research data management Europe • ‘Data Management Expert Guide’ of the Consortium of European Social Science Data Archives: www.cessda.eu/Training/Training-Resources/Library/Data-Manage ment-Expert-Guide • German Data Forum: https://www.ratswd.de/en/content/start • GESIS Leibniz Institute for Social Sciences: https://www.gesis.org/en/research/resea rch-data-management • GO FAIR initiative (France, Germany and the Netherlands): www.go-fair.org/ • King’s College London Guide on ‘Managing Research Data’: https://www.kcl.ac.uk /researchsupport/managing

5 For more details on WECO3, see RCOS (2017b). As of April 1, 2020, the Japan Consortium for Open Access Repository (JPCOAR) (2020) lists 630 member universities on its website. 6 Rikkyo University’s Data Archive (RUDA) for social research provides a list of Social Science data archives in Japan (RUDA n.d).

435

Cornelia Reiher and Cosima Wagner U.S. • American Political Science Association’s DA-RT-initiative: https://politicalscienceno w.com/data-access-and-research-transparency-initiative-da-rt/ • Purdue University Library Subject Guide on Sensitive Research Data Management: https://guides.lib.purdue.edu/sensitivedata Japan • Tokyo University, Centre for Social Research and Data Archives (CRSDA): https://c srda.iss.u-tokyo.ac.jp • National Institute of Informatics (NII), Research Center for Open Science and Data Platform, GakuNin-Framework: https://rcos.nii.ac.jp/en/service/rdm As we write this chapter, the discussion on good RDM practices in the context of Area Studies and Japan-related research has only just begun. But three main issues are currently being critically debated: First, the debate about what long-term, preservable research data in qualitative research actually are—especially in anthropological research (Imeri 2018; Imeri 2019; Huber 2019)—is important when thinking about RDM. Because of the complexity of research data in the Humanities and Social Sciences (field notes, transcripts, coded data, interview recordings, software, etc.), there is no one-size-fits-all solution when storing data and their digital representation. However, organisations responsible for creating infrastructures for RDM might not be aware of this (Edmond et al. 2019). Therefore, researchers should communicate about the plurality of research data to funding organisations, universities and libraries. This will help them to co-develop solutions when it comes to good ‘digital’ research practice together with the researchers by taking into account that in Area Studies as well as in the Humanities and Social Sciences research in general, research data are socially produced in a specific context (Imeri 2018, p. 222; Knorr-Cetina 1988). Thus, when such data are stored, ‘adequate procedures and strategies for the documentation of contexts’ (Imeri 2018, p. 229) of these different materials are needed. Second, this plurality of research data is intimately linked to the question of the ‘reuse’ of qualitative research data in a methodologically sensible and ethically correct way. Many funding organisations now expect researchers to produce reusable data. But as noted above, because data—especially in qualitative research—are co-created by the researcher and the research participants in a specific place, time and constellation, it is difficult for other researchers to simply use them again outside the original context. In this respect, ethical questions of protecting research participants’ privacy and the aim of more openness and transparency by sharing data conflict with each other. This raises questions about whether reusing qualitative data is possible at all and what kind of informed consent would be necessary to cover the various ways of reusing the data, or whether the reuse of anthropological research data should be declined per se (Imeri 2018; Imeri 2019; Huber 2019). Would there be self-censorship by the researcher when she co-produces data with their reuse in mind? And how does software used for curating and analysing data change them?7 Questions like these increasingly concern Japan scholars as well and should be discussed further in order for them to make informed decisions when confronted with the issue of reuse.

7 For a critical self-assessment about using coding software in a PhD-project, see Müller (2019).

436

Chapter 16 How to conduct reliable and fair research Third, questions and criticism surround how the FAIR principles of Japan-related research data can be achieved in research infrastructure environments that are not prepared for the use of non-Latin scripts like Japanese. Research infrastructures for Area Studies in the Anglophone IT-sphere of higher education institutions and libraries in Europe and the U.S. often do not (sufficiently) support the use of multilingual data curation tools and the discovery of metadata in non-Latin scripts (Asef et al. 2019). This makes not only the discovery of and access to research data from countries like Japan difficult, but also searching for data sources in Japanese outside Japan (see Peucker/Schmidtpott/Wagner, Ch. 9). Furthermore, Area Studies researchers have also reported on practices of academic journals discriminating against the use of non-Anglophone references in papers and therefore making research in other languages invisible and undetectable. This reveals again the dominance of Anglophone research in the digital sphere. More lobbying for multilingualism in academia from Area Studies scholars to their home institutions, funding organisations and journal editors is necessary. Against this backdrop of emerging debates on RDM, it is important to be aware of its growing importance in order to protect your data, collaborators and research participants and to position yourself when confronted with demands to make your data reusable and to be ready to develop a data management plan for your next scholarship or research grant application.

3.3 Presenting research results: Things to remember before and while writing Once your data have been collected, safely stored and analysed, it is time to write down your results. In publishing, fair and reliable research means finding ‘the appropriate balance of different voices, and translating grounded data into theoretically engaged […] knowledge’ (Gagné, this chapter, Ch. 16.1). Thus, the first thing you should make sure of is to reference and cite other scholars’ publications about your research topic (see Zachmann, Ch. 4). This also includes selecting scholarly works and data that do not support your argument. It also means that you must not ignore academic work or data that have already addressed the issues you are discussing, possibly making your own work look less original. We understand research as a collective endeavour in which researchers acknowledge each other’s work and try to contribute to previous work by adding knowledge rather than focusing on (often artificially created) research gaps they are trying to fill, sometimes by ignoring work that has been done in the field. An easy way to realise this kind of scholarship is through referencing other scholars’ work. Although everybody knows that references are important, let’s reconsider their functions: References provide authority for cited arguments and enable readers to check the accuracy of and further information about the point cited. They can also point readers to additional sources and give credit to the original author of ideas, arguments or words (Language Log 2007). If you fail to provide adequate references for your sources, this is called plagiarism. Plagiarism is defined as ‘the unacknowledged borrowing of someone else’s idea or the unacknowledged borrowing of someone else’s words […] without giving proper credit to the original source’ (ALLEA 2017, p. 8). Because such behaviour violates the original authors’ rights to their intellectual outputs (ibid.), it does not matter whether this happens on purpose or by accident. It can result in the withdrawal of scholarships or funding (JSPS 2019, p. 2) or expulsion from your university. Yet, plagiarism happens more often than you might think.

437

Cornelia Reiher and Cosima Wagner Why would people plagiarise? Plagiarism might occur due to time pressure (a fast-approaching deadline, end of funding, pressure to raise funding, for example), problems during the research process (insufficient data, a lack of data analysis skills) or simply because people realise that research is actually a lot more work than they thought and they are not willing or able to devote that much energy and time to it. Sometimes, international journals are sceptical of research conducted in languages other than English, and researchers don’t want to jeopardise their chances of getting published, so they use the ideas of foreign colleagues anyway without citing them. However, plagiarism severely harms other researchers (emotionally and sometimes even their careers), relationships between researchers and the research itself. It damages the reputation of an institution or academia as a whole and can expose ‘research subjects, users, society or the environment to unnecessary harm’ (ALLEA 2017, p. 9). Thus, to avoid accidental plagiarism, we will highlight the different types of plagiarism and give advice on how to avoid them.8 Direct plagiarism You copy someone else’s work word for word without using quotation marks. You can avoid it by using quotation marks! Paraphrasing without a source Although you don’t copy other authors’ texts word for word, you use key ideas from different works, but don’t cite the sources of those ideas that are not your own. Accidental plagiarism is frequently caused by paraphrasing without a source but can be avoided by properly citing the sources (Streefkerk 2018). Sometimes, people paraphrase content they have read in Japanese (or other foreign languages) sources without citing the source, because plagiarism software cannot detect it easily. However, your instructors, supervisors and reviewers can, so don’t try it! Copy-and-paste plagiarism Similar to paraphrasing without citing a source, but through copying and pasting different texts together, you create a new text. This can include ‘rewording pieces of sourced material while keeping the structure of the original texts. This type of plagiarism requires a little more effort and is more insidious than simply paraphrasing a source’ (Streefkerk 2018). So why don’t you use this extra time to just properly do your work instead? Citation plagiarism Citation plagiarism refers to ‘the citation of a reference without acknowledging that it came from another source’ (Language Log 2007). This means that you must not pretend to have read all the literature somebody else has nicely summarised in her literature review, for example, without mentioning that you owe parts or all of it to this author. But you can easily solve this problem by reading the source yourself. If you really cannot obtain the original source because it is not available, cite like this: ([Author xyz] cited in [Author xyz] [year]: p. xx).

However, plagiarism is not only something every researcher should avoid, but also a phenomenon researchers can fall victim to. Plagiarism can happen to PhD students whose ideas 8 Self-plagiarism is another, yet controversial, type of plagiarism. The discussion about whether self-plagiarism is possible at all is ongoing. Self-plagiarism refers to a situation in which an author uses parts of her own work in another publication without citing the previous work. This can be particularly problematic for PhD students who don’t know whether they can include parts of already published material in their PhD thesis. When in doubt, check the regulations for obtaining a PhD at your university. In addition, the University of Glasgow (n.d.) provides a good overview of the types and possible consequences of self-plagiarism.

438

Chapter 16 How to conduct reliable and fair research appear in their supervisors’ publications without them properly acknowledging their origin, but PhD students can also plagiarise senior colleagues’ work. Imagine there is a junior colleague who works on similar topics and you send your publications to this colleague to discuss them. When you are asked to sit on this person’s PhD committee and read the thesis, you find out that your colleague has plagiarised your (and other scholars’) work in several ways (from the research question and research design to an unreferenced patchwork copy of at least three of your articles). Although you voice your concerns to the committee and to an ombudsperson for good research practice, nothing happens, and your junior colleague is awarded a PhD. Such an experience of being unsuccessful in protecting your rights, however, should not prevent you from reporting breaches of good research practice. If you become a victim of plagiarism, contact the ombudsperson of your university or funding organisation and/or your supervisor to ask for help. Don’t just let the perpetrator get away with it. This can lead to supervised mediation with the perpetrator or even a warning and will certainly affect the perpetrator’s career. Experiencing plagiarism can also be an important motivation to teach research ethics to undergraduate and graduate students and to write about it. How to prevent plagiarism? Organise your research process in a transparent way. Begin when you start reading and taking notes. Always add all the bibliographical information to every note you log on your computer. You can use software like Citavi or Zotero. Create a system that enables you to write down page numbers and bibliographical information and to distinguish other people’s ideas and arguments from your own. This will save you so much time when you write out your work (see Gerteis, this chapter, Ch. 16.3) and don’t have to look up every source you read again because you cannot remember where the information you are using came from. Give credit to other researchers whose ideas or results you are (re)using (Tóth-Czifra 2020a).

4 Closing remarks: Towards open Japan(ese) Studies Once you finish your paper or thesis, you might want to publish it. Among the many publication avenues available (see Farrer/Liu-Farrer, Ch. 17), open access books and journals as well as institutional repositories that aim to distribute research outputs online without access barriers have been increasing in number. These journals are part of a larger open science/open scholarship movement that is trying to liberate academic scholarship from paywalls built higher and higher through the market power of a few publishing houses which therefore became an increasing burden for university library budgets. At the same time, it aims at increasing the accountability and reliability of research, for example, by making research results, research data and software accessible to everybody. We want to close this chapter by briefly introducing the open science/open scholarship movement as food for thought about how it can nudge the Japanese Studies community and its research culture towards more openness and fairness and address some ethical challenges related to open scholarship.

439

Cornelia Reiher and Cosima Wagner Open science is being discussed in the international scientific world, including Europe and Japan (FOSTER portal; RCOS 2017a), and covers a wide range of topics ‘ranging from the democratic right to access publicly funded knowledge (e.g. open access to publications) or the demand for a better bridging of the divide between research and society (e.g. citizen science) to the development of freely available tools for collaboration (e.g., social media platforms for scientists)’ (Fecher/Friesike 2013, p. 1). In its most general sense, open science is about creating new commons of knowledge in the digital sphere (from sharing preprints of papers and open access publications of dissertations in institutional repositories to open peer review or more open research data and copyright licences), but so far mainly in the (Natural) Sciences. Discourse about open scholarship in the Humanities and Social Sciences started just recently (Knöchelmann 2019, pp. 1, 9–13). Critics, however, point to the limitations of the open science/open scholarship movement in changing academia into a community focusing on good arguments rather than on the ‘mad run rush for more publications, conferences and researchprojects’. They identified the latter as the ‘major force to prevent a more open research culture’ (Rosa 2010, p. 55). We can already access resources for research through public data sets and software from the Center for Open Data in the Humanities (CODH) or data archives like the SSJDA or GESIS. The opportunities the open science movement offers to researchers in Area Studies and Japanese Studies in particular to communicate outside traditional publication avenues are vital. For example, research blogs like Hypotheses (see below) seem to be a promising new medium of open communication in scholarly communities. But we also think that the potential of open scholarship, including open research data, should be carefully discussed with an eye on the characteristics of the Japan Studies community and its possible ethical challenges (see GIDA’s CARE principles in text box below). This includes thinking about how we want to collaborate with each other within and across national and disciplinary communities in the future and how we want to make research results more widely accessible. Against this backdrop, it is also important to ‘ensure that crucial decisions about knowledge creation and sharing such as to whom it will be accessible, for how long, how interactions take place, etc. remain indeed in the hands of those who create this knowledge’ (Tóth-Czifra 2020b) and not with proprietary platforms like ResearchGate or Academia.edu (ibid.).

• • • • •

Open science/open scholarship resources FOSTER portal (introductory e-learning platform for open science): https://www.fo steropenscience.eu Global Indigenous Data Alliance (GIDA): CARE Principles9 for Indigenous Data Governance: https://www.gida-global.org/care Hypotheses (Blog): https://hypotheses.org, i.e. https://dariahopen.hypotheses.org National Institute for Informatics (NII), Research Center for Open Science and Data Platform (RCOS): https://rcos.nii.ac.jp/en/ NII/The Institute of Statistical Mathematics: Center for Open Data in the Humanities (CODH) http://codh.rois.ac.jp

9 The CARE principles advocate researchers’ commitment to reciprocity towards research participants, especially when Indigenous Peoples’ rights and interests are concerned, and two-way open communication beyond academic circles. They call for C-ollective benefit, A-uthority to control, R-esponsibility and E-thics of research data.

440

Chapter 16 How to conduct reliable and fair research Open access publishing of a preprint copy of a paper or a dissertation in one’s institutional repository or writing blog posts are attractive to some, but sometimes conflict with the requirements in tenure procedures or traditional publisher’s copyright requirements. Therefore, more scholarly community awareness and action are needed, because openness could also foster transparency in tenure procedures and could encourage the promotion of junior scholars. These are only a few issues regarding openness in Social Science research in/on Japan and Japanese Studies that could and (as we believe) should be discussed. In summary, considering the diverse contexts in which researchers make the ethical decisions introduced above, we want to stress that—although ethical standards are guidelines that should be adjusted to the individual research context—there are some hard rules when it comes to preventing harm to research participants, your colleagues and yourself that you must follow. These include reliability, honesty, respect and accountability at all stages of the research process. When adhered to, these rules help to preserve academic freedom.

441

16.1 Fairness in research and publishing: The balancing act of cultural translation Isaac Gagné

What is fair and reliable research practice, and what is particular about it when studying Japan? While this is usually discussed in relation to institutional research and publishing ethics and regulations, this chapter explores two somewhat different perspectives through my experience as an anthropologist with religious groups, and my experience working with peer-reviewed Japanese Studies journals. ‘Fair and reliable research’ involves negotiating and striving for fairness in both research and publishing as part of the dynamic process of ‘cultural translation’. This requires recognising how our research is part of a (fieldwork) process where we enter into the local worlds of our informants, and extends to the writing process, which ‘is inevitably enmeshed in conditions of power—professional, national, international’ (Asad 1993, p. 190). Specifically, 1. in fieldwork, this means understanding the ‘local moral worlds’ and interpersonal dynamics in our respective field sites, and establishing fair relationships with informants. 2. In publishing, this means finding the appropriate balance of different voices, and translating grounded data into theoretically engaged (anthropological) knowledge.

Navigating ‘local moral worlds’ In Anthropology, practising fair research starts from the moment that you begin interacting with potential field sites, and especially the ways in which you negotiate with your field sites. This is the beginning of your enmeshment in the local networks of your informants—the ‘local moral worlds’—that consist of the values and stakes of individuals within their local networks of relationships. Arthur Kleinman (1992, pp. 129–130) defines ‘local moral worlds’ as the ‘the social psychological and moral processes’ that anchor individuals’ personal experiences, including the patterns of life, sociality, ethics and values, to their social worlds. As researchers, of course, we must follow the ‘institutional moral worlds’ of our funding agencies and research ethics boards, which are concerned with issues of transparency, liability and safety. Yet as fieldworkers working on Japan, where trust, long-term relationships and social context are of prime importance, we must also negotiate with and within the ‘local moral worlds’ of our informants (see Alexy, Ch. 7.3; Brumann, Ch. 7.1; Gagné, Ch. 6.1; Yoshida, Ch. 5). Thus, in addition to following institutional research ethics, it is equally important to recognise the local dynamics in our field sites and be able to adapt to local ethics in interacting with informants. In my fieldwork with members of ‘new religions’ (shin-shūkyō) and ‘self-cultivation organisations’ (shūyō-dantai) in Japan, I spent the first twelve months doing participant obser442

Chapter 16 How to conduct reliable and fair research vation with several different organisations, participating in daily, weekly and monthly activities, rituals and informal get-togethers (Gagné 2011; Gagné 2017). As my informants themselves reminded me, religious beliefs and spiritual practices cannot be grasped just through words, but they have to be experienced through practice. I was frequently told, ‘you don’t have to believe, just experience it.’ According to them, even without professing belief at first, if one subjects one’s body to the practice, one’s heart will eventually follow. This echoes observations that belief in Japan is based on orthopraxy, ‘right practice,’ rather than orthodoxy, ‘right doctrine’ (Shields 2010). Thus, to enter the local moral worlds of my informants I had to first subject myself to the same routines and daily experiences. Specifically, through spending time with other participants, waking up early for 5 a.m. meetings, taking long-distance pilgrimage bus trips to religious headquarters, participating in intensive overnight ‘training camps’ and kneeling (seiza) for hours on tatami mats during meetings, the local members and the leaders came to see me as part of their ordinary contexts and their local networks. This was especially important given that participation is gendered as well as private and discreet in Japan—roughly two-thirds of the active members are women, and many never disclose their participation to their own family. In addition, joining what informants called ‘difficult’ and ‘demoralising’ activities such as door-to-door solicitation not only diffused the burden but also blurred the distinctions between the members and me as a researcher. Consequently, by subjecting myself to the same experiences, not only did this help me to understand relationships and social dynamics in the field, but it also helped me to gain access to their private worlds of beliefs by immersing myself in their orthopraxy. While participating in rituals and meetings alongside other members was important for me to gain access to local moral worlds, religious participation also brings out irresoluble tensions between belief and money. This raises the question: How much are you expected to participate, and how far are you expected to be involved in rituals—especially their monetary entailments? In Japanese religious practice, belief and money are often intricately connected and reinforce each other. Thus, the money you spend also reflects your degree of respect for the religion and the other members, and this brings conflicting challenges for a fieldworker. While you must enfold yourself within the local moral worlds of their relationships, these consist of ‘contestations and compromises that actualise values both for collectives and for individuals’, and engaging in these contestations and compromises requires careful navigation and setting limits (Kleinman 1999, p. 71). In my fieldwork, it seemed unethical to ‘buy’ their trust by paying my way into their organisations. This required constant negotiation, as I had to explain my interest in understanding their beliefs while maintaining my position as an ‘outside’ researcher. Participation by payment might seem unethical to some; however, I found that fair participation in this context requires payment. Thus, it would have been contrary to my informants’ cultural logic for me to not pay anything to participate in their ritual practices and spaces. Indeed, members constantly solicited my payments, and I needed to set a limit, so I paid the standard fees for rituals as ‘entrance fees’. Yet, members frequently paid more than was expected in order to demonstrate and deepen their commitment and participation. Moreover, as members grew to know me, I found that some even paid voluntary donations in my name by proxy without consulting with me. Of course, members are aware that one cannot buy belief, and instead they pay as a sign of gratitude for their connection (go-en) with fellow members. In this sense, whether they are paying for themselves or for others, their money became a cur443

Isaac Gagné rency of sincerity which demonstrated their belief, gratitude and commitment to their peers and the organisation, while my own payments were likewise necessary to demonstrate my respect for their values.

