Categorising Meat Alternatives: How Dominant Meat Culture Is Reproduced and Challenged Through the Making and Eating of Meat Alternatives

Home , Chipolata, Grain, Hain Celestial Group, Linda McCartney, Mock duck, Nut roast

Categorising Meat Alternatives: how dominant meat culture is reproduced and challenged through the making and eating of meat alternatives

A thesis submitted to the University of Manchester for the degree of Doctor of Philosophy in the Faculty of Humanities

2019

Malte B. Rödl

Alliance Manchester Business School

Contents

Contents 3

List of Figures 9

List of Tables 11

Abstract 15

Declaration 17

Acknowledgements 19

Preface 21

1 Introduction 23 1.1 Why Meat Alternatives Matter ...... 24 1.2 Categorisation and Carnism ...... 26 1.3 Scope and Aim of this Thesis ...... 28 1.4 Overview of this Thesis ...... 31

2 Situating the Research 33 2.1 Foundations of Category Studies ...... 34 2.1.1 Category Identity ...... 35 2.1.2 Category Relations ...... 36 2.1.3 Strategic Categorisation ...... 37 2.2 Category Formation ...... 38 2.3 Substitution and Categories ...... 40 2.4 Changing Culture and Challenging Cultural Dominance ...... 42 2.5 Conceptual Framework ...... 44

3 Contents

2.6 Methodologies to Study Categories ...... 49

3 Methodology 51 3.1 Research Design ...... 51 3.1.1 Overview of the Research Design ...... 51 3.1.2 Assumptions and Considerations ...... 53 3.1.3 (Qa) Studying Emergence ...... 54 3.1.4 (Qb) Investigating Category Identity ...... 57 3.1.5 (Qc) Analysing Cultural Change Through Frames . . . . . 59 3.2 Methods ...... 61 3.2.1 Content Analysis ...... 61 3.2.2 Frame Analysis ...... 67 3.3 Data ...... 68 3.3.1 Contextual Data ...... 68 3.3.2 Print Advertisements for Meat Alternatives in Vegetarian Magazines ...... 71 3.3.3 Semi-structured Expert Interviews with Manufacturers and Stakeholders ...... 76 3.3.4 Taste-Test Videos of Meat Alternatives from YouTube . . . 79 3.3.5 Tweets on Replacing Meat during National VegetarianWeek and World Meat Free Week ...... 83 3.3.6 Summary of Data Sources ...... 87 3.4 Limitations and Validity of this Research Project ...... 87

4 Contextualising Meat Alternatives in Past and Present 89 4.1 Description of the Data ...... 89 4.2 What are Meat Alternatives? ...... 90 4.3 A Brief History of Commercial Meat Alternatives in the UK . . . . 92 4.3.1 Early Meat Alternatives ...... 92 4.3.2 Vegetarianism Becomes Visible (1985–1993) ...... 94 4.3.3 The Healthy Choice (1994–2001) ...... 95 4.3.4 Slowly Towards Mainstream (2002–2011) ...... 96 4.3.5 Arising Competition (2012–2015) ...... 97 4.3.6 Market Boom (since 2016) ...... 98 4.4 Meat Consumption in the UK ...... 99 4.5 The Market for Meat Alternatives in the UK ...... 100 4.5.1 Consumers ...... 100 4.5.2 The Market ...... 102 4.5.3 Ownership Structures ...... 103 4.5.4 Supply Chain Networks ...... 104 4.5.5 Laws, Regulations, and Certifications ...... 105 4.5.6 Marketing and Consumer Engagement ...... 107 4.5.7 Relations to Meat and its Industry ...... 109 4.6 Other National Markets for Meat Alternatives ...... 110 4.7 Conclusion ...... 113

5 Print Advertisements for Meat Alternatives 115

4 Contents

5.1 Description of the Data ...... 115 5.2 (Qa) Longitudinal Analysis of Changes in Advertising (1985 to 2017) 117 5.3 Contemporary Advertisements (2016 to 2017) ...... 129 5.3.1 (Qb) What Are Meat Alternatives? Promoting New Foodstuffs 129 5.3.2 (Qc) Is there Meat in a ‘Chilli Non Carne’? Suggesting Al- ternatives and Re-inventing Meat ...... 132 5.4 Conclusion ...... 135

6 Expert Interviews with Meat Alternatives Stakeholders 137 6.1 Description of the Data ...... 137 6.2 (Qb) What Are We Selling? Identifying Meat Alternatives . . . . . 139 6.3 (Qc) On Copying, Reducing, and Reinventing Meat ...... 146 6.4 Conclusion ...... 148

7 Taste Test Videos of Meat Alternatives 151 7.1 Description of the Data ...... 151 7.2 (Qb) Valuing Meat Alternatives ...... 153 7.3 (Qc) The Meanings of Meat (Alternatives) ...... 158 7.4 Conclusion ...... 164

8 Tweets on Replacing Meat 167 8.1 Description of the Data ...... 167 8.2 (Qb) Describing Meat-Free: How Meat Matters ...... 171 8.3 (Qc) Talking Meat: How Tweets Frame Normality and Disruption . 175 8.4 Conclusion ...... 180

9 Meat Without Animals: An Analysis 181 9.1 An Established and Coherent Category ...... 181 9.2 Emerging Understandings of Meat Alternatives (1985–2017) . . . . . 183 9.2.1 Origins in Vegetarian and Meat Categories ...... 184 9.2.2 Declining Differentiation from Meat ...... 185 9.2.3 From Characteristics to Meanings ...... 186 9.2.4 Emancipation from Meat ...... 188 9.3 The Meat in Alternatives: Current Understandings ...... 188 9.3.1 Format Flexibility ...... 189 9.3.2 Meat as a Reference ...... 191 9.3.3 Signposting Deviance ...... 192 9.3.4 Opposing Meat ...... 194 9.4 Breaking Down Meat Alternatives ...... 195 9.5 The Cultural Politics of Meat Alternatives: Mechanisms of Chal- lenge and Reproduction ...... 200 9.5.1 Stigmatising and Invisibilising Deviant Ways of Eating . . . 201 9.5.2 Knowing Meat and Trusting Meat ...... 201 9.5.3 Counting Consumption and Creating Moral Obligations . 202 9.5.4 Eating Experiences and Expectations ...... 203 9.5.5 Navigating Experiences: When Food is Just Tasty ...... 204 9.5.6 Reclaiming Language ...... 205

5 Contents

9.6 Theoretical Implications ...... 205 9.6.1 Insights on Category Formation ...... 206 9.6.2 Limited Substitutability and Categorisation Stability . . . . 206 9.6.3 Categorical Generativity and Hegemonic Accommodation . 208 9.7 Conclusion ...... 209

10 Discussion 213 10.1 Responses to the Research Questions ...... 214 10.2 On Meat Alternatives ...... 216 10.2.1 Just the Same: Meat Alternatives as Continuity ...... 216 10.2.2 Violating Norms: Meat Alternatives as Diversity ...... 217 10.2.3 Plant-Based Counterculture: Meat Alternatives as a Political Project ...... 217 10.2.4 Food Activism: Meat Alternatives between Values and Money 219 10.2.5 More Meat from Less Feed: (Eco-)Efficient Contingencies . 220 10.3 On Theory ...... 222 10.3.1 Insights from Category Studies ...... 222 10.3.2 Limitations of the Theory in Context of this Research Project 223 10.3.3 Reflections on the Proposed Conceptual Framework . . . . 224 10.3.4 Further Considerations for Category Studies ...... 225 10.4 Contributions of this Thesis ...... 227 10.5 Limitations and Generalisability of this Research Project ...... 228 10.6 Rethinking Meat Alternatives? Implications and Recommendations 230 10.7 Conclusion and Future Research Directions ...... 232

Epilogue 235

References 237 Academic References ...... 237 Archival References ...... 256

Appendix

A Further Details of the Contextual Data 269 A.1 Detailed Overview of the Sampling Procedure ...... 269 A.2 Key Companies and their Histories ...... 270 A.2.1 Direct Foods ...... 271 A.2.2 Haldane Foods ...... 273 A.2.3 Quorn Foods ...... 276 A.2.4 Linda McCartney Foods ...... 280

B Further Details of the Advertisements 283 B.1 Detailed Overview of the Sampling Procedure ...... 283 B.2 Magazine Selection ...... 286 B.3 Software used for Coding ...... 287

6 Contents

B.4 Statistical Considerations ...... 288 B.5 Bibilographic Details of the Collected Advertisements ...... 289 B.6 Coding Schedule ...... 307 B.7 Statistical Evaluations ...... 312

C Further Details of the Interviews 317 C.1 Detailed Overview of the Sampling Procedure ...... 317 C.2 Overview of Interviewees ...... 318 C.3 Interview Guide ...... 319

D Further Details of the Taste Test Videos 323 D.1 Detailed Overview of the Sampling Procedure ...... 323 D.2 Overview of Taste Test Videos ...... 325 D.3 Structure of Taste Test Videos ...... 327

E Further Details of the Tweets 329 E.1 Detailed Overview of the Sampling Procedure ...... 329 E.2 Food Words Searched in the Data Set ...... 331 E.3 Foods Mentioned in the Tweets ...... 332 E.4 Reprint of Tweets Used for the Qualitative Analysis ...... 333 E.4.1 During #NationalVegetarianWeek ...... 334 E.4.2 During #WorldMeatFreeWeek ...... 350 E.5 Coding Schedule ...... 356

F Data Summaries Across Data Sets 359 F.1 Category Identity ...... 359 F.2 Frames ...... 360

Word Count: 84,530

List of Figures

1.1 Visual outline of this thesis...... 32

2.1 Visualisation of the conceptual framework...... 48

3.1 Visualised overview of data choices and research design. On the top, the choice of data and the different analyses used are mentioned. On the bottom, a timeline of data collection and analysis is presented, and the different research questions and their use are summarised briefly. . . . . 54

4.1 Listings of various meat alternatives brands by store as identified in online shops in July 2017...... 91 4.2 Comparison of the per capita (a) expenditure and (b) consumption of meat alternatives based on market data from 2001 to 2018. Kantar calculates their estimates from consumption-level data, Mintel and Euromon- itor use aggregated retail data; categorisation differences are likely to be responsible for the remaining differences in the estimates; each continuous represents data published at different times. The price adjustment of Mintel data was reversed; price index and population data were obtained from the Office for National Statistics...... 98 4.3 Upstream supply chains...... 105 4.4 Advertising expenditure from 1995 to 2016, including the share of Mar- low Foods (owner of Quorn; the Cauldron brand was bought in 2005, but without major advertising investment). Multiple lines indicate numbers from different reports. Compiled from: Nielsen Ad Intel/Mintel (2000, 2001, 2004, 2006, 2008, 2010, 2011, 2013, 2017)...... 108

9 List of Figures

4.5 Comparison of the markets for meat alternatives in UK, US, Germany, the Netherlands, and Western Europe. (a) Total annual expenditure (Re- tail Service Price); (b) total annual sold volume; (c) retail volume of meat alternatives relative to total volume of fresh and processed meats (including meat alternatives; no data for Western Europe); (d) annual expenditure per capita; (e) annual sold volume per capita; and (f) unit price of meat alternatives. All currencies were converted using 2018 exchange rates. 111

5.1 Overview of the sampled advertisements by year and magazine. The vertical lines separate the different time periods used in the analysis. Note that no advertisements in issues published after November 2017 are included in the sample...... 117 5.2 Development of the use of the labels and headlines in advertisements across the time periods...... 121 5.3 Development of the use of the descriptions in advertisements across the time periods...... 124

8.1 Timeline of tweet frequencies throughout the two theme weeks. Note again that WMFW is likely to have a more international audience (see again Appendix E.1), which can be observed in the distribution of tweets. 168 8.2 Types of foods mentioned in the tweets by theme week...... 170

9.1 Overview of the combinations of the three frames and the observed patterns and rhetorical strategies related to these...... 198

10.1 Core-periphery model of the category changes through the entrant of a substitution product...... 226

B.1 Exemplary view of the coding software: (a) Coded advertisement with its attributes; (b) overview of codings for a specific code; (c) use of the coding schedule...... 288

F.1 Frames used for meat alternatives across the data sets. Building on the characteristics identified, the frames for meat alternatives within single texts have been mapped into this scheme. The letters in circles denote the data set this frame occurred in; the elliptic lines identify which items represent which of the three frames...... 360

10 List of Tables

3.1 Summary of the research design. The table covers all research questions, data, methods, and a brief summary of the operationalisation...... 62 3.2 The procedure for the content analyses for the different data sets (inspired by Elo & Kyngäs, 2008; and Hsieh & Shannon, 2005)...... 66 3.3 Overview of the data sources used in this research project, including their timely coverage and amount of collected texts...... 86 3.4 Overview of the data sets, including where findings are presented, and where additional information can be found, how they are referenced in the thesis, and where the index is located...... 87

4.1 Overview of time periods in the category narrative and the events identified that signposted changes...... 95

5.1 Brands in the advertising sample with fifteen or more printed adverts, including the duration of their activity, the amount of printed advertisements per year, and their recent advertising activity...... 118 5.2 Amount of advertisements across the time periods which are printed by the two and four most active advertisers ...... 118 5.3 Overview of labels and headlines used for the products in the advertisements (Qa1). An advert could be coded with up to two labels...... 119 5.4 Amount of average descriptions coded per advert in each of the time periods; each advert could have up to 4 codes for product descriptions (Qa1)...... 121 5.5 Overview of the descriptions used for the products (Qa1)...... 122 5.6 Overview of the different distinctions drawn between products (Qa2). . 125 5.7 Advertised novelty in advertisements across time (Qa3)...... 126 5.8 Overview of different themes presented in the advertisements (Qa3). . . 128

11 List of Tables

5.9 Overview of descriptions used in recent advertisements (Qb1). An advert could be coded with up to four descriptions, see also Table 5.5 for an aggregation of the same code in the longitudinal sample...... 130 5.10 Overview of the identified frames for meat identified in the recently printed advertisements (Qc2)...... 133 5.11 Overview of how advertisements position meat alternatives in relation to meat (Qc1)...... 134 5.12 Summary of the frames observed for meat and meat alternatives in the recently printed advertisements (Qc). Statements in brackets are not made explicit in the advertisements...... 134

6.1 Overview of descriptions used for meat alternatives in the interviews (Qb1). 141 6.2 Overview of frames identified from the manufacturer interviews (Qc). . 147 6.3 Summary of the frames...... 149

7.1 Overview of characteristics and evaluations used to describe products in the videos (Qb1, Qb2)...... 155 7.2 Overview of comparisons made in the videos (Qb2)...... 158 7.3 Overview of the YouTubers’ motivations to make a taste test video, to eat meat alternatives, and what they suggest to their audiences (Qc1). Statements in brackets are not explicitly expressed in the videos...... 160 7.4 Overview of the frames used (Qc). Statements in brackets are not explicitly expressed in the videos...... 162 7.5 Summary of the frames identified in the taste test videos and how they understand meat and meat alternatives (Qc)...... 164

8.1 Overview of the collected tweets...... 168 8.2 Overview of the participation of selected meat-free brands and other stakeholders in the two theme weeks...... 169 8.3 Overview of descriptions for meat alternatives in the tweets (Qb1); as many tweets contained multiple characterisations, counts are based on the number of tweets mentioning a specific characteristic and do not add up...... 172 8.4 Overview of labels used for meat alternatives in the tweets (Qb3); combinations of these were possible, so counts do not add up...... 174 8.5 Overview of frames used in the tweets (Qc)...... 177 8.6 Summary of the frames including the issues and solutions marked in the tweets (Qc). Statements in brackets are not explicitly expressed in the tweets...... 179

9.1 Summary of the identified frames and their function (Qc)...... 195

12 List of Tables

9.2 Summary of the frames and mechanisms with which meat alternatives reproduce and/or challenge dominant meat culture (Qc)...... 200

B.1 Overview of the food magazines which were screened initially, and how many ads were found per screened issue...... 286 B.2 Results of the χ2-tests for the labels/headlines of advertisements (Qa1). . 313 B.3 Results of the χ2-tests for the descriptions of advertisements (Qa1). . . . 313 B.4 Results of the χ2-tests for the comparisons drawn in the advertisements (Qa2)...... 314 B.5 Results of the χ2-tests for the themes of the advertisements (Qa3). . . . 315 B.6 Results of the χ2-tests for the novelty with which products are advertised (Qa3)...... 316

C.1 Overview of the interviewees who kindly gave their time to participate in this study...... 318

D.1 Overview of the YouTube videos used for analysis...... 325

E.1 Mentions of various food categories in the tweets...... 333

F.1 Summary of the different evaluations found across the data sets (Qb2). . 359 F.2 Summary of the different labels and categorisations found across the data sets (Qb3)...... 359 F.3 Summary of the most prominent characteristics by data source (Qb1). . 361 F.4 Summary of the three overarching frames for meat alternatives identified, and their breakdown into diagnostic, prognostic and motivational components; these frames do hardly occur independently, but are usually used in combination (Qc1)...... 362 F.5 Frames used for meat across the data sets. These are from the different texts, whereby specific frame might be split into its components to match the three frames identified (Qc2)...... 363

Abstract

Meat alternatives are foodstuffs made to resemble animal flesh, but manufactured from other ingredients, that have become increasingly popular in British mainstream diets. These products are on the one hand appreciated as an opposition to animal products and their associated issues; on the other hand, they are often used to replace meat in an indistinguishable manner, thereby providing an alternative to this core constituent of British food culture. This thesis explores this apparent contradiction between challenging and reproducing hegemonic meat culture. To achieve this, this thesis assesses qualitatively (a) the history of meat alternatives in the United King- dom, (b) their description in texts of producers and consumers, and (c) the cultural embeddedness of meat alternatives by their advocates in relation to meat. From a theoretical perspective, this analysis builds on the category studies literature, which allows understanding meat alternatives as a set of goods with shared properties and meaning systems, that emerge from other previously existing categories. Specifically, it is argued that as categories become established, they may ultimately influence the original understandings they were built upon, such as meat culture. Grounding findings in diverse and detailed perspectives from various actors, this research utilises interviews with manufacturers and related stakeholders, print advertisements, taste test videos, and tweets. Furthermore, market research, industry news, and academic literature is reviewed to contextualise the research. It is found that historically both meat (through its use of animals and the associated externalities) and vegetarian foods (through its alleged inconvenience and inap- propriateness for a ‘normal’ life) are opposed. However, as meat alternatives became mainstream, arguments against meat were internalised as well as outsourced to dietary choices. These choices are made convenient and nearly invisible through meat alternatives which are rendered as a tool for a smooth transition, as they afford the same meanings and occasions as their meat counterparts. Conversely, the requirement or desire for meaty tastes or textures is individualised. The findings suggest that understandings of meat alternatives strongly build on hegemonic meat culture. Specifically, meat culture is reproduced through the reliance of meat alternatives on the practices, occasions, and meanings of meat-like foods. Conversely, dominant meat culture is challenged by an increased exposure to non- meaty tastes, textures, or formats; thus opening the possibility to create new norms in food consumption. As they are morally and environmentally preferable, meat alternatives are largely understood as ‘better meat’ rather than as exclusive to meat, albeit subject to individual approval of sensory desirability. It is argued that meat alternatives may help to reduce the consumption of animal meats in the short term as they simplify access to meat-free diets through existing tacit and practical knowledge. However, they reinforce the notion that meat eating is normal, pleasurable, or convenient, which delegitimises other non-‘meat’ food choices. As such, meat alternatives may present a long-term barrier to the elimination of non- human animal meat in human diets, both practically and symbolically.

Declaration

No portion of the work referred to in the thesis has been submitted in support of an application for another degree or qualification of this or any other university or other institute of learning.

i. The author of this thesis (including any appendices and/or schedules to this thesis) owns certain copyright or related rights in it (the “Copyright”) and s/he has given The University of Manchester certain rights to use such Copyright, including for administrative purposes. ii. Copies of this thesis, either in full or in extracts and whether in hard or electronic copy, may be made only in accordance with the Copyright, Designs and Patents Act 1988 (as amended) and regulations issued under it or, where appropriate, in accordance with licensing agreements which the University has from time to time. This page must form part of any such copies made. iii. The ownership of certain Copyright, patents, designs, trademarks and other intellectual property (the “Intellectual Property”) and any reproductions of copyright works in the thesis, for example graphs and tables (“Reproductions”), which may be described in this thesis, may not be owned by the author and may be owned by third parties. Such Intellectual Property and Reproductions cannot and must not be made available for use without the prior written permission of the owner(s) of the relevant Intellectual Property and/or Reproduc- tions. iv. Further information on the conditions under which disclosure, publication and commercialisation of this thesis, the Copyright and any Intellec- tual Property and/or Reproductions described in it may take place is available in the University IP Policy (see http://documents.manchester.ac.uk/display.aspx?DocID=24420), in any relevant Thesis restriction declarations de- posited in the University Library, The University Library’s regulations (see http://www.library.manchester.ac.uk/about/regulations/) and in The Univer- sity’s policy on Presentation of Theses.

Acknowledgements

Ten years ago, I cannot imagine that anyone I know would have thought that I would do a doctorate related to meat, culture, or just words in general, instead of one related to numbers, symbols, and formulas. The hard work of my academic apprenticeship finished, I am delighted that I took this path which has given me an understanding of the world and its complexity far surpassing what any formula or any mathematical proof could ever bring, even though reality may not be as beautiful. For this, I am particularly grateful to my supervisor Prof. Frank Boons, who I first met in Septem- ber 2012 at the TU Delft where he taught a course titled ‘Social Systems — Policy and Management’. In the first lecture, Frank told an anecdote about how he had in all his life living in the Netherlands never seen anyone leaving the train barefoot, while I sat there in the fourth row of the way too big lecture theatre just grinning at him, and completely ignorant of the fact that reading social sciences, policy, or organisation studies papers was really hard work. Our paths crossed again in 2015, when I got the opportunity for a studentship under Frank’s supervision at the Sustainable Con- sumption Institute in Manchester. I am extremely grateful for this opportunity that Frank gave to me; thank you for allowing me to explore some crazy research ideas, for trying to make me write papers, for showing me how to do research, for understanding when it was time to push or pull me, and when it was better to let me do my own things, and for growing a thorough awareness for proper methodology and operationalisation in me. And for engaging with eProg once in a while. Big thanks also go to my second supervisor Dr. Josephine Mylan for providing emotional, methodological, topical, and practical support where ever necessary, and for making sure I know that things always take longer than I thought they would. Thank you both for never losing trust, and if you did thank you for never telling me across all the probably torturous meetings. I apologise for all the new ideas around culture and meat alternatives that I initially tried to explain in engineering language with definition lists or decision matrices. Thank you both also for giving me the opportunity to see a clash of methodological traditions firsthand, for allowing me to choose my path, and for being so wonderfully complementary to each other. I am further extremely grateful to the Sustainable Consumption Institute: First for the studentship it generously provided to me, allowing to live a wonderful life in Manchester. The embedding as a PhD researcher into the SCI permitted a deep and invaluable insight into the workings of such extraordinary research institute, and over the years offered plenty of insightful seminars and workshops, discussions with senior academics, as well as fabulous office space. I have also enjoyed the embedding into an active PhD community here. I would also like to thank Susan Hogan and even more so Katrina Farrugia for their invaluable continuous administrative and

19 Acknowledgements practical support in all matters imaginable. Without you I would probably still wait for something. This research would not have been possible without all the people who are interested in meat alternatives professionally or in their spare time; in particular I would like to thank all my interviewees for their dedication, excitement, and most importantly their willingness to share some of their valuable time to contribute to my research. — Besides the amazing facilities and support of the University Library in Manchester, special thanks also go to the library of the Vegetarian Society as well as the British Library. These acknowledgements specifically also cover the people who not only spend a lot of time creating catalogues and allowing access, but also search for and find all the documents in their archives that were so important for this work. Never will I forget the envy I felt at the moment I was asked in the British Library whether the reason why I looked at a few hundred vegetarian magazines a day was because I was a chef. — I wished. Apart from the great office which the SCI provided me, I am very grateful for the uncountable hours I was allowed to spend on the 4th floor of Arthur Lewis Building, as well as in university libraries in Freiburg, Valparaíso, and Düsseldorf. My thanks also go to the Swing dancing and in particular the Collegiate Shag community in Manchester, my bikes, my running shoes, my camping gear, and my rain jacket for the numerous hours I spent with you distracting myself from work and keeping myself sane. A great support were also my parents who not only provided shelter and food for some of my intense writing months, but more importantly im- planted enough confidence in me to finish this massive project. I would also like to express my gratitude to my brothers who supported me in their own ways, and all the other friends who made me stay a human being–at least once in a while. Further, I would like to thank my laptop, external hard drive, cloud storage, and version control system, as well as all the amazing people dedicating their time to create an ecosystem of fantastic open source software. This includes in particular the people relentlessly improving Linux, Manjaro, Gnome, Atom, Pandoc, Zotero, Fire- fox, Git, R, Python, LaTeX, Lua, LibreOffice, Inkscape, and an even larger number of extensions and plugins. While sometimes being a good way to procrastinate, all these things made my work so much nicer, easier, and more enjoyable. I would like to thank explicitly my colleague, friend, and office mate of choice, (now) Dr. Ulrike Ehgartner who has always been there for guidance, motivation, in- depth discussion of mine or other people’s research, food, drinks, and practical and emotional support throughout all fun, boring, or stressful phases of the PhD process. Special thanks also go to (soon-to-be-Dr.) Shashwat Pande, with whom I shared so many inspirations, ideas, stories, references, insights into the world, and uncountable nights in front of ever empty glasses. This work would have been very different without the helpful comments and thoughts of Ulrike, Yannick, Wouter, Sterling, Jemàl, Godwin, Harald, Steffen, and Philippe, which I gratefully acknowledge. Lastly, I would like to thank Dr. Grace ‘Pufferfish’ Whitaker for all your emotional and practical support. Your love and care enlightened my life like a constant ray of sunshine that I had never imagined, sometimes asked for, but that I could always be certain of–even on cloudy days or at night, with a smile, with candles or fairy lights, or when you send sunshine from the other end of the world.

20 Preface

“I don’t eat meat.” — “What do you eat then? Do you eat, like Quorn?” — When I came to England first, this conversation happened regularly to me. And I did not even know what this British institution of Quorn was. Even though I grew up a vegetarian in a meat-eating household, I did not know that a thing like meat alternatives existed until I moved away to study. And even then, all that I discovered was that tofu sausages or somewhat meat-like burgers were not my type of food. In fact, I never understood what people liked about it, as I had never liked meat myself, and never really eaten much of it before I stopped altogether. I was just not attached to the idea of eating meat, neither culturally nor for hedonic reasons. Over the years since coming to England, I have trialled some of these foods, enjoyed them, did not enjoy them, but I always had to suppress the feeling of eating meat, and the reactions my body had developed to eating ‘flesh’ over eating meat-free for so long. At least for me, there was nothing magical about meat alternatives. Coming to England was certainly a change in my relations to meat alternatives. I had grown a bit more accustomed to the thought of vegetarians eating meat-like things, but I was not quite prepared for the diversity, choice, and following which meat alternatives had here. Whether they liked it or not, people had tried Quorn and other brands, and had an opinion on it. And they expected me to have one as I was not eating meat. This was only reinforced when they learned that this was my area of study. Originally we had intended to study the evolution of meat alternatives and the related industries over time from a socio-technical systems perspective. But in the first few months into my PhD, it became clear that this was neither what I was interested in, nor did I feel comfortable researching such topic if the starting point was not clear: ‘What are meat alternatives?’, I wondered. What made them increasingly hyped across the world soon after I had started my PhD? And why did they seem such powerful products in the first place? These questions crossed our debates in supervision meetings, and gradually they changed my research project. In this way, my research project became somewhat autobiographical–born out of my own puzzlings of the world (as I suppose is the way for many or most research questions). As such, the ideas in this thesis were originally developed from my inability to see the ‘magic’ of

21 Preface meaty pleasures, and curiosity about how technology and culture work together to form something new–maybe even having counterintuitive effects. Returning to the topic in question, it seemed somewhat clear to me that meat alternatives were comforting to people. But I also wondered what kind of ‘comfort’ they provide, considering that they are intended to replace meat with something ideally indistinguishable. Is that really change at all? Is it any effort? Does change need to be an effort, or might it be better if it is easy? — I wondered whether people would be more likely to go back to eating meat, if they would continue eating meat alternatives. How does this impact meat eating? I wondered whether all the ethical or gender connotations of meat eating would continue with meat alternatives, just like tastes and flavours seemed to. Where are the differences between meat and meat alternatives, and how do they play out? And if these foodstuffs soothed mind and taste buds alike, had I misunderstood what veg*anism meant? As my research progressed, the market for meat alternatives was booming. Food industry newsletters regularly brought new announcements: Meat processors in Ger- many re-invented their brands as they introduced meat-free versions of their cold cuts and sausages; vegan and vegetarian numbers were on the rise; companies from the US and the Netherlands joined the UK market with highly acclaimed products; bleeding burgers and steaks were introduced; being flexitarian or meat reducer became mainstream; some meat alternatives were sold in meat shelves in supermarkets; big agri- food corporations joined the market; and globally even meat processors invested in or manufactured their own meat-free meats. In one of these news articles, I stumbled across maybe the most explicit enunciation of my thoughts in a daily newsletter from the food industry: In a linked article, “advances in replicating [meat-like characteristics] in plant-based products” were promised. These products, it was claimed, would be able to “ensure that meat remains a staple in the vast majority of the nation’s shopping trolleys” (foodmanufacture.co.uk, 2018). — Hold on! Could meat and meat alternatives be in fact all the same? I have found some answers throughout the course of my research, whilst other questions continue to puzzle me. In this dissertation I will take you on a long journey, that might feel a bit surreal after the almost 3,000 times I carved the word ‘meat’ into this thesis. It may sometimes seem self-evident once you start thinking about it, and sometimes it is truly astounding. But looking back on it in a quiet minute, my mind is blown by the long way I have come, the things I have learned, and just by how interesting it has been. If you feel somewhat similarly puzzled–or even not–the perspectives on meat alternatives served up in this dissertation should be an insightful and worthwhile read.

22 Chapter 1

Introduction

The ingestion of flesh from non-human animals1 is deeply ingrained in the evolution and structure of human societies (Leroy & Praet, 2015). Commonly referred to as eating meat, this practice2 is central to British culinary identity. Embodied in meals such as bangers and mash, roast dinner, toad in the hole, fish and chips, or the contemporary national favourite of chicken tikka masala–the presence of meat is traditionally indicative of a ‘proper meal’ (Douglas & Nicod, 1974; Murcott, 1982). Historically associated with affluence and luxury, the introduction of industrial animal farming led to increased productivity and made meat cheap and accessible–and progressively over- consumed (Mitchell, 1999; Smil, 2002). However, over the last decades, a counter- trend seems to be emerging, as an increasing amount of red meat is substituted with white meat (e.g. beef with poultry, Mitchell, 1999) and more people reduce or reject eating meat altogether (Beardsworth & Bryman, 2004; Vegetarian Society, 2001). The desire for a reduction of red meat consumption is often related to increasing insights that excessive rearing and eating of animals might be disadvantageous for individuals and societies as a whole (Ruby, 2012). For example, evidence has emerged that red and processed meats cause cancer, heart diseases, and other major impairments to human health (e.g. Bouvard et al., 2015; Battaglia Richi et al., 2015). Furthermore, the modes of production and increased consumption of these foods are established as major contributors to environmental and ecological challenges, such as anthropo- genic climate change, deforestation, desertification, or eutrophication (e.g. Steinfeld et al., 2006; Nijdam, Rood, & Westhoek, 2012). And while some people reject the idea of eating animals altogether, others are opposed to the conditions under which animals in intensive farming are raised, killed, and further processed (e.g. Armstrong & Botzler, 2016; Ruby, 2012; Verbeke & Viaene, 2000). With increased visibility and spreading of such insights, these issues are cited as reasons for the reduction or rejection of meat consumption (Mintel, 2017, see also chapter 4). So-called ‘meat alternat-

1Unless explicitly stated otherwise, throughout this thesis the term animals will refer to domesticated non-human animals that are reared for the purpose of slaughter and subsequent human ingestion. Similarly, the term meat is used to refer to parts of these animals when eaten. 2The term practice is used here in dictionary terms as something that is on a regular, habitual, or customary basis, and does not refer to the practice theory literature.

23 1. Introduction ives’ have been identified as an alleviation for these issues associated with meat, and will be discussed herein.

1.1 Why Meat Alternatives Matter

Meat alternatives are foodstuffs intended to be similar to meat either in appearance, taste, texture, smell, or cooking properties–but are not made from animal flesh. Re- search has suggested that these alternatives are healthier than meat (e.g. Sadler, 2004), better for the environment (e.g. Nijdam et al., 2012; Smetana, Mathys, Knoch, & Heinz, 2015), and although some products are not entirely free from animal products3, they are arguably more ethical. Consumer studies report that some consumers find meat alternatives insufficiently similar to meat (Elzerman, van Boekel, & Luning, 2013; Mintel, 2011, 2017). However, more regular consumers have more favourable attitudes towards meat alternatives or even prefer these over meat (Hoek et al., 2013; A. C. Hoek, Luning, et al., 2011). Fur- thermore, technological developments are claimed to be able to close these sensory differences to meat (e.g. Wild et al., 2014; Zorpette, 2013). Regardless of sensory credentials, the consumption of meat alternatives has further been found to be motivated by allowing people to reduce their meat consumption without significantly changing their preferences and practices (e.g. Nath & Prideaux, 2011). Meat alternatives thus imply continuity and stability, as people may continue to cook the same recipes, eat similar foodstuffs, and participate in the same meat-centred customs and traditions (e.g. christmas dinners, barbecues, or even daily meals). As this simplifies and facilit- ates meat reduction or elimination, sometimes meat alternatives are referred to as a ‘transition food’ (e.g. in the food industry, see Michail, 2018). Given the associated benefits, meat alternatives may thus be seen as beneficial for individuals, societies, animals, and the planet. While producers of meat alternatives used to target vegetarians and vegans, over the last decade the products have become increasingly popular among a general audience (e.g. Mintel, 2017), and sales have soared (Euromonitor International, 2017a, see also chapter 4). In some cases, meat alternatives have been established as a foodstuff that appears to be a saviour to the world: For example, popular outlets have titled that “We could end factory farming this century, thanks to meat alternatives” (Piper, 2018), or that “In the Future, the Meat You Eat Won’t Come From Living Organisms” (Gohd, 2017). Business-oriented outlets suggested that “A Bright Future is Germin- ating for Alternative Proteins” (Hartman, 2018) or “Our Meatless Future: How The

3Some products may contain milk protein, or use egg albumen as a binder; but many products are largely made from gluten, soy, or other plant-based proteins (see section 4.2).

24 1.1. Why Meat Alternatives Matter

$90B Global Meat Market Gets Disrupted” (CB Insights, 2017). Titles like these emphasise meat alternatives as nearly identical, or even a continuation of meat and the associated understandings and industries. This is further illustrated by a sponsored article for manufacturing technology on a website for UK food industry news:

He [the author of a study] identifies the crucial factors in new product development to capture the opportunities as vegetarians, vegans and those who are looking to reduce their intake of red meat search for innovations which can offer the tastiness and texture of meat. He draws on his own experience of a tasting session in which the technology employed offered significant advances in replicating in plant-based products the characteristics which ensure that meat remains a staple in the vast majority of the nation’s shopping trolleys. (foodmanufacture.co.uk, 2018, highlight added)

In this example, ‘meat’ is not necessarily portrayed as a product of animal origin but instead regarded as resembling a cultural identity that needs to be preserved in and beyond “the nation’s shopping trolleys.” While meat alternatives are undis- putedly better for environment and animals than their model counterparts, meat is an essential component in many diets and meals (e.g. Murcott, 1982; Yates & Warde, 2015). Meat is also understood as having superior sensory (Frank, Oytam, & Hughes, 2017) and nutritional (Biesalski, 2005; McAfee et al., 2010) qualities compared to other foodstuffs. On the other hand, criticisms of the cultural connotations of animal meat consumption exist, which might be reflected in meat alternatives: Meat has, for example, been related to exhibiting and perpetuating social hierarchies and power imbalances on a level of gender (Adams, 1990) and species (Potts & Adams, 2010). Meat eating is regularly justified as being nice, normal, natural, and even necessary (which Piazza et al., 2015 term the ‘4N’). The dissonance induced by meat eaters being confronted with critique therefore needs to be reduced with mental protection mechanisms, such as a perceived change of behaviour, dissociation between eating meat and animals, or denial of animal pain (Rothgerber, 2014). The consumption of meat as a safeguarded ‘natural symbol’ (Fiddes, 1992) has been long established over time and across societies (e.g. Fairlie, 2010; Joy, 2009; Leroy & Praet, 2015; Murcott, 1982), so that the belief system surrounding meat may be understood as a cultural hegemony. Therefore, Joy (2009) introduces the term ‘carnism’ in juxtaposition to ‘vegetarianism’ or ‘veganism’ to denote a diet containing meat. The term emphasises the ideological nature of meat consumption, with its inherent inconsistencies and unspoken assumptions. Considering the prominent and often invisible status of carnism, there have been arguments against the replication of meat through technology: For example, from a

25 1. Introduction philosophical perspective Bramble (2017) suggests that it may allow society to conveniently ignore the “moral failing” associated with (intensive) animal agriculture. More emotionally, US-American vegetarian activist Maude Sharpe (see Cronin, 2018 and section 4.3) argued in their vegetarian cookbook published the early times of meat alternatives:

There will also be found lacking from the index such dishes as “Veget- arian Hamburg Steak,” “Pigeon Pie, Vegetarianstyle,” etc., which should repel rather than attract, by bringing to mind what Bernard Shaw has graphically spoken of as “scorched carcasses.” (Sharpe, 1908, p. 12)

The fact that meat originates from animals is strongly related to its cultural meaning, not only for individuals (Piazza et al., 2015; Rothgerber, 2014) but also societies (Joy, 2009; Leroy & Praet, 2015). Thus, if meat alternatives ought to be the same for people but include no animals, meat alternatives may build on and re-negotiate the relationships between their consumers and dominant culture at large. Accordingly, this thesis critically examines the seeming tensions between disruptive material change, and soothing practical continuity–which both appear to be associated with meat alternatives. Building on these thoughts, the following research question is investigated in this thesis:

How does a category of ‘meat alternatives’ reproduce and/or challenge the dominant meat culture?

1.2 Categorisation and Carnism

Research on categories and categorisation encompasses an emergent body of literature within organisational studies. Category studies is rooted in literature on psychological sense-making on the one hand, and discursive and sociological understandings on the other (Vergne & Wry, 2014). Prevalent research deals with questions such as how categories come into place (e.g. Durand & Khaire, 2017; Navis & Glynn, 2010); how category members are identified, and how their categorisation may be influenced by audiences, producers, and intermediaries (e.g. Granqvist, Grodal, & Woolley, 2012; Pontikes & Kim, 2017); or how category membership is evaluated by audiences (e.g. Zuckerman, 2017; Paolella & Durand, 2015). Building on this research, in this thesis ‘meat alternatives’ are understood as a socio-cultural category: Their categorisation draws on established meaning systems and cultural elements, which existed previously to the category, and are in constant flux and re-negotiation (Glynn & Navis, 2013). Categorisation is furthermore to be understood as reflecting ‘social realities’ through the occasions and meanings which are embodied in categories (Kennedy &

26 1.2. Categorisation and Carnism

Fiss, 2013). Any speech act is categorising, as it voices a specific understanding of a category instead of others, which can be understood in terms of framing (Pontikes & Kim, 2017; see also Goffman, 1974). I suggest three sub-questions to this research project, which are described in detail in chapter 3:

a. How did the meat alternatives category evolve whilst becoming mainstream? (1985–2017) For the research project, it firstly must be established if the category exists, and whether the proposed conceptual framework (see section 2.5) represents category emergence and the associated developments appropriately. According to previous literature, such extended insight into category origins also helps to contextualise and interpret findings about the present. b. How are meat alternatives described and related to in the present? (2016–2018) The research project must then establish (using more diverse data) whether this proposed category can currently be understood as a coherent entity that warrants further investigation; furthermore, it must be asserted in detail whether and how meat is relevant in the understanding of meat alternatives. c. How are meat alternatives (and meat) framed, and how does this relate to dominant meat culture? (2016–2018) Approaching an answer to the main research question, this third part investigates how understandings of the category concern the relevance and interpretation of meat alternatives and of meat. The identified frames are used to investigate the implicit and explicit relations between meat alternatives and meat, and thus how meat alternatives may reproduce or challenge dominant meat culture.

Although most previous research on meat alternatives has dealt with manufacturing technology (e.g. Davies & Lightowler, 1998; Wild et al., 2014) or conducted consumer experiments (e.g. Hoek et al., 2013; Elzerman, Hoek, van Boekel, & Lun- ing, 2011), the suggested understanding of meat alternatives is not unique: Previous research has investigated the meanings of meat alternatives for consumers, such as by studying consumer identity and societal participation (Nath & Prideaux, 2011); gender embodied in meat alternatives (Flail, 2006; Nath, 2011); the meat norms perpetuated in advertisements for meat alternatives (Rödl, 2018a; see also Fuentes & Fuentes, 2017 for an analysis of the marketing of dairy alternatives); or the semiotics of meat alternatives (Stahl, 2017). Recently, research has increasingly also focused on discursive politics enacted through meat alternatives or their absence; this includes analyses of meat-free Monday campaigns (Morris, 2018), or instituting meat alternatives as edible (Sexton, 2018). Especially the example of meat grown in cell cultures, so called in-vitro meat, has also led scholars to discuss emerging and changing ontologies (Jöns- son, Linné, & McCrow-Young, 2019; Stephens, 2013), or to investigate the promises

27 1. Introduction and expectations around these products (e.g. Chiles, 2013; Sexton, Garnett, & Lor- imer, 2019). Some scholars have also looked at alternatives to meat from a practice theoretical approach (e.g. Kanerva, 2016; Twine, 2018; see also House, 2018 for the hypothetical case of insects). Previous research, however, exhibits a number of shortcomings: Some research conflates or treats as similar plant-based meat alternatives, in-vitro meat, and/or insect-based proteins (e.g. Buscemi, 2014; Sexton, 2018; Stahl, 2017); relies on single groups of actors such as consumers (e.g. Nath & Prideaux, 2011; Elzerman et al., 2011) or producers (Sexton, 2018); and/or focuses on single issues concerning meat alternatives (e.g. Flail, 2006; Hoek et al., 2013). Building on previous conceptualisations of meat alternatives that relate to existing ontologies (Nath & Prideaux, 2011; for the case of in-vitro meat, see also Steph- ens, 2010; Jönsson et al., 2019), categorisation studies and the approach put forward here allow the investigation of the meaning systems of meat alternatives, both as originating from as well as contributing to the understanding of meat (see section 2.5). Specifically, meat alternatives are understood as a socio-cultural category related to meat through culture (e.g. Glynn & Navis, 2013). Such category is co-constructed by a multitude of producer and audience perspectives (e.g. Quinn & Munir, 2017), established through discourses (Grodal & Kahl, 2017; Khaire & Wadhwani, 2010), and negotiated by frames (Schiller-Merkens, 2017; Weber, Heinze, & DeSoucey, 2008). This research project can therefore provide a nuanced and contextualised account of meat alternatives as a contradictory foodstuff, whilst the various perspectives allow for more generalisation. Although links between category studies and social movement research exist (cf. Negro, Koçak, & Hsu, 2010), ideological struggle in general and cultural hegemony in particular has been under-theorised in category studies. As meat eating is hypothesised to be a central aspect to a category of meat alternatives, this is accounted for in the developed conceptual framework (see section 2.5).

1.3 Scope and Aim of this Thesis

At large this research project relies on academic literature in organisation and innovation studies, and should be read as a contribution to the broad field of science, technology, and innovation studies (STIS). Meat alternatives are therefore not primarily treated as a foodstuff but as a novel technology that conflicts with and integrates with existing cultural understandings of ‘good food’. Specifically, this thesis investigates the relationship between meat alternatives and meat in order to explore and understand the apparent tensions between continuity and change that meat alternatives–as outlined in this chapter–appear to embody. Furthermore, this thesis aims to identify some mechanisms of reproduction and challenge involved. Concretely, this is done by exploring the emergence of a category of meat alternatives, its current valuation,

28 1.3. Scope and Aim of this Thesis and its overlaps to dominant meat culture. This is achieved by qualitatively investigating the understandings related to commercial meat alternatives by consumers or producers of such products in the United Kingdom. In the following paragraphs, the precise scope and aims of this work, as well as a few limitations of this approach are outlined. This section concludes by listing the concrete contributions of this thesis to existing academic and practical knowledge and literature. In this thesis, ‘meat alternatives’ are understood as commercial products that are appreciated for their (subjectively identified) similarity to meat, but are free from animal tissue (for a more comprehensive overview of what meat alternatives are, see section 4.2). Whilst the term ‘meat substitutes’ tends to be used similarly, ‘meat analogues’ are sometimes understood as any meat-like textured protein material (Ma- lav, Talukder, Gokulakrishnan, & Chand, 2015), or as highly similar to meat in their sensory properties (e.g. Zorpette, 2013). Meat analogues are therefore part of what is understood here as meat alternatives. Not covered in this research project are meat extenders (meat products that are stretched with textured non-meat protein), or in- vitro meat (meat grown from cell cultures). As this study focuses on products and foodstuffs, it does not allow any insights into meat-free diets or broader practices. Furthermore, this study will focus on meat alternatives in the United Kingdom. This includes any product sold in the UK4, because via availability in stores, also products from foreign producers participate in categorisation by audiences. Although a cross-cultural study would have extended possible implications of this work, such analysis would be confounded or impeded by the different cultural connotations of meat, and the differing market histories and developments (see section 4.6). All data for this research project was collected as purposive samples with further limitations on accessibility and availability, and the resulting issues to generalisability this creates. The data further deals exclusively with meat alternatives as an object of discussion bearing three major delimitations to potential outcomes: Firstly, no direct conclusions regarding meat and its perception can be drawn, as meat is never a direct focus of the sources. Conclusions regarding meat are limited to the meanings and relations which are asserted, implied, or absent in the talk surrounding meat alternatives. Secondly, findings are limited to those types of meat alternatives which are discussed in the data but not the entirety of meat alternatives available. In the data, there is a much stronger focus on burgers, sausages, mince, and other replicas of processed meats, but only seldom on steaks or maybe pieces. Therefore, no specific consideration has been given to the potential differences in their cognitive evaluation,

4The UK was primarily chosen because I lived and researched in this country. The rich history and strong acceptance of meat alternatives in the UK were considered to allow for interesting insights when the research began (see chapter 4 and specifically section 4.6 for a brief overview of some other national markets for meat alternatives).

29 1. Introduction which has previously been established in non-vegetarians but not in consumers of meat alternatives (A. C. Hoek et al., 2011). Thirdly, all data sources used in this research project cover meat alternatives as part of their own commercial or private initiative. Concretely, the data sources cover advertisements for meat alternatives and expert interviews with manufacturers, but also social media users who independently write or speak about these foodstuffs. These delimitations are necessary to retain a sufficient amount of detail on the topic of interest. Any deviance from the focus on meat alternatives and self-initiated reporting would arguably give a wider range of insights, yet may obscure the identification of mechanisms. The chosen approach thus implies a perspective of meat alternatives as a legitimate foodstuff, which by proxy relates to meat and its cultural connotations. While the diversity of data employed should reduce any shortcomings within the limitations outlined, possible generalisation is limited to the interactions of meat alternatives and dominant culture through discourse of producers and audiences which see meat alternatives affirmatively. Furthermore, despite the variety of sources from which data has been drawn, this work covers no systematic analysis of agency. Although occasional remarks on agents and agency are made, differences across the data sets (and thus groups of agents) are only discussed when they become apparent or contribute to the analysis of the mechanisms and frames. Whilst the importance of agency can not be understated, its detailed discussion in this research project would have exceeded the manageable extent of this research project. This research contributes to literature and knowledge on multiple levels: Firstly, previous category studies have not (to my knowledge) been explicitly concerned with categories that may be described as a direct substitute, or as challenging a dominant order. The theoretical and empirical insights obtained from this study can help establish and illustrate the suitability of category studies as social or cultural change. As an outcome of this thesis, it should be possible to judge whether category studies are suitable to deal with hegemony, ideology, and challenges to them. Secondly, this research project contributes to the growing literature on meat alternatives, providing an in-depth account of its origins, its current understandings, its developments, and its cultural entanglements. Thirdly, businesses and retailers involved with meat alternatives may be interested in the outcomes of this research project, which provides a snapshot of the discourses in which meat alternatives are constructed, and which these ventures participate in. The perspectives offered in this thesis may, for example, help to identify business opportunities for the food industry. Lastly, I hope that my research contributes a cautionary tale on the recent hype around meat alternatives. While certainly there are many benefits on practical levels, as a result of engaging with this work, interested readers may reflect on their own involvements in meat, meat al-

30 1.4. Overview of this Thesis ternatives, and their individual roles in reproducing and challenging dominant social and cultural norms related to consumption.

1.4 Overview of this Thesis

As illustrated in Figure 1.1, the remainder of this thesis is structured as follows: In the next chapter (chapter 2), the extant literature on categories is reviewed. This includes previous findings in relation to culture and cultural change, as well as methodologies applied in the field. In chapter 3, the research project is outlined in more detail, including its motivations, epistemological underpinnings, and the interactions between and progression across the research questions. An overview of methods and data sampling procedures is presented, and the individual methodologies used to gain insights for the separate research questions across the data are described. Furthermore, the limitations and validity issues to this research are outlined. Chapter 4 introduces the empirical context to this research project through a historic narrative, as well as an overview of the more recent developments in the market. This chapter is mostly informed by secondary literature, a bibliography, market research, and news articles. Chapters 5 to 8 introduce the data: Print advertisements, expert interviews, taste test videos, and tweets. All these chapters firstly offer a description of the data, followed by the findings relating to each of the research questions. Chapter 9 provides a rich analysis of the data along the lines of the three research questions. These discuss category emergence, current category identity and understandings, and the cultural connotations of meat alternatives. Hereby, the findings from the different data sets are synthesised, before they are discussed in detail, and theoretical implications are presented. Lastly, in chapter 10 the findings are discussed more broadly. This includes a summary of responses to the research questions, further reflections and implications of the empirical insights, as well as a discussion of conceptual framework and theoretical insights. Limitations of the findings and their potential generalisation are discussed, and the practical significance of the research is highlighted. The chapter closes in a brief summary of the contribution of this research project, and some directions for future research.

31 Figure 1.1: Visual outline of this thesis. Chapter 2

Situating the Research

In order to introduce the concepts and theoretical underpinnings used in this research project, this chapter summarises the relevant theoretical literature. Underlying the analysis throughout this thesis, a conceptual framework is proposed, and previously applied methodologies for similar research projects are summarised; these both support the construction of a methodology to answer the research questions in the following chapter. The specific conceptual understanding employed in this thesis belongs to the emerging literature of ‘category studies’. Part of the organisation studies literature, it has roots in cognitive and psychological literature on the one hand, and in sociological literature on the other hand (see e.g. Negro et al., 2010; Durand et al., 2017b, as well as the following section 2.1; Vergne & Wry, 2014). In this research project mostly findings and concepts from the sociological strand of research are used, as it understands categories as socially constructed and historically contingent. This firstly allows understanding meat alternatives as product categories and thus as technologies or concepts rather than exclusively a foodstuff. According to the literature, both valuation and categories are discursively negotiated and that those discourses build on historical or preexisting concepts and valuations. This then makes it possible to understand meat alternatives as deriving from and engaging with understandings of meat (and vegetarian foods) as existing concepts. In the conceptualisation proposed here, this is augmented with understandings of cultural hegemony and cultural change (see section 2.4), allowing meat to be understood this way for the purpose of this work. The literature review presented here is purposive, as the concepts relevant for this research project are identified, and a conceptual framework on the basis of these reviewed concepts is developed. For this, a systematic literature review (Vergne& Wry, 2014), and two introductions to books on categorisation (Durand et al., 2017a; Negro et al., 2010) were used to identify topics and issues relevant to this research project, after which an in depth review of each of these topics was conducted. Papers were identified for review if they were referenced in the aforementioned publications, or cited any of these contributions (as identified via Google Scholar) and matched the focus of this literature review; further references in identified papers were followed up and reviewed if their content appeared relevant to this work.

33 2. Situating the Research

In the following sections, firstly a broad overview of the literature and its main trends is presented, before three aspects of category studies and cultural change relevant to this thesis are reviewed. Thereafter, the conceptual framework underlying this work is outlined, and previous methodological choices are reviewed.

2.1 Foundations of Category Studies

At large, categories can be understood as grouped together entities which are somewhat similar to each other. As part of a wider evaluation framework, mental representations of categories allow classifying unknown objects. Comparisons between different objects within one category can focus on the relevant differences as the category asserts some foundational similarity (Durand et al., 2017a). Categories thus help us navigate our lives: On a fundamental level they do so by distinguishing environmental stimuli into for example harmful, beneficial, or neutral ones; on a more concrete level, categorisation can be employed for example to identify and evaluate different offerings in a supermarket. The academic investigation of categories as something that gives structure to transactions emerged since the 1980s in the fields of sociology of markets, organisational theory, and strategy research (Vergne & Wry, 2014). The research area focuses on organisational, product, or market categories as the respective research questions and empirical focus determine the scope and level of analysis (Vergne & Wry, 2014). While originally much work has focused on self-categorisation of actors and the effects of this, more recent work is concerned with understanding categorisation either as an exogenous mechanism with certain effects on category members, or as a social construction (Durand et al., 2017a; Vergne & Wry, 2014). These approaches are thus about how categories are evaluated in cognitive processes on the one hand, and how categories are constructed in social processes on the other. Therefore, “categories both affect and reflect organization and market behaviour” (Kennedy & Fiss, 2013, p. 1140). While these two areas are not necessarily exclusive on a conceptual level, they commonly differ in the assumptions, research questions, and approaches used. The cognitive approach understands categories as a framework for market exchange (Durand et al., 2017b; Pontikes, 2012; Zuckerman, 1999). Within this framework, there are two major models of categorisation: Prototypical evaluation compares an actual entity against a prototypical entity; whereas goal-based evaluation ranks entities based on how well they match goal- and need-oriented ideals (Durand et al., 2017a; Vergne & Wry, 2014). Durand and Paolella (2013) further suggest a mechanism on causal grounds. Generally, there is an understanding that when entities deviate too much from an ideal category member they may be sanctioned or disregarded in

34 2.1. Foundations of Category Studies an evaluation; this is influenced for example by different evaluative contexts, evaluation mechanisms, audiences, or expectations (Zuckerman, 1999, 2017). Although the cognitive evaluation of categories used to be the main concern of category research (see Vergne & Wry, 2014), an increasing amount of work highlights the limitations of this approach to understand where categories come from and why they are evaluated in specific ways. This research instead suggests understanding categories as more flexible entities that can be shaped and navigated by actors (e.g. Glynn & Navis, 2013; Durand & Paolella, 2013; Kennedy & Fiss, 2013). This led to stronger focus on the social process approach, which understands categories as contextually produced and enacted (e.g. Glynn & Navis, 2013; Khaire & Wadhwani, 2010). This approach focuses on the assumption that categories and their boundaries and meanings emerge and are collectively shaped, negotiated, and constructed. Research concerned with the social process approach focuses among others on strategic agency for (self- )labelling (Granqvist et al., 2012; Pontikes & Kim, 2017), the role of intermediaries (Khaire, 2017), or the discursive enactment of categorisation (Grodal & Kahl, 2017). For the purpose of this work and in line with the questions asked, categorisation is understood as a social process. To give the reader a better understanding, the overview within the next few subsections illustrates both streams of literature, whereas the review thereafter focuses on the social approach to categorisation. Besides the basic concepts of category studies which are summarised in the following subsections, the investigation of a product category in general, and of a foodstuff (like meat alternatives) in particular requires some further literature to be reviewed: Firstly, different conceptual propositions and insights into the formation of new categories, as well as category origins in discourse and culture are described in section 2.2. Secondly, as some products do not create new demand, but instead tap in to existing markets by offering a substitute, the relevant literature on substitution and related category demise is discussed in section 2.3. Lastly, as products and their consumption are strongly related to everyday practices and thus prevalent mindsets or ideologies, section 2.4 discusses the relations between categories and socio-cultural change. Thereafter, a conceptual framework building on these insights will be outlined, and methodologies to study categories presented.

2.1.1 Category Identity The construction of organisational identities can be motivated by two different ideas: Either blending in as part of universal isomorphism (DiMaggio & Powell, 1983), or sticking out as part of an individual differentiation (Deephouse, 1999; Pedersen & Dobbin, 1997, 2006). Category identities for markets and products follow the same pattern, as they emerge from member identities that share similar symbolic and mater-

35 2. Situating the Research ial resources despite their individual differentiation (Hannan, Pólos, & Carroll, 2007; Negro et al., 2010; Vergne & Wry, 2014). Category identities are negotiated by producers, audiences, and intermediaries alike, and therefore can be influenced by any of these (Grodal & Kahl, 2017; Pontikes & Kim, 2017). They are often understood to be constructed by selectively borrowing, combining, or modifying elements from other institutional logics, categories, or generally cultural elements (Glynn & Navis, 2013; Jones, Maoret, Massa, & Svejenova, 2011). Although category membership is granted by the evaluating audiences or intermediaries (Glynn & Navis, 2013; Khaire, 2017), category membership and identity can be suggested or asserted by attaching labels to entities (Granqvist et al., 2012; Vergne & Swain, 2017). Category membership can be both enabling and restricting (Glynn & Navis, 2013; Negro et al., 2010; Vergne & Wry, 2014): A foundational set of codes is assumed, common courses of action are implied, immediate rivalry and a collective identity are established, and audiences evaluate role conformity. Violations of these category codes may be sanctioned by audiences, which is often referred to as a ‘categorical imperative’ or an ‘illegitimacy discount’; these terms describe the legitimacy acquired by conformity to established category ideals and the illegitimacy of deviant identities respectively (Zuckerman, 1999, 2017).