Writing ethnographies and balancing voices Fair research practice extends beyond the field, as the next step in understanding local lifeworlds is to ‘translate’ them to a wider audience. When you are analysing, writing and publishing your findings, you are no longer dealing with just informants and local moral worlds, but you are dealing with very different interlocutors—informants, scholars and reviewers— and negotiating with how to translate across different moral worlds with very different stakes. This phase can be referred to as ‘cultural translation’ (Asad 1993): making local lifeworlds and cultural logic intelligible to a broader scholarly public. Cultural translation is a complex process that includes identifying often implicit meanings in informants’ practices and words (rather than just official doctrine or explicit discourse) during fieldwork, and then theorising the implicit patterns and meanings within cultural practices for an audience that may not necessarily care about your informants’ local moral worlds, but which is looking for academic meanings and insights. As Talad Asad (1993, p. 159) notes, [w]hen the anthropologists return to their countries, they must write up ‘their people,’ and they must do so in the conventions of representation already circumscribed (already ‘written around’, ‘bounded’) by their discipline, institutional life, and wider society. To take the example of money and belief mentioned above, a cultural translation of this money–belief dynamic involves contextualising members’ self-understanding of such transactions through the theoretical concept of ‘representational economy’ (Keane 2003), in which money acts as a ‘tangible operator’ for the transvaluation of internal states (i.e. members’ sincerity in their prayers) into recognisable material forms, reinforcing the former in the process. For Asad (1993), this process of cultural translation was made difficult by the fact that anthropological subjects often had no power to speak on their own, and thus he reminds us to be sensitive about becoming the sole voice for other cultures. Thus, writing ethnographic articles involves multiple voices: engaging with existing literature and theories, reflecting on ethnographic findings, and developing your own analysis and contribution—often wrestling with the tensions between your ethnographic material and scholarly presumptions or established theoretical paradigms. Moreover, researchers on Japan inevitably encounter informants and popular media using cultural idioms to explain certain social dynamics, cultural practices and individual psychological processes in Japan. While we might hear explanations for behaviour or practices based on cultural idioms like honne/tatemae (inner feeling/social face), shūdanshugi (collectivism), amae (indulgent dependency), etc. and while there is cultural saliency in those idioms within their moral worlds, either taking these terms at face value or dismissing our informants’ use of these terms as explanatory mechanisms altogether is equally problematic. The former risks falling into the trap of nihonjinron—culturally essentialist and tautological explanations of ‘why the Japanese do things’, while the latter ignores the importance of local logic and personally meaningful explanations by our informants—the stories that they tell themselves to justify or explain beliefs and practices. 444

Chapter 16 How to conduct reliable and fair research Critical analysis involves analysing why such idioms are invoked, how they help informants to make sense of their local worlds and the limits of such explanations. The meaning and contexts behind such usage are more important as they give more clues about what informants mean than the semantic content of the idioms themselves. Yet, there are many cases in which authors take cultural idioms at face value—or reject them outright—and reviewers are often quick to identify this fatal flaw and tear apart a paper, even if it has good ethnographic data and academic potential. In sum, fair research includes critical engagement with cultural idioms within the context of moral worlds that shows the nuances on the ground as well as the implicit cultural logic and dynamic processes that make cultural idioms so powerful for individuals in Japan. Ultimately, the most powerful ethnographies are those which translate local meanings and logic into theoretical analyses that force us to interrogate our assumptions about structural influences, individual agency and creative action in cultural fields. This is how we ‘test the tolerance of [our] own language for assuming unaccustomed forms’ (Asad 1993, p. 190). To rephrase it simply, how does the ethnographic evidence force us to reconfigure our own language of knowledge?

Final thoughts: The balancing act of cultural translation Writing on the process of cultural translation, Asad (1993, p. 180) suggests ‘the anthropologist’s translation is not merely a matter of matching sentences in the abstract, but of learning to live another form of life and to speak another kind of language’. To extend his insights, we can say that this is true of the two registers discussed above. As researchers engaged directly with our informants’ moral worlds, we must grasp the ‘form of life’ of our informants, engaging with the logic and entailments that animate their local worlds and relationships, while also recognising the relations of power and our responsibility to represent local lifeworlds and cultural logic intelligibly. As scholars, we must translate these experiences into a new language of academic knowledge production by weaving and reanimating our informants’ narratives into the scholarly register of disciplinary and theoretical debates. Doing both requires a constant, delicate balancing act of translating ‘local moral worlds’ into global discourses of knowledge production, while staying attuned to our informants’ sensibilities within the ongoing conversations in our fields.

445

16.2 Digital oral narrative research in Japan: An engaged approach David H. Slater, Robin O’Day, Flavia Fulco and Noor Albazerbashi

As long as our research goals are primarily focused on collecting data for purely academic publication, our research priorities usually involve efficiency and verifiability. But once we open the goals and audiences up to something larger than academia, the idea quite quickly becomes complicated. This is a short outline of some of the considerations that arise when researchers work not only with and for other academics, but also work as participants in their own research site, as members of a collaborative process of collection, and on analysis and dissemination of material relevant to a wider audience around topics of shared social and scholarly significance. We might call this ‘engaged research’ (Van de Ven 2007). This article outlines some of the ways an ongoing research team based at Sophia University, Tokyo, has attempted to develop some consistent and productive practices over the last ten years. We draw upon multiple related projects under the collective name Voices from Japan project. These include an oral narrative archiving project with disaster survivors from the 2011 triple disasters in Tohoku, Voices from Tohoku (2019); anti-nuclear activists from Mothers Against War and youth protesters in the SEALDs movement in Tokyo (Slater et al. 2015); homeless men in Yotsuya; and currently refugees coming to Japan seeking asylum. While this work is primarily qualitative and ethnographic, and somewhat specific in its own conceptualisation and execution, we hope that the underlying principles can be applied to a wider range of approaches, disciplines and research themes on Japan.

Selection of topic The first and most basic question involves the focus and goals of the research itself. Does your data, or some part of it, address relevant social issues in Japan around which people have a stake in the outcome? One way to conceptualise a topic is to begin with society, broadly conceived, and to identify issues that are of importance to individuals and groups in your field site and beyond. Our Voices from Tohoku is a digital video archive of more than 500 hours of first-person oral narratives of the 2011 triple disasters in Tohoku, with hundreds of shorter clips on an open website that tell the story of the disaster and recovery to other survivors and to wider audiences of policymakers and the public at large. Since there was no similar sort of research work done on these topics when we began, it was easy to see how we might make some scholarly contribution. But our priority was to engage in serious work that was a direct response to urgent and relevant social and political events. The point of identifying the particular desired social contribution at the start is often important in order to build the scholarly outcomes into our research design, especially through the specification of data we target for collection and for the way we plan to disseminate this data to a wider audience.

446

Chapter 16 How to conduct reliable and fair research Methodologically, it was somewhat challenging. We knew that part of the research output would be a public website with our narrators’ voices on it, which we called nama no koe (genuine voices). Thus, we decided to use the far more expensive and cumbersome collection method of digital video (rather than just audio) because it would be more impactful as a public site. This was made possible due to the close physical proximity of us being in Tokyo, having a large undergraduate research team of interviewers coming into the project each term, and some budget. Nevertheless, it was only because we had conceptualised our research as ‘engaged’ in some way that we were able to produce the archive at all.

Modes of engagement Within this larger framework, there are many modes of engagement that might guide the conceptualisation of our work, as different parts of our ongoing project illustrate. Some work is unabashedly advocacy aimed at the development of a self-consciously ‘politically committed, morally engaged’ (Scheper-Hughes 1995, p. 415) research agenda (see Ganseforth, Ch. 4.2). This sort of advocacy includes the scholarly dissemination of some research findings to political activists seeking a transformational end. So, for example, when we researched political movements (mothers and youth), we shared our interviews with the activists for them to use for their own purposes (of course, only with the written consent of the individual narrators). Alternatively, we might imagine ourselves as a sort of ‘supporter’ or ‘volunteer’ for the individuals and groups with whom we are working. We often bring some useful labour, specialised linguistic or technical knowledge or skill to the field, sometimes with which we barter for research information. In Tohoku, we did disaster relief work (in rough chronological order: delivering food, digging debris, and rebuilding structures) as we collected data. When we work with homeless people and with refugees in Tokyo, we do volunteer work in the soup kitchens or ‘refugee café’ respectively (Ando 2019). Because we work through the regular curriculum of an undergraduate programme, we established this as ‘service learning’, but the same principles apply to individual graduate or professional projects: we seek to ‘give back’ to a community or group in exchange for their taking the time to be interviewed for our project (see Prochaska, Ch. 17.2). Some sort of reciprocity is necessary in any engaged research. However we imagine our particular mode of engagement, it requires us to identify stakeholders and their interests, and to articulate how our research, in both process and product, might address these interests. Thus, rather than the ideal of ‘disinterested’ research that might be used by others in whatever way they wish in the future, we extend our own responsibility for our research outcomes, to ask how we ourselves address the interests of different stakeholders. Often it is not entirely clear who all of the stakeholders might be, and where your topic involves conflict, different stakeholders could have different and equally valid conflicting interests. Engaged research requires some tolerance of such messiness.

Contact and research design We always involve our community stakeholders in the design stages of the project. The advantages of such an approach are often a plan that is more clearly relevant to those involved,

447

David H. Slater, Robin O’Day, Flavia Fulco and Noor Albazerbashi which is one of the requirements of any engaged research, but also more robust scholarly work. The more participants in your field site know about your research, the more they can direct you to issues, data and people you might never have thought about. Moreover, we share preliminary findings, allowing us to identify alternative interpretations, and even refocus midproject. For example, showing some of our video footage to volunteers in the soup kitchen for homeless men using this facility allowed us to generate a number of areas that required further study. Unexpectedly, this data sharing session also established some dialogue between the volunteers and the men themselves about how to improve the service provided by the soup kitchen—the first time that had happened. This sort of collaborative effort also led to a more durable and committed set of relationships as our project developed in new ways. In our case, these same support agencies introduced us to refugee populations in need.

Relationships with stakeholders, collection of data and written release forms It is our personal commitment to make some substantial portion of our own research as collaborative as possible, and since 2011 we have only selected projects where this is possible. These sorts of collaborative arrangements blur the distinction between researcher and subject in productive ways, but in many disciplines and national traditions, this can be a cause of logistical anxiety, scholarly confusion and even academic discredit. In some scholarly contexts, there is an assumption that the line between researcher and research subject must be maintained to preserve the ‘objectivity’ or even the independence of the scholarly practice. In some cases, collaboration and/or the production of some engaged research can lead to questions about the degree of credit the researcher can claim for her or himself. In part to address these concerns, it is necessary to be unambiguous about the legal ownership, control and responsibility of data from the moment the collection process begins. For us, the interviewee retains all rights to the content of the interview, and as such can withdraw permission for the use of this data at any time (during, just after or weeks after the interview). We always use a written release form (dōisho), a sort of contract that gives us explicit but limited permission to use some parts of the data, in some specified forms and context, to a particular audience and under clear attribution. Use of a release form is an opportunity to clearly communicate the goals of the research project to research subjects, to explain that we acknowledge the value of their stories and experiences, and that we take their privacy seriously. While some may think providing informants with the full rights to their own data could make the researcher vulnerable, we can only say that in thousands of hours of interviews, we have only once had someone ask us to withdraw their data from our archive. Many Japanese and foreign researchers seem to have misunderstood research relationships, which has led them to suggest one cannot use written consent or release forms with informants in Japan. This is untrue. In most cases, difficulty with release forms occur when scholars have not spent the time required to develop strong enough relationships of trust with informants. When used correctly, a formal release form is an important research tool that ensures clear communication, facilitates honesty and serves as a licence to speak freely.

448

Chapter 16 How to conduct reliable and fair research

Dissemination of research results While most ethics reviews require that the production of scholarly articles be shared with informants, the final publication of scholarly articles rarely has much impact on this audience. Such articles are usually written in English or other languages besides Japanese, in a style that even most native speakers cannot decipher, in journals in libraries, often beyond digital paywalls. Even if these articles could be accessed, they are oriented towards a set of scholarly rather than social issues. Taken together, these considerations comprise social relevance or meaningful consumption by those from your field site or other stakeholders. Today, more than ever before, there are a large number of publishing options that are open and accessible, so if researchers learn how to write in open and accessible ways on topics of broad social importance, our work will be recognised by a larger Japanese public as relevant and important (see Farrer/Liu-Farrer, Ch. 17; Steger, Ch. 17.3). In Voices from Japan, we have gone a step further by making our original source material accessible. Of our 500 hours of digital oral narrative interviews in Tohoku (and about 150 hours in each of our other projects), where people talk about their community, the disaster and its survivors, and their hope for the future, we shared all of them with the public. From the larger data—which we share with other researchers working on these topics—we have selected hundreds of short clips of one to three minutes, arranged according to geography and narrator, and tagged by theme and time for easy navigation. These have so far gained more than 80,000 hits, exponentially more than any researchers’ article is usually read. Making a public archive might seem like an extreme step, but it has already proved to be hugely useful for scholars young and old, the NPO and other community groups. While there will surely be some who are reticent about us making ‘our data’ accessible to all, once we see this data as the result of collaborative work between us and those who we once thought of as ‘research subjects’ engaged in something larger than the scholarly project, we might imagine this not only as an obligation but even as an opportunity.

449

16.3 Writing for publication: Eight helpful hints Christopher Gerteis

Research writing is central to the career of all scholars. While there are increasingly diverse formats in which to present your research, all rely on your ability to present your thoughts in writing. Indeed, good writing is central to the practice of research, and while styles of presentation differ, good research writing practices will be central to your success as a scholar. This essay will lay out some basic advice I ask less experienced authors to consider as they prepare their research for publication. I serve as an editor for both an academic journal and a research monograph (book) series. Most manuscripts that cross my desktop conform to disciplinary conventions in terms of style, content and methodology. However, I occasionally come across a piece of scholarly writing that transgresses good research practice. Such instances are rare and usually rectifiable so long as the intervention is early enough during the review process. Occasionally, however, I am surprised by authors who decline to revise their manuscript despite the advice of their colleagues, peer-reviewers and editors. 1. The quality of your writing counts. Be sure that your writing is clear and helps the reader come to your conclusion. The number of poorly edited manuscripts that come across my desk is stunning. While I have myself made the mistake of submitting a poorly edited manuscript or two, allow me to advise you to avoid that mistake. Edit, edit again, then edit one more time. And if you are still not sure, engage the services of a professional copywriter. It is very much worth the investment. The right copyeditor brings fresh eyes to your project and can see where you are not being clear. Your publisher is not going to do that work for you. The burden is on you to make it read well and align with conventions. A good scholar must write in a style with which the reader is familiar and trusts. 2. Whatever your discipline, absolutely make certain you are compliant with established research ethics and have overtly identified any potential conflicts of interest. Social scientists and scientists most commonly work within an awareness of the interrelated issues of ethics and conflicts of interest. Yet, in an increasingly privatised university research environment, all scholars, in the Humanities especially, need to clearly identify potential conflicts of interest that can arise from sources of funding and family connections. Failure to do so greatly undermines your credibility as a scholar and causes serious harm to the profession broadly. 3. Accurate source attribution is key. One of my earliest influences as a scholar gave me the best advice I have yet heard: ‘always cite the source you used, not the source that you wanted to use.’ At its core, this advice identified one of the greatest temptations for any writer: the desire to look smarter than you are. This can lead to serious breaches of research ethics, which is why established writing-up practices help keep all of us honest. Showing the shoulders your work stands on establishes your credibility. Each author cited serves as a mnemonic that helps your reader understand where your argument fits into the research literature they have already read. It also helps you explain your thought process in 450

Chapter 16 How to conduct reliable and fair research

4.

5.

6.

7.

8.

a logical and coherent fashion by your using secondary literature as signposts to make it easier for the reader to follow the development of your argument. Showing a logical linkage between what has come before, and where you are going, also lays the foundation of credibility upon which you build the totality of your write-up. Your job as a writer is to persuade the reader of your point of view. Ours is an evidence-driven profession, but presenting evidence of the human condition in written form is a Sisyphean challenge. My doctoral thesis supervisor spilt a great deal of red ink on my draft manuscript, and one recurring phrase was ‘show me, don’t tell me.’ A simple laundry list of assertions will fall flat, and a simple narrative description of events will make no significant contribution to the field. Scholarship is empirical. Make sure the evidence lines up with and leads to the conclusion you intended. Your primary job is to show the reader with evidence and well signposted how you arrived at your conclusions. Keep your notes well organised by using a consistent and easy to access filing system. I often advise my students to use the software tools Evernote and Zotero to help keep their work organised, but there are a number of equally robust software packages that are just as good. The key thing is to have a set of digital tools that allow you to organise your ongoing research notes and evolving bibliography. In a book length research project, your data set is going to become unwieldy very quickly. So, make sure you start your project with a clear organisational system in place, and be certain to make regular backups to a cloud storage account or an external hard drive. You do not want to lose your research to a system crash; it happens more often than you want to know. Do not exaggerate a gap in scholarly literature. As a journal editor, I have mentored many first-time authors who felt obliged to carve their niche in the field by overstating their case. A minor gap can still be significant enough to warrant an article—indeed article length projects often do examine the smaller problems within the human condition. While your colleagues do not want to meditate on minutiae, it is nonetheless important for us to know when something small is important. Show us how that is the case, but do not feel like you have to make it bigger than it is. Criticise, critique or revise, but do not attack. It can be remarkably attractive to want to use your manuscript as a vehicle to attack a scholar or public figure. Discretion is often the better part of valour, especially for an early-career scholar, so keep your discourse civil. You may even find that your thinking changes over time and it will be easier to look back at your work. One final piece of advice: write with passion. Try to write about subjects which you are interested in and feel strongly about. Passion can inform writing and help to keep the reader’s attention. While you want to avoid flowery language—adjectives and adverbs especially—precise imagery and vivid narrative can engage and hold a reader long enough to hear you out. You do not want to manipulate the reader, but you do want their attention. And skilful, passionate prose can help you do that.

The advice I have laid out in this essay will help you establish a good basic framework for thinking about writing, and leaves room for you to develop your own approach to research writing along the way. Your career is yours, and your personal style—your approach—should be distinctive and unique. Following established writing practices is going to be central to your career, and establishing good practices early on will make it all the easier to succeed, but it is just as important that you blaze new paths and explore new intellectual territory. Be yourself. You are your own best asset.

451

Further reading Corti, Louise (2020): Managing and sharing research data: A guide to good practice. Los Angeles: Sage. Haviland, Carol Peterson/Mullin, Joan A. (eds.) (2009): Who owns this text? Plagiarism, authorship, and disciplinary cultures. Logan, UT: Utah State University Press. Iphofen, Ron/Tolich, Martin (2018): The SAGE handbook of qualitative research ethics. London: Sage. Suber, Peter (2016): Knowledge unbound: Selected writings on open access, 2002–2011. Cambridge, MA: The MIT Press. Whiteman, Natasha (2012): Undoing ethics: Rethinking practice in online research. New York, NY: Springer US.

References AAA (American Anthropological Association) (2012): AAA statement on ethics. https://www.americanant hro.org/LearnAndTeach/Content.aspx?ItemNumber=2645&navItemNumber=652, [Accessed 30 August 2019]. Alexy, Allison/Cook, Emma E. (2019): Reflections on fieldwork: Exploring intimacy. In: Alexy, Allison/ Cook, Emma E. (eds.): Intimate Japan: Ethnographies of closeness and conflict. Honolulu, HI: University of Hawai‘i Press, pp. 236–259. ALLEA (All European Academies) (2017): The European code of conduct for research integrity. https://ec. europa.eu/research/participants/data/ref/h2020/other/hi/h2020-ethics_code-of-conduct_en.pdf, [Accessed 24 June 2019]. Ando, Rei (2019): A Sophia University student group is building the bridges that refugees need. In: Japan Times. August 14. https://www.japantimes.co.jp/community/2019/08/14/issues/sophia-university-studen t-group-building-bridges-refugees-need/, [Accessed 08 February 2020]. AoIR (Association of Internet Research) (2019): Internet research: Ethical guidelines 3.0. https://aoir.org/r eports/ethics3.pdf, [Accessed 29 May 2020]. Asad, Talal (1993): Genealogies of religion: Disciplines and reasons of power in Christianity and Islam. Baltimore, MD: The Johns Hopkins University Press. Asef, Esther/Wagner, Cosima/Lee, Martin (2019): Workshop report ‘non-Latin scripts in multilingual environments: Research data and digital humanities in Area Studies’. https://blogs.fu-berlin.de/bibliotheken/ tag/non-latin-scripts, [Accessed 4 April 2020]. Bestor, Theodore/Steinhoff, Patricia/Lyon-Bestor, Victoria (2003): Introduction: Doing fieldwork in Japan. In: Bestor, Theodore/Steinhoff, Patricia/Lyon-Bestor, Victoria (eds.): Doing fieldwork in Japan. Honolulu, HI: University of Hawai‘i Press, pp. 1–17. Calvey, David (2008): The art and politics of covert research. Doing ‘situated ethics’ in the field. In: Sociology 42, No. 5, pp. 905–918. CAO (Cabinet Office) (2019): Kenkyū dēta ripojitori setsubi unyō gaidorainu (Guideline for research data repository construction and operation). https://www8.cao.go.jp/cstp/tyousakai/kokusaiopen/guideline.p df, [Acessed August 27, 2020]. CESSDA (Consortium of European Social Science Data Archives) Training Team (2017–2020): CESSDA data management expert guide. https://www.cessda.eu/DMGuide, [Accessed 5 April 2020]. DFG Deutsche Forschungsgemeinschaft (German Research Foundation) (2013): Safeguarding good scientific practice. Weinheim: Wiley-VCH. DFG Deutsche Forschungsgemeinschaft (German Research Foundation) (ed.) (2019): Guidelines for safeguarding good research practice. www.dfg.de/download/pdf/foerderung/rechtliche_rahmenbedingungen/ gute_wissenschaftliche_praxis/kodex_gwp_en.pdf, [Accessed 5 April 2020]. Edmond, Jennifer/Nugent-Folan, Georgina/Doran, Michelle (2019): Reconciling the cultural complexity of research data: Can we make data interdisciplinary without hiding disciplinary knowledge? In: CODATA Data Science Journal. Preprint. http://hdl.handle.net/2262/83156, [Accessed 4 April 2020]. Embree, John F. (1946): A Japanese village: Suye mura. London: Kegan Paul, Trench, Trubner & Co. ERC (European Research Council) Scientific Council (ed.) (2019): Open research data and data management plans. Information for ERC grantees. https://erc.europa.eu/sites/default/files/document/file/ERC_in fo_document-Open_Research_Data_and_Data_Management_Plans.pdf, [Accessed 4 April 2020]. Fecher, Benedikt/Friesike, Sascha (2013): Open science: One term, five schools of thought. RatSWD_WP_ 218. http://dx.doi.org/10.2139/ssrn.2272036, [Accessed 4 April 2020].

452

Chapter 16 How to conduct reliable and fair research Fulco, Flavia/O'Day, Robin/Slater, David H. (2019): Voices from Tohoku: From a digital archive of oral narratives to scientific application in disaster risk reduction. In: Digital Archive—Basics Vol. 2: Higaikiroku o mirai ni ikasu. Tōkyō: Bensei. Gagné, Isaac (2011): Spiritual safety nets and networked faith: The ‘liquidity’ of family and work under late modernity. In: Contemporary Japan 23, No. 1, pp. 71–92. Gagné, Isaac (2017): Religious globalization and internal secularization in a Japanese new religion. In: Japan Review 30, pp. 153–177. Gerster, Julia (2018): The online-offline nexus: Social media and ethnographic fieldwork in post-3.11 northeast Japan. In: ASIEN—The German journal on contemporary Asia 149, pp. 14–32. GESIS—Leibniz Institute for Social Sciences (2018): Rules for safeguarding good scientific practice. https:// www.gesis.org/fileadmin/upload/institut/leitbild/Gute_Praxis_GESIS_engl.pdf, [Accessed 20 July 2020]. Gill, Tom (2014): Radiation and responsibility: What is the right thing for an anthropologist to do in Fukushima? In: Japanese Review of Cultural Anthropology 15, pp. 151–163. Goldfarb, Kathryn E. (2019): Beyond blood ties: Intimate kinships in Japanese foster and adoptive care. In: Alexy, Allison/Cook, Emma (eds.): Intimate Japan: Ethnographies of closeness and conflict. Honolulu, HI: University of Hawai‘i Press, pp. 181–198. Huber, Elisabeth (2019): Affektive Dimensionen von Forschungsdaten, ihrer Nachnutzung und Verwaltung. Working Paper SFB 1171 Affective Societies 01/19. https://refubium.fu-berlin.de/bitstream/handle/ fub188/24721/SFB1171_WP_01-19_Huber.pdf, [Accessed 4 April 2020]. Imeri, Sabine (2018): Order, archive, share: Research data in the ethnological disciplines. In: Journal of European Ethnology and Cultural Analysis 3, No. 2, pp. 215–240. Imeri, Sabine (2019): ‚Open Data‘ in den ethnologischen Fächern. Möglichkeiten und Grenzen eines Konzepts. In: Klingner, Jens/Lühr, Merve (eds.): Forschungsdesign 4.0—Datengenerierung und Wissenstransfer in interdisziplinärer Perspektive. Dresden: ISGV Digital, pp. 45–59. JPCOAR (Japan Consortium for Open Access Repository) (2020): Sanka kikan. https://jpcoar.repo.nii.ac.j p/?page_id=40, [Accessed 5 April 2020]. JSPS (2019): JSPS international fellowships for research in Japan. https://www.jsps.go.jp/english/e-fellow/d ata/guideline_2020/2020_applicationguideline_e.pdf, [Accessed 2 July 2019]. Keane, Webb (2003): Semiotics and the social analysis of material things. In: Language & Communication 23, No. 3–4, pp. 409–425. Kleinman, Arthur (1992): Local worlds of suffering: An interpersonal focus for ethnographies of illness experience. In: Qualitative Health Research 2, No. 2, pp. 127–134. Kleinman, Arthur (1999): Moral experience and ethical reflection: Can ethnography reconcile them? A quandary for ‘the new bioethics’. In: Daedalus 128, No. 4, pp. 69–97. Knöchelmann, Marcel (2019): Open science in the Humanities, or: Open Humanities? In: Publications 7, No. 65. DOI: 10.3390/publications7040065. Knorr-Cetina, Karin (1988): Das naturwissenschaftliche Labor als Ort der ‘Verdichtung’ von Gesellschaft. In: Zeitschrift für Soziologie 17, No. 2, pp. 85–101. Language Log (2007): Citation plagiarism? http://itre.cis.upenn.edu/~myl/languagelog/archives/004608.ht ml, [Accessed 2 July 2019]. Mai, Bo/Repnikova, Maria (2018): Ethical issues in Internet research: The case of China. In: Foucault Welles, Brooke/González-Bailón, Sandra (eds.): The Oxford handbook of networked communication. Oxford: Oxford University Press. MEXT (2014): Guidelines for responding to misconduct in research. https://www.mext.go.jp/a_menu/jinz ai/fusei/__icsFiles/afieldfile/2015/07/13/1359618_01.pdf, [Accessed 18 July 2020]. Moeran, Brian (1997): Folk art potters of Japan. Beyond an Anthropology of aesthetics. Honolulu, HI: University of Hawai‘i Press. Müller, Andreas W. (2019): MAXQDA or ATLAS.ti? How software shapes research. https://methodos.hyp otheses.org/1575, [Accessed 5 April 2020]. NESH (The [Norwegian] National Committee for Research Ethics in the Social Sciences and the Humanities) (2019): A guide to Internet research ethics. https://www.etikkom.no/globalassets/documents/publik asjoner-som-pdf/forskningsetisk-veileder-for-internettforskning/a-guide-to-internet-research-ethics.pdf, [Accessed 18 July 2020]. Nishiyama, Toshiki/Tokiwa, Takuji/Suzuki, Ryoko (2015): Jitchi chōsa nyūmon: Shakai chōsa no dai ippō. Tōkyō: Keiō Gijuku Daigaku Shuppan. Numazaki, Ichiro (2012): Too wide, too big, too complicated to comprehend: A personal reflection on the disaster that started on March 11, 2011. In: Asian Anthropology 11, No. 1, pp. 27–38. O’Reilly, Karen (2012): Ethnographic methods. Abingdon: Routledge. Pels, Peter/Boog, Igor/Florusbosch, J. Henrike/Kripe, Zane/Minter, Tessa/Postma, Metje/Sleebom-Faulkner, Margaret/Simpson, Bob/Dilger, Hansjörg/Schönhuth, Michael/von Poser, Anita/Castillo, Rosa Cordillera

453

References A./Lederman, Rena/Richards-Rissetto, Heather (2018): Forum discussion: Data management in Anthropology: The next phase in ethics governance? In: Social Anthropology/Anthropologie Sociale 26, No. 3, pp. 1–23. Purdue University Library (2019): Sensitive research data management: Overview—A guide for addressing issues with sharing research data involving human subjects or other sensitive data sets. https://guides.lib. purdue.edu/c.php?g=352785&p=2377925, [Accessed 04 April 2020]. RCOS (Research Center for Open Science and Data Platform) (2017a): Mission. https://rcos.nii.ac.jp/en/ab out/mission, [Accessed 07 April 2020]. RCOS (Research Center for Open Science and Data Platform) (2017b): WEKO3 (Repository Platform). https://rcos.nii.ac.jp/en/service/weko3/, [Accessed 05 April 2020]. Reiher, Cornelia (2014): Lokale Identität und ländliche Revitalisierung. Die japanische Keramikstadt Arita und die Grenzen der Globalisierung. Bielefeld: Transcript. Reiher, Cornelia (ed.) (2018): Fieldwork in Japan: New trends and challenges. Special Issue, ASIEN—The German Journal on Contemporary Asia 149. Reiher, Cornelia (2020): Embracing the periphery: Urbanites’ motivations to relocate to rural Japan. In: Manzenreiter, Wolfram/Lützeler, Ralph/Polak-Rottmann, Sebastian (eds.): Japan’s new ruralities: Coping with decline in the periphery. London: Routledge, pp. 230–244. Reuters (2018): Japan chastises Nobel laureate-led research team after data fabrication. 23 January. https:/ /www.reuters.com/article/us-japan-ips-scandal/japan-chastises -nobel-laureate-led-research-team-afterdata-fabrication-idUSKBN1FC10C, [Accessed 18 July 2020]. Robertson, Jennifer (2007): Ethics and Anthropology: Reality check. In: Critical Asian Studies 39, No. 4, pp. 507–508. Rosa, Hartmut (2010): Alienation and acceleration: Towards a critical theory of late-modern temporality. Malmö: NSU Press. RUDA (Rikkyo University’s data archive for social research) (n.d.): Domestic links. https://ruda.rikkyo.ac.j p/dspace, [Accessed 05 April 2020]. Scheper-Hughes, Nancy (1995): The primacy of the ethical: Propositions for a militant Anthropology. In: Current Anthropology 36, No. 3, 409–440. Shields, James (2010): Beyond belief: Japanese approaches to the meaning of religion. In: Studies in Religion/Sciences Religieuses 39, No. 2, pp. 133–149. Slater, David/O’Day, Robin/Uno, Satsuki/Kindstrand, Love/Takano, Chiharu (2015): SEALDs (Students Emergency Action for Liberal Democracy): Research note on contemporary youth politics in Japan. In: The Asia-Pacific Journal 13, 37, No. 1. https://apjjf.org/-David_H_-Slater/4375, [Accessed 16 August 2019]. Streefkerk, Raimo (2018): Types of plagiarism. http://www.scribbr.com/plagiarism/types-of-plagiarism/, [Accessed 02 July 2019]. Tōhoku kara no Koe (2019): Voices from Tohoku. https://tohokukaranokoe.org/, [Accessed 16 August 2019]. Tóth-Czifra, Erzsébet (2020a): Ten practical tips to fight against the culture of non-citation in the Humanities. https://dariahopen.hypotheses.org/747, [Accessed 18 July 2020]. Tóth-Czifra, Erzsébet (2020b): One more word about ResearchGate/Academia.edu and why using these platforms will never be equal to proper self-archiving. https://dariahopen.hypotheses.org/878, [Accessed 18 July 2020]. University of Glasgow (n.d.): Self-plagiarism. https://www.gla.ac.uk/research/ourresearchenvironment/prs/ self-plagiarism/, [Accessed 04 July 2020]. University of Sheffield (n.d.): What is research data management? https://www.sheffield.ac.uk/library/rdm/ whatisrdm, [Accessed 18 July 2020]. Van de Ven, Andrew H. (2007): Engaged scholarship: A guide for organizational and social research. Oxford: Oxford University Press on Demand. Wilkinson, Mark D./Dumontier, Michel/Aalbersberg, IJsbrand Jan et al. (2016): The FAIR guiding principles for scientific data management and stewardship. In: Scientific Data 3, Article no. 160018. DOI: 10.1038/sdata.2016.18. Yamashita, Shinji (2012): The public Anthropology of disaster: An introductory note. In: Asian Anthropology 11, No. 1, pp. 21–25.

454

Chapter 17 How to present findings: Presenting and publishing James Farrer and Gracia Liu-Farrer

1. Introduction: Reaching your audience Research is intended not only to produce results but to convey them to someone. But who is your audience? And how do you best reach them? As important as these questions are, the answers—usually multiple—are often not well thought-out. In our experience, a great many wonderful research projects—including many MA and PhD theses—are never published beyond the initial thesis, and a chief reason is that young scholars fail to consider their audience beyond their committee members. For the purposes of this chapter, we will divide audiences into two types: the professional and the public. Both of these are important, but they also pull your project in different directions, and we urge young scholars to think clearly about their relative importance at different stages in their careers.

2. Think about your audience(s)! Defining a desired audience can be daunting. Should I write for sociologists? For Japanologists? For the general public? Considering that years or even decades of your life will be spent on a project, thinking dynamically and flexibly about different audiences can keep a project alive and relevant. In our experience, it is all too easy to lose sight of the goal of publishing during the research project. To stay on target, it is best to begin presenting and publishing very early in the research process. The professional audience is nearly always the primary one for academic researchers, and is usually a requirement for career advancement or landing a first academic job. Therefore, young scholars need to familiarise themselves with the ‘tricks of the trade’ of international academic journal publication. Mentors and senior colleagues are good sources of advice in this process, but sometimes young scholars are on their own. Beyond the academic audience, academic research typically addresses problems of public importance. Scholars who receive public support for their research often have an obligation to present findings of interest to a broader public. Giving a public face to their research may also help scholars integrate their academic research more closely with teaching and community service. Mentors and senior colleagues are often less help here. The Internet offers many tools for reaching new far-flung audiences, but it is not foolproof. For example, we have found our-

455

James Farrer and Gracia Liu-Farrer selves facing the headache of keeping online scholarship alive beyond the few years in which a particular online platform survives. One lesson of this chapter is that publishing and research should be thought of as a recursive process in which earlier publications—including conference presentations, working papers and preliminary journal articles, but also non-conventional formats such as blogs and public data sets—are all different ways of framing and reframing an ongoing academic project. A project may have a much longer life, and reach many more interested people, if all these outlets are part of a process of developing ideas.

3. Reaching an academic audience Most scholars who start doctoral programmes aim to become career academics or professional researchers, i.e. working for a university or research institute. The core audience for their research is therefore fellow academics and researchers. This section focuses on three typical academic outlets for research, ordered along the most common temporal sequence: conferences and workshops, academic journals and books.

3.1 Conferences and workshops There are numerous conferences and workshops, ranging from gigantic to miniature, available to researchers around the world. While all conferences are places to present our research findings, they also assume different functions for participants. In our experiences, they can be roughly categorised into three broad types: disciplinary, Area Studies and thematic conferences. Because Japanese Studies is interdisciplinary, Japanese Studies scholars often seek to contribute to a discipline while also reaching an interdisciplinary Japanese Studies audience. It may be necessary therefore to present different versions of their work at these two types of conferences. Both disciplinary and Area Studies conferences announce their conference themes, venues and dates one or two years ahead of time through their organisational websites. The application deadlines are often eight months to even a year before the conference dates. For young scholars, large disciplinary and Area Studies conferences are exciting, because they are opportunities to be exposed to the best and most cutting-edge research in the disciplines as well as the area. They also confer a sense of professional belonging. Presenting at such conferences initiates one’s membership in the academic community. Such conferences are also places where one networks, identifies potential collaborators, and receives information related to research and career development. Sometimes job recruitment activities are conducted alongside the conferences. Because of their sheer sizes and diversity, however, such big conferences are often less productive in terms of receiving feedback for our work. There are simply too many parallel panels to choose from, and we might not see as big an audience for our own presentations or have a chance for deeper discussions. The purposes of large conferences is often professionalisation, whereas focused academic discussion is often more productive, in our experience, at smaller thematic workshops and symposiums.

456

Chapter 17 How to present findings Participating in smaller thematic conferences, on the other hand, is a very productive way for scholars to establish themselves in a field as well as to foster publication projects such as edited volumes and journal special issues. However, it is not always easy to find them. Once we start to participate in academic events, we tend to receive ‘Call for Papers’ (CfPs) for thematic workshops and symposiums circulating via various mailing lists we have signed up for. There are some big thematic conferences; the ones we receive the most insightful feedback, however, are the smaller symposiums and workshops organised by an institution or programme with specific interests. They are often the most productive in terms of producing publications and building long-term research collaborations. Such workshops are also more likely to fund participants’ travel: the conveners make the call often because they have received a grant for the event they propose to host with the goal of advancing a particular research agenda. The conveners are also interested in producing journal special issues or edited volumes out of such workshops. Both authors have had several publications produced through such workshops and symposiums (Farrer 2010; Liu-Farrer 2010). Information on smaller workshops and symposiums are posted through professional association websites and mailing lists (see this chapter, North, Ch. 17.1, for some of these websites).

•

•

• •

•

Checklist for conference presenters Articulate a puzzle: The most effective presentations are those that pique the audience’s interest with a puzzle and then focus on presenting the findings that solve the puzzle. Minimise preliminaries: Many people, veteran researchers included, spend too much time on literature reviews and methodology, and run out of time before they can say anything exciting. Time yourself: Conference presentations usually last 15 to 20 minutes. Filter advice: Conferences are where new concepts and ideas are circulated. Audience members will throw all sorts of theoretical advice at you, or recommend new jargon. Stay calm and grounded. Seek advice that truly enhances the interpretive power of one’s own empirical findings. Do not waste a conference paper: Conferences are places where we get feedback on our ongoing research. We should utilise such opportunities to improve a paper and try to get it published in a professionally recognised journal

In short, though their functions and productivity vary, conferences and workshops are necessary places to present our research findings, to be known as researchers specialised in particular subjects and areas, to build professional identities, and to enter relevant research networks. Young scholars should try to participate in all these different types of conferences and workshops.

3.2 Academic journals When we choose a journal, we are also choosing an audience. Sometimes, conveners of thematic workshops or symposiums will organise journal special issues, saving the authors’ time and energy in locating the appropriate journals. More commonly, however, we make indepen-

457

James Farrer and Gracia Liu-Farrer dent submissions to international journals. Given the increasing number of journals, how do we find the right one to submit to? For a Japan Studies scholar, typical journal outlets can be lumped into the following five broad categories: • • • • •

Subject journals—interdisciplinary journals that deal with the specific subject areas of our research, such as International Migration Review and Sexualities. Area Studies journals—in this case, interdisciplinary journals focusing on Asia and Japan, such as Journal of Asian Studies, Contemporary Japan and Japan Forum. Disciplinary journals—such as American Journal of Sociology. Journals that are dedicated to specific methodological approaches, such as Journal of Contemporary Ethnography. Crossovers—journals that are combinations of methodological and disciplinary approaches, such as Qualitative Sociology, or of a specified discipline and subject, such as Medical Anthropology. Social Science Japan Journal, as the name suggests, combines both disciplinary and area specifications.

Most articles could potentially fit into any of these categories of journals. The choice of journal is a judgement about how well our research speaks to a specific audience. By choosing a specific journal, we are expected to engage the intellectual conversations that are typical of that genre of journal. In other words, one needs to reference the literature in that field, and often in that journal. Different journals call for attention to different types of literature (see McMorran, Ch. 15). For example, when Gracia drafted a paper about the emigration of rich Chinese, she saw this migration phenomenon as something emerging in a changing China and decided to submit it to a China journal (Liu-Farrer 2016). As a result of this choice, she needed to include the body of literature on class structure and social stratification in Chinese society as well as that on the logic of migration. If she were to publish it in a migration journal, much more effort would be needed to engage literature on emerging patterns of migration. If she had aimed to write for a sociological journal, she would have needed to consider how the empirical study could make original contributions to the discipline of Sociology, which would have resulted in a different theoretical discussion emphasising disciplinary significance. As empirical researchers who have been working in specific subject fields with a geographic focus, we find subject as well as Area Studies journals easily correspond to our work. Your direction may depend on your professional goals. If you are aiming at employment in a particular discipline, you may need at least some publications in recognised disciplinary journals. Regardless of whether you favour area-based or discipline-based publications, in the era of academic globalisation, it is generally wise to consider publications that are regarded as having broad recognition in the field (either through quantitative or qualitative measures).

458

Chapter 17 How to present findings Checklist: Preparing a journal article The main aim of a journal article is to engage in the conversations with specific academic communities by indicating how our original research fills an empirical gap and advances theoretical development. It is therefore always important to ask yourself the following questions: • Who is the audience of this article? Does it correspond to the audience of the journal we are considering? • What is the state of the field and how does your article contribute to it? • What is your main argument? Keep in mind that you have only several thousand words, so ONE clear argument is preferred. • Does your evidence support your argument? Young scholars always wonder how much data, especially the number of interviews, is needed for a journal article. There are no fixed criteria. It depends on both your methods and the field you contribute to. Instead, ask whether you are confident that the evidence you present illustrates your argument. When preparing a journal article, make sure that it incorporates the following before sending it out: • A clear and informative abstract that indicates the main research question and main argument: The abstract is of great importance because it is the first thing the journal editors and the anonymous reviewers read. A paper is more likely to be reviewed when it piques a reviewer’s intellectual interest. A well-written abstract attracts attention and also indicates the quality of the paper. • A clear structure: When receiving a paper submission, the first decision a journal editorial board makes is whether it is worth sending out for reviews. A paper that does not look like a journal article is most likely to be rejected without even entering the review process. • The style that fits the type of journal it is submitted to: Some journals might want policy analysis, while others prioritise theoretical contributions. Make sure you are familiar with recent issues of the journal. The above procedures should be applied to ensure that your paper does not end at desk rejection, and actually gets sent out to external reviewers. Once it enters the anonymous review stage, you have an increased chance of having it published because the editors have already invested in your paper. Upon receiving review results, as Scott North (see this chapter, Ch. 17.1) explains, do not get flustered but read the reviews again and again. A major revision means that the editors hope to eventually publish it. Even a rejection does not mean the end of the paper. You will receive valuable comments from experts on the paper and revise it accordingly to submit it again or somewhere else. Sometimes, it is okay to disagree with the reviewers. Then, when resubmitting, state the reasons why you disagree with the reviews. It is through such (repeated) revision processes that our scholarship improves. The whole process from submitting a journal article to receiving the notice about its final status can take somewhere between six months to a year. The initial review process might take three to four months. If you do not hear from the journal editors after three months, do not 459

James Farrer and Gracia Liu-Farrer hesitate to enquire. The author is then given six to ten weeks to revise their work. Depending on whether another round of reviewing is required, it might take another two to three months to hear back from the editors. Depending on the editorial decision, the paper might need further revision. When it is officially accepted as a publication, the online version often appears much earlier than the paper version. Many popular journals are so backlogged that an article accepted in 2021 might be intended for an issue planned for 2023. If you are building a career, do not just work on one piece at a time. It is best to have multiple pieces ‘in the pipeline’.