2.1.2 Category Relations Commonly, categories are seen to be related to other categories on various levels. In an evaluative approach, cognitive representations of categories are linked in vertical and horizontal manner, such as in form of umbrella, sub-, or rival categories (Vergne & Wry, 2014). On a horizontal level, categories can contrast each other or be more similar, which is usually related to shared attributes (Vergne & Wry, 2014). Larger differences between categories increase mutual competitiveness, while smaller differences are more likely to convolute category memberships (Kovács & Hannan, 2010). If one entity can be equally classified into different horizontally related categories, this is termed category spanning or category straddling and can cause the entity to be understood as illegitimate in either category (Vergne& Wry, 2014). However, when category systems are emergent or under heavy contestation, spanning categories is understood to be less problematic (Negro et al., 2010). Also vertically, different categorisations can influence audience evaluation. For example when an entity spans multiple categories or needs to appeal to a certain audience, a higher order label is more inclusive and may thus be advantageous (Negro et al., 2010; Vergne & Wry, 2014), such as a nano-technology venture which might not be as promising to a specific group of investors as a generic technology venture (Granqvist et al., 2012). In socio-cultural approaches, category boundaries and relations are less important,

36 2.1. Foundations of Category Studies also because cultural resources and discourses are subjective and shared by different categories. These approaches focus on category meanings; specifically on how they are attained from existing meaning systems (e.g. Glynn & Navis, 2013; Kennedy & Fiss, 2013) and negotiated as part of a broader discourse (Grodal & Kahl, 2017; Khaire & Wadhwani, 2010). Category relations therefore exist in the lineages of different meaning systems (Glynn & Navis, 2013), their impact on each other through re-imagination or translation (Kennedy & Fiss, 2013), or the interaction of these meaning systems over time, for example as they become unified or rearranged (Kennedy, Lo, & Lounsbury, 2010).

2.1.3 Strategic Categorisation Evaluative research in category studies considers established categories as reflecting a consensus view defined by middlemen or intermediaries and therefore as exogenous to the category members (e.g. Kovács & Hannan, 2010; Ruef & Patterson, 2009; Zuckerman, 1999). However, a social approach treats categorisation as dynamic and thus subject to strategic action by all involved actors (see Negro et al., 2010; Vergne & Wry, 2014). In fact, Pontikes and Kim (2017) propose that any act of categorisation is inherently strategic as it “convey[s] information about the producer and/or the market and . . . promote[s] a categorical system that is favorable to the organization or person” (p. 73). This inevitability implies that categorisation does not need to be deceptive; instead it is based on frames as ways to generally view and articulate things (Goffman, 1974), to “present [a] product, or [to] define a category, in a plausible and targeted way” (Pontikes & Kim, 2017, p. 75). Previous research has identified categories as targets of strategic action: Cat- egory membership or labels (Granqvist et al., 2012; Vergne, 2012; Zhao, Ishihara, & Lounsbury, 2013); used codes and valuation criteria (Khaire & Wadhwani, 2010; Navis & Glynn, 2010); or attempts to evade categorisation efforts altogether (see Granqvist et al., 2012; Vergne & Wry, 2014). Purposes of such strategic action are promotion and demotion through the influence of categorisation or codes, changing intermediaries, but also to evade certain legislation or other implications of categorisation (Granqvist et al., 2012; Kornberger, 2017; Vergne & Wry, 2014). Understanding any act of categorisation as inherently strategic resonates with an understanding of categories as discursively constructed (Glynn & Navis, 2013; Pontikes & Kim, 2017). In this way, both producers and audiences refer to the same discursive resources for sense-making, meaning-making, and evaluation, and negotiate understandings and expectations together. But because only audiences and intermediaries can grant membership in categories, the success of strategic action is subject to the individual perceptions of its targets. Specifically, Kornberger (2017) suggests

37 2. Situating the Research that strategic action for categorisation has to focus on valuation practices, which are “concrete practices through which the work of valuing, weighing, comparing, ordering and categorising is accomplished” (pp. 1767–1768). Thus, competition between different entities can be influenced by the development of valuation practices that become actualised and circulated, and subsequently (co-)mediate between producers, products and audiences (Kornberger, 2017).

2.2 Category Formation

While traditionally, categorisation research was concerned with evaluation within established categories, more recently scholars have investigated how new categories become established (e.g. Khaire, 2017; Alexy & George, 2013; Durand & Khaire, 2017; Khaire & Wadhwani, 2010; Navis & Glynn, 2010; Rhee, Lo, Kennedy, & Fiss, 2017). For a new category to come into place, it firstly needs to be recognised as such: Han- nan et al. (2007) suggest that this is the case when there is a collective recognition of similarities amongst member entities, and a basic agreement about the conditions required to be included in this set, for example concerning the (cultural) codes applied (see also Simões Braga & Silva e Meirelles, 2014). Focusing on artefacts as elements of material exchange, Porac, Rosa, Spanjol, and Saxon (2001) see a category as established when these material objects are connected with shared conceptual systems by buyers and sellers alike. Category codes are abstracted into a shared set of, for example, physical attributes, uses, or values (Negro et al., 2010). Finally, Navis and Glynn (2010) see a category as established when “two or more products or services are perceived to be of the same type or close substitutes for each other in satisfying market demand” (p. 440). For a nascent category to be successful, resources and legitimacy but also associated meanings and occasions are required (Kennedy & Fiss, 2013). Understanding claims for legitimacy as discursive, actors related to a category have been found to employ framing (Khaire & Wadhwani, 2010), analogies (Alexy & George, 2013), identity claims and politics (Granqvist et al., 2012; Navis & Glynn, 2010), certification (Quinn & Munir, 2017), labels, or the (non-)engagement in nonverbal practices such as participation in industry events (Granqvist et al., 2012). The emergence of category identities has been proposed to be the result of distributed agency in collective interpretive work (Grodal & Kahl, 2017; Khaire & Wadh- wani, 2010). Glynn and Navis (2013) describe categorisation as “a cooperative venture between category members and their relevant audiences, rooted in cultural understandings and expectations” (p. 1125). When in this process meaning is asserted, actors may borrow and reconfigure established concepts (Khaire & Wadhwani, 2010; Quinn & Munir, 2017); may appropriate logics, practices, and identities from related fields to

38 2.2. Category Formation normalise a category, or hope for a legitimacy spillover (Alexy & George, 2013; Navis & Glynn, 2010); or may re-interpret existing categories and establish a new order of worth (Durand & Khaire, 2017; Khaire, 2017). An order of worth is essential for audiences to be able to comprehend a category, as it outlines an understanding of significance and quality, suggests how entities are made comparable, and how their values are assessed and compared (Khaire, 2017; Khaire & Wadhwani, 2010; Kornberger, 2017). Related to these origins of new category meanings, Durand and Khaire (2017) suggest distinguishing category creation and category emergence: An emergent category consists of elements originating outside of an existing category system. Conversely, when a category is created, existing entities within a system are re-interpreted and new boundaries drawn. The authors suggest, that more generally category emergence relies on physical or material innovation, in which attributes from an unrelated category are ‘imported’ into another in order to become distinguishable from similar categories (Durand & Khaire, 2017; Wry, Lounsbury, & Jennings, 2013). Therefore, category emergence often brings to light new actors, while on the contrary, category creation tends to preserve the social structures of markets. This is because actors supporting a newly created category are likely to be established as they have the resources required to induce such change, can impart their legitimacy on the created category, and exert their control (Durand & Khaire, 2017). Although the rigid ontology of this approach is different from the constructivist approach taken in this thesis, its implications are not too dissimilar. Glynn and Navis (2013) suggest that new categories build on existing ones by recombining previously existing meanings. Building on an approach of “culture as a ‘toolkit’ or reservoir of meanings” (Glynn & Navis, 2013, p. 1134; see also Swidler, 1986; Giorgi, Lockwood, & Glynn, 2015; Rindova, Dalpiaz, & Ravasi, 2011; Weber, 2005), culture is not a unified and distinct set of values that influences and gives meaning to action; instead it can be understood as a ‘grab bag’ from which private or public new meanings can be assembled or various contradictory behaviours aligned (Giorgi et al., 2015) in equal or hierarchical ways (e.g. Rindova et al., 2011; Wry et al., 2013). Accordingly, categories and their meanings can be recombined from previously existing elements. These new categories may include the unification of seemingly contradictory or mutually exclusive meaning systems (Glynn & Navis, 2013; Quinn & Munir, 2017). In a similar understanding of categories, Kennedy and Fiss (2013) identify categories as ‘social realities’, which emerge from and necessarily relate to specific occasions and motivations. As these usually originate from existing category configurations, categories can emerge as a re-interpretation of familiar occasions or purposes, or as an analogy of a familiar meaning in a new context. In both conceptualisations of category forma- tions, any category and its meaning systems only exists through continued practical or symbolic actualisation that grants legitimacy (cf. Kennedy & Fiss, 2013; Noe & Alrøe,

39 2. Situating the Research

2011). Thus, the acceptability of a given category combination increases as categories or elements are used together more frequently (Kennedy et al., 2010; Paolella & Dur- and, 2015). This in turn requires collectively constructed ‘culture’ shared by many audiences and intermediaries (Giorgi et al., 2015; Small, Harding, & Lamont, 2010; Swidler, 1986), which highlights the importance of culture and discourse on the one hand, and actions and practices on the other. Although this depends on the category and the ways in which producers and audiences interact, intermediaries are generally found to be powerful: They can be gatekeepers (e.g. retailers, see Kornberger, 2017), trusted experts holding symbolic power (e.g. film festivals, see Khaire, 2017), or can prescribe rules for categorisation and valuation (e.g. awards, see Hsu, Roberts, & Swaminathan, 2012). Intermediaries may have direct or marginal financial interests in the categories and may be tied to their own rules or practices of categorisation as they grant legitimacy, provide resources, and can function as a quality credential (Khaire & Wadhwani, 2010; Kornberger, 2017; Navis & Glynn, 2010; Rosa, Porac, Spanjol, & Saxon, 1999). Audiences similarly help to cooperatively build a category in public discourse (Navis & Glynn, 2010), whereby any engagement with products shapes interpretations of emerging categories, leading them to select and reproduce discourses that may fit better with the new category than previously existing ones (Grodal, Gotsopoulos, & Suarez, 2015). It has been suggested, that once a category is established some of its interpretive flexibility is lost and it is evaluated more strictly (Granqvist & Ritvala, 2016; Navis & Glynn, 2010). Navis and Glynn (2010) suggest that while an emergent category is compared as a whole to other categories, in an established category, member entities are compared to each other as the category has rather refined boundaries. For a product category, the authors suggest that it needs to be widely commercialised and available, and commonly seen as legitimate. Because certain attributes are required for category membership, they become internalised and taken-for-granted (Hannan et al., 2007; Navis & Glynn, 2010); thus, entities can be compared on their individual, category- specific differences. Accordingly, it was found that as categories stabilise, evaluation tends to focus on differentiation within the category rather than on differentiation between categories (Navis & Glynn, 2010; Rosa et al., 1999).

2.3 Substitution and Categories

With a rather fixed demand for food products because people can only eat so much, any novel category of foodstuffs inevitably functions as a potential threat to existing categories. While new foodstuff categories can nonetheless be formed, often they might be considered as a substitute for others, if they are deemed sufficiently similar by consumers for the specific reasons the products are consumed. In their conceptual

40 2.3. Substitution and Categories

framework on consumer responses to novel food items, Michaut (2004) outlines an understanding of food substitutability, which similar to evaluative category studies draws among others on psychological and cognitive research on categorisation. They find that perceived resemblance and therefore substitutability of products can be achieved by means of similarity and/or analogy: Similarities are shared attributes, such as colour or texture which are physical aspects or codes; these may be meaningful in themselves, but also in relation to previous experiences with other food categories. Analogies are conceptualised as shared relations, such as goals or scripts. They are thus representative of the meaning systems surrounding foodstuffs which focus for example on practices, rituals, or subjective significance (cf. Kennedy & Fiss, 2013). In category studies the substitution of different categories with each other is not frequently discussed. If substitution is mentioned, it is usually bound to members within the same categories because substitutability is an often employed definition of category membership. For example, Porac, Thomas, and Baden-Fuller (1989) argue in one of the earlier studies on categories that “firms are competitors when they produce products that can substitute for one another in the satisfaction of customer needs” (p. 406). Conversely, in an industry-centred perspective technological similarity is crucial to define competition (Porac et al., 2001). Similarly, as cited before, Navis and Glynn (2010) understand a category as established “when two or more products or services are perceived to be of the same type or close substitutes for each other in satisfying market demand” (p. 440). It has been noted, that demise in one category often aligns with growth in other categories as resources are retracted and established categories are contested and dele- gitimised (Grodal & Kahl, 2017). Kuilman and van Driel (2013) have suggested that this can be analysed as a “sub-category of a broader, more general category” (p. 536), so that disappearance can be treated analytically a result of direct competition (see also Polos, 2002). For example, the demise of global railroad systems in the 20th century (Roberts & Geels, 2018) can be treated as changes within broader categories concerning mobility or transport. This again builds on the idea that substitution can only occur within a category. Kennedy and Fiss (2013) argue that new categories can make existing ones obsolete “not only by adding categories that describe newly recognized social realities, but also by forgetting existing categories or relegating them to history” (Kennedy & Fiss, 2013, p. 1146). Opposite to how categories can come into place in the authors’ framework from new meanings or new occasions (see above), categories can “fall out of common knowledge” when there is no broad necessity for it any longer and “the category label then declines in use and perceived value” (p. 1146); such as horse carriages or spurs which have been outmoded by automotive transport. Another means for change according to the authors is “when changing tastes make labels and what they stand

41 2. Situating the Research for seem old-fashioned or out of date, leading to the gradual disappearance of related ways of doing things” (p. 1146), such as the use of instant messengers instead of SMS or e-mail. Both these ways of category demise can be caused by a newer category focusing on similar occasions or meanings. Accepting that categories can change, Kennedy et al. (2010) outline a variety of ways in which interacting categories can alter their meaning systems: Given two categories with related meanings, either one category can be subsumed by another (i.e. its meaning is incorporated into the other), or substituted (i.e. the newer category extends on expense of the older one). The authors highlight, that this substitution results from ‘substitute relationships’ between these product categories and markets without further defining this. Extending this to three or more categories, category meaning systems can recombine into a single new category by dissolving individual identities, preserving them, or combining elements of the merged categories.

2.4 Changing Culture and Challenging Cultural Dominance

As elaborated earlier, meaning systems are a central symbolic resource in categorisation as they establish “institutionalized definitions that indicate what attributes of a good have value and what attributes do not” (Khaire & Wadhwani, 2010, p. 1297). Meaning systems are historically embedded into a broader cultural context, collectively shaped via discourse and in constant flux, because categories are in a “close and recursive interrelationship between markets, society, and culture” (Khaire, 2017, p. 288). Here, broader cultural change is linked to widespread “beliefs about what attributes of the goods are appropriate and desirable” (Khaire, 2017, p. 288). Similarly, Kennedy and Fiss (2013) have suggested that “systems of categorization and their associated meanings capture and reflect what societies view as social realities, or ontologies” (p. 1138). Acknowledging distributed agency in which actors build on others’ discursive contributions and understandings, Khaire and Wadhwani (2010) suggest that any analysis needs to focus on understanding the “collective interpretive roles of various actors . . . in shaping the types of historical discourses that surround a market” (p. 1298, emphasis in original). Cultural change is thus the large-scale effect of a multitude of adjustments to individual systems of valuation and meaning. However, Rhee et al. (2017) suggest that categories building on hegemonic thought are more durable than those building on ideologies, as hegemonies are understood as unconscious and thus less ‘accessible’ to questioning. This means that actors need to envision a next step as desirable, but also have to be able to imagine it in the first place.

42 2.4. Changing Culture and Challenging Cultural Dominance

In combination with models of category emergence–which propose that categories are based on existing cultural toolkits or existing meanings and occasions (see section 2.2 above)—distributed agency makes radical cultural change and the rapid deconstruction of a hegemony hardly practicable. In fact, theorisations of cultural hegemony suggest that cultural change is usually incremental: An alternative ideology to the cultural hegemony often represents a reconfiguration and adjustment of a dominant ideology (Filippini, 2017; Gramsci, 1971). To be successful, this alternative ideology is required to provide a more enticing meaning system than the current thought; under normal circumstances a too ‘radical’ change would be nearly unthink- able, and audiences would not easily alter their belief systems (Filippini, 2017; Gram- sci, 1971). Therefore, cultural change induced by categorisation or otherwise needs to be incremental and collectively agreeable. Change would therefore focus on altering ‘visible’ ideological categories and their codes rather than on scrutinising ‘invisible’ hegemonic ones (Rhee et al., 2017). Social movement research is often referred to in category studies (Vergne & Wry, 2014), and conversely categorisation is relevant to social movement frames (e.g. Rao, Monin, & Durand, 2003; Weber et al., 2008; see also Negro et al., 2010). Frames are understood as the ways in which attention is limited and guided, for example like a ‘picture frame’ (Giorgi et al., 2015; Goffman, 1974). They have been defined as “sig- nifying work or meaning construction” (Benford & Snow, 2000, p. 614), whereby “to frame [means] to select some aspects of a perceived reality and make them more salient in a communicating text” (Entman, 1993, p. 52). Thereby a particular view on the world is promoted, which “can transform meanings and define a range of acceptable behaviours” (Giorgi et al., 2015, p. 12). Snow and Benford (1988) distinguish the frames employed by social movements as drivers for change into diagnostic frames focusing on problem identification and attribution, prognostic frames suggesting solutions, and motivational frames delineating reasons for action (see also Benford & Snow, 2000; Entman, 1993). Because of their capacity to highlight issues and promote a particular view of the world, framing is also a means of strategic categorisation (Khaire & Wadhwani, 2010; Pontikes & Kim, 2017). Social movement research also appears to support the suggestions made earlier: Wider success of movements is dependent on the desirability and congruence of frames with established cultural, political, or even legal systems. Movements were found to be flexible at the interplay between a movement’s immediate aims, the overall intended achievements, and the actual frames utilised to reach a goal (Benford & Snow, 2000; Snow & Benford, 1992; see also Negro et al., 2010). For example, Noonan (1995) shows that women’s rights activism in Chile during the 1950s failed in their struggle for equal rights under a feminist frame, but succeeded under a maternal frame, which resonated with the values of the political elite. Similarly, Weber et al.

43 2. Situating the Research

(2008) in a study on the emergence of a category for grass-fed beef found that movement success related to the resonance between the movement and the targets; and specifically to whether the mobilisation frames “were tailored to the target’s interest and identity” (p. 560). The access of movements to political institutions, and the level of organisation and influence of the dominant order were also found important de- terminants of success (Negro et al., 2010; Rao, Morrill, & Zald, 2000): For example, Schiller-Merkens (2017) found that over time ethical fashion designers at a prominent fashion show increasingly presented themselves as business-oriented rather than ethics-oriented. This suggests that only widely held, publicised, and agreeable beliefs have the capacity to induce successful socio-cultural changes. Both, frames and categories, are ways in which culture has been conceptualised in context of organisation studies (Giorgi et al., 2015). Although collective action frames conceptualised for social movements and categories might not easily integrate, it has been suggested that strategic categorisation–intentional or not–can be conceptualised as “employ[ing] framing techniques where people present their product, or define a category, in a plausible and targeted way” (Khaire & Wadhwani, 2010; Pontikes & Kim, 2017, p. 75; see also Goffman, 1974). As such, strategic or inadvertent framing conveys information on what the speaker understands as category origins, meaning systems, valuations, or orders of worth, and whether there are issues, solutions, or motivations relating to a category. Therefore, while the impacts and findings from the social movement literature are relevant, frames themselves can be seen as the inevitable actualisation and legitimation of categories in discourse.

2.5 Conceptual Framework

In this section, the reviewed literature is synthesised into a conceptual framework. The description firstly covers a number of ontological and conceptual assumptions and clarifications, and then continues to discuss how categories emerge from a variety of visible and invisible cultural repositories, how categories change, and how the categories contribute to cultural change. Epistemologically, this research builds on a critical realist paradigm (e.g. Sayer, 2000). It is assumed that categories exist as collective constructs related to cognitive processes, and that their manifestations are shaped through social and cultural condi- tioning in interaction with the inevitable learning processes of the human brain (see Shepherd, 2011). As such, categories are not coherent or stable but the sum of individual interpretations and perceptions, whose exact shape depends on individual histories, values, motivations, desires, needs, and the context of categorisation. Category perceptions are thus altered through thought processes and interaction with practices and discourses, such as by means of suggestions, propositions, practical examples, or

44 2.5. Conceptual Framework more generally through texts and actions. As a way to filter, hierarchise, and attribute value, valuation and categorisation are therefore cultural performances that are shaped through the mundane experiences manifested in occasions and meanings (Kennedy & Fiss, 2013)—or analogously in valuation practices (Kornberger, 2017) and discourse (Grodal & Kahl, 2017). In the context of this work, this covers among others what is legitimate to eat, what is tasty to eat, or what is the most appropriate thing to eat in certain situations. From valuation, category boundaries and membership follow as a more or less fuzzy and prominent set of distinctions to other categories. Although valuation is acquired by and continuously updated through the environment, actors may assert different valuations or boundaries to the same category, or describe different categories despite similar valuations. When actors participate in categorisation through discourse, they draw upon their own representations of categories. In doing that, they shed light on the valuation and thus also meanings and occasions they associate with it. As categories are perceived yet contextual realities, any claims inevitably frame and therefore filter and make salient certain elements of the underlying cultural repositories while sidelining or ignoring others. By exposing their own social reality or strategically influencing discourse, they contribute to the development and change of collective repositories. Valuation is therefore individually and collectively historical and contingent. A change in individual valuation can be caused by internal realignments of inconsistencies, or through external influences. On the extreme ends, such influences could result from increased opposition through induced dissonance (e.g. Rothger- ber, 2014) or life-changing tipping point events (e.g. Schäfer, Herde, & Kropp, 2007). A possible leverage point for strategic actors–those consciously aiming to alter others’ categorisation and valuation system–is for example to influence intermediaries’ valuation practices, which include for example the physical availability and arrangements of products in grocery stores, reports in food blogs, or newspaper coverage. Influences on valuation and meaning systems of audiences might focus on direct or indirect communication and signposting, for example through in-store promotions, social media, packaging, or advertising, which could propose how or in which contexts products ought to be evaluated. Further influence can be achieved by labelling (e.g. Schiller-Merkens, 2017; Granqvist et al., 2012) which (un-)claims labels and thus category membership, and thereby suggests specific valuation criteria. Generally, any description can be categorising (Grodal & Kahl, 2017) and thus propose or influence certain meaning systems. However, as specific meanings need to be actualised or reproduced through the audience, a variety of meanings can emerge from the same entity (Noe & Alrøe, 2011). Therefore, in case of foods, actors continuously engage in categorisation on the basis of their consumption, the associated cultural repositories, and the frames utilised to rationalise these. — In the following paragraphs, a process

45 2. Situating the Research of cultural change is put forward, which builds on the outlined relations between categories and agents. The framework is illustrated in Figure 2.1.

(1) At any time there are numerous cultural resources embedded in established category systems. They carry meanings and occasions which are actualised through discourses and practices and thus uphold and represent certain valuations (Glynn & Navis, 2013; Khaire & Wadhwani, 2010). Notably, not all cultural resources are acknowledged as such: Some are more visibly marked as a specific, subjective, and potentially deviant way of thinking or doing things (i.e. an ideology), while others are hidden as they are understood to be ‘normal’ or a dominant way of doing things (i.e. a hegemony, cf. Rhee et al., 2017).

(2) Any new category needs to draw on these established categories. As part of their generative capabilities, they can be recombined and sprout new categories in a variety of contexts and purposes (Glynn & Navis, 2013). Through this generative process, existing or emerging tensions and inconsistencies amongst a wider array of categories, cultural repositories, and social realities can be resolved (Khaire & Wadhwani, 2010).

(3) As new categories invariably rely on previously existing categories and their embedded meanings, occasions, discourses, or practices, they can not deviate too much from existing ways of thinking. In an attempt to appeal to specific audiences, the resolved tensions are legitimised. Category meanings will invariably tend towards less ‘friction’ to meaning systems established amongst its target audiences. In doing so, certain aspects will be highlighted, and others dropped or hidden as an internalised membership criterion to the category. When categories ought to appeal to larger or mainstream audiences, those category conceptualisations with less ‘friction’ and more resemblance to dominant understandings are more likely to persist and succeed. Hegemonic culture is therefore firmly protected simply by being prevalent amongst most actors’ internal meaning systems and valuation frameworks which shape the modes of physical, discursive, and symbolic exchange.

(4) Any category contributes to cultural repositories. Through their generative capabilities, categories become resources themselves that can be drawn upon by other or new categories, and thereby contributing to cultural change. Then, the new category itself may be updated, merged, amended, or introduced to different contexts.

(5) As occasions, meanings, discourses, or practices have evolved with the category and are actualised somewhat differently in different actors, they also have vary-

46 2.5. Conceptual Framework

ing influences on the cultural elements they build on. These different understandings of a category reflect perceived realities and strategic interests of actors through their emphasis on some frames1 over others, which themselves are based upon categorical meaning systems. As described above, the actualisation of categorisation in text or action frames and therefore highlights some elements of a category most relatable or most amenable to an actor’s own cultural understanding. This may be employed strategically, because particular understanding of a category is made more salient and thus legitimised. As these voiced categorisations then participate in a wider discourse, any actualisation not only affects categories themselves, but also updates and rephrases original cultural repositories and their associated meaning systems.

Part of an actualisation is what and how aspects of a new category are made salient as a new cultural resource. For example, if a more visible cultural resource is combined with a more invisible one as displayed in Figure 2.1, they can be mixed and realigned in several ways. To align with dominant culture, many actualisations of the emerging will tend towards invisibility within mainstream cultural understandings. Regardless of the engaged agents, this may be understood as an incremental amendment to or support for the hegemonic order. Less likely is that more visibility is retained through categories’ deviance, which in turn may cause them to be branded as ‘ideological’ and to be less likely to acquire mainstream support. However, continuous exposure even to such deviant, new or not-yet amenable categories can make them more accepted. Any of these configurations and frames promote a different identity that becomes amenable to different individual identities: By choosing a certain symbol and language in specific ways, they underline their position and view on the world, which may for example be agreeable, business-as-usual, or radical (e.g. Chong & Druckman, 2007; Pontikes & Barnett, 2015). As conceptualised here, categories can be analysed and understood only in view of their individual and collective historic contingency (cf. Glynn & Navis, 2013; Khaire & Wadhwani, 2010). Throughout their emergence they can acquire and drop elements from various categories while the focus shifts from external to inherent legitimation (Navis & Glynn, 2010). In this light, the codes2, category claims, and references made throughout are important to evaluate a category’s origins, relations, and underlying

1The understanding of frame employed here does not resemble a collective action frame from the social movement literature (e.g. Snow & Benford, 1988; Giorgi et al., 2015) in a sense that it politicises and pushes a specific collective agenda. While a collective effect may still be observable, this project understands frames as reflective of perceived realities and personal experiences (Goffman, 1974) which for some actors may be more strategic and calculated than for others (Pontikes & Kim, 2017). 2Note that in the following chapters to avoid confusing with the concept of codes in content analysis, instead of ‘codes’ I use the term ‘characteristics’ or occasionally ‘descriptions’.

47 2. Situating the Research

alignment with hegemony 1. cultural resources carry 4. legitimised categories meanings and occasions extend existing cultural repositories and can be recombined into new categories visible challenge 5. frames legitimise invisible / and make salient ‘hegemony’ speciﬁc cultural components and visible / thereby amend ‘ideology’ existing repertoires

2. tensions can be observed and exploited 3. categories develop to form new categories towards less friction with dominant ways of thinking when approaching mainstream

Figure 2.1: Visualisation of the conceptual framework. meaning systems. The frames that portray categories in a specific context can therefore be analysed to understand an actor’s individual positioning towards a category, its multiple possible configurations, and its implicitly or explicitly retained category references. The proposed conceptual framework may be useful to scholars of categories for the following reasons: Firstly, it allows conceptualising cultural change as both a source and result of categorisation; secondly, it understands categories as continuously related and malleable through shared cultural repositories beyond horizontal competition or vertical nestedness; and thirdly, it integrates frames and meaning systems as mutually constitutive in the socio-cultural study of categorisation. Assuming the suggested framework was valid, the following observations should be made (the numbers correspond to Figure 2.1 again; (1) is an underlying assumption and thus not discussed further): (2) Firstly, ideas rooted in previously existing categories should be visible in the creation of new categories. This can be observed in references and comparisons from the analysed category to other categories as meanings and practices are asserted (Navis & Glynn, 2010); a failure to observe such references to one or more other categories would refute this foundational assumption of the conceptual framework. If the category originates from the resolution of tensions, this is likely to be observed in reported achievements or progress. (3) Throughout its emergence process, it should be observed that the category’s visible deviance to dominant cultural ideas and their categories increasingly diminishes. Following the findings

48 2.6. Methodologies to Study Categories of Navis and Glynn (2010) it should be expected that a category becomes independent, but in the context of this research project it is also plausible that similarities to the hegemonic category are highlighted to ensure invisibility. What should not be observed is that a highly visible and thus deviant category becomes mainstream. (4) Once a category has acquired foundational legitimacy, its generative capabilities may be exhibited in that other categories or understandings draw on the established understandings. Assuming that legitimacy has been granted, this may or may not be observed in a given sample, but cannot be refuted because it is assumed in the framework (cf. Glynn & Navis, 2013). (5) Lastly, the observed frames may highlight different aspects of categories, but most frames should be observable or implicit in every representation of the category. Influences of the new category on existing categories can be observed in the negotiation of existing boundaries, ongoing comparisons between the categories, and frames which are utilised to (re-)present the new category. For this, comparisons and frames are essential to identify interactions between the categories. Conversely, if the newly established category and its originators are kept separate entities and are not compared or evaluated alongside each other, and they exhibit no shared understandings or meanings, it can be assumed that no influences exist.

2.6 Methodologies to Study Categories

The early work on categorisation in a cognitive evaluation framework relied heavily on quantitative data, and insights were gained by proving hypotheses (e.g. Zucker- man, 1999; McKendrick, Jaffee, Carroll, & Khessina, 2003; see also Vergne & Wry, 2014). Even constructionist arguments are regularly validated or supported by quantitative theory testing or assessments (e.g. Kennedy, 2008; Navis & Glynn, 2010; Pachucki, 2012), while few publications are based on purely qualitative methodologies (e.g. Jamerson, 2009; Quinn & Munir, 2017; Schiller-Merkens, 2017). Most studies also draw heavily on contextual data to construct case studies and guide interpretations (e.g. Ruef & Patterson, 2009; Khaire & Wadhwani, 2010; Paolella & Durand, 2015). The quantitative data broadly cover transaction catalogues, monetary values or performance measures, contextual figures, or assessments of evaluations or rankings (e.g. Khaire & Wadhwani, 2010; Durand, Rao, & Monin, 2007; Paolella & Shar- key, 2017). Sometimes that data is not originally numeric, but instead represents frequency counts or is based on content analysis, for example of interviews or press releases (e.g. Navis & Glynn, 2010; Pachucki, 2012). Qualitative data used include, among others interviews, archival resources, news, biographies, magazines, catalogues, press releases, industry reports, journal articles, books, or online discussion forums (e.g. Khaire & Wadhwani, 2010; Navis & Glynn, 2010). As many studies employ a

49 2. Situating the Research diversity of data, the need for integrative designs of multiple data sources and approaches is often highlighted. Data and/or method triangulation is often applied to obtain “complete, holistic, or contextual” (Jick, 1979, p. 603; cited in Khaire & Wadh- wani, 2010) findings and interpretations, as theorised relationships and narratives can be confirmed, rejected, cross-validated, or contradicted (see also Flick, 2018). As meaning systems are central to the study of categories from a social perspective (Kennedy & Fiss, 2013), the studies most similar to this research project frequently employ discursive readings of their data sources. They understand data as narratives, frames, targeted stories, or parts of a wider discourse (e.g. Khaire & Wadhwani, 2010; Quinn & Munir, 2017; Schiller-Merkens, 2017; Weber et al., 2008; see also Giorgi et al., 2015). Analytical strategies vary with the research questions and the gathered data, and usually involve multiple steps of analysis and abstraction. Approaches used in relevant studies include inductive and deductive coding (Granqvist & Ritvala, 2016), case study research (Quinn & Munir, 2017), grounded theory (Granqvist et al., 2012), discourse analysis (Khaire & Wadhwani, 2010), content analysis (Navis & Glynn, 2010), frame analysis (Schiller-Merkens, 2017), or semiotics (Weber et al., 2008). The conceptual framework and its implications outlined above suggest themselves to favour particular data and analytical approaches over others. Firstly, a ‘thick description’ of history and contemporary context is required to interpret any findings meaningfully and evaluate validity and transferability of the findings (e.g. Quinn & Munir, 2017; see also Langley, 1999). Secondly, category emergence may be investigated using a longitudinal content analysis similar to Navis and Glynn (2010), which the authors have shown to be able to uncover category dynamics and relations. Thirdly, an approach combining a variety of data sources, both from producers and consumers, can increase the validity and usefulness of the outcomes, as actors contribute and are essential to categorisation (see Granqvist & Ritvala, 2016; Grodal & Kahl, 2017). Fourthly, it has been demonstrated that frame analysis can uncover category relations in the interplay between business and activism (e.g. Schiller-Merkens, 2017; Weber et al., 2008), whilst in literature review and conceptual framework, frames have been suggested as a meaningful way to study category relations. Lastly, the framework outlined above and the variety of required data to obtain reasonable insights increases the need not only for triangulation, but also for inductive, iterative approaches (e.g. Langley & Abdallah, 2011), which allows interpretations of various contexts and stages of this work to influence one another (Granqvist & Ritvala, 2016).

50 Chapter 3

Methodology

This research project investigates how dominant meat culture is challenged and reproduced through meat alternatives. In this chapter, the research questions posed in the Introduction are operationalised and methodology, methods, and data are elaborated in detail, to answer this research question on the basis of the conceptual framework proposed in the previous chapter. The findings of the data sets introduced here are presented in the following chapters 4 to 8. These are then synthesised concerning the three research questions and presented in chapter 9. A final discussion and reflection of the findings is offered in chapter 10. The remainder of this chapter proceeds as follows: In the next section the research design is introduced, which covers a broad summary, underlying assumptions for this research project, and detailed individual research designs for each of the three research questions. In the next section, the main methods used are described. Thereafter, each of the data sets is discussed in detail, which includes a detailed rationale for its usage, contextualisation in the literature, sampling procedure, as well as methodological considerations and limitations. Supporting material is referenced throughout and can be found in the Appendix. Finally, limitations to the overall research design are discussed.

3.1 Research Design

3.1.1 Overview of the Research Design The research design was devised on the basis of the research questions outlined in the introduction, the conceptual framework, and the identified previous methodologies to study categorisation. Broadly, meat alternatives are understood here as commercial foodstuffs which are identified as similar to meat in the texts drawn upon, for example because they fulfil the same functions or have similar characteristics (see again section 1.3). As previous academic literature has not specifically investigated the issues studied in this thesis, this research explores qualitatively the meanings of meat alternatives and their capacity of cultural change in relation to meat.

51 3. Methodology

To achieve exhaustive, generalisable insights on meat alternatives, data sources relating to various consumer and producer perspectives have been chosen: One source respectively covers data with few data points which are detailed (manufacturer and stakeholder interviews, and taste test videos from YouTube), and a second source each covers many data points with less depth but likely a higher diversity of represented attitudes (print advertisements, and tweets related to meat alternatives during two meat-free theme weeks). In order to contextualise and give depth to the insights, this was complemented with contextual data to provide an in-depth account of a narrative of historical developments in the category, as well as recent insights from the academic literature and market research on meat alternatives. To investigate all data to similar levels, each data set was initially analysed independently using a version of qualitative content analysis appropriate for the data (see Figure 3.1 and section 3.2 below). After this, insights were merged by research question and discussed in detail. The first research question concerns the emergence of the meat alternatives category and aims to identify whether it is an established category, and how it has changed or remained stable. This is based on the historical narrative and a longitudinal study of advertisements as a timely point of interaction between producers and consumers; for this, the advertisements have been coded, and time periods in the development distinguished. A χ2-test was used to identify which trends and developments would warrant in-depth investigation. The second research question concerns the current category identity, and is motivated by getting a better understanding of how meat alternatives are described, evaluated, and labelled. The analysis is based on all four data sets1 (although only advertisements printed in 2016 or 2017 were used; the other data sets present a recent perspective) and comprises a largely descriptive assessment of the characteristics, labels, and comparisons made when describing meat alternatives. This research question is relevant to identify whether the category is described in ways coherent enough to make any generalising claims, and to assess whether and how the category does relate to meat. The third research question builds on the previous insight that the category exists, is coherent, and relates to meat, and concerns how meat alternatives and by proxy meat are framed in texts about meat alternatives. To achieve this, prominent (diagnostic, prognostic, and motivational) frames are identified in each data set and discussed; these different frames were then gradually integrated and underlying legitimation strategies discussed in order to understand whether and how meat alternatives reproduce and challenge dominant meat culture. In Figure 3.1 the research design is

1Obviously, the contextual data is a data set in itself as well. Because it stands apart by not directly being subject to the main research questions, I refer to data sets or four data sets for simplicity and clarity.

52 3.1. Research Design summarised. In the following section, the assumptions relating to the research design and the research question will be recapitulated. Thereafter, the research design will be discussed in more detail in relation to the three research questions (sections 3.1.3, 3.1.4, and 3.1.5), after which methods (section 3.2), data (section 3.3) and limitations of the project (section 3.4) are elaborated on.

3.1.2 Assumptions and Considerations From a theoretical perspective, the research conceptualises meat alternatives as a socio- cultural category that has relations to at least one established category–meat—and its associated meaning system. The outlined conceptual framework assumes that categories exist as a cognitive and social construct for individual and collective sense-making. Categorisation is understood to be emergent from discourse and practices while categories themselves are interpreted as assemblages of previous cultural elements that are re-conceptualised and re-contextualised in a new category (e.g. Glynn & Navis, 2013; Kennedy & Fiss, 2013). Category studies further assume that a category can be observed, identified, and analysed from a third-party researcher like me through texts on the basis of talk about category members (see again the assumption made for the conceptual framework in section 2.5). The overall research question broadly assumes that meat alternatives are considered as an established category, that a meat alternatives category is coherent enough to be investigated as a whole, and lastly that understandings of consumers and producers retain links to a meat category. These assumptions have to be dealt with in preparation to answering the research question, and are represented by research questions Qa (see section 3.1.3) and Qb (see section 3.1.4). The research empirically and conceptually depends on the fact that British food culture strongly focuses on meat and may be considered hegemonic. The prominent position of meat in British diets and meals has long been studied or identified (e.g. Warren, 1958; Douglas & Nicod, 1974; Murcott, 1982; Yates & Warde, 2015). Cur- rently, around 85% to 95% of the population consume meat on a regular basis, while an average British consumer eats around 60 kg of meat annually, and spends around a quarter of the total food expenditure on meat (see section 4.4 for a summary of meat consumption in the UK). Meat consumption has been attributed with a strong symbolic status in Western cultures (Fiddes, 1992), and was identified to exhibit many invisible assumptions and value systems defining what is legitimate to eat (Joy, 2009). Breaking with at least the animal origin of meat products, meat alternatives may be considered political, and a hypothesised category of meat alternatives becomes part of an ideological struggle. Contrary to previous studies on categorisation (cf. Quinn & Munir, 2017; Schiller-Merkens, 2017), this ideological struggle is exerted through

53 3. Methodology

Data Choice Analysis Producers Consumers

Detailed I V open, iterative content analysis

Diverse A T exploratory coding, guided content analysis using a coding schedule I nterviews V ideos A dvertisements T weets Timeline Research Design A ¹+ Context A (Qa) Category Emergence I narrative developments, interpretation of changes and constants of coded advertisements (supported by χ²-test) V informs Collection for each RQ A ² I V T

T separate Analyses (Qb) Category Identity descriptive assessment of labels, characteristics, and 2017 2018 product comparisons based on content analyses informs A ² I V T (Qc) Frames and (Meat) Culture A ¹from 1985–2017 ² frame analysis, identiﬁcation of salient issues A from 2016–2017 legitimising meat alternatives and (by proxy) meat allows to answer the main Research Question How does the category of ‘meat alternatives’ reproduce and/or challenge the dominant meat culture?

Figure 3.1: Visualised overview of data choices and research design. On the top, the choice of data and the different analyses used are mentioned. On the bottom, a timeline of data collection and analysis is presented, and the different research questions and their use are summarised briefly. everyday food choices, which are suggested to engage in cultural politics through their challenge and reproduction of dominant meat culture.

3.1.3 (Qa) Studying Emergence How did the meat alternatives category evolve whilst becoming mainstream? (1985–2017)

Studying the emergence of a meat alternatives category is relevant to understand its origins in discourses (Khaire & Wadhwani, 2010), as cultural or institutional roots from category antecedents or cultural repositories (Bajpai & Weber, 2017; Glynn & Navis, 2013), or as a process of acquisition of legitimacy and identity (Navis & Glynn, 2010). The approach to seeing meat alternatives in context of the category’s development builds on the literature on category emergence (see section 2.2); deriving from the review, I identified the following as essential developments as categories become established: (1) descriptions used for category members stabilise; (2) members are more

54 3.1. Research Design strongly differentiated within the category, instead of categories from each other; and (3) legitimacy of the established category is unquestioned and does not need to be claimed. Deriving from the conceptual framework it should further be observed that (4) category antecedents are highlighted in early stages of the emergence process; (5) if the category antecedents are in tension, there are references to achievements or progress as resolutions of the tensions; (6) the ‘friction’ with hegemonic categories decreases over time; (7) visible links to the hegemony may prevail. From these understandings of category emergence, indicators guiding the analysis were operationalised, which can be summarised in three guiding questions:

1) How have labels and descriptions used for meat alternatives changed over time?

• attributes used for describing products, e.g. succulent, healthy, vegetarian, frozen, tasty; relates to (1) as characteristics2 should stabilise towards later time periods • category labels used to denominate the product, e.g. meat-free alternatives, frozen foods, family classics, bbq bangers; relates to (4) with labels as proxies for category antecedents that should be more highlighted in earlier time periods

2) What other products have meat alternatives been compared with over time?

• references to other categories and competition, e.g. like meat, better than vegetarian food, better than other meat alternatives; relates to (2, 4, 7) exposing intra- and inter-category references throughout time, with (2) suggesting more intra-category references in later time periods, (4) suggesting references to other categories in earlier time periods, and (7) suggesting continued links to meat as a hegemonic reference

3) How have suggestions about the supposed relevance and context of products changed?

• theme of the text as evaluative guideline or context, e.g. christmas, taste, animals, barbecues, agriculture, brand quality; relates to (6) as indication of identified issues, contexts, or other propositions, which should reflect a change from highlighting issues to highlighting affordances • portrayed achievement or novelty of the product, e.g. environment- friendly products, animal rights concerns, unprecedented taste, particip-

2Note that as argued in the previous chapter, the term characteristics is used to denote empirical codes

55 3. Methodology

ation at christmas dinners; relates to (3, 5) with stated achievements being indicators for acquiring legitimacy and a notion of progress, whereby both predictions suggest more statements of progress in earlier time periods

To answer these research questions a longitudinal content analysis is employed. Following the approach outlined by Navis and Glynn (2010), the content analysis is supplemented with a narrative in order to identify time periods3 and a start dates (here 1985, see section 4.3), and to contextualise emerging interpretations. The narrative is constructed from reporting of the market and its developments, such as in news, market research, academic research, or relevant books and bibliographies; changes in reporting were understood as indicative for new time periods, a detailed elaboration of this can be found in section 3.3.1. The above mentioned indicators are therefore analysed regarding their behaviour throughout different time periods related to the respective predictions. As a comprehensive longitudinal data set, advertisements for meat alternatives were collected. They were found to span a sufficiently long time period, are reflective of social realities of meat alternatives at the interplay of producers and consumers, and allow investigating the sense-making and legitimacy claims of meat alternatives. Conversely, news or industry reports which are also involved in meaning making are an intermediary perspective and likely aimed at a rather general audience without always claiming legitimacy. Furthermore, they might not cover a niche market to sufficient detail to study it appropriately. Advertising is frequently analysed using content analyses both comparative and longitudinally (e.g. Fay, 2013; Pollay, 1985; Riffe, Lacy, & Fico, 2014; see also Navis & Glynn, 2010). The advertising was collected from three vegan/vegetarian magazines as vegetarians represented the target audience of early meat alternatives; in other, not targeted magazines, hardly any advertisements for meat alternatives could be identified in recent issues. A detailed approach of the sampling and analysis of the advertisements can be found in section 3.3.2. Within the scope of the outlined indicators, exploratory coding was performed on a subset of the collected advertisements to devise a coding manual which was then used to code all advertisements. The detailed codes resulting from the content analysis were aggregated, and their timely development was assessed on the basis of the time periods established in the narrative. Developments and changes were assessed using χ2-tests for each code across all time periods. This statistical test is often used for content analyses and calculates the likelihood of observed patterns emerging by chance (e.g. Riffe et al., 2014; see also McHugh, 2013). The outcomes of these stat-

3Note that Navis and Glynn (2010) identify two time periods, one before they deem a category as established, and one after. They use t-tests to compare the two based on a number of hypotheses.

56 3.1. Research Design istical test were used to assess the non-randomness of observations across the time periods; to avoid over-interpretation, only clear, non-zig-zag trends were further interpreted. Coding and analysis are further described in section 3.3.2 below. The resulting analysis in section 9.2 provides a rich analysis of these insights from content analysis (see section 5.2) and the developed narrative (see section 4.3)

3.1.4 (Qb) Investigating Category Identity How are meat alternatives described and related to in the present? (2016–2018)

Category identities imply a number of characteristics and criteria for their member entities, so that the foundational description of a category is taken for granted. A category identity is the result of general agreement for what category membership needs to constitute, and which symbolic or material resources are shared despite differences between entities (e.g. Negro et al., 2010; Vergne & Wry, 2014). This research question is required for the third part of the research, as it needs to be identified whether category links to meat persist. Compared to the other two research questions, this one is rather descriptive; it extends and contrasts the findings from the previous research question with more data and in the present; this should lead to a detailed understanding of the category of meat alternatives, covering among others how it is used today, and whether it relates to other categories. Therefore, the concepts to be observed here are not substantially different from those in the previous section. Building on the literature review, of particular interest is: (1) how category members are described; (2) whether products are related to other products or categories; (3) how products are valued and evaluated; and (4) whether strategic categorisations occurs and what labels are used. Relating to the conceptual framework, it should further be observed that there are: (5) contested boundaries from meat alternatives to meat; and (6) ongoing comparisons or references to the meat category. Based on these considerations, this part of the research project is operationalised into three guiding questions:

1) What characteristics are used to describe the products?

• characteristics that describe products, e.g. taste, texture, appearance, smell, usefulness, cooking; relates to (1) and allows to reconstruct a category identity

2) How are products and their characteristics evaluated and compared? What are they compared to?

57 3. Methodology

• characteristics used to establish product differences, e.g. taste, texture, environment, looks; relates to (3) and allows inferring what is valued in products • means to evaluate these characteristics, e.g. tastes good, tastes better than other products, tastes like meat; relates to (3, 6) to investigate how valuation is done and in which ways these relate to other categories • other categories referenced, e.g. tastes like sausages, make spaghetti Bo- lognese with it, as many vitamins as fresh vegetables; relates to (2, 6) in order to identify symbolic dependencies and links to other categories 3) Are there any implied or strategic category claims or labels? • labels used for meat alternatives, e.g. part of a vegetable category, deli slices, just a burger; relates to (4) to understand what category labels are used for the products • suggested ways to see other categories in relation to meat alternatives, e.g. meat and meat alternatives are both protein, it’s the same thing; relates to (5) as it allows investigating how boundaries are claimed and re-considered

Unlike during the emergence process, more recently there is no shortage of coverage on meat alternatives in all data sets as the products have gained a wider popularity. As illustrated before, an approach combining various data sources by producers and consumers increases the usefulness of prospective findings. On top of a recent subset (2016–2017) of the already collected advertisements–to not impact the findings with historical advertisements–, producer data were generated by conducting semi- structured expert interviews with meat alternatives manufacturers and stakeholders. These account for a manufacturer perspective, as well as further and background understanding of the market, brands, histories, and visions. For consumers, I decided to use user-generated content related to meat alternatives from social media platforms. Sampling from these platforms has the advantage that the data is available and accessible, that users have some interest in the products as they by their own choice refer to it in their content or messages and are thus the texts of interest (see section 1.3). Further- more, any researcher bias as an interviewer or focus group facilitator does not need to be accounted for in the data itself but only in sampling and analysis (cf. Miyazaki & Taylor, 2008). In particular, for the user-generated content I chose firstly to analyse taste test videos from YouTube. In these, users eat and then comment on foods (here: meat alternatives) in front of their camera. As speech, they have a similar information density as the interviews, although most videos are shorter than the conducted interviews. Secondly, I decided to use short messages from the Twitter platform as they would provide a breadth of different perspectives from various consumers and

58 3.1. Research Design possibly producers in condensed form. Detailed accounts of the rationale to use these different data sets and how the data was sampled can be found in section 3.3. Following common approaches to study complex cases (cf. Eisenhardt, 1989), analyses for each research question were approached separately for each data set before they were combined. Furthermore, due to the different nature and content of the data, I used different inductive content analytic approaches; section 3.2 describes in detail the methods used, and section 3.3 the different specific analyses of the data. Tweets and advertisements were analysed using a guided content analysis based on the operationalised indicators, while a coding manual was devised through exploratory coding. Videos and interviews were analysed using more open coding whereby the guiding questions and indicators were used to identify relevant text passages, which were then summarised and analysed later. Following a more general inductive approach–and due to resource and complexity constraints through the variety and diversity of the data–, focus was limited to the more important and more prominent themes emerging in the data sets (see Thomas, 2006). After all data sets were coded and interpreted individually, they were combined so that the different underlying themes and understandings could be complemented. While points of congruence can be seen as strengthening validity of the insights, divergence may on the one hand invalidate findings, yet on the other hand “can often turn out to be an opportunity for enriching the explanation” (Jick, 1979, p. 607). Throughout the analysis a digital journal was written to hold on to ideas and challenges involved. Based on the findings from the separate data sets, the analysis in chapter 9 was approached iteratively so that emerging interpretations evolved as codes, raw data, research questions, or the literature were revisited; this allowed interpretations of various contexts and stages of this work to influence each other (see also Granqvist & Ritvala, 2016).

3.1.5 (Qc) Analysing Cultural Change Through Frames How are meat alternatives (and meat) framed, and how does this relate to dominant meat culture?

The third and last part of this research project investigates the relations between meat and meat alternatives, whose existence is implied in the main research question: How do the ways in which meat alternatives are framed reproduce and challenge the norms established by the dominant meat culture? In their wording, these questions already reflect the insight that frames are a suitable and helpful way to understand culture, particularly in this research project (see again sections 2.5 and 2.6). According to initial assumptions of the research topic, and as the previous research question should

59 3. Methodology have shown, there are ongoing references to a meat category for the category’s identity. This third part of the research builds on these findings and investigates closer the meanings attributed to meat alternatives, and how those relate to or represent meat culture. The concept of frames is used here to investigate perceived individual realities in relation to meat alternatives; building on the nearly ubiquitous references to meat across meat alternatives, these frames for meat alternatives also convey a specific understanding of meat or a meat category (see again the conceptual framework in section 2.5). Framing–strategic or inadvertent–is understood here as a means “to select some aspects of a perceived reality and make them more salient in a communicating text” (Entman, 1993, p. 52), which can at large be separated into solution-oriented (prognostic), problem-oriented (diagnostic), and motivational frames (Benford & Snow, 2000; Snow & Benford, 1988). As further argued in section 3.2.2 below, this conceptualisation was amended with an alleged absence of these frames, whereby a specific ‘normality’ is displayed. Building on the conceptual framework, the following observations should be made: (1) the frames used for a category relate to an understanding of linked cultural repertoires, that means that a specific understanding of meat alternatives implies a specific understanding of meat; (2) counter-hegemonic (problem-oriented) frames had become part of the implicit category identity throughout the emergence possibly because ‘friction’ was reduced. These can be conceptualised into two guiding questions:

1) How are texts about meat alternatives framed?

• frame for the text, e.g. tasty food, healthier alternative, childhood food, alleviation of guilt; relates to (1, 2) as a means to identify the different issues, solutions, motivations, and normalities involved

2) How is meat framed in texts about meat alternatives?

• frame for meat, e.g. (un)healthy, proteinaceous, causing animal suffering, tasty, traditional, socially accepted; relates to (1) as a means to investigate implicit or explicit assumptions about a hegemonic issue or ideal

Returning to the original research question, these identified frames allow inferring challenges and reproduction of dominant culture: Meat could be framed as something positive, such as sentimental or tasty, but it could also be framed for example as a problem in itself, a cause, or may imply moral judgement. Similarly, meat alternatives could be framed as a solution to a ‘meat problem’, as a reconciliation of conflict- ing interests, or as a solution to a problem unrelated to meat, such as the quality of meat alternatives on the market. A challenge would thus be implied if frames focused on non-human animal meat as problem, cause, or subject of moral judgement, with

60 3.2. Methods solutions emphasising difference to meat; conversely, reproduction of dominant culture would be identified if frames highlighted other issues that require products to be more similar to meat or reiterated a normality of meat consumption. As the two proposed questions are difficult to disentangle and may be seen as the opposite side of the coin, they are not analytically disintegrated in the interviews and the tweets. Just like in the previous study, the same methods and the same data can be used, albeit the focus on frames requires a more interpretive analytical lens whereby a frame was identified for a whole text: For the advertisements and the tweets as a coded ‘dominant meaning’ that was expanded using all similar meanings into the three frame component listed above. Conversely, for videos and interviews, each component was identified per text; the specific analytical procedure is further described in section 3.2.2, and eventual peculiarities in the respective analysis sections of each data set. As the analysis builds on the findings from the previous study, the analysis was performed only afterwards, so that it could build up on any findings and emergent themes; however, for tweets and advertisements the frames had been coded in parallel to the previous research question. Generally, the overall analysis and synthesis followed a similar approach as the previous research question. Specific differences are that the frames are compared and aggregated on the basis of the three frame components, as well as their portrayal of meat alternatives and meat respectively (see section 3.2.2). The emerging frames are then disintegrated, and discussed, and thereafter some mechanisms elicited about how meat alternatives are used to engage with and thereby reproduce and challenge dominant culture. — Table 3.1 summarises all research questions, the used data, and an indication of the operationalisation.

3.2 Methods

In this section the methods used in this research project are grounded in the literature and their specifics outlined: content analysis, and frame analysis. Generally, qualitative content analyses were applied throughout this thesis. In case of the advertisements and the tweets, the frame analyses carried out as part of the third research question (Qc) built on the same approach of guided coding. For the interviews and videos, frame analysis was conducted on the overall text, and not on text snippets like the previously conducted open content analysis.

3.2.1 Content Analysis Content analysis is a widely used approach to make inferences from texts (Weber, 1990). It is frequently applied in research about media (e.g. Krippendorff, 1980; Riffe

61 3. Methodology , employed , -tests 2 product description , Operationalisation asserted novelty , used label/headline comparisons to other products/categories themes product characteristics characteristics used for comparison means to evaluate these codes references labelling and category claims suggestions how to interpret the category in relation to others frames for meat alternatives frames for meat adverts adverts, interviews, videos - archival data narrative - interviews, videos open content analysis - relations: interviews Summary of the research design. The table covers all research questions, data, methods, and a brief summary of the operationalisation. (Qa1) How have labels and descriptions of meat alternatives(Qa2) changed What over other time? products have meat alternatives been compared with over time? adverts (Qb1) What adverts codes are used to describe the products?(Qb2) How are products and their characteristics evaluated and compared? What are they (Qb3) Are there any implied or strategic category claims or labels?(Qc1) How are texts about meat alternatives framed?(Qc2) How is meat framed in all texts about meat alternatives? - claims: adverts, tweets all all (Qa3) How have suggestions about the supposed relevance and context of products changed? Research Questioncompared to? dominant meat culture? Data Method / - interviews, videos open content analysis, frame analysis (Qa) How did the meat alternatives category evolve whilst becoming mainstream? - adverts (1985–2017)(Qb) How are meat alternatives described guided and content related analysis, to χ in the present? - adverts, tweets(Qc) How are meat alternatives (and guided meat) content framed, analysis and how does this relate to - adverts, tweets guided content analysis, frame analysis Table 3.1:

62 3.2. Methods et al., 2014), the study of advertisements (e.g. Pollay, 1985; Fay, 2013; Sixsmith & Fur- nham, 2010), social media (Basáñez, Majmundar, Cruz, & Unger, 2018; So et al., 2016), interviews or focus groups (Graneheim & Lundman, 2004; Kondracki, Wellman, & Amundson, 2002). Vergne and Wry (2014) note that content analysis has also been increasingly used in category studies “to examine the cultural meanings associated with specific categories” (p. 65). A standard approach to content analysis was outlined by Krippendorff (1980) and proposes data analysis in a replicable and valid manner through systematic and guided coding, for which operationalised indicators are used to develop coding categories and a coding manual which is applied to each datum. Al- though the authors acknowledge that any content analysis is inherently qualitative as researchers actively read and interpret the texts in order to arrive to codes, dedicated approaches for qualitative content analyses have been suggested which account for interpretive flexibility (e.g. Hsieh & Shannon, 2005; Elo & Kyngäs, 2008). Qualitative content analyses have been previously distinguished for example into deductive and inductive (Elo & Kyngäs, 2008), or into conventional (descriptive), directed (theory- driven), and summative (focused on word use) approaches (Hsieh & Shannon, 2005). Codes elicited in the content analysis can then be used for further qualitative or quantitative assessment. Quantitatively, codes are understood as metrics and used to investigate and test differences using statistical measures, sometimes exploratory, but more often to test hypotheses. Depending on the type of data coded and the underlying ideas to be tested, these include t-tests, χ2-tests, or regressions, and depend on data, codes, and hypotheses (Riffe et al., 2014). In qualitative research, content analyses are used “for the subjective interpretation of the content of text data through the systematic classification process of coding and identifying themes or patterns” (Hsieh & Shannon, 2005, p. 1278). For quantitative approaches it is suggested to increase the reliability of the coding process through the use of detailed coding schedules, training of coders, and the assessment and discussion of coding differences (Krippendorff, 1980, 2004; Riffe et al., 2014; see also Drost, 2011). However, Graneheim and Lundman (2004) suggest that qualitative content analyses need to be open to individual interpretive flexibility, and coders could legitimately come up with different interpretations; they suggest that qualitative content analyses and associated processes of interpretation need to be ‘trustworthy’, which they understand as “establishing arguments for the most probable interpretations” (p. 110) and allowing the reader to identify alternative interpretation if they wish so. Graneheim and Lundman (2004) summarise trustworthiness as being credible, dependable, and transferable. Credibility focuses on the confidence in data collection and analysis for the given purposes. The authors highlight in particular (1) an adequate choice for meaning units or units of analysis, which should neither be too broad nor too narrow as otherwise the codes lose some of their

63 3. Methodology meaning; (2) a large diversity of experiences covered in texts to allow investigation of the research questions from different angles; and (3) the extent to which the derived coding categories adequately represent the data, that is “no relevant data have been inadvertently or systematically excluded or irrelevant data included” (Graneheim & Lundman, 2004, p. 110), and that the identified similarities within coding categories and differences between them are highlighted and adequately documented, for example with representative quotes. The authors further suggest regular exchange during the analysis process, for example within the research team, or with experts or participants. Qualitative content analyses also need to be dependable, with which Graneheim and Lundman (2004) mean that inconsistencies during data collection and analysis should be avoided; the authors however appreciate that for example participant interviews may change over time as insights develop and narrower follow-up questions may be used. Lastly, research should be transferable to other contexts or settings. Graneheim and Lundman (2004) suggest that “authors can give suggestions about transferability, but it is the reader’s decision whether or not the findings are transferable to another context” (p. 110). They suggest that transferability may be facilitated by detailed descriptions of contexts, sampling and analysis procedures, as well as “a rich and vigorous presentation of the findings together with appropriate quotations” (p. 110). In this thesis qualitative content analyses were applied throughout, as they were directed by the research questions and thus the literature. However, as no clear hypotheses were made, and the research topic was primarily meant to be established and explored in this project, no coding categories or codes had been suggested. Thus, within the scope of the operationalised research questions, all content analyses were conducted from emergent themes in the data (cf. Hsieh & Shannon, 2005). Fur- thermore, due to the variety of data used in this thesis–few transcribed verbal texts on the one hand, and many advertisements and short messages on the other hand— different approaches to content analysis (summarised in Table 3.2) are used in this thesis to ensure meaningful results and optimal trustworthiness. For the advertisements and the tweets, coding manuals were devised after exploratory coding4 and applied throughout the analysis, which led to a rather guided and coherent analysis of the data (cf. Krippendorff, 1980); each advertisement and each tweet were treated as a single unit of analysis. Each indicator derived from the research question was subjec- ted to exploratory coding, after which codes were revised, possibly merged or distinguished, and a coding manual devised. Along the lines of the coding manual, all data was then coded, and observations assorted to one code were revised and rearranged to

4For exploratory coding, 15% (n=34) of tweets, 16% (n=50) of the full advertising sample, and 34% (n=19) of the recent advertising sample were chosen at random.

64 3.2. Methods make coherent and meaningful categories. From these, groups of codes were created (Graneheim & Lundman, 2004 term these ‘themes’ or ‘sub-themes’) on the basis of similar themes in previous considerations (e.g. sensory and habitual properties), the conceptual framework (e.g. meanings and occasions), and similar findings in the other data sets. For the purpose of analysing category emergence (Qa), the analysis of the codes across time periods was supported by the use of χ2-tests. This was done to identify codes that could be subject to further interpretation given their meaningful or substantive change (or stability; see section 3.3.2 for a detailed overview of the analysis process). However, the interpretation was also informed by the historical narrative, the actual trends over time, and the latent content of the advertisements which had not been captured in codes. For the other two research questions, the emerging codes have been used qualitatively as providing structure to the data and guiding the analysis. Specifically, for the last research question, frames were identified as ‘dominant meaning’ (see also the following section) and captured in one code for the tweets and two codes for the advertisements. The videos and the interviews due to their longer and more complex use of language were investigated using an open inductive content analysis (cf. Elo & Kyngäs, 2008), whereby one or more sentences, but not whole text passages were understood as a unit of analysis. Because I have transcribed and verified all data myself, there was no need to be further familiarised with it. Along the lines of the different coding categories, text passages were assorted to the different research questions and codes (see Hsieh & Shannon, 2005), as emerging themes were noted in a coding log. In an inductive and iterative approach, text snippets relating to one coding category were then re-read, rearranged, and using the coding log and the emergent themes grouped and put in context with each other. Often, themes emerged to which other, not previously coded text passages could be related and were thus added. Contrary to the more guided strategy which used coding manuals for the other data, this approach allowed meaning to emerge from deeper yet targeted engagement with the texts. This allowed investigating text passages in their contradiction to and resemblance of the same or other texts. To identify the frames for research question Qc, frame components were identified per text (see the following section). A summary of the approaches taken for these different data sets is provided in Table 3.2. The main differences between these two approaches are in the use of different units of analyses, as well as the lesser degree of flexibility allowed by the use of coding schedules.

65 3. Methodology

Guided Approach Open Approach

Table 3.2: The procedure for the content analyses for the different data sets (inspired by Elo & Kyngäs, 2008; and Hsieh & Shannon, 2005).

Guided Approach Open Approach

Data adverts, tweets interviews, videos Preparation Unit of Analysis whole text Qb: text snippets, Qc: whole text Coding Categories from research questions from research questions Exploring Data exploratory coding through transcription and verification Organising - developing structured coding - open coding according to manual (coding schedules in categories Appendices B.6 and E.5) - coding according to the manual - inductive, iterative rearranging in random order of all codes in a category - grouping - grouping - categorisation and abstraction - categorisation and abstraction Frame Analysis manual holistic approach hermeneutic approach (Matthes & Kohring, 2008) Reporting description of themes, code description of themes frequencies supportive Software Used custom-made software (see RQDA (R-based qualitative Appendix), spreadsheet data analysis tool, Huang, 2016) (LibreOffice)

Concerning research credibility (see again Graneheim & Lundman, 2004), units of analysis were adjusted based on the specific data sets. The variety of data chosen should lead to both in-depth and broad findings. Furthermore, conversations with research colleagues and supervisors were sought as often as possible to discuss emergent coding and ensure a reasonable research process. All sampling and analysis procedures for specific data sets are described in detail in section 3.3 to ensure further credibility for the processes applied. For dependable research outcomes, sampling of tweets and videos followed a predefined methodology, as advertisements were collected with very broad inclusion criteria which were then narrowed down before the analysis when all data had been collected. Interview partners were approached based on conveni-

66 3.2. Methods ence and purposive sampling (see section 3.3.3). The interview guide remained largely stable throughout all interviews, although specific emergent themes from the analysis were asked for or focused on if participants had not mentioned them yet. Analyses were conducted systematically, as described throughout this chapter. Transferabil- ity is facilitated by in depth descriptions of the respective findings for each data set, and many overview tables with codes and representative quotes. To support or reject certain theories, Hsieh and Shannon (2005) suggest that thorough descriptions may be extended with reporting on frequency counts and rank order, which are reported as supplementary data in tables alongside detailed descriptions of the findings. Fol- lowing what Thomas (2006) termed a ‘general inductive approach’, I omitted less frequent codes in detailed descriptions unless they appeared relevant.

3.2.2 Frame Analysis Frame analysis is widely used in the social sciences to study how meanings are constructed and conveyed (Cornelissen & Werner, 2014; Goffman, 1974; Snow & Ben- ford, 1992; Vliegenthart & van Zoonen, 2011). For studying categorisation, frames have been analysed as a device to convey meaning in context of identity claims and social values “thereby not only lending legitimacy to their offering but also making it credible and resonant to others” (Schiller-Merkens, 2017, p. 219; Weber et al., 2008). Grounding their frame analysis on a modified understanding of frames compared to the one used here (cf. Entman, 1993), Matthes and Kohring (2008) summarise five commonly used approaches to study frames: Hermeneutic, linguistic, manual holistic, computer-aided, and deductive. Of interest in the context of this work are the hermeneutic and the manual holistic frame, which both build up on an interpretive induction of frames based on close reading and reasoning. In a hermeneutic approach, frames are described in detail and without quantification, whereas in the manual holistic approach, frames are identified on a subset of the data, converted into a coding manual and the analysis is scaled up using a content analysis. Similar to the qualitative content analysis, the accountability and reliability of the research outcomes is important (cf. David, Atun, Fille, & Monterola, 2011; Matthes & Kohring, 2008), which is why a similar methodology as for the content analyses was applied. Notably, an identification of frames does not focus on the descriptive but instead the latent content (cf. Graneheim & Lundman, 2004) and is thus strictly speaking not a content analysis. For the interviews and the videos the frames were identified using the differentiation by Snow and Benford (1988; Benford & Snow, 2000) into prognostic (solution- oriented), diagnostic (problem-oriented), and motivational frames (see again section 2.4). These frames were identified from a close reading of the whole texts, thereby

67 3. Methodology identifying the overarching diagnostic, prognostic, and motivational frames ascribed to meat alternatives consumption. On the contrary, for the advertisements and the twitter data, frames were identified in a condensed way as a single, most salient ‘dominant meaning’ (cf. Entman, 1993) through exploratory coding, and then identified across all data points. This dominant meaning was understood here as theme which provided a view of the world that represented a particular problem, solution, or motivation (see above) or a combination thereof; some dominant meanings also highlighted an absence of problems, solutions, or motivations, and instead suggested a specific ‘normality’. In case of the tweets, the operationalised frames for meat and meat alternatives were assessed in one code. For the advertisements, the themes identified for research question Qa3 were recapitulated as indicative of a dominant meaning of the tweet, and frames for meat alternatives concurrently assessed in relation to how they frame meat. After all of these dominant meanings were coded and aggregated, they were analysed in detail and expanded to describe the different underlying prognostic, diagnostic, and motivational frames for a given group of codes. After frames were identified for all data sets, they were summarised into brief captions of how meat alternatives are understood and how meat is understood respectively. Based on these summaries, as well as diagnostic, prognostic, and motivational frames, commonalities were identified concerning the findings of the previous content analysis. The elicited frames were then interpreted on the basis and through examples of the original data.

3.3 Data

3.3.1 Contextual Data Data describing the market context of meat alternatives was collected to contextualise and support the analysis of category emergence, and to provide background to the study overall, such as by reporting on supply chains, actors, finance, legislation, competition, marketing, and the consumership. Collecting such data to narrate the context of the particular case is frequently done in category studies, not only to introduce, expand on or help interpret findings (e.g. Granqvist et al., 2012; Durand et al., 2007), but also to narrate historic developments (e.g. Vergne & Swain, 2017; Quinn & Munir, 2017).

Sampling

Drawing on previous literature and references, library and online searches, recommendations by colleagues, as well as accessibility in university library subscriptions and other libraries, the main data collected originated from five sources: (1) Relevant

68 3.3. Data academic literature was reviewed to understand current academic discourse, and if applicable its congruence with market research. (2) A systematic sample of industry news coverage from The Grocer (1985–2017) was collected in order to identify developments in the market such as company news and their developments, as well as market reviews or new product introductions; The Grocer was chosen as one of the most important UK-focused food industry newspapers, and its availability from 1985 through a computer-searchable newspaper archive. (3) Market research reports— largely from Mintel (2000–2017) and Leatherhead (1990–1995)—were used as comprehensive overviews of changes both in reporting and in the reported; furthermore, a custom market research by Kantar Worldpanel was kindly provided by one of my interviewees. (4) Market data from Euromonitor International (2003–2018) allowed extending and/or verify the data from the collected market research reports. And (5) UK-related entries from an annotated bibliography on meat alternatives by Shurtleff and Aoyagi (2014), which was identified in a scoping search and helped to construct a narrative covering the late 19th century until the 1990s. If business data or histories were missing, data was further augmented with news coverage from other media outlets, or detailed company data, such as from the Com- panies House business registry or the FAME business database. The data collection followed an approach of ‘data saturation’, whereby I sampled data until no new information could be obtained from it (Saunders et al., 2018). An overview of the data is listed in Table 3.3, while the sampling procedure and further justifications for specific data is described in more detail in Appendix A.1. In order to get an overview of the variety of brands sold in the UK, I further reviewed offerings of online stores of major supermarkets and wholesalers in July 2017.

Analysis

The construction of the narrative proceeded as follows: All UK-related entries from the annotated bibliography were extracted, read, and assorted into story lines and networks based on described market contexts or companies; as the data quality of the bibliography decreased towards the end of the 20th century, these story lines were then amended and connected with the data gathered from the industry newspaper The Grocer (from 1985), again on the basis of market contexts and companies (cf. Boje, 2001). Resulting networks were converted in multiple iterations into narratives as emerging discrepancies were sought to be resolved. The narratives were among others built to cover company histories (some of which printed in Appendix A.2), technological advances, coverage of consumers, company takeovers, and market developments; they were later combined into one overarching narrative of market developments. Amended with figures and further insights from market research reports, a narration of the developments in the UK meat alternatives market was devised. Follow-up

69 3. Methodology searches in news databases and academic, historic, and historical literature were undertaken to fill gaps in the narrations. On the basis of the narrative, time periods were sectioned by category assertions and reporting of pivotal developments in The Grocer: In particular, these were reports on changes within the industry as well as increasing investments or revenues, the formation of industry bodies, or altered category labels. For example, the used category labels changed over time from frozen, healthy, and vegetarian food towards meat alternatives. As the magazine can be considered an important intermediary to the food industry, changes in reporting can be seen as general indication for changes in the category, both observed by and possibly also fuelled through The Grocer. These identified points in time were then augmented with further insights, such as changes in market research and advertising activity. While the constructed time periods are not definite divisions of developments, they provide a solid account of the category’s recent history. Furthermore, interpretations of the advertisement codes which depend on the duration of the time periods focused on observed progressive or U-shaped trends over the time periods; therefore, flexibility in time periods might alter significance levels but the directionality of the overall trajectory. A detailed discussion of other caveats of this approach can be found below in section 3.3.2. The overview of the current market was based on the topics described above as well as on recurring sections in market research reports. Insights from these reports were combined with relevant academic literature, business and company databases, and information from press releases and other news sources; the review was continued until no further novel or surprising information could be obtained, and gaps had been filled. Both, the creation of historical narratives and the exploration of the current market were undertaken concurrently, sharing gained insights and emergent issues. For an overview of meat alternatives brands available in the UK, in July 2017 the online shops of most major UK retailers and two health food wholesalers5 were reviewed. For this, firstly via a search mask, products of market leader Quorn were identified in the category structure of the web shop, and the category tree was searched from there, including sibling and parent categories; secondly, the category tree was searched top-down for other likely occurrences of meat alternatives. The search excluded tofu, tempeh, and seitan in non-sausage, and non-burger formats, as well as un- shaped dehydrated textured protein material, such as TVP. The search was amended by screening the offerings of a convenience sample of UK-based or UK-targeting ve-

5This sample includes: Sainsbury’s, Tesco, Waitrose, ASDA, Morrisons, Iceland, as well as online supermarket Ocado and wholesalers Suma and Goodness Foods. Aldi and Lidl could not be included as they lacked an online shop at the time.

70 3.3. Data gan online shops until two successive websites did not offer any brands not previously identified. The approach taken to construct the narrative history relies on accessibility and availability of data, as no funds were available to purchase additional market research data. While precise statistics and details of the results differed across sources, content and trends were sufficiently similar across the various reports, and were validated by occasional summaries in The Grocer of other market research. The constructed narrative relies on a very small set of secondary sources at certain stages in time; yet, to keep the project manageable I have decided not to focus on extensive further inquiry for primary sources. Further, the analysis may be subject to an ‘end of history’ bias, which describes the human tendency to see the present as some sort of climax after which no more meaningful changes are to be expected (Quoidbach, Gilbert, & Wilson, 2013, see also section 3.4). In order to limit the impact of this, I continued following food industry news and evaluated my findings and understandings in context of the developments.

3.3.2 Print Advertisements for Meat Alternatives in Vegetarian Magazines In order to study the emergence of meat alternatives, a data set was required which would cover at least 20–30 years, be consistent, and quantitative or quantifiable; this data had to be widely accessible to target audiences, and thus participating in categorisation within the connection between producers and consumers. It further needed to allow tracking how meat alternatives are described, contextualised, and legitimised. As advertisements have been found to be interlinked with culture (cf. Fowles, 1996) and highly influential for the perception of food (cf. Gunter, 2016), they were considered suitable for this purpose. Much other data such as media reports or patents would not function at the interplay of producers and consumers, but instead as intermediaries or business-focused texts respectively, therefore participating in somewhat different categorisation discourses (see e.g. Grodal & Kahl, 2017). Because this data had been collected for the first research question and was suited for the other analyses, a subset of advertisements published recently (2016–2017) was further used for the remainder of this research project. Not all data was used, because the other research questions are explicitly about the present, as the category was found to be in substantial change over time. A few publications on meat alternatives have used advertisements in their arguments before (e.g. Stahl, 2017; Rödl, 2018a); and some publications related to categorisation have used advertisements to strengthen narratives (e.g. Rao et al., 2000; Weber et al., 2008). Grodal and Kahl (2017) acknowledge that advertisements may be one of

71 3. Methodology the main ways for producers to contribute to categorisation discourse (next to press releases and regulatory documents), and thus as a means to perform strategic categorisation (see Pontikes & Kim, 2017). One model to understand advertising (see Gunter, 2016) relates effectiveness to repeated exposure to brands or products, whereby increased exposure leads to a normalisation and thus attractiveness of the food product, both for the advertiser and their products but also for competitors (e.g. Ehrenberg, 1988; Hoek & Gendall, 2006). Gunter (2016) links this to socio-cognitive theories which suggest that repeated exposure can build up strong attitudes and beliefs about products, even without an individual’s awareness. This model has similarities to socio-cultural categorisation where adverts can function as a way to suggest a favoured interpretation of a producer for sense-making of novel food products. Through the suggestive function of advertisements and their continuous exposure to audiences, advertisements can therefore contribute to and participate in collective meaning-making and categorisation (see again Grodal & Kahl, 2017).

Sampling

Since the mid 2000s, the advertising market for meat alternatives was dominated by market leader Quorn which often spent more than 95% of the total advertising expenditure in the market on their extensive TV advertising campaigns (Mintel, 2013, 2017). However, as print advertising is rather inexpensive and highly targeted, Mintel (2004) noted that many companies advertised meat alternatives in specialist magazines, for example focusing on health or vegetarianism, while a few medium or larger brands also published advertisements in broader magazines, such as the “women’s or mainstream food press.” While a lack of TV advertisements in the sample may be considered a shortcoming, there would have been no more than two different advertisers since 2000, which is why they were subsequently ignored (see also section 4.5.6 for further details on marketing of meat alternatives). Initial sampling of magazines revealed that the long-running magazine The Vegetarian—the member’s magazine of the Vegetarian Society of England–included advertisements for meat alternatives even from the first sampled year in 1985 (which was decided upon through the narrative in section 4.3). As no other long-running vegetarian magazines with advertisements could be identified, food magazines available in the UK were reviewed systematically. Another three magazines were subsequently identified which printed a high frequency and a large diversity of meat alternatives advertisements (see Appendix B.2 for an overview of all considered magazines). Due to limited availability in the British Library and other archives, one of these had to be dropped. The magazines were sampled in February and November 2017 at the British Library and the library

72 3.3. Data of the Vegetarian Society (see Appendix B.1 for more details on the sampling procedure). The following issues became part of the sample:

• The Vegetarian: from January/February 1985 to Winter 2017; 152 issues (not continuously unnumbered), 76 with at least one sampled advert • Cook Vegetarian! : from November 2008 to June 2015; 78 issues (1–69, 71–79), 77 with at least one sampled advert • Veggie (successor of Cook Vegetarian! ): July 2015 to October 2017; 28 issues (80–108), all with at least one sampled advert • VeganLife: January 2015 to November 2017: 27 issues (5–11, 13–32), 21 with at least one sampled advert

The readership of those magazines largely consists of female vegetarians and vegans, and the magazines focus on a broad range of issues surrounding vegetarianism, such as cooking, lifestyle, or vegetarian-related debates (see also Appendix B.1). At first, high-resolution pictures of all advertisements and promotional content featuring remotely-related food products were taken. After all advertisements were collected, this was narrowed down to adverts of at least a quarter page which represented products for which there is a flesh-based counterpart, including sausages or steaks, or derivatives thereof if they were intended for human consumption6. Brand advertising was included when it was clear from the advertising content that the brand sells meat alternatives, or when the reader of the respective magazine was likely to know that that specific brand sold meat alternatives, for example through editorial content or previous advertisements7. The final sample consisted of 557 printed advertisements, or 319 unique advertisements when ignoring reprints. The subset for the analysis of research questions Qb and Qc covered 91 adverts (56 unique) printed in the years 2016 and 2017; the time period emerged from the segmentation of the narrative (see again section 3.3.1). Toen- sure that the adverts sampled from recently published magazines were representative of a larger population of meat alternatives advertisements, all adverts found during the scoping in other magazines not sampled were compared. But because no advert with substantially different content could be identified (i.e. altered messages or content beyond cosmetic differences), the sample may be considered representative of the diversity of print adverts in UK food magazines in 2016–2017 (see Appendix B.1 for details). As the advertisements collected are copyrighted material, they could not be reprinted; a bibliographic index of all collected advertisements and their respective

6Excluded were ‘plain tofu’, ‘vegan pizza’, or ‘party pieces’, as well as meat-free pet food, while ‘tofu mince’ or ‘vegan pizza with chorizo-style slices’, or ‘meat-free party sausages’ were included. 7This covered advertisements of the brands Fry’s, Linda McCartney’s, VBites, Quorn, and Vegusto.

73 3. Methodology headlines can be found in Appendix B.5, and I welcome interested researchers and individuals to ask for a copy of this and the other data sets.

Analysis

The sampled advertisements were analysed using content analysis, which is commonly used in the study of advertisements (e.g. Pollay, 1985; Riffe et al., 2014). For the analysis of research question Qa on category emergence the complete data set was used, whereas the other two research questions would only refer to the subset of advertisements printed between January 2016 and November 2017. Based on the operationalised research questions, and on exploratory coding of 16% (n=50, Qa) and 34% (n=19, Qb and Qc) of adverts respectively, a detailed coding schedule was developed and can be found in Appendix B.6. All guided coding was solely undertaken by the author of this thesis in random order, and each advertisement was coded in relation to one research question before the next advertisements was coded; this was usually done in blocks of two to five hours a day, resulting in 25 to 200 coded adverts. To further avoid systematic changes in coding behaviour, all coding was conducted in a short period of time, and observations were reviewed after the coding process to ensure the coherence and consistency of codes. Coding was undertaken in a custom-made software that allowed for multiple codes per category and coding of specific details of the adverts (see Appendix B.3 for a rationale). As the initial exploratory coding might have not captured the full diversity of the data, in this structured phase more specific codes could be added, which were reviewed later and many were removed again and merged into pre-existing ones. After coding, codes to investigate category emergence (Qa) were aggregated into three to six semantic groups; these were formed on the basis of similar themes or topics based on previous considerations and the conceptual framework, such as by separating meanings and occasions, or sensory, technical, and emotional properties; this was further influenced by similar findings and aggregation in the other data sets. For aggregation, I started with the most frequent codes and added codes in decreasing order of coding frequency, eventually forming another category for low occurrences (i.e. N << 31 as based on the preliminary considerations for using the χ2-test, see below). This led for example to groups concerning contents (e.g. ‘plant-based’, list of ingredients), sensory characteristics (e.g. appearance, taste), or specific occasions (e.g. barbecues, festivities). To all these groups, as well as to each code individually, a χ2-test was applied on the basis of a given aggregated group of codes across all time periods. As the first two time periods covered few advertisements, they were combined, resulting in a total of four time periods. Assumptions of the χ2-test (cf. McHugh, 2013) could be met when

74 3.3. Data a code occurred at least 31 times across all time periods and when it was assumed that adverts in different time periods were independent of each other. This latter assumption is technically not valid in case of reprints or advertising campaigns; but the frequency of these is low and conceptually they are to be counted in both time periods as they represent an influence on the discourse at the time of the respective publication. P-values resulting from the χ2-test were used to evaluate which changes in codes warrant further interpretation to understand the emergence process. In particular, I focused on the interpretation of those patterns which were revealed by the χ2-test as significant and showed a rather linear trend or (inverted-)U-curve throughout the time periods; less importance was attached to patterns showing two or more turns of direction (i.e. zig-zag), as these would be more dependent on the exact cuts of the time periods. These trends were then interpreted concerning the propositions outlined in the specific research design (section 3.1.3) above; this covered for example an indication for a changes in codes used to describe products, such as the disappearance of ‘friction’ or of proclaimed novelty, an increase in protein-related advertisements, or decreasing fluctuation towards later time periods. Further details on the rationale to choose the specific statistical test, as well as further explanation on the assumptions mentioned can be found in Appendix B.4. Since the coding was solely undertaken by the author of this study, who was at the same time involved in creating the coding schedule and creating the software, no training for programs or coding was required. As the coding process was interpretive, trustworthiness was sought to be ensured along the suggestions of Graneheim and Lundman (2004) described in detail in section 3.2.1. In particular, the research process and analytical decisions were documented in detail, and results and emergent themes were frequently discussed with supervisors or colleagues to ensure meaningfulness of the emerging findings. Potential issues to the approach taken here are the use of calendar years to cut time periods, the exact shape of the time periods, and the handling of duplicate advertisements. To address the first concern, advertising does not follow a fixed calendric structure; instead, advertisement activity is said to relate to financial years and may be booked a month or a year ahead (personal communication with the Veggie magazine, 3 May 2018), or may even be spontaneously sought after by the magazine (as interviewee Daniel reported). Thus calendar years seem as reasonable as any other choice. The use of time periods further limits the applicability of this criticism to three instances with plausibly 3–4 months overlap. Nonetheless, when a high amount of adverts with a specific code is at the edge of a time period, their exact cut used in the χ2-test may lead to differing statistical outcomes. Criticism on whether to include duplicate adverts in the statistical evaluation per print or per advert is also plausible, as this makes up a significant part of the data; reasonably, this would mean that advert-

75 3. Methodology ising campaigns which use approximately the same content on many adverts would need to be excluded as well–which leads to questions what is similar enough and thus further issues with the representativeness data. As participation of the adverts in the discourse is of interest in this research question, every printed advert was included in the statistical evaluations. In turn, the diversity across adverts might not be adequately accounted for in the analysis for Qa, which is acknowledged and sought to be corrected for in the present account of the advertisements (Qb and Qc). For the analysis of category identity (Qb) and cultural reproduction (Qc), a guided content analysis was performed as outlined above. The findings build on the codes, and are discussed qualitatively to answer the research questions using representative examples. Alongside the analysis and findings, coding frequencies have been tabulated to illustrate the prominence of groups of codes. The analysis procedures are explained in more detail in section 3.2. The coding schedule can be found in Appendix B.6. Note that for research question Qc, themes used in research question Qa were used indicatively for the frames for the texts (Qc1); instead, another code concerning the relation between the product and meat was coded.

3.3.3 Semi-structured Expert Interviews with Manufacturers and Stakeholders In a discursive understanding of categories, a large variety of texts by consumers and producers alike contribute to collective sense-making. In addition to advertisements, which are analysed as means to position a product in the marketplace, interviews were conducted to obtain a more general and holistic account of producer narratives, goals, and ideas. Unlike websites or archival interviews, they can be tailored to the research endeavour. Thus, semi-structured expert interviews–a common method to generate data in category studies (e.g. Granqvist & Ritvala, 2016; Quinn & Munir, 2017) and the social sciences in general (e.g. Vogt, Gardner, & Haeffele, 2012; Kvale, 2007)—were conducted with representatives of meat alternatives manufacturers, related consultancies, and third sector organisations. The interviews served to identify how those involved in manufacturing meat alternatives label and attach meaning to their businesses and their products by investigating their valuation and categorisation, as well as their understanding and relation to meat consumption. The interviews were further used to aid contextualisation and interpretation of the market through their narrations of market and business history, and their projections for the future of the market; one interviewee even provided a custom market research report for reference. Interviews have been used before to investigate meat alternatives, for example as consumer interviews (Nath & Prideaux, 2011), or as part of auto-ethnographic approaches (Sexton, 2016).

76 3.3. Data

Sampling

For the interviews, UK-based or UK-selling manufacturers of meat alternatives and a number of food-related NGOs and consultancies were identified on the basis of the investigations of listed brands in UK supermarkets (see section 4.1), contacts made at exhibitions and talks, and contacts they suggested speaking to. Contacts at the companies were identified via LinkedIn, company websites, or business databases; if no other possibility was found, companies were contacted using contact forms. After three unanswered emails, contact via phone was attempted and if unsuccessful dropped from the list of potential interviewees. While in some cases I spoke to the person I had approached, some organisations willing to participate suggested a different interview partner related to the focus of my questions. As no compensation could be offered to the participants, the interviews represent a convenience sample of respondents willing to participate. The interview process was approved by the Ethical Review Panel of the Manchester Business School under reference number 2017-0684-4085. Interviewees were provided with a participant information sheet in advance and were required to give their consent via an approved form; after the interview, respondents received the interview transcript and were given sufficient chance to retract their participation or point out corrections they would like to make. The interviews were conducted between November 2017 and May 2018 via phone, via online video calls, or in two cases in person in a café. All interviews were recorded, transcribed verbatim, and sent to the interviewees. In total, I approached 29 manufacturers and ten other organisations for interviews, leading to nine interviews with eight different manufacturers, and two interviews with civil society organisations. A number of other contacts declined, never replied, offered to reply in writing but never did so, or did not manage to arrange an interview. Three further interviews were conducted but permission for use in this research project never granted or retracted. Given that previously twelve interviews have been shown to result in data saturation (Guest, Bunce, & Johnson, 2006), and the fact that other data is included in the research project the amount of interviews conducted can be considered sufficient. The data is limited strongly by its arguably limited coverage of different brands. Only one interviewee could be won who works for one of the companies represented in the 2016–2017 advertising subset, as all others declined; therefore, this data can not be seen as an in-depth extension of the findings from the advertisements but as a separate sample of manufacturers. Furthermore, Quorn is the only major brand on the UK market which is represented it the sample, with another two strong international brands Vivera and Tofurky, whose standing on the UK market has expanded since the interviews. The sample further represents five brands which are available nation-

77 3. Methodology wide, two still young companies, and one whose production has been discontinued. Since they sold no products themselves, the two interviews with third sector organisations are actually interviews with intermediaries and used supplementary to the other data. However, because this research aims to identify a diversity of categorisations instead of a representative view on the industry, these shortcomings of the sample of interviewees should not impact the results. All interviews were conducted in a semi-structured format. The interview guide was devised on the basis of the research question and emergent themes from the contextual data, and covered four thematic sections on the history of the company, its visions, its products (in general and in relation to meat), as well as the market of the company. Confirmation was sought to voice record the interview and if not yet clari- fied a time limit for the interview was established; the structure of the interview was briefly outlined, with opportunity for the interviewees to ask questions in the end. As an ice-breaker question, I asked what their job was, how they got into their job, or alternatively why they founded their company. In the first block of questions, I investigated more on the history of the company focusing on the identity they attempted to construct with their company and how this resonated with consumers and the wider world. Next, the company vision was discussed, aiming to find out how companies see their role and identity now and in the future, as well as how companies tried to achieve this vision and whether consumers and stakeholders could relate. Thereafter, I asked about the product ranges of the interviewee’s company. Hereby I was particularly interested in how products are presented and described, how they are justified, and how strong interviewees would relate their products to meat by themselves and when explicitly asked about it. Lastly, I inquired about the market of the company, such as their target groups, their interactions with other businesses, their relationship with retailers, or their thoughts on technological change. This section aimed at generally identifying trends in the marketplace, and to augment the construction of narratives (see section 3.3.1). I closed the interview by asking about the perceived strengths and weaknesses of their company over time, asked for clarification such as for dates or names I had not clearly understood throughout the interview; I further inquired about any recommendations for colleagues or business partners for me to speak to. The full interview guide can be found in Appendix C.3 and was discussed with and trialled on two colleagues, and further revised after the first four interviews–two of these could not be used in the analysis, see above. Interviews with non-manufacturer stakeholders were adapted and covered the general market rather than a specific company. Interviews were transcribed verbatim by the researcher, whereby difficult to understand passages were double-checked by a native English speaker, and names and dates were verified using online sources if possible; if no clear answer could be found,

78 3.3. Data

I again asked for clarification by the interviewees. All transcripts were revised again before sending them for approval to the interviewees.

Analysis

The analysis was conducted as an iterative and open inductive content analysis (see section 3.2.1) and aimed at identifying and exploring the diversity represented by the participants regarding the different research questions. In a first attempt half-way through data collection, seven interviews were firstly coded and analysed (two of which could not be used in the end). Guest et al. (2006) have found this number of interviews to cover elements of all meta themes, which is why this was considered a suitable point for preliminary assessment. Based on this preliminary analysis, emergent themes were explored, and similarities and differences among their representation identified. This informed the other ongoing or yet-to-come interviews and analyses, and similar to grounded approaches led to refinement of the research questions and the theoretical constructs (see e.g. Conrad, 1978). In the coding process, which led to the analyses outlined here, passages relating to either of the operationalised indicators were identified (using the software RQDA, Huang, 2016); thereafter, all text segments in a specific coding category were iteratively revised and rearranged (see again section 3.2.1). This approach allowed on the one hand to be directed and follow the proposed research questions, while on the other hand themes and findings could emerge from the juxtaposition or opposition of different statements and opinions within and across interviews (Hsieh & Shannon, 2005; Kondracki et al., 2002). Throughout the analytical process I kept a diary of emergent themes, issues, and insights, while equally engaging with colleagues about the emerging findings. Throughout the analysis process, a certain level of reflexivity had to be kept, as any statement implicitly could have been a sales pitch towards me as a means to have their companies ‘adequately represented’ in my research. This however, may be seen as an advantage to the interview process as the interviewees would thus argue in similar ways to how they would participate in general discourse on meat alternatives. Yet, possible attempts to manipulate or at least convince me of their perspective need to be acknowledged; they were reflected as part of the analysis process, which became easier as I dealt with all transcripts together and conducted the analysis a few months after the interviews were conducted.

3.3.4 Taste-Test Videos of Meat Alternatives from YouTube Requiring audience perspectives of consumers of meat alternatives, I decided to utilise user-generated content from social media; among others, such platforms had been studied for their food-related subculture (Rousseau, 2012), or for ‘digital food act-

79 3. Methodology ivism’ (Schneider, Eli, Dolan, & Ulijaszek, 2017). Specifically, I decided to analyse taste test videos from the largest online video platform, YouTube. In these videos, users eat and comment on a food they (usually) have not tried before: Here, content creators value and evaluate food, expose their attitudes to eating, food practices, and their kitchens, and contextualise them in their everyday and purpose-based consumption; by describing and evaluating products in their own perceived reality for their specific audience, users contribute to categorisation discourse as they might be considered a so-called ‘influencer’ (see Kolo & Haumer, 2018). Providing a momentary and unscripted–albeit many videos are heavily cut–account of an immediate taste experience, implicit and explicit but also verbal and non-verbal engagement with the foodstuff allows identifying assumptions, relations, and identity of the product debated. In opposition to interviews, focus groups, and other means to generate data (cf. Miyazaki & Taylor, 2008), the videos allow for an unsolicited insight into interactions with and reactions to the food, which is why these videos are highly suited to observe and investigate discursive categorisation by audiences for other audiences. YouTube is an online social network and platform, where users can upload and share video content for a private or a public audience. The biggest video sharing platform on the internet, it is used both by individuals and corporations for a variety of purposes, including marketing, information, or documentation; generally, the platform provider and the content uploader share the advertising revenue, making it a profitable for the content provider when large audiences can be reached. It has been described as having “multiple roles [such] as a high-volume website, a broadcast platform, a media archive, and a social network” bearing a high degree of “cultural ‘generativity’ ” (Burgess, Green, Jenkins, & Hartley, 2009, p. 5; see also Zittrain, 2008). This generativity originates from YouTube’s role as both a commercial platform and a social network; in its network capacity, users can engage with each other via comments, video responses, up- and down-votes, private messages, or even their willingness to watch an advert to support the content provider. The ‘participatory culture’ established on YouTube (Burgess et al., 2009) is prone to activism (e.g. Askanius & Uldam, 2011; Ekman, 2014; Thorson, Ekdale, Borah, Namkoong, & Shah, 2010) and counter-activism (Youmans & York, 2012), but also food-related community building (Kierans, 2017), or identity work (Forchtner & Tominc, 2017). Similar to Pace (2008) who sees YouTube videos as a source for consumer narratives that can be analysed in the social sciences, I understand the introduction, description, and negotiation of meat alternatives in user-generated content as a contribution to negotiate a category identity. Dolbec (2015) acknowledges in their doctoral dissertation the important role of among others video content for categorisation, both as user-generated and as marketing material, “for the creation of shared meanings and the transmission of tacit

80 3.3. Data knowledge” (p. 101). Otherwise, user-generated content or videos have not been used in category studies or research on meat alternatives before. However, there are certain similarities between the taste test videos and experimental taste tests (cf. Hoek et al., 2013; Wild et al., 2014), as well as focus groups and taste test videos with more than one person (cf. Elzerman et al., 2013). Outside of the organisation studies literature, user-generated content has been used to study categorisation in a cultural context as self-identification (Hall, Gough, & Seymour-Smith, 2012) or assertion of social category membership (Halonen, 2015). Despite a different focus, these studies highlight that categorisation is regularly undertaken within and through user-generated video content.

Sampling

As taste test videos had not been studied academically before, I established through watching a number of these, that they were indeed suitable for this research project and identified relevant search terms for this study: To obtain a systematic sample, any of the twelve combinations of vegan/vegetarian, meat/burger/sausage, taste test/tasting, and UK were entered in YouTube’s search mask on 17 May 2018, and from each result list presented in ‘order by relevance’ the first five UK-based consumer taste test videos were noted. This excluded videos from news agencies, videos about meat alternative ready meals or fast food, and recipe, cooking, or ‘food haul’ (i.e. shopping review) videos. Geographic credentials were identified on the basis of the description, my knowledge of UK products, and by accents and the autobiographies offered on linked websites. From the resulting set of 60 videos, duplicates were removed which led to twelve unique taste test videos. As the amount was very similar to those of the interviews and thus probably satisfying to cover the broad themes (see again Guest et al., 2006), no further videos were sampled. Metadata was retrieved from the YouTube API and the videos were stored. I then transcribed the videos in verbatim, included remarks on actions in case they are relevant for the content, and assigned random names to the YouTubers. To be more comparable with the other data, all following analyses are based on the audio content of the video while contextual descriptions are offered. Analyses have been conducted with the qualitative data analysis tool RQDA (Huang, 2016). As the videos essentially represent publicly available information, which was published with the intent to be watched and possibly even redistributed, no ethical approval was obtained (cf. Townsend & Wallace, n.d.). For a more detailed overview of the sampling process, see Appendix D.1; for details of the sampled videos, see Appendix D.2. A potential caveat is that the sample is limited to people who are conducting a

81 3. Methodology taste test, film and cut it, and then publish it online. The YouTubers furthermore have highly diverse reach, as Lucy’s channel has over 450,000 subscribers and the video almost 100,000 views at the time of sampling, whereas Robert’s has less than 25 subscribers and 100 views, which may further be skewed by a variance in publication dates. So while some YouTubers might be described as ‘influencers’, most product reviews appeared motivated by an idea of sharing information or impressions, and maybe promoting meat-free eating to their respective audiences, but not by promoting particular products8. While it may appear unjustified to give all videos the same representation in the analysis, no correction for the magazine circulation of respective adverts was undertaken, and consumer interviews are usually also conducted no matter whether people have ten or a thousand friends. Because the videos are treated as a set of diverse ideas and participation in discourse whose opinions may or may not converge, no video is given precedence over another. Similarly, while the sample does cover non-vegetarians and people who do not like meat anymore, through their active engagement they nonetheless represent people engaged in shaping, negotiating, and valuing a category of meat alternatives. However, it needs to be noted that people who do not eat meat alternatives can of course still influence categorisation, for example by not offering their valuation or not buying any such products.

Analysis

While a variety of both quantitative (e.g. Vergani & Zuev, 2011) and qualitative approaches (e.g. Benson, 2015; Gibson, 2016) including content analysis (e.g. Chang & Chang, 2018) have been used to study the content of videos on the YouTube platform, Laurier (2015) asserted that there is “no single appropriate method or form of analysis for investigating YouTube” (p. 488). Given that both videos and interviews were used as transcribed speech, also videos were analysed using an open inductive content analysis, whereby all text passages relating to a specific research question were coded, and then later iteratively sorted, grouped, and analysed. The themes were then linked amongst each other, and the analysis followed the prominence of certain themes as well as the links made in between them. Research question Qb3 on labelling was not evaluated using the YouTube videos, because they often only concerned one product which was named after the label on the packaging. To elicit the frames for meat and meat alternatives in the videos, I firstly identified and discussed motivations of indi-

8Notably, some reported how they did not like specific (previously) trialled products or had abandoned previous favourites. Furthermore, while view counts and subscriptions are high, no YouTuber reported to have been sponsored to promote particular products. However, Lucy reported the video to be sponsored by a social media app likely paying for their purchase. Charlotte promotes a vegan subscription box which should however not affect their evaluation of the particular product, particular because they sample these with their friends.

82 3.3. Data vidual YouTubers. This covered why people wanted to sample a specific meat alternatives and film it, why they might eat meat alternatives more generally, and how they engage with or speak to their audience. The motivation for this intermediate step was to identify how private or public their video was, and whether there was an actual problem, solution, or motivation (and thus frame) involved in the video. Only afterwards were those insights abstracted into the three different frame components per video, and later aggregated. The general analyses are described in more detail in section 3.2. Although I as a researcher was certainly biased in my analytical capacity through the continuous engagement with the other data sets, the analyses of the taste test videos do not need to account for the researcher’s influence on responses. The videos may still conform to a social acceptability bias, which is likely to be reflective of the YouTubers’ self-representation, of their online communities, of some political stances they want to transmit in the video, and/or of not wanting to affront audiences (e.g. Fox & Warber, 2015; Gustafsson, 2010). Further, as no prior research has been conducted on the topic of taste tests as a phenomenon in the general media or on YouTube, no comparisons can be made or offered on the specificity of the findings.

3.3.5 Tweets on Replacing Meat during National Vegetarian Week and World Meat Free Week Similar to YouTube, also the short messaging platform Twitter has attracted considerable interest by social scientists (Giglietto, Rossi, & Bennato, 2012; Proferes & Zi- mmer, 2014), among others in relation to politics (Jungherr, 2016), health (Laranjo et al., 2015; Sinnenberg et al., 2016), brands (Liu & Lopez, 2016), or the reception of academic and scientific outputs (Newman, 2017; Vainio & Holmberg, 2017). It has equally been identified as a platform for negotiating food and food practices, which Mann (2017) termed ‘hashtag activism’, referring to the prominent use of hashtags as a searchable and community-building or theme-defining topic marker. Investigating the prominence of different topics on Twitter, Koylu (2018) identified that approximately 10% of all tweets mentioning other users relate to food. Other researchers focusing on food on twitter investigated for example the spread of news and opinions in food scandals (Brummette & Fussell Sisco, 2018), or the dissemination and contestation of food-related health claims (Declercq, Tulkens, & Van Leuven, 2018). Some scholarly work on meat alternatives touches the role of social media and twitter to increase its prominence (e.g. Kalamus, 2019), while none has engaged with it in detail. Short messages from this platform, so called ‘tweets’, were collected in larger quantity to broaden the insights on audience categorisation of meat alternatives. Jungherr (2016) proposed that twitter may be an “informal barometer” for public

83 3. Methodology opinion which however “might offer no true picture of reality but instead one mediated by the attention, interests, and motives of politically vocal Twitter users” (p. 80). Twitter can therefore be understood to match well the scope of discursive categorisation (cf. Grodal & Kahl, 2017). While they often only comprise a single message, their quantity might exhibit a rather diverse set of ideas and opinions, which makes it a suitable data set to triangulate the limited amount of consumer perspectives gained from the taste test videos.

Sampling

The sampling aimed at collecting a targeted, concentrated data set of tweets that are likely to cover meat alternatives. Lacking adequate hashtags despite brand names (which would be partial), I decided to track two different vegetarian-related events in May and June 2018: National Vegetarian Week (#NationalVegetarianWeek) organised by the Vegetarian Society of England with support of retailers and manufacturers since 1992 (The Grocer, 1992), and World Meat Free Week (#WorldMeatFreeWeek), which originated as a single day in 2015 and extended to a whole week in 2018. While the website of the latter is run from the United States and a US-based PR agency listed on the contact form, the event is also sponsored by UK-market leader Quorn, which suggests some UK involvement. As both of these were top-down facilitated events rather than emergent properties of a conversation or of events in world history, it was considered sufficient to restrict sampling to these two hashtags (cf. Lorentzen & Nolin, 2017). Using the Python programming language and the ‘tweepy’ package, data was collected from the Twitter API with one day padding (from the perspective of British Summer Time) before and after the announced dates. Because tweets were stored at the time of posting, various popularity metrics (e.g. likes, favourite counts, etc) were not useful, as they build up over time. According to the behaviour of the API, retweets were still recorded as they covered the same hashtag, and replies to and quotes of tweets with a particular hashtag need not include that hashtag themselves. While a variety of ways to filter the data for UK-only messages were attempted, only filtering for English language tweets led to meaningful results; preliminary results of these filtering attempts, as well as the timely distribution of the tweets indicate that still many tweets actually originate in the UK (see also Appendix E.1 for a detailed overview of the sampling process) which is a major limitation to this data set. Using the programming language R, the data set was explored in order to assess some generic attributes of the collected data. Exploration was also done to assert whether there were substantial differences between the two different weeks, or whether it was justifiable to cover them in the same sample later on, which is described in section 8.1.

84 3.3. Data

To obtain a sample of tweets related to meat alternatives, the following qualitative analysis considered all messages which were not retweets and contained terms that resemble meat or types of meat; these words were obtained from the semantic lexical database WordNet (Miller, 1995) and included words such as ‘beef’, ‘hamburger’, ‘roast’, ‘chop’, ‘venison’ or more specific terms like ‘guinea hen’, or ‘liver’; a full list is reproduced in Appendix E.2. From the resulting 1,675 tweets, those tweets which related to meat alternatives–generically, through specific products, food service, or recipes–or replacing/swapping meat were identified manually. Duplicate tweets were removed, which were identified when the same account re-posted the same message, if they were retweeted without being identified as such by the API, or one where only a few contextual words differed. This resulted in 225 tweets, which were anonymised and then reproduced in Appendix E.4. The reduction of data for such a qualitative analysis may be considered necessary to perform such task. Although the results from the different theme weeks appear very similar, the tweets remain a subset of the specific theme week; therefore they either celebrate vegetarianism or eating meat-free, which needs to be considered when generalising on the basis of this data set.

Analysis

A guided qualitative content analysis was performed on the filtered subset (see e.g. Sjøvaag & Stavelin, 2012; Vainio & Holmberg, 2017, for previous uses of content analysis for twitter data). The tweets were coded in a LibreOffice spreadsheet, whereby metadata was collected and coding categories were defined in advance on the basis of the operationalised indicators outlined in section 3.1. Trial coding was performed on 34 tweets (15%), to devise a coding manual and to identify potential difficulties to cover each research question in one code. Research question Qb2 on evaluation and references was ignored for this data set, so that Qb is covered in two coding categories–each allowing multiple codes. The research questions for Qc were together abstracted into only one coding category relating to an overarching frame as covering the ‘dominant meaning’ of the tweet (see again section 3.2.2 on the frame analysis), and only later expanded analytically to cover the different elements of frames. The analysis followed the procedure outlined in section 3.2, and the coding schedule is reproduced in Appendix E.5. All codes were checked for consistency, and some codes were merged or re-coded. In the rare instance that tweets did not appear meaningful by themselves or seemed out of context, their appearance on the twitter platform was checked, where they were embedded into conversations or joined by pictures.

85 3. Methodology

Table 3.3: Overview of the data sources used in this research project, including their timely coverage and amount of collected texts.

Texts Number Years

Expert Interviews with meat alternatives manufactures 9 2017–2018 with NGOs 2 2017–2018 Print Advertising 557 The Vegetarian 156 1985–2017 Cook vegetarian! 263 2008–2015 Veggie 104 2015–2017 Vegan Life 34 2015–2017 sample used for Qb and Qc 91 2016–2017 YouTube Taste Test Videos 12 2015–2018 Tweets #NationalVegetarianWeek ~35k May 2018 #WorldMeatFreeWeek ~8k June 2018 selected sample 225 Contextual Data Shurtleff and Aoyagi (2014) bibliography 965–2014 bibliography entries (UK-related) 163 product introductions (UK-related) 131 The Grocer articles 459 1985–2017 business data Euromonitor sales data 2017 Companies House FAME database market research Leatherhead 3 1990–1995 Mintel 9 2000–2017 Kantar Worldpanel 1 2016 other news sources academic literature

86 3.4. Limitations and Validity of this Research Project

Table 3.4: Overview of the data sets, including where findings are presented, and where additional information can be found, how they are referenced in the thesis9, and where the index is located.

Data Findings Appendix Example Reference Index/Sample

Contextual Data Chapter 4 A e.g. The Grocer (1995b) Archival References Advertisements Chapter 5 B e.g. 2005-10 Appendix B.5 Interviews Chapter 6 C e.g. interviewee Tony Appendix C.2 Taste-Test Videos Chapter 7 D e.g. YouTuber Pippa Appendix D.2 Tweets Chapter 8 E e.g. N15, W27 Appendix E.4

3.3.6 Summary of Data Sources In Table 3.3 the data collected for this research project is summarised; Table 3.4 provides an overview of how data is referenced in this thesis, where data and findings are described, and where an index for the data can be found.

3.4 Limitations and Validity of this Research Project

There are some major limitations and caveats to this research project beyond the issues discussed in the separate data and research design sections. Firstly, the selection of 2017 as an end date for studying emergence needs to be treated with caution. Although the choice of this specific date can be justified, of course no later date could have been chosen if this project was supposed to be brought to an end in the time given. Likewise, any earlier date would have temporally disconnected the study of emergence (Qa) from the remainder of this research project (Qb and Qc). So although I did not know the developments in the category when I started this research project, I am likely to be subject to the so called ‘end of history’ bias; in personality research, this describes the human tendency to assert most importance to developments leading to the present and after which no more meaningful changes are to be expected (see Quoidbach et al., 2013). In context of this research project, this means that the role of the present may be overly emphasised in its relevance for a meat alternatives category. In order to alleviate the impact and ground the research in contemporary developments, I followed food industry news regularly throughout my dissertation, even though much of this insight could not be used in this thesis.

9All tweets are either referenced by N for National Vegetarian Week, or W for World Meat Free Week.

87 3. Methodology

The choice of dates also leads to a second caveat: At the moment, nationally and internationally markets for meat alternatives and thus also the categories seem to be in high flux (see chapter 4). There are hopes connected to the market, and high attention brought to it (see e.g. CB Insights, 2017). Their different histories firstly limit the transferability of any findings to cultural contexts other than the UK. Despite usually slow cultural change, this flux may render results meaningless by the time of publication or soon after. The fact that the twelve sampled YouTube videos and the ‘recent’ advertisements span more than three years in the past from the point of publication thus need to be considered. If however the findings on category emergence align well with findings from the other data sets, this may be not such a major issue as then data may at least be considered historically coherent. A major challenge and possible limitation further originates from the amount and diversity of different data sets. In anticipation of the potential issues, the methodology was constructed in a way which aims at balancing consumer and producer perspectives, and integrating analyses and insights across data sets. But one may challenge for example whether one interview has the same importance as twelve tweets, eight adverts, or one YouTube video; although each represents an eleventh of the respective data set, the quality of the data and the insights possible vary greatly. If differences in findings can be explained by their specific sampling methods and the composition of the data set as such, there should be no issue to the validity of the findings beyond the scope of the individual data sets. On the other hand, if findings cannot be compared or can not be triangulated in meaningful ways, this would invalidate any claims beyond individual data sets. Any findings from this research project are thus limited to the geographic and cultural context of the UK, and–unless otherwise shown–to the state of the category around 2016 to 2018. Within the limits of the individual data sets and as outlined in the scope for this research (section 1.3), findings are further limited to producers and audiences who assert meaning to meat alternatives in one way or the other; findings are therefore not generally transferable for example to audiences who dislike meat or who dislike the idea of substituting animal meat with plant-based facsimiles (although the YouTube sample has one example of both groups). Beyond these major points, there are a number of further limitations, more specifically related to individual ways of sampling, analysing, or interpreting data, which have been discussed in the previous sections. A final reflection of limitations of the findings is offered again in the final discussion in section 10.5.

88 Chapter 4

Contextualising Meat Alternatives in Past and Present

In this chapter, the research topic is contextualised. Using among others academic literature, bibliographies, historic market research, and food industry news, a longitudinal narrative of the historical developments in the meat alternatives market is constructed to provide background and context to the developments concerning meat alternatives in the UK. Further to providing a historical narrative, time periods were identified to separate the analysed advertisements for the assessment of temporal change (see chapter 5). Both, time periods and context support the analysis of category emergence (see section 9.2). Using market research, industry news, business data, and academic literature, the current market environment of meat alternatives is described. These descriptions allow to contextualise any findings in the current market, consumer, and product environment, as well as in recent and ongoing trends. In the remainder of this chapter, first the data used is briefly described. In the following section, previous definitions and a detailed description of meat alternatives is presented. Next, the history of meat alternatives is narrated in six time periods starting in Victorian England and finishing with the most recent market boom since around 2016. After meat consumption in the UK is briefly discussed, the current market for meat alternatives in the UK is introduced in detail focusing on consumers, brands, companies, networks, regulations, marketing, and relations to the meat industry. Be- fore a conclusion is offered, three other national markets for meat alternatives are briefly described.

4.1 Description of the Data

Accessed via the Factiva database, in The Grocer 459 news articles relating meat alternatives in general could be identified between 1985 and 2017; with supplementary company searches, 404 articles for Quorn, 211 for Linda McCartney’s, 93 for Tivall,

89 4. Contextualising Meat Alternatives

86 for Goodlife, and 69 for Redwood and VBites could be identified1. In the annotated bibliography, out of 3650 items in total, there were 163 bibliography entries and 131 product introductions tagged as related to the UK. The data density in the bibliography declined towards the end of the 20th century; according to the author this is because by then many other people or organisations were collecting and providing similar data (personal communication, 4 March 2016). Apart from this, 13 market research reports were collected (three by Leatherhead, 1990–1995; nine by Mintel, 2000–2017; one by Kantar Worldpanel, 2016). When required, the data was supplemented with other news sources, other related market research (e.g. on protein or meat), business data, and company data (see again section 3.3.1 and Appendix A.1 on the sampling process). Reviewing offerings of online shops of retailers and wholesalers revealed that around 60–70 different brands (including own label) of meat alternatives were sold in the UK in July 2017. The uncertainty results from retailers Aldi and Lidl not being part of the sample, as well as from out-of-date websites, possibly changed branding, smaller companies selling only on their own website, and the general uncertainty induced through the depth of the internet. Figure 4.1 displays the product listings per online shop and highlights the brands with the most listings.

4.2 What are Meat Alternatives?

Various definitions of ‘meat alternatives’—sometimes also called meat analog, meat substitute, mock meat, faux meat, or imitation meat2 (e.g. Sadler, 2004)—exist in the academic literature. What all of them have common is that they describe food items which relate to meat on a level observable by consumers. Definitions include, among others, “something that looks, tastes, and feels like meat” (Zorpette, 2013, p. 64), or having “approximately the same taste, appearance, and texture of a related food made from meat, poultry, fish or shellfish” (Shurtleff & Aoyagi, 2014, p. 5). They “should have a certain resemblance to meat in order to replace meat on the plate” (Hoek et al., 2011, p. 371), thus not only being a similar product but also aiming to replace a function. As well as their less visible constitution, some definitions include nutrition as an important factor, stating that “nutritional value is, in general, approximately

1All company searches are limited to entries before 16 August 2017, while the overall search includes all articles from 2017. Single news may be counted multiple times as searches were conducted independently, and there was some duplicate indexing in the database. Details of the search terms used can be found in Appendix A.1. 2In the 1970s and early 2000s, terms such as novel or new protein foods were quite popular in research. More recently, alternative proteins or plant-based proteins have emerged as terminologies for products such as meat alternatives. For a delineation of terminology relevant in this research project, see again section 1.3.

90 4.2. What are Meat Alternatives?