3.3 From dissertations to books For qualitative researchers, the first book often develops out of their dissertation.1 When considering publishing monographs, we face the choice of university presses versus commercial presses. The former tend to be more selective, have fewer titles and take a longer time to publish. The commercial ones, in comparison, publish many more titles, are more aggressive in soliciting submissions and publish more quickly. However, in order to make a profit on the limited number of copies they sell, their book prices are usually much higher than those from university presses. In some academic circles, university presses are considered superior because of their selectivity. However, good works have emerged from both forms of presses. Sometimes, excellent work is rejected by cash-strapped university presses because of the concern that it lacks a sufficient market. The choice of commercial press also has to do with time constraints, the broadness of the field, as well as what is required to advance one’s career. Regardless of all this, one rule of thumb by which to judge suitability is to take a look at your own bookshelf: which press publishes most of the books in your field? To publish a book, one needs to send publishers a book proposal (or prospectus) together with sample chapters and, in the case of most university presses, a complete manuscript. One can find samples for such proposals on the Internet, or ask one’s colleagues who have published books. Each press might also specify different content. Upon receiving successful anonymous reviews, the author is then offered a book contract. From the time one submits a full manuscript, it takes more than a year for a book to be published. Academic presses take longer. Gracia delivered the full manuscript of her book Immigrant Japan to Cornell University Press in July 2018. A year prior to that, she had contacted the Asian Studies editor about her book project, and showed him several sample chapters and received comments. She received both anonymous reviews by the end of October 2018, and the book contract in late November. She turned in the revised manuscript in February 2019. The book then entered production, and was published in early 2020 (Liu-Farrer 2020). It took over one and a half years from submitting the manuscript to the book finally coming out, and much longer if her initial contact with the press is taken into account.

1 Please note that the rules and procedures for publishing dissertations vary in different national contexts and across universities.

460

Chapter 17 How to present findings

3.4 Beware of predators Because of the importance of all these formats for academic careers, businesses have emerged which prey on scholars by inventing conferences, journals and presses. With the increasing pluralisation of academic outlets, sometimes it is difficult to tell the legitimate ones from those that are not. When in doubt, play it safe. Stick to conferences that have a longer running history, and stay away from those that have several disciplines or fields in the same title. Publish in journals that have been around for more than ten volumes (years). Shun presses that you have never heard of. It is always a good idea to consult with veteran researchers.

3.5 Promoting your publications Your work is published. Congratulations! But you should also think of sharing your publications (including the ones to a broader public, which is covered in the next section). Even to reach a greater professional audience, we recommend using websites such as Academia or ResearchGate to promote pre-publication drafts or links to your publications. Social media outlets such as Twitter and Facebook are also effective tools. Nando Sigona (2019), a researcher studying migration and refugee issues, provides useful advice and practical tips for promoting research through social media. Updating institutional web pages or personal academic web pages also increases visibility and provides additional content such as photographs or blog posts that cannot be included in academic formats. James runs a web page (2017) in order to publish images from ongoing fieldwork as well as links to publications. Consistently promoting one’s publications online will greatly increase their availability to other academics. However, note that for young scholars, non-refereed papers published on such web pages are generally not counted as academic publications.

4. Reaching a broader audience Nearly every academic research topic has an audience beyond professional academia, and it is rarely reached simply by writing an academic book with a catchy title. Different approaches are called for to reach a non-specialist audience. Moreover, non-specialist writing can typically reach a broader audience of academics as well, and may be very useful in undergraduate teaching. Broadly speaking, there are two ways of finding this larger circle: letting others do it for you or doing it yourself. We will start with the former, but focus on the latter.

4.1 Talking with journalists For many topics, sitting for an interview with a journalist or blogger may be the most effective way of getting your ideas across to a larger audience of readers or listeners. This may be true even for topics that are not already in the news, a situation described by Brigitte Steger in relation to her research on Japanese sleeping practices in her essay for this chapter (see this chap-

461

James Farrer and Gracia Liu-Farrer ter, Ch. 17.3). The benefits of letting journalists write about your work are obvious: minimal effort from you and greater impact through professional writing, editing and marketing. Good interviewers may even find ways of framing your ideas that you would not have thought of yourself, or bring out elements of one’s research that are buried inside a long monograph. Given the brevity of many media interviews, however, we may worry that journalists will misreport our findings or garble our ideas. For scholars working on politically or socially sensitive topics, this is a common concern. To control the quality of the messenger, Steger recommends working with reporters from established or reputable journalistic outlets. Also, for reports in which you are the main source, she recommends checking over texts before publication. Both authors of this chapter have been interviewed numerous times about our research in media formats ranging from mainstream television and newspaper reporting to community-centred blogs and papers. We rarely refuse an interview. In our experience, even the journalistic questions we do not like—because they seemed biased or presumptive—may be an important indicator of ideas or findings that we should highlight more clearly in our academic writings. It is always possible to help the journalist rephrase the question to better match your expertise and to make sure the answers you provide are what you intend. The most productive way of working with journalists, we have found, involves bringing them more closely in contact with our research, and discussing the project over multiple meetings. James did a 30-minute television special with NHK World based on his Japanese neighbourhood food research (NHK World 2019). The first meetings with the producer aroused his suspicions, since the producer seemed to want to focus the programme on his personal life as a foreigner in Japan. However, in discussions with the producer he refocused the segment on James’s research project on neighbourhood foodways. Working with the producer at all stages of production from editing the script, choosing people to interview and (to a more limited extent) discussing the final product, ensured that the production represented the ideas of the research project extensively and fairly, while still achieving the goals of the producer (to create an entertaining introduction to a Tokyo neighbourhood).

4.2 Publishing for a general audience While working with journalists may be one of the most efficient ways to reach a broad audience, a deeper approach to reaching the public may require producing your own texts. Publishing books for the general reader used to be considered a career-killing move—particularly among North American social scientists—but it has always been the case in Japan that scholars write for a broad reading public. And especially when one is writing in the Japanese language, popular writing may be a more effective way of reaching a Japanese audience than aiming at narrow Japanese-language academic journals. Young scholars are generally cautioned against jumping into popular book writing, but some have used popular writing to further their academic research agendas. For example, Patrick Galbraith, an anthropologist who teaches in the School of Law at Senshu University in Tokyo is the author and editor of many books on Japanese media and popular culture, both in popular and academic formats (Galbraith 2014). When asked about the advantages of appealing to a broader audience, he replied:

462

Chapter 17 How to present findings By translating your academic research into popular publications and vernacular, you can reach more people, meet them where they are and potentially impact ongoing discussions and debates. I feel that this is incumbent on those working on issues of public interest and social importance. We really shouldn’t just be talking to one another inside the institutions of the university and the adjacent publishing industry (Galbraith, email interview June 2019). Galbraith suggests that we social scientists have an ethical obligation to reach a broader audience with our ideas, and not limit them to a ‘handful of people’ with institutional library access. In short, Galbraith argues, not only can we reach a broad audience as academics, but we should also try to do so. The practice of writing for a general audience may actually help us clarify our academic writing as well.

4.3 Film and video Another method of reaching a broader audience is through film and video, an approach with even greater potential impact through the rise of YouTube and online streaming services. Isabelle Prochaska-Meyer (see this chapter, Ch. 17.2) writes in detail about the process of producing a documentary on ageing people in rural Japan. Even early career scholars also find this approach rewarding. One Japanese Studies scholar who took this approach is Dipesh Kharel, who in his PhD research at the University of Tokyo blended traditional ethnography with videography in studying the transnational lives of Nepali restaurant workers in Japan and their connections with their family in Nepal. This resulted both in his PhD dissertation and his documentary film Playing with Nan (2012). When asked about the advantages of this approach to presenting his data, Kharel replied: Certainly, conducting visual ethnographic fieldwork, recording data and editing more than 600 hours of ethnographic footage into a coherent ethnographic film was very timeconsuming and took me several years, which is one of the shortcomings of this approach. However, I have learned that ethnographic film is the best way to disseminate research findings to the larger audiences (Kharel, email interview June 2019). As Kharel’s case shows, documentary films may not only be created by senior scholars with substantial institutional resources but also by postgraduate students as part of their dissertation research. In some academic degree programmes, a documentary film may be presented as an alternative to a traditional thesis. For example, Miki Dezaki produced the documentary film Shusenjo (2019) on the controversies surrounding the wartime ‘comfort women’, which completed his MA degree in Global Studies at Sophia University. We asked Dezaki about the advantages of this approach to presenting data: A comment I often hear regarding my film is that they have read quotes by the people in my film before, but seeing them say these things on screen is much more powerful and impactful. It seems that one of the biggest advantages to making a documentary film over a written thesis is that video interviews when coupled with music and other types of visuals can make the project more appealing to a wider audience compared to a written thesis (Dezaki, email interview June 2019). Dezaki also pointed out, however, that this type of project can be quite challenging for a young researcher with limited funding and time. Echoing Meyer-Prochaska in this volume, 463

James Farrer and Gracia Liu-Farrer Dezaki replied that producing a documentary can be like an additional ‘full-time job while you are taking classes’ (Dezaki, email interview June 2019). Dezaki, Kharel and Meyer-Prochaska all find their films have reached a broad audience, which most scholars will never achieve through academic writing alone. This exposure brings accolades and (sometimes critical) attention, though also costs a lot in terms of time spent and the stress of production and publicity. Translating visual data to textual formats also poses challenges. As Jamie Coates, another young academic film-maker, wrote back to us, it may be ‘difficult to translate your observations and findings into more widely recognised formats’ (Coates, email interview June 2019; see also Coates, Ch. 3.2). Therefore, as intriguing as filmmaking is, whether visual data can be simultaneously used for producing traditional publications (transcribing and coding may be necessary) is a concern for academics. It is also incumbent upon senior academics to inform junior colleagues how a film, video or podcast will be treated in career evaluations.

4.4 Sharing data Social Science research inevitably produces far more data than can be represented in even a long book or film. Publicising your research, therefore, can also go far beyond a final authored product; it may also involve sharing the data itself, either in its raw or a highly curated form, or somewhere in between. David Slater, a Japan Studies scholar and anthropologist at Sophia University in Tokyo, created a curated project on the aftermath of the Tohoku Earthquake, in which interviews conducted by faculty members and students have been shared online, allowing researchers all over the world access to it. The resulting Voices of Tohoku is an oral narrative project of more than 500 hours of semi-structured interviews captured on digital video from survivors of the 2011 earthquake, tsunami and nuclear disasters in northern Japan (Tōhoku kara no koe 2019; see also Slater et al., Ch. 16.2). Slater explained the goals and principles behind the online project: We developed a collaborative principle whereby anyone who contributed to the archive would be entitled to gain access to the full corpus of data, making it a very valuable resource to many MA, PhD and post-doctoral students who used this data for their own research projects (Slater, email interview June 2019). Despite its academic and public benefits, Slater cautioned that producing this type of public scholarship may be costly for younger scholars and is ‘best led by a senior scholar with more resources and job security’ (Slater, email interview June 2019). In short, as with all forms of public scholarship, early career scholars should consider their audience early and consult with institutional leaders about whether this type of public-benefit scholarship is recognised in their institution.

4.5 Connecting with a community Another goal of public scholarship is not only to share data and findings, but also to foster interaction and dialogue with the community one is studying. This is the idea behind James’s (2019) Nishiogiology Urban Food Studies project. This project centres on a webpage that doc464

Chapter 17 How to present findings uments the experiences of small-scale culinary entrepreneurs in the Nishi-Ogikubo district of Western Tokyo. The web page is produced in both English and Japanese so that local residents can also easily access the information, which also provides some benefits for the interviewees in terms of providing information and publicity about their businesses. Many members of the local community follow the web page through social media. Creating the project has not only allowed James to connect with informants but also other people involved in community journalism and place-making. The web page thus functions not only as a form of academic knowledge in urban studies, but also as a project to support the community that it studies. One consideration in public scholarship in Japanese Studies is the language of publication. For Nishiogiology, publishing in Japanese was essential but also very challenging. For James, working very closely with a Japanese editor and research assistant was key to the success of the bilingual project.

5. Conclusions: Balancing goals In this chapter we have covered publications aimed at a professional and a broader general audience. Both may be essential for fulfilling careers as academics, though only the former is systematically recognised by institutions. This presents a dilemma for researchers that we can only point to here. In the era of increasingly strong pressure to either ‘publish or perish’, it is important for young scholars to inform themselves of what forms of publication are expected. Increasingly, peer-reviewed journal articles are preferred by academic institutions, sometimes with specific requirements for the types or ‘ranks’ of journals that are acceptable. Therefore, if data is gathered in a video format, we have to think not only of visual presentations, but also how we can code the data for presentation in a standard article. If we are collecting copious interview data, we should consider how this can be properly stored as a future public data archive that will benefit many researchers. Yet, early career scholars also need to be clear about how their participation in the group project will lead to publication. Finally, in the age of electronic publication, there are many new technical dilemmas. We have to consider not only archiving our raw data but sustaining archives and publications that we have created online. For any online publications or data we create, it is also best to back them up in various ways to ensure their survival if the original platform goes offline. Ironically perhaps, paper may still be the most reliable medium with which to ensure our publications will be available not only in a year, but in a hundred years.

465

17.1 Finding an audience: Presenting and publishing in Japanese Studies Scott North

I am a sociologist interested in labour, gender, law, social movements and quality of life in Japan. The issues I study are common in industrialised societies. Like most academics, I have concentrated on presenting my findings at professional conferences and publishing in peer-reviewed publications because they are the coin of the intellectual realm. But my research topics are also news, so there are parts of my research that I can hive off as newspaper articles or opeds. Regardless of discipline or field, I think publicly funded scholars are obliged to try to reach multiple audiences and to put things happening now in a theoretical–historical context.

Choosing where and how to publish or present It is best to have a long-term plan for getting your research into print. Writing with a particular journal in mind is probably the way to proceed. Each journal has a particular mission and audience. Unless your paper suits the journal, it will not be reviewed. Also, each journal has its own style guide. Preparing an article to meet the style requirements for publication is timeconsuming. So knowing the style and writing the piece in line with those guidelines from the start is efficient. That said, even if you do not have a plan, the world of academic publishing is an unexpectedly forgiving place. Do not be afraid to submit your work because you fear rejection. Conversations with journal editors have taught me that journals sometimes struggle to find enough decent submissions. I submitted a paper based on my doctoral dissertation to a major Japanese Studies journal in 2005. It was rejected. Chastened by the referees’ comments, I put the paper away for a year or so while I licked my wounds. Then the editor wrote to ask if I was still working on it. It seems that the editorial board thought the paper showed more promise than the reviewers did. Energised by the idea that the journal felt there was an audience for the paper, I buckled down, doing more research and drastically changing the content of my piece. It was re-reviewed and, since finally being published (North 2009), it has become one of the most frequently read articles in that journal. I have to credit both sets of reviewers for their clear-eyed appraisal, and the editors for their continuing interest, without which I might have ‘abandoned the manuscript to the gnawing criticism of the mice’, as Marx and Engels said of their failed effort to publish The German ideology (Tucker 1995, p. 146). Here is another example of how failure sometimes works in a writer’s favour. As a doctoral student, I submitted a paper to The Journal of Japanese Studies. It was based on research for my Master’s thesis about karōshi, death caused by overwork. Although the paper was rejected, one of the referees sent helpful comments, which succinctly identified a very common struc-

466

Chapter 17 How to present findings: Presenting and publishing tural weakness in graduate student papers and offered an elegant solution. Despite this advice, again I put the paper away for a time. Then came a chance to participate in a small workshopstyle conference organised by one of the International Sociology Association research committees. The papers had to be submitted prior to the workshop. As in the first example, the anticipation of an audience for the work provided the motivation needed to revise the manuscript, which, in its improved form, eventually found a home in a Sociology journal (North 2011). One lesson from these examples is that when reviewer’s comments arrive, we need to remember that they are suggestions rather than orders. Do not be dismayed by the words ‘major revisions’. Instead, negotiate your way to publication. First, negotiate with your feelings to reach a dispassionate understanding of reviewers’ comments. With each successive read, emotions diminish and the constructive aspects of the criticisms shine through. Remember also that no matter what reviewers say, the paper is yours: you do not have to address every criticism or meet demands for revisions that you disagree with. But you must justify your response. When you resubmit the paper, provide a letter in which you explain which comments you addressed and how, or did not address and why. Resubmissions may be read by editors instead of blind peer reviewers. Their incentives differ. Whereas blind peer reviewers doing unpaid labour are sometimes harsh in defence of disciplinary integrity or ideas that they may favour, the editors want to put out a journal. Editors may be less likely than reviewers to act as disciplinary gatekeepers because they want well written, stimulating contributions that will attract a broad audience to the journal. Should you publish only in peer-reviewed journals? I say, ‘Yes.’ And you should aim at higher quality journals because, even if you are rejected, their reviews are more likely to show you the way forward. Also, the better journals tend to be able to offer quality editorial processing (proofreading and copyediting) and they are more experienced at negotiating compromises. That said, however, some open-access, online-only journals are also worth considering as outlets for your work. Examples include The Electronic Journal of Contemporary Japanese Studies and The Asia-Pacific Journal: Japan Focus. Both are non-profit organisations that review manuscripts quickly and upload articles to their sites continuously, ensuring speedy publication of your work, in full text and with free access. Because they are online publications, these journals are more flexible about inclusion of graphics and colour plates. And do not forget publishing in newspapers or magazines; they occasionally pay their contributors. Wherever you decide to publish, if you are asked to pay a fee to see your work in print, it is probably best to walk away.

Conference presentations Most of my published papers started as conference or workshop presentations. Preparing for professional meetings causes you to focus on some portion of your research, and submission deadlines concentrate your attention on getting words on paper. In addition to the large, annual disciplinary conferences in Europe and North America, there are numerous intimate opportunities throughout the year. I think you are more likely to get helpful feedback at workshops, roundtables and smaller conferences. To have conference opportunities delivered to your desktop, subscribe to Japanese Studies mailing lists or website feeds (e.g. the British Association for

467

Scott North Japanese Studies, the European Association for Japanese Studies, H-Japan, Social Science Japan Forum and the Society for East Asian Anthropology). The bar for conference presentations is lower than for publication, so conferences are good places to try out new ideas and learn how to position and market your work. Moreover, conference attendees, such as journal editors, scholars with plans for edited volumes, publishers looking for scholars with manuscripts in hand, organisers of panels for future conferences and university job search committee members may see your presentation or find the title of your paper in the programme. Networking at conferences sometimes yields invitations to contribute to forthcoming projects. Browsing the book exhibits is another way to sense how broad the audiences for your work might be.

Some publishing problems: Edited volumes, language concerns Edited volumes may take longer to appear than journal articles. In 2000, I eagerly participated in a panel with other authors of a proposed edited volume on law and social change in Japan. By time the book came out (Steinhoff 2014), from a cash-strapped university press, the cases were history. Delay is not ideal for young scholars, who need to publish in order to find work or gain promotion. Two papers on Japanese fatherhood in edited books offer an illustrative contrast. The first example (North 2012) shows how edited volumes can obscure your work. The publisher’s description states, ‘Taking an international focus, Men, wage work and family contrasts [...] paid work and non-work domains in industrialised countries in Europe, North America, and Australia’, but although my contribution is entitled The work-family dilemmas of Japan’s salarymen, Asia is omitted from the blurb! The second publication (North 2014), however, was superbly organised. The editors had a strong vision for the volume, for which they recruited draft chapters and secured funding to bring the authors together for two days of critical discussion. Following revisions, the volume was submitted to the publisher for blind review, leading to another round of revisions, professional proofreading and editing and, finally, publication, all in just two years. My experience with edited volumes in Japanese, however, has been uniformly good. I have no hope of ever writing perfect Japanese. This is strangely liberating, even if the writing is quite difficult and time-consuming. I do not make an English draft, but write directly in Japanese from an outline consisting of key words. All of my contributions to edited volumes in Japanese have gone from draft to publication in a single fiscal year.