100

50 Listed Products

Suma Ocado Tesco ASDA Waitrose Iceland GoodnessFoodsSainsbury's Morrisons Wholefoods Brands with more Quorn Linda McCartney's Other than 5% of all listings VBites Fry's own−label

Figure 4.1: Listings of various meat alternatives brands by store as identified in online shops in July 2017. equal to (or sometimes greater than) that of the related food, including essential vitamin B-12” (Shurtleff & Aoyagi, 2014, p. 5). Wild et al. (2014, p. 45) understand meat alternatives in a product development context as “products that fulfil consumer de- mands of healthy and tasty products which both replace the function of meat in a dish and contribute a similar high protein nutritional value.” While definitions vary, key components remain: Meat is assimilated on the basis of sensory and cognitive features. Commonly appreciated similarities between meat and meat alternatives include among others taste, texture, nutritional and protein content (e.g. Wild et al., 2014), similar cooking properties and affordances, and the usage in particular social contexts or meals (e.g. Nath & Prideaux, 2011). These definitions may help the reader to understand the ideas which have driven academic inquiry into meat alternatives. They also helped me form a foundational understanding of meat alternatives and therefore have an impact on this research project. However, in order to avoid any preconceptions or bias through sampling in this thesis, meat alternatives are empirically identified as meat-free foodstuffs which are (a) either compared to meat for any sensory or cognitive feature in the text in question; (b) or are labelled like a meat format3 or in a context commonly associated with meat (e.g. steak, burger, sausage, but also shepherd’s pie or toad in the hole).

3The term ‘format’ is used throughout this thesis to refer to a specific group of ‘meatstuffs’ or their facsimiles which share a common denominator, such as sausages, burgers, steaks, wings, nuggets, roasts, bacon, and others.

91 4. Contextualising Meat Alternatives

Most meat alternatives sold in the UK are made from plant protein. Particu- larly common are soy protein and gluten (wheat protein), while proteins from other legumes become increasingly frequent. One brand makes their products on the basis of fungal protein grown in large fermentation vessels, and products primarily based on egg or milk protein exist as well. In the future, other protein crops, such as lupin, peas, or algae are suggested to gain traction (Frost & Sullivan, 2016). Traditional meat alternatives also include tofu (coagulated soy milk), tempeh (fermented whole soy beans), or seitan (cooked wheat gluten); for modern meat alternatives, the preprocessed protein such as gluten, protein isolate, de-fatted soy flour, or fungal biomass are formed into an emulsion using fat, oil, water, and flavourings. This emulsion is ex- truded under pressure and heat in order to create a fibrous texture, after which usually thermal treatment is undertaken to increase durability. While extrusion is the most common method applied, spinning or shearing of the protein is also possible (Davies & Lightowler, 1998; Krintiras, Gadea Diaz, van der Goot, Stankiewicz, & Stefan- idis, 2016). Depending on the exact ingredients and process characteristics, more or less meat-like mouthfeel and texture can be created, as manufacturing technology improves continuously (e.g. Frost & Sullivan, 2017; Krintiras et al., 2016; Zorpette, 2013). These are also regularly seasoned, marinated, or shaped to be sold as meat alternatives. For some products these foodstuffs are packaged frozen, refrigerated, or as dehydrated bulk good such as Textured Vegetable Protein (TVP); for other products, these ingredients would be shredded, blended with other ingredients, glued together, marinated, or embedded into more complex foot items or meals. Examples for this further processing include meat-style balls, coated schnitzels, chicken-style nuggets, marinated seitan steaks, or pork-style pies. Other developments associated to meat alternatives but not further discussed in this research project are insect-based protein products (e.g. Verbeke, 2015) and so-called in-vitro meat, which is muscle tissue grown outside of animals in cell cultures (e.g. Heffernan, 2017; Stephens, 2010).

4.3 A Brief History of Commercial Meat Alternatives in the UK

4.3.1 Early Meat Alternatives The history of meat alternatives is commonly told to have started in ancient China: The first recordings of tofu being used as a meat alternative originate from almost three millennia ago, and around 700 years ago seitan4 was first mentioned in context

4The term seitan was only coined in the early 1960s in Japan; under different names, the use of cooked wheat gluten as a food product in China goes back to at least the sixth century (Shurtleff & Aoyagi,

92 4.3. A Brief History of Commercial Meat Alternatives in the UK of meat reduction; Shurtleff and Aoyagi (2014) report that particularly when meat was scarce or forbidden, their replacement character was highlighted. In the West, considerations for replacing meat with plant-based alternatives in vegetarian diets have been identified to originate in Victorian England and 19th century USA (Gregory, 2002; Shprintzen, 2012). While cookbooks with ‘mock’ products existed (cf. Sharpe, 1908), there was an increasing amount of political, religious, and health reform vegetarians who against criticism from purist vegetarians tried to make vegetarian diets more appealing to a wider audience. This led to the creation of a diverse range of vegetarian foodstuffs including nut-based meat-like products (Gregory, 2007). In the UK, these products were also influenced and partly imported from members of the American health reform movement, particularly from the creation of the nutritionist Dr John Harvey Kellogg (Shurtleff & Aoyagi, 2014). While successful in their niche, in the mainstream perception alternatives to meat were only taken seriously in war times: In WW1, many wholemeal or fortified foodstuffs (including bread) were labelled ‘meat alternatives’, while nut meats became extensively promoted as a good protein source (Owens, 2016); and in WW2, many products were fortified with soy flour or soy protein isolates, making them more proteinaceous to combat malnutrition. However, as soy processing was still in its infancy, foods had with strong off-flavours which led to a negative public perception of soy products (see Shurtleff & Aoyagi, 2014, n. 2795; Collingham, 2011). In the 1950s and 1960s, increasing meat prices, world hunger, and food inequal- ity were seen as global threats, which fuelled research into alternative protein sources (e.g. Altschul, 1967), such as so called single-cell protein (SCP). For this, microorgan- isms such as bacteria or algae were fed with various input streams, which ideally resulted in a proteinaceous matter that was usually used as animal feed (Ugalde & Castrillo, 2002). One British research project concerned with this, aimed to develop a fungus- based ‘mycoprotein’ product for human use. After a decade of research and almost another decade of food safety assessments, it was first sold in 1985 as ‘Quorn’ (Sadler, 1988; Trinci, 1992). Research into soy continued, and processes were developed that allowed to reshape protein isolates into bundled fibres through spinning or extrusion (Johnson, Myers, & Burden, 1992). Resulting products such as textured soy protein (TSP) or textured vegetable protein (TVP) were brought to the markets throughout the late 1950s and 1960s, and were sold in chunks of varying sizes to the food industry. They were mostly used by food manufacturers in ready meals, or to increase margins of processed meats by adding cheaper plant protein (Collins, 1975). In 1969, the first company started to sell flavoured TVP to UK consumers, and was soon followed by

1994). I use the term seitan here to refer to all cooked gluten-based products as this is the common English name for such foodstuffs.

93 4. Contextualising Meat Alternatives other small businesses selling meat alternatives (Shurtleff & Aoyagi, 2014, nn. 533, 2686, 2345; see also Appendix A.2.1, and Rödl, 2019a).

4.3.2 Vegetarianism Becomes Visible (1985–1993) The year 1985 was identified as the beginning of the modern meat alternatives category in the UK for two reasons which will be illustrated in detail in the following paragraphs: Firstly, products of the now leading brand Quorn were first released in ready meals, making it some of the first or even the first foodstuffs in mainstream grocery chains based on meat alternatives; secondly, another company formative for the category for the next 15 years was created. For further reference, Table 4.1 summarises the different time periods, their changes, and their relevance. In 1985, grocery multiple Sainsbury’s released own-label vegetarian ready meals that included the newly approved Quorn as a replacement for meat in these dishes. Over the next few years, Quorn became ingredient to many more products from other supermarket chains and to products by other brands, usually including the Quorn brand name on product packaging. Starting in the early 1990s Quorn-branded products and ingredients for home cooking were sold; and by the mid 1990s, the brand had almost entirely shifted towards producing under their own brand and selling directly to consumers (The Grocer, 1995c). Also in 1985, UK-based soy processor British Arkady started buying small companies (many of which were its B2B-customers) that sold meat alternatives such as TVP directly to consumers; in 1988 they formed the Haldane Foods Group, which was continued to be expanded to serve the growing vegetarian and health foods market (Shurtleff & Aoyagi, 2014, n. 2686). Quorn and Haldane catered for a growing interest in vegetarian food products in the grocery multiples, and a few brands joined in the early 1990s; for example, Linda McCartney’s, which is runner-up in the market in 2018, was launched by frozen foods producer Ross Young’s in collaboration with vegetarian advocate and activist Linda McCartney (The Grocer, 1993d). Benefiting from being part of an established food manufacturer as well as being supported by a celebrity, the Linda McCartney’s5 brand soon became market leader in the “meat-free frozen meal centre market” (The Grocer, 1993c). The amount of people not eating meat increased steadily from 2% in the mid 1980s to about 4–5% in the early 1990s (The Grocer, 1994b; Vegetarian Society, 2001), and ‘health foods’ sections were established in all major supermarkets, with a small variety of meat alternatives nationally available (Shurtleff & Aoyagi, 2014, n. 2687). Just as vegetarianism, meat alternatives increasingly attracted businesses. In 1991, the market was valued at £25m, broken

5For more detailed narratives of the Haldane Foods Group, Quorn, and Linda McCartney’s, see Ap- pendix A.2.

94 4.3. A Brief History of Commercial Meat Alternatives in the UK

Table 4.1: Overview of time periods in the category narrative and the events identified that signposted changes.

Time Period Starting Events Developments

1985–1993 arrival of Quorn in major dedicated vegetarian food producers are retailers; beginning of the founded; vegetarianism is increasingly formation of the Haldane Foods catered for by retailers and food processors Group 1994–2001 The Grocer (1994b) publication a consolidation of the vegetarian market; “focus on vegetarian foods” growth also stimulated by meat scandals; arguing that vegetarianism had increasing awareness of limits to the market become socially acceptable when only vegetarians are targeted 2002–2011 category gains increasing Quorn is the only major advertiser and legitimacy in the food industry; targets mainstream audiences; increasing increased targeting of focus on health-conscious consumers; mainstream consumers increased interest across the food industry, but largely stagnant market value 2012–2015 increasing recognition of meat increasing competition from newly alternatives as a mainstream founded brands, with both more meat-like food choice (Phillips, 2012); and less meat-like products; increasingly sales slowly starting to grow profitable brands since 2016 booming sales (see Figure 4.2); competitors from abroad, large food growing interest in meat-free corporations, and the meat industry eating joining the meat alternatives market; meat alternatives visible to mainstream audiences; strong market growth down into the categories Quorn (£15m), TVP (£6m), and tofu (£4m, see The Grocer, 1993b).

4.3.3 The Healthy Choice (1994–2001) This increasing diversity for a designated vegetarian target group led to The Grocer publishing its first ‘focus on vegetarian foods’ article, which was identified to mark the beginning of the second time period. The same meat alternatives products which appeared in this article had previously been part of the magazine’s ‘focus on healthy foods’ or ‘focus on frozen foods’ reviews. The article argued that vegetarian foods had lost its “beard and sandals image” and was instead served by a “consumer pre- occupation with healthy eating” (The Grocer, 1994b), thus indicating a major shift

95 4. Contextualising Meat Alternatives in the perceived target groups of these products. The healthy eating trend covered mainly young and educated women, which also made up a large share of the vegetarian population (The Grocer, 2001); increases in vegetarianism and meat reduction may also have been related to a number of meat scandals that happened in the 1990s (see e.g. Beardsworth & Bryman, 2004). Many brands tapped into this arising mis- trust towards meat to promote meat-free foodstuffs and fought over market leadership with marketing activity such as a strong TV presence (e.g. The Grocer, 1996). This was related to an increasing recognition that the market would primarily target health-conscious people who reduced their meat intake in growing awareness of its downsides (The Grocer, 1999d). In the late 1990s, EU legislation about the labelling of genetically-modified organisms (GMO) in foods caused issues to brands with soy-based products. Consumers were generally not in favour of GMO, which in turn led retailers to threaten companies with a boycott if they continued using GMO. The public’s trust in soy-based meat alternatives dropped consequently, and Linda McCartney’s announced that they would switch to products based on wheat protein (Buckingham, 1999). Even some new and promising market entrants using soy, such as Khero aiming at health- conscious consumers, were said to be unable to sustain themselves in this emerging consumer scepticism (Dibb & Simkin, 2001). Product quality and market size had increased throughout the 1990s. By the end of the millennium, “cardboard-like soya sausages” were called a thing of the past (Brockes, 2000), as companies offered a wide range of products based on re-hydrated and further refined textured soy or wheat protein. Although the category had seen some ups and downs throughout the 1990s, it had outgrown initial expectations. As the food news platform Just Food noted in 2000, there were “just [not] enough vegetarians to be creating all this demand” for meat alternatives (Just Food, 2000). This led them to suggest that a distinction between people who liked to eat meat and those who did not was ‘outdated’. As limits to growth among vegetarian consumers were noticed, companies instead began to target health-focused meat reducers alongside meat-liking vegetarians (The Grocer, 2003a), which leads into the mainstreaming of meat alternatives in the next time period.

4.3.4 Slowly Towards Mainstream (2002–2011) The first Mintel report specifically on meat-free foods had been published in 2001 (Mintel, 2001) and in 2002, within the Food and Drink Federation (the main trade body of the food and beverage industry in the UK) a “Vegetarian (Meat-Free) Group” was founded (The Grocer, 2002d). Both these developments display an increasing recognition of the legitimacy and impact of the category, as meat alternatives had ar-

96 4.3. A Brief History of Commercial Meat Alternatives in the UK rived in all major supermarkets. However, according to Euromonitor statistics the market share increased on average at the level of inflation throughout the first decade of the 20th century (Euromonitor International, 2018). Still of minor significance to the overall food industry, the sector slowly gained interest as an investment: Quorn, for example, was sold to Montagu Private Equity in 2003, before becoming part of Premier Foods in 2005, who shortly afterwards also bought Cauldron–a manufacturer of premium tofu-based meat alternatives–and merged the two companies. And Linda McCartney and the assets of Haldane Foods were both acquired by Hain Ce- lestial (now Hain Daniels) in 2006. A few new competitors, such as Grassington’s, joined the market but could not persist for long, and the market leadership of Quorn remained unchallenged. The Linda McCartney brand built a strong third place behind own-label products, and after that a few smaller brands in varying proportions shared the rest. Throughout the decade, promotional activities to mainstream consumers were largely Quorn’s effort as the only company left to regularly advertise on TV (e.g. Mintel, 2010, see also section 4.5.6 below on marketing). This allowed the company to shape the public impression of the category to their ambitions.

4.3.5 Arising Competition (2012–2015) Only around 2012—together with an increasing awareness and appeal of vegan, vegetarian, meat-free, and meat-reduced diets–the market started to slowly grow again. Indicating another major shift in the understanding of the category, The Grocer noted that “the meat-free sector is ahead of total food in growth and smart NPD means it is becoming an increasingly mainstream market” (Phillips, 2012). The emerging competition and new companies in the market claimed to challenge the boredom and lack of quality of products from the market establishment. Products were introduced which were supposed to taste more like meat than those of brands already available; some other brands popularised meat-like foods but filled vegetables, pulses and other ingredients (Brooks & Brown, 2014, see also interviewees Tony and Rachel). In 2014, The Grocer printed its first report for a ‘meat-free category’ (Brooks & Brown, 2014), which can be seen as the final acknowledgement of this emerging market. An identified lack of vegan products was slowly catered for, both by incumbents who reformulated recipes and by new entirely vegan entrants (Brooks, 2015). The scale of this growth and more so the growing interest during this time period may be illustrated at the example of Quorn, which had been sold to a private equity investor in 2011. They pushed for major investments into marketing, and Quorn was sold off again in 2015 for more than twice its purchase price, whereas their market share remained almost stable just above 40% and the retail value associated with Quorn products increased by less than 40% (Euromonitor International, 2018).

97 4. Contextualising Meat Alternatives

a) 6 b) Published by 700 Euromonitor 5 Kantar Mintel 600 4

500 3 Per Capita Expenditure (£/year) Per Capita Consumption (g/year) 400 2

2001 2005 2010 2015 2018 2005 2010 2015 2018

Figure 4.2: Comparison of the per capita (a) expenditure and (b) consumption of meat alternatives based on market data from 2001 to 2018. Kantar calculates their estimates from consumption-level data, Mintel6 and Euromonitor use aggregated retail data; categorisation differences are likely to be responsible for the remaining differences in the estimates; each continuous represents data published at different times. The price adjustment of Mintel data was reversed; price index and population data were obtained from the Office for National Statistics.

4.3.6 Market Boom (since 2016) At some point between 2014/15 and 2016/17, the slow growth of the market that had been observed for a few years, became substantial (see Figure 4.2 later on in this chapter for values reported by different market research outlets), which was identified as the most recent major change for the category. Since then, vegan and vegetarian cooking and lifestyle magazines sprouted in large numbers, and mainstream media showed an increasing interest in meat alternatives, and campaigns such as Meat-free Mondays or Veganuary were increasingly taken seriously (e.g. Morris, 2018). Starting in the US and the Netherlands, there had been growing investment and interest in meat alternatives and new technologies to develop these (e.g. Zorpette, 2013; see also CB Insights, 2017). These new technologies, such as in-vitro meat or high-moisture meat alternatives, fostered expectations of increasingly meaty facsimile products (cf. Chiles, 2013), and around the world meat businesses and the big companies in the food industry started to invest into meat-free products (e.g. The Economist, 2015; CB Insights, 2017; Maurin, 2015). In the UK, this interest by the meat industry had come by a year or two delayed compared to other countries (see Rödl, 2019b), while interest by other multinational meat alternatives brands in the UK market caught up at a similar time.

6Ready meals, meat-free snacks, and pastry-based products were excluded from Mintel’s otherwise much higher estimates.

98 4.4. Meat Consumption in the UK

Interviewee Geert-Jan suggests that strong foreign meat alternatives manufacturers may have avoided targeting the UK market with its strong incumbents, but with increasing growth prospects this changed.

4.4 Meat Consumption in the UK

Without a brief overview of meat consumption, no study on meat alternative would be credible. Meat consumption in Britain has grown from around 20 kg to just under 80 kg per capita and year over the last two centuries (Smil, 2002). Although there have always been people abstaining from the consumption of meat (see section 4.3 above), around 90–95% of the British population eat meat on a more or less regular basis (Mintel, 2017; The Vegan Society, 2016; Waitrose & Partners, 2018); it has been reported that around 10% of the population considers themselves vegetarian or vegan, as another third reported reducing their meat consumption (Mintel, 2017; Waitrose & Partners, 2018). A survey in 2012 found that 70% of main dishes included meats or meat products as a centrepiece (Yates & Warde, 2015), while for 2016 Kantar World- panel (2016) reports that around 63% of evening meal occasions in October 2016 have involved meat, acknowledging a steady decline throughout the year prior. Similarly, as argued in section 4.5.1 below, the amount of vegetarians and meat reducers has increased over the last decade. Since 2010 the total per capita volume of fresh and processed meat and poultry7 has declined from 66 kg to 60 kg per capita and year in 2018, with processed products being significant less consumed (Euromonitor Interna- tional, 2018). Building on estimates by Kantar Worldpanel (2016), throughout 2016, the expenditure for fresh meat and poultry declined by 1.5% (£1.3 per capita and year) and by 3.7% for processed meat products (£2.1 per capita and year). According to this data, only poultry and poultry products, as well as fish were increasingly sought after, underlining a declining interest in red meat (Kantar Worldpanel, 2016). For domestic consumption, DEFRA (2017) estimated for 2015 an average expenditure for meat and meat products of £5.91 (929 grams) per week and household; this equates 2.4% of the total household expenditure, and around a quarter of the total food expenditure.

7The statistic includes the consumption of meat alternatives, whose change is negligible compared to these numbers.

99 4. Contextualising Meat Alternatives

4.5 The Market for Meat Alternatives in the UK

4.5.1 Consumers Consumers of meat alternatives and consumers who cut out meat of their diet are more likely to be female or young (Hoek, Luning, Stafleu, & de Graaf, 2004; Min- tel, 2013), with 25% of women between 16 and 24 not consuming meat (Mintel, 2017). Meat avoiders are also more commonly educated and live in urban areas. The proportion of vegetarians and meat reducers has been growing throughout the last decade: In 2004, a Mintel study found that 7% of women and 4% of men described themselves as vegetarian, while in 2017 Mintel found that 9% and 6% respectively followed meat-free and fish-free diets (Mintel, 2004, 2017). A considerably larger study by the Vegan Society in 2016 found that 3.25% of the UK population identified as vegetarian in 2016, while around 1.05% identified as vegan (The Vegan Society, 2016). Discrep- ancies in these numbers may originate in the questions asked: A report by Waitrose & Partners (2018) found that 9.5% of their respondents described themselves as vegetarian, and 3% as vegan; however, they note that only 40% of those report that they never eat meat, with others eating meat occasionally, on weekends, or when there are no other options (see also The Grocer, 1994b where a similar issue is emphasised). Meat reducers–conceptualised in a questionnaire by Mintel (2017) as those who “limit/reduce the amount of red meat/poultry . . . eaten in the last 6 months”—, were found to have increased from 17% of women and 8.6% of men in 2004 to 30% and 27% in 2017; a further 17% planned to reduce their meat consumption (Mintel, 2004, 2017). Generally, meat avoiders but also people cutting back on meat are more likely to consume meat alternatives (Mintel, 2017): The proportion of people who ate meat alternatives in the six-month prior to the study grew from 15% in 2006 (Min- tel, 2006b) to half of the surveyed population. Around 17% of consumers reported eating Quorn ‘weekly’, with other types of meat alternatives such as soy-based ones consumed weekly by 10% to 14% of people (Mintel, 2017). The most prominent reason to avoid meat is concern over animals, but environmental reasons, health, and food safety concerns related to animal agriculture (e.g. the use of hormones) were also prominent reasons for meat avoidance in 2017. Meat reducers are more commonly motivated by concerns over health and weight management (Mintel, 2017; see also Izmirli & Phillips, 2011; Tobler, Visschers, & Siegrist, 2011). Other reasons for the substitution of meat with meat alternatives are costs, concerns about food quality and safety (for example through food scandals like BSE, or the use of hormones and antibiotics in industrial animal farming), concerns about the social acceptance of the own diet, and meat alternatives as a transition aid to become vegetarian (Euromonitor In- ternational, 2017b; Mintel, 2004, 2006a, 2017).

100 4.5. The Market for Meat Alternatives in the UK

Some consumers have been found to value the neutral or positive taste of meat alternatives, although others see them as bland, uniform, boring, or not having meat- like tastes (Elzerman et al., 2013), and products were generally evaluated more positive when meat alternatives are used as ingredients in a dish (Elzerman, Hoek, van Boekel, & Luning, 2015). The texture of meat is often seen as one of the key features of meat (Kubberød, Ueland, Rødbotten, Westad, & Risvik, 2002), and is also often found a key criterion for consumers of meat alternatives. A. C. Hoek, Luning, et al. (2011) found that consumers who rarely or never consume meat alternatives but only meat, would like these products to resemble meat more closely; in contrast, those who regularly consume meat alternatives prefer them to be more distinct from meat, which is similarly reflected in attitudes concerning product name, preparation and packaging. Another study supports this perspective that certain consumers prefer meat alternatives which do not remind them of meat too closely (Elzerman et al., 2013). The consumer groups potentially consuming meat alternatives are quite hetero- geneous and with a variety of different tastes and motivations. McIlveen, Abraham, and Armstrong (1999) already suggested that only a high diversity of products and promotional activity could satisfy all consumers under this broad banner. Similarly, interviewee Eddy suggested that a meat alternatives category must cover the whole diversity of consumer desires in order to be valuable for retailers. To summarise, literature and market research suggests that generally consumers are prone to eating meat for sensory properties as well as for social or cultural reasons. However, an increasing amount of them opts out of eating animals for a variety of reasons; criticism on meat consumption is largely focused on the amounts of meat eaten as well as on the processes and side effects of animal agriculture8. This may indicate that audiences consuming meat alternatives, including those from the social media samples and target consumers of companies, are likely to be subject to a conflict of liking meat but not wanting to eat it. Furthermore, the consumption of meat alternatives seems to be highly modulated by experiences, expectations, and embedding of the foodstuffs; this may specifically be relevant in the consumer-focused studies of this research project.

8In fact, none of the reviewed surveys distinguish between not eating animals for moral reasons and not eating animals for the conditions they are reared in. Reported survey items of “concerns over animal welfare” (Mintel, 2017) or “animal-welfare concerns” (Waitrose & Partners, 2018) indicate for the surveys that not eating animals by principle is not an option.

101 4. Contextualising Meat Alternatives

4.5.2 The Market Measured by sales and volume, the UK market for meat alternatives is, after the US market, the second largest in the Western world. The total market value for ‘meat alternatives’ in the UK was estimated by Euromonitor International (2018) to be £393m in 2018. After New Zealand, the UK also has the second largest per capita consumption with 764 g per capita and year, and with £5.93 per capita and year the third largest per capita expenditure after Israel and Sweden (Euromonitor International, 2018, see also section section 4.6 for an international comparison). With most recent estimates of up to 20% annual growth in expenditure and volume (e.g. Mintel, 2018; Euromonitor International, 2018), meat alternatives in the UK are increasingly popular. As Figure 4.2 (see again above) shows, some sources indicate that this substantial increase started between 2014 and 2016. Before this, the annual per capita expenditure largely stagnated. As discussed earlier, these changes in market growth have been paralleled with increased business interest. For the future, there appears to be a general optimism in the market from my interviewees as well as from market analysts (e.g. Euromonitor International, 2017b; Frost & Sullivan, 2016; Mintel, 2017). In a report by Kantar Worldpanel (2016) which uses consumption-level data, increases in market value are attributed to existing consumers buying more, but also to more new consumers starting to buy meat alternatives. Prices for meat alternatives have increased recently and range around £7–10 per kilo (Euromonitor Interna- tional, 2018; Kantar Worldpanel, 2016), being only marginally bigger than prices for processed meats (Euromonitor International, 2017b). Kantar Worldpanel (2016) explicitly highlights considerably lower prices for the established brands such as Quorn and Linda McCartney’s, for which the average price paid per kilo was around £6.5 and £5.5 respectively in October 2016. Kantar Worldpanel (2016) suggests that meat reducers eat around half of all sold meat alternatives with a constantly increasing share; and Mintel (2017) found that every second consumer had eaten meat alternatives in the six months prior to their study, with for example Quorn being consumed by 17% of their respondents on a weekly or even more frequent basis. Although this finding makes averages less meaningful, according to Euromonitor International (2018) data, on average a British household spent £14 on meat alternatives in 2018, amounting to 1.98 kg per household per year (around £5.8 and 0.8 kg per capita and year). For 2016, Kantar Worldpanel (2016) using consumption-level data estimated the volume to be 4 kg on average per household that buys meat alternatives regularly. In 2018, the sales of meat alternatives are at around 1.2% of the total volume (retail and food service) of meat products (Euromonitor International, 2018, see also Figure 4.5)

102 4.5. The Market for Meat Alternatives in the UK

The brands Quorn and Linda McCartney’sare the leading brands by retail volume with 46.3% and 5.7% respectively of the total market in 2017. Considering their below market pricing, their volume shares can be considered to be significantly higher. Both brands are listed in all grocery multiples (see again Figure 4.1). With a reported own- label share of 15.6% in 2017, no other brand listed by Euromonitor International (2018) exceeds 3% of the market share. This leaves the other around 60 brands to share the remaining third of the market. Historically, Quorn has been continuous market leader, ranging from 40% to 46% of market share in between 2007 and 2017. Their main competitor Linda McCartney’s peaked in 2010 and 2011 with over 8% of value sales (Euromonitor International, 2018; Mintel, 2010 related this to ‘Quorn’s “weak” period’ and a national advertising campaign for Linda McCartney’s). However, the growing market and the appearance of more and more brands in UK retailers is likely to have major impacts on this balance. That the UK is one of the leading countries concerning meat alternatives consumption may be attributed among others to a long history of vegetarianism and veganism in the UK, as well as the over 30 years since meat alternatives first entered big retail chains. Likely, the availability of such products during the meat scares in the 1990s and 2000s, and the constant TV promotion of Quorn in mainstream TV throughout the last decade (see section 4.5.6) have contributed to these developments. As indicated by consumption figures, there are strong differences across consumer groups. Together with major differences in expenditure for meat and meat alternatives, this diversity of consumers gives an idea of how representative findings or implications of this research project might be.

4.5.3 Ownership Structures As the distribution of market share revealed, most brands in the market are considerably small. Such smaller businesses are usually owned by one or two individuals and have been founded as individual or family ventures. Some of such businesses have received financial backing by investors or partnered with larger companies, while others are reluctant to take out big loans or be invested in during their first years of business (any of these have been reported by my interviewees). Also, some larger, globally operating manufacturers are independent, such as the South African manufacturer Fry’s, or the US manufacturer Tofurky. Other ownership models exist, such as Ger- man tofu and meat-free brand Taifun which is owned by a trust since 2014, aiming to secure the independence of the business and its promotion of meat-free eating (Tai- fun Tofu, 2014). Most of the larger brands and a substantive portion of recent entrants to the UK market are however owned by larger food corporations. The more established ones

103 4. Contextualising Meat Alternatives have regularly changed ownership, including larger food corporations and private equity investors. Quorn and Cauldron, for example are part of Philippine food cor- poration Monde Nissin Corp, but have been owned by Astra Zeneca, Premier Foods, and two different private equity investors before. And while the trademark of Linda McCartney’s is family owned, the products are produced in licence by Hain Celestial Group, a publicly traded natural and organic foods producer from the United States; they were previously also produced by food companies such as Heinz, or United Bis- cuits. The convenience brand Dalepak which sells both meat and meat-free products is owned by ABP Food Group, the largest beef processor in the UK and Ireland; in early 2019, ABP also launched their own designated meat-free brand to the market. And Nestlé among others owns the Tivall brand from Israel and has other meat-free brands in UK supermarkets. Among others in the United States, there has been growing interest in meat-free alternative products by venture capital, which among others is granted by a number of the largest meat processors in the world (e.g. CB Insights, 2017); as of early 2019, some of these products have already been introduced to UK supermarkets, such as Beyond Meat’s burger.

4.5.4 Supply Chain Networks Apaiah and Hendrix (2005) describe the supply chain of manufactured protein products as containing primary production, ingredient preparation, product processing, distribution, retail, and finally consumption. Depending on the manufacturer, different degrees of vertical integration exist (which is illustrated in Figure 4.3). Quorn for example is supplied with wheat starch, a by product of wheat gluten manufacturing, as well as further nutrients and minerals, and grows and ferments a fungus in large vessels on site (Trinci, 1992). This primary protein is then further processed and shaped to a variety of pieces. Conversely, primary protein sources for all other products, such as legumes, grains, or whey, are an agricultural product. Sometimes these would be used on a contractual basis and be used directly as input to manufacturing, such as for German tofu processor Taifun; but mostly there would be an intermediate to produce a refined agricultural product, such as gluten or soy protein isolate. While these would mostly be further processed to textured protein material before being sold to the meat alternatives manufacturer, some companies such as Tiv- all make the protein material themselves (e.g. The Grocer, 1998b). These textured protein materials are then spiced, glued together, shaped, and fabricated into more value-added products, such as sausages, burger, pieces, mince, or ready meals (e.g. Dav- ies & Lightowler, 1998). While some companies such as Quorn rely on a single protein ingredient, according to Leyland (2008) it was seen as an advantage at Linda McCart- ney’s that various protein ingredients with their differing qualities could be used to

104 4.5. The Market for Meat Alternatives in the UK fabricate a variety of different products. Notably, some products such as vegetable or bean burgers often do not include any textured protein material, but may include soy protein isolate or flour. Lastly, there are own-label products in the market, which are fabricated under licence; furthermore, many smaller brands do not own manufacturing equipment but outsource production to other food processors.

primary ingredient protein material preparation production processing manufacturing & packaging vertically integrated e.g. Quorn

integrated manufacturing e.g. Cauldron

integrated production e.g. Tivall

food preparation e.g. Meat the Alternative labelling contract farming, supply agreements, e.g. own-label contract production, etc

Figure 4.3: Upstream supply chains.

Although textured protein materials tend to be dehydrated, most meat alternatives available to buy are fully re-hydrated and spiced so that they can be fried, roasted, or cooked immediately, or may be already fully cooked as well (cf. Davies & Lightowler, 1998). While dry products such as TVP are shelf-stable, most meat alternatives are sold frozen or chilled to appeal to consumers. Products used to be sold via mail order, and today many products of smaller brands are exclusively or mainly sold online. Health food shops historically offered a larger variety of meat alternatives than grocery multiples, and many brands are also listed by distributors. Due to the competition, low prices, and tight margins, smaller companies often do not sell in the major retailers; increasing competition through larger food businesses may arguably lead to even tighter margins. Meat alternatives are also becoming increasingly popular in food service, fast-food chains, or single ventures dedicated to meat-free fast-food, which is not explicitly covered in this thesis.

4.5.5 Laws, Regulations, and Certifications Neither the British Food Standards Agency (FSA) nor European legislation has any specific legal requirements surrounding meat alternatives. Nonetheless, all meat alternatives need to adhere to the legal frameworks valid for the production, sale, or import of any food item. The FSA (2016) has a special ‘Trade Information Sheet’ particularly focusing on meat alternatives, summarising the legal situation for imports of such products. For imports and manufacturing products of animal origin–which

105 4. Contextualising Meat Alternatives includes non-vegan meat alternatives–specific regulations apply. The FSA further issues ‘Salt Targets’ which include the categories “plain meat alternatives,” “meat-free products,” and “meat-free bacon” (2017); all of these are on the lower end of what is allowed for meat, while the upper limits can be more than 50% bigger for animal- derived meat. At the time this chapter was written, European legislation was still valid in the UK, and it was unclear whether this would continue to be. EU food legislation is concerned with food safety and consumer protection, stating that misleading consumers is prohibited. Novel food items that have not been sold before May 1997 are subject to EC Regulation 258/97, which requires extensive testing for ingredients newly isolated, modified on a molecular level, originating from algae or fungi, or subject to a process altering ingredient structure significantly. This would concern meat alternatives if the products were based on previously unused crops or manufactured using radically different processes. In case novel ingredients are intended to replace formerly existing ingredients, their consumption should also not be disadvantageous for the consumer (Food Standards Agency, 2015). This concerns aspects such as ingredients, allergens, claims of origin, health claims, or product descriptions (Avery, 2014; Food Standards Agency, n.d.). There is no legally binding definition of the terms ‘vegan’ or ‘vegetarian’ (either EU-wide or nationally), but third-party certifications do exist (European Vegetarian Union, 2016). These certifications are not standardised, and may be stricter or laxer than the common understanding of the term by consumers. For example, the Veget- arian Society does not certify products as ‘vegetarian’ which use eggs from battery hens (Vegetarian Society, n.d.-b), while this might not be an issue for ‘suitable for vegetarians’ declaration on supermarket packaging. For organic certification, which some meat alternative products also adhere to, there is EU-wide legislation as well as stricter certifications from third-parties (Janssen & Hamm, 2011). The EU further requires that the use of genetically modified food is clearly indicated on product packaging (European Commission, n.d.), which has been an issue for soy-based products when it was introduced (see again section 4.3). Other legislated substances, such as additives, preservatives, aromas, or flavourings (including so-called ‘E-numbers’), need to be indicated and may be further restricted in certified vegan, vegetarian, or organic products. Beyond ingredients, also naming of meat alternatives has been subject to considerable debates across the world, which are primarily calling for consumers not to be con- fused to think they eat meat when buying products such as ‘vegetarian meat’ or ‘bean burger’. For example, in France such legislation restricting the use of traditional meat terms in vegetarian or vegan products has been introduced in 2018, and in the Nether- lands national food authorities have asked one manufacturer to point out clearer that

106 4.5. The Market for Meat Alternatives in the UK their products are not of animal origin (Rödl, 2018b). In Germany according to 2018 guidance for the food industry, vegetarian products may carry generic format terms such as ‘sausage’, but not more specific meat-related terms such as ‘salami’ (Stähle, 2018). And in the United States, a petition was filed to the Department of Agriculture, which called for the words ‘meat’ and ‘beef’ to be forbidden on products that do not contain animal flesh in general or of cattle in particular (U.S. Cattlemen’s Association, 2018); this petition targeted both, in-vitro meat and plant-based meat alternatives. While in the European Union general regulations exist for non-dairy products to use common dairy food names (e.g. ‘milk’, ‘butter’, or ‘yoghurt’), this ruling has no effects on the labelling of meat or fish alternatives (European Court of Justice, 2017). The European Commission stated in December 2016, that they had no intentions to address the subject (Andriukaitis, 2016); however, there attempts by some EU parlia- mentarians to change this (e.g. Boffey, 2019). According to the European Vegetarian Union (n.d.), in countries where meat alternatives are rare, the meat industry is particularly opposed to ‘meaty names’ for vegetarian products. There are ongoing debates about potential regulations or political attempts to curbing meat consumption. While no national legislator has yet taken action to reduce meat consumption, plans to facilitate this–for example with a meat or a carbon tax–have been reported for China, France, or the Netherlands. These regulations would likely affect consumption of meat alternatives positively (Euromonitor Inter- national, 2017b).

4.5.6 Marketing and Consumer Engagement In terms of marketing, especially the market leader Quorn stands out: It engages in regular national TV advertising campaigns, and has dominated and determined the total advertising expenditure of the meat alternatives market for over a decade (e.g. Mintel, 2017, see also Figure 4.4). The total advertising expenditure of the market is largely related to their TV advertising campaigns, in which they regularly promote their product with the support of UK athletes (Mintel, 2004, 2011, 2017; The Gro- cer, 1995d). The only other TV commercials for meat alternatives in the last decade were run by Linda McCartney’s in 2011 and 2013 (Mintel, 2011, 2013). Mintel (2001, 2006b) explain spikes in advertising expenditure around 1996 and 2002 as responses to ongoing food scares, such as BSE; conversely, the dip in advertising activity in 2010 has been attributed to Quorn-owner Premier Foods reduced investments before the brand was sold to a different investor (Mintel, 2011; Watson, 2011b). By the mid-2000s, Quorn was left as the only major advertiser, while advertising expenditure for Linda McCartney’s declined just like the brand’s market share (Mintel, 2004). Analysts have argued, that the market would benefit from more investment by other companies, to

107 4. Contextualising Meat Alternatives

Advertiser 7.5 Total Marlow Foods

5.0

2.5 Advertising Expenditure in £m 0.0 1995 2000 2005 2010 2016

Figure 4.4: Advertising expenditure from 1995 to 2016, including the share of Mar- low Foods (owner of Quorn; the Cauldron brand was bought in 2005, but without major advertising investment). Multiple lines indicate numbers from different reports. Compiled from: Nielsen Ad Intel/Mintel (2000, 2001, 2004, 2006, 2008, 2010, 2011, 2013, 2017).

“help to build on the current momentum in the market” (Mintel, 2017), and become more independent of only one company’s advertising strategy.

Other companies with smaller budgets may run print advertising (Mintel, 2004; The Grocer, 2016), which tends to be printed in specialist magazines focusing on health or vegetarianism, but occasionally also target the “women’s or mainstream food press” (Mintel, 2004). Many smaller brands rely on word-of-mouth, loyal customers, and at present also digital marketing strategies. Sponsoring of vegetarian events such as the National Vegetarian Week or the World Meat Free Week (see also chapter 8) is a popular advertising investment. In larger supermarkets, promotions for meat alternatives are frequent, for example with cross-promotion, 3-for-2 deals, recipe advertising, promotional leaflets, freebies, discounts, or coupons (Mintel, 2004, 2006b, 2008, and also interviewee Eddy). By now, all brands have a website, with some smaller brands relying on their own web shops for selling products. Websites also allow these brands to inform their consumers about products, stockists, recipes, company values or missions, health or lifestyle content, and contact details. Social media and networks are used to promote products, share content such as recipes, or spread health, lifestyle, or activist information, while also being a convenient way for consumers to get in touch and interact with the brand.

108 4.5. The Market for Meat Alternatives in the UK

4.5.7 Relations to Meat and its Industry As the names for products such as ‘meat alternatives’, ‘meat substitutes’, or ‘meat-free foods’ suggest, there might be links between meat and meat alternatives, which is a major concern of this thesis. For example for consumers, sensory and functional elements of products may matter (see above). Yet, there are also links between the meat industry and meat alternatives. For example, many large corporations which own meat-free brands, such as Nestlé or Heinz, also process meat products; and some brands that are now manufacturing meat alternatives are primarily known for their meat products, such as the UK brand Dalepak, or the German brands Rügenwalder Mühle, or Wiesenhof which are part of major poultry processors. More generally, an increasing investment of big meat processors into meat alternatives start-ups as well as established businesses can be noted (e.g. CB Insights, 2017). Apart from Perkins Foods’ failed attempt around the turn of the millennium (with the Khero brand, Dibb & Simkin, 2001), no prominent UK-based meat processor had ventured into branded meat alternatives products for a long time. In early 2019, UK meat processor ABP Food Group launched their own meaty meat-free brand (Rödl, 2019b), after they had produced vegetable-based meat alternatives for the UK own-label market before. Besides meat-free health food shops or vegan distributors, meat alternatives and meat are stocked in similar retail outlets and in meat reducers share a large target group (see above). However, in most UK supermarkets they tend to be shelved differently, but that seems to be slowly changing (e.g. Smithers, 2018). Looking at the upstream supply chain, soy is both, main feed ingredient in animal agriculture, and the most commonly used vegetable protein for meat alternatives; in 1998, it was estimated that 1–3% of the total soybean protein in the US was for human consumption (Frost & Sullivan, 2001). In fact, in the 20th century, soybeans became increasingly common as a feed, thus increasing availability and lowering market prices of soybeans, which allowed for the consideration and development of soy-based meat alternatives (Johnson et al., 1992). Some meat products are fortified with plant protein (called meat extenders, see also Asgar, Fazilah, Huda, Bhat, & Karim, 2010), and some ‘flexitarian’ ranges merge meat and vegetables or vegetable proteins into novel products (Mintel, 2017). In fact, Frost & Sullivan (2010) report that in 2009 in the EU, around 32% of all refined soy protein (isolates, concentrates, textured) was used the meat processing sector (Frost & Sullivan, 2010). Making similar products, the industries for meat and meat alternatives are also connected in terms of research and development. Common overlapping themes and technologies are for example extrusion technology or emulsions. For processed food products, such as sausages and cuts, production of meat-based and meat-free products can be similar. Despite its much higher revenues (see also Figure 4.5c), in 2008 the marketing ex-

109 4. Contextualising Meat Alternatives penses in the meat industry have been reported to be only slightly higher than for the meat alternatives sector (Mintel, 2008). The challenges for these industries are nonetheless fairly similar, as consumers demand healthier and more sustainable products; meat consumers are sought to be convinced by lean and premium cuts with lower fat levels, or with specific animal welfare standards (Key Note, 2015). Therefore, the increasing interest in meat alternatives by established food corporations in general, and meat processors in particular should not come at a surprise given the recent developments in the market. Essentially, most meat reducers and meat avoiders are or used to be customers of the meat industry, and thus provide a huge potential. The power that the meat industry can develop here is not only based on available funds, or on negotiating power with supermarkets, but also ideational as German poultry processor Rügenwalder Mühle (n.d.) argues in the FAQ on their website: “If anyone has the competence to produce vegetarian alternatives so that they taste like meat and sausage, then it should be us.”

4.6 Other National Markets for Meat Alternatives

The UK is the second largest market for meat alternatives in the Western world (measuring total retail value and volume) after the USA and before Germany, Canada, and the Netherlands. Per capita, the UK is the second largest market in terms of volume after New Zealand and before Israel and Sweden; and by retail value per capita, the UK is third largest market, after Israel and Sweden, and before The Neth- erlands and Denmark (Euromonitor International, 2018). In this section, three of the strongest markets for meat alternatives next to the UK–the USA, the Netherlands, and Germany–are briefly introduced. A timeline of the total and per capita volume and retail value, the ratio between sales of meat and meat free products, and development of unit prices in the compared markets is visualised in Figure 4.5. From panels (a) and (b) It can be firstly noted that meat alternatives are on the rise in all four countries. Germany’s market had started from near nothing around a decade ago, with all other markets having substantial presence before. Compared to the Netherlands (see panels d and e), it becomes clear that in the UK growth on the meat alternatives market happened before 2004, while the Netherlands showed more growth in the early 2000s. As noted in previously concerning, the strong standing of the UK might be related to a long vegetarian and vegan history, as well as strong brands throughout the last decades. Panel (f) reveals that meat alternatives in the UK are considerably cheaper than in other countries; however, as comparatively low 2018 exchange rates were used for currency conversion, the real difference at the time can be expected to be lower. Trends are obvious nonetheless, with UK prices slightly increasing, whereas considerably higher Dutch

110 4.6. Other National Markets for Meat Alternatives 2004 2010 2018 2004 2010 2018 8 12 10 1% 0%

1.2% 0.2%

0.4% 0.8% 0.6% of fresh and processed Meat processed and fresh of

Alternatives / Total Volume Total / Alternatives Unit Price (GBP / kg) / (GBP Price Unit Retail Volume of Meat of Volume Retail c) f) Data from Euromonitor International. Currency conversion using 2018 exchange rates. 2004 2010 2018 2004 2010 2018

7.5 2.5 5.0 0.2 0.4 0.8 0.6

0.0 0.0 12.5

10.0

(Retail Volume in million kg) million in Volume (Retail

(Retail Volume in kg per capita) per kg in Volume (Retail

Annual Consumption Annual Annual Consumption Annual b) e) United Kingdom Germany Netherlands USA Western Europe 2004 2010 2018 2004 2010 2018 2 4 6 0 0 Comparison of the markets for meat alternatives in UK, US, Germany, the Netherlands, and WesternEurope. (a) Totalannual expenditure

500

1000

(Retail Value in mGBP) in Value (Retail

(Retail Value in GBP per capita) per GBP in Value (Retail

Annual Expenditure Annual Annual Expenditure Annual d) a) Figure 4.5: meat alternatives; no data foralternatives. Western All Europe); currencies (d) were converted annual using expenditure 2018 per exchange rates. capita; (e) annual sold volume per capita; and (f) unit price of meat (Retail Service Price); (b) total annual sold volume; (c) retail volume of meat alternatives relative to total volume of fresh and processed meats (including

111 4. Contextualising Meat Alternatives prices were decreasing. In line with these growth patterns, Dutch consumption might have increased due to price reductions, while on the UK according to interviewee Eddy, market prices needed to be kept low to keep customers. In comparison, the US market which increased unit prices by around a quarter throughout the covered time period, as consumption stagnated. The less linear developments in the German market are likely related to different market entrants and rearrangements of the still young market. In panel (c) it further becomes obvious that in none of these countries the consumption of meat alternatives is remotely as profitable as the meat industry; while the relative volume of meat alternatives increased across all countries, the UK benefits from cheap prices making volume comparisons much better than in other countries (cf. panel d and e). In the West, most meat alternatives in absolute terms are sold in the USA. By per capita sales, the market ranges 10th, and by volume 14th globally (Euromonitor Inter- national, 2018). Historically, the US have been central to the development of meat alternatives globally (e.g. Johnson et al., 1992). In the UK, these developments have at times been looked at with fear or a lack of understanding; for example an article in The Guardian from 1969 titled “Meatless eat–synthetic way to the US heart” (Raphael, 1969), where the author describes recent trends in the industry. Since the late 2000s, the Californian Silicon Valley is host to an increasing number of food and biotech businesses, a number of which are concerned with developing meat alternatives, and are not seldom backed by the meat industry (CB Insights, 2017; The Economist, 2015) In 2016, a number of companies formed a lobby organisation called the Plant Based Foods Association, which gives its members among others a common voice in policy making. The Netherlands has the fifth largest market for meat alternatives in the West by retail value (per capita: 4th), and the seventh largest by volume (per capita: 6th, Euromonitor International, 2018). The meat alternatives market in the Netherlands exists at least since the early 1990s, when some brands like Vivera were founded and the government supported elaborate research into reducing meat consumption and development of so called Novel Protein Foods (see Quist, Thissen, & Vergragt, 2011). Meat alternatives are a common sight in supermarkets, can sometimes be found in meat aisles, and some supermarkets even stock insects and insect-based foods as alternative protein products. Cooperation in the industry is among others supported and facilitated by The Protein Cluster, which encourages and supports collaboration among different manufacturers, or the Green Protein Alliance, which aims to facilitate a ‘protein transition’. The German market for meat alternatives is the third biggest globally by volume and by retail value, and ranks eleventh and tenth in the West concerning per capita retail value and volume respectively (Euromonitor International, 2018). Historically,

112 4.7. Conclusion in Germany meat alternatives have had much lower visibility albeit tofu, seitan and derivative products could be bought in larger retailers. As the market grew from nearly no offerings in the early 2000s, in 2015/16 two of the largest German poultry processors and producers of sausages and sliced meats launched vegetarian versions of their products (e.g. Maurin, 2015). Afterwards the market gained traction and attention, and many brands known for their meat products followed. Between 2012 and 2017, per capita meat alternatives retail volume has increased by 179% and value by 140%, making Germany have the sixth largest relative growth in the Western world over this time period; per capita sales increased by around USD 1.8 (4th) and 170g (5th) in this time period (Euromonitor International, 2018). Many of these products are made from egg and milk protein and have the same branding, packaging, and similar naming as their meat counterparts and tend to be placed in the same supermarket shelves. This visibility has likely helped the rapid growth of the market.

4.7 Conclusion

In this chapter, the empirical background to this study was reviewed both longitudinally and from a current perspective, as well as in international comparison. With a long history and almost a decade of stagnation, it seems that since around 2016, the market for meat alternatives in the UK has been growing substantially. On top of the traditional vegetarian audience, an increasing amount of vegans as well as meat reducers have to be catered for. Particularly meat reducers because of their large numbers provide business opportunities (cf. Kantar Worldpanel, 2016) which likely attracted an increasing amount of funds and larger food corporations to the market. Although descriptions were sought to be as recent as possible, the thesis at large and most reporting in this chapter ends in 2017. Likely, by now there are more than the estimated 60–70 brands, as a more diverse offering can be expected particularly by major grocery retailers. Generally, there should now be more competition involved in the market than ever since the 1990s when companies fought in expensive advertising campaigns about the growing vegetarian audiences. This increasingly strong competition from abroad, supported by the meat industry or large food corporations, might slowly dissolve the traditional power of the Quorn and Linda McCartney’s brands. These have been and likely still are certainly the most visible brands for a mainstream audience. How contested the market has become recently, is among others indicated by the retraction of Nestlé’s Garden Gourmet brand in early 2019 (e.g. Askew, 2019; Graham, 2019). The findings from this chapter have a number of implications for the following data analyses: The social media data is likely–but not without exceptions–to originate from more engaged audiences with a broader awareness of their options. Following

113 4. Contextualising Meat Alternatives the findings of A. C. Hoek, Luning, et al. (2011), this means that they are likely to be satisfied with less meat-like products or do not endeavour their consumed meat alternatives to be more like meat, which may also cause them to overstate the meatiness of their consumption experience. Similarly, many of the interviewed manufacturers and the advertisements identified from vegetarian journals do not represent these larger companies which appear to gain traction in the market. Readers of vegetarian journals might not represent the prime target group of all brands and throughout all episodes, thus being only an approximation of meat alternatives in advertising discourse. The interviews might have been conducted slightly too early to account for the growing competition in the market, which in the UK has been starting slightly later than elsewhere in the world.

114 Chapter 5

Print Advertisements for Meat Alternatives

This chapter is the first of four chapters which present the collected data, and outlines findings for the respective research questions. In this chapter, the sample of collected advertisements of meat alternatives is described in detail. Advertisements have been chosen as they represent a meaningful and unmediated point of interaction between producers and consumers, and span long time periods (see again section 3.3.2). The analysis presented in this chapter includes firstly changes in advertisement during the hypothesised emergence of the meat alternatives category (Qa), which spans the whole data set from 1985 to 2017. The results of this are merged with the narrative of meat alternatives in the UK, which was presented in the previous chapter. Secondly, a subset of advertisements (printed in 2016-171) is analysed regarding the present category identity (Qb), and the relation between meat alternatives and meat (Qc). This subset was taken as the adverts were published at the same time as the interviews, videos, and tweets to be presented in the following chapters. Their coordination in time allows for more accurate synthesis of their content. The advertisements were analysed using a guided content analysis (Qa, Qb) and frame analysis (Qc) as described in section 3.3.2. The remainder of this chapter proceeds as follows: Following a description of the collected data, the findings concerning category emergence (Qa) on the full data set are described. Based on the advertisements printed in 2016 and 2017, the findings on category identity (Qb) and relations to meat (Qc) are then presented. Finally, a brief summary and conclusion of the findings are offered.

5.1 Description of the Data

The final sample consists of 557 printed and 319 unique adverts. They originate from three different magazines, namely Cook Vegetarian! (succeeded by Veggie in 2015; 367 adverts in total, 3.5 adverts per issue), Vegan Life (34, 1.3), and The Vegetarian (156,

1Note again that data collection stopped in November 2017 and all issues available by that time were included in the data.

115 5. Print Advertisements for Meat Alternatives

1.0). However, in the latter magazine, in only half of all issues advertisements for meat alternatives could be identified. For a longer period since 2015—possibly because of an editorial change that reduced the amount of advertisements overall–as well as in the mid-1990s2, the magazine has not printed any advertisements for meat alternatives. More generally, The Vegetarian had less overall advertisements per issue than the other two sampled magazines. In Figure 5.1, an overview of the adverts identified in the different magazines is presented. More than two-thirds of advertisements (n=385, 69%) were full page advertisements; seven advertisements covered two pages; 82 adverts were printed on half a page (15%) either vertically or horizontally split; nine adverts covered a third of a page; and 74 adverts covered a quarter of a page (13%; the minimum size to be included in the sample). On average an advert covered 0.83 pages (SD=0.321). Adverts were particularly concentrated in the beginning and the end of magazines, with 13% of all adverts being printed on page two (the inside of the front cover), 15% on the last page (the back cover), and 7% on the inside of the back cover. The remainder of the advertisements are distributed throughout the magazine with certain page numbers being more prominent than others, probably due to established content guidelines; the most frequent page outside of front or back cover was page 19 (n=15, 3%). Both, the generally large size of the adverts and a high concentration of adverts on front and back cover, are indicative of a generally larger amount of money paid for the adverts and the prestige sought to be acquired in comparison to other advertisers in these magazines. In total 41 advertisements (7%) are advertorials, which is paid for advertising content in a style similar to the magazine. These are often reports of brand or product novelties, and are labelled with ‘advertisement’, ‘promotion’, or similar in order to be identifiable by the audience. All 557 advertisements sampled belong to 46 different brands, one of which carried two different names throughout the sampling period. The median amount of advertisements per brand are four, and twelve brands only have one advert in the sample. By far the most frequently advertised brands are Fry’s with 116 adverts (21%), and VBites with 109 adverts (named Redwood until 20133; 20%); all brands with more than ten advertisements in the sample are listed in Table 5.1. While some brands were advertised over long time periods, others had focused advertising campaigns over short time

2To my best knowledge, during the mid-1990s there were no other UK-based meat-free or vegetarian magazines of considerable size. It is unlikely that this gap represents a lack of innovation or advertising activity in general, as in the mid-1990s many vegetarian-focused TV advertising campaigns were run, which Mintel (2001) relates to increasing BSE reporting on the opportunities for vegetarian products. This gap is therefore likely to be a result of the magazine’s advertising practice. 3Two advertisements in Cook Vegetarian! in June and July 2013 announced this name change: “The Redwood Co are changing their name to VBites.” Rewdood was bought by the owner of the VBites trademark in July 2009 (The Grocer, 2010).

116 5.2. (Qa) Longitudinal Analysis of Changes in Advertising (1985 to 2017)

Number of Adverts 20

0 '85 '90 '95 '00 '05 '10 '15 '17 Year The Vegetarian Cook Vegetarian! Veggie Vegan Life

Figure 5.1: Overview of the sampled advertisements by year and magazine. The vertical lines separate the different time periods used in the analysis. Note that no advertisements in issues published after November 2017 are included in the sample. periods; the average number of adverts printed per brand and year in which the brand printed adverts is 2.7 (SD=1.94). Later time periods showed a larger concentration of brands; highlighting this change, Table 5.2 lists the relevance of the largest brands for the overall advertisement count. Concerning the products sold, around a third of advertisements showed a wide variety of products that the brand offered (n=195; 35%), while another 17% (n=96) adverts promoted a range of products that were related, for example through a shared naming such as ‘deli products’. The most common product formats advertised in isolation are burgers (n=70; 13%), sausages (n=28; 5%), roasts (n=21; 4%), and fish (n=20; 4%). 55 advertisements (10%) promote other products such as slices, breaded formats (schnitzel or cordon bleu), mince, pieces, steak, ready meals, nuggets, or bites. 60 advertisement (11%) promoted a brand but no particular products or product ranges.

5.2 (Qa) Longitudinal Analysis of Changes in Advertising (1985 to 2017)

(Qa1) Firstly, up to two labels and headlines per advertisement were coded. Advert- isements for meat alternatives were often headlined or labelled in ways different from the actual product names or what was written on packaging. More than every other

117 5. Print Advertisements for Meat Alternatives

Table 5.1: Brands in the advertising sample with fifteen or more printed adverts, including the duration of their activity, the amount of printed advertisements per year, and their recent advertising activity.

brand adverts years active per year in 2016–2017

Fry’s 116 15 7.7 28 Redwood / VBites 109 19 5.7 18 Goodlife 43 9 4.8 2 Quorn 27 11 2.5 – Linda McCartney’s 25 5 5.0 22 Dalepak 22 8 2.8 – Grassington’s 17 2 8.5 – Vegusto 16 3 5.3 –

Table 5.2: Amount of advertisements across the time periods which are printed by the two and four most active advertisers

Time Frame Advertisers Adverts Top 2 Advertisers Top 4 Advertisers

1985–2001 15 47 19 (40%) 27 (57%) 2002–2011 23 222 71 (32%) 120 (54%) 2012–2015 19 197 106 (54%) 134 (68%) 2016–2017 10 91 50 (55%) 74 (81%) advertisement (53%) related to meat either directly–with labels like meat alternative or alternative to meat, meat substitute, meat-free or fish-free, beef-style and similar, or as fish–or via meat formats–such as sausages, roast, burger, schnitzel, or bacon, but also meat-like food categories such as ‘deli ranges’. Slightly less than a quarter of adverts respectively are labelled with dietary terms (usually ‘vegetarian’ or ‘vegan’; 22%), or purposive labels such as ‘food for all occasions’, ‘Fajita Strips’, or more illustrat- ively “The ‘Penny’s latest boyfriend doesn’t eat meat’ pak” (2002-05; 23%). Contents were used by 15% of adverts as part of a headline, which covers ingredients in general, but also ‘protein’ and ‘plant-based’ labels. Another 23% of adverts described the advertised products generically such as ‘everyday food’, ‘good food’, ‘the choice’, or ‘our range’, or classify the food within supermarket categories such as “Chilled and Frozen” (2008-10) or “Storage Cupboard Essentials” (2010-03). 10% of adverts focused exclusively on the brand and not on products, and 7% of adverts did not have a prominent headline that elaborated on the product but instead featured a large-scale image of a

118 5.2. (Qa) Longitudinal Analysis (1985 to 2017)

Table 5.3: Overview of labels and headlines used for the products in the advertisements (Qa1). An advert could be coded with up to two labels.

2 Label nC χ p Examples

format 164 33.5 0.000 “The Best Burger Quest is over!” (2015-13); “Take a pack of tofu mince” (2011-17); “As always, a new roast from Granose is something rather special.” (1989-05); “Delicious Vegan Food From Our Vegideli Range” (2012-03) meaty 159 26.7 0.000 “The great taste of bacon without the pig” (2001-01); “World class Swiss . . . meat alternatives” (2013-12); “Fish Style Fingers Sandwich” (2016-09); “Be one of the first to try a NEW meat substitute which tastes as good as it looks!” (2008-12); “All the taste of the ocean but with none of the fish!” (2012-10) purpose 128 41.2 0.000 “The ‘Penny’s latest boyfriend doesn’t eat meat’ pak” (2002-05); “Back to school lunches” (2013-20); “New Decade Diet?” (1990-01) others 124 36.1 0.000 “Food Made Good” (2015-04); “Good food, no faff.” (2000-05); “Freeze your imagination” (2004-05); “Storage Cupboard Essentials” (2010-03) dietary 125 29.2 0.000 “100% Vegan” (e.g. 2006-04); “Simply Vegetarian! Simply Irresistible!” (2003-05); “100% vegetarian, 200% taste” (2011-04) contents 81 10.7 0.013 “Take a fresh look at vegetables” (1989-03); “Nature’s Plant Proteins” (2016-01); “We’ve got a vegtastic new range!” (2014-27) brand 56 15.2 0.002 “Linda McCartney’s — Deliciously Vegetarian” (e.g. 2017-01); “So easy, so tasty, so spicy! Quorn Fajita Strips” (2005-02) “Enjoy the taste fo Spring with Redwood Foods” (2013-07) visual 40 13.1 0.004 — serving suggestion. An overview of the labels and representative examples is presented in Table 5.3. The distribution of all aggregated labels proved to be independent of the time periods (p < 0.05, for values see Table 5.3); in terms of individual codes with numbers above the minimum threshold of 31 coded advertisements (see again section 3.3.2), only labels such as ‘meat alternative’ as ‘meat-style’, and those concerning ingredients do

119 5. Print Advertisements for Meat Alternatives not significantly differ across time periods (see Appendix B.7 for all values). Because of uncertainties about interpretation and limits to trustworthiness, all groups of codes which represent patterns which are not linear or U-shaped (e.g. zig-zag) are therefore ignored in the following analyses (see again section 3.3.2) and only noted. In case of labels, this concerns dietary, purposive, and format-based labels which all oscillate. The following observations4 could be made:

1) Labels relating to contents, brands, and visual focus peaked in the latest time period (in Figure 5.2 these developments are illustrated across time periods). 2) Advertisements labelled with meat-like terminology, and those with generic food-focused headlines peak in the time period from 2002 to 2011 in an inverted- U shape.5 Their relative frequencies in the first and the last time period are approximately equal and below half of their maximum relative frequencies.

A total of 21 different descriptions could be identified in the data, whereby each advert could have 0 to 4 codes for descriptions (see Table 5.4). These codes were grouped into six categories: Ingredients and health credentials (64%), focus on the food format and its appearance (53%), sensory aspects (46%), dietary credentials (42%), descriptions on use and function of the product (31%), and the worldly context of foods (11%). Ingredients and health credentials cover what ingredients are included (20% of total), what is not included (e.g. cholesterol or saturated fats; 29%), but also health claims (15%), claims for naturalness (27%), product quality (6%), mentions of protein content (8%) and the plant-based origins of the product (2%). Product appearance in large scale images is covered in 23% of adverts, while appearance is covered in writing in only 3% of adverts, for example by a Grassington’s advert for “a NEW meat substitute which tastes as good as it looks!” (2008-12). More central than appearance is the food format which is explicitly used to provide a description of the advertised products in 37% of advertisements. Sensory aspects mentioned in advertisements cover taste at large (46%) with a few also relating to product texture, such as by describing the product as ‘succulent’ or ‘juicy’. Descriptions of potential use and function of the product emphasise among others convenience (16%), storage temperature (10%), the usefulness of this product for traditional or established practices (6%), or price (2%) A small share of adverts also embeds the advertised product or meat into wider ‘worldly’ contexts, such as by referring to sustainability issues (5%), animal rights/welfare (3%), authentically made products (for example raw material “harvested from local farmers”, e.g. 2013-24; or products made by “the Vegenistas!”,

4For clarity, these observations and their numbers will be referred to again in chapter 9. 5This peak can be attributed to manufacturer VBites (then Redwood), as they printed many advertisements relating to meat through specific meat ‘styles’. When the impact of this company is removed, labels relating to meat are distributed more evenly.

120 5.2. (Qa) Longitudinal Analysis (1985 to 2017)

40.0% others meat−related only visual focus on the brand focus on the purpose 20.0% focus on the format ingredients vegetarian / vegan food

0.0% 1985−2001 2002−'11 2012−'15 2016−'17

Figure 5.2: Development of the use of the labels and headlines in advertisements across the time periods.

Table 5.4: Amount of average descriptions coded per advert in each of the time periods; each advert could have up to 4 codes for product descriptions (Qa1).

Time Frame Adverts Average Descriptions SD

1985–2001 47 2.32 0.86 2002–2011 222 2.47 0.77 2012–2015 197 2.32 0.73 2016–2017 91 2.11 0.90

2015-06; 3%), or statements that products are “British manufactured” (e.g. 2016-09; 1%). Lastly, dietary credentials with descriptions of the foodstuff as vegan or vegetarian are covered in 42% of all adverts. All aggregated codes and their trajectories across the time periods are illustrated in Figure 5.3; statistical measures and examples are presented in Table 5.5. The following observations can be made:

3) Statistical evaluation on the amount of coded descriptions (see Table 5.4) revealed that this decreased significantly in the later time periods (using Welch’s two sample t-test in R: t(415.54) = 2.0248, p < 0.05 from 2002–2011 to 2012–2015; and t(146.12) = 1.9983, p < 0.05 from 2012–2015 to 2016–2017). 4) Descriptions of use and function of the products have declined since the first time period, as convenience was most prominently reported in 1985–2001, and

121 5. Print Advertisements for Meat Alternatives

Table 5.5: Overview of the descriptions used for the products (Qa1).

2 Description nC χ p Examples

ingredients 356 25.2 0.000 “free from animal ingredients, cholesterol, hydrogenated and health fats, . . .” (2012-14); “Organically grown” (2009-12); “the food your body would choose” (2014-26); “get more of this healthy superfood in your diet” (2009-07); “a blend of plant protein and wholefoods” (2016-08) focus on the 295 37.8 0.000 “Organic Wholefood Burgers” (2011-15); “Vegetable format Kievs” (2003-04); “Chicken-Style Louisiana Tenders” (2012-05); “makes them cook like traditional sausage” (2014-15) focus on 256 4.7 0.199 “which tastes as good as it looks” (2008-12); “kids love sensory them!” (2011-08); “enjoying the big burger texture and aspects taste” (1991-02); “The real secret is the sauce inside. Spicy Korma for the Indian style variety, delicious Sweet & Sour for the Chinese style . . .” (2003-02) vegetarian / 232 7.0 0.071 “Vegetarian Style Tuna” (2004-01); “100% vegan” vegan food (e.g. 2012-16); “not everyone knows how to cook for vegetarians” (2002-05); “Go Veggie!” (2014-09) use and 170 18.9 0.000 “delicious and convenient” (2014-16); “our frozen range” function (2013-01); “being meat-free shouldn’t mean compromising on great festive food” (2016-30); “excellent value for money” (2010-02) worldly 61 5.2 0.155 “the very first to guarantee a clean plate as well as a clean context consciousness” (2009-06); “50% less packaging” (2007-01); “great news for turkeys” (2002-10); “leave it to the vegenistas!” (2015-06); “Best of British Barbecue” (2014-19)

storage temperature and price in 2002–2011; the description of the product as a traditional or established food did not differ significantly across time. 5) The focus on ingredients and health has peaked in 2002–2011, and declined since. However, constitutive of it this group, most recently protein and plant- 2 6 based credentials (jointly χ (nC = 49) = 90.0, p < 0.001) became increasingly important, whereas focus on product quality, naturalness, and what’s not in-

6Note that because the χ2-test throughout this chapter has been applied with the same data set, I do not report the degrees of freedom (df = 3) or the number of valid cases (N = 557) specifically; instead I report the number of total adverts which were coded in the reported way using nC.

122 5.2. (Qa) Longitudinal Analysis (1985 to 2017)

2 cluded in a product declined since 2002–2011 (jointly χ (nC = 261) = 17.1, p < 2 0.001). Notably, mentions of ingredients (χ (nC = 110) = 1.8, p > 0.1) and 2 health aspects (χ (nC = 82) = 0.4, p > 0.1) did not differ significantly across time periods. 6) Changes in the advertised relation between products and worldly contexts did not significantly differ. However, within this group focus on sustainability and animals proved to change with time, having peaked in the 2002–2011 time 2 period and only one advert each in the first and last time periods (jointly χ (nC = 39) = 0.6, p < 0.05). Conversely, the Britishness of produce is exclusively mentioned in the most recent time period and covers up this decline in the

coding category (nC = 6). 7) The share of advertisements focusing on the dietary credentials of products did not significantly change across the time episodes. 8) Descriptions concerning sensory aspects did not change significantly either.

(Qa2) More than half of advertisements (53%; see also Table 5.6) relate to other products or product groups both explicitly (e.g. “vegetables disguised as sausages”, e.g. 2014-22) and implicitly (e.g. “all taste. . . no sacrifice”, 2017-13). The remainder does not use comparisons in their advertising (19%) or make absolute statements about the advertised products whereby the reference remains unclear (e.g. “. . . eat the Healthy Way”, 2016-07; “Easy to cook and taste great!”, e.g. 2011-16; 28%). 5% of adverts proclaim a state that is now better than before–largely in relation to their own range but sometimes compared to the competition–thanks to the introduced product; this covers for example “an eggciting announcement” (2005-01) about Quorn gaining Vegetarian Society approval after switching their full range to free-range eggs, as well as a variety of other announcements about for example changed recipes. Another 4% of adverts compare their products to a general food culture, which covers for example an advert headlined “best of British barbecue” (2014-19), or referring to the common use of allergens in most food products “at last something everyone can enjoy!” (2017-09) Of all comparisons to more specific foodstuffs, those related to meat are the most common (25%). Most of these adverts represent the products explicitly as an alternative to meat (10%), or at least as like meat or equivalent to meat (e.g. “designed to be the ideal substitute for meat in any healthy diet”, 1992- 01; 9%). Taking the opposite approach, 7% suggest that their products are different from meat, proclaiming for example that a product “leaves meat standing” (1987-01) or that it has “90% less fat . . . than normal sausages” (e.g. 2013-24). Not relating to meat but instead to meat alternatives and the immediate market competition are 13% of adverts. The larger share of these (7% of total) relates to some diffuse competition with which meat alternatives are likely to be meant but not explicitly referred to, such

123 5. Print Advertisements for Meat Alternatives

80.0%

60.0% worldly context use and function 40.0% vegetarian / vegan food ingredients and health focus on the format sensory aspects 20.0%

0.0% 1985−2001 2002−'11 2012−'15 2016−'17

Figure 5.3: Development of the use of the descriptions in advertisements across the time periods. as by hinting at the context such as: “There’s a new way to make a meal of tofu, and we’re loving it!” (2011-17), or as framed through an award (“multi-award winning vegan cheese and meat-free alternatives”, 2015-07). The remaining 5% explicitly relate to meat alternatives in statements such as “the healthier alternative to meat” (2009- 03), the “first Fairtrade product to launch in meat-free” (2009-01), or “Psst! If you’re sick of the same old vegetarian sausages, read on . . .” (2013-25). Lastly, 7% of all adverts relate to vegetarian food and 1% to vegetables. Those adverts relating to vegetarian food exclusively point out differences (e.g. “Who ever thought vegetarian food could be that tasty?”, 2010-15), whereas half of the few adverts relating to vegetables point out similarities (e.g. “Vegetables disguised as sausages — Stealthy Healthy!”, e.g. 2014- 22). Because of low code counts (N << 31), no conclusions can be drawn about the changes in advertisements about food culture and adverts suggesting a state better than before; summarised in a category ‘other’, they have not changed significantly (see Table 5.6). Looking at the remaining results, the following observations can be made:

9) Adverts pointing out differences or similarities to vegetarian food and vegetables have significantly declined over time, with the highest prominence in 1985–2001 (17% of all advertisements). 10) Adverts referencing meat have not significantly changed throughout the time periods, and neither have those not making references or value statements at all.

124 5.2. (Qa) Longitudinal Analysis (1985 to 2017)

Table 5.6: Overview of the different distinctions drawn between products (Qa2).

Distinction 2 from nC χ p Examples absolute 154 26.3 0.000 “delicious vegan food” (2012-03); “If you’re looking to liven up your meal times, you can’t go wrong with our new VegeBangers.” (1993-01); “Give your family 100% tasties with 0% nasties” (2015-16); “these taste incredible” (2015-10) meat 137 1.5 0.675 “this delicious, sustainable alternative” (2011-10); “dust off the BBQ and enjoy all your favourites the Fry’s way” (2016-17); “the fabulous frozen meat-free selection” (2010-12); “discover BBQ meatfree” (2010-10) meat 70 35.0 0.000 “winner of the best vegan sausage in the UK” (2016-14); alternatives “meet the new king of mock’n’roll” (2015-33); “first fairtrade product to launch in meat-free” (2009-01); “probably the best vegan quarter pounder money can buy” (2009-10) vegetarian 45 11.1 0.011 “The choice for vegetarians!” (2010-02); “take a fresh look food and at vegetables” (1989-03); “which is exactly what you’d vegetables expect from . . . veg, pulses, herbs & spices” (2016-26); “vegetables disguised as sausages” (2014-22) others 45 4.1 0.253 “As customers become more and more discerning . . .” (2000-03); “Roast Dinners are not just for Sundays.. they’re for Everyday!” (2015-09); “an eggciting announcement” (2005-01); “can brighten up any barbecue bash or afternoon picnic” (2016-12) none 106 1.3 0.722 “good food swoops in” (2017-12); “4 new products for 2017” (2017-05); “food that speaks for itself” (2014-12)

11) References to meat alternatives changed significantly over time, being most prominent in 2002–2011 with 23% of all adverts, while otherwise not having much prominence (<9%). 12) Those advertisements using absolute statements have increased since 2002–2011, representing 41% of all advertisements in 2016–2017.

125 5. Print Advertisements for Meat Alternatives

Table 5.7: Advertised novelty in advertisements across time (Qa3).

new product others none

1985–2001 12 19 16 2002–2011 70 51 101 2012–2015 61 37 99 2016–2017 22 6 63 Total 165 113 279

(Qa3) As summarised in Table 5.7, half of all advertisements (50%) do not report any novelty, and almost a third (30%) advertise a product they suggest being new. Of the remainder, 6% of adverts highlight new or unprecedented taste, such as by illustrating the use case of products as part of meals (e.g. “helping you make delicious meat-free meals”, 2012-15), or a difference in taste to previous solutions (e.g. “They’re our best ever bangers, not only beating our old range in taste tests, but outperforming some other vegetarian sausages on the market too!”, 2011-12). Other adverts point out a way of leading societal change through the consumption of the products (6%), which largely focus on fairtrade, animal rights/welfare, or the unsustainability of fishing. 3% of adverts suggest that unlike previous solutions they could increase the social acceptance of the own diet or of vegetarian products for others, illustrated by headlines such as “How to please everybody at meal times” (2000-01) or “New Tricks — Redwood’s latest products are about to open even more doors for veggies and vegans” (2011-06). Below 10 adverts each are mentions of online shops for meat alternatives or by meat alternative brands, previously not achieved naturalness of products, novel health credentials, or the possibility to ‘help’ animals through consuming the advertised products. Given the distribution of the three groups across the time periods, the following observations can be made:

2 13) Fewer adverts have focused on novel features of products (χ (nC = 113) = 23.6, p < 0.001), with the share of these adverts dropping from 40% in the first to 7% in the last time period. 14) Advertisements suggesting taste has been improved have not significantly 2 changed over time (χ (nC = 33) = 3.5, p > 0.1). 15) Advertisements focusing on new products have not significantly changed 2 (χ (nC = 165) = 2.2, p > 0.1; which means that adverts which do not advertise 2 any novelty have increased continuously: χ (nC = 279) = 20.1, p < 0.01).

Ascribing one overall theme to each advert, 24 different themes emerged, plus a few minor themes which I clustered as other (3%) and none (1%). The remaining codes were grouped into six different categories (see Table 5.8), which related

126 5.2. (Qa) Longitudinal Analysis (1985 to 2017)

to product characteristics, commercial ideas, regular use, special occasions, values, or people. The most prominent themes are those constituted by product characteristics or descriptions (38%). These themes of characteristics cover in particular taste (17%), suggested ‘naturalness’ of products (8%), health (5%), and themes relating to the dietary profile of the product (e.g. vegan or vegetarian; 2%). 22% of adverts have themes related to the commercial environment of products, such as advertising focusing on online shops (2%), won awards (4%), current offers (2%), or new product introductions (15%). Almost a third of adverts employ themes related to the suitability or applicability of products in the lives of its audiences: On the one hand, these are themes related to the products suitability for regular use (16%), which include among others seasonal uses without a particular purpose (e.g. “winter warmers”, e.g. 2003-10; or “enjoy the taste of spring”, 2013-07; 5%), recipes (4%), product convenience (3%), everyday use of the product such as in sandwiches or for breakfast (3%), or school lunches (<1%). On the other hand, special occasions are a theme in 13% of adverts, focusing most prominently on barbecues (5%), christmas (3%), other celebrations such as parties, Sundays (i.e. roast), or halloween (3%), picnics (1%), and a suggested equivalence between the advertised product and a fine dining experience (e.g. “so tasty, they make eating in like dining out”, 2005-04; 0%). Values are used as a main theme for 8% of advertisements, which covers specifically animal rights and welfare (5%), food safety such as in relation to allergens or antibiotics (2%), and global stewardship such as fair trade products (2%). Lastly, themes surrounding people are covered in 5% of adverts, whereby 3% focus on the self (e.g. “new year, new you”, 2015-05; “for the discerning”, 2000-03); and themes concerning family or friends are emphasised in 2% of adverts, for example as “a tasty and convenient meal for all the family” (1985-02) or “find out why Tivall is a top choice for busy mums” (2010-12). The development of regular use themes and of characteristics-related themes oscillated over time and were thus excluded from further analysis. Apart from this, the following observations can be made (see also Table 5.8 for detailed values):

16) Advertisements with themes focusing on values such as related to environment or animals have steadily decreased since the first time period. 17) Similarly, advertisements with people-oriented themes have likely decreased since the first time period (note though that there are a less than the suggested 2 minimum of 31 coded advertisements: χ (nC = 28) = 24.7, p < 0.001). 18) Advertisements focusing on special occasions have increased throughout time periods, with the strongest difference observed between 2002–2011 and 2012–2015. This is highly related to more advertisements on barbecuing and 2 picnics (χ (nC = 37) = 30.4, p < 0.001), which are increasingly prominent after 2012. Conversely, special occasions, such as christmas, fine dining, and other

127 5. Print Advertisements for Meat Alternatives

Table 5.8: Overview of different themes presented in the advertisements (Qa3).

2 Theme nC χ p Examples

characteristics 173 30.4 0.000 “100% vegetarian, 200% taste” (2011-04); “healthy eating for vegetarians” (1991-06); “100% natural ingredients” (2010-18) commercial 125 10.4 0.016 “voted best food product in The Vegan Society Awards 2008 & 2009” (2010-11); “Vegusto has a new website” (2013-16); “March special offers” (2004-01); “just launched 3 new products” (2014-11) regular use 90 11.0 0.011 “winter warmers” (2012-20); “what will you use in your sandwich?” (2015-17); “everyday food, fit for every occasion. . .” (2014-25); “for healthier eating, take a leaf out of our [cooking] book” (2010-14) special 70 23.1 0.000 “barbecue with Boss Burgers this summer” (2017-13); occasions “of all the surprises under the tree that year, none brought more amazement than the Tofurky” (2002-08); “a world of ideas for festive food and party treats” (2013-28); “ideal for picnics, lunchboxes & outdoor eating” (2014-21); “so tasty, they make eating in like dining out” (2005-04) values 46 12.3 0.006 “make it a fish free Friday!” (2017-19); “free from 14 allergens” (2017-09); “the ingredients are sourced from small farmers to whom we guarantee a fair return and are grown exclusively on organic land” (2009-12); “naturally meat & dairy free & now fairtrade!” (2011-21) people 28 24.7 0.000 “New year, new you” (2015-05); “how to please everybody at meal times” (2000-01); “food for sharing” (2015-15) other 25 4.3 0.232 “10,000 people are trying vegan this January — dare you?” (2017-04); “leave it to the Vegenistas!” (2015-06); “self-catering holiday? . . . pack it in” (1990-02)

celebrations have not significantly changed. Adverts focusing on products as being suitable for regular use peaked in 2012–2015, but oscillated throughout time periods. 19) Those adverts with a commercial theme peaked in 2012–2015, while they considerably dropped afterwards. As part of this commercial theme, adverts about 2 new products have not changed significantly (χ (nC = 82) = 3.3, p > 0.1,

128 5.3. Contemporary Advertisements (2016 to 2017)

see also the same observation concerning advertised novelty above), leaving responsibility for the peak with numerous advertisements for awards and online shops from 2002 to 2015.

5.3 Contemporary Advertisements (2016 to 2017)

5.3.1 (Qb) What Are Meat Alternatives? Promoting New Foodstuffs (Qb1) Overall, two-thirds of the analysed adverts in the 2016–2017 time period feature a large scale picture in which they display a product that appears like meat to me; the remainder picture things that either does not look like meat or they only display product packaging instead, which is not analysed here. In 15% of all adverts this pho- tograph acts as a main description of the product; any text is small print, if at all. In these adverts, the meat-like depiction of the products becomes the primary description. In Table 5.9 all descriptions found in the 91 advertisements are presented alongside some examples. The most common textual description for products is related to taste (56%); while most adverts only mention buzzwords such as ‘taste’ or ‘delicious’, a small share of them suggests that the advertised products taste like meat (10% of total). The only other reference to sensory characteristics that might be related to meat is one advert referring to a ‘succulent’ falafel (2016-16). Although all products are printed in magazines for a vegan and/or vegetarian audience, almost half of all adverts (43%) prominently describe that the advertised products conform to such diets, with another six adverts claiming to be plant-based. A fifth of advertisements mention or list some of the product’s ingredients, and a similar amount of adverts mention what is not part of the product or is included less than in comparable products, such as cholesterol, unsaturated fat, or GMOs; 14% of adverts do both illustrate what is not included and give some insights into ingredients that are included. More than a third of adverts describe the advertised product as protein, as containing protein, or as ‘high in protein’, which however is very concentrated on the Fry’s and the More than Meat brands. 13% of adverts make health statements or touch upon health topics in their descriptions, either shallowly with statements such as “good for you” (e.g. 2016-08), but also by mentioning vitamins or other nutrients. Another 23% of advertised products are described as natural through statements like “with only natural flavourings” (2017-14). Ten adverts describe the product as convenient, three adverts describe their product in somewhat traditional contexts (e.g. “being meat-free shouldn’t mean compromising on great festive food”, 2016-30), one adverts describes the product as authentic (after describing taste and in-

129 5. Print Advertisements for Meat Alternatives

Table 5.9: Overview of descriptions used in recent advertisements (Qb1). An advert could be coded with up to four descriptions, see also Table 5.5 for an aggregation of the same code in the longitudinal sample.

Description n Examples

taste 51 “just great tasting food” (2017-16); “bursting with flavour” (2017-06); “ludicrously tasty” (2016-13); “100% taste” (2016-17) dietary 39 “vegan summertime sensations” (2016-12); “whether you’re vegan or vegetarian” (2016-30); “all 100% vegan” (2017-14) plant-based 6 “plants on fire” (2016-17); “new plant-based recipes” (2016-16) ingredients 19 “soy & quinoa country roast” (2016-31); “great-tasting veg, pulses, herbs, and spices” (2017-02); “a blend of . . . wholefoods” (2016-25) exclusion 18 “free from 14 allergens” (2017-06); “fish-free” (2017-19); “no GMOs, no hydrogenated fats, no hydrogenated oils” (2016-04) health 12 “with our healthy Chilli Non Carne” (2016-04); “eat the healthy way” (2016-07); “designed to be delicious and healthy” (2016-08); “A source of protein — high in B12 — low fat — low salt” (2016-25) protein 31 “plant proteins” (2016-25); “high in protein” (e.g. 2016-16); “nature’s plant proteins” (e.g. 2016-27) natural 21 “with only natural flavourings” (2017-14); “with nothing added and nothing taken away” (2017-02); “no artificial additives” (2016-14) convenient 10 “plant-based eating has never been easier” (2017-03); “fit for every or versatile occasion” (2014-25); “are remarkably versatile” (2016-13) traditional 3 “a new take on a classic” (2017-18); “being meat-free shouldn’t mean compromising on great festive food” (2016-30) authentic 1 “. . . and bursting with taste. No compromises!” (2016-20) sustainable 1 “make a meat-free diet a . . . sustainable proposition” (2016-04) ‘British’ 5 “British manufactured” (e.g. 2017-16; 2016-09) visual 14 — gredients, the adverts proclaims “no compromises!”, 2016-20) and one as sustainable. Five adverts, all by VBites, prominently feature a flag or tag with the Union Jack and a “British manufactured” statement next to it (e.g. 2016-09). (Qb2) Almost half of adverts suggest that the products ought to be evaluated either as meat or very similar to meat (31%), or as a meaty food format (14%) such as sausages or bacon. Six adverts contextualise the advertised products as vegetable or wholefood derivatives (e.g. “Try these delicious new additions to our vegetable fam-

130 5.3. Contemporary Advertisements (2016 to 2017)

ily”, 2015-337). Ten adverts primarily suggest an evaluation as a generic food product without creating strong ties to any food category, for example by stating that “Good food swoops in!” (2017-02). An evaluation primarily as vegan food is suggested by seven adverts, including for example mentions of a won “best vegan sausage” (e.g. 2016- 14) award, or an advert with a big “100% VEGAN” stamp (2016-07) on it. Five adverts suggest an evaluation in context, for example for a celebration or festivities such as for “a festive meat-free feast” (2016-30). Almost a quarter of the adverts, which are exclusively Linda McCartney’s adverts (24%), is not focused on promoting particular products, but instead focuses on the brand. With a large scale meat-like picture in the background and the slogan “Linda McCartney’s — deliciously vegetarian” (e.g. 2016- 23) or “25 years of a food pioneer” (e.g. 2016-32), these highlight choice, taste, and being a trusted brand. (Qb3) The advertised products were suggested as part of a variety of categorisations and labels such as diets, formats, purpose, protein, and meat-free. Dietary labels (23%) cover a number of Linda McCartney’s brand advertising, in which the main slogan reads “deliciously vegetarian” (e.g. 2016-23), or a few Goodlife adverts for their “vegtastic new range” (2015-33). 15% of advertisements primarily focus on purposive food, which is constituted largely by barbecue-related advertisements (e.g. “Set your plants on fire!”, 2016-16), with two further adverts for snack food and one for sandwich toppings; although they also usually mention the product format, the purpose sticks out. Format categorisations (25%) cover burgers, sausages, and roasts; 12% of adverts each categorise the products as meat-free (or fish-free) or as protein product respectively. The protein adverts exclusively cover Fry’s adverts under the headline of “Nature’s Plant Proteins” (e.g. 2016-01). In the meat-free and fish-free adverts which are all but one printed by VBites, emphasis is placed on equivalence of use and normality, such as in their “Chili Non Carne” (2017-10) or their “make it a fish-free Fri- day!” (2017-19) advert. Another eleven adverts (12%) suggest a product categorisation as a foodstuff, such as Gosh¡s “good food swoops in!” (2017-02) advert for their new range of products “free from 14 allergens,” or just wholefoods’ “everyday food, fit for every occasion” (2014-25, reprinted in 2016) advert for vegetarian burger and banger mixes.

7Instances of pre-2016 indices mentioned in these sections have been reprinted in 2016 or 2017; bibliographic details of these reprints can be found in Appendix B.5.

131 5. Print Advertisements for Meat Alternatives

5.3.2 (Qc) Is there Meat in a ‘Chilli Non Carne’? Suggesting Alternatives and Re-inventing Meat (Qc2) Firstly, to investigate the frames used for meat alternatives and meat, the themes of the advertisements as coded for research question Qa3 are recapitulated. Being indicative of ‘dominant meanings’, the majority of adverts is either concerned with what the product is or how it can be used: Specifically, almost a half of adverts focus on characteristics (taste: n=27, health and naturalness: n=15), and a third concerns the use of context of the foodstuffs (at special occasions like barbecues or christmas: n=20, regularly or in everyday life: n=13). The remainder of advertisements emphasises the brand (n=5), new products (n=8), or less frequently awards (n=2) or allergens (n=1). Similar patterns can be observed concerning how meat is understood or referenced in the advertisements. Of all adverts a bit less than half frame or refer to meat in writing (46%), while an additional 22% evoke associations with meat through their imagery. This leaves just about a third (32%) of advertisements not referring to meat beyond the specific food format (e.g. a burger or a sausage). In the remaining advertisements, meat is largely represented as something normal, habitual, traditional or otherwise common (37% of total), for example through the context that is established in dishes, labels, or events (see Table 5.10 for examples). Another seven adverts relate to meat as a hedonic item by referring to an enjoyable taste or experience, and only one advert published by Fry’s utilised a rather critical framing of meat, as they refer to their product as “a non GM product with no hormones, antibiotics, or chemicals” (2016-10). (Qc1) Extending the understandings put forward of meat, this paragraph investigates what kind of relation advertisements put forward concerning the frames for meat. For those 43% of adverts which relate to meat (excluding recipe advertisements) broadly three different relationship types were identified (see also Table 5.11): (1) The advertised products are as established and normal as meat; (2) the advertised products are an alternative to meat that is largely unchanged; or (3) the advertised products are a ‘new version’ of an established foodstuff which happens to be meat. Firstly, being as established as meat is covered in 12 adverts, which includes largely VBites adverts (and two by just wholefoods) in which the products are framed as the most normal thing in the world. Examples for this are an advert for meat-free slices, which is titled “Happiness is . . .an absurdly delicious vegan sandwich” (2016-22) and surrounded by seven pictures that display these meat-free slices in buns, bagels, bread rolls, or sandwiches. Yet without the captions for turkey-style slices or beef-style slices or knowledge about the brand, the audience would be unable to distinguish the images from meat-based sandwiches. This advert, similar to the other adverts in this category, equates “delicious” vegan food with meat facsimiles.

132 5.3. Contemporary Advertisements (2016 to 2017)

Table 5.10: Overview of the identified frames for meat identified in the recently printed advertisements (Qc2).

Frame of Meat n Examples

normal, 34 labels such as “More than Jerk Burgers” (2016-25); contextual such habitual, or as “get set for a tasty vegan christmas” (2016-30); “make it a fish-free traditional Friday” (2017-19); “a new take on a classic” (2017-18); dishes or recipes such as “Chilli Non Carne” (2017-10); “fish-style fingers sandwich” (2016-09); “chicken style garlic salad” (2016-07); hedonic (taste) 7 “all taste. . . no sacrifice” (2017-13); “enjoy all your [barbecue] favourites the Fry’s way” (e.g. 2016-17) critical of meat 1 “a non GM product with no hormones, antibiotics, or chemicals” (2016-10) visual only 20 — none / format 29 “delicious sausages and falafel” (2015-01); “Nature’s Plant Proteins” only (e.g. 2016-01)

Secondly, 16 adverts present the advertised product as an alternative to meat: not quite alike, but neither too dissimilar. These cover Fry’s and VBites advertisements, which at large emphasise that meat can be legitimately replaced with something else. For example, Fry’s points out a different way of barbecuing by asking consumers to “dust off the BBQ and enjoy all your favourites the Fry’sway” (e.g. 2016-18), and VBites describes their company in a number of adverts as “producing delicious, tasty, creative alternatives to meat, fish, dairy & egg” (e.g. 2016-22). Thirdly, in 11 adverts the products are presented as a new version of an established foodstuff; they reference meat through naming, format, or convenience aspects, but otherwise highlight the advertised product’s uniqueness which remains in context of the meat product. For example, Gosh! describes their burger as “a new take on a classic” (2017-18) whereby they emphasise the traditional aspect of the burger format yet strip it of the associated expectations towards taste or texture. Keeping taste associations, Moodley Manor advertises their product range with a subtitle hinting at animal cruelty by suggesting: “all taste. . . no sacrifice” (2017-13). (Qc) Summarising these insights, it can be noted that at large the reasons for why people would reduce meat, or why they might need to be motivated to eat meat alternatives are not featured prominently in the advertisements: While they may invariably draw on awareness for these issues amongst their audiences, only in rare cases are such reasons hinted at. In the only advertisement which was identified to frame meat as an issue, audiences nonetheless need to connect ‘hormones’ or ‘antibiotics’ with animal husbandry and thus meat consumption. Here, the advertised products become

133 5. Print Advertisements for Meat Alternatives

Table 5.11: Overview of how advertisements position meat alternatives in relation to meat (Qc1).

Meat Alternatives as. . . n Examples

any 39 (43%) similarly established 12 “backyard BBQ — celebrate summer with awesome food” (2017-15); “10,000 are trying vegan this January — dare you?” (2017-04); “fish style fingers sandwich” (2016-09) alternative to meat 16 “dust off the BBQ and enjoy all your favourites the Fry’s way” (e.g. 2017-17); “a delicious and healthy alternative to ‘normal’ meat-based foods to everyone, regardless of their choice of diet” (2016-04); “Soy and Quinoa Country Roast is the perfect centrepiece for the family table” (2016-30) reinvented meat 11 “a new take on a classic” (2017-18); “all taste. . . no sacrifice” (2017-13); “Fry’s will be inspiring meat-free grills up and down the country” (2016-16) a resolution to the advertised tensions. Many adverts instead highlight a normality in which meat alternatives can become as normal as meat, and may be used in occasions and regular use. While this normality largely focuses on dishes, meals, and contexts, a number of adverts highlight that meat alternatives are pleasurable or healthy. Not- ably, meat alternatives can not only be as pleasurable as meat, but serving different consumer interests they can also be pleasurable in their own right as a re-imagination of meat. Even though they may not be exactly alike, they claim to draw on the same cultural resources, and highlight issues such as taste, health, or naturalness. Table 5.12 summarises these three frames of impacts, normality, and pleasure which were identified in this section. Table 5.12: Summary of the frames observed for meat and meat alternatives in the recently printed advertisements (Qc). Statements in brackets are not made explicit in the advertisements.

Frame Meat as. . . Diagnostic Frame Prognostic Frame Motivational Frame

impacts harmful some foods have meat alternatives can (meat should not be but negative impacts alleviate any issues eaten) replaceable normality entrenched (meat is normal) meat alternatives are (people may not normal, or can be as want to eat meat) normal

134 5.4. Conclusion

Frame Meat as. . . Diagnostic Frame Prognostic Frame Motivational Frame pleasure tasty, (meat is tasty) meat alternatives (people may not pleasurable taste identical, or want to eat meat) tasty in their own right

5.4 Conclusion

In this chapter, the collected advertisements from three different vegan and vegetarian magazines were reviewed. Firstly, they were reviewed longitudinally to investigate category emergence. Secondly, a subset of the more recent advertisements was reviewed to study category identity (Qb) and frames (Qc). Concerning the emergence of the meat alternatives category (Qa), it was found that the prominence of values declined throughout the time periods, as focus on health and taste remained largely stable. More recently, there has been an increasing number of adverts focusing on protein and plant-based credentials, as well as on barbecuing. The current category identity (Qb) as represented in the advertisements focuses on taste, health, ‘naturalness’, and food formats. In particular the brand Fry’salso has a strong focus on protein, while the brand Linda McCartney’s appears to be widely enough known that they solely advertise the brand. In the studied advertisements, meat alternatives display a complex relationship to meat (Qc): Overall focusing on taste, health, consumption occasions, and everyday practices–meat alternatives were portrayed as products just as established as meat, and in exactly the same context; as legitimate replacements to meat; or as a new version of a traditional foodstuff that builds on meat but is framed as detached from it. From the findings of the longitudinal analysis, it appears that links to meat consumption persist, both linguistically and structurally (as observed in format and context). As health characteristics and continuously improving taste were advertised similarly throughout all time periods, it may be suggested that the category has a certain set of fixed connotations. The disappearance of most criticisms of meat consumption from advertisements may be one of the most prominent findings, which might be related to reducing ‘friction’ to appeal to mainstream consumers (see again section 2.5). Conversely, it appears plausible that the increased contextualisation of products in specific occasions and everyday life highlights an ongoing struggle for legitimacy: Al- though meat has been argued in earlier adverts to be reduced, audiences may need to be convinced regularly that meat alternatives are suitable, tasty, and equivalent substitutes. Any of these aspects will be further elaborated and expanded in section 9.2, where the emergence of the category is analysed in detail.

135 5. Print Advertisements for Meat Alternatives

In the recent adverts, focus on the issues of meat consumption disappeared almost completely. However, while many adverts suggest that products conform to the standards set by meat, a diverse range of aspects are highlighted: Many advertisements suggest normality through the consumption of the products, and a few cover a similarity in taste. A substantive number of adverts however also highlight that sensory experiences do not need to match completely. These adverts build on previous experiences or expectations with a foodstuff, but adapt the cultural and symbolic underpinnings to different contexts. Although they remain culturally consistent through the format and hedonic goods because otherwise they would not sell, some meat alternatives may therefore transcend the tastes and textures commonly associated with meat. However, most adverts seem to represent ‘normal’ foodstuffs which are meat-like, tasty, healthy, and fit for every occasion.

136 Chapter 6

Expert Interviews with Meat Alternatives Stakeholders

For this research project, semi-structured expert interviews have been conducted as a second producer-originating data set besides the advertisements. The interviews can provide additional context and in depth accounts of the issues at stake in this research project. In this chapter, the interviews are at first briefly described, and then the findings concerning the category identity of meat alternatives (Qb), and frames for meat alternatives and meat (Qc) are presented. The interviews were analysed using an open inductive content analysis and frame analysis (see again section 3.3.3). In the remainder of the chapter, firstly the interviews are described in general and each interview is briefly summarised. Thereafter, the findings to the two research questions on category identity (Qb) and frames (Qc) are presented in one section each. Finally, the chapter is summarised and concluded.

6.1 Description of the Data

Building on the sampling and the interview process as described in section 3.3.3, the analysis is based on eleven interviews with a total recording time of 9:17 hours, as individual interviews lasted between 16 and 122 minutes (median 42 minutes). Of these interviews seven are with UK-based manufacturers, two manufacturers are based abroad, and two represent civil society organisations. Two of these interviews were conducted with Quorn employees in different roles, therefore each providing unique perspectives. An overview of the interviewees and their organisations can be found in Appendix C.2; each interview is also briefly described later in this section. The sample can be considered a cross-section of organisations in the market in late 2017, while it does not represent the latest trends as identified in chapter 4. More smaller than larger companies are part of the sample, possibly as they are more willing to engage in external activities, and because there are more of them. Of the eight companies in the sample, five have at least one advertisement in the overall sample from 1985–2017, but only one of these companies is covered in the most recent sample

137 Expert Interviews with Stakeholders

(see again section 5.1); one further company had adverts printed, but these were not included in the earlier sample as they were smaller than a quarter page. At the time of interviewing, products of at least three brands could be bought in at least one major supermarket in the UK, and two further brands were available in health foods shops nationwide; one brand was only regionally available at the time of the interview, one was only available online, and a third one had discontinued their production short before the interview was conducted. In the following paragraphs each represented organisation will be introduced briefly in alphabetic order, and the main arguments summarised. The eating better alliance is a civil society organisation that was formed as an umbrella organisation by UK organisations involved in sustainable diets. Interviewee Sue discussed changing diets, the role that plant-based (protein) products play towards more sustainable diets, and the ‘less and better’1 approach to meat consumption which the organisation advocates. LoveSeitan is a Greater London-based seitan manufacturer which was founded in 2017. Their aim is to make seitan a prominent and well-known meat alternative similar to tofu or tempeh. Interviewee Steve discussed seitan as a food product and as a meat alternative, scaling as a start up, and food activism. Since the interview, the brand has started a partnership with manufacturer VBites and has released a number of new convenience products. Meat the Alternative is a Wales-based manufacturer for meat alternatives founded in 2012 by a former butcher. Interviewee Tony pointed out the importance of high-quality foods with textures differing by product type: They2 discussed market trends and opportunities related to meat reducers in the market, as well as the difficulties of market entrance in the UK. More than Meat was a small manufacturer for wholefood meat alternatives based in Greater London, that stopped business in September 2017. Interviewee Daniel particularly focused on the role of and the difficulty for small enterprises to establish themselves, as well as about health and ingredient quality in meat alternatives. PETA (People for the Ethical Treatment of Animals) is an international animal rights organisation that advocates against animal husbandry and promotes animal- free food practices. Interviewee Dawn works with food services and among others pointed out the importance of vegan choices and what role meat alternatives can play in there.

1In this and other interviews, as well as in a report from May 2017 (Eating Better, 2017), this was referred to as the ‘less but better’ approach. As a report released in April 2018 (after most interviews were conducted) terms this ‘less and better’ (Eating Better, 2018), and will be used in this thesis instead. 2Throughout this thesis I use they instead of gender-specific pronouns, because the preferred pronoun of interviewees or YouTubers was not discussed or assessed.

138 6.2. (Qb) What Are We Selling? Identifying Meat Alternatives

Quorn is the UK market leader for meat-free products and manufactures fungi- based products; it was founded in 1985 and is the biggest meat alternatives company internationally. Interviewee Louise is Sustainability Manager and discussed the role of sustainability and certification as an opportunity for meat alternatives. The other interview was conducted with Insight Manager Eddy, particularly focusing on market and consumer data, Quorn’s market position, category management, and an overarching category strategy for meat alternatives. Sgaia Foods is a Scotland-based manufacturer for vegan meat and cheese alternatives that was founded in 2015 and specialises in meat alternatives to traditional Italian foods. Interviewee Hilary particularly discussed the role of food as culture and identity, and the necessity for animals in food and food names. Tofurky is an Oregon-based family owned manufacturer for meat alternatives that originated from a small tempeh business in the 1980s and also sells in the UK. Interviewee Seth pointed out the role of values for companies in the meat alternatives business, as well as the merits of being an iconic brand and selling internationally. Vegetables in Disguise is a Scotland-based manufacturer for vegetable-based sausages under their Secret Sausages branding that was established in 2013. Interviewee Rachel discussed the role of meat-like flavour, allergens, and texture in meat alternatives, the diversity of meat alternatives, and purposive foods. Since the interview, they have partnered with an external food processor. Vivera was founded in 1990 and is a Dutch manufacturer for mostly vegan meat alternatives; today, it is part of the Enkco Foodgroup and trading in the UK since 2016, as well as in many other markets in Europe. Interviewee Geert-Jan discussed being an international manufacturer for meat alternatives, market entrance to the UK, global challenges, and the relation of meat and meat alternatives now and in the future. Since the interview, Vivera had more successful product launches on the UK market.

6.2 (Qb) What Are We Selling? Identifying Meat Alternatives

(Qb1) Meat alternatives are characterised by my interviewees at large by the use of sensory aspects, their suitability to match established practices, formats, health, environment and ethics, and product pricing. Two put a ranking for the characteristics forward: Seth suggests that taste, texture, and value are the most important aspects of foodstuffs, whereas Tony highlights the importance of taste, health, and environmental aspects in this order. Because they were mentioned infrequently, the following findings do not discuss characteristics such as dietary product credentials, conveni-

139 Expert Interviews with Stakeholders ence, product preparation, product origin, food safety, allergens, and availability. In Table 6.1, an overview is presented and discussed in more detail in the following paragraphs. Sensory characteristics are the most prominent characteristics discussed by my interviewees. They mostly discuss taste and texture, but some also mention appearance and smell. Particularly taste but also texture is understood as something vital for meat alternatives, as Eddy and Tony both understand taste and texture as mutually dependent. But across the interviews, different opinions on what constitutes good taste or texture, and how to describe it were voiced. On the one hand, there are those manufacturers whose idea it is to make products as meat like as possible, such as Tony and Geert-Jan. For example, Geert-Jan defines ‘good’ meat alternatives as those that are “good in terms of taste, mouthfeeling, structure, . . . and appearance of meat.” Tony, who suggests that their products should be stocked next to meat, finds that most other products available are only insufficiently resembling meat, suggesting among others that “you buy a piece of Quorn chicken, you buy a piece of Quorn beef, it’s the same texture.” On the other hand, Hilary suggests that the best descriptions for foodstuffs are not referring to meat but use “the language of flavour and texture.” Seth tries to put into practice arguing that on their

packaging, we’re using savoury, and . . . zesty, and, sort of, words, that invoke flavour, because . . . in the end of the day, you know, you got to have that, in order to have repeat sales.” (Seth)

Steve and Rachel focus on a comparable experience to meat. For example, for Rachel it is important to assert that their sausages have a different texture and flavour from meat; the similar experience of their sausages however does not only relate to the format, but also to their sausage casing

called nutracasing which is made of rice and seaweed, and it basically holds the product together as it cooks up, and then when you cut it, it obviously gives a little bit of that springiness, like traditional meat sausages with collagen. However the actual sausage lacks the sponginess of the meat, but has a more fresh vegetable interior to–as a person who enjoys vegetables–far more enjoyable. (Rachel)

While this quote emphasises that sensory differences to meat are seen as crucial, some haptic and physical features of meat sausages are framed as positives. And Hilary suggests that all people, who do not know the traditional Italian meat products their meat alternatives are based on, “just like them for what they are, which is also a great positive: Like they’re not comparing it to meat, they’re just liking the product, because

140 6.2. (Qb) What Are We Selling? Identifying Meat Alternatives

Table 6.1: Overview of descriptions used for meat alternatives in the interviews (Qb1).

Characteristic Examples

sensory “taste is king, value is queen, and everything else is marketing” (Seth); properties “replicate . . . the spongy, chewy, texture that meat delivers” (Rachel); “a chicken flavor, which is vegan, we used chicken spices” (Steve); “if I’m gonna make a change from eating meat, I still like the succulence I get from meat, I like the aftertaste, I like the flavour” (Tony); “of the meat alternatives out there it’s the one that best mimics the fibres of meat, in terms of the texture, so it gives it a better appeal, it gives it a better look, and it gives it a better taste” (Eddy) formats and “they’re still producing meals without mince, for example, so they’d say practices they’d produce a spaghetti Bolognese or a lasagne, don’t tell the rest of the family, but they satisfy the vegetarian” (Tony); “we’ve taken all of these traditional recipes, and . . . re-invented it, and applied that to our protein-base” (Hilary); “the fact that it’s a vegan version of what they’re famous for makes it very interesting” (Dawn); “it is very easy to use, compared to Tofu, for example, you can cut it up, and use it in a stir-fry or grill it, or roast it” (Steve) health “compare the quality of our product ingredients, the health benefits, the nutritional statistics with anything else that’s on the market. And our products were by far on the way the best nutritionally” (Daniel); “the most healthy vegan alternatives to the meat products” (Geert-Jan); “health benefits are hugely important to companies at the moment” (Rachel); “we see plant-based is becoming a really big message, associated to health more than anything else” (Tony); “from a nutritional perspective . . . the assumption that meat alternatives are necessarily going to be healthier does need to be looked at on a case-by-case basis” (Sue) environment “because of, you know, ethical reasons, environmental reasons, the fact that and ethics animal products are slowly killing us, in many ways” (Hilary); “good for you in inverted commas, ‘good for you and the planet’ ” (Daniel); “these plant based foods are a lot smaller footprint, and water usage, and land, and all of that” (Seth); “We tend to say it’s just as good as to meat but a lot better for, for people, animals, and that the planet.” (Geert-Jan) product “the price point needs to be in line with the meat alternatives that are pricing typically churned out by the thousands and are very cheap” (Daniel); “be parable with the meat prices, premium meat” (Tony); “one of those sustainability choices, which unusually can save you money” (Sue)

141 Expert Interviews with Stakeholders it’s good.” It may be concluded that each interviewee in their own way builds on past consumer experiences and accounts for arising expectations. This may be even more important when assuming like Steve that “arguably, meat doesn’t really have any flavour, it’s flavoured with what you add to it anyway.” Ideas that meat alternatives need to match established practices, such as in traditions or customs, are frequently mentioned, as well. The commonly used argument suggests that people have established practices, and if they can retain their practices while adjusting or changing their diet, this is a great advantage. Dawn suggests in general that “vegan chicken nuggets, vegan burgers, vegan sausages, all of these things make it [dietary change] easier.” Similarly, both Seth and Hilary emphasise the popularity of their meat-free bacon products, which Hilary suggests is “because people just enjoy the ritual of bacon, you know, having a bacon butty in the morning.” This statement is representative to all my interviews and identifies the product format as fundamentally important to meat alternatives. However, the significance of formats has different interpretations: Daniel, for example, suggests that expectations prior to the eating experience are inevitable leading to disappointment, if sensory properties of meat are not matched closely. Hilary even suggests that manufacturing methods are important as they convey “cultural accuracy” and thus need to be at least approx- imated for an “authentic” experience. Seth similarly points out the ‘historic’ relevance of their meat-free turkey in the 1990s

as a peacemaker in many families, because people will serve Tofurkys and Turkeys on the same table . . . because up until that point, vegans . . . and vegetarians were left out of the celebration (Seth)

and according to Seth had to eat salads and similar inappropriate thanksgiving foods instead. Steve and Rachel, on the other hand, oppose a necessity for accurate copies and point at functional equivalence, as they seek to distinguish the format as a cultural entity from sensory elements. As practices are considered key to meanings of meat alternatives, new food formats adopted from meat are considered the main innovative capacity for their companies by Geert-Jan and Seth; contrarily, Eddy believes that replicating “[what] the meat market [is] doing . . . isn’t a sustainable sort of approach” for the category. According to basically any of my interviewees, health is a major, if not the main reason for people to reduce meat consumption. Underlining this claim, Seth and Tony told me anecdotes about doctors suggesting to their clients that for health reasons they should not eat (too much) meat anymore. That products need to be healthier than meat or the competition is considered a unique selling point. Accordingly, Geert-Jan for example aims at producing “the most healthy vegan alternatives to meat

142 6.2. (Qb) What Are We Selling? Identifying Meat Alternatives

products,” and Steve claims that their products are “a whole lot more healthy” than meat. However, what healthiness means for products remains unclear: Tony sees it in the amino acid composition of the respective crops used as a protein source, as well as in the idea to “maintain a as natural product as possible” by not using additives; on the contrary, Geert-Jan highlights that their company adds vitamin B12 and iron “to make a perfect copy of the nutritional values that are in . . . beef.” And while Rachel brushes over health with a more general claim about “the other benefits that you can have when you’re not eating meat,” Daniel’s primary motivation for starting their business was to encourage people to “really [take] control of your health.” Op- posing a general stigmatisation of meat, Tony argues that “poultry and fish are still classed as . . . a healthier protein” even though “plant-based is . . . associated to health more than anything else.” Distrust towards rather ‘artificial’ or ‘unnatural’ competitor foods is also highlighted, insinuating that these would be unhealthy; for example, two interviewees tell me stories of Quorn’s history in an attempt to make the product appear artificial, which do not match my or the academic literature’s understanding of it. And Tony voiced concerns about the meat industry driving scaremongering about processed foods. Sue is generally more cautious regarding claims about the health benefits of meat alternatives in general, pointing out that “from a nutritional perspective, you know, the assumption that meat alternatives are necessarily going to be healthier does need to be looked at on a case-by-case basis.” Environmental issues and ethical considerations are identified as another important reason why people would reduce their meat consumption. Hilary, for example, finds that meat consumption needs to be curbed “because of, you know, ethical reasons, environmental reasons, the fact that animal products are slowly killing us, in many ways.” Yet, most interviewees neither engage with the causes of ethical and sustainability issues in explicit ways, nor with the validity or extent of their proposed solutions. Instead, they rather remain vague or use generic phrases like “good for you in inverted commas, ‘good for you and the planet’ ” (Daniel). In these arguments, meat alternatives are never argued to be ‘good’ by themselves, but only in contrast to meat. In some cases, environmental issues may be discussed more pragmatically; Louise claimed that they

understood that there was a massive benefit where [we] look up publicly available data for meat [and compare them to meat alternatives]. . . . But . . . we never went externally with much of that data. (Louise)

Similarly, Steve wonders about the trade-offs of ‘eco-friendly’ packaging, weighing up a necessity to add preservatives or shorter shelf life. While everyone could agree with environmental benefits through the substitution of meat with meat alternatives, ethical grounds were more ambiguous. For Hilary and Steve, being a vegan company

143 Expert Interviews with Stakeholders is part of their business ethos, with Steve proclaiming their business to be “food activism.” On the other hand, in the interviews, Tony or Geert-Jan appeared hesitant to highlight the wrongdoings of the meat industry, with Tony suggesting that “I re- cognize the consumers are recognising: they need change.” Reasons for their different attitude might be that their brands explicitly target meat reducers or that they themselves appeared to eat meat. Product pricing is discussed by around half of the interviewees as a relevant factor. One important thread in this debate is the role for price as a gatekeeper for the large retailers, who with their considerable margins create barriers to entry particularly for small businesses. However, even outside of retailers, Daniel points out that adjustment to the “category pricing, [was still necessary] ’cause if we didn’t we wouldn’t have sold them.” Tony points out the price competition of established brands and suggests that this was mainly aimed at driving competition out of the market, but would also reduce consumer willingness to pay for premium products. Conversely, Eddy as a representative of the market leader Quorn highlights the importance of low prices and regular offers as a way to attract new consumers and to drive repeat purchases. (Qb2) As discussed above, food formats like sausages or burgers are strongly associated with specific occasions and rituals; the importance of this for their products is highlighted across all interviews. But some interviewees are not satisfied with this: For example, Hilary wants to “[remove] the concept of animal from food,” Rachel would like to dissociate sausages from meaty textures, and Steve wants to make ‘seitan’ a widely understood foodstuff in an attempt to make an explanation of their product as a meat alternative obsolete. However, they all still utilise or benefit from the foundational understanding of meat in their business. Apart from meat, some references also concern other food categories, such as vegetarian foods, including vegetables, or tofu. To Steve a comparison to tofu may carry its own difficulties and may need explanation, and thus providing no advantage over labelling their food as ‘meat alternative’. In such logic of format-based meat references, single words such as ‘sausage’ do not only evoke consumer expectations tastes and textures, but more importantly carry meanings in terms of use cases and affordances. However, references to products other than meat have a less strong connotation as pointed out above. This is also illustrated by Tony who claims that other brands have a difficulty to attract meat eating consumers because of “the baggage of being a vegetarian brand.” Beyond product comparisons, some interviewees also referred to the immediate competition in the market place which is largely manifested through fixtures defined by the retailers. Comparisons with the competition are then declared on whatever my interviewees think they are best or better in, such as being more ‘natural’, containing

144 6.2. (Qb) What Are We Selling? Identifying Meat Alternatives

fewer additives, being a trusted brand or vegan, having better texture or taste, or offering food formats that competitors cannot offer. (Qb3) Unlike in the consumer data sets, where labelling may be considered a reflective exercise, my interviewees likely follow more strategic aims. And unlike advertising, they had a reasonable environment to express their views and discuss labels in detail. However, at large it needs to be noted that categorisation is an important aspect in the food industry carried out by intermediaries who define supermarket fixtures, delineate competition, and provide a foundational set of references to the audiences. As is further pointed out by some interviewees, also consumers follow their own purchasing routines by following accustomed shopping routes, and show selective attention in supermarket aisles; here, meat eaters allegedly avoid the meat-free shelves, and vegetarians avoid the meat aisle. Regarding category labels, Eddy positions category labels as potential barriers which “can also be very off-putting to someone.” They report that the most common naming for the category where meat alternatives tend to be stocked is ‘meat-free’; occasionally it is also titled ‘vegetarian’ and in Sainsbury’s it is called ‘healthier choices’. When discussing categorisation of their products, opinions differ: Rachel, for example, understands their market segment as a vegetarian category, and Steve would like an independent seitan category for their products. And Tony constructs a dichotomy between a vegetarian and a meat category that their supposedly more meat-like product can bridge:

We’re trying to disrupt what’s going on in the meat category, but we’re also in a way, trying to disrupt what Quorn are doing in the vegetarian category. (Tony)

When asked about an ideal category name instead of meat-free, Eddy believes that ‘protein’ would be a good candidate, which both Tony and Hilary agree with: “You [can] think of it in terms of protein,” Hilary suggested. These claims all use the concept of protein as reason to explain and legitimise the transferability of established food formats through different protein bases. Opposing this, Steve declares protein as a recent food hype, although they acknowledge that outside of the UK the term ‘protein’ may be more common when referring to meat alternatives. My interviewees from established and iconic brands, such as Quorn or Tofurky do not have the endeavour to create a new or different category. This might be because they suggested that the existing category was hugely influenced and defined by their brands. Underlining this power, Eddy reports that Quorn as the market leader engages with retailers in order to develop a category vision, which should guide and steer retailers’ understanding of the category; this is said to include enough opportunities for other entrants as the market is thought to be growing.