General advice To sum up, do not hesitate to submit a more or less completed manuscript, or to consult journal editors about a proposal for a paper based on your in-progress research. Editors encourage writers and help them understand how to position their work to find an audience. So take some risks: write a fast draft of your paper, and then submit the abstract or first section of the paper with a letter introducing the work to the editors of a journal that seems to publish the sort of thing you are working on. If they express interest, then slowly and carefully rewrite 468

Chapter 17 How to present findings: Presenting and publishing and edit your draft to meet the journal’s style. Follow the excellent advice of Adam Pzreworski and Frank Salomon (1995) about appealing to the inner needs of reviewers by making the contribution of your paper clear: in the title if possible, but certainly by the end of the first page. If your paper is sent out for review, it may take months for the comments to come. While you wait, turn your attention to other projects; find a conference or workshop in which to present the work. Be patient. If after six months no news is forthcoming, send a polite inquiry to the editors. Study the academic art of Social Science writing. Read classic texts about long-form writing to learn how others pursue their craft (e.g. Becker 2007; Bolker 1998; Lamott 1995). Treasure your graduate school papers. With revision, they become the basis for future publications. Join (or create) a writing group that corresponds to your needs. Finally, write simply and avoid jargon. Remember Reinhardt Bendix, who said, ‘You know, a little bit of theory goes a long way’ (Stinchcombe 1968, p. v). And always think of your readers—the audiences that you hope to reach.

469

17.2 Ethnographic film and fieldwork on active ageing in rural Japan Isabelle Prochaska-Meyer

One thing that I have always found fascinating about fieldwork is that it affords the researcher access to people and places that they may not have been able to experience otherwise. Similarly, documentary filmmakers often gain access to environments that lie outside their usual surroundings, enabling them to observe and comment on the conditions, conventions, views and feelings of members of a group or setting. This essay outlines the process of making the documentary film 65+ Being old in rural Japan (Kieninger/Prochaska-Meyer 2014). It was produced as part of the ethnographic project Aged communities and active ageing—A case study of rural villages in the Japanese Alps, headed by Sepp Linhart. The project was carried out by Pia Kieninger and me and funded by the Oesterreichische Nationalbank’s Anniversary Fund. The research project focused on the experiences of elderly people in rural municipalities in Japan and analysed how different actors deal with the issue of ageing and depopulation. We wanted to produce a film during our fieldwork to portray the situation in our research sites and to show the life of senior citizens in rural Japan to a broader audience in and outside Japan. Our research included two fieldwork stays of four months in total (two months in autumn and two the following summer) in three municipalities in the Nagano and Yamanashi prefectures. Through these two stays we hoped to develop closer relationships to our informants and gain a better insight into the topic. Starting from the pre-production stage, through shooting and post-production until the movie was finally released, this essay summarises both the challenges encountered during the fieldwork phase, the making of the ethnographic film and what we learned from this project.

Pre-production: What story did we want to tell? Before the first shooting phase, we only decided on the style of the documentary and left the more specific film pitch for later in order to carry out our research inductively. Nevertheless, we decided on a few things before going to the field: For example, we agreed that we would not put ourselves into the picture and defined the approximate length of our film. It turned out to be a great advantage that we had planned two fieldwork and shooting phases. During the first fieldwork phase, we were able to be open to a broad spectrum of topics and locations in order to get to know the characteristics of the villages and the activities of the elderly people there. For the second fieldwork phase, we already had a clearer image of how our film would look and had established trust relations with the informants. The tentative script that we designed for our second stay focused on the portrayal of active senior residents in different living situations in order to discuss the concept of active ageing (WHO 2002). We met elderly people living on their own or in three-generation households, as well as retired men who had returned 470

Chapter 17 How to present findings: Presenting and publishing to their rural home village in order to care for their mothers. After the first fieldwork phase, we applied for funding for the post-production of the documentary film. We already had visual references for our film and a clearer outline of the planned documentary’s working title, length, target audience, main focus and goal. Luckily, we were able to rely on close cooperation with the municipal offices at the respective study sites. Before meeting senior residents for interviews, we participated in communal activities or observed specific organised courses (e.g. a gymnastics course, chorus rehearsals, etc.) to meet elderly community members. During these activities, the course leader always knew in advance that we would be present, and we asked for consent to observe and film the activities. For the first few days, we spent time exploring the hamlets of the municipality and filming daily scenes: people working in their fields, the shop that sold everyday food items and agricultural tools, or scenes of nature like kaki fruits hanging from leafless trees, etc. In these shooting situations, we were also able to spontaneously meet people and have informal talks. In total, we conducted interviews with 32 elderly people. The interviews outlined several main points of interests, such as participation in municipally organised group activities for the elderly, mobility, social interaction, etc. In addition, we also conducted expert interviews with municipal administration and health care representatives. Our initial film pitch was to portray senior citizens in various living conditions. However, in the end, we selected only two single senior residents, who lived on their own, as the main protagonists for our film: 84-year-old Shimako-san from Minamiaiki, who talks with a husky voice, grows vegetables and whose passion is gateball (a Japanese team sport inspired by croquet); and 93-year-old Genichi-san from Kitaaiki, the oldest villager with a driving licence, who composes short poems on daily events. We decided to focus on their stories during our second stay in the two neighbouring municipalities. Both Shimako-san and Genichi-san were very hospitable and open, so we were able to get an intimate glimpse into their daily lives. We accompanied them to their vegetable plots, to the supermarket or at home drinking tea with neighbours.

Post-production: What story were we going to tell? After screening all the interviews and scenes of participant observation, we transcribed and translated all the conversations with our main informants and identified important topics that provided hints as to a potential structure for the film. We then realised that we wanted a narrator in the film in order to explain the context of the research and to show the community life of senior villagers in general. The narrator would describe the overall situation in the study sites so that the focus would periodically shift between the individual perspectives (Shimako-san and Genichisan) and the total picture (group activities, daily routine in the villages). Rather than commenting on the situation in our study sites from our subjective perspective as researchers, we chose a professional ‘neutral-voiced’ narrator, who recited our text in German. We added subtitles in English and Japanese as well as in German when the two protagonists spoke on camera. All other interviews, as well as the expert interviews, were recorded and translated, but in the post-production phase, we decided that only the voices of our two main informants and the narrator would be included in the film. The expert interviews contributed to an understanding of the topic, but in the end we did not include them as original comments in the film.

471

Isabelle Prochaska-Meyer An important process during the post-production phase was getting feedback at test screenings. This helped us to become aware of potential misunderstandings. Our opening scene is an empty street in the morning. In the background, municipal loudspeakers broadcast the morning melody, which announces the time of day to local residents. After a preview of the rough cut, a colleague did not recognise the melody as an original sound in the background, but rather asked us why we had chosen such a ‘childish melody’, which emphasised the image of the elderly as childlike. This was a very important remark for us, because in our view the context seemed clear as we had also explicitly shown a close-up shot of the loudspeaker. Consequently, we included a commentary in the subtitles which explained that the melody in the background was being broadcast through the municipal loudspeakers.

Challenges and what we learned As amateurs in the film-making world, we were not aware of the workload an ethnographic film involves. On the one hand, this was good because we approached the film-making with a broad interest and shot on any occasion possible. On the other hand, we returned from our fieldwork with an enormous amount of visual and sound material, totalling more than 4,000 files with ca. 1.3 terabytes, from which we could only use a fraction for the final documentary. A more defined film pitch could have facilitated the shooting process so that we would have decided more precisely during the second shooting phase which scenes we would have liked to include in the film. The schedule for our first fieldwork stay was definitely too full. On one day, for example, we started with the senior citizens’ chorus rehearsal in the community centre. Afterwards we drove to the day-care centre, where we conducted an interview with the director and then with a nurse. We then drove to another hamlet where we had previously arranged an interview with an 83-year-old man who was still engaged in commercial farming. In the evening, we visited the community centre again, this time to observe the hula dance course, in which all the participants were in their 70s or older. The schedule for this day was obviously too intense, and as a result we did not have enough time to reflect and recharge our batteries—both literally and figuratively. We therefore decided on a maximum of three activities per day. During our second fieldwork stay, we also planned days off as part of our schedule. We allowed more time to transcribe or at least organise the interview memos while on location and to keep the research journal up to date and more reflective. During interviews there were also some points we had to keep in mind while filming. Contrary to standard qualitative interviews, where affirmative comments and reactions help the interviewee to keep on talking (see Kottmann/Reiher, Ch. 7), we had to switch to ‘silent affirmation’ by nodding or employing facial gestures that expressed surprise or amusement in order not to have our comments or distracting sounds picked up in the recording. When filming our main informants during informal conversations, we also needed to keep the camera running and capture silent moments. On the other hand, we deliberately switched off the cameras and microphones in situations when our interpersonal relationship with the informant was in the foreground rather than the intended subject matter of the ethnographic film project. Examples of this included when we helped Genichi-san with his farm work or when we cooked a Western meal together with Shimako-san in her kitchen.

472

Chapter 17 How to present findings: Presenting and publishing Sometimes, while on location, you tend not to see the wood for the trees. For example, in both the research and fieldwork phases, we encountered the problem of abandoned houses in rural areas. However, back in the editing studio, we had difficulty finding cuts showing an abandoned house, because although we saw abandoned houses every day on location, we had simply forgotten to intentionally film such a house. It would have been helpful to make an explicit list of important topics and terms related to our research beforehand and to collect related visual material, just in case.

Conclusion: It’s a wrap! The documentary film project made me more aware of the subtle techniques you can use in editing, with visuals, cuts and sounds in order to trigger a certain reaction (e.g. smiles) among the audience. A documentary film is always a ‘subjective construct’ (Rabiger 2009, p. 18), not only through the editing process, but also with the camera (Engelbrecht 1995, p. 153). Therefore, at all stages, from pre-production until editing, the film-making process (as in every other project) demands reflection and discussion. Having a camera during fieldwork also allowed us to print out and give pictures to our informants. While this strengthened our relationships with people, it was also our way of showing gratitude for their cooperation (see Gagné, Ch. 6.1; Klien, Ch. 8.1). We realised the powerful function of photographs as memories of events and people when we learned during our second visit that one of our informants had passed away a few months before. The documentary was shown at film festivals, Japan-related events, lectures and conferences. After the screening, we were also able to discuss the wider context of the research project, and we received feedback from colleagues who had shown the film in class. One Austrian viewer found Shimako-sans remark that she was looking forward to her old age in order to join the gateball group very surprising and impressive—as in our Western society getting old is rarely associated with something positive and desirable. In general, audiences in Europe thought the film was positive and that the senior citizens were portrayed as physically and mentally very active, while we also received remarks from some Japanese viewers that it was sad to see elderly people living alone in the countryside. They were rather worried about the future of elderly people in rural areas ten or twenty years ahead. In one study site, the documentary was broadcast on the municipal TV channel. Our main informants also received a DVD copy of the film and were satisfied with the results. We uploaded the film to YouTube, where it can be watched and commented on by an ‘unlimited audience’ anywhere, any time. Our aim of introducing the topic of active ageing in rural Japan to a broader audience had therefore been realised.

473

17.3 Weird and wonderful: Popularising your research on Japan Brigitte Steger

Do you want to be in the news? In the 1990s, that question was tricky. Aspiring academics feared for their reputation when their name appeared in the popular press. Academia is supposed to be respectable. That attitude reminds me of the neo-Confucian scholar Ogyū Sorai, who, in 1715, criticised a new entertaining lecturing style that ‘mixes in a few funny stories to wake up sleepers’ as unsuitable for the training of serious scholars. The practice ‘degrades the teacher’s character and corrupts the student’s intellect’ (Dore 1965, p. 140). It was thought to be impossible to dumb down serious research results, and if you could do it, that was hardly a sign of sophistication. However, today, the mood has certainly changed. No one now doubts the importance of getting your research into the news. After all, universities have to legitimise their use of taxpayers’ money. To secure further funding, we need to make an impact. Making our research known in popular media also benefits us directly. Most students—and even professional academics—first learn about academic work outside their narrow speciality from the media or via a Google search. Of course, media responses and public comments will often be thoughtless or plain stupid, but constructive feedback can be very helpful. Moreover, relying on academic journals alone to get the word out is not ideal as it requires expensive subscriptions that not all libraries are able to buy; and in some countries, access to academic literature is further restricted by censorship. How do you get started? What format should you choose? Or perhaps I should say, how do the media choose us? Anyone can, of course, publish their work on the Internet; and it is very useful to have information on your research projects and publications online. Make sure that journalists googling for experts find you if they suddenly become interested in your topic. However, not every topic is equally ‘sexy’ for a general readership. Since anyone can publish on the Internet and much of what we do is not obviously topical, it is hard to get attention for your latest findings when you want or need it. Do not underestimate the reach of university newsletters and magazines or academic society websites. For instance, the JAWS website japananthropology.org welcomes reports by young scholars introducing their research on Japanese society, both on published work and work in progress. They might not have popular appeal at first sight, but good academic journalists regularly monitor these magazines and might recognise the ‘newsworthiness’ of your story. These outlets are also trusted, and do not require you to write for dummies or reduce sophisticated arguments to sensationalist headlines. It makes a lot of sense to pay attention to the quality of an article you might publish in such formats. In fact, some of my most important media appearances were based on university publications. When I was invited to give a lecture at the Technische Universität in Berlin in 2004, my recently published PhD thesis on social and cultural aspects of sleep (Steger 2004) was reviewed in their newsletter. A journalist from the weekly newspaper Die Zeit read this, did her research and contacted me. She read the whole book and travelled from Hamburg to Vienna for an interview, which resulted in a full-page article introducing me and my research

474

Chapter 17 How to present findings: Presenting and publishing (Etzold 2004). Not many media organisations can afford such expenses, but Die Zeit is subscribed to by other newspapers, magazines or TV stations, so the story spread quickly. More recently, in 2016, I was asked by the Cambridge University alumni magazine CAM to write something about my research on public napping. Over the years I had learned to present my findings on inemuri (to be present and sleep) to general audiences, so I complied. Several of my colleagues and former students told me that they had read the article; for the first time they knew what I was doing. That was nice, but a few months later, an editor of BBC Future acquired the rights to publish the article on the BBC Worldwide network (Steger 2016). In hindsight, I should have negotiated a higher fee for that article, as it proved to be immediately popular and was selected to be part of the ‘Best of Future 2016’. By November 2018 it had received about 1.4 million individual browser hits in English alone. BBC Worldwide is a huge network, and the article was translated into Albanian, Arabic, Azeri, Chinese, Greek, Hausa, Indonesian, Persian, Portuguese, Russian, Spanish, Swahili, Turkish and Vietnamese. I also found summaries or excerpts of it in many other languages, which were copied and discussed in private blogs and which inspired people to design napping devices, put together quizzes about napping habits or use the term inemuri as their music band’s name. True, most of this did little to further my career, but it is nevertheless good to see that the ideas inspired other people. A number of media outlets, such as the New York Times, the Guardian, Dagens Nyheter or Russian RenTV, have since contacted me for interviews. The most common way for your research to get into the news is by means of an interview when a journalist has found you as a useful source of information or to provide ‘sound bites’ for their radio programmes. Interviews are usually edited to a few seconds or minutes at most. Some journalists ask questions through which they obviously want you to feed their stereotypes. It can be difficult to get your own message across. You need to be careful and selective. How serious is the request? Some journalists send a catalogue of questions by e-mail; you sit down and think of thoughtful answers, only to find out that they have just a line or two to write about what you say. Ask how they will use the interview; if it is longer and you are their main informant or the story is about you, ask them to send you a draft to confirm. Generally speaking, it is better to concentrate on well-known quality media; not only do they have more space to write and a readership which will be happy to read sentences exceeding eight words, but these journalists are also better trained and ask interesting questions. Some of them can even be inspiring, making you think of a new angle on your research. When you write the story yourself, you have ‘message control’, but there are a few basics to keep in mind, in particular, the structural difference in writing academic essays and stories for a general public. Academics try to find ‘the truth’ about certain phenomena, and then they think of how to present their results. Journalists are looking for good and topical stories and information that help to explain. They talk about the ‘7 W’ questions: who, what, where, when, how, why and where from. When we write a story, we should always check whether we have answered all these questions. In fact, this is not only a requirement for more popular writing. We academics work for many years on one topic and thus tend to take the most obvious information for granted. So it helps to remind yourself of those questions. Journalists also ask: Does this article fulfil the ‘breakfast criterion’? In other words, would you continue reading this article if you came across it while flipping through the newspaper and having your morning coffee? If anything, in the age of clickbait, this has become crucial—and demanding.

475

Brigitte Steger While we usually think of the title right at the end of the writing process, to a reader it is the start; the hook. A good title has to catch the attention of a potential reader, and the first paragraph (as well as the rest of the text, of course) needs to be clear and engaging. I once heard the story of a man who researched the most popular words in bestseller titles; they were ‘cat’, ‘golf’ and ‘sex’. He subsequently published his own book, including all three terms in the title. Lo and behold, the book sold very well. Of course, your title needs to deliver on the argument of your research, but it also needs to reach the readers on an emotional as well as an intellectual level. I understand that not every topic is easy to sell and it is a question of timing. The global interest in the 3.11 disasters has declined, in diametrical opposition to the excellent research that has been produced. The intricacies of Heian period (794–1185) court protocol or of recent changes in Japanese waste disposal law might sound rather dry, catering to minority interests. It is the researcher’s job to show the significance of these seemingly detailed questions and make it relevant for local and international audiences today. A royal wedding or abdication in Japan provides a great opportunity to ‘sell’ your research, as does any kind of environmental discussion. It is as much about how you write as what you have to say. The same goes for visuals. Beautiful, witty, puzzling, ironic or iconic, there are many reasons a picture speaks to us. A good picture should make potential readers curious of what you have to say. Of course, copyright issues also need to be considered. There is also the ‘grandma criterion’. Can your grandmother understand your article? It is always better to write for one specific person you care about than trying to make everyone happy. Your text will get some immediacy and it will be easier for your readers to connect with it. Being able to write for an intelligent but non-specialist audience is a useful skill to acquire, in particular for job and grant applications, but even for your academic work. It can be quite a challenge to simplify what you have to say. Sometimes seemingly boring details are necessary, and not all audiences are able and willing to follow lengthy and detailed arguments. Keep those for your peer-reviewed articles. Moreover, in everyday language, fancy theories sound far less sophisticated. Getting rid of all the 50 dollar words, we may look like the emperor in his new clothes. However, simplifying and abbreviating force you to think precisely and do not mean that you need to avoid all technical terms. If your Japanese terms are the most precise with which to describe an issue, explain them well. In fact, your audiences will like to learn intriguing Japanese terms. When you write for a general audience, make sure that it is clear and ‘flows’. In fact, this is useful advice for any kind of text. Reading academic work—and especially students’ essays—I sometimes feel like I am chasing a rabbit. When rabbits try to escape a fox or dog, they dodge and jink right or left very suddenly. The dog falters and then turns to take up the chase again. Jinking is a useful skill for a rabbit, but it is better if your readers can catch you and follow your arguments; you can and should captivate them. Finish your thought before starting a new one and do not cram several thoughts into one sentence! I am quite aware how difficult this can be, especially when you are struggling with a word count. Read your article out aloud and listen to yourself. Of course, even today, working with the media and popularising your knowledge is not without risks. Japan-related topics are often published under the category ‘weird and wonderful’. The media cater to stereotypes, both by feeding them and by provoking contradiction (such as in my case, about diligent Japanese who take a nap at work). Not everything is in our control, but trying to get the stories out ourselves as clearly written as possible and being aware of how 476

Chapter 17 How to present findings: Presenting and publishing the media work do help us to avoid the most serious pitfalls. Go for quality media outlets; they ask better questions and allow you more time and space to develop your arguments. Talk about work that is already published, so that people can buy your books or read your articles, and no one can simply steal your research ideas! The main problems and biggest risks are really the demands on your time and how you organise and prioritise. But neglecting media work is no longer an option.

477

Further reading Becker, Howard S. (2007): Writing for social scientists: How to start and finish your thesis, book, or article. Chicago, IL: University of Chicago Press. Curry, Mary Jane/Lillis, Theresa (2013): A scholar’s guide to getting published in English: Critical choices and practical strategies. Blue Ridge Summit, PA: Multilingual Matters. Klingner, Janette K./Scanlon, David/Pressley, Michael (2004): How to publish in scholarly journals. In: Educational Researcher 34, No. 8, pp. 14–20. Moxley, Joseph Michael (1992): Publish, don’t perish: The scholar’s guide to academic writing and publishing. Westport, CT: Greenwood Press. Pzreworski, Adam/Salomon, Frank (1995): On the art of writing proposals. www.ssrc.org/publications/vie w/7A9CB4F4-815F-DE11-BD80-001CC477EC70/, [Accessed 18 August 2020]. Rosenthal, Alan/Eckhardt, Ned (2016): Writing, directing, and producing documentary films and digital videos. Carbondale, IL: Southern Illinois University Press. Wellington, Jerry J. (2003): Getting published: A guide for lecturers and researchers. London: RoutledgeFalmer.