145 Expert Interviews with Stakeholders

6.3 (Qc) On Copying, Reducing, and Reinventing Meat

(Qc) As listed in Table 6.2, the manufacturers employ a variety of different frames towards meat alternatives and meat. The main diagnostic frames put forward are concerned with health or animals, or generally the idea that meat consumption is not ‘sustainable’. For some interviewees, this appears implicit to consumer decisions to buy meat alternatives. Beyond this, different issues are identified: Meat alternatives are found unhealthy, not meat-like enough, or too meaty. Thus, there are problems identified with meat, but also within the meat alternatives category. Diagnostic frames cover at large the provision of meat alternative products which aim at alleviating the issues identified in the prognostic frames. This includes products which are suggested to be healthier, meatier, or less meaty. Daniel further suggests more generally, that going meat-free can be a solution to individual health issues, and Hilary suggests that food culture should be reclaimed from meat and applied to vegan foodstuffs. The motivational frames covered by the interviewees vary depending on the acknowledgement of the harm done through eating meat; for example, Tony and Geert-Jan do not suggest a moral imperative to reduce meat generally, but either assume it as given by prospective audiences, or only in the case that vegan and animal-based meat can be considered equivalent. Conversely, Hilary, Seth, and Steve consider it moral imperative to reduce or stop eating animals, which their products can help to achieve. Daniel finds meat-free consumption advantageous but not imperative, and Eddy and Rachel abstain from this debate. Instead, Eddy’s frame can be summarised that products should be bought from a trusted brand, while Rachel merely points at the enjoyable and functional formats which meat items provide, as long as they do not taste like meat. From these various viewpoints, three overarching frames for meat can be identified: At large, meat can be copied, replaced, or reinvented. These frames will be discussed in more detail in the following paragraphs, and are summarised again in Table 6.3. “We just want to be a vegan meat alternative, that copies meat products till the 100% exact copy, and to, to appeal to meat lovers. . . . [And then] why should I still take the meat variant of this one.” (Geert-Jan) In this frame, the focus is on copying meat, as no questions are asked of consumers and no assumptions are made about why people may want to eat a meat-free meal or diet. However, Tony suggests that people with a meat-free diet “accept [that] the compromise is: difference” whereas meat lovers or meat reducers would only change if the meat-free alternative is perfectly equivalent. Thus, more so than in any of the other frames, meat represents

146 6.3. (Qc) On Copying, Reducing, and Reinventing Meat (Qc). 3 take control of your health buy from a trusted brand if an alternative tastes identical, choose the better version food is constitutive to culture; everyone should be vegan compromise going meat-free can re-claim health can help; provide healthier meat alternatives provide a variety of products appealing to meat lovers but also those less fond of it copy” of animals; provide ‘authentic’ vegan equivalents provide a good/better alternativeprovide more ‘meaty’ meat alternatives eating tasty but not harmful food eating meat-free should not mean to Overview of frames identified from the manufacturer interviews Table 6.2: as most meat alternatives are not ‘healthy’ shoppers in the meat-free category meat enough IntervieweeDaniel Diagnostic FrameEddy there is a public health problem; meat as well Geert-Jan there is an increasing diversity of different meatHilary is not future-proof Prognostic Frame meat is harmfulRachelSethSteve existing meat alternatives provide are vegan too versions meaty of meats to a “100% — people struggle to provide find sausages good that alternatives Motivational are Frame to not meaty reclaim food culture in vegan ways, and rid it meat formats are enjoyable and functional provide tasty “protein alternatives” create a world without animal (ab)use Tony existing meat alternatives are not meaty Interviewee Louise has been omitted in this table as no consumer-focused frames could be identified. 3

147 Expert Interviews with Stakeholders a normality whose extraordinary taste is supposed to be met. Just as any of the other options, the food formats promise continuity in everyday life and special occasions.

“For us, every block of seitan we sell, for us, is, you know, potentially a piece of meat that isn’t being purchased or consumed.” (Steve) In this frame, the focus is on replacing meat because it is considered inappropriate for a growing world population, or generally bad for the environment, animals, and health. While all acknowledge these reasons for meat reduction, only few interviewees make them explicit; for example, Hilary, Steve, Daniel, or Seth explicitly declared it their business aim to proactively help people to reduce their meat consumption. Here, meat alternatives are not the goal in itself, but the easiest means to an end, as Steve proclaimed: “we want to be this alternative that’s easy to find and easy to use.” “And those people, whilst they might want to enjoy the occasion of eating a sausage or a burger, from a functional point of view, doing a barbecue, or just enjoying a sausage format, they didn’t want the meat . . .” (Rachel) In this frame, meat is reinvented as it is understood as an important cultural constituent for broad aspects like identity but also for lived everyday experiences and the alleged convenience of meat. As emphasised by most interviewees, meat is something through which consumers know what to expect from meat alternatives, and know how to deal with the product. Seth, for example, illustrates the special role of their meat-free turkey for American thanksgiving and christmas, and Steve highlights that “you can use it [Seitan] exactly like you would [in] a meat dish, and it’s a direct replacement for meat in a lot of dishes.” Notably however, in this frame meat alternatives do not need the same material constituents as it suffices that alternative products are similar enough and can fulfil the same purpose in a dish, and therefore are still able to evoke experiences. Accordingly, Hilary, Steve, and others acknowledge that their products can never be exactly like meat, but they also do not need to be; as Eddy suggests, “people will want something different, and they [companies] have to have some products . . . that cared for people who want something different.”

6.4 Conclusion

This chapter dealt with the analysis of the eleven semi-structured interviews conducted. Concerning the category identity (Qb), it was found that interviewees attribute the most importance to sensory aspects, and use cases which are associated with meat- like formats, but also naturalness and health. Environmental and ethical benefits are often also mentioned or assumed implicit in consumer decisions. Meat as a foodstuff is the most commonly used reference point to explain products, which particularly relates to the product formats replicated in meat alternatives. However, when arguing

148 6.4. Conclusion

Table 6.3: Summary of the frames.

Frame Meat as . . . Meat Alternatives as . . . copying meat tasty normality identical (but better) replacing meat inappropriate and harmful, but the easiest alternative replaceable reinventing meat convenience and culture reclaiming/reinvention of meaty formats why it should be bought, they largely position their products in comparison to other meat alternatives, while strategic aims hint towards a rearrangement of the category towards a protein umbrella which would allow it to be integrated with meat products. The relations to meat (Qc) are highlighted by all interviewees, as they allow adapting taste expectations, dishes, food practices, and traditions conveniently to a meat-free context; yet, the degree to which meat alternatives should resemble meat products is contested among the interviewees. Three frames emerge from the analysis, which highlight meat as a tasty normal foodstuff, meat as something whose consumption ought to be reduced because of its negative impacts, and meat as a convenient cultural concept which can be reinvented. Unlike in other data sets, in the interviews the role of intermediaries is much more prominent: Retailers but also distributors are considered important gatekeepers to audiences, with their power to dictate prices, as well as to define competition in shelves, online shops, or catalogues. Accordingly, interviewees have a stronger sense for categories as the selection of key competitors and evaluation characteristics are a primary means to promote one’s brand. Noteworthy here is a prominent idea that meat as well as meat alternatives ought to be understood as ‘protein’. This allows seeing meatiness across different products and competitors as a design choice, whereby a focus would shift from meat as a foodstuff for comparison to meat as one way to satisfy a nutritional goal. This essentially allows for the cultural backgrounds and concepts of meat to be reinvented. Furthermore, varying degrees of activism and business-focus can be observed as some interviewees trust that their consumers know why they want to reduce meat, whereas others want to strengthen the case for not eating meat anymore. Although these two approaches may arguably target different people, they allow for meat alternatives to ignore the issues surrounding meat entirely and outsource conscience to audiences.

149

Chapter 7

Taste Test Videos of Meat Alternatives

In order to analyse consumer discourses participating in categorisation of meat alternatives, taste test videos from the YouTube social media platform have been chosen as a detailed data set (see again section 3.3.4). Similar to the previous chapter, an open inductive content analysis and a frame analysis were conducted (see again section 3.2). In the remainder of this chapter, firstly the sampled videos are described and introduced. Thereafter, the findings concerning category identity (Qb) and frames (Qc) are described in one section each. Finally, the findings are briefly summarised and a conclusion presented.

7.1 Description of the Data

The videos were uploaded between May 2015 and May 2018 (median March 2017), and no YouTuber was represented twice in the sample. Their duration is between 2:53 and 36:12 minutes (median 7:11 minutes), each having between one and 519 comments (median 10.5) and 100 to 97,327 views (median 1,544) on 17 May 2018. The most common tags associated to the videos are ‘vegan’ (7 videos), and ‘taste test’ (5 videos), with another four videos each featuring the tags ‘vegetarian’, ‘linda mccartney’, and ‘no bull burger’ (for all shared tags, see Appendix D.2). The videos cover a range of different demographics, settings, and circumstances, with presumably British people of an age range of approximately 15 to 50 although most appear to be in their 20s and 30s; among the YouTubers are eleven females and six males, one person non-white, and one reported a non-binary sexual orientation in the video. Contexts in which the videos are placed cover vegans having their friends sample, meat eaters trying vegan food, vegans sampling a single product, vegetarians comparing products, and combinations thereof. Appendix D.2 gives an overview of the sampled videos; and in Appendix D.3 a general structure of taste test videos is described based on my observations. In the following paragraphs, each video and the YouTubers are briefly summarised in alphabetic order of their (made-up) names. Alice is a teenager who appears to generally talk about cruelty-free cosmetics and

151 7. Taste Test Videos of Meat Alternatives reports to be unable to become vegetarian because they1 are too fond of Southern Fried Bites and other meatstuffs. In the video, Alice reports that their vegetarian and activist mother had decided to buy various meat alternatives including aforementioned Souther Fried Bites to be reviewed. Alice likes all the products, contemplates to become vegetarian, and urges their audience to at least try some meat alternatives. Charlotte is a vegan student, and received a vegan subscription box for which they had become affiliate. In the first part of the video, they unbox the contents, and describe and comment on the products, which include some vegan barbecue strips. In the second part of the video, they meet some of their meat eating friends who all eat and discuss the products. Elliot is vegan, and mental health and animal rights activist who likes the taste of meat. They had heard of Iceland’s new vegan burger online and wanted to sample and review it. Emily and Finlay sample the No Bull Burger from Iceland. Emily takes offence at the red colour of the burger, trying to establish that no meat burger would ever be that ‘bloody’. They compare the burger to other ones they had eaten before and speculated how other hyped burgers which they had not tried would compare to this. Georgia likely follows a meat-free diet but this is not entirely clear in the video. In the video which has slapstick influences, they sample some Chorizo and Red Pep- per Sausages from Linda McCartney’s with the aim to find better tasting vegetarian sausages as they find most others boring. Jessica highlights many times that they are a meat eater although being okay with eating ‘vegetarian food’ once in a while. Motivated by a friend who had done a similar taste test video on YouTube, they acknowledge people may find their opinion on Linda McCartney’s vegetarian sausages interesting. Jessica finds that the sampled sausages are a bad equivalent to meat sausages, but acknowledges that in comparison to hot dogs or frankfurters (which they do not like) they are quite tasty. Lucy is a student and may be considered a fitness and lifestyle influencer, and the video is sponsored by a company unrelated to meat alternatives. They are motivated by their recent decision to become vegan, after which they had ordered online some vegan versions of their favourite foods, including burgers, Parmesan, and pizza. Over the course of the video, Lucy elaborates that becoming vegan is not difficult, but that they would like to continue their established food practices. Lucy also acknowledges that they do not like taste and texture of meat anymore since becoming vegan. Lydia and Robert are vegetarian and meat eater respectively, and made with over 30 minutes by far the longest video in the sample. In the first part of the video, the

1Throughout this thesis I use they instead of gender-specific pronouns, because the preferred pronoun of interviewees or YouTubers was not discussed or assessed.

152 7.2. (Qb) Valuing Meat Alternatives viewer can join them on their shopping tour to various food retailers where they buy every meat-free sausage they can find in order to identify the ‘best’ one. Afterwards, in one long shot, they film the whole frying and trying session. Notably, meat eater Robert sometimes distinguishes between whether a sausage is tasty, or whether a sausage tastes like a (pork) meat sausage. They also consider in their evaluation how their children might react to the different sausages. Mia and Ollie try Iceland’s No Bull Burger together as ‘everyone’ seemed talking about them. They film and discuss their preparation of the burger, which Ollie eats in a bun, while Mia eats it plain as they try to avoid gluten and did not have gluten-free buns. Mia reports that they did not eat meat alternatives for a while after becoming vegan, but now actually like them for taste and nostalgia. Mia and Ollie among others discuss their surprise that Iceland would stock a vegan burger (and vegan products in general), and suggest that this is the best vegan burger they have eaten. Phoebe and Steven are a vegan couple trying to find the best vegan bacon. For this, Phoebe had bought a variety of products in a vegan online shop; they also emphasise their home-made marinade to further refine one of the bought products as well as tofu and mushrooms. In the evaluation they settle for the marinated product as the best. Afterwards Phoebe marinates other leftover bacon and prepares a vegan BLT (i.e. bacon, lettuce, tomatoes) sandwich. Pippa is a newly pescetarian parent, who had bought Linda McCartney’s vegetarian sausages after seeing an offer in a shop. They hope to find a vegetarian protein source as they do not want to return to eating meat. In contrast to the other videos, Pippa does a review of the product and not actually a live taste test; all narration in the video is a voice over which appears scripted. Pippa emphasises the good taste and versatility of the sausages. Zach buys and prepares a No Bull Burger from Iceland which they had heard about on social media to sample together with Sophia. They both like the burger, although Sophia is slightly appalled by the food colouring and Zach notes that they do not need to eat meat-like foods anymore but are happy to do so.

7.2 (Qb) Valuing Meat Alternatives

(Qb1) The descriptions used for meat alternatives in the taste test videos can be aggregated into three groups: Context, pre-tasting descriptions, and post-tasting descriptions. Firstly and least prominent across the videos, is a presentation of context. It is usually established in the beginning, when the product packaging is presented. Here, the YouTubers talk about ingredients, allergens, calories or nutrients, how items can be cooked, how many items are in a pack, or whether it is vegan or (less commonly)

153 7. Taste Test Videos of Meat Alternatives vegetarian. These are usually just mentioned or listed, but in rare circumstances may be further discussed; for example, Lydia declares to like that a sausage was vegan (as compared to vegetarian); and Pippa in response to the suggested preparation methods announces that they usually oven cook sausages “just because it’s easier for me.” Mia also found in line with online comments that the recommended preparation was not entirely satisfactory and thus comments on their preferences to cook burgers during the tasting. Secondly, once the products are unwrapped and presented on a plate or similar, products are started to be evaluated with descriptions of appearance and smell. When opening or touching a product, this may also be described in terms of appearance or texture, and these may also be used as a summary for the products. Appearance and smell are often discussed in terms of their positive or negative appeal to the YouTubers (e.g. “they are looking good”, Zach; “smell delicious”, Alice). There are also attempts to qualify these descriptors (e.g. “looks a bit wrinkly”, Lydia; “smells not as burnt as barbecue usually is”, Emily), and references to other products are drawn (e.g. “looks like raw meat”, Mia; “they smell quite sausage-y”, Jessica). Thirdly, during and after the sampling of the product, taste and texture are most commonly discussed. Both texture and taste are evaluated similarly to smell and appearance in terms of their appeal, a reference to another foodstuff, or a description. While likely to be mediated by texture, smell, and appearance, product taste is used throughout as a primary criterion to establish a hierarchy between products; this is made clear for example when Phoebe identifies the second best bacon which “looked the best, had the best texture, and was very, very close” to being the best but did not have the best taste, possibly because of their homemade marinade. The importance of taste and texture is also highlighted when after their first bite YouTubers enthuse about the products (e.g. “Woow. Ooh, these are bomb.”, Georgia). Taste but not texture is furthermore discussed in very vague (e.g. “very flavourful, definitely not bland”, Finlay) or absolute terms (e.g. “tastes like a really good marinated mushroom [but not like bacon]”, Phoebe). A summary of all codes and evaluations including examples is presented in Table 7.1. (Qb2) Product evaluations which rely on comparisons with other products usually compared the foodstuff to meat (see Table 7.2 for an overview with examples). Sampling the No Bull Burger, Ollie justifies this by stating that “when you’re eating a burger like that, you kind of naturally have to compare it to beef.” But even if meat alternatives are compared amongst each other, the reference value remains a meatstuff, as products are asserted to be “more realistic” (Ollie), or “as close as meaty as you’re gonna get” (Elliot). Quite insightful is for example what Emily has to say about their new favourite the No Bull Burger:

154 7.2. (Qb) Valuing Meat Alternatives

Table 7.1: Overview of characteristics and evaluations used to describe products in the videos (Qb1, Qb2).

Description n Evaluation Examples appearance 11 - appeal “they are looking good” (Zach); “this doesn’t look as appealing” (Alice) - reference “looks quite realistic” (Lydia); “don’t really look like sausage” (Jessica); “looks like raw meat” (Mia); “looks just like some fried chicken” (Alice) - description “looking a bit red” (Sophia); “looks a bit wrinkly” (Lydia) taste 11 - reference “they don’t taste exactly like meat” (Emily); “they taste more like hot dog sausages” (Jessica); “it doesn’t taste vegetarian” (Lydia) - appeal “Woow. Ooh, these are bomb” (Georgia); “Oh, my, god! Mhmmm. . .” (Alice); “that’s not the most disgusting vegetarian food I’ve ever tasted” (Jessica) - absolute “they have a nice, their own kind of taste” (Emily); “tastes like a really good marinated mushroom” (Phoebe); “doesn’t taste like anything like a sausage . . . but I really like it” (Robert) - description “tastes like it’s been barbecued” (Zach) - vague “there’s so much flavour” (Georgia); “very flavourful, definitely not bland” (Finlay) vegan 9 - mentioned “they’re basically a new vegan burger” (Mia); “Suitable for vegans just thought I’d double check.” (Georgia); “these vegan foods are made from seitan” (Phoebe) - appeal “it’s vegan as well, which I like” (Lydia) smell 8 - reference “they smell quite sausage-y” (Jessica); “it just smells just like KFC” (Alice); “smells like pasta, like tomato pasta” (Mia) - appeal “smell lush” (Robert); “smell delicious” (Alice) - description “smells . . . not as burnt as barbecue usually is” (Emily); “it’s got this like, Spanish, kind of scent” (Georgia) texture 7 - reference “if you’re not anti, obviously meaty textures, this is as close as meaty as you’re gonna get” (Elliot); “paté sort of texture” (Robert) - appeal “had the best texture” (Phoebe); “really enjoyed the texture as well” (Ollie)

155 7. Taste Test Videos of Meat Alternatives

Description n Evaluation Examples

- description “it’s not dry or anything, so, it crumbles very easily” (Jessica); “on, like, touching these, [you] get a lot of oil, look at that” (Georgia); “I like the texture, and you can taste onions and everything” (Robert) price 7 - appeal “the price is knocking [the rating] down a bit” (Emily); “it’s quite expensive but it’s worth it if you can get it” (Elliot); “1.25, that’s not bad a price” (Lydia) - mentioned “they do usually retail at about two pounds” (Pippa) - reference “we’ll compare like a standard quarter pounder . . . because a lot of people say that veganism is expensive” (Mia) preparation 6 - appeal “I always go for things that are really quick to cook” (Lucy); “I usually oven cook mine, just because it’s easier for me” (Pippa); “it cooks really really well” (Lydia) - description “they didn’t cook, they cooked properly, but the texture bit in the middle was a bit soft still” (Ollie); “only takes like fifteen minutes to cook” (Zach) packaging 6 - mentioned “they are actually frozen, you do get six in a pack” (Pippa) - appeal “I feel like the punchline should have been something else” (Mia) vegetarian 3 - mentioned “it’s vegetarian” (Alice); “these are vegetarian” (Lydia) (others) - mentioned ingredients (4); allergens (2); calories (2); nutrients (2)

And then they [Iceland] bring out like the best vegan burger ever! I still like the Linda McCartney burgers though, I think they’re kind of like their own separate thing, aren’t they? But, yeah, my parents, like, my stepdad, was really into–meat. And they tried it, and they didn’t like it ’cause they thought it was too meat-like! (Emily)

Goodness, is here established on the basis of how meat-like a meat alternative is. As this however might imply that less meat-like products are less valuable, Emily is eager to acknowledge that they like both products and see value in eating either. Similarly, when comparing sausages Robert asserts that one sausage “doesn’t taste like anything like a sausage, like a meat pork sausage. But I really like it, it’s nice . . . I like the texture, and you can taste onions, and everything.” Conversely, Ollie— with the same constellation of products as Emily but in a different video–notices a similarity between these products but states that the No Bull Burger was better

156 7.2. (Qb) Valuing Meat Alternatives without suggesting that Linda McCartney’s burger was nice in its own right. At the example of Jessica it can be seen that meat is however not always meat but in itself may be a contested reference point: Tasting vegetarian sausages and upon the first taste being a bit repelled by what they eat, they finally evaluate their sampled meat-free sausages in “comparison to actual sausage . . . [as] a two out of ten. As a comparison to a Frankfurter, they can have a good eight out of ten,” as Jessica appears not fond of Frankfurters. So while meat alternatives ought to be like meat, some acknowledge that they may just be nice or tasty in themselves, and that meat is not equal to meat. This highlights that all tasting and evaluation is constructed through experiences as well as expectations. These expectations are related to whether the individuals like meat or not, whether they have previously eaten meat alternatives or products similar to what they sample, and what they have heard about the products they sample. Maybe most notably, many YouTubers in the sample describe themselves as vegan or vegetarian themselves, making meat appear to be a memory-based reference: Finlay for example claims that they can not remember how meat tasted, but later asserts that their sampled burger is “way better” than the “packaged vacuum burgers” they used to eat. And although Georgia does not compare their meat-free sausages to meat, they acknowledge that they have never eaten Chorizo prior to sampling a sausage named after Chorizo. Voicing expectations based on others’ evaluations, Phoebe comments that they had selected one of the bacons in their comparison video because “this one I’ve seen so much raved about on Instagram, so, I’m hoping it’s gonna live up to my expectations.” Similar expectations based on social media accounts are also voiced by Elliot, Ollie and Mia, and Sophie and Zach, all concerning the No Bull Burger; and concerning the No Bull Bur- ger, Elliot, Ollie, and Finlay also make references to a meat-free burger from the US which has been hyped in various media: Although none of them tasted the product before, they wonder whether they would taste similar, or how much better that Bey- ond Burger might still taste. While what they actually expect is otherwise hardly ar- ticulated, expectations are likely to go along the lines of ‘good’ meat alternatives as established above. Illustrating this idea of ‘goodness’ from the opposite end, Jessica highlights their very low expectations when they conclude that “that’s not the most disgusting vegetarian food I’ve ever tasted.” ‘Meatiness’—for many of the above an indicator of goodness–may also be evaluated negatively in individual narratives. While not for themselves, Elliot acknowledges that their sampled burger “would probably put a lot of people that obviously don’t like the taste and texture of meat off.” And new vegan Lucy mentions that they have “got to the stage where I don’t like the taste of meat,” while Mia said they had been through that stage but now enjoy the nostalgic feeling together with their good taste and their animal-friendliness:

157 7. Taste Test Videos of Meat Alternatives

Table 7.2: Overview of comparisons made in the videos (Qb2).

Comparison to n Examples

meat and meatstuffs 7 “[my parents] didn’t like it ’cause they thought it was too meat like” (Emily); “Do any actual usual burgers come out looking that bloody?” (Emily); “tastes slightly gristle-y like bacon” (Steven); “that looks like meat, and, no, no joke, that smells like meat” (Charlotte); “kind of the same texture as pulled pork, like barbecued pulled pork” (Charlotte’s friend) meat alternatives 5 “I’m trying to think if I’ve had a more realistic burger, you know?” (Ollie); “I can taste the Quorn in it, which is, remind[s] me of a vegetarian sausage” (Robert); “this is as close as meaty as you’re gonna get [with a vegan burger]” (Elliot) untasted products 5 “have never had a Chorizo before in my life” (Georgia); “No Bull burgers from Iceland, these things are so so meaty, a lot of people have been absolutely raving about them, so these to rival the Beyond Burger.” (Elliot) (others) pulled Jackfruit (Lucy); jam (probably because of the colouring, Finlay); paté (Robert); “like a vegetable patty that you would make from the leftovers on the dinner sort of thing” (Robert); “there isn’t skin, so it kind tastes like you’re eating the inside of a sausage” (Alice); “as it goes on you taste it more planty” (Lydia)

When I first went vegan, I was so turned off from the taste of meat, and I really didn’t want meat replacements. I found replacements really gross, and I used to not enjoy them, and I wouldn’t really ever buy them. . . . Then as time goes on, you realise that there are these amazing alternatives that, are nostalgic to you, and make you kind of go back to your childhood, and they taste really nice, and there’s no animals being harmed in the process. So it’s all round just a good thing. And now I really quite like meat replacements. (Mia)

7.3 (Qc) The Meanings of Meat (Alternatives)

(Qc1) Because motivations, problems, or solutions are a core component of frames, firstly the motivations reported by the various YouTubers were identified. Before frames are thus elicited below, in the following paragraphs motivations to make the respective videos or to try a, another, or a specific meat alternative are described, and the YouTubers’ engagement with their audiences illustrated. At large, three reasons

158 7.3. (Qc) The Meanings of Meat (Alternatives) for making a video could be identified (for a detailed overview, see Table 7.3): Recom- mendations from social media, a specific dietary credential of a product, or finding a better product. Social media has informed them about a product (such as the No Bull Burger) or in case of Jessica brought the idea of doing a taste test, which was briefly discussed in the previous section. Diet is a reason in terms of dietary change (envisioned by Alice, and undertaken by Pippa and Lucy), or because products can be assorted to a different diet (vegan in case of Charlotte’s subscription box, and not meat but ‘vegetarian’ for Jessica). Pippa, who recently became a pescetarian, tries to consume more protein “without having to go back to eating meat”; new vegan Lucy wants to find vegan alternatives for their established (meat- and dairy-based) food practices; and Alice is considering a vegetarian diet but does not dare to transition because they like certain meat formats too much. Lucy and Alice also suggest that they would not appreciate if a change of diet meant a change in food practices, as Pippa makes clear after their review of vegetarian sausages that “I cannot say I miss meat sausages.” The idea to find a better product is explicitly part of Georgia’s argumentation–who believes that “all the like vegan, vegetarian, -like sausages, are just kind of plain Jane”—, but also implicit in everyone trying a product they have seen on social media. Further to that, Phoebe and Steve, as well as Lydia and Robert film their declared quest to identify the best product of a category, which are vegan bacon and meat-free sausages respectively. Regarding motivations to eat meat alternatives, ‘cruelty-free’ food seems to be the most important aspect for many. As this alone does not warrant the consumption of meat-alternatives, the YouTubers make arguments concerning taste (e.g. Phoebe, Alice, Elliot), nostalgia (Mia), convenience and lifestyle (Lucy), or nutrition (Pippa). Many others, despite not voicing motivations, say they are vegan and thus likely to be motivated by similar thoughts of wanting to eat animal-free foods. Only for Jessica, eating meat alternatives appears to be a means rather than an end, as they provide no personal narrative behind eating meat alternatives other than for the sake of curiosity in context of their video. As YouTube is a social medium and network, the YouTubers also address their audiences, thus not only speaking about meat alternatives but also encouraging people to take action. Charlotte specifically samples their meat-free food with and for a number of friends, thus putting the focus on the offline community and their responses to vegan food instead of an online community. All others who address their online audiences (see Table 7.3) make a plea to their audience in varying intensity: For example, Alice is rather explicit towards their audience, asking them to “go to your supermarket, and just get an alternative!” And when summarising their experience of sampling a burger, Finlay finds that “there is literally no excuse to not go vegan!” and thus implies that good facsimile products necessitate or enforce dietary change.

159 7. Taste Test Videos of Meat Alternatives become meat-free eat and share the burger, maybe at barbecues environment and animals veganism can be convenient — Motivations to Eat Meat cruelty-free, likes taste consider trying meat alternatives and/or and convenience nostalgia intake Alternatives Plea to Audience vegetarianism Overview of the motivations YouTubers’ to make a taste test video, to eat meat alternatives, and what they suggest to their audiences (Qc1). CharlotteElliotEmily & FinlayGeorgia recipient of vegan subscription box —Lucy hype on the internet thinks — most meat alternatives areLydia & bland RobertMia & Ollie (find the best new — vegan/vegetarian vegan, sausage) finding equivalentsPhoebe & Steven hype on the internet — (find the best vegan cruelty-free,Pippa bacon) loves meat (try vegan stuff) try the keeping burger the same lifestyle Sophia & Zach (not missing meat) — offer at hype on the social shop, media new pescetarian — (try the — burger/veganism) cruelty-free, taste, worried about protein — eat vegan versions because of health, — Jessica watched a taste test by a peer trying vegetarian food — YouTubers Motivation for Video/Sampling Alice mum bought products, interested in Table 7.3: Statements in brackets are not explicitly expressed in the videos.

160 7.3. (Qc) The Meanings of Meat (Alternatives)

Conversely, Phoebe who does not “see what harm it is to have the vegan version” if people liked taste and texture of vegan counterparts, as Phoebe subsequently touch upon (animal) cruelty and environmental issues, and then elaborate on health benefits. Mia acknowledges that a lot of their audience is interested in veganism, and Lucy offers to do more videos on veganism, such as kitchen or fridge tours in order to help the viewers to make a change if they were interested. (Qc) In their discussions of meat alternatives, the YouTubers identify problems and causes, impart moral judgement, provide specific solutions, and generally propose a specific view of the world and of meat alternatives. Building on the different motivations to eat meat alternatives or to make their videos, Table 7.4 lists the different identified frames for those videos that exhibit any motivation or issue. At large, there are differences in how the YouTubers utilise the frames, as some use them outwards oriented to make a public point to others, while other YouTubers make a private point to themselves. On the basis of these different frames, five lines of argument can be distinguished. Somewhat apart from the others and thus ignored in the following is that Georgia’s framing which exclusively focuses meat alternatives: They found that meat alternatives tasted bland, and through the recorded trial identified one that they liked. The other four frames all involve meat: At large there are two frames dealing with taste, and two with the context of meat consumption. In the following paragraphs, each of these four frames will be introduced with an exemplifying quote and then discussed in detail. The frames are summarised in Table 7.5. “I dead love my meat, but I just don’t love the cruelty and the suffering of the animals it was involved.” (Elliot) Elliot is championing a private, meat- loving frame which weighs up morals and pleasure. They find that meat is tasty but cruel, giving a negative meaning to what they understand to be meat and using meat alternatives to alleviate the perceived ‘sacrifice’. Such perspective inevitably brings enthusiasm for “a really, really meaty taster” (Elliot), which “tastes like a proper” or “realistic burger” (Mia), sausage, or similar. Acknowledging the limits of such ideas, Elliot finds that the eaten burger “is as close as meaty as you’re gonna get,” as Ollie points out the inevitability of such comparisons: “when you’re eating a burger like that, you kind of naturally have to compare it to beef. And when it tastes that good.” “If you like the flavour and the texture, I don’t see what harm it istohave the vegan version.” (Phoebe) Somewhat similar to the meat-loving frame, Phoebe proposes a reversed argument to Elliot as they identify meat eating as a problem and thus place satisfactory taste as a precondition to action in this vega-positive frame. Not- ably, Phoebe champions this as a public and thus outward-oriented frame, while they do not give information on why they are vegan themselves. This approach causes less friction in the audience, as it does not require full compliance, but instead reiter-

161 7. Taste Test Videos of Meat Alternatives animals if it tastes the same, choose vegetarianism is better for veganism is better ‘better’ — — eat meat alternatives, because they are great eat vegan meatthe sampled sausage is better — eat vegan versions of what eating you meat is unethical like eat (and share) the sampled burger meat alternatives are tasty, so please eat them meat-free sausages are found to be proteinaceous used (Qc). Statements in brackets are not explicitly expressed in the videos. 2 tasty; but eating meat is bad for animals because of animal suffering alternatives tasteless meat is not because meat eating is the norm and health) stopped eating meat Overview of the frames Table 7.4: CharlotteElliot publicGeorgia private + publicLucy (eating loves meat meat but causes private does environmental issues) not eat animals Mia & try Ollie becoming vegan privatePhoebe considers & previously Steven tried public meat public — Pippa veganism may be inconvenient, but eating barbecue can be difficult for new vegans, (meat is bad for animals, environment, private worries about protein intake as they YouTubers Orientation Diagnostic Frame Prognostic Frame Motivational Frame Alice private + public struggles to be vegetarian because meat is Emily & Finlay, Jessica, Lydia & Robert, and Sophia & Zach have been left out in this overview as no meaningful frames could be identified. 2

162 7.3. (Qc) The Meanings of Meat (Alternatives) ates supposedly shared issues–animal cruelty, environment, and health–as a reminder to act whenever possible. Alice and Mia also take this approach when making suggestions or propositions of gradually trying and/or adopting meat-free alternatives. Alice impersonates the tensions as they acknowledge before tasting meat-free versions of their favourite meat products that they “would never be able to be a vegetarian,” arguing that liking meat too much is “why people aren’t all vegetarians.” There are two different solutions proposed: As indicated by Phoebe, taste should be about “the flavour and texture [of] . . . a vegan version.” And as described above good flavour and texture can exist independent of meatiness, as long as the meaning of the product format, the ‘vegan version’ is retained. The other solution is to establish equivalence between meat and meat-free versions of a product: Asserting such equivalence, Fin- lay’s observation that “there is literally no excuse [anymore] to not go vegan” however assumes that burgers are the only meat people would eat. More cautiously, Robert believes that “if you were to give that [sausage] to somebody and not tell them it’s vegetarian sausage, it’ll, it’ll pass the test!”, leaving the proof of equivalence with the individual but not praising any less. “I’m not being like, oh, I won’t do it if it’s not convenient, but . . . it needs to be just as convenient as a normal diet.” (Lucy) In this frame, meat is simply normal. It is a meat-living frame which is entrenched as a supposedly ‘normal’ way of life in discourse and practices; this covers habitual practices such as the meal after the night out (Lucy), social gatherings such as a barbecue (Mia), or the meat-free Bolognese (Alice). However, transferability is not always given: Ollie for example suggests that bean burgers despite their good taste “don’t barbecue very well,” thus emphasising that not all meat alternatives are equally suited for all occasions. Associ- ations of foods, occasions, emotions, and ingrained practices can be observed throughout the videos, such as by Mia who finds that meat alternatives may be “nostalgic to you, and make you kind of go back to your childhood, and they taste really nice.” Putting this nostalgia into practice, Phoebe notes that “’cause I had a heap-loads of bacon left, it would have been rude not to have made a BLT.” Similarly, all the burger- focused videos exhibit a strong emotional bond between burger (patty) and bun, and Lydia “can imagine dipping that [cocktail sausage] into, some sauces, and, you know when you got friends over, and, you wanna do a few nibbles.” Somewhat reversed, Alice argues that they should try the foods without their staple condiment for meat products at first, realising that “then the ketchup can’t take away all the flavours and stuff.” In all this, not only meanings are transferrable, but also skills: For example, Mia argues that their burgers usually do not cook thoroughly in a pan and reports that they already “used to have that problem when I was a meat eater,” and instead usually bakes burgers in the oven. “I . . . wanted to look at other ways to get to some protein into my diet

163 7. Taste Test Videos of Meat Alternatives without having to go back to eating meat.” (Pippa) As a consequence of this desire for protein, Pippa identifies meat-free sausages as a solution. More so than nour- ishment, this is a meat-aware frame which encompasses tacit knowledge associated with meat. While this also implies normality, in this frame it appears to shape contextual knowledge more than it influences practices. In Table 7.5 the four frames are summarised with a brief notion of what meat and meat alternatives are understood as. Table 7.5: Summary of the frames identified in the taste test videos and how they understand meat and meat alternatives (Qc).

Frame Meat as . . . Meat Alternatives as . . .

meat-loving tasty but cruel (almost as) tasty remedy vega-positive harmful but replaceable (independently) tasty alternative meat-living entrenched filling the same functions in life meat-aware context anchor in times of change

7.4 Conclusion

In this chapter, the taste test videos were introduced, and findings presented. These videos exhibit a consumer-based description, evaluation, contextualisation, and therefore also categorisation of meat alternatives. When meat alternatives are described in context of the videos (Qb), sensory attributes are the most prominent. These attributes are often related to meat, but still, whenever meat alternatives are compared, the reference to establish an order of worth persists to be meat. These evaluations are mediated by expectations and experiences, and can be highly different based on what meat is used as a reference baseline. When one product was found to be less meaty than another but is still likeable, it becomes a ‘thing in itself’ valued for its good taste rather than its meatiness. The prominence of meat as a reference for the evaluation of meat can therefore be asserted. The frames used for meat alternatives (Qc) cover meat as something tasty yet harmful or cruel, but also as an entrenched way of life, and as related to tacit knowledge. For some YouTubers, meat is a contradiction between nostalgia, a way of life, or a love for a certain taste on the one hand, and cruelty and harm on the other hand. This contradiction is not always made explicit, as not all YouTubers state their own reasons for eating meat-free or insinuate why others might follow such diet. Nonethe- less, in many of the videos it is implied that the abstention from eating animals is a goal people want to pursue. In these lines of thought, meat alternatives symbolise normality and convenience,

164 7.4. Conclusion as practices and discourses are upheld by consuming meat alternatives. They also give a sense of nostalgia by allowing to transfer occasions and meanings which have long accompanied people. Notably, meat is also framed beyond its meaning as something known to fulfil certain needs or functions. For many of these YouTubers who are not opposed to eating meat alternatives, dominant culture thus becomes a carrier for meat alternatives, as they are able to reconcile dominant culture and broader consciousness. As such, meat alternatives provide a sense of certainty and continuity while being able to satisfy desires for change.

165

Chapter 8

Tweets on Replacing Meat

Alongside the YouTube videos discussed in the previous chapter, tweets have been chosen as a second consumer-based data set because of the large variety of different consumers and their perspectives represented. In this chapter the individual findings regarding category identity (Qb), and frames for meat alternatives and meat (Qc) are presented. Note that due to the specifics of the data, question Qb2 on evaluation and references could not be answered, and question Qc is treated as a single question (see again the relevant research design in section 3.3.5). The tweets were analysed using a guided content analysis and frame analysis (section 3.2). This is the last of four chapters primarily concerned with presenting the findings for each of the data sets. In the following chapter, findings from all data sets are analysed along the lines of the research questions. The remainder of the chapter proceeds as follows: The overall sample of tweets collected, as well as the subset used for analysis are described in detail. Thereafter, the findings regarding category identity (Qb) and frames (Qc) are presented in corres- ponding sections. Lastly, the chapter is summarised and concluded.

8.1 Description of the Data

Including padding at either end of the weeks, the sample collected during #Nation- alVegetarianWeek (NVW)1 comprises 32,264 tweets, and the sample collected during #WorldMeatFreeWeek (WMFW) 8,013 tweets. They are published by a total of 22,116 different users resulting in an average of 1.83 tweets per user across both theme weeks; only 716 users have posted at least once in each week. In both weeks, Monday midday (11:30–12:30 BST) had the highest amount of activity; daily peaks of activity and tweets per day declined towards Sunday, with British nighttime showing the least

1In the following, the abbreviations NVW and WMFW will be used to identify the specific theme week on twitter. So ‘during NVW’ means that it was during National Vegetarian Week and included in the sample as a relation to the relevant hashtag was created; it does not account tweets during that time which were posted unrelated to the theme week, even though they may have dealt with meat alternatives.

167 8. Tweets on Replacing Meat activity. Figure 8.1 shows the overall distribution of tweets throughout the two weeks; Table 8.1 gives an overview of a range of statistics about the two samples.

Table 8.1: Overview of the collected tweets.

NVW WMFW Total % of Total

total 32264 8013 40277 100% cropped2 31405 6762 38164 95% in English 31058 7088 38146 95% retweets 21931 5917 27848 69% replies 1148 129 1277 3% quotes 2090 808 2898 7% users 17514 5268 22066 1.83 per user tweets of top 20 users 3314 1024 3556 177.8 per user peak activity (hourly) 989 228 3% time of peak (in BST) Mon, 12pm Mon, 12pm National Vegetarian Week 600

400

200

0 World Meat Free Week 150

Tweets per Hour 100

0 Mon Tue Wed Thu Fri Sat Sun Day of the Week Type of Tweet original quote or reply retweet

Figure 8.1: Timeline of tweet frequencies throughout the two theme weeks. Note again that WMFW is likely to have a more international audience (see again Appendix E.1), which can be observed in the distribution of tweets.

The organiser of National Vegetarian Week, the Vegetarian Society, was the most active (n=1424) during NVW (they posted another 16 tweets for WMFW); the second

2This covers the amount of tweets posted from midnight British Summer Time (BST) of the starting date to midnight BST on the end date of the respective week.

168 8.1. Description of the Data most active user tweeted 218 times in total. The ‘World Meat Free Week’ account tweeted a total of 70 times and exclusively during that week. Among the 50 most active users during both weeks, are many commercial accounts named for example ‘The Flexitarian’, ‘LOHAS Top Stories’, ‘The Cook & Him’, or ‘Chicken Shop’. Yet, the only meat-free brands in the top 50 are Swedish meat alternatives brand Oumph! and Danish frozen food producer Daloon Foods, both of which have a marginal stake in the UK market. Retailer Tesco is also among the top 50. The activity of other well- known UK brands and their rank are listed in Table 8.2. The user distribution indicates that both theme weeks are not primarily motivated and driven by promotions of big businesses, but instead by individuals (who may however also express their loyalty to brands, see for example Paquette, 2013) and smaller businesses.

Table 8.2: Overview of the participation of selected meat-free brands and other stakeholders in the two theme weeks.

Twitter Account Tweets Rank Notes

Vegetarian Society 1440 1 mostly during NVW — 218 2 Oumph! 72 20 (Swedish) World Meat Free Week 70 21 only during WMFW Tesco 48 35 UK retailer Meat Free Mondays 38 42 campaign group Daloon Foods (UK) 36 44 (Danish) — 36 50 Quorn (UK) 30 73 UK market leader Linda McCartney’s 14 248 Fry’s (UK) 10 423 (South African) VBites 6 840

Food words were mentioned significantly more often during NVW (χ2(1, N = 12429) = 62.521, p < 0.001, V = 0.071)3. The most prominent food categories from WordNet (see again section 3.3.5) in the sample were vegetables (29% of not retweeted messages), fruits (11%), meat (14%), baked goods (4%), and dairy products (4%); ‘ingredients’ were mentioned in 9% of the sample. Words related to dishes (15%) and meals (10%) were also common. Mentions of meat-related words were significantly

3V stands for Cramér’s V and can be understood as the effect size for χ2-tests. It is often used for large sample sizes, as significant outcomes of χ2-tests are more likely in these cases. Results can be interpreted as a correlation ranging from 0 (no correlation and equal distribution) to 1 (completely unequal distribution as a strong correlation), with results below 0.3 indicating weak differences (see also McHugh, 2013, and Appendix E.3).

169 8. Tweets on Replacing Meat more frequently used during WMFW (p << 0.001, V = 0.190). The remaining main food categories were either not significant or significantly more common during NVW. The strongest difference apart from meat was observed in vegetable words (V = 0.171). Albeit the theme week hashtags were removed, the stark contrast is likely to originate from the name of the respective week, as the term ‘veggie’ is sometimes used synonymously to vegetarian, but is also a vegetable-related word in WordNet. However, as illustrated in a comparison between the two weeks in Figure 8.2, the contrast about different food objectives remains throughout the other categories such as fruit, baked goods or dairy products. For absolute counts, and more information on the less frequent categories see Appendix E.1.

vegetables 30% meats

20%

dishes others

10% ingredients meals Percentage of Tweets

Mentioning a Type of Food baked goods fruits

0% National Vegetarian Week World Meat Free Week

Figure 8.2: Types of foods mentioned in the tweets by theme week.

From the 1,675 tweets containing meat-related words, tweets were manually identified for further analysis if they related to meat alternatives or replacing or swapping meat. Of these 225 tweets, 168 were posted during NVW, and 57 during WMFW. The share of tweets that originate from WMFW is significantly higher than if both weeks were to be represented proportionally (χ2(1, N = 12429) = 3.86, p < 0.05). In the sample, 13 authors contributed two, two authors three, and one author four tweets, while all others contributed only once. In 19% (n=42) of the tweets, users reported about themselves, for example what they had eaten or how long they had been vegetarian; 61% (n=138) engaged in some sort of promotion, mostly of recipes (n=44), products (n=27), or eateries (n=33). In 9% (n=21) of tweets, authors primarily gave their opinion on specific products or meat alternatives in general; and in 8% (n=19) ‘information’ was spread, which covered among others supermarket stockings, or health effects. The remaining tweets, were

170 8.2. (Qb) Describing Meat-Free: How Meat Matters attempts to make conversation by asking about audiences’ favourite meat alternatives (n=2), as well as humorous contributions to the debate (n=3). Quorn is the most prominently referred to brand in the sample (n=39, 17% of total), followed by Linda McCartney’s (n=13, 6%). The most popular meat formats to be replaced are burgers (n=32), chicken (n=24), sausages (n=14), wings (n=11), steak (n=11), roast (n=8), kebab (n=7), hot dogs (n=6), and bacon (n=6).

8.2 (Qb) Describing Meat-Free: How Meat Matters

(Qb1) 60% (n=135) of all tweets described the meat alternative featured in the tweet. Many twitter users talk about the pleasure invoked by a product, others mention health, food preparation, taste, environment, animals, and product texture. An overview of the different characteristics and their occurrences is tabulated in Table 8.3. As characterisations about the subjective pleasure induced by products are most frequent, many tweets remain shallow and vague, such as by describing the foodstuffs as ‘delicious’ (n=23) or ‘tasty’ (n=16) or otherwise pleasurable or good (n=48), while actual taste is hardly described (n=9). However, many taste experiences and expectations are also embodied in specific format attributions: Items such as “Buffalo Cauliflower Wings” (e.g. N8, W16), “Peri Peri Vegetarian Sausage Rolls” (N3), or “Kentucky Fried Cauliflower” (N70) connect meatstuffs with spice blends and means of preparation. These observations indicate that the taste of meat alternatives may just like meat be more reliant on seasoning or marinade. Similar to descriptions of taste, the few mentions of texture of meat alternatives blur description of texture with specific expectations of what meatstuffs ought to be like: For example, the two tweets mentioning the term ‘crispy’ both concern ‘mock duck’ (N11, N130), while a chicken is shredded (N167), and there are “thick cut chunky strips” (W40). Descriptions of smell are completely absent, and appearance is discussed in similarly shallow ways such as “look delicious” (N152), or “look like meat” (N15, W43). This indicates that in many tweets, sensory product features are understood in reference to meat. Accordingly, there some tweets (n=11) emphasise that also meat eaters may like a foodstuff; for example as products may be “popular even with meat lovers” (N43), that “not even the most hardened meat eaters could tell the difference” (W31), or products “that even fooled a few meat eaters” (N114). These descriptions suggest that an adequate language for describing meat alternatives is missing in the tweets. Instead, they rely on binary good/bad schemes and on what may be expected from similar meatstuffs, which is possibly reinforced by the length restrictions of these messages. Some tweets highlighted the ease and versatility of preparation for the featured foodstuffs (n=12). For example, it was stated “that Quorn is a very simple and easy

171 8. Tweets on Replacing Meat

Table 8.3: Overview of descriptions for meat alternatives in the tweets (Qb1); as many tweets contained multiple characterisations, counts are based on the number of tweets mentioning a specific characteristic and do not add up.

Characteristic n Details

any 135 (60%) pleasure 87 delicious (22); tasty (16); even meat eaters would like this (11); great (10); amazing (4); best (4); flavour (3); awesome (3); favourites (2); others (21) health 22 healthy (10); protein (6); fibre (2); low calories (2); low fat (2); vitamins (2); others (5) preparation 12 easy (7); versatile (5) taste 9 bbq (4); smoky (3); beefy (1); meaty (1); spicy (1); tangy (1) environment 9 save the planet (3); save resources (2); not bad for environment (1); sustainable (1); ecosystems (1); palm-oil free (1) animals 7 ethical consumption (2); stop eating animals (2); cruelty (1); suffering (1); help animals (1) texture 5 crispy (2); chunky (1); shredded (1); squeaky (1) appearance 5 like meat (2); tasty/delicious (2); bleeding (1) others 24 homemade (4); naming (3); creative (3); price (2); previous products (2); ubiquitous (2); others (9) replacement for meat” (N168), that one can “incorporate this tasty, meat alternative [tofu] into a variety of salads, stir-fry’s and pastas” (N112), or that one can make a “jerky . . . snack . . . from loads of different fruits and vegetables” (N125). These em- phases may counter established preoccupations that meat-free cooking may be difficult, cumbersome, or time-consuming, or that one needs many different ingredients. Concerns about animals and environmental issues are not frequently mentioned directly in relation to products. Tweets among others suggested that swapping meat for other foodstuffs would be “good for the planet too” (N53), or “of benefit for Planet Earth” (W25). Others noted that it would save resources (e.g. “you will feel better, save tonnes of water, crops, and energy”, N55), or even entire ecosystems (“soya . . . driving destruction of forest, savannah & marshland, is embedded in meat, dairy, eggs & fish”, W57), with meat being “really bad for the environment” (W48). It was also noted that “the meat substitutes on the market are . . . sustainable” (N151), thereby emphasising a positive credential of the product itself. Generally, there is no clear distinction between animal rights and animal welfare, with some mentioning generally that the featured products are ‘ethical’ (W4, N151) or “help the animals” (W42), while others suggest more directly “extra veg instead of the dead animal” (N21),

172 8.2. (Qb) Describing Meat-Free: How Meat Matters or meatstuffs “minus the #animalcruelty” (N117). Focus on environment and on animals both function by either highlighting destructive capacities of prevalent diets, or pointing out the alleviation or eradication of issues through changing consumption. Often these changes are presented as effortless, thus suggesting an implicit moral duty for change. Claimed health credentials of meat alternatives remain either vague by use of the term ‘healthy’, or list a variety of contents or the absence thereof, such as high in protein or fibre, or low in calories or fat. As most tweets do not make a direct connection between these (non-)contents and health, this has to be interpreted by the audience, and accordingly also by the researcher. These claims are usually broad, such as “low in saturated fat and very high in protein” (W4), or “reduce ur calories intake by a 3rd & ur total fat intake by over two-thirds” (W10), and may resonate with specific audiences. Other characteristics used to describe the products or recipes are for example homemade or creative, ubiquitous, natural, or fresh; some tweets comment on the naming of products, as others make links to meat alternatives when they first became vegetarian, comment on the suitability of products for celebrations, or the “sizzling” sound (W21) products make when cooking. (Qb3) In Table 8.4, category labels used for the different products are listed. The most prominent labels are those concerning meat in general, such as meat alternatives, substitutes, meat-free, or meat replacements, and specific formats, such as sausages, burgers, roasts, or steaks. This prevalence is not surprising as meat-related words were used to filter the overall data set. Meat-related labels are the only labels that are generally used independently of others, with the exception that brands might be mentioned as representative of their products (e.g. “Shout out @LindaFoods”, N58), or when products are reduced to protein (e.g. “Switch your daily dose of tough to cook protein to green and crunchy vegetables”, W50). Meat-related labels often function as a general remark over a range of products such as in statements like “creative and tasty alternatives to meat and fish dishes” (N84), “more fantastic meat alternatives” (N5), or “I don’t miss meat, [as] there are so many yummy alternatives out there” (N37). All other labels require a combination of labels to establish an understanding of what the foodstuff represents. Labels highlighting a brand or suitable diets are commonly used as attributes to products, such as in combination with formats like ‘Quorn sausages’ or ‘vegan sausages’. Less frequent are combinations with meat- related labels (e.g. “I find that Quorn is a very simple and easy replacement for meat.”, N168). Similarly, contents can hardly function independently as cauliflower or tofu alone are not considered a meat alternative (see again the scope of this work in section 1.3). However, cauliflower wings, cauliflower steak, or tofu sausages are, and through their labelling incur some meaning from meat.

173 8. Tweets on Replacing Meat

Table 8.4: Overview of labels used for meat alternatives in the tweets (Qb3); combinations of these were possible, so counts do not add up.

Label n Examples

format 120 “chicken fried seitan” (N160); “vegetarian sausage rolls” (W47); “Linda Mccartney vegetarian sausages . . . and Quorn Mince (meat free)” (N46) meat 97 - alternative 26 “meat alternatives” (W35); “swapping out meat in your diet for some tasty alternatives” (N104) - free 23 “with meat free menus” (W41); “meat free sausage and leek hash for tea” (N74) - substitutes 12 “ideas for meat substitutes” (N85); “which is the best meat substitute?” (N164) - swap 10 “it’s easy to swap meat” (N14); “swapping out meat for a #plantbased meal” (N147) - alternative to 7 “there are loads of alternatives to meat that we can eat” (N56); “delicious vegetarian alternatives to our meat based canapés” (W33) - fake 4 “your range is poor & full of Fake meat” (N113); “The fake burger test” (N95) - replacement 4 “replace the meat” (N123); “a very simple and easy replacement for meat” (N168) - others 17 e.g. “pulled”chicken” hoagie” (N1); “so delicious that you won’t miss the meat?” (N136); “instead of meat” (N88); “without the meat” (N89) diet 83 - vegan 30 “can you believe this burger is vegan” (N114); “delicious vegan sausage rolls” (W47); “#Vegan jackfruit ‘duck’ pancakes” (N47) - vegetarian 28 “try our new vegetarian option hot dog” (N30); “everything in our restaurants is Vegetarian, even our Chicken style burgers” (N63) - veggie 27 “our veggie sausage and butternut squash traybake” (N71); “I really dislike most of the veggie substitutes” (W2) brand 45 “@QuornFoods chicken” (N10); “Shout out @LindaFoods” (N58); “made with @quornfoods” (W24) contents 43 - specific 28 “Cauliflower Steaks” (N76); “Seitan Duck” (N130); “beet burger” (W1)

174 8.3. (Qc) Talking Meat: How Tweets Frame Normality and Disruption

Label n Examples

- plant-based 14 “#plantbased alternative” (W11); “#plantbased #vegetarian alternatives” (N110); “we have been looking at the 100% plant-based steak” (N143) - others 1 “something delicious and plant powered on the menu” (N12) dish 32 “Vegan Paella” (W7); “Quorn ‘Cowboy’ on the menu today” (N39); “Mixed Mushroom Stroganoff” (N77) others 31 - option 7 “21 Vegetarian Main Meal options available” (N98); “up your ‘veggie option’ game” (N107) - protein 7 “Nut, seed proteins more heart healthy than meat” (N61); “protein without the meat” (N89) - food 6 “meat free food” (W36); “super delicious plant based food” (N137) - products 4 “reflecting back there was hardly any vege products” (N90); “I hate veggie products that look like meat.” (N15) - purpose 2 “Our Buffalo Cauliflower Wings . . . are a perfect party nibble” (N72) - version 2 “a veg-version of our latest special” (N120) - others 6 “to eat less soya” (W57); “I think won’t feel meat free with these bad boys” (W20)

Dishes are frequently also combined with any of the other labels. They too would not be properly identifiable as based on or containing meat alternatives, if they did not resolve the tension between a meat names or ideas and plant-based preparation. Examples for this are “Vegan Paella” (W7), “Al Pastor ‘pork’ ” (N116), or “Mediter- ranean vegetable ragu or roast vegetable lasagne” (N12), as opposed to Paella, pork, or lasagne. Notably, in some dishes meat alternatives appear to be added as choice, but not because the dish would require this; examples are a “Lentil and Sweet Potato stew made with @quornfoods” (W24), or a “@QuornFoods Rainbow Salad with your Chicken Pieces” (N51).

8.3 (Qc) Talking Meat: How Tweets Frame Normality and Disruption

(Qc) As tweets are short by requirement, only one frame resembling the ‘dominant meaning’ was coded in each tweet. This was then expanded into diagnostic, prognostic, and motivational frames. Most tweets however, do not feature an issue but instead concern the normality represented in meat alternatives, or highlight meat-free foodstuffs as a positive choice. Generally, tweets emphasise a change or a swap of

175 8. Tweets on Replacing Meat products for a meat-free version, which might not be exactly the same. Other tweets frame meat as something to be reduced with the help of meat alternatives, and some of these tweets also highlight why meat ought to be reduced. An overview with examples is listed in Table 8.5; each frame is discussed in detail in the following paragraphs. The most frequent frame concerns normality: In these tweets, a business-as- usual case is drawn out, where meat alternatives are accepted as normal and established, either by themselves or as a replacement for meat. A question such as “what’s your favourite meat replacement item?” (N34) already implies that there a wide array of options which may be evaluated very differently. Many tweets assert normality by discussing specific recipes, such as when a tweet suggests to “bring on the buffalo without the wing with a big bowl of tempting cauliflower bites” (N13), and another one proposes “Spicy and meat free! Vegetarian curries are where it’s at. Try my Quorn Keema for #NationalVegetarianWeek” (N153). Other tweets emphasise meat-free choices in eateries, such as in this tweet: “Did you know we have a range of delicious vegetarian food in our Bistro . . . [which] is solely vegetarian? Even our staff can’t get enough of the exquisite nut roast” (N66). Normality can also cover meat alternatives, such as for someone who “buy[s] Quorn nuggets instead of chicken nuggets as they actually taste great” and not because they are vegetarian (N149); or when “Quorn ‘Cowboy’ [is] on the menu today, one of the children’s favourites. Great replacement for chicken” (N39). In all these tweets, it is implied that meat alternatives are a normal foodstuff which–whether homemade or bought–are suited for parties or as snacks, for dinner or for eating out; this includes individual occasions as well as habitual use. Notably though, in many cases, meat alternatives remain a replacement for meat in the dish and also people’s mind. This symbolic relevance of meat is reflected, when tweets cover quotation marks around meat names, such as pulled or fried “chicken” (N1, N157), cauliflower “chicken” (N23), or “Hot Nachos Topped with Al Pastor ‘pork’ ” (N116). So while these tweets assert normality to the use of meat alternatives, many equally normalise the idea of eating meat in the first place. Two other frames propose a choice between supposedly equal foodstuffs, or a swap between two possibly different foodstuffs. They promote products or recipes as a suggestion is made to the audience to consider exchanging one food item for another, or as singular efforts for one meal (or one week) at a time. Both frames emphasise the theme week as a potential reason to swap or choose differently, such as when something was cooked “in honour of #nationalvegetarianweek” (N73) or a dish is offered “to celebrate #WorldMeatFreeWeek” (W27). When asking their audiences to swap to a different product, tweets suggest that products may not be equal but require a change, such as using “meaty shiitake mushrooms instead of pork” (W3), or replacing “chicken for tofu for a vegetarian treat” (N83). Notably, such swap appears effortful and might be considered unusual, such as indicated by the following tweet

176 8.3. (Qc) Talking Meat: How Tweets Frame Normality and Disruption

Table 8.5: Overview of frames used in the tweets (Qc).

Frame n Examples

normality 81 “@QuornFoods my Veggie salad with bulgar wheat/spelt lettuce/quinoa and your chicken pieces” (N145); “We feel like keeping it healthy this weekend so why not join us and tuck into this amazing tofu steak with Snow Pea, Rocket and Radish salad with a Wasabi dressing – it definitely packs a punch!” (N62) choice 47 “have you tried our veggie sausage and butternut squash traybake recipe” (N71); “Tried something new tonight vegetarian sausage pasta using @LindaFoods vegetarian sausages” (W46) swap 35 “why not swap your Sunday Roast for a veggie alternative” (N138); “It’s #FathersDay and of course Dad deserves his favourite meal tonight. Swop out the meat in his curry for a veggie version, and his normal rice for brown. . . We bet he’ll love it just as much.” (W13) reduction 20 “How hard is it to give up meat in the UK?” (N6); “If you’re worried about going veggie, trust me you can do it! I’ve done it for almost 2 years and I can honestly say I don’t miss meat, there are so many yummy alternatives out there and I’m definitely healthier now I think more about what I’m cooking!” (N37) impacts 20 “Reducing meat consumption is good for you, animals, and the environment” (N45); “Id rather have the version that isnt carcinogenic, requires no suffering, and isnt really bad for the environment. #Vegan Jerky all the way!” (W48) others 22 from an eatery: “If you have a favourite dish, please ask and we may be able to prepare it specially for you or replace the meat with a vegetable substitute” (N123). Or a tweet which suggests that for a party “you need to cater for the vegetarians too” (N57). Conversely, choice frames position replacements as equivalent options (e.g. “From Chicken Green Curry to King Prawn Jungle Curry, we can guarantee we have your Thai favourite! All can be made Vegetarian”, N41; “Who needs meat when you can have Buffalo Cauliflower wings? Check out our tasty recipe”, W16) or an independently good option (e.g. “our homemade vegetable lasagne is set to up your ‘veggie option’ ”, N107; “Sink your teeth into these meat free burgers”, N133). Although less prominent than in swap frames are suggested, also choice frames highlight that meat eating is an established norm; for example in tweets such as “just because we’re a chicken shop don’t think we’re not all over #nationalvegetarianweek” (N92) or “Try these crispy mock duck pancakes this #NationalVegetarianWeek for a #vegan twist

177 8. Tweets on Replacing Meat on a takeaway favourite. . . #mockmeat” (N11). Either way, as swap or choice, replacements are an option for the audiences as these tweets do not use coercion or suggest that eating meat-free might be preferable in any way. Apart from a few exceptions which are meant rather humorously, the remainder of tweets are framed in relation to a reduction of meat consumption either in general or for specific reasons. In both cases meat can be ‘reduced’ because its consumption is made quantifiable in suggestions such as “u don’t have to fully cut out meat to make a difference with little changes like that” (N124); in this tweet the idea or implications of ‘making a difference’ remain unexplored4. Opposing previous frames, these two frames imply a rather long-term perspective, as dietary change is largely seen as a goal. In the reduction frames, a rationale is often implicit, while reasons why meat-free eating may be more desirable are not discussed. For example, accessibility to meat alternatives in stores or take-aways is considered a “great way to get people to try them and realise that they can reduce and or stop eating meat” (W35), and it is suggested to “try easing into meatless meals” (W44) with one meal at a time. Only few tweets consider that not all audiences may be interested in reducing meat consumption, such as by asking “Do you want to try and cut more meat out of your #diet?” (N129) External impacts are used as a frame to motivate the reduction of meat. Motiv- ations cover environment, animals, and health, and are either attributed as positive changes when meat consumption is reduced (e.g. “even if you just stop eating AS MUCH meat instead of completely going veggie it helps the animals and the environment and your general health!”, N163), or highlighted as negative impacts of meat (e.g. “Id rather have the version that isnt carcinogenic, requires no suffering, and isnt really bad for the environment.”, W48). When tweets further emphasise the normality of meat alternatives or assert the equivalence of meat and meat-free foodstuffs, a change of diet becomes a a moral responsibility to audiences. For example, “these 30+ #veganrecipes all provide the authenticity of #bacon — minus the #animalcruelty” (N117), or “not even the most hardened meat eaters could tell the difference and we saved the carbon equivalent of boiling 2330 kettles” (W31). However, even if a reduction is underlined with justification, the reasons often remain unexplained and vague, so that the audiences themselves have to know or infer the connections between meat and among others cancer, animal cruelty, or environmental destruction. A summary of all discussed frames is provided in Table 8.6.

4It should be noted that this ‘difference’ to be made for whatever reason often scales disproportionately to the reduction as a cause of the difference, such as: “Little changes like this can make a big difference on your way to eating less meat” (W3). Difference can already be asserted after one meal (e.g. “Even swapping one meal to meat free makes a difference!”, W50), indicating that ‘making a difference’ may be largely used to motivate audiences.

178 8.4. Conclusion preferable to meat as any reduction matters foodstuff non-standard, legitimate choice reducible preferrable) duty to reduce harmful helps to replace — countable and substitute with meat alternatives substitute with meat alternatives to be reduced to be reduced because meat is ‘harmful’ Summary of the frames including the issues and solutions marked in the tweets (Qc). Statements in brackets are not explicitly expressed in Framenormality Diagnostic Framechoice (meat is normal)swap (meat-free for Prognostic a Frame week)reduction meat alternatives (meat-free are for the a same week) choose meat-free options meat consumption needs —impacts change individual meals — Motivational Frame meat consumption needs Meat as . . (meat-free . might be Meat Alternatives as . normality . . foodstuff identical equivalent, legitimate choice Table 8.6: the tweets.

179 8. Tweets on Replacing Meat

8.4 Conclusion

In this chapter, tweets concerning meat alternatives posted during #NationalVegetari- anWeek or #WorldMeatFreeWeek were introduced, and findings presented. Describ- ing meat alternatives in the context of these tweets (Qb), they were mostly associated to a pleasurable experience, while less prominently they were related to health, environment, or animal suffering, and few tweets described taste, texture, or appearance. The labels used for the different products are concerned with formats (e.g. sausage or steak), around meat (e.g. meat alternatives or replacement for meat), or a dietary identification (e.g. vegan, vegetarian). Other labels were concerned with the brand, contents, or the dish a foodstuff is embedded in. Apart from those labels concerning meat, it was found that no label is used independently–because they can not clearly designate a foodstuff which is like a meatstuff but does not include meat. The frames used across the tweets (Qc) focus on meat alternatives as a lived normality, as a plausible choice or swap for a single meal, or as something to reduce meat consumption. Meat alternatives as a normality resemble an idea that meat eating is normal, but also meat alternatives are understood as normal and legitimate by the respective twitter users. When meat alternatives are understood as a choice or a swap, the focus is usually to make different choices for a single meal, as tweets imply the necessity of a swap or suggest free choice. In those frames where meat has to be reduced, meat alternatives become an easy and convenient facilitator to such change, although only some tweets highlight the reason why this might be desirable. As expected, the ideas which can be communicated about meat alternatives are somewhat limited within single tweets, while there is a large diversity of different perspectives. Meat alternatives are most often seen hedonistically as something nice, tasty, or otherwise pleasurable. Although many tweets positively emphasise the quality of meat alternatives available for purchase, also products deviating from meaty sensory experiences such as ‘chickpea burgers’ or ‘cauliflower wings’ are described as pleasurable. Furthermore, through their contextualisation as specific meat formats such as wings, steaks, or burgers–they attain established meanings and become suitable for specific occasions (e.g. ‘party nibbles’). Even if meat is replaced with foodstuffs such as lentils, the used frames focus on individual interests such as health or protein intake. For those twitter users describing their own meat alternatives consumption, these products are established in individual lives, and may even exist alongside meat as regular or one-off treats, for example because they taste better. Further to being hedonistic and normal, meat alternatives can be understood as a tool able to quickly and comfortably reduce individual meat consumption in general as well as in specific dishes; the reasons for the necessity to reduce or eliminate meat consumption are often undiscussed, possibly because audiences are expected to know about these.

180 Chapter 9

Meat Without Animals: An Analysis

This research project engages with the question of how discourse around a hypothesised category of meat alternatives reproduces and/or challenges dominant culture. Whilst in the previous chapters, the collected data–contextual data, advertisements, expert interviews, taste test videos, and tweets–were reviewed, contextualised, and integrated by data set, in this chapter the findings are combined into a rich analysis. Broadly along the lines of the original research questions, this chapter will firstly establish the foundations of the category and why–in category studies terms–it can be treated as an established and coherent category and thus analysed as a single entity throughout this chapter (Qa, Qb, section 9.1). Thereafter, emerging understandings of meat alternatives will be discussed (Qa, section 9.2) in order to contextualise the current situation, as well as to explore some suggestions made in the conceptual framework; this is primarily based on the longitudinal study of advertisements (section 5.2) and the contextual data. Next, current understandings of meat alternatives are discussed (Qb, section 9.3): Building on all data sets used in this thesis, a rich analysis is provided which explores the use and references to meat in talk about meat alternatives, specifically to assert that there are links to meat which warrant further investigation. The different frames used for meat alternatives are then summarised and broken down into its constituent parts as well as their combinations (Qc, section 9.4). In the final analytical section, an analysis of the mechanisms is presented through which talk about meat alternatives reproduces and/or challenges dominant culture (overall research question, section 9.5). Before concluding the chapter, an overview of the theoretical implications of the findings is given (section 9.6).

9.1 An Established and Coherent Category

For any analysis within this research project to be valid, a category of meat alternatives had to be identified as existing and somewhat established so that the theoretical framework could be applied. Looking at the historical and recent developments (see chapter 4), changes in the advertisements (see chapter 5), and the insights gained from the collected data sets, this can be confirmed. Despite its substantial changes, within the time

181 9. Meat Without Animals: An Analysis periods identified earlier, a category of meat alternatives can be considered established at some point between 2012 and 2015: The market is already visible in the mainstream as products are available in most major supermarkets. Around 2012 a number of new companies joined the market, lots of new products were developed; the market grew again after almost a decade of stagnation, and press coverage increased. These new companies brought new ideas to the category, which can be observed in narrative and advertisements, in the directions of both more meaty and less meaty products, thus diversifying the offerings. The market has also been differentiated with certified offerings such as organic or fair trade; this is among others indicated by the declining share of adverts presenting ‘novel’ product features (observation O-13 in section 5.2), which suggests that a foundational legitimacy has been granted to meat-free products as they are able to unify meat-based food practices and animal-free diets and all market niches could be filled. Market research suggests that also meat alternatives have also been increasingly appreciated and taken for granted by mainstream consumers, with more than half of respondents reporting that they had consumed meat alternatives in the six months prior to Mintel’s 2017 study. When looking specifically at the observed changes in advertising, this trend towards an establishment of the category can be confirmed: The increasing share of brand advertisements and those with primarily visual features suggests that brands consider consumer awareness as taken for granted (O-1): Brands may on the one hand highlight that they have been trusted for a long time (“25 years a food pioneer”, e.g. 2016-32) thus setting them apart from the emerging competition, and on the other hand they may aim to focus on their general visibility and prominence. Ad- vertisements focusing on either brand or product picture sideline product descriptions, suggesting that the respective advertisers believe that the target audiences of the magazines are aware of meat alternatives and their affordances in general, or their brand in particular. Furthermore, product use and function (O-4) is described less over time, suggesting that products are increasingly known and understood by target audiences. There is also an increasing use of absolute statements of worth to describe products (O-12), suggesting that valuation criteria have largely been instituted, and comparisons amongst category members are based on established and shared understandings. This is also supported by the fact that advertisements on average describe products less (O-3) which is in line with the existing literature in category studies suggesting that the category identity is increasingly taken for granted. Notably though, despite some constants, the codes to describe meat alternatives do not stabilise over time, which was suggested in the literature as well. Across the collected data sets, no major differences in attributed characteristics or product evaluations could be identified, suggesting that what was sampled as ‘meat alternatives’ is a broadly coherent set of products. Most differences can in fact be

182 9.2. Emerging Understandings of Meat Alternatives (1985–2017) explained with different focus of the respective data sets (see also the summaries in the Appendix in Tables F.3 and F.1). Thus, generally it can be concluded that there is an understanding of what a meat alternative is, what its important characteristics are, and how it is evaluated. However, in recent phases of category emergence as well as throughout the data sets, diverse levels of ‘meatiness’ were discussed. Whilst in producer texts these are occasionally played out against each other (e.g. “a lot of meat eaters have tried a lot of products in the vegetarian section, and through disappointment on the two key attributes of texture and taste, they don’t return”, interviewee Tony), amongst audiences differing preferences are usually accepted (e.g. “if you’re not anti, obviously meaty textures, this is as close as meaty as you’re gonna get”, You- Tuber Eliott). The coherence of the category is thus not impaired but instead diverse taste and texture experiences are an aspect of in-category differentiation. The established nature of the category is why it can be increasingly interpreted in new contexts. For example, trends in the advertising data point at an increasing diversity of products (O-13) but also at the highlighting of ‘protein’ and ‘plant-based’ credentials (O-5). This diversity is resembled among others in the variety of products that the interviewees successfully sell. So while the category still appears in flux, it is nonetheless established, allowing this research project to proceed.

9.2 Emerging Understandings of Meat Alternatives (1985–2017)

Besides identifying whether a category of meat alternatives can be considered established, the historical analysis of advertisements and markets (Qa) was concerned with the origins of the category and the developments it had undergone in recent decades. A few aspects are popular throughout the sampled period, including a continued as- surance that products are vegan/vegetarian (O-7), tasty (O-8), or healthy (O-5). Whilst health concerns have always been associated with both vegetarian consumers and meat alternatives (e.g. Fox & Ward, 2008; Freeland-Graves, Greninger, Graves, & Young, 1986; Rödl, 2019a), dietary credentials may further support self-identification for consumers as well as alleviate subconscious doubts about whether something that meat- like could actually be vegetarian. Explicit health and taste claims may also routinely be required when products should be made appealing; yet, an emphasis of taste credentials might indicate a constant struggle to be similar to meat and emphasise their ‘better than before’, ‘better than others’, or ‘tasty’ nature (see section 9.2.2 below). Between 2002 and 2011 this focus on health claims peaked, as natural or health credentials, and that certain contents are not present (O-5) were most frequently mentioned. This finding may indicate a wider aim to attract a health-conscious mainstream audi-

183 9. Meat Without Animals: An Analysis ence who might be worried about the processed and artificial nature of meat alternatives (see section 9.3.4 below for a deeper discussion of health and ‘naturalness’ of current meat alternatives).

9.2.1 Origins in Vegetarian and Meat Categories In fact, in the early time periods, the category antecedents–a vegetarian food category (O-9) and a meat category (O-10)—have been prominently referenced. However, advertisements refer to vegetarian food categories significantly less over time, whilst references to meat remain broadly stable. This can be interpreted as an initial attempt at sense-making to audiences, which requires the new category to be distinguished from existing ‘vegetarian foods’ as something unprecedented. This distinction might have been internalised by audiences across time, possibly because mainstream vegetarianism and meat alternatives in conventional supermarket shelves were still a new phenomenon in the 1990s and early 2000s. Broadly illustrative of this relationship to vegetarian food are adverts with titles such as “Now you’ll never be stuck for ideas, or for ages in the kitchen!” (2000-05), “you’ll never find meal times a chore again” (2001-02), or more recently “Who ever thought vegetarian food could be that tasty?” (2010-15). These examples position vegetarian food as unsuitable for everyday life or a generally dissatisfying experience. A rather constant differentiation to meat products–usually with meat as a role model to be likened or a certain ‘quality’ to be achieved–may indicate that constant emphasis of ‘being comparable to meat’ is required. This is for example emphasised in an advertisement for a meat-free burger in 1991, which proclaimed that “unless you tell them, [children] won’t realise that . . . [our products] contain no meat” (1991-02). And an advert from 2016 suggests that manufacturer VBites’ mission is “to offer a delicious and healthy alternative to ‘normal’ meat-based foods to everyone, regardless of their choice of diet.” (2016-04) In both quotes, meat and meat-based foods are highlighted as a normality and a point of reference for ‘good’ food, whilst meat eating and its rejection represent a ‘choice’. With similar statements of reference or affirmation visible in all time periods, meat–as representing a hegemonic cultural concept— remains a prominent implicit reference point in a substantial share of advertisements. Unlike predictions made in the literature, comparisons with other meat alternatives do not generally increase (O-11). References to the competition exclusively point out differences across products and are most prevalent in the time period from 2002 to 2011, when the UK market was consolidating and slowly entering the mainstream. A majority of these comparisons feature won awards, such as by company Fry’s whose products were “recommended as the best buy by The Ethical Consumer” (e.g. 2010- 22, see also O-19 on the peak in awards). These awards might grant legitimacy to

184 9.2. Emerging Understandings of Meat Alternatives (1985–2017) specific products and the evaluated product category through the names of trusted intermediaries, and they provide guidance by establishing an order of worth and setting evaluation criteria. In both functions, awards are more relevant in earlier time periods than in later ones. Non-award comparisons also introduce valuation criteria to the audience, such as a “clean conscience” (2009-12) or “probably the best vegan quarter pounder money can buy” (2009-10). This lack of comparisons across meat alternatives can also be explained with the ‘novelty’ advertised: In particular, it was observed that products were always seen as somewhat insufficiently resembling meat and that new products would be better. For example, a Grassington’s advert promotes “a NEW meat substitute which tastes as good as it looks” (2008-12), and Quorn “Best Ever Bangers,” which are “our best ever bangers, not only beating our old range in taste tests, but outperforming some other vegetarian sausages on the market too!” (2011-12) Both examples highlight taste insufficiencies in product ranges of the advertising brand and competitors alike, and seek to overcome these. Taste is not the only characteristic where meat rather than meat alternatives or vegetarian food is the main reference. For example, a Tivall advert from 2009 emphasises that other meat alternatives are insufficient, but also that meat eating is normal at barbecues and that vegetarians did not have any sufficient food options at barbecues:

Where in the past meat has always been the main affair — especially at barbecues — the fab range of veggie burgers, sausages, frankfurters, nuggets, garlic butter kievs and much, much more, means that vegetarians are now spoilt for choice! . . . Far from the sad-looking veggie sausages that we have all experienced at barbecues, which sit limp and ashen atop the flames, Tivall’s sausages and burgers positively sizzle with flavour. (2009-14)

9.2.2 Declining Differentiation from Meat A declining proportion of adverts mention animals, environment, or social issues (O-6, O-16, O-17), which all implicitly position products as different from (but importantly not oblivious to) meat. This aligns with a change in consumers or their interests, away from largely vegan/vegetarian audiences in the 1990s (who were and still are largely concerned with animals, see e.g. Beardsworth & Keil, 1992) towards a more mainstream meat reducer or lifestyle audience (which is more concerned with health, see e.g. Mintel, 2017). Also, implicit references to animal agriculture focusing on sustainability claims have nearly completely disappeared (O-6). The messages of many advertisements align with a struggle of vegetarians to be accepted in what they cook and eat: They are left out from shared dining experiences

185 9. Meat Without Animals: An Analysis due to their choice of diet (Beardsworth & Keil, 1992), and the promoted products promise reconciliation (which also interviewee Seth suggested their Tofurkydid in the 1990s for thanksgiving celebrations). Specifically, people-focused advertisements (O- 17) portrayed a desire but lack of knowledge, skill, time, or products to appeal to both meat eaters and vegetarians at the same time. Among others, adverts acknowledged that children will “be the first to tell you if [burgers are] anything less than 100% pure perfection” (1991-02), or promised “a real family favourite” (a ‘Toadin the Hole’, 1993- 03) in meat-free or “to please everybody at meal times” (2000-01).1 In all cases, the adverts suggest that the displayed meat alternatives could bring peace to the dining table and promise relief from everyday struggles in families. In these advertised value statements (but also in notions of progress, see O-13) and people-focused advertisements, products are presented as solutions to technical, social, or moral problems. Since a focus on values is over time replaced by emphasis on everyday experiences and special occasions such as barbecues or christmas (O-18, see the following section), advertisements then focus on what is conveniently possible: Eating meat-free ‘meat’ as if it was normal. Thus, instead of highlighting and then solving restrictions, advertisements promise affordances in the lives of consumers. Fol- lowing insights from the category studies literature, texts originating after a category has emerged might presume that consumers are aware that meat alternatives can solve health, ethical, or environmental issues. Thus, these aspects become internalised into the category identity. In line with suggestions made in the conceptual framework, links to a hegemonic category (i.e. meat) persist, whilst references to other antecedent categories disappear (i.e. ‘vegetarian foods’). This supports a view that a prospectively successful category needs to reduce its ‘friction’ to mainstream views, which is reflected in declining shares of ethical, moral, sustainable, or conflict-resolving advertisements. The category further matures through an increased shared understanding, which also builds on prior legitimation granted through awards and within-category differentiation.

9.2.3 From Characteristics to Meanings In place of advertising themes focusing on values and people, occasions and everyday life situations are highlighted increasingly more often after 2012 (O-18). Generally, these advertisements portray existing meat-free food items as unsuitable or inadequate for a concrete purpose or context as they may not be tasty, traditional, accepted, normal, or meat-like enough. Yet, instead of emphasising product characteristics such as good taste or a perfect sensory experience, advertisements establish a social context in

1In fact, many of these adverts depict (and thus target) women or mothers as the supposed chef of a family as well as the demographic group most likely to be vegetarian.

186 9.2. Emerging Understandings of Meat Alternatives (1985–2017) which products should be evaluated. They highlight the occasions and opportunities arising when it is more or less acceptable to eat meat-free, as opposed to the ‘chores’ of everyday life and the engagement with the rather new phenomenon of vegetarianism as discussed above. So instead of highlighting difficulties, these adverts highlight affordances, which builds on and reproduce a ‘normality’ of meat eating. All meanings associated with these meat-centred occasions are thus retained, as advertisements assert that similarity ought to be contextual and social experience more important than a perfect taste experience (see also Elzerman et al., 2011 who find that meat alternatives are evaluated more positively in a meal context). This integration into dominant culture can be understood as reducing ‘friction’ with mainstream views. Differences between occasions can nonetheless be observed, for example between christmas and barbecues, the most prominently advertised occasions: While contexts of christmas and celebratory meals can be regularly identified across all time periods, barbecues have been featured more frequently in recent time periods. This might indicate an increasing focus on men: A smaller target group than women, but also a yet untapped potential for advertisers. Exemplary of this change is an advertisement from 2010 to “discover BBQ meatfree” by Goodlife (2010-10), which on their illustration pictured a barbecue and a woman with a carrot on their fork. In contrast to this, a Fry’s advertisement for somewhat meat-like burgers suggests to “set your plants on fire” (2016-16) and appears to target men as the (self-)declared masters of barbecues and fire (see e.g. Nath, 2011). Another explanation for this increase in barbecue-centred advertising is the respective framing of the adverts themselves: Re- cent barbecue-related advertisements focus on the social context of barbecuing and sharing food, whereas christmas adverts seem to focus on the individual as an outlier in their social context. At christmas, meat alternatives thus enable individuals to participate in festivities without having to rely on anyone’s inclination to cook meat-free. Meat alternatives are then coping mechanism in a meat-dominated society. In contrast, the barbecue advertisements seem to focus rather on the collective experience of eating and sharing in a society where it is accepted or possibly even normal not to eat meat2—which of course required meat-free diets to become wide-spread and normal. However, in both cases, the social gathering and its meaning is more prominent than the foodstuff itself and its sensory characteristics.

2To underline this claim, a few examples of recent adverts shall be mentioned. For christmas, roasts are advertised with statements such as “When the unexpected guests drop by . . .” (2016-30); or “Christmas [is] . . . also a time that can be tricky for vegetarians. If you’re eating at someone else’s house . . . Or if you’re doing the cooking, it can be tricky to know what to serve in place of meat” (2015-34). On the other hand, for barbecues, we can read of products “. . . ideal for outside buffets and are certain to get your guests gathering around the serving platter” (2016-12); or “In reality, barbecue food has never been so eclectic and varied . . . [and] many meat-eaters wouldn’t even question the meatless burger they’re feasting on” (2015-20).

187 9. Meat Without Animals: An Analysis

9.2.4 Emancipation from Meat As elaborated above, the focus on use cases allowed matching meat alternatives with their model. Advertisers could then frame products as distinctly flavoured but still useable in the same ways as their model. This is increasingly apparent in the last two time periods where advertisements for meat alternatives can be considered reflective of wider food trends, including an increased focus on plant-based foods, as well as on protein and health credentials (O-5). However, with the available data it cannot be ascertained whether the meat alternatives category and its advertisements (co-)produced this trend or whether they are merely responding to external developments. Since around 2012, some companies have produced wholefoods, vegetable-based, or pulse-based meat alternatives; so even though references to a vegetarian food category have disappeared, an increasing amount of products is created from “just a riot of great-tasting veg, pulses, herbs, and spice” (2017-02) or “fresh garden vegetables harvested from local farmers” (e.g. 2014-22). Thereby its plant-based roots are reclaimed and genuinely new tastes and textures are made possible. Whilst similar products had already existed in earlier time periods, those new products were advertised as modern and exciting. This image of progress is aided by an increasing focus on protein credentials more broadly, which allows meat alternatives to be seen as equivalent to meat in its provision of protein. This changing perspective on meat alternatives indicates that meat alternatives are increasingly less understood as a copy of meat, but instead as a ‘swap’ in a level relationship where all foods are protein. A focus on plant-based credentials gives meat alternatives a supposedly clean image so that meat alternatives further become a positive affordance–instead of discrediting meat as was done in earlier time periods, see section 9.2.2 above. This again reduces ‘friction’ to dominant meat culture, with similar effects to what has been discussed concerning values above. Possibly in order to broaden understandings of meat alternatives, they firstly required an established and appreciated category meaning (see also section 9.6.3) and acceptance for meat-free eating in general.

9.3 The Meat in Alternatives: Current Understandings

Besides establishing the coherence of the meat alternatives category using the multiple data sets (as discussed in section 9.1), the analysis of the category identity (Qb) was focused on whether and how meat is relevant for the identity of the meat alternatives category. Specifically, the evaluation follows the ways in which meat is signposted or referred to in the description, evaluation, and labelling of meat alternatives. While other aspects of these issues are still interesting in their own right, they are ignored in

188 9.3. The Meat in Alternatives: Current Understandings

the following as they do not contribute to this thesis. An overview of all data sets in relation to the sub-questions for research question Qb can be found in Appendix F.1.

9.3.1 Format Flexibility Most labels for the products reviewed in the data follow linguistic references to meat, including terms like meat alternatives, meat swaps, meat substitutes, meat-free foods, or meat-related food formats.3 Interviewee Hilary clarifies, why such labelling is done: “The steak is a steak because that’s what it is, [and] the burger is a burger because that’s what it is.” Besides product labelling on packaging–which is hugely significant for audience evaluation (see e.g. Piqueras-Fiszman & Spence, 2015)—, product appearance is a heuristic frequently used by audiences and producers alike to establish a specific similarity between meat alternatives and their model counterparts: the food format. The format is generally associated to specific meanings and occasions familiar to the audiences, and evoke expectations of specific sensory experiences. They promise to audiences the same or similar emotions and experiences, on a sensory, functional, and contextual level. For example, burgers and sausages are understood as convenient to cook and eat–which is also related to specific contexts in which they are best to eat. Audiences appear to know how to prepare these foods, and sometimes know this even better than the product packaging, such as Mia who reports that “online people have been saying to cook it for a bit longer than it says on the back, because it’s cooked on the outside, but not on the inside.” Formats also relate to knowledge about the meal contexts in which products are usually eaten. For example, Interviewee Tony and YouTuber Alice both claim that Bolognese made with meat-free mince is hardly distinguishable form one made with meat-based mince. The only data set where formats feature relatively less often are the advertisements: Here, food formats become a means to an end rather than the reverse, as experiences are directly highlighted through occasions or purposes, which for example include barbecues, sandwich toppings, or finger foods. Across data sets, formats furthermore appear to be associated with perceived normality and food traditions (and thus identity), with occasions and functions of the product, and with versatility and convenience. These all build on existing beliefs held in relation to meat, such as that it is convenient, easy to prepare, or versatile. Notably, in most cases these affordances of meat formats are understood as self-explanatory and not explicitly highlighted. The only notable exception is interviewee Steve, who emphasises that their seitan products cover all these aspects that meat affords; this is

3This prominence should not be too surprising, as meats or meat formats served as inclusion criterion for much of the data, and I used the term meat alternatives to approach my interviewees.

189 9. Meat Without Animals: An Analysis possibly done to facilitate sense-making of their product which they try to not directly label as a meat alternative. Particularly in the advertisements but also in the other data sets, these affordances are highlighted as a counterpoint to stereotypical meat-free food, which is not considered convenient or versatile, and not adequate for traditional occasions, such as barbecues or christmas4. As such, meat alternatives can–through their seamless integration into existing meat-based practices–alleviate the perceived disruption to individual lives caused when meat consumption is reduced on a singular or regular basis. In most interviews the range of legitimate formats for meat alternatives is equivalent to and bounded by existing meat formats (but nothing beyond this); only interviewee Eddy has an explicitly contrary perspective as they suggest that doing “just what’s the meat market doing . . . isn’t a sustainable, sort of approach,” yet without offering further visions. Once asserted through labelling or appearance, meat formats are however quite flexible. For example, vegetable-based sausages or seitan steaks are as legitimate as their more meaty counterparts. How far format-based labels can extend connotations of meat is illustrated when vegetables or mushrooms are labelled like meatstuffs through a bit of marinade, spices, or coating: This covers for example “BBQ cauliflower wings” (tweet W1; that become “a perfect party nibble with a bit of spice”, tweet N8), “cauliflower steaks” (e.g. tweet N121), or “marinated mushroom bacon” (YouTuber Phoebe). This flexibility in formats disregards the tastes and textures of meat, which means that meat alternatives are primarily understood through the use of meat formats and their social and cultural affordances: They transfer experiences and expectations, and are found to relate to food-related identities and practices (see also Nath & Prideaux, 2011 who suggest that shared occasions are the reason why consumers adopt meat alternatives). Formats can therefore function regardless of the source material. This vast interpretive flexibility may however lead to different interpretations and thus potential conflicts: On the one hand, products may be ‘mislabelled’ and evaluated differently by audiences; for example, YouTuber Jessica finds that a vegetarian sausage tastes more like a good hot dog instead of a bad sausage; and YouTuber Alice suggests that their sampled meat-free bacon “is more like ham, with the taste of, smoke and bacon”. On the other hand audience expectations may be much more concrete than what the label suggests, such as for YouTuber Robert who in their quest to find the best sausage continuously compares foods to “pork sausage”. A notable exception to the compliance of formats are a number of adverts that do

4Note however, that in the advertisements this attitude towards meat alternatives had been expressed more explicitly and more often in previous time periods, when meat alternatives needed to be established as foodstuffs different from vegetarian food (see again section 9.2.1).

190 9.3. The Meat in Alternatives: Current Understandings not label products explicitly and instead propose somewhat generic food, for example using statements like “good food swoops in” (2017-02), or “everyday food, fit for every occasion” (2014-25). This allows to position the foodstuff as part of wider food culture, so that manufacturers can propose new ingredients or format combinations without restrictions and stigmas of other labels. In a few instances, labelling products as ‘protein’ is also attempted, which reduces both meat and meat alternatives to a substantial protein supply to the body. Meat alternatives are here positioned as plant protein, and therefore allow for an easier rhetorical substitution, which is exemplified by phrases such as “protein alternatives” (interviewee Seth), or “veggie alternatives to meat/protein sources” (tweet N84; see again section 9.2.4; see also interviewees Hilary and Tony).

9.3.2 Meat as a Reference The relevance of meat as a reference to understandings of meat alternatives is clearest in relation to the different food evaluations performed across the data set (Qb2). At large, they can be summarised into evaluations by (a) comparison, (b) subjective appeal, or (c) description. Comparisons and subjective appeal were most prominent and usually used positively (e.g. “looks like a real burger”, “tastes like meat”, “smells amazing”) rather than negatively (e.g. “but it doesn’t taste like a pork sausage to me”, Robert; “it looks disgusting”, Charlotte’s friend). The description of products was less frequent, although interviewees Hilary and Seth had suggested that this would be preferable, because “using the language of flavour and texture, you can get a long way without having to use many [meat-related] words” (interviewee Hilary). However, established ideas and vocabulary related to meat may still be used to evaluate meat alternatives as food-related vocabulary is reflected through experiences in evaluating meatstuffs and cannot be detached from it. Furthermore, given the frequent descriptions in tweets or videos, such as “Tangy BBQ Cauliflower Wings” (tweet W1), or “it’s got this like, Spanish, kind of scent” (YouTuber Georgia), it is unclear whether this ‘language of flavour and texture’ is sufficiently refined in consumer’s minds. Conversely, a lack of relatable language might obstruct engagement with products; this is exemplified for example by YouTuberGeorgia who acknowledges that they “have never had a Chorizo before in my life” before sampling a meat-free version, and interviewee Steve points out that “very few people have heard of [seitan]” making it more difficult to explain to prospective customers. Previous literature has found that foodstuffs are often evaluated on the basis of taste or expectations thereof (e.g. Zald, 2009; Enax, Krajbich, & Weber, 2016). Thus, connotations of situated meats–and less often specific meat formats or other meat alternatives–is a frequent component to most product evaluations: For example a fast

191 9. Meat Without Animals: An Analysis food joint is making “a vegan version of what they’re famous for” (interviewee Dawn), an advertisement suggests that “being meat-free shouldn’t mean compromising on great festive food” (2016-30), or a twitter user suggests to “try it out n u might be surprised to find you like it, i know i was” (N124). Even if expectations were triggered by other people via word of mouth in real life or on social media (as was the case for a number of YouTubers), the framing of these expectations might still be related to established understandings of meat. Notably, there are different aspects for which meat is a positive or negative reference. Many comparisons establish similarities between meat and meat alternatives, such as that a foodstuff has the same taste as meat, that it would cook like normal sausages, or that it can deliver the same experience in a Bolognese sauce. Conversely, when contrasts to meat are drawn, they often include nutritional differences and a suggested health benefit of reducing or eradicating individual meat consumption; animal ethics and environmental aspects are also mentioned, and occasionally meat alternatives are said to not have the same taste or texture as meat. When discussed products are compared to other meat alternatives, the focus tends to be on sensory attributes, on convenience, and sometimes on animal rights/welfare topics. Particularly in the videos and the adverts, many comparisons are also based on the respective food format, so that a burger-like foodstuff is compared with other ‘burgers’ regardless of whether they are meat-free or not (cf. Hoek et al., 2011 who find that ‘processed’ meat alternatives are evaluated by format not by animal origin). These comparisons are often mediated by the experiences and expectations that individuals relate to a specific food format (see the previous section), so that meat remains at least an implicit baseline for what may be considered a ‘good’ meat alternative.

9.3.3 Signposting Deviance Generally, when discussed positively (which was the case in most data) descriptions of ‘tasty’ or ‘delicious’ meat alternatives–as judged by smell, taste, or appearance— usually imply that something tastes meat-like (e.g. interviewees Tony or Geert-Jan; YouTubers Jessica or Robert). But as some manufacturers have popularised non- meaty alternatives, these ‘alternative alternatives’ are sought to be dissociated from understandings of meat by almost everyone. On the disproving end, terms such as “planty” (YouTuberLydia) or “vegetarian” (e.g. interviewee Tony) are sometimes used to describe an unappealing taste to the speakers. Interviewee Tony even suggests that “people have made a conscious decision to be a vegan or a vegetarian, invariably, they’ll accept it, that the alternative won’t taste that good.” However, meat alternatives may also be ‘good’ or ‘tasty’ without being like meat, as interviewees Hilary or Rachel, or YouTuber Lucy argue. These products exhibit a

192 9.3. The Meat in Alternatives: Current Understandings careful negotiation of value precisely because only certain characteristics of meat are rejected while others are reproduced. Interviewee Steve, for example, relies on understandings of meat alternatives and their cultural, social, and functional connotations to explain their seitan products. And interviewee Rachel points out the skin around their vegetable-based sausages which adds “a little bit of that springiness, like traditional meat sausages”—although they otherwise propose that their sausage is good because it does not have meaty tastes or textures. Similarly, YouTuber Lucy proclaims that a vegan diet “needs to be just as convenient as a normal [meat-based] diet,” and therefore they opt for foodstuffs which have the same format and the same affordances but otherwise taste and feel very different. When meat alternatives are different from meat, they are frequently signposted as being nice despite not tasting like meat. For example, YouTuber Robert tries to identify the best “pork sausage” and finds that a specific sausage does not taste like meat, “but I really like it, it’s nice . . . I like the texture, and . . . the flavour is nice too.” So whilst the label ‘sausage’ does not match Robert’s expectation of ‘sausage’, these are still evaluated positively. Similarly, upon trying a different and more meaty burger than what they had known before, YouTuber Emily proclaims that “I do still like the Linda McCartney burgers as their own thing, like they don’t taste exactly like meat, they have a nice, their own kind of taste.” The good taste of those meat alternatives which do not taste like meat is thus explicitly narrated and thereby legitimised. An exemption from meat norms is also achieved by an advertisement for a vegetable and pulse-based burger suggesting “a new take on a classic” (2017-18). This signposting of deviation exempts expectations of meatiness, allowing established meat formats to be re-imagined. While rhetorically such ‘new take’ can therefore escape sensory evaluations of meat, the ‘classic’ affordances imposed by the food format persist across all deviations. So even in those cases where meat alternatives are not supposed to be meaty, meat remains an important reference. This includes YouTuber Lucy’s reference to “actual veggie burgers” as opposed to more vegetable-based ones, interviewee Steve’s seitan, or the skin around interviewee Rachel’s vegetable sausages as discussed above. Unless taste experiences conform with expectations, legitimacy as a ‘good’ foodstuff is thus not granted to meat alternatives. Interviewee Tony generalises this attitude arguing that “meat reducer[s’] . . . request from protein is: I want texture, I want to bite, I want to taste it . . . — and they don’t get that with [most meat alternatives].” Conversely, it is to be expected that animal-based meats can be evaluated negatively because they taste unexpected, but they would not lose their legitimacy. These observations suggest that meat in general and meat formats in particular behave like category prototypes that legitimise category members–with the crucial difference however, that meat is not part of the meat alternatives category (see section 9.6.2).

193 9. Meat Without Animals: An Analysis

9.3.4 Opposing Meat Whilst the previous sections were all concerned with how meat is navigated as a role model, in this section opposition to meat in talk about meat alternatives is reviewed. Specifically, this covers health and values, which were originally identified as product characteristics, but upon closer investigation, these appear to be associated to meat- free diets more than to specific products. For example, in the taste test videos, these aspects are only mentioned in the introduction or at the end, when YouTubers talk about why they eat meat-free themselves or why others should eat less meat. A similar pattern can be observed in the tweets including their focus on swaps, choice, and meat reduction. Conversely, the interviewees associate dietary choice to their products, and thus blur the boundaries between products and diets. In the literature, ‘health’ is identified as the main reason for people to reduce meat consumption, while ethical or environmental arguments to eliminate meat consumption (e.g. Mintel, 2017). As briefly mentioned in section 9.2, one important aspect of health claims are so called ‘clean labels’, which suggest in a ‘good for you/better for you’ scheme (interviewee Daniel) that ‘bad’ or unwanted contents are not contained in a foodstuff (Asioli et al., 2017) which includes cholesterol, fat, salt, preservatives, or ‘chemicals’. This requires the audience to know that meat (or meat alternatives) might be unhealthy because of these contents. In the case of meat alternatives, this is extended to what is commonly included in meat, such as antibiotics, growth hormones, or animals themselves, which YouTuber Elliot delightfully observes when discussing the ‘No Bull Burger’: “There is no bull in this, absolutely no bull, that is definitely not a bull burger, there’s no bull-crap in this burger.” Here, Elliot marks both the animal and the supposedly negative contents of animals as absent. Simil- arly, an advertisement promoting “a non GM product with no hormones, antibiotics or chemicals” (2016-05 and 2016-10) clearly marks non-meatiness through an opposition to conventional animal agriculture, but not in opposition to meat. Sexton (2016) calls this absence of meat and its potential other contents in meat alternatives “the non-stuff that distinguishes this product from conventional meat” (p. 70) and thus promotes health through alleged differences. Related to health and also concerned with ingredients are debates about what constitutes a ‘natural’ product, for example in relation to food safety, additives, or the level of processing. The meaning of naturalness is variegated across the data, with advertisements asserting naturalness as being a ‘clean label’ product, or interviewees suggesting the vegetable origin of products (Rachel), low- or unprocessed foods (Daniel, Steve), containing no additives and thus no added iron (Tony), or in reverse having added iron and vitamins to be a natural replacement for meat (Geert-Jan). Controlling the idea of naturalness is thus more important than “controlling the naturalness” (inter-

194 9.4. Breaking Down Meat Alternatives viewee Tony) itself. For example, some interviewees point out how unnatural meat production processes are, and others attack market leader Quorn as unnatural for its use of fermentation vessels to grow fungi–whereas conventional meat alternatives are based on chemically isolating and then emulsifying and extruding proteins from plants. Animals, ethics, and slaughter are often associated with each other and with animal agriculture, whilst environmental issues remain largely unspecified: Those few texts that mention the environment fall short of highlighting explicit connections between animals, meat, and planetary health, usually attributing effects to meat as a symptom and not animal agriculture as a cause. Twitter users claimed that “our planet needs less meat and more plants!” (W51), or that “swap[ping] out meat for veggies . . . will . . . save tonnes of water, crops, and energy.” Furthermore, some manufacturers refrained from ethical or environmental debates as they are understood as consumer decisions (and thus a business case, e.g. Tony, Geert-Jan, a discussion on that can be found in section 10.2.4). Although value statements might be implicit in the understanding of a meat alternatives category, they are not explicitly mentioned, so that further audience knowledge and understanding is presumed or required. Non- etheless, they participate in framing meat alternatives as they propose an advantage of a particular diet or food choice over another, instead of being used as an attribute of a specific foodstuff. Thereby, meat alternatives are considered as a solution or practical help to a concrete moral dilemma.

9.4 Breaking Down Meat Alternatives

The research project set out to take apart meat alternatives in order to investigate their relations to dominant culture. Building on the insights gained across the data sets (specifically questions Qc1 and Qc2, see also Appendix F.2 for summary tables) on how meat alternatives are discussed, valued, and signified, three major frames to discuss meat alternatives and meat could be identified: A consumption frame, a cultural frame, and a political frame. These will firstly be described independently, before afterwards combinations of these are discussed; their main their features are summarised in Table 9.1.

Table 9.1: Summary of the identified frames and their function (Qc).

Frame Consumption Cultural Political

Focus hedonic product established practices scrutinisation and dietary attributes change

195 9. Meat Without Animals: An Analysis

Frame Consumption Cultural Political

Meat as. . . a usually tasty and a socio-cultural deriving from animals enjoyable consumable entity as bearer for and thus a social and traditions and technical entity culture responsible for externalitives Meat recreate and offer diversity able to function in (indirectly: eradicate Alternatives to meaty attributes place of meat issues) as. . . Purpose diversity continuity disruption

Consumption frames focus on individual hedonic consumption experiences and preferences. This includes whether one likes meat or does not like meat, and how an individual understands their own health and diet. A choice of products in this frame is thus individual and usually accepted as an individual choice. Meat and meat alternatives in this context are framed as consumables, which can be enjoyed or not enjoyed for various reasons, and that have some relation to the individual well-being. In this frame, meat alternatives are valued on the basis of individual preferences and expectations, which is most prominently attributed to taste. Notably, an implicit relation to meatiness remains as deviance from animal meat is explicitly signposted (see section 9.3.3). Cultural frames focus on established collective and individual consumption patterns in form of rituals, habits, practices, or traditions. Here, meat is a cultural entity that is valuable through its embedding into individual and collective identities. Con- versely, meat alternatives through the equivalence in format are framed as being able to occupy the same or at least a sufficiently similar spot that allows a seamless substitution in either direction: They are positioned as convenient, normal, and straight- forward, and often in opposition to ‘vegetarian food’ (see section 9.2.1). This frame is not bound to an immediate dining experience but more generally to its associations and the multitude of ways in which the foodstuffs can be used. Cultural frames thus focus on continuity, stability, and established patterns as meat alternatives allow continued participation in collective meat-centred rituals and the preservation of individual practices regardless of dietary choice. Political frames highlight the wider, usually negative effects of food consumption–which however are often unspoken or implicit (see section 9.3.4) In case of meat, these include animal rights/welfare, environmental and ecological issues, and concerns about public health (as opposed to individual positive health impacts when switching to meat alternatives). In this frame, meat is understood as deriving

196 9.4. Breaking Down Meat Alternatives from animals and thus as an entity that is embedded into wider production and consumption systems; therefore, the detrimental effects of excessive production and consumption of meat become visible. Part of the political frame are thus instances where audiences are sought to be urged, motivated, asked, or convinced to reduce their consumption of animals or stop it completely. Thus, most criticisms voiced in the political frame include animal welfare, environmental, and health concerns relating to the dominant production measures of (industrial) animal agriculture— which includes rearing, slaughtering, and manufacturing, as well as the effects thereof. However, in most cases where political frames are evoked, the underlying assumption that it is acceptable to eat animals is not questioned. Thus, meat alternatives are arguably not part of a political frame, but instead provide a material solution or a tool for seamless dietary change. This dietary change then can eradicate the issues with meat, because meat alternatives are able to align the identified negative impacts with the positive aspects of the consumption and cultural frames. By itself, the political frame promotes change and disruption of established systems. However, this frame and thus also the category identity is dependent on these established systems, as it can be argued that without the destructive capacity of intensive animal agriculture which enable and require the political frame, the category would be considered a duplicate of meat, which may not be able to match its model. The political frame therefore extends the meaning of the category to be a resolution of individual cognitive tensions concerning meat consumption (e.g. Buttlar & Walther, 2018; Rothgerber, 2014). The insights which these frames offer match previous findings about meat alternatives: They are foodstuffs with which an easy transition to a flesh-free diet can be achieved, as they offer the same or similar cultural embedding and sensory experiences without ‘guilt’ (e.g. Nath & Prideaux, 2011). Depending on the text, one, two, or all three frames may be employed to motivate and discuss meat alternatives, which need to be actualised in framing of texts and actions. Interactions function in all combinations and may differ on temporal or causal grounds, which are illustrated in Fig- ure 9.1, and discussed in the following paragraphs. Rhetorical strategies or patterns that could be found across many data sets which engage with, build on, or–as will be argued–inadvertently reproduce dominant meat culture are described in detail in section 9.5 below. The overlap between consumption and cultural frames is a ‘business as usual’ context; hereby, all motivating factors except for potential individual health benefits remain unspoken and thus rely on audiences’ previous knowledge and willingness to substitute meat alternatives for something different. This covers for example the approach of interviewee Tony’s business to “become part of standard part of a weekly shop for people who love meat and aspire to be healthy.” However, contrary to a fixation on meaty tastes, the consumption frame enables a narrative of diversity so

197 9. Meat Without Animals: An Analysis

MPTION NSU CO sensory and other product going beyond attributes meat norms healthy meat, and formats business as usual, enabling diversity meat alternatives as ‘better meat’ criticisms of meat established practices P reclaiming O L meat words, IT IC re-imagining A L L meat RA TU CUL

Figure 9.1: Overview of the combinations of the three frames and the observed patterns and rhetorical strategies related to these. that products may have different tastes and textures to appeal to a variety of audiences. Particularly, when people emphasise that not everyone likes meat, this is highlighted and explicitly scrutinises some norms established by meat culture, such as YouTuber Elliot who says that the burger “would probably put a lot of people that obviously don’t like the taste and texture of meat off.” However, adherence to and focus on the cultural norms established through formats and occasions may internalise different taste and texture experience into a general understanding of meat eating. This is for example emphasised by interviewee Rachel’s attempt to provide different taste and texture experiences to meat, but still opting a casing that “gives a little bit of that springiness, like traditional meat sausages with collagen.” The combination of cultural and political frames allows to re-imagine the concept of meat onto new foodstuffs–regardless of their tastes or textures–as long as they fulfil similar functions as meat. This covers for example seitan which interviewee Steve emphasises is “very easy to use, [as] you can eat it straight from the pack, you can cook it, fry it, stick it in a sandwich, or whatever else.” This combination of frames furthermore sets the scene to reclaim meat words as intended by interviewee Hilary, as cultural ideas and the formats they manifest in are intended to be (re-)ap- propriated from associations with animals. As the functions or linguistic contexts of meat are continuously highlighted and re-imagined into different products and contexts, meaty or non-meaty tastes may become irrelevant. This allows to essentially transcend meat and challenge dominant meat culture, if they manage to extend

198 9.4. Breaking Down Meat Alternatives

the normality of practices and discourses to other non-‘meaty’ ways of eating. Con- versely, these aspects might reproduce dominant meat culture if carnist exclusivity and normality is perpetuated (see section 9.5 below). The overlap between consumption and political frames, is probably the rarest, as it requires going beyond the norms and formats of meat, thus breaking any underlying cultural narrative and possibly removing any grounds to call it a meat alternative. Examples are YouTuber Phoebe’s “marinated mushroom bacon” that retains the mushroom’s appearance but has bacon flavour through a marinade; or a twitter user’s suggestion to “add tamari when cooking to give it [sauces with meat-free mince or lentils] a slightly ‘beefy’ flavour” (N103) which highlights meaty taste but does not require specific meat-related practices. Similarly, YouTuber Mia highlights that nostalgic feelings may be evoked by the sensory experience of eating or smelling a foodstuff, and not exclusively by the modes of consumption it is embedded in. High- lighting the importance of going beyond meat formats, interviewee Eddy suggests that it may not be a sustainable strategy for the industry to only replicate meat products. All of these highlight that meaty taste experiences are possible regardless of the format of meat, as they rather ignore cultural contexts. Thus, these examples manage to also challenge meat consumption and its association between taste and texture experiences, established formats and consumption contexts, and animals. Highlighting the boundaries of this overlap between political and consumption frames, a suggestion that “lentils make for a great alternative to meats like beef, pork and chicken” (tweet N129) is not an example of meat alternatives as defined in this research project–although it also would likely not reproduce dominant culture in the understanding taken here. A combination of all frames merges the restrictions of the cultural frame, the newly gained allowances and health credentials of the consumption frame, and the changes demanded by the political frame. The similar cultural and sensory experience has led some texts to argue that the consumption of meat alternatives is a moral imperative, as will be described in the following section. Through the diversity of the consumption frame, some space is created for different flavours and textures to be developed within the confinements of the same cultural concepts, which is for example emphasised by Gosh’s advert for “a new take on a classic” (2017-18), with which they advertise a vegetable-based burger. Thus, meat alternatives can be an anchor in times of change, simply an alternative to meat, or even ‘better meat’, because its ethical, environmental, and health impacts are seemingly eradicated, while becoming more inclusive.

199 9. Meat Without Animals: An Analysis

9.5 The Cultural Politics of Meat Alternatives: Mechanisms of Challenge and Reproduction

In order to illustrate how meat alternatives are used to reproduce and challenge dominant meat culture–or inadvertently do so through their framing in relation to the symbolic power of dominant meat culture–, in this section, mechanisms for the challenge and reproduction of dominant culture are discussed. Most of these mechanisms place meat alternatives in a ‘business as usual’ scenario based on a combination of cultural and consumption frame, or as ‘better meat’ through a combination of all three frames. Thus, a majority of mechanisms lead to a reproduction of dominant meat culture. While I do not claim that this overview is exhaustive, it is illustrative of the diversity of mechanisms in which dominant culture is challenged and reproduced. A summary of these mechanisms is reproduced in Table 9.2.

Table 9.2: Summary of the frames and mechanisms with which meat alternatives reproduce and/or challenge dominant meat culture (Qc).

Effect Mechanisms

challenging dissociating meat from animals-based features – reclaiming language and terminology – detaching sensory or cultural elements respectively from the symbolic and sensory norms of ‘meat’ reproducing positioning meat alternatives as similar to meat as a normal, tasty, and healthy foodstuff – invisibilise or stigmatise other ways of eating, for example making them not normal, pleasurable, adequate, or convenient – sensory experiences that are different from the norms established by the dominant meat culture are explicitly emphasised as such – democraticising animal meat reduction by building on tacit and practical knowledge, thereby reducing the need for knowledge about other ways of eating both meat alternatives as ‘better meat’, transcending established understandings of meat – meat alternatives as an indvidual and conditional moral obligation as meat needs to be reduced

200 9.5. The Cultural Politics of Meat Alternatives: Mechanisms of Challenge and Reproduction 9.5.1 Stigmatising and Invisibilising Deviant Ways of Eating As discussed in the literature review, particularly an emergent category needs to be differentiated from its competitors. Especially in the advertisements as a direct means to legitimise and make palatable meat alternatives to audiences, common (mis-)understandings of meat-free diets were reiterated. Advertisements for example–as briefly discussed in section 9.2.1—highlighted that meat-free evening meals were difficult and cumbersome, celebrations without roasts inadequate, and much vegetarian food not tasty (e.g. “being meat-free shouldn’t mean compromising on great festive food”, 2016-30). This mechanism can also be found in other texts, such as when YouTuber Lucy proclaims that a vegan diet “needs to be just as convenient as a normal diet;” or when a Twitter user suggests that “there are so many alternatives and substitutes, you won’t feel like you’re missing out at all” (tweet N67). These examples perpetuate stigmatisation related to ways of eating that do not conform to the ways prescribed by dominant meat culture–because these are not normal and by extension not nice, convenient, or similar. Similarly, food without ‘meat’5 may also be invisibilised: Interviewee Tony reports that some of their customers make meat-free Bolognese or lasagne, and “don’t tell the rest of the family, but they satisfy the vegetarian,” thereby hiding that people may make different choices in the first place. These ways of engaging with meat alternatives therefore perpetuate a normality of ‘meat’-based food culture by delegitimising alternatives. This might bring about barriers for further behaviour change to a genuinely ‘meat’-free diet, as tacit or practical knowledge for other ways of eating is still missing (cf. Nettleton & Green, 2014).

9.5.2 Knowing Meat and Trusting Meat How important knowledge is, can be seen across the data. For example, YouTuber Pippa decided to try meat-free sausages “to look at other ways to get to some protein into my diet without having to go back to eating meat.” And virtually every You- Tuber knows how to cook their meat alternatives, and might even ignore the cooking instructions on the packaging because they prefer different means or directions of preparing such foodstuff. People know how to cook Bolognese regardless of whether it is meat-free or meat-based mince (e.g. interviewee Tony), and they know that ‘wings’ might be “a perfect party nibble with a bit of spice” (tweet N72). Meat alternatives are trusted to fulfil the same societal functions, they are accepted on the barbecue (e.g. YouTuber Mia), they are accepted as “great festive food” (e.g. advertisement 2016-

5When I put ‘meat’ into tick marks, I refer to non-human animal meat and its animal-free alternatives, to symbolise their shared meaning systems as established throughout this thesis.

201 9. Meat Without Animals: An Analysis

30), and “in [a] lot of ways [can be] a peacemaker in many families, [for example] because people will serve Tofurkys and Turkeys on the same table” (interviewee Seth). As has been argued previously (see section 9.3.1, and e.g. Nath & Prideaux, 2011), the tacit and practical knowledge that meat alternatives afford and tap into, quite literally means that people do not have to worry about their food choices: These choices are risk-free, because their properties are known, the skills are available, and the foods will always be appropriate and usually accepted. This extends not only the practices and affordances of meat onto meat alternatives as argued before, but also re-utilises established knowledge–this highlights the cultural generativity of the meat category, or more specifically its formats (see also section 9.6.3). Because knowledge is shared and transferable, meat alternatives make the process of replacing meat democratic, because they are easily accessible, nearly universally appealing, and allow for an instantaneous switch since existing knowledge can be meaningfully and efficiently utilised, without requiring any extra resources beyond adjusted shopping choices. This legit- imises meat alternatives on the sole basis that they reproduce meat formats and the cultural and social aspects embodied therein–they thereby reproduce and entrench these elements.