References Becker, Howard S. (2007): Writing for social scientists: How to start and finish your thesis, book, or article. Chicago, IL: University of Chicago Press. Bolker, Joan (1998): Writing your dissertation in fifteen minutes a day: A guide to starting, revising, and finishing your doctoral thesis. New York, NY: Owl Books. Dezaki, Miki (dir.) (2019): Shusenjo [film]. 120 min., No Man Productions LLC. Dore, Ronald Phillip (1965): Education in Tokugawa Japan. Berkeley, CA: University of California Press. Engelbrecht, Beate (1995): Film als Methode in der Ethnologie. In: Ballhaus, Edmund/Engelbrecht, Beate (eds.): Der ethnographische Film. Berlin: Dietrich Reimer, pp. 143–186. Etzold, Sabine (2004): Die Globalisierung des Nickerchens. In: Die Zeit. August 12. https://www.zeit.de/2 004/34/P-Steger/komplettansicht, [Accessed 18 August 2020]. Farrer, James (2010): ‘New Shanghailanders’ or ‘New Shanghainese’: Western expatriates’ narratives of emplacement in Shanghai. In: Journal of Ethnic and Migration Studies 36, No. 8, pp. 1211–1228. Farrer, James (2017): Global Japanese Cuisine Project. www.global-japanese-cuisine.org, [Accessed 24 March 2020]. Farrer, James (n.d.): Nishiogiology. www.nishiogiology.org, [Accessed 16 October 2019]. Galbraith, Patrick W. (2014): The moé manifesto: An insider's look at the worlds of manga, anime, and gaming. Clarendon, VT: Tuttle Publishing. Kharel, Dipesh/Saito, Asami (dir.) (2012): Playing with Nan [film]. 88 min., RAI Film. Kieninger, Pia/Prochaska-Meyer, Isabelle (dir.) (2014): 65+: Being old in rural Japan [film]. 35 min., www.youtube.com/watch?v=GDyPwiVObzg, [Accessed 15 October 2018]. Lamott, Anne (1995): Bird by bird: Some instructions on writing and life. New York, NY: Anchor Books. Liu-Farrer, Gracia (2010): The absent spouses: Gender, sex, race and the extramarital sexuality among Chinese migrants in Japan. In: Sexualities 13, No. 1, pp. 97–121. Liu-Farrer, Gracia (2016): Migration as class-based consumption: The emigration of the rich in contemporary China. In: The China Quarterly 226, pp. 499–518. Liu-Farrer, Gracia (2020): Immigrant Japan: Mobility and belonging in an ethno-nationalist society. Ithaca, NY: Cornell University Press. NHK World (2019): I love Tokyo! A professor explores Nishi-Ogikubo: TOKYO EYE 2020. https://www 3.nhk.or.jp/nhkworld/en/ondemand/video/2053145/, [Accessed 11 July 2019]. North, Scott (2009): Negotiating what’s ‘natural’: Persistent domestic gender role inequality in Japan. In: Social Science Japan Journal 12, No. 1, pp. 23–44. North, Scott (2011): Deadly virtues: Inner-worldly asceticism and karōshi in Japan. In: Current Sociology 59, No. 2, pp. 146–159. North, Scott (2012): The work-family dilemmas of Japan’s salarymen. In: McDonald, Paula/Jeanes, Emma (eds.): Men, wage work and family. London: Routledge, pp. 17–33.

478

Chapter 17 How to present findings: Presenting and publishing North, Scott (2014): Hiding fatherhood in corporate Japan. In: Inhorn, Marcia/Chavkin, Wendy/Navarro, Jose-Alberto (eds.): Globalized fatherhood: Emergent forms and possibilities in the new millennium. New York, NY: Berghahn, pp. 53–80. Pzreworski, Adam/Salomon, Frank (1995): On the art of writing proposals. www.ssrc.org/publications/vie w/7A9CB4F4-815F-DE11-BD80-001CC477EC70/, [Accessed 18 August 2020]. Rabiger, Michael (2009): Directing the documentary. Amsterdam: Focal Press. Sigona, Nando (2019): Thinking impact. https://prezi.com/803jgrg0sudo/ node-ukjapan-masterclass/, [Accessed 24 March 2020]. Steger, Brigitte (2004): (Keine) Zeit zum Schlafen? Kulturhistorische und sozialanthropologische Erkundungen japanischer Schlafgewohnheiten. Münster: LIT. Steger, Brigitte (2016): The Japanese art of (not) sleeping. www.bbc.com/future/story/20160506-the-japan ese-art-of-not-sleeping, [Accessed 08 February 2020]. Steinhoff, Patricia G. (2014): Going to court to change Japan: Social movements and the law in contemporary Japan. Ann Arbor, MI: University of Michigan Press. Stinchcombe, Arthur L. (1968): Constructing social theories. Chicago, IL: University of Chicago Press. Tōhoku kara no koe (2019): Voices from Tohoku. https://tohokukaranokoe.org, [Accessed 18 August 2020]. Tucker, Robert (1995): The Marx-Engels reader. New York, NY: W.W. Norton & Company. WHO (World Health Organization) (2002): Active ageing: A policy framework. http://whqlibdoc.who.int/ hq/2002/WHO_NMH_NPH_02.8.pdf?ua=1, [Accessed 15 October 2018].

479

Notes on contributors Shinichi Aizawa is an Associate Professor at Sophia University. His main research interests include the Sociology of Education, social mobility and social class. He recently published High school for all in East Asia (Routledge, 2019) and Sōchūryū no hajimari: Danchi to seikatsujikan no sengoshi [The rise of all-middle-class society: Postwar history of public housing complex and time-use] (Seikyūsha, 2019, co-edited with Watanabe). Noor Albazerbashi is a MEXT Fellow at the Graduate Program in Global Studies at Sophia University. Her work is on Syrian refugees and immigration support in Tokyo. Daniel P. Aldrich is Director of the Security and Resilience Studies Program and Professor of Political Science and Public Policy at Northeastern University. He received his PhD from Harvard University and his main research interests include the role of social networks during crises and disasters, nuclear energy and environmental politics. He recently published Black wave: How networks and governance shaped Japan’s 3/11 disasters (University of Chicago, 2019). Allison Alexy is an Assistant Professor at the University of Michigan. She received her PhD from Yale University and her main research interests include intimacy, family, law and divorce in contemporary Japan. She has recently published Intimate disconnections: Divorce and the romance of independence in contemporary Japan (University of Chicago Press, 2020). Celeste L. Arrington is the Korea Foundation Assistant Professor of Political Science and International Affairs at George Washington University. She holds a PhD from the University of California, Berkeley. Her main research interests include law and social change, legal professionals, social movements and comparative policy processes. She recently published ‘Hiding in plain sight: Pseudonymity and participation in legal mobilization’ in Comparative Political Studies (2019) and ‘The mechanisms behind litigation’s “radiating effects”: Historical grievances against Japan’ in Law & Society Review (2019). Verena Blechinger-Talcott holds a PhD from Ludwig-Maximilians-Universität München and is Professor of Japanese Politics and Political Economy at Freie Universität Berlin where she was also one of the founding directors of the Graduate School of East Asian Studies. Her research interests include Japanese politics and international relations in comparative perspective, particularly the role of institutions. She recently co-edited a special issue on Dimensions of SinoJapanese rivalry in a global context in The Pacific Review (2019) (with Schulze). Christoph Brumann is Head of Research Group at the Max Planck Institute for Social Anthropology, Halle and Honorary Professor of Anthropology at Martin Luther University HalleWittenberg. His most recent research has focused on the urban Anthropology of Kyoto, in particular urban conservation and development and the current condition of Buddhist temples. He published Tradition, democracy and the townscape of Kyoto (Routledge, 2012) and coedited World heritage on the ground: Ethnographic perspectives (Berghahn, 2016). His monograph on the UNESCO World Heritage Committee is currently under review.

Notes on contributors Genaro Castro-Vázquez is Professor of Sociology at Kansai Gaidai University where he teaches courses related to Medical Sociology and Sociology of Education. He obtained a PhD from the University of Tsukuba and undertook postdoctoral studies at Keio University. His research interests include sexuality, gender, health and education and Latin Americans living in Japan. Recent publications include Male circumcision in Japan (Palgrave, 2015), Intimacy and reproduction in contemporary Japan (Routledge, 2017) and Masculinity and body weight in Japan: Grappling with metabolic syndrome (Routledge, 2020). David Chiavacci is Professor in Social Science of Japan at the University of Zurich where he also obtained his PhD. His research covers political and economic Sociology of contemporary Japan from a comparative perspective. His recent publications include Japanese political economy revisited: Abenomics and institutional change (Routledge, 2019) and Civil society and the state in democratic East Asia: Between entanglement and contention in post high growth (Amsterdam University Press, 2020). Jamie Coates holds a PhD from Australian National University and is a Lecturer in East Asian Studies at the University of Sheffield. His research interests include the Anthropology of media, mobility and imagination in China and Japan. He recently published ‘The cruel optimism of mobility: Aspiration, belonging and the “good life” among transnational Chinese migrants in Tokyo’ in positions: asia critique (2019). Emma E. Cook holds a PhD from the University of London and is an Associate Professor in the Modern Japanese Studies Program at Hokkaido University. Her main research interests include issues related to food, health, risk, emotion, gender, family and intimacy. She recently published a special issue with De Antoni titled Feeling (with) Japan (Asian Anthropology, 2019) and an edited volume with Alexy titled, Intimate Japan: Ethnographies of closeness and conflict (University of Hawai‘i Press, 2018). Laura Dales obtained her PhD from the University of Western Australia where she is a Senior Lecturer in Asian Studies. Her main research interests include agency, sexuality, friendship and dating across Asia, as well as singlehood and marriage in contemporary Japan. She recently published the co-edited collection Configurations of family in contemporary Japan (Routledge, 2015) with Dasgupta and Aoyama. Andreas Eder-Ramsauer is a PhD Candidate and Junior Research Fellow at the Freie Universität Berlin. He holds an MA in Japanese Studies from the University of Vienna. His main research interests include Japanese politics, democratic thinking in postwar Japan and populism in Japan. James Farrer is Professor of Sociology at Sophia University in Tokyo. His research focuses on the contact zones of global cities, including ethnographic studies of sexuality, nightlife, expatriate communities and food. His recent publications include Shanghai nightscapes: A nocturnal biography of a global city (co-authored with Field, University of Chicago Press, 2015), the edited volume Globalization and Asian cuisines: Transnational networks and contact zones (Palgrave, 2015) and International migrants in China’s global city: The new Shanghailanders (Routledge, 2019).

481

Notes on contributors Flavia Fulco is Assistant Professor at Tohoku University in the International Research Institute of Disaster Science (IRIDeS). She has a PhD in American Studies and has conducted research on the cultural memory of the 3.11 disaster. She is a collaborator on the Voices from Japan digital oral narrative project based at Sophia University. She recently published ‘Voices from Tohoku: From digital archive of oral narratives to scientific application in disaster risk reduction’ (with Slater and O’Day) in Digital Archive Basics 2 (Bensei Ed., 2019). Isaac Gagné is a Senior Research Fellow at the German Institute for Japanese Studies (DIJ) in Tokyo and Managing Editor of Contemporary Japan. He holds a PhD from Yale University. His research interests include morality and ethics, religion and globalisation, mental health and social welfare, gender and popular culture. His publications include ‘Religious globalization and reflexive secularization in a Japanese new religion’ (Japan Review, 2017) and ‘Dislocation, social isolation, and the politics of recovery in post-disaster Japan’ (Transcultural Psychiatry, 2020). Nana Okura Gagné is a cultural anthropologist and Associate Professor at The Chinese University of Hong Kong. Her research explores how global capitalism impinges upon local ideologies and social and gender relations, family relations, socio-economic and class relations and individual subjectivities. She holds a PhD from Yale University and has published in American Ethnologist, Anthropological Theory, Ethnography and the Journal of Contemporary Asia. Her book, Restructuring and resilience: Changing men at work and play in neoliberal Japan is forthcoming (Cornell University Press). Sonja Ganseforth is a Principal Researcher at the German Institute for Japanese Studies (DIJ) in Tokyo and received her PhD from Leipzig University. Her research interests include the social and economic geography of the globalisation of agrifood systems, rural livelihoods, development politics and discourses and social movements in Japan. She recently published Occupations: Japanese development spaces in Palestine (Transcript, 2016). She is the co-editor of a volume on the socio-politics of sub-national spaces in Japan (Routledge, forthcoming). Sheldon Garon is Nissan Professor of History and East Asian Studies at Princeton University. He completed his PhD at Yale University and his research interests include state-society relations in modern Japan and transnational history that spotlights the flow of ideas and institutions between Japan, Europe and the United States. His publications include Beyond our means: Why America spends while the world saves (Princeton University Press, 2012). Julia Gerster is Assistant Professor at the International Research Institute of Disaster Science (IRIDeS) at Tohoku University. She holds a PhD from Freie Universität Berlin and her main research interests include the dynamics of social relations after disasters, recovery processes, negative heritage, identity and community building. She has recently published ‘Hierarchies of affectedness: Kizuna, perceptions of loss and social dynamics in post-3.11 Japan’ (International Journal of Disaster Risk Reduction, 2019).

482

Notes on contributors Christopher Gerteis is an Associate Professor of Modern and Contemporary Japanese History at the University of London. He is the founding series Editor of the scholarly monograph series SOAS Studies in Modern and Contemporary Japan, published in association with Bloomsbury, and he was Chief Editor of Japan Forum. He currently holds a five-year residence as Associate Professor and Academic Editor for the International Publishing Initiative in Humanities and Area Studies at the Institute for Advanced Studies on Asia, at the University of Tokyo. Roger Goodman is Nissan Professor of Japanese Studies at the University of Oxford where he also obtained a DPhil. His main research interests include Japanese education and social policy. He recently published (with Breaden) Family-run universities in Japan: Sources of inbuilt resilience in the face of demographic pressure, 1992–2030 (Oxford University Press, 2020). Markus Heckel holds a PhD from the University of Duisburg-Essen and is a Senior Research Fellow at the German Institute for Japanese Studies (DIJ) in Tokyo. His research interests include monetary policy, the political economy of central banks and labour economics. He recently published an article about labour contracts in Japan and the World Economy (with Genda and Kambayashi, 2019). Steffen Heinrich received a Doctorate from Heidelberg University and is Visiting Professor of Japanese Politics and Political Economy at Freie Universität Berlin. His main research interests include welfare and labour market policies in Japan from a comparative perspective. He recently published the book chapter ‘The politics of balancing flexibility and equality: A comparison of recent equal pay reforms in Germany and Japan’ (Springer, 2019). Joy Hendry holds a DPhil from the University of Oxford and is Professor Emerita of Social Anthropology at Oxford Brookes University. Her research interests include rural Japan, marriage and family life, child-rearing, politeness and presentation (wrapping) and cultural display. She recently published a fifth edition of Understanding Japanese society (Routledge, 2019), a collection of her many articles appeared in An anthropological lifetime in Japan (Brill, 2017) and she is working on an illustrated account of forty-five years of fieldwork entitled An affair with a village. Swee-Lin Ho is Associate Professor in the Department of Sociology at the National University of Singapore. She received her PhD from the University of Oxford and her research interests include the neoliberal transformations of work, gender, friendship and urban space in Japan; the globalisation of Korean popular culture; and the growing importance of western classical music to Asia. Among her recent publications are Friendship and work culture of women managers in Japan: Tokyo after ten (Routledge, 2018), and Women managers in neoliberal Japan: Gender, precarious labour and everyday lives (Routledge, 2020). Barbara Holthus holds PhDs from the University of Hawai‘i and Trier University. She is a sociologist and Deputy Director of the German Institute for Japanese Studies (DIJ) in Tokyo. Her main research interests include marriage and the family, childcare, happiness and well-being, media and demographic change. Her publications include Life course, happiness and well-being in Japan (Routledge, 2017, co-edited with Manzenreiter) and Japan through the lens of the Tokyo Olympics (Routledge, 2020, co-edited with Gagné, Manzenreiter and Waldenberger).

483

Notes on contributors Carola Hommerich is Associate Professor at Sophia University in Tokyo since 2019. She received her PhD from the University of Cologne and her research focuses on how people evaluate their place in society, on whether their objective living situation matches their subjective experience and how that in turn affects their subjective wellbeing, attitudes and behaviour. Her recent publications include ‘Movement behind the scenes: The quiet transformation of status identification in Japan’ (with Kikkawa) (Social Science Japan Journal, 2019). Katharina Hülsmann is a PhD Candidate at Heinrich Heine University Düsseldorf where she previously taught Japanese Culture at the Department of Modern Japanese Studies. Her main research interests include Japanese popular culture, representations of gender and sexualities and fan cultures. She recently co-edited the volume Japanische Populärkultur und Gender [Japanese popular culture and gender] (Springer VS, 2016). Jun Imai is Professor of Sociology at Sophia University in Tokyo. Before joining Sophia, he gained a PhD from Stony Brook University and taught at the University of Duisburg-Essen, Tohoku University and Hokkaido University. His major field of research is the development of employment relations and its consequences on social inequalities. He recently published ‘Struggling men in emasculated life-course: Non-regular employment among young men’ in Heinrich and Galan (eds.) Being young in super-ageing Japan (Routledge, 2018). Hanno Jentzsch is Assistant Professor (postdoc) in East Asian Studies at the University of Vienna. His research interests include central-local relations, agricultural politics and the welfare regime in Japan. He holds a PhD from the University of Duisburg-Essen and has recently published on the local origins of national farmland consolidation policies (Social Science Japan Journal, 2017) and regional revitalisation in Yamanashi Prefecture (Routledge, 2020). He is the co-editor of a volume on the socio-politics of sub-national spaces in Japan (Routledge, forthcoming). Aya H. Kimura is Professor of Sociology at University of Hawai‘i-Mānoa. She received her PhD from the University of Wisconsin-Madison and her research interests include agrifood, sustainability and gender. Her book Radiation brain moms and citizen scientists: The gender politics of food contamination after Fukushima (Duke University Press, 2016) was the winner of the Rachel Carson Award from the Society for Social Studies of Science. She recently published Science by the people: Participation, power, and the politics of environmental knowledge (Rutgers University Press, 2019) with Kinchy. Emi Kinoshita is Research Assistant (postdoc) at the Faculty of Education, Leipzig University, where she also obtained her PhD. Her main research interests include pedagogical professionalism and biographical approaches in the field of comparative education, history of education and qualitative teaching research. She recently published ‘Modern education and biography’ (Comparative education: bulletin of the Japan Comparative Education Society, 2010, in Japanese) and ‘Reception of lesson study (jugyō kenkyū) abroad’ (Annual Journal of the Asian Cultures Research Institute, 2020).

484

Notes on contributors Susanne Klien is Associate Professor at Hokkaido University. She obtained her PhD from the University of Vienna and her main research interests include the appropriation of local traditions, demographic decline and alternative forms of living and working in post-growth Japan. She recently published Urban migrants in rural Japan: Between agency and anomie in a postgrowth society (State University of New York Press, 2020). Nora Kottmann is a Senior Research Fellow at the German Institute for Japanese Studies (DIJ) in Tokyo and received her PhD from the Heinrich Heine University in Düsseldorf. Her research interests include the interrelation of intimacy and space/mobility, personal relationships, gender, (not) belonging and lifestyles abroad. She is author of Marriage in Japan: Romantic and solidary relationship worlds in flux (Springer VS, 2016, in German). Recent publications include book chapters on relationship worlds of unmarried individuals (2019) and on Japanese women working abroad (2020). Gracia Liu-Farrer obtained her PhD from the University of Chicago and is Professor of Sociology at the Graduate School of Asia-Pacific Studies and Director of the Institute of Asian Migrations at Waseda University. Her main research interests include cross-border migration and social mobility in Asia and Europe. She recently edited (with Yeoh) the Routledge handbook of Asian migrations (Routledge, 2018) and published the monograph Immigrant Japan: Mobility and belonging in an ethno-nationalist society (Cornell University Press, 2020). Patricia L. Maclachlan holds a PhD in Political Science from Columbia University. She is Professor of Government and Mitsubishi Heavy Industries Professor of Japanese Studies at the University of Texas. She currently conducts research on the political economy of Japanese agriculture. She has published Consumer politics in postwar Japan: The institutional boundaries of citizen activism (Columbia University Press, 2002) and The people’s post office: The history and politics of the Japanese postal system, 1871–2010 (Harvard University Asia Center, 2011). Wolfram Manzenreiter is Professor of Japanese Studies at the Department of East Asian Studies, University of Vienna where he also obtained his PhD. His main research interests include physical culture, happiness, mobilities and the diversity of lifeways in the peripheries of a globalised world. He recently published a co-edited volume with Lützeler and Polak-Rottmann on Japan’s new ruralities: Coping with decline in the periphery (Routledge, 2020). Kenneth Mori McElwain is Professor of Comparative Politics at the Institute of Social Science, University of Tokyo. His research focuses on comparative political institutions and, most recently, on differences in constitutional design across countries. His work has been published in a number of edited volumes and journals, including American Journal of Political Science, Journal of East Asian Studies, Social Science Japan, Chūō Kōron and the Journal of Japanese Studies. Levi McLaughlin is Associate Professor at the Department of Philosophy and Religious Studies, North Carolina State University. He is co-author of Kōmeitō: Politics and religion in Japan (IEAS Berkeley, 2014) and author of Soka Gakkai’s human revolution: The rise of a mimetic nation in modern Japan (University of Hawai‘i Press, 2019).

485

Notes on contributors Chris McMorran holds a PhD from the University of Colorado and is Associate Professor of Japanese Studies at the National University of Singapore. He is a cultural geographer of contemporary Japan with research interests in tourism, gendered labour, mobilities and the geographies of home. He co-edited Teaching Japanese popular culture (Association for Asian Studies, 2016) and co-produces Home on the Dot, a podcast about the meaning of home in Singapore through the lives of NUS students. Caitlin Meagher is Visiting Assistant Professor of Anthropology at Skidmore College in Saratoga Springs, New York. She obtained a DPhil from the University of Oxford and her main research interests include changes to the home and family in contemporary Japan. Her book Inside a Japanese sharehouse: Dreams and realities is forthcoming (Routledge, 2020). Lynne Y. Nakano is Professor at The Chinese University of Hong Kong where she has been teaching since 1995. Her main research interests include gender, family, disability and experiences of marginality. She holds a PhD from Yale University, is the author of Community volunteers in Japan: Everyday stories of social change (Routledge, 2004) and has written on the experiences of single women in Shanghai, Hong Kong and Tokyo. She is currently conducting research on special education in Japan. Scott North obtained a PhD from the University of California at Berkeley and is a sociologist of family life and labour. His publications on Japan include studies of work styles, overwork and karoshi, fatherhood, leisure, the gendered division of household labour and reforms to employment and labour law. He was Professor of Sociology in the Graduate School of Human Sciences at Osaka University from 2002 to 2021. Robin O’Day is Assistant Professor of Anthropology at the University of North Georgia. He holds a PhD from the University of British Columbia and his main research interests include Japanese social movements. He recently co-authored the chapter ‘Mass media representations of youth social movements in Japan’ with Slater and Uno in Social movements and political activism in contemporary Japan edited by Chiavacci and Obinger (Routledge, 2018). Kaori Okano obtained her PhD from Massey University and is a Professor at La Trobe University. Her research interests include education, social justice and the politics of difference, multiculturalism, gender, life course, sociolinguistic variations, ethnography and longitudinal designs. Her recent publications include Nonformal education in civil society (Routledge, 2016), Discourse, gender and shifting identities in Japan (Routledge, 2018), Rethinking Japanese Studies: Eurocentrism and the Asia-Pacific regions (Routledge, 2018) and Education and social justice in Japan (Routledge, forthcoming). Robert J. Pekkanen is Professor at the Henry M. Jackson School of International Studies at the University of Washington. His research interests lie in electoral systems, political parties, interview methods and nonprofits or civil society. He has published articles in Political Science journals such as The American Political Science Review and The Journal of Asian Studies. He has published ten books in English on electoral systems, American nonprofit advocacy, Japanese civil society and Japanese elections and political parties.

486

Notes on contributors Saadia M. Pekkanen is the Job and Gertrud Tamaki Endowed Professor at the University of Washington in Seattle. She works at the intersection of international relations and international law, specialising in the economic, legal, and security policies shaping the space industry. She also investigates contemporary geopolitical change through the lens of infrastructure investment relations among countries. Her regional expertise is in the foreign affairs of Japan and Asia. She is interested in the teaching and practice of qualitative research methods and is working to extend them to big data studies in the Social Sciences. Theresia Berenike Peucker obtained her Doctorate from Martin Luther University Halle-Wittenberg. As a trained librarian, she has been serving at the Campus Library of Freie Universität Berlin since 2012. She manages the Japanese studies collection and assists with library and archive searches and academic writing. Her research interests include memory politics as well as information retrieval and information literacy. Isabelle Prochaska-Meyer is an Assistant Professor (postdoc) at the Department of East Asian Studies at the University of Vienna where she also obtained her PhD. Her main research interests include religion (especially religion in Okinawa), rural Japan and ageing. She recently published the documentary movie (co-directed by Pia R. Kieninger) ‘65+. Being old in rural Japan’ (2014). Cornelia Reiher is Professor of Japanese Studies at Freie Universität Berlin and holds a Doctorate from Leipzig University. Her main research interests include rural Japan, food, globalisation and Science and Technology Studies. Her recent publications include a special issue on Fieldwork in Japan: New trends and challenges (2018) and book chapters on transnational protest movement(s) against preferential trade agreements in Asia (2019), the governance of radionuclides in food in post Fukushima Japan (2020) and urban-rural migration (2020). Nancy Rosenberger is Professor Emeritus at Oregon State University. Her main research interests in Japan are women’s changing lives, organic farmers and rural revitalisation. Her publications include Gambling with virtue (University of Hawai‘i Press, 2000) Dilemma of adulthoods (University of Hawai‘i Press, 2013) and ‘Young organic farmers in Japan: Betting on lifestyle, locality, and livelihood’ in Contemporary Japan (2017). Richard J. Samuels is Ford International Professor of Political Science and Director of the Centre for International Studies at the Massachusetts Institute of Technology. In 2015–2019 he was Albert Einstein Visiting Fellow at the Freie Universität Berlin, where he completed his most recent book Special duty: A history of the Japanese intelligence community (Cornell University Press, 2019), named by the Council on Foreign Relations’ journal Foreign Affairs one of the ‘Best of Books, 2019’. Annette Schad-Seifert is Professor of Japanese Studies at Heinrich Heine University Düsseldorf and holds a PhD from Freie Universität Berlin. She has served as Vice President of the German Association of Social Science Research on Japan and Specially Appointed Professor at the Institute for Gender Studies at Ochanomizu University. Her major research areas are Japanese society, demographic change, family policy and gender issues. She recently published Family life in Japan and Germany: Challenges for a gender-sensitive family policy (Springer, 2019).

487

Notes on contributors Katja Schmidtpott is Professor of Japanese History at Ruhr-Universität Bochum. She specialises in the social and economic history of modern Japan, and her main research interests include urbanisation, modernisation of everyday life and the economic relations between Japan and (West) Germany. She recently published The East Asian dimension of the First World War: Global entanglements and Japan, China and Korea, 1914–1919 (Campus, 2020, co-edited with Schmidt). Tino Schölz obtained a PhD from Martin Luther University Halle-Wittenberg and is a Postdoctoral Research Fellow at Freie Universität Berlin. His main research interests include the political and military history of modern Japan. His publications include ‘Die Gefallenen besänftigen und ihre Taten rühmen’: Gefallenengedenken und politische Verfasstheit in Japan seit der Mitte des 19. Jahrhunderts (De Gruyter Oldenbourg, 2016). Kai Schulze is a Postdoctoral Researcher at Freie Universität Berlin where he also obtained his PhD. His main research interests include foreign policy analysis, interstate rivalry, cross-regional relations, Japan's foreign and security policy, Sino-Japanese relations and East AsiaMiddle East relations. He recently co-edited a special issue on Dimensions of Sino-Japanese rivalry in a global context in The Pacific Review (2019) (with Blechinger-Talcott). Kay Shimizu holds a PhD from Stanford University and is a Research Assistant Professor in the Department of Political Science at the University of Pittsburgh. Her research focuses on comparative politics, especially fiscal and financial politics, of Japan and China. Her publications include Cultivating institutional change: Economic liberalization, demographic decline, and the reform of Japan Agricultural Cooperatives (forthcoming, co-authored with Maclachlan), as well as articles in Socio-Economic Review, Journal of East Asian Studies, Current History and Social Science Japan Journal. Karen Shire holds the Chair for Comparative Sociology and Japanese Society at the Institutes of Sociology and East Asian Studies at the University of Duisburg-Essen. Her research compares the transnationalisation of labour in Asia and Europe. She holds a PhD from the University of Wisconsin-Madison. Recent publications include the Transnationalization of labour (Springer, 2018, co-editor), ‘The social order of migration markets’ (Global Networks, 2020), and ‘Who are the fittest? The question of skills in national employment Systems’ (Journal of Ethnic and Migration Studies, 2020, co-author). David H. Slater is a Professor of Cultural Anthropology at Sophia University, Tokyo. He has recently worked on the practice and politics of post 3.11 nuclear disasters in northern Japan and runs the Voices from Japan project. He is now researching homeless and refugee populations in Tokyo. Celia Spoden is a Research Associate (postdoc) at the Institute for History, Ethics and Philosophy of Medicine at Hannover Medical School and holds a PhD in Japanese Studies from Heinrich Heine University Düsseldorf. Her research focuses on bioethical questions using a qualitative approach. She recently published ‘Deciding one’s own death in advance: Biopower, living wills, and resistance to a legislation of death with dignity in Japan’ (Contemporary Japan, 2020).

488

Notes on contributors Brigitte Steger is a Senior Lecturer at the University of Cambridge. She studies the social and cultural embeddedness of seemingly natural, bodily matters and daily life such as sleep, time, cleanliness, waste disposal and shelter life. Steger holds a PhD in Japanese Studies and recently published ‘The stranger and others: Life and legacy of the ethnologist Oka Masao’ (Vienna Journal of East Asian Studies, 2019) and Beyond kawaii: Studying Japanese femininities at Cambridge, co-edited with Koch and Tso (Lit, 2020). Nicolas Sternsdorff-Cisterna is an Assistant Professor of Anthropology at Southern Methodist University. He holds a PhD from Harvard University and his main research interests include the study of food, risk, science and technology. He is the author of Food safety after Fukushima: Scientific citizenship and the politics of risk (University of Hawai‘i Press, 2019). Christian Tagsold is Professor at the Department for Modern Japan at Heinrich Heine University Düsseldorf. He has been researching Japanese gardens in Europe and the USA since 2006. His book Spaces of translation: Japanese gardens in the West (Pennsylvania University Press, 2017) was awarded with the Abbott Lowell Cummings Award by the Vernacular Architecture Forum in 2019. His other research interests include the aging society in Japan, the Tokyo Olympics 1964/2020 and the Japanese diaspora in Düsseldorf. Akiko Takeyama is an Associate Professor of Women, Gender, and Sexuality Studies at the University of Kansas. Her research interests include changing gender, sexuality and class dynamics in contemporary Japan. Her first book, Staged seduction: Selling dreams in a Tokyo host club (Stanford University Press, 2016) was a finalist for the 2017 Michelle Rosaldo Book Prize (Association of Feminist Anthropology). She is currently working on her second book project Involuntary consent: Sexual labor and violence in Japan’s adult video industry. Katrin Ullmann is a trained systemic counsellor and works at the University of Applied Sciences in Düsseldorf. She obtained her PhD from Heinrich Heine University Düsseldorf, where she worked at the Institute for Cultural Science and Media. Her research interests include generational and family research, Global and Mobility Studies, as well as counselling research that integrates aspects of Cultural Studies. She recently published the monograph Generationscapes: Conceptual framework and empirical findings of a global generation (Transcript, 2017). Gabriele Vogt holds the Chair of Japanese Studies at Ludwig-Maximilian-Universität München. Her main research interests include international labour migration and demographic change, local politics, social movements and Okinawan Studies. She holds a PhD from the University of Hamburg and has recently published Population aging and international healthcaregiver migration to Japan (Springer, 2018) and co-authored the article ‘Identity politics in Okinawan elections: The emergence of regional populism’, with Hijino (Japan Forum, 2019).

489

Notes on contributors Cosima Wagner has a PhD from Goethe University Frankfurt and is academic librarian at Freie Universität Berlin, serving as liaison to the East Asian Studies faculty with a special focus on Digital Humanities, Research Data Management and Open Science. Her research interests include a Science and Technology Studies approach to library infrastructure management, multilingualism and non-Latin scripts in the digital space, Area Studies librarianship as well as critical algorithm studies and social robotics in Japan Monograph: Robotopia Nipponica. Recherchen zur Akzeptanz von Robotern in Japan (Tectum, 2013). Daisuke Watanabe is an Associate Professor at Seikei University. His main research interests include the Sociology of aging, health and life course. He recently published Sōchūryū no hajimari: Danchi to seikatsu jikan no sengoshi [The rise of all-middle-class society: Postwar history of public housing complex and time-use] (Seikyūsha, 2019, co-edited with Aizawa) and ‘Older adults’ integration in the labour market: A global view’ in Ageing and society (2019). Daniel White is a Senior Researcher in the Graduate School of East Asian Studies at Freie Universität Berlin and a visiting scholar in the Department of Social Anthropology at the University of Cambridge. He obtained a PhD from Rice University and conducts work on the intersections of politics, technology and affect, with a current research emphasis on artificial emotional intelligence in Japan and the U.K. His publications and research projects can be found at www.modelemotion.org. Anna Wiemann is Assistant Professor (postdoc) at Ludwig-Maximilians-Universität München since 2019. Her main research interests include social movements, social network theory, Japan’s foreign and domestic policy and collective memory. Her dissertation titled Networks and mobilization processes: The case of the Japanese anti-nuclear movement after Fukushima was published in 2018 by Iudicium. Tomiko Yamaguchi is a Professor of Sociology at International Christian University in Tokyo. She studies social controversies surrounding safety, such as issues related to GMOs and to hazards such as radioactive nuclides in food. Her current project deals with discourses surrounding the use of gene editing technologies. She has published numerous articles in journals like Science, Technology and Human Values and Food Policy. Her most recent publication is a book entitled Simulation, prediction and society: The politics of forecasting (2019). Akiko Yoshida holds a PhD from the University of Oklahoma and is an Associate Professor of Sociology at the University of Wisconsin-Whitewater. Her main research interests include singlehood, gender role views and parental involvement in childrearing. She recently published a monograph entitled Unmarried women in Japan: The drift into singlehood (Routledge, 2017) and a refereed article, ‘Gender role attitudes: An examination of cohort effects in Japan’ (Journal of Marriage and Family, 2019, co-authored with Piotrowski et al.).

490

Notes on contributors Urs Matthias Zachmann is Professor of Modern Japanese History and Culture at Freie Universität Berlin. Prior to that, he was the inaugural Handa Professor of Japanese-Chinese Relations at the University of Edinburgh. His main research areas are the cultural, intellectual, legal and diplomatic history of modern Japan within the East Asian regional context. He is the author of China and Japan in the late Meiji period: China policy and the Japanese discourse on national identity (Routledge, 2009) and Völkerrechtsdenken und Außenpolitik in Japan, 1919– 1960 (Nomos, 2013).

491

Index abstract 36, 78, 85–87, 153, 177–179, 186, 242, 278, 300, 303, 341, 351, 355, 357, 371, 378, 386, 396, 415, 416, 421, 428, 445, 459, 468

Area Studies 36, 40, 42, 78, 109, 112, 114, 117, 121, 122, 238, 260, 264, 283, 334, 360, 367, 369, 377, 408, 416, 417, 436, 437, 440, 456, 458

academic career 72, 75, 114

audience – academic 124, 160, 161, 330, 455, 456 – broader/wider 174, 444, 446, 461–463, 470, 473 bias 96, 110, 175, 221, 230, 240, 265, 274, 277, 307, 319, 320, 327, 365

actors 35, 41, 46, 62, 121, 122, 147, 175, 179, 184, 227, 231, 232, 243, 253, 286, 293, 294, 301, 304, 345, 349, 354, 364, 365, 371, 378, 381–383, 385–387, 390, 433, 470 affect 30, 39–42, 66, 71, 137, 167, 211, 219, 221, 232, 234, 343, 360, 393–396, 416, 432, 439 age 31, 32, 55, 65, 88, 133, 136, 140–142, 187, 192, 219, 238, 239, 271, 274–276, 289, 326, 339, 344, 390, 434, 465, 473, 475 agency 33, 35, 36, 147, 148, 157, 232, 257, 287, 310, 354, 386, 387, 409, 416, 445 ambiguity 124, 319, 400 American Anthropological Association (AAA) 112, 159, 217, 430 analysis – comparative 83, 303, 316 – qualitative 71, 306, 308, 312, 352, 356 – quantitative 278, 284, 285, 289, 315, 318, 349 – secondary 259, 260, 273, 277 – thematic 85, 302 approach(es) – actor-centred 62 – analytical 301, 303 – interpretative 374 – positivist 229, 353 archive(s) 80, 118, 119, 157, 238, 242–255, 259, 260, 274, 361, 391, 423, 440, 446–449, 464, 465 – business 248–250 – data 243, 259, 260, 274, 440, 465 – newspaper 391 – private 251 – public 238, 244–247, 449

blog(s) 232, 349, 354, 421, 440, 441, 456, 461, 462, 475 buisiness card(s) (meishi) 56, 107, 162, 165, 189, 202, 216 buzzword(s) 73, 385, 389–391, 397, 398, 400, 401, 418 Call for Papers (CfP) 457 CAQDAS (computer assisted qualitative data analysis software) 301, 305–307, 359 CARE principles 440 career 46, 72, 75, 87, 102, 114, 127–129, 139, 229, 264, 272, 287, 332, 337, 377, 405, 422, 439, 450, 451, 455, 456, 460, 462–465, 475 case selection 68, 71 case study, case studies 37, 54, 60, 78–82, 85–87, 91, 95, 97, 98, 122, 125, 127, 128, 283, 285–287, 290, 318, 336, 365, 408, 409, 416, 470 – multiple 79, 82, 85 – single 82, 85, 87 citation(s) 118, 193, 194, 241, 311, 361, 414, 438 co-author(s) 71, 95, 174, 246, 323, 361, 416 code(s) 86, 212, 233, 251, 303–305, 309, 315, 324–332, 335–337, 339–342, 344, 345, 352, 355, 358, 359, 361, 364, 365, 372, 393, 436, 465 – key 328, 329, 331, 332

493

Index coding 267, 301, 305, 323–336, 339–341, 343–346, 349–351, 353, 355, 357–366, 434, 464 – axial 339 – focused 325, 329, 331, 334 – in vivo 324, 329, 330, 333, 344 – inductive 364, 365 – initial 323–325, 328, 329, 331, 333, 334, 339 – open 327, 328, 332, 339 – selective 323, 331, 332, 340 – substantive 324, 332 – theoretical 323, 332, 334

context – contextualisation 110, 123, 124, 169, 179, 287, 288, 355, 372 – historical 37, 295, 380, 466 – sociocultural 328, 378

collaboration 71, 124, 147, 174, 218, 281, 309, 423, 440, 448

curiosity 54, 67, 125, 212, 216 data – accuracy/uniformity of 288 – empirical 73, 86, 177, 179, 258, 270, 279, 301, 341 – OECD 146, 244, 258, 273 – qualitative 74, 79, 132, 134, 141, 177, 265, 267, 268, 277, 289, 300, 301, 303, 305–309, 323, 324, 328, 344, 345, 359, 363, 365, 366, 436 – quantitative 45, 63, 79, 96, 97, 137, 256, 259, 264–268, 271, 277–280, 293, 310, 311, 319, 323, 328, 361 – research 135, 218, 310, 434–437, 439, 440 – sharing 436 – storing 164, 436 – UN 146, 258 – World Bank 258

communities 35, 36, 44, 46, 91, 110, 112, 126, 128, 152, 157, 165, 166, 190, 194, 217, 226, 233, 258, 281, 330, 343, 344, 433, 440, 459, 470 – academic 110, 157, 233, 459 – online 165 – scientific 281 comparison 36, 73, 89, 90, 144, 148, 151, 153, 198, 248, 256, 267, 268, 288, 345, 352, 369, 423, 460 – comparative studies 95 concept development 355 conference(s) 36, 58, 71, 112, 113, 151, 158, 175, 200, 256, 312, 415, 440, 456, 457, 461, 466–469, 473 consent 81, 185, 193, 198, 217, 228, 277, 311, 352, 400, 430, 432, 433, 436, 447, 448, 471 – informed 228, 311, 352, 430, 432, 436 – written 193, 432, 433, 447, 448 constructivism 302 content analysis 81, 134, 135, 303, 305, 306, 318, 319, 349–355, 357, 359, 360, 362–366, 368, 371–374, 380, 383, 386 – qualitative 349, 350, 353–355, 357, 360, 362–366, 371, 372, 374 – quantitative 349, 359, 363

494

contingency 97, 124 continuum 85, 86, 185, 303 copyright 239, 361, 440, 441, 476 counselling 220, 412, 413 court record(s) 251 cultural baggage 31 cultural translation 342, 442, 444, 445

data analysis 58, 59, 63, 134, 194, 267, 268, 277, 278, 285, 286, 300–309, 314, 319, 320, 337, 353, 366, 368, 373, 381, 438 data collection 58, 59, 63, 81, 84, 94, 132, 134–138, 141, 179, 266–268, 273, 274, 277, 281, 300–307, 323, 324, 344, 349, 351–353, 358, 361, 362, 368, 381, 385, 387, 428, 430, 431, 434 debate(s) 30, 40–42, 57–59, 64, 67, 73, 74, 85, 102–104, 112–114, 125, 129, 217, 292, 341, 342, 349, 356, 369, 390, 398,