9.5.3 Counting Consumption and Creating Moral Obligations Particularly in the tweets–possibly reinforced by the context of the theme weeks as encouraging different meals–the capacity of instantaneous substitution was used to motivate singular or regular action, which is ‘countable’. For example twitter users report that “so far during #WorldMeatFreeWeek the week I had 14 meat free meals” (W5), others suggest to their audiences to “consider going meatless one day a week” (W44), or that “even if you just stop eating AS MUCH meat instead of completely going veggie it helps the animals and the environment and your general health!” (N163) When meat consumption can be counted, there is no dichotomy between eating meat and not eating meat, but substitution becomes a gradual process (assuming there is a target) or simply a scale. The idea that something needs to be reduced, signifies a compromise for the sake of animals, environment, or individual health, because otherwise no reduction would be necessary. Arguing for such compromise, YouTuber Phoebe suggests that: “If you like the flavour and the texture, I don’t see what harm it is to have the vegan version.” On the one hand, this argument functions on a case-by-case basis, and on the other hand it requires likeability, but not necessarily equivalence. This makes swaps opportunistic, and promotes animal-free eating if it is convenient or sufficiently tasty. Interviewee Geert-Jan employs a similar argument, suggesting that “there’s no

202 9.5. The Cultural Politics of Meat Alternatives: Mechanisms of Challenge and Reproduction use to . . . take the meat variant, if the plant-based variant is just as good.” While this sets a moral imperative, both Phoebe and Geert-Jan make substitution conditional and countable–this explicitly accepts that eating meat remains a legitimate choice. Nonetheless, this strategy implicitly and sometimes explicitly highlights a need to reduce meat consumption in line with the political frame; even though it relies on an individualised understanding that meat alternatives may indeed allow for a ‘business as usual’. They are thus reminders of the necessity to action, but also individualise responsibility.

9.5.4 Eating Experiences and Expectations A favourable evaluation is a precondition to singular or regular swaps as discussed in the previous section. Because the meat alternatives category builds on and strongly relates to meat, it should be no surprise that meat therefore plays an important role in its evaluation. Even comparisons amongst meat alternatives often treat meat as the baseline for what would constitute a better experience (see section 9.3.2). Compared to cultural aspects of meat alternatives which attain symbolic meanings of meat, taste experiences are even more so mediated by individual experiences and preferences–they are thus only meaningful for the individual. Particularly taste test videos highlight a variety of caveats that mediate experiences and expectations of foodstuffs: There seems to be a fundamental awareness that something can taste different from meat or less like meat than another product, but it can nonetheless be nice, such as voiced by the YouTubers Robert or Emily. In other taste test videos, we can recognise an awareness of how the context of a food would influence the experience: For example by Alice who decides not to add ketchup, and by Ollie who tries a burger by itself and in a bun context. And through their discussion of sausages and hot dogs, You- Tuber Jessica inadvertently highlights the issue that sensory experiences of meat are constructed in themselves. Previous literature has frequently highlighted the importance of external food cues such as labels for taste evaluation (see Piqueras-Fiszman & Spence, 2015 for a review). This suggests that labelling is crucial for meat alternatives to navigate the individual experiences and expectations: Are sausages supposed to be like pork sausages, Lincolnshire sausages, hot dogs, or vegetable-based sausages? Whatever label, consumer knowledge that products are animal-free will attract scrutiny on a level that meat products would not get: Meat may taste good or bad, but it would remain meat because of its animal origin. Meat alternatives in contrast can be meaty or not meaty, as well as tasting good or bad; they can meet the individual expectations of a sausage or a burger, or fail to meet them. Curiously, what might have been a legitimately tasty

203 9. Meat Without Animals: An Analysis meat-free burger may need to be re-legitimated, such as by YouTuber Emily after tasting a ‘more meaty’ burger:

I do still like the Linda McCartney burgers as their own thing, like they don’t taste exactly like meat, they have a nice, their own kind of taste. (YouTuber Emily)

The prominent role of expectations and labels and the different level of scrutiny applied inadvertently cause the sensory evaluation for meat alternatives to reproduce dominant culture–and if it is just by explicitly pointing out differences to it.

9.5.5 Navigating Experiences: When Food is Just Tasty As outlined in the previous section and in section 9.3.2, meat consumption is usually a reference experience to meat alternatives. However, many products are not meant to be meat-like and thereby subvert the sensory norms established by dominant meat culture. Yet, these products often still reproduce these norms through their formats, certain design elements, or their external framing. For example, interviewee Rachel emphasises the authentic sausage casing which “gives a little bit of that springiness, like traditional meat sausages,” and complements their equally emphasised vegetable filling; and cauliflower wings become “a perfect party nibble with a bit of spice” (tweet N8), thereby providing similar experiences to established understandings of ‘wings’. Interviewee Steve balanced these tensions with their seitan products for which they emphasised the context and the usage most, for example by “not pretending to be chicken but . . . an alternative for people to chop up and put in their stir-fry.” Yet, instead of describing seitan they resort to the concept of meat alternatives: “When somebody comes up to our stand . . . we describe what it is, which is basically, it’s a wheat protein-based vegan meat alternative.” This way Steve’s brand makes seitan part of the concept of meat alternatives not because they understand their product primarily as such, but because the concept gives a quick understanding of the uses and functions of their foodstuff. So while taste and texture experiences of dominant meat culture are undermined or ignored in these examples, they still retain specific elements of it; these include the links between format and/or labels and use cases, or highlighting specific sensory elements that are similar or different to meat. These products go beyond the confinements of dominant culture and can be a challenge to certain aspects of it, yet their reliance on certain aspects of meat to claim further legitimacy selectively supports such hegemony.

204 9.6. Theoretical Implications

9.5.6 Reclaiming Language There are few texts in the data which explicitly focus on the language concerning meat alternatives. Interviewee Hilary for example suggests that the act of “just reclaiming some of the little words that are used in animal-based products, without . . . involving the use of animals” may remove the focus on animals in food culture. Specifically, they argue that when steaks or burgers are generally understood as steaks or burgers regardless of their animal-origin, the terminology can be reclaimed. They compare this to the etymology of meat, which “shouldn’t actually mean animal muscle or animal tissue, it just means food.” Language also plays a role in branding, such as for interviewee Tony’s brand that they just had renamed from ‘Meet the Alternative’ to ‘Meat the Alternative’ to reposition the brand’s claim as an equivalent to meat that should be positioned in meat shelves. YouTuber Elliot appreciates the branding of Iceland’s ‘No Bull Burger’ as a word play, whereas YouTuber Mia finds that they should have more explicitly emphasised in their punchline that the burger was not actually meat. Other texts use meat-based language is inadvertently, possibly because it is known and associated to the specific context and alternative ways to speak about foods are missing. The findings remain inconclusive on whether language can actually be ‘reclaimed’ because it might still be associated with dominant culture. An increased understanding that such meat words do not require animal ingredients might however challenge dominant culture–even if it were absorbed into an understanding of ‘meat’. However, reclaimed language may still reproduce dominant culture in different ways as noticed throughout the identified mechanisms engaging with dominant culture.

9.6 Theoretical Implications

In this section, an overview of the theoretical implications of the findings is presented. Specifically, this concerns the congruence of findings with projections from the literature (chapter 2) and the conceptual framework (section 2.5). While these implications presented here are very concrete and close to the data, a broader discussion of the theory will be presented in section 10.3. These issues discussed here specifically concern insights on (1) category formation and emergence as derived from the analysis in section 9.2; (2) limited substitutability and the stability of categorisation as building on the discussion of format flexibility and the difficulty to identify clear, universal category boundaries in section 9.3.1; and (3) generativity of the category and its increasing accommodation to hegemonic culture building on section 9.4.

205 9. Meat Without Animals: An Analysis

9.6.1 Insights on Category Formation The theoretical literature and the conceptual framework helped to identify the roots of the category both in vegetarian food and in meat consumption (see section 9.2.1). By these means, tensions between the two categories are suggested to have been resolved as they form a new category. While this resolution of tensions had been highlighted in early time periods, it almost completely disappeared later. In accordance with the literature, certain characteristics internalised into the category definition and the references made to the vegetarian antecedent category disappeared. However, references to the meat category remained, possibly to ensure amenability to dominant culture. Thereby, ‘friction’ has been reduced whilst the category attracted an increasingly mainstream audience. In recent time periods, the generativity of the category could further be observed through new meanings, labels and associations emerging, as some conceptualisations of meat alternatives became increasingly positive and at par with meat, rather than dependent on its sensory evaluations (see also section 9.6.3 below). The literature reviewed in section 2.2 has led to inaccurate predictions in relation to the specific observations made when categories have emerged. The observations made here suggest that earlier the category promised to resolve tensions, whereas later, positive visions and affordances are highlighted. Over time, niches and subsets of the category are occupied by different products; as the category reaches maturity, its scope is broadened so that new experimental designs exploit the generative capacity of categories as they build on the core concepts of the category. Responsible for the theoretical mis-alignment between the literature and the presented findings could be specific texts and their function, such as press releases as compared to advertisements, as well as the category in focus. Future research might investigate in more detail the differences which categorisation from different texts asserts for different product groups. Of particular relevance might also be the differences between tools or services, which a large share of categorisation literature seems to be concerned with, and consumption goods, which appear to be evaluated more on individual taste and experiences than on generalisable criteria. Further consideration should be given to the importance of a broadening of audiences when categories emerge: The growth throughout the emergence phase inevitably requires new consumer groups to be attracted, which as outlined may affect valuation criteria.

9.6.2 Limited Substitutability and Categorisation Stability Recalling the category studies literature, category members were suggested to be “perceived to be of the same type or close substitutes for each other in satisfying market demand” (Navis & Glynn, 2010, p. 440). Although at large, sensory characteristics

206 9.6. Theoretical Implications feature more prominently than practices across the data sets, meat alternatives only require meat formats but not necessarily meaty tastes or textures (section 9.3.1). Because meat formats as an implicit reference point are almost inescapable, food formats assert almost global substitutability. Conversely, emphasis of vegan or vegetarian credentials asserts differences as it distinguishes one diet from another and caters for a specific target group, but such material distinction into meat-based and meat-free products is not relevant to everyone alike. This suggests an inconsistency in the current categorisation for meat alternatives: While formats are substitutable and tastes might be more or less equivalent, the used ingredients might be exclusive to each other for certain audiences but not for others. This in turn limits substitutability of these categories at large. Formerly, most consumers of meat alternatives followed meat-free diets, but the recent trends towards flexitarianism and meat reduction expose these limitations (see sections 4.3 and 4.5.1), which makes the categorisation historically contingent. Grocery fixtures therefore cannot be representative of all audience realities at the same time, and suggestions for re-categorisation like those made by the interviewees highlight questions concerning the future development of the meat alternatives category. These tensions between different realities of intermediaries and audiences might be one explanation why boundary contestation by the interviewees can be observed. It thus needs to be noted that despite the category being considered established (section 9.1), its generativity causes a broader array of categories to emerge, such as those focusing on ‘protein’ or ‘plant- based’ credentials. The predictions concerning the meat alternatives category made on the basis of the conceptual framework (see again section 3.1.4) can however be confirmed: Firstly, while the technical boundaries relating to meat as animal flesh are uncontested, some interviewees suggest that meat and meat alternatives should be understood jointly as ‘protein’ or more generally by format, thereby trying to remove prominent boundaries. Other data such as tweets and advertisements may be understood to support this, as they often see meat and meat alternatives as identical, equivalent, or at least interchangeable. These attempts to alter boundaries or the social realities which understand them as negligible are again founded on the widespread use of meat-related food formats for meat-free products: These formats allow substitutability in relation to the affordances associated with the formats, or more specifically the terminology used for these.

207 9. Meat Without Animals: An Analysis

9.6.3 Categorical Generativity and Hegemonic Accommodation Kennedy and Fiss (2013) asserted that category effects are strongly linked to the “familiarity of occasions and motivations for their usage” as well as the “meanings and ontologies they carry” (p. 1138). Both are at the core fulfilled through the cultural frame, as occasions and meanings of meat-centred gatherings were incorporated during the emergence of the category (see section 9.2) and persisted thereafter. Broadly speaking, the cultural frame can thus be understood as the foundational meaning system of the meat alternatives category. Meat formats thus provide an ontology which is filled with meaning through their societal and cultural functions, attributed by historically contingent individual and collective understandings of ‘meat’. Furthermore, as also emphasised in the conceptual framework, the category has become generative in that other or different foodstuffs relate to and build on understandings of meat alternatives for their sense-making. Many narratives found in the data across this thesis indicated that differences between meat and meat alternatives are increasingly invisible. When considering the increasing interest of incumbent actors from the food and meat industries in meat alternatives (see section 4.5), and the tensions identified in their mutual substitutability (see section 9.6.2), a subsumption (Kennedy et al., 2010) of the meat alternatives category into a meat category is possible. A similar understanding of change has been termed ‘hegemonic accommodation’ whereby elements of societal pressure are internalised by a hegemonic power (Levy & Scully, 2007). In this particular case, this meant on the one hand that a dominant meat culture would encompass more different tastes and embrace different raw materials not derived from animals. On the other hand, if the issues fuelling the political frame would be perceived to be resolved in reality and/or in discourse, it is imaginable that meat alternatives lost their seductive power. Thus, a meat alternatives category would not be the challenger, but dominant meat culture would imbibe previously deviant ideas into cultural and societal norms. One may thus identify the meat industry to have the most power over the meat alternatives category, both ideologically and practically. While other ways to enact meat-free diets are still possible, and maybe even more likely through a greater acceptance of fleshless eating supported by meat alternatives, the dominant meat culture limits what is ‘normal’ while extending understandings of what is ‘nice’, regardless of whether animals are involved or not. This general trend is in line with research on social movements, which found that their wider success depends on the desirability and congruence of the framing with established cultural, political, or legal systems (e.g. Snow & Benford, 1992, see also section 2.4). Essentially, meat alternatives developed from a counterpoint of the dom-

208 9.7. Conclusion inant order in the emergence phase, to a convenient alternative offer at present, and possibly a universally acknowledged equivalent in the future. Although this stretches the empirical intentions of the author, this process of assimilation to the dominant order can be understood as the ‘categorical imperative’—whereby market forces pun- ish deviance from established category codes and identities. In fact, Zuckerman (2017) argues that “recent research . . . supports the idea that categorical codes (and their dis- ciplinary effects on offerings) are endogenous to the application of a particular theory of value” (p. 49). In the case of meat alternatives, this theory of value is the dominant cultural way of eating and thinking about food: carnism.

9.7 Conclusion

In the conceptual framework it was proposed, that the frames used to describe meat alternatives have an effect on our understandings of meat, as they make certain aspects of meat eating or their difficulties more salient. Specifically, it was proposed that meat alternatives were in tension between being like meat and being different from meat— or between challenging meat culture and reproducing meat culture. In this chapter, the different data sets were merged into a rich analysis. The findings suggest that meat alternatives have emerged from understandings of ‘vegetarian food’ and ‘meat’ alike (see section 9.2). But whilst meat alternatives have usually been proposed as a more convenient, normal, or nice counterpoint to ‘vegetarian food’, meat remained an ambiguous reference. Originally, animal ethics and health impacts of meat were explicit in the category identity, but these have increasingly been incorporated into implicit understandings of meat alternatives or outsourced to meat-free eating at large. For meat-free eating, meat alternatives are a tool for a smooth transition as via meat formats they embody meanings and occasions and can transfer them to other products through labels or appearance (see section 9.3). This way, meaty experiences can be maintained whilst a requirement or desire for meaty tastes or textures is individualised. Acknowledging this foundational understanding, meat alternatives can be broken down into three constitutive frames: a consumption frame encompassing the hedonic and sensory eating experience with individualised tastes; a cultural frame including the individual and collective experiences, occasions, and meanings associated to meat and its facsimiles; and a political frame through which meat alternatives become a tool for dietary change. The combination of these three frames positions meat alternatives as ‘better meat’, which allows everyone to consume healthy, guilt-free, and according to individual taste and texture preferences without having to change the ways of eating. The mechanisms with which discourse on meat alternatives integrates with, challenges, or reproduces dominant culture were identified (section 9.4). Broadly speaking, meat alternatives appropriate and reproduce dominant meat culture by perpetu-

209 9. Meat Without Animals: An Analysis ating an understanding of a specific ‘normality’ which defines which food products can legitimately be eaten when and why, and which ones may not. Specifically, three mechanisms could be identified, which also draw on the sensory norms of dominant meat culture: Firstly, meat culture is reproduced as other ways of eating are occasionally stigmatised or invisibilised as they are not understood to be as normal, pleasurable, appropriate, or convenient. Secondly, sensory norms of dominant meat culture are the baseline to evaluate meat alternatives, and any deviance from this is explicitly emphasised, thereby highlighting what the norm is. Thirdly, dominant meat culture is reproduced as meat alternatives build on the same tacit and practical knowledge and therefore remove necessities for knowing about other ways of eating. Conversely, challenges to dominant meat culture were expected in criticisms of the health, environmental, and ethical issues involved with it. However, due to an internalisation of issues related to meat culture into understandings of meat alternatives (see section 9.2), meat alternatives became largely depoliticised as a challenge to dominant meat culture; instead, they became a tool within a wider discourse on reducing meat consumption. Challenges to meat culture were identified when either cultural or consumption frames were intentionally broken; for example, when new tastes and textures are associated to traditional meat products. In this manner, the concept of meat could be detached from understandings of animal-based tastes and textures, from its cultural understandings and relevance, or from the language used. Elements of both, reproduction and challenge, were unified in an understanding of meat alternatives as ‘better meat’. Here, the established understandings of meat may be challenged through more diverse understandings of what good taste and texture constitutes, but at the same time there would be an emphasis on the moral superiority of meat alternatives over meat. When presented as superior, meat alternatives become an obligatory choice over meat, albeit one which is conditional to what audiences are willing to compromise, and therefore individual responsibility. However, all challenges outlined here rely on the fact that dominant meat culture is associated with animals, opposing which crucial to the ‘political frame’. Therefore, it is possible that a dominant meat culture might simply be extended onto a more diverse understanding of what ‘meat’ means–including that it does not derive from animals and that it might not have the taste and texture of animal flesh. Should this be the case, then commercial meat alternatives in the understanding of this thesis are unable to challenge dominant meat culture. This is, unless an understanding of meat alternatives is extended on, for example, lentils or beans that are explicitly sold or consumed as meat alternatives, which is the case in some tweets. A summary of the ways in which meat alternatives reproduce and challenge dominant meat culture is presented in Table 9.2 earlier. Any challenge to dominant meat culture might also need to be assessed on the

210 9.7. Conclusion basis of consumption figures. Dominant meat culture–according to the findings— can only be quantitatively challenged if the combined consumption of meat and meat alternatives were to decline, both in everyday life and on meat-centred occasions. While the share of meat alternatives by volume has reached more than 1.2% of all meats, the overall consumption of all ‘meat’6 in the UK has declined over the past years, from around 66 kg per capita and year in 2010, to 60 kg in 2018 (Euromonitor International, 2018). In this way, the reproductive mechanisms of dominant meat culture as outlined in this study may not be so dramatic. Nonetheless, environment, animals, and public health will certainly benefit from a reduction in animal meat that is reared and eaten.

6In fact, Euromonitor counts meat alternatives in their category of processed meats.

211

Chapter 10

Discussion

In this thesis it was argued that the talk about and the use of meat alternatives selectively borrows, reiterates, affirms, and contrasts elements of existing food categories— here specifically those of ‘vegetarian food’ and ‘meat’. Throughout the category’s journey into mainstream food culture, friction with dominant ways of thinking or doing things would disappear, as otherwise the new category would not be amenable to broader audiences. Furthermore, every interaction with meat alternatives inevitably frames the category, but also participates in wider signification and valuation of food, specifically of those food categories which meat alternatives emerged from. Therefore, meat alternatives cannot be understood independently; instead, as done in this thesis, these foodstuffs need to be understood as drawing on and influencing how society values food in general, and meat in particular. To undertake such described analysis of meat alternatives, in this research project the relationship between meat alternatives and meat was analysed. The project was concerned with firstly whether and how dominant meat culture would be reflected in a category of meat alternatives, and secondly how meat alternatives then reproduce and/or challenge dominant meat culture. This was achieved by employing a qualitative approach using inductive content analyses on four different data sets related to meat alternatives: Print advertisements, manufacturer and expert interviews, taste test videos, and tweets. This was supported by a historical narrative and contextual data. The research project was motivated by the increasing popularity of meat alternatives in ‘the nation’s shopping trolleys’, and the tensions that were suspected in how meat alternatives related to dominant meat culture. The remainder of the thesis proceeds as follows: Firstly, responses to the research questions will be presented. Then, a synthesis of broader implications of the empirical insights are discussed. This is followed by remarks and reflections on the proposed conceptual framework and the theoretical literature of category studies in the context of this research project. The chapter proceeds by summarising the contributions of this work, highlighting and discussing the limitations of this research project, and illustrates a number of practical implications and recommendations. The chapter is then concluded by suggestions for future research.

213 10. Discussion

10.1 Responses to the Research Questions

The first research question concerned how the meat alternatives category changed throughout its development process. The analysed time period was designated to commence in 1985, as informed by a review of historical documents such as industry news and market research. It was found that meat alternatives likely formed an established category after 2012. The category was found to be taken for granted, and widely commercialised, as well as being a legitimate food category for its audiences and having an established valuation and meaning system. Three important changes throughout the emergence could be identified: (1) Value statements disappeared in descriptions of products, as these might have become implicit to the understanding of the category and might otherwise alienate mainstream audiences; (2) food occasions were found to be increasingly prominent ways to advertise products, possibly in order to continuously highlight meat alternatives as an adequate replacement for meat; (3) more recently, meat alternatives were found to be more diverse as products with different tastes and textures appeared on the market; they created positive visions through aspects of plant-based and/or protein foods. It was further found that while comparisons to vegetarian food at large decreased, comparisons to meat products persisted in the advertisements. By way of distinguishing the category from vegetarian food, advertising for meat alternatives was further found to reproduce existing stigmatisation of diets or food choices free from meat (or alternatives to these). The second research question called to investigate the current identity of the category in order to identify its consistency and persistent links with other categories such as a meat category. Generally, descriptions of meat alternatives focused most prominently on sensory experiences, followed by their embedding into life and practices, as well as positive health credentials. While descriptions also featured an alleviation of issues surrounding animals and environment, these were usually either acknowledged implicitly, or tied to dietary change which was supported by products rather than specific products themselves. Extending the findings from the previous chapter, it was found that meat is inherent to the category identity of meat alternatives on the basis of meat formats, which–together with being meat-free–appear as the primary criterion for category membership. Meat formats, such as sausages, burgers, or roasts, function as a carrier for experiences and expectations, as they are linked to food-related practices and identities. Meat was further found to be the explicit or implicit baseline for what ‘good’ meat alternatives are, whereby from established ideas is accepted but often explicitly signposted or emphasised. For the third and final research question it was investigated how meat alternatives are framed and how this in turn frames meat. In particular, considering the main research question, it was suggested that meat alternatives may reproduce and/or chal-

214 10.1. Responses to the Research Questions lenge dominant meat culture. At large, meat alternatives and meat were framed in three interrelated frames: (1) Consumption frames highlight both, meat alternatives and meat, as a hedonic consumable as everyone can have their own taste and texture preferences, albeit it is usually assumed that consumption experiences for meat alternatives indistinguishable from meat. (2) Cultural frames highlight meat as a socio- cultural entity which is normal to eat at specific occasions and in everyday life, and is valuable through the individual and collective experiences, expectations, and associations concerning the food; meat alternatives are thereby an extension of meat that can be used in the same ways and contexts. (3) Political frames highlight the wider context of food consumption, specifically that meat-free diets are preferable for animals, environment, and public health, and that meat alternatives can help to make a switch away from meat consumption. The main research question asked how meat alternatives may reproduce and challenge dominant meat culture. Direct criticisms on meat consumption through meat alternatives turned out to have disappeared from meat alternatives: In particular, it was found that while previous advertisements for meat alternatives explicitly criticised meat consumption, at present, meat alternatives are a tool to facilitate dietary change which is motivated by these reasons. The main challenge of dominant meat culture through meat alternatives has been identified in a potential dissociation of tastes, textures, or affordances and ‘meat’. Although the norm still implies that meat alternatives ought to be meaty, it is increasingly acknowledged that not everyone enjoys the taste or texture of meat, and that products can be enjoyable without or with less meaty sensory or cultural experiences. A potential further threat to meat consumption may originate from the use of language and the reclaiming of specific meat words through animal-free alternatives. Both these–diversified sensory experiences and reclaimed language–may however become part of an understanding of dominant meat culture and extend what may be seen as ‘better meat’. Choosing this ‘better meat’ is sometimes presented as a moral obligation, even though one that is conditional to individual liking. Dominant meat culture was found to be reproduced through the cultural frame and its understanding that meat is a normal, convenient, and tasty foodstuff to eat as part of regular meals as well as the essential centrepiece for specific occasions. Specifically it was found that the use and legitimation of meat alternatives might stigmatise or invisibilise other food choices; that deviance to sensory norms of dominant meat culture are explicitly emphasised as such; and that meat alternatives eliminate the need to build or amend tacit or practical knowledge beyond what dominant meat culture requires. Meat alternatives may thus both reproduce and challenge dominant culture; it depends on how they are framed.

215 10. Discussion

10.2 On Meat Alternatives

The findings of this thesis as summarised above suggest that meat alternatives need to be understood as a category that is inherently dependent on meat. It reproduces dominant culture broadly through its replication of the formats and in turn the affordances of meatstuffs; the challenges to dominant culture lie largely in rejection of sensory norms. The underlying political issues that are pointed out in meat through the consumption and framing of meat alternatives scrutinise methods of ‘meat production’; hereby, meat alternatives are often put forward as a tool to reduce or eliminate meat- issues. In the following few subsections, the major findings will be contextualised in more depth: At first, section 10.2.1 offers concluding thoughts on the continuity and consistency which is embodied in the cultural frame of meat alternatives; next, in section 10.2.2 the extension of meat alternatives towards non-meaty tastes as reflected in the consumption frame is discussed as a project of diversity and social change. The two following sections then deal with the political frame of meat alternatives: Sec- tion 10.2.3 investigates the degree to which meat alternatives may be understood as a ‘counterculture’, which is then extended to the tension between food activism and big business in section 10.2.4 which became apparent in the interviews. In the final section (10.2.5) these insights are combined and meat alternatives are placed into a larger context of agro-industrial development.

10.2.1 Just the Same: Meat Alternatives as Continuity It should not be surprising that a food category which is often referred to as ‘meat alternatives’ can be conceived as somewhat similar to meat. This similarity has been shown in this thesis to be based on experiences associated with the context of meat consumption. Originating from meat-based food formats, occasions and associated meanings are adopted in the meat alternatives category. This finding is in line with previous research: Using interviews with Australian consumers, Nath and Prideaux (2011) highlight the importance of meat alternatives for new meat avoiders, as they are said to make dietary change easier through the familiarity of occasions and meanings. Providing more depth to the findings of Elzerman et al. (2011, 2015) who suggest that meat alternatives are evaluated more positively in a meal context, this thesis suggests that meat alternatives are in fact legitimised by the social norms and cultural occasions surrounding meat consumption. Although meaty tastes and textures are often considered important in product development (e.g. Wild et al., 2014; Zorpette, 2013), these were found here to be not tied to the category identity, but are instead subject to individual choice. Given that probably almost any British consumer has once enjoyed meat and many still do, the legitimacy of meaty meat alternatives is unchallenged by this. While the share of meat-free meals in the UK appears to be increasing, meat

216 10.2. On Meat Alternatives alternatives make sure that established understandings of normal, appropriate, or convenient food will at least not be forgotten.

10.2.2 Violating Norms: Meat Alternatives as Diversity As illustrated, meaty sensory experiences are not a requirement for membership in the meat alternatives category. Nonetheless, they are occasionally criticised for being not meat-like enough or too meaty, which make meatiness a product design choice rather than in inherent property. This design choice may become an explicit violation of norms and expectations, when it is understood and framed as reclaiming ‘meat’ or normalising meat-free pleasures (see also section 9.5). However, this may also be understood as the pervasiveness of meat culture that prescribed meat-free foodstuffs to adhere to its cultural norms. Nonetheless, meat alternatives yield diversity through a generally increased choice for consumers concerning not only tastes and textures, but also regarding food allergens, or ‘raw materials’ used. Notably, any deviance from meaty tastes is often highlighted or mentioned upfront as a violation of norms and expectations. While diversity enables tasty experiences for everyone, different experiences also allow comparing and weighing up experiences, and this includes meat. Therefore, meat and meat alternatives can be swapped or reduced, but also explicitly chosen. This leads to the discussed issues of quantifying meat reduction (see section 9.5) which imply that meat should but need not be replaced by a meat-free version. Thereby, meat remains a legitimate choice. As products increasingly attract meat reducers, the choices emerging from this align with two approaches currently championed by the third sector: Firstly, the ‘less and better’ approach for meat consumption which highlights that less meat should be eaten and that meat ought to be of higher quality (Eating Better, 2018); and secondly, considering the focus on protein, an understanding of ‘protein diversity’ in the diet which argues that a variety of proteins instead of only meat protein should be consumed (Carbon Trust, 2015). Both approaches promote choice rather than restricting it, and therefore making the political frame subject to individual opinions and evaluations (see also Ehgartner, 2019 for a further discussion on protein diversity in context of sustainable consumption and production systems).

10.2.3 Plant-Based Counterculture: Meat Alternatives as a Political Project Starting from the first meat alternatives in Victorian health food communities, meat alternatives had always been a compromise between certain political ideas and concerns, and being amenable to meat eaters in order to win them for this political project (see

217 10. Discussion section 4.3.1). The current market for meat alternatives accordingly focuses on meat reducers by emphasising health and hedonism while the choice for meat alternatives as a ‘better meat’ is left to audiences to decide. However, meat alternatives still exhibit capacity for change, as the highlighted issues–largely health, but also animal rights, animal welfare, and more recently environmental sustainability–persist. According to the understanding of its political advocates (e.g. interviewees Steve and Hilary, You- Tubers Phoebe or Mia), any meat alternative consumed is one that benefits animals or another cause. But what are the prospects of this political project? In news reports on the UK meat-free Monday campaign, Morris (2018) identified similar patterns as this research project: Reduced meat consumption was largely framed as positive, convenient, somewhat unpolitical, and not requiring much com- mitment. Similarly, while some texts frame a reduction of meat consumption as a moral obligation (section 9.5), they remain conditional to individual preferences. This argument follows a utilitarian understanding that informed consumers will make the choices best for them. This is turn individualises health and ‘care’ aspects more broadly. It works with the idea that health needs to be maximised, whereby consumers “are seen as . . . both the cause and the solution to potential health problems and thus are made to be accountable for their own health” (Ayo, 2012, p. 104). Similarly, studies on environmental discourse have discussed ‘green guilt’ (Plevin, 1997) which can be ‘offset’ and thus alleviated (Kotchen, 2009). The frequent positioning of meat alternatives as a choice or an offer (such as the frames used in tweets, see section 8.3) conforms to this logic of individualised responsibility, as no other actors appear to push regulation against excessive meat consumption (see section 4.5.5). Instead, meat alternatives present a market-based solution which occasionally receives government funding for research (see e.g. Quist, 2008) but is otherwise left to the market whilst its ‘competitor’, the meat industry, benefits from direct and indirect subsidies (cf. Welles- ley, Froggatt, & Happer, 2015). So while meat alternatives may present a political project for individuals and individual organisations, they do so as an individualised market-based solution within the context of economic exchange. The suitability for meat eating to be reduced by market forces may be further questioned when considering psychological insights: Rothgerber (2014) has shown that even the mental presence of vegetarians increases denial of animal mind in order to reduce cognitive dissonance and further justify one’s meat eating. The presence of meat alternatives alongside meat in the shelves may help convenience purchases as among others interviewee Tony hoped, and are likely to attract those people who seek to reduce their dissonance–such as to offset their own guilt or improve their own health–and are confronted with this convenient swap. However, in other people it may have the contrary effect as it could strengthen individual dissonance-reducing belief systems. Therefore, as argued in the conceptual framework, there needs fun-

218 10.2. On Meat Alternatives damental amenability to reducing meat consumption or swapping meat with meat alternatives. This limits the usefulness of meat alternatives as an origin of change rather than its role as a facilitator or tool as discussed previously. However, not only on an individual level a substitution may backfire: In general, meat alternatives are a foodstuff which can facilitate dietary change, and changing diets may increase pressure on meat eaters to reduce their consumption of meat. But as discussed in the previous chapter (section 9.5), frames used to legitimise meat alternatives often delegitimise, stigmatise, or invisibilise other ways of eating. This finding paired with understandings of ‘less and better’ meat and dietary ‘protein diversity’ allow meat consumption to persist as a legitimate choice. In fact, meat then remains a ‘normal’ choice, if not also ‘necessary’ (such as a convenient food to barbecue) and ‘nice’ (see also Piazza et al., 2015, on the ways in which meat consumption is justified). This normalisation of meat-like foodstuffs may on the long run impact complete removal of non-human animal flesh from human food supply. This is because a perpetuated normality of meat-like diets keeps transitions between animal-based and animals free ‘meats’ easy. So while meat alternatives may reduce the need to develop tacit and practical knowledge about other ways of eating (see section 9.5), this reliance on existing tacit knowledge makes meat alternatives an inherently democratic food choice in a meat-centred society: As exemplified for example by YouTubers Pippa or Lucy, meat alternatives thus remove all barriers to adopting a meat-free diet.

10.2.4 Food Activism: Meat Alternatives between Values and Money Particularly amongst the interviewees, two contrary self-identification mechanisms could be observed: On the one hand, Geert-Jan or Tony understand their companies as a facilitator of consumer desires to reduce meat. On the other hand, Hilary or Steve see their businesses as offering an increased choice to convince people to go meat-free. Despite the former group claiming that their products are more meaty than competitors, another substantial difference can be observed in the emphasis of the political frame. Hilary and Steve both consider themselves food activists (cf. Counihan & Sin- iscalchi, 2013), who through their business enable dietary transitions by increased choice. Their underlying motivation is that animal exploitation in line with other meat-related issues needs to stop sooner rather than later. Although every product is said to count, the anecdotes I was told focus on how people managed to start a meat- free diet supported by these products. They understand their audiences as food lovers, people who care, and people who would like to change their diets but do not feel able to. Their products may taste like meat but need not, because their consumers are

219 10. Discussion considered to like good foods, which does not need to mean meaty ones. Meat alternatives are used here as part of an activist toolkit which is contextualised as a business case. Tony and Geert-Jan on the other hand understand their respective businesses as ‘high-tech manufacturers’ that surpass the competition in an attempt to provide the meatiest experiences on the plant-based market. Albeit the positive effects of less consumed meat are not rejected, they voiced no deeper motivation beyond sales and making the ‘best’ products. Consumers are conceptualised as wanting to reduce meat eating for health reasons but only if there is no difference on taste, texture, or nutrient/protein intake to actual meat; in this case, the businesses are facilitators to individual choice. The anecdotes I was told are accordingly concerned with quality credentials, such as when people can not distinguish their products from animal meat or when they were told of meals cooked who satisfied both meat eaters and vegetarians. They take no moral or moralising stance, and health credentials are only emphasised to educate consumers about how nutritionally good or better substitutes could be. For these companies, manufacturing meat alternatives is thus an ever-growing, apolitical business case which benefits from individual consciousness of meat reducers. As the increasing popularisation of meat alternatives and meat-free diets brings investment from the powerful actors in the agricultural and food industries, this might lead to an increasing number of such companies in an apolitical, consumer choice-focused market. This will be discussed in the next section.

10.2.5 More Meat from Less Feed: (Eco-)Efficient Contingencies In the last centuries, meat became increasingly removed from the process of hunt- ing or slaughtering (Buscemi, 2014). Since the mid 20th century the industrialisation of animal agriculture has not only contributed significantly to climate change, but also to environmental degradation, the destruction of ecosystems, or eutrophication (e.g. Steinfeld et al., 2006; Nijdam et al., 2012). Interviewee Hilary accordingly expresses the urgency for “how the future may be plant-based, because . . . animal products are slowly killing us, in many ways.” As illustrated above, it may be considered an achievement for the social movements highlighting these issues, that meat alternatives slowly become a widely accepted solution to replace meat. Across the data sets, the functional equivalence of products is highlighted, and given the pressing issues, meat alternatives are identified as a moral yet conditional choice (see again section 9.5). Acknowledging that meat alternatives represent a more open, encompassing version of the status quo (see section 10.2.1), meat alternatives can be considered an extension of what meat is currently understood as.

220 10.2. On Meat Alternatives

In opposition to the political aspects of meat alternatives illustrated above, meat alternatives may also be considered a technologically incremental step away of meat, which is among others emphasised by interviewee Geert-Jan. Meat alternatives are just the same, but healthier, more environment-friendly, do not rely on animals, and that most importantly could be cheaper. The strongest argument to understanding meat alternatives as a logical, incremental successor of meat is possibly the material change induced by meat alternatives, which is largely dependent on the disruptive practices in industrial animal farming. They are a solution to a sustainability issue in which meat is overthrown by the impacts of its own success and replaced by a better, yet more complex system (cf. Tainter, 2011). Meat alternatives do not rely on animals in its supply chains, and therefore can avoid the highly inefficient food conversion ratios that can be found in animal agriculture (see Pimentel & Pimentel, 2003). They therefore need less agricultural raw material, less energy, and less land (Nijdam et al., 2012; Pimentel & Pimentel, 2003; Smetana et al., 2015). Replacing meat with meat alternatives is thus not only a reasonable choice but also the most efficient and cheapest choice to produce ‘meat’—if it is not already, then it will be cheaper with the returns on efficiency and scale. However, because products may nonetheless be priced for consumers as equivalent to premium meat (as suggested by interviewee Tony), this could increase profit margins. Goodman, Sorj, and Wilkinson (1987) in their theory on agro-industrial development, suggest that any commercial development of the ‘agri-industrial complex’ aims at increasing profits by overcoming the limitations of organisms and their metabolisms and by reducing the risk involved in natural systems. This is achieved by reducing “the importance of nature in rural production” (p.3, appropriationism, e.g. horse to tractor) and by “eliminat[ing] the rural product” (p.4, substitutionism, e.g. fabricated foods from generic food components, or the use of chemical additives). Although a rural product is still required as feed to the machines that make the textures we term ‘meat alternatives’, it substitutes the animal organism with a factory1. As the animal has already been substituted, “controlling the naturalness” (interviewee Tony) of manufacturing processes and associated additives or the absence thereof remain important. With increasing success of meat alternatives, and an increasing consumer perception that meat and meat alternatives are somewhat equivalent, this development might become reality. In the grand narrative of the agro-industrial complex–not the least

1Conversely, in-vitro meat replaces the animal organism and stables with petri-dishes and clean rooms. Although in their conclusion, Goodman et al. (1987) treat tissue cultures as substitutionism (p. 186), in-vitro meat is still constrained to metabolisms through cell growth. However, as the input feeds differ between in-vitro meat and plant-based meat alternatives, the details of the processes might affect efficiencies. Further uncertainty originates from often undefined specifics of the technology (Mattick, Wetmore, & Allenby, 2015; Welin, 2013), high consumer scepticism towards in-vitro meat (Verbeke et al., 2015), and regulatory challenges (Johnson, 2018).

221 10. Discussion represented by the investments into meat-free by the likes of US-based meat processor Tyson Foods, Nestlé, or ABP (see again sections 4.5.3 and 4.5.7)—meat alternatives might thus just be an efficient continuation of the status quo that happens to be more environment-friendly. As carnism appears to be perpetuated through manufactured, plant-based food products, this promises another victory of technology over nature. In fact, this prime example of ecological modernisation (Hajer, 1996) allows us to continue ignoring the moral injustice of industrial animal agriculture (Bramble, 2017).

10.3 On Theory

In this research project, an understanding of categories as culture (Glynn & Navis, 2013; Kennedy & Fiss, 2013) was applied to the investigation of commercial meat alternatives in the United Kingdom. The overall approach taken in this research project proved useful for the research endeavour: The different data sets, both detailed and diverse, and both from producers and consumers, could be combined beneficially to generate a variety of insights. The division of the research project was beneficial, and allowed for insights which continuously integrated and built on each other; however, while this would not have reflected the logical necessities of the research progression, it might have been more insightful to first assess the current state of the category, and only then analyse its historic trajectory. This could have allowed to specifically address focal points in the current identity to understand where they might have originated from. In the following subsections, I firstly discuss the merits and then the difficulties when applying the category studies literature to the empirical case described in this thesis. Thereafter, I reflect briefly on the conceptual framework put forward and discuss possible further considerations for category studies which I encountered in my research.

10.3.1 Insights from Category Studies At large, the category studies literature provided an adequate and useful theoretical baseline with which to conduct this research project. In particular the possibility to combine insights both from an individual perspective of cognitive evaluation and understanding, and from collective perspectives of socio-cultural meanings and inter- relations. The broad conceptualisations of category emergence and category identity proved to be empirically useful, as did an understanding of categorisation as ‘social reality’ (see Kennedy & Fiss, 2013). The conceptual framework built largely upon the previous literature on category meanings including their origin and proliferation (e.g. Glynn & Navis, 2013; Kennedy

222 10.3. On Theory

& Fiss, 2013), which gave an important conceptual guideline to understanding categorisation; in this context, category lineages in historical development could be meaningfully investigated, whereby an understanding of explicit references and distinctions in texts proved useful (Navis & Glynn, 2010). Also, understandings of both strategic categorisation (e.g. Schiller-Merkens, 2017; Pontikes & Kim, 2017) and categorisation through discourse (e.g. Grodal & Kahl, 2017) sensitised this work for the inevitability of categorisation at large. In conceptual framework and empirical work, this could be abstracted into an application of frames which through the concept of saliency use- fully captured this notion of inadvertent categorisation (cf. Goffman, 1974; Entman, 1993).

10.3.2 Limitations of the Theory in Context of this Research Project As was suspected from the outset of the project and what led to the development of the conceptual framework, the concept of an independent category identity did not hold true in this research project. For the concrete empirical findings, one explanation would be that the category is simply not yet established, but more likely is that categories are in constant flux, and an analysis can only investigate snapshots in a complex web of category attributions (see Kennedy et al., 2010). In particular, it was found that the category became independent of some antecedent categories for its category identity, but not from all: On the one hand and in line with previous literature, a broad understanding of differences to ‘classic’ vegetarian food has been found to be not essential anymore (see section 9.2.1), but is still implicit when legitimacy is sought (see section 9.5). On the other hand, understandings of meat have continuously been reproduced and highlighted within the category, and continued to be a reference (see section 9.3). This strong discrepancy to existing literature may simply highlight the special case studied here, either as a ‘political’ good, as a fast-moving consumer good, or as a substitute product. It may however also be related to different data used, or a different conceptual understanding of what category independence actually means. Plausibly, satellite radio might be distinguished from other types of radio (Navis & Glynn, 2010), and modern Indian art might be distinguished from other types of art (Khaire & Wadhwani, 2010) but not necessarily independent of the concepts of radio or art as such. This aspect of independence leads to category prototypes, which I found a misleading conceptualisation in the study at hand. In particular, it appears an assumption that the category prototype is a category member (Durand & Paolella, 2013; Vergne & Wry, 2014), because everything else would be understood as category spanning. Al- though notably the prototype concept is based on a cognitive evaluation approach to

223 10. Discussion categorisation, based on the data prototypes could be identified. As this might relate to an understanding of categories as recombining elements from cultural repositories, the conceptual and empirical overlap between category spanning and such understandings of categories could be investigated in future research. — Specifically, in the research project meat emerged as a clear prototype against which other category members (i.e. meat alternatives) needed to be evaluated2 as deviance from the prototype was usually explicitly highlighted and legitimised (see section 9.3.3). Notably, those category members arguably most distant from the prototype were rhetorically the most active to legitimise their category membership using the codes provided by the prototype. As explicit differences between a meat category and a recent meat alternatives category can be understood as exclusively ideational, meat alternatives build their legitimacy exclusively on trust–for example through ingredient lists, certifications, mediators, and the retailer. This likens the findings of Weber et al. (2008) in their study on the emergence of grass-fed beef, with the difference that beef is understood as a subcategory of beef, not as a separate entity to it. These discrepancies between the findings from this study, and the findings made elsewhere, support claims by interviewees that meat alternatives and meat should be treated as more alike as they are.

10.3.3 Reflections on the Proposed Conceptual Framework It was noted at the outset of this research project, that the tools and perspectives provided by category studies may be of limited use for applications to meat alternatives, as these appeared to bear a strong ideological component. At large, the findings from the literature referred to throughout the discussions could however be utilised fruitfully. One of the main contributions of the conceptual framework put forward here, is that derivative categories may actualise and reconfigure understandings of antecedent categories. It needs to be noted that this study was not designed as a general proof to this proposition on cognitive levels. Instead, this study explored this claim and outlined mechanisms, both empirically and theoretically, how this might function. Nonetheless, further research (see also section 10.7) needs to look into these claims; for example, by comparing implicit and explicit understandings of meat across individuals who do or do not consume meat alternatives. Assuming that category antecedents can indeed be reconfigured by other categories, the conceptual framework allows putting forward a more explicit model of cat-

2Brands such as Quorn or Tofurkyhave prototypical properties in discourse, for example through their widespread availability and brand awareness. However, they do not serve as a reference baseline for what a meat alternative ought to be like and how they should taste like, but rather what meat alternatives are and how they do taste like.

224 10.3. On Theory egory substitution: A model category may be understood like an undifferentiated monolithic block with a fixed identity that sanctions any deviance (cf. Zuckerman, 2017)—in this case on the basis of sensory, social, or material differences. However, as shown in this thesis it appears that certain elements of a model category are more important than others, which here lie the meanings and affordances essential to constructing social and cultural experiences (cf. Kennedy & Fiss, 2013). In early stages of emergence, meat alternatives highlighted differences to animal agriculture, as well as to vegetarian food on grounds of poor taste or acceptance; more recent approaches such as a ‘less and better’ emphasise political consumerism whilst eating meat. While what has been identified as the ‘political frame’ became implicitly associated with the category identity and persists more or less enunciated in the views of the audiences, both emphasising and silencing this frame can be considered strategic categorisation. Viewing categories as ‘social realities’ fits with this selective representation of the political frame as it represents the narratives and justification mechanisms used to uphold or limit a hegemonic idea: As we silence, accept, or emphasise various frames, politics is made. In understandings of ‘protein diversity’ or simply consumer choice, any consumption of meat or meat alternatives therefore becomes political as they juxta- pose different issues, ideas, or notions of what ‘good food’ is, and more importantly what is not. Notably though, as meat-like sensory experiences do not appear to be a requirement, such new category captures deviant tastes in a new understanding of hegemonic meat culture; as argued before, this means that the substitution process outlined here may never actually be complete as meat consumption is accepted and persists as a model. This change process is illustrated in Figure 10.1. Lastly, the strong dependence on the meat category as argued in the previous section, has led to the prototype of meat alternatives being member of a different category. Building on this, it was illustrated in section 9.6, that the cultural hegemony of meat as identified in this study therefore functions like the often referred to ‘categorical imperative’ that sanctions deviance (cf. Zuckerman, 2017). Accordingly, incremental change is favoured over radical change; and moreover, legitimacy for a challenger category needs to be established prior to incremental change. In this research project, this can for example be observed in the considerably late focus on reframing product identities as protein and plant-based foods.

10.3.4 Further Considerations for Category Studies A few issues and aspects have emerged over the course of this research project, which could not be answered with the given literature and are upon other scholars to investigate further. For example, empirically, there appear to be differences between the categorisation of organisations, technologies, tools, or services which have all gained

225 10. Discussion

circle size = audiences which can political state- align with the category ments through reduction + choice

political social, cultural, statements and consumption through experience ingredients social, cultural, (tied to tastes) + choice and consumption diversity experience ‘less but better’ in taste (tied to tastes and social and cultural political experience ingredients) statements through con- ingredients consumption meat sumption experience experience compared to social and vegetarian meat and meat alternatives cultural food as part of ‘protein diversity’ experience

early meat alternatives model category intermediate categories substitution category

Figure 10.1: Core-periphery model of the category changes through the entrant of a substitution product. a lot of attention in categorisation research. Fast-moving consumer goods and particularly the role of supermarket fixtures provide a different set of challenges for the literature. Differences in categorisation for these various classifiable entities may need to be investigated more systematically. Next, substitutability emerged as fluid and context-dependent in this thesis, as some audiences may view entities as substitutable whereas others would not. The definitions of category membership which employ the concept of substitutability may need further exploration and investigation of boundary cases; in particular, this might focus on cases where single codes can lead different audiences or intermediaries to identify something is a ‘close substitute’ or not. While there appears to be an understanding of category formation and stability, little research has systematically investigated the fluidity of categories once established. Particularly, I found it difficult to investigate, assess, and make sense of the changes and tensions within the category that appeared to persist in the texts studied throughout the latter parts of this thesis. While Kennedy et al. (2010) has provided an extensive understanding of category dynamics, it appears as if those aspects can only ever be studied in hindsight, which makes category studies less useful in the context studied here. Overlaps with, for example, the transition studies literature might be explored in more detail (see e.g. Dolbec, 2015). Lastly, the conceptual framework outlined in this thesis, as well as the empirical

226 10.4. Contributions of this Thesis case have shown the relevance and insights which can be gained through a study of cultural constrains and norms in context of categorisation. Previous research has outlined the different ways in which strategic interests or power relations are intertwined with categorisation (e.g. Pontikes & Kim, 2017; Quinn & Munir, 2017), and many aspects in which categorisation is a relevant constituent of social, economic, or legal power (e.g. Schiller-Merkens, 2017; Vergne & Wry, 2014) When understanding categorisation as culture and discourse (e.g. Glynn & Navis, 2013; Grodal & Kahl, 2017), categorisation becomes a continuous struggle of association, dissociation, and actualisation of identities and valuation, as has been highlighted previously (e.g. Granqvist et al., 2012). Considering the potential impact of such insights, future work in categorisation studies may seek to enhance its societal relevance by investigating topics among others relating to social, cultural, ethical, or environmental challenges.

10.4 Contributions of this Thesis

Very broadly, this thesis contributes a critical appraisal of ‘technologies’ which have recently been hyped as a solution to one of the main sustainability challenges of our times: animal agriculture. Whilst this work does not engage with questions on whether they are more (environmentally) sustainable than the status quo, it provides a rich analysis of the origins, valuations, and significations of meat alternatives and how they engage with meat–the consumed product of animal agriculture–as a dominant aspect of contemporary British food culture. Meat alternatives almost exclusively obtain their valuation and meaning through appropriating (or seldom opposing) meat. This thesis highlights the ways in which this happens. Dominant meat culture is specifically reproduced through the occasions, practices, and individual meanings and memories associated to meat; conversely, contemporary understandings of meat are extended (and sometimes challenged) through novel tastes, textures, or formats. Whilst meat alternatives are often proposed as a sustainability technology, the analysis revealed that meat alternatives are in fact a voluntary, democratic tool to achieve dietary change. However, meat alternatives are not inherently political and often not exclusive to meat but instead seen as an addition or a transitional technology. This positions meat alternatives as some sort of ‘better meat’, which through common legitimation mechanisms reproduces understandings or dominant food culture such as what is good, normal, convenient, or appropriate food. It ultimately questions the ‘sustainability’ of meat alternatives in broader terms: Whilst sustainability gains on the short-term are likely, on the long-term meat may stay at the centre of dining plates as the stereotype of good food. Methodologically, this thesis builds on a balanced view of producers and audiences to the study of categorisation, and applied category studies to a food product.

227 10. Discussion

Furthermore, user-created taste test videos were identified as a novel and insightful data set. The insightfulness of this kind of data for this study indicates that it could be worthwhile paying more attention to this user created data in future studies on food, specifically in relation to food activism, practice theory, or food culture. On a theoretical level, this thesis has contributed to categorisation studies by emphasising the role of dominant culture in the category development process, thereby integrating insights on social change with the social process approach to categorisation. Specifically, it has been proposed that recursive amendments of categorisation keep categories in flux and interdependent. The implications of this approach were explored and illustrated in this thesis.

10.5 Limitations and Generalisability of this Research Project

This work comes with some shortcomings and limitations. While many constraints and limitations have already been identified at the outset and in the respective methodologies, this section briefly reflects on the possibility to generalise the identified findings. However, as Graneheim and Lundman (2004) argued, “the [author] can [only] give suggestions about transferability, but it is the reader’s decision whether or not the findings are transferable to another context” (p. 110). Although the results might appear tautological in parts, simply because meat alternatives are alternatives to meat, I hope that I could convince the reader of foundational trustworthiness (see again section 3.2)—through the methodological rigour applied throughout, as well as transparent and elaborate descriptions of research design, data, findings, and thought processes. An important limitation is that by design this is not a study of consumer acceptance, but quite the opposite: This research only engages with voices that are at least drawn towards meat alternatives or replacing meat. In this respect, this study must not be seen as a monitor of the state of meat alternatives in the population, let alone their perceptions of them. It is thus no complement to studies of consumer acceptance, but instead an insight into the meaning systems constructed by and through meat alternatives by consumers and producers of these foodstuffs. Thus, this study may be generalised towards social realities of consumers and producers similar to the ones studied. The focus on products instead of individual identities, diets, or practices might further bear issues concerning the generalisability of the results. In particular, one may argue that the relation of product categories and empirical observations is not always valid–for example, because these observations rather reflect diets or practices, which

228 10.5. Limitations and Generalisability of this Research Project may also be considered more important in challenging dominant culture. However, it has been shown that endeavours to maintain practices and change diets are strong motivations for the adoption of meat alternatives; in this respect, meat alternatives as artefacts or tools mediate people’s relation to diets and practices in ways that might not be identified with that clarity through a different empirical focus. So even though a focus on practices or diets might have led to fruitful results as well, and the approach used does not allow for generalisations about meat-free diets and practices in general, these concerns do not invalidate the relevance or validity of this research project. This study only investigated the case of meat alternatives. Therefore, findings— such as the identified mechanisms or challenge and reproduction–may need to be validated in follow-up studies before wider generalisation to other topic areas is possible. Particularly this might concern substitute products (such as e-cigarettes compared to conventional cigarettes), or morally contesting products (such as free-range in relation to conventional beef, see also Weber et al., 2008), albeit these two aspects may often overlap. A few methodological considerations also limit the power of this study. In particular, this accounts for the integration of the various data sources, as they not only carry different meanings but each data sets covers a different amount and complexity of texts; weighing between the data and extrapolations thereof are thus always influenced by the interpretive flexibility of the researcher. While it cannot be excluded that a different researcher with the same data might have come up with different conclusions, I hope to have convinced the reader at this point, that the findings are plausible, reasonable, and trustworthy, and might even warrant further investigation. Finally, this work would probably need to be reassessed in a few years; although the research has shown that meat alternatives may be considered an established category, the institutional dynamics in which the category is embedded appeared to change dramatically. For example, the influences of protein and plant-based food trends, as well as a commercial interest of the meat industry are equally to be taken into account for future studies, as the prospective shake up of category boundaries through the establishment of in-vitro meat in the market place. As such, the findings presented here only represent a snapshot of the cultural dynamics that surround meat alternatives. Whether the findings persist or dissolve, and whether the concerns raised and issues highlighted prove legitimate for society is upon follow-up studies to investigate.

229 10. Discussion

10.6 Rethinking Meat Alternatives? Implications and Recommendations

In this research project, I have shown that meat alternatives can take up a variety of flavours, textures, and levels of meatiness, as long as they are relatable to meat on the basis of its format definitions. These formats allow meat alternatives to fulfil the same niche on the plate, and therefore provide a disruption-free reduction, transition, or substitution of ‘normal meat’. However, meat alternatives through this process were also shown to reproduce dominant meat culture, and furthermore entrench carnism’s hegemonic status by delegitimising other ways of eating. In section 10.2.2, I suggested that the normalisation of meat alternatives deflects criticism on meat eating under the umbrella of ‘protein diversity’. I hope we have learned from the past, that changing times need changing ideologies, for example through the challenges of ecological crises (e.g. deforestation, eutrophication), environmental crises (e.g. climate change), ethical crises (e.g. the industrialisation of animal husbandry), economic crises (e.g. the continued non-regulation of banks after the 2008/2009 crisis), or a continuation of colonialism through means of free trade agreements. While it is hard to recognise such situation, whenever we are part of a dominant culture, counterculture, or group of people with a shared ideology, it is reasonable to remain critical however harmless or even beneficial such ideology may seem at first. Thus, in the following paragraphs I summarise a few implications and suggestions in relation to meat alternatives that may help to avoid any lock-in to dominant meat culture, or at least reduce its impact. Of course any of these can only be a thin description and vague ideas; because if we knew how to change people’s minds, someone would have done so to make us eat something unreflectedly. . . oh, wait! For people who eat meat alternatives—whether meat-eating, flexitarian, or vegan— it may be fruitful to think about meat alternatives critically once in a while. Why do you eat meat alternatives? Do you feel pressured by society to do so? Does engagement with meat alternatives make you more or less confident that you could live without them? How important are your past food experiences, and how conditioned is your liking? Have you tried eating foods that are less meat-like? — While I would hope that no one’s diet, hosting of dinner parties, or access to fridges would need to exclude anyone based on their food choices, there also needs to be some consideration to the patterns we are caught in. While freeing one’s mind from the inequalities of eating animals and the replicas which defend its status as normal and nice, it may help to reclaim meat terms, rephrase animal-based sayings, have roast-free christmas dinners or vegetable skewers on the barbecue, and–yes—maybe an ideally plant-based burger once in a while. Because considering the results of this research, what is to be

230 10.6. Rethinking Meat Alternatives? Implications and Recommendations avoided is that tacit knowledge and decade-old norms keep us symbolically hooked to carcasses. Somewhat contrary to my advice to consumers and any issues with incremental solutions, I suggest supermarkets and retailers for the sake of sustainability to dissolve the traditional distinction into meat-free and meat-based foods. While not everyone (me included) might appreciate going through the meat aisle, they may well do so if they really want to have meat alternatives; and certainly, a lot would be won for the environment if people ate root vegetables instead. And even better were, if people chose a plant-based counterpart over bovine flesh once in a while. It was suggested in section 9.5 that the taste of meat alternatives only matters if something like meat is expected. But if the opportunities by ‘less and better’ and ‘protein diversity’ approaches are taken seriously, maybe some people might accept a more diverse diet after all— just like stocking tinned beans next to meat. This could hopefully build more tacit knowledge and frees us of the negative preconceptions towards ‘meat’-free eating. I suggest that marketing of meat alternatives should refrain