Index 399, 407, 409, 415, 418, 431, 436, 437, 445, 463 deduction/deductive 63, 79, 85, 86, 292, 303, 305, 316, 357, 359, 364, 365 development of theoretical models 301 disaster 44, 66, 188, 194, 224, 330, 343–345, 354, 364, 377, 379, 382, 387, 408, 423, 428, 432, 433, 446, 447, 449 disciplines 32, 36, 37, 42, 58, 61, 65, 78, 91, 93, 110, 114, 125, 185, 195, 200, 265, 312, 323, 349, 361, 415, 429, 434, 435, 446, 448, 456, 461 – disciplinary conventions 185, 450 discourse 73, 74, 81, 84, 106, 121, 123, 124, 179, 287, 288, 303, 315, 317, 318, 320, 340, 341, 350, 354, 368, 369, 371, 373, 377–392, 394, 395, 397, 398, 401, 423, 444, 451 – counter-hegemonic 387 – economic 317 – political 371, 377, 380, 397 – popular 288 – threads of 390 discovery system 240 dissertation 29, 41, 44, 92, 102–105, 108, 113, 115, 116, 121, 132, 177, 179, 285, 313, 314, 343, 345, 367, 405–408, 412, 413, 430, 434, 441, 460, 463, 466 document analysis 81, 292 documentation 250, 251, 289, 352, 434, 436 editor(s) 56, 63, 117, 174, 355, 360, 362, 385, 398, 417, 437, 450, 451, 459, 460, 462, 465–469, 475 efficiency 134, 318, 446 elevator pitch 38, 56, 57, 158, 160 engaged scholarship 124, 387 ethical concern(s) 134, 157 ethical principle(s) 167, 434 ethical review 429, 431

ethics 81, 137, 165, 193, 198, 217, 233, 290, 310, 428–430, 432, 433, 439, 442, 449, 450 – fieldwork 165, 428, 430, 432, 433 – research 290, 428, 439, 442, 450 ethnicity 32, 55, 139–142, 377 ethnography 66, 92, 94, 134, 135, 157, 173, 176, 214, 219, 221, 227, 229, 230, 266, 323, 335, 354, 407, 416, 463 – affective 219, 221, 227, 229, 230 – multi-sited 92, 176 – virtual 214 – visual 94, 134, 135 evidence 42, 70, 74, 78, 81, 85, 127, 148, 185, 255, 284, 285, 300, 314–316, 331, 344, 346, 351, 352, 354, 355, 358, 399, 434, 445, 451, 459 expressions (of interviewees) 123, 133, 151, 184, 190, 191, 193, 194, 225, 226, 258, 311, 319, 330, 340, 341, 344, 369, 370, 396, 399 fairness 428, 439, 442 fellowship 44, 45, 405 field site 47, 92, 93, 169–171, 177–180, 214, 218, 224, 225, 344, 432, 433, 446, 448, 449 fieldnotes 49, 92, 104, 111, 112, 115, 135, 139, 162, 164, 165, 191, 193, 194, 211, 213, 220–222, 304, 311, 324, 325, 327, 343, 344, 385, 405, 414, 418, 436 – interview 414, 424 – notetaking 164, 326 – taking notes 49, 111, 112, 191, 193, 197, 198, 211, 213, 439 fieldwork 157–159, 161–166, 168–176, 178–180, 231–234, 393–396, 470–473 – long-term 91, 166, 171, 198 – multi-sited 169 – qualitative 121, 289, 290, 307, 355 film 93, 463, 464, 470–473 – documentary 463, 470, 471, 473 – ethnographic 93, 463, 470, 472 flexibility 65, 141, 169, 294, 295, 406, 420 focus group 133, 134, 186, 200, 302, 350

495

Index frame analysis 349, 350, 352–357, 359, 360, 362, 380, 383, 386, 387 framing 206, 349, 350, 355–357, 360, 389, 400, 401, 406, 414, 456, 462 – diagnostic 400 functionalism 314 gatekeeper 187, 188, 225 gender 30–32, 40, 55, 65, 79, 82, 83, 134, 136, 139–142, 187, 192, 215, 219, 225, 227, 231, 258, 271, 274, 276, 289, 294, 310, 314, 315, 325, 326, 331, 344, 389, 392, 408, 466 Gender Studies 287, 312, 389 generalisation 85 gift(s) (omiyage) 40, 41, 45, 48, 139, 163, 171, 172, 189, 196, 206, 216, 409, 410, 422 grounded theory 63, 80, 81, 86, 187, 270, 305, 323–325, 329, 331, 339, 344, 368 hegemony 379, 380 hierarchy 140, 160, 162, 172, 250, 336, 373 history – global 252–254 – life 133, 136, 201, 367, 368 – oral 142, 368 – transnational 244, 252, 254 honesty 441, 448 Human Research Ethics Committee 289 hypothesis, hypotheses 40, 60, 63, 80, 81, 86, 87, 97, 136, 186, 251, 265, 268, 270, 276, 278, 285, 314, 328, 332, 353, 440 ideology 379, 382, 386, 387, 466 inequality 33, 73, 267, 286, 292, 295, 379, 397, 400 informed choices 225 Institutional Review Board 167 interdiscursivity 386 interlocutor(s) 47, 127, 158, 159, 161–164, 166–168, 189, 191, 192, 196–199, 205, 206, 224–226, 327, 395, 432, 433, 444 (inter)personal relationships 287, 327

496

interpretation 85, 102, 105, 106, 191, 193, 229, 249, 267, 268, 292, 293, 296, 300, 309, 330, 342, 350, 354, 356, 359, 361, 363, 369, 370, 372 interviewee(s) – interview partner 191, 193 interview(s) – biographical 367, 368 – couple 133, 186 – episodic 186 – expert 178, 186, 187, 293, 294, 471 – group 133, 134, 186, 193, 225 – in-depth 133, 200, 227, 278 – narrative 186, 304, 339, 369, 449 – online 188 – qualitative 73, 132, 133, 136–138, 140, 141, 151, 184, 185, 195, 200–202, 280, 290, 301, 302, 308, 472 – questions 79, 178, 198, 233, 277, 414 – semi-standardised 186 – semi-structured 143–145, 148, 186, 287, 305, 343, 365, 464 – structured 34, 93, 94, 133, 143–145, 148, 186, 192, 287, 305, 343, 365, 464 – unstructured 34, 133, 169, 186 introduction(s) 31, 45, 48, 49, 97, 98, 107, 112–114, 118, 119, 132, 157–161, 163, 171, 187, 190, 202, 206, 211, 213, 243, 248, 250, 253, 256, 264, 267, 276, 300, 301, 307, 309, 341, 377, 406, 408, 420, 428, 431–433, 435, 462 – self- (jiko shōkai) 158, 161 iteration 332, 358, 362 Japan Studies 117, 118, 120, 128, 146, 148, 150, 323, 330, 440, 458, 464 Japanese Studies 29–31, 36–38, 40, 41, 53, 79, 105, 109–114, 121, 150, 152, 175, 239, 241, 242, 247, 260, 310, 312, 346, 407–409, 413–415, 417, 422, 425, 428, 439–442, 456, 463, 465–468 Japanology 31, 36 journalist(s) 35, 41, 187, 286, 288, 320, 341, 382, 390, 392, 461, 462, 474, 475

Index journal(s) – disciplinary 122, 458 – journal article(s) 117, 118, 241, 360, 406, 408, 412, 413, 456, 459, 468 – peer-reviewed 109, 465, 467 key documents 254

methodology 38, 39, 63, 78, 111, 173, 195, 196, 232, 286, 300, 309, 316, 323, 363, 368, 381, 422, 429, 450, 457 – cosmopolitan 78 – methodological nationalism 146 – methodological triangulation 265

knowledge production 78, 85–87, 102, 124, 169, 171, 392, 445 Latin scripts 238, 240, 241, 437 – non-Latin scripts 238, 240, 241, 437

methods – comparative 260 – interview 152 – mixed 34, 37, 97, 132, 264–270, 272, 278–282, 287, 292 – qualitative 34, 45, 63, 132, 137, 142, 179, 264, 265, 272, 283, 284, 286, 306, 309, 416 – quantitative 34, 94, 132, 223, 264, 267, 268, 270, 271, 273, 275–278, 280, 283, 284, 286, 353, 355, 362

library 43, 45, 57, 97, 106, 107, 112, 117, 118, 120, 122, 238–243, 247, 249, 251, 253, 259, 434, 439, 463 life story 133, 191, 368 literacy – digital 238, 239, 247 – information 239 literature – Japanese 48, 108, 128, 367, 368 – secondary 59, 103–106, 108, 111, 112, 117, 118, 124, 147, 284, 451 literature review 54, 57, 64, 102–104, 106, 112–117, 125–127, 129, 406, 407, 409, 438 location 47, 92, 93, 121, 138, 144, 145, 157, 167, 175, 188, 189, 472, 473 magazine(s) 120, 135, 254, 255, 287, 391, 421, 467, 474, 475 management 104, 113, 232, 233, 244, 292, 294, 359, 361, 405, 423, 428, 429, 434, 435, 437 market research company 289 MAXQDA 199, 345, 352, 359, 365 measurement levels 271 MeCab 320 mechanisms 35, 68, 109, 167, 283, 354, 444 media work 477 memorial sites 250, 251 memory 104, 164, 193, 198, 212, 220, 250, 304 mental health 166, 412, 413

mobilities 128, 146, 173–175, 214, 294, 408 mobility 33, 91, 92, 126, 146–149, 173–175, 260, 292, 294, 400, 408, 424, 471 narrative(s) 37, 66, 79–81, 103, 113–115, 122, 126, 142, 186, 191, 196, 205, 224, 225, 279, 280, 292, 295, 304, 310, 325, 327, 333–335, 339, 343, 354, 360, 361, 369, 394, 408, 418–420, 445, 446, 449, 451, 464 National Diet Library (NDL) 107, 118, 242, 257, 259, 400 native speaker 110, 158, 162, 164, 204, 308 – non-native speaker 110, 308 network(s) – social 44–46, 139, 153, 160, 178, 187, 194, 273, 350, 358, 364, 417 – systemic 315 newspaper 40, 117, 119, 120, 318, 350, 351, 355, 356, 360, 362, 391, 398–400, 414, 419, 462, 466, 474, 475 note cards 424, 425 optical character recognition (OCR) 98, 359 oral narrative 446, 449, 464

497

Index Orientalism 121, 288 originality 56, 57, 59, 60, 64 othering 234 outreach 412 participant 34, 45, 79, 88, 93, 133, 137, 138, 142–145, 159, 163, 169, 170, 200, 204, 205, 211–214, 216–221, 223, 227, 228, 286, 287, 294, 302, 311, 314–316, 325, 327, 343, 385, 414, 442, 471 patience 54, 116, 118, 143, 188, 211, 213, 215 personality 138, 141, 142, 190, 197, 206, 207, 217–220

publishing 36, 109, 158, 280, 391, 430, 437, 439, 441, 442, 444, 449, 455, 456, 460, 463, 465–468 – open access 119, 243, 435, 439, 440 Python 319, 320 quality 57, 66, 69, 104, 108–110, 113, 120, 154, 162, 163, 194, 199, 241, 281, 301, 302, 307–309, 316, 364, 412, 416, 417, 422, 429, 450, 459, 462, 466, 467, 474, 475, 477 questionnaire(s) 34, 79, 133, 184, 188, 190, 196, 198, 274, 276, 277, 279, 335, 434 race 31, 61, 65, 140, 141, 219, 285, 408 rapport 134, 203, 212, 216, 223, 228, 230

perspective(s) – comparative 41, 70, 73 – historical 293, 423

reasoning 63, 133, 136, 143, 196

plagiarism 111, 430, 431, 437–439

recorder 50, 163, 193, 194, 205, 206 – recording device 164, 189–191

politics 30, 31, 40–42, 45, 53, 54, 68–71, 82, 96, 119, 121, 123, 153, 170, 205, 256, 258, 349, 352, 371, 382, 385, 389, 394, 397, 398, 400, 401, 412, 423 – comparative 68, 71, 349, 352 – Japanese 40, 41, 53, 54, 68, 69, 71, 371, 397, 398, 401 – party 68, 70 positionality 65, 78, 124, 140, 141, 192, 204, 207, 229, 230, 326 post-modernism 30, 302 practice – good research 54, 428–432, 435, 439, 450 – writing 104

reciprocity 172, 194, 218, 221, 432, 447

recordings 105, 162, 198, 199, 206, 304, 344, 436 reflexivity 30, 93, 124, 192, 215, 285, 301, 307, 326, 334, 342, 387 – reflexive 30, 58, 145, 171, 396 reform 40, 41, 55, 60, 68, 69, 74, 95, 177–179, 283, 284, 292, 293, 318, 385, 397–400 relevance 56, 57, 60, 64, 67, 103, 104, 114, 122, 140, 186, 187, 238, 240, 241, 256, 327, 328, 331, 381, 385, 416, 417, 449 – sociological 416 reliability 137, 274, 276, 280, 307, 361, 364, 439, 441

pragmatism 265, 304

repository 151, 273, 435, 441

process tracing 286, 354

representation(s) 63, 69, 105, 111, 234, 243, 275, 315, 337, 371, 373, 383–385, 387, 389, 393, 394, 398, 436, 444

public opinion 70, 71, 385, 399 public space 231, 248, 250, 251 publication 73, 109, 240, 241, 247, 256, 280, 308, 323, 367, 372, 429, 439, 440, 446, 449, 450, 455, 457, 460–462, 465–468

498

research – action 134, 135, 142 – archival 132, 134, 135, 169, 244, 247, 248, 255, 351 – assistant(s) 164, 175, 192, 290, 352, 361, 465 – budget 120, 145

Index – comparative 44, 70, 71, 95, 169 – design 54, 58, 61, 64, 73, 74, 78, 84, 90, 93, 95–98, 111, 146–148, 239, 264, 270, 272, 273, 278–281, 319, 344, 349, 353, 355, 365, 377, 380, 387, 428, 439, 446, 447 – empirical 40, 42, 323, 367, 434 – engaged 446–448 – exploratory 54, 136 – gap 57, 72, 122, 301 – multi-method (MMR) 283, 355 – observational 132, 135, 137, 141 – participatory 204 – process 59, 75, 84, 86, 117, 118, 194, 217, 218, 250, 266, 267, 278–281, 300, 301, 303, 306, 307, 333, 349, 357, 363, 367, 373, 381, 410, 419, 428–431, 433, 434, 438, 439, 441, 455 – puzzle 29, 38 – qualitative 34, 55, 63, 125, 128, 129, 132, 135–137, 141, 144, 146, 150, 184, 187, 200, 256, 264, 267, 273, 277, 279, 281, 289, 290, 300–302, 305–309, 323, 352, 357, 359, 360, 364, 366, 368, 419, 436 – quantitative 34, 55, 63, 135, 136, 147, 153, 264, 266, 267, 270, 275, 278, 280, 289, 305, 307, 308, 319, 419 – question 53, 54, 56–65, 67, 72–75, 78, 81, 84, 103, 114, 137, 138, 141, 184–188, 191, 200, 215, 256, 265, 267, 270, 272, 273, 276, 281, 283–286, 296, 301–304, 306, 342, 349, 350, 352, 353, 355, 357, 358, 360, 362–366, 377, 379, 381–383, 385, 431, 439, 459 – survey 271, 355 – team 147, 148, 280, 430, 446, 447 – topic 29, 40, 59, 68, 78, 97, 105, 121, 185, 187, 190, 215, 301, 326, 329, 343, 381, 382, 384, 385, 431, 437, 461 – traditions 267, 302 – transnational 146, 147, 150 respect 138, 140, 151, 152, 160, 202, 233, 234, 244, 251, 286, 315, 363, 367, 389, 428, 431, 436, 441, 443, 444

responsibility 124, 129, 172, 194, 217, 218, 233, 337, 428–431, 445, 447, 448 review – ethical 429, 431 – literature 54, 57, 64, 102–104, 106, 112–117, 125–127, 129, 406, 407, 409, 438 – peer 57, 109, 241, 440 rigour 54, 56, 58–60, 64, 110, 124 salience 68, 70, 288, 398, 399 sampling 71, 91, 96, 174, 175, 187, 188, 232, 268, 274–277, 279, 281, 339, 342, 349, 352, 353, 355, 359, 361, 362 – convenience 276 – non-probability 276 – sample size 137, 267, 275, 277, 355 – snowball 187, 232, 276 saturation 296, 302, 305, 325, 331, 332, 339 – theoretical saturation 325, 331, 339 scholarship in Japanese 129, 465 science 102, 114, 124, 212, 246, 265, 314, 381, 387, 416, 417, 435, 439, 440 – open 246, 435, 439, 440 Scrivener 420 self-reflexivity 342 seminar (zemi) 113, 158 serendipity 47, 48, 84, 91–94, 169, 215 silence 191, 193, 197 social change 61, 62, 72, 73, 258, 294, 468 social media (platforms) 93, 138, 144, 157, 159, 162, 165, 166, 188, 194, 221, 310, 332, 351, 354, 357, 410, 433, 440, 461, 465 source(s) – information 178 – primary 102–106, 119, 122–124, 165, 251, 284, 371, 372 – secondary 106, 110, 117, 118, 120, 122–124, 238–240, 243, 244, 251, 285, 286, 288 statement of support 157

499

Index statistics 79, 88, 108, 110, 137, 146, 147, 178, 228, 243, 256–259, 273, 277, 279, 308, 320, 355 – descriptive 277, 355 – official macro 259

test(s) 55, 62, 63, 66, 68–71, 78–80, 85–87, 95, 98, 136, 186, 198, 199, 229, 250, 267, 268, 270, 271, 277, 278, 285, 286, 312, 328, 331, 353, 364, 382, 419, 445, 472

Statistics Bureau 259, 273, 274

text analysis 200, 355, 359, 377, 386

stereotypes 215, 216, 234, 475, 476

text mining 319

stories 93, 98, 119, 133, 135, 136, 142, 152, 173, 175, 184, 185, 191, 192, 195, 196, 201, 202, 204, 225, 336, 349, 367, 410, 418–421, 432, 444, 448, 471, 474–476

text(s) – elicited 325, 326 – extant 324–327

story 54, 59, 60, 93, 120, 133, 139, 173–175, 191, 203, 214, 218, 253, 335–338, 356, 368, 369, 418–421, 446, 470, 471, 474–476 storyline 119, 418, 420 strategies 41, 43, 53, 64, 66, 106, 107, 109, 116, 137, 140, 157, 167, 185, 187, 188, 197, 215, 254, 307, 309, 318, 323, 325, 329, 353, 362, 380, 383, 386, 387, 394, 401, 416, 436 – rhetorical 394 stress 56–59, 113, 186, 197, 199, 204, 272, 312, 373, 387, 409, 412, 441, 464 study – descriptive 54, 179 – ethnographic 81–83, 314 – longitudinal 83, 271 – multiple case 79 – panel 79, 83 – qualitative 97, 268 – single case 82, 85, 87 – wave 79, 83 subcode(s) 339 subcultures 136, 310 survey 46, 71, 72, 108, 117, 126, 133, 151, 153, 198, 252, 257, 259, 260, 264, 266–268, 271–274, 276–280, 289, 290, 294, 351, 355, 398 – national 289 – original 274, 280 symbolic interactionism 380 tabulation 349

500

theme(s) 85–87, 93, 133, 224, 258, 268, 288, 290, 302, 304–306, 325–333, 337, 338, 344, 352, 390, 394, 407, 411, 414, 446, 449, 456 – theme maps 337, 338 theoretical concept(s) 186, 328, 339, 340, 351, 368, 370–372, 444 theorising 40, 315, 316, 323, 324, 341, 354, 444 theory building 55, 79, 85–87, 266, 334 theory testing 85, 86 theory (theories) 33–35, 37, 45, 53, 68, 69, 71, 72, 80, 81, 85–87, 89, 95, 98, 104, 121, 125–128, 211, 213, 215, 266, 270, 303, 305, 308, 313, 323, 334, 339, 341, 353, 354, 357–359, 367–369, 377–381, 388, 407, 409, 416–418, 444, 476 – grounded 63, 80, 81, 86, 187, 270, 305, 323–325, 329, 331, 339, 344, 368 thick description 40, 212, 360 time – frame 272, 290 – management 113, 405 timeliness 67, 74 timing 188, 201, 267, 339, 341, 476 tool(s) – digital 239, 312, 345, 451 – reference 238, 242, 247 transcription(s) 174, 184, 193, 199, 240, 241, 280, 281, 290, 293, 301, 304, 327, 352, 365, 369, 411 – transcribing 193, 194, 352, 359, 365, 464

Index transcultural 310, 367, 369, 370

verbal expression(s) 133, 151, 184, 190, 191, 193, 194 – non-verbal expression(s) 133, 191, 193, 194

translation(s) 111, 162, 174, 216, 240, 255, 277, 280, 281, 330, 339, 341, 342, 354, 358, 367, 369, 419–421, 428, 442, 444, 445

verifiability 124, 446

transnational 92, 127, 137, 146, 147, 150, 190, 219, 244, 252–255, 358, 384, 463

video(s) 134, 135, 138, 162, 188, 193, 304, 344, 345, 392, 434, 446–448, 463–465

transparency 124, 218, 239, 280, 359, 361, 366, 434, 436, 441, 442

visual image(s) 135

triangulation 132, 265, 266, 283, 288, 292, 293, 295 trust 63, 80, 141, 145, 152, 157, 159, 171, 172, 187, 194, 198, 216, 217, 228, 233, 234, 312, 406, 430–433, 442, 443, 448, 470 turn(s) – linguistic 418 – spatial 122 validity 124, 137, 280, 283, 307, 316, 361, 364, 371

vulnerability 432 White Paper 256–258 work-life balance 258, 318, 405, 412, 413 workshop(s) 58, 71, 112, 113, 359, 367, 370, 432, 456, 457, 467, 469 writing 227, 229, 230, 252, 405, 406, 408–418, 420–422, 424, 425, 450, 451 writing group(s) 411, 415, 469 writing habits 405, 410 writing routines 410

variables 31, 32, 40, 55, 63, 67, 85, 267, 270, 271, 274, 276, 294, 338, 354, 419 – independent 271

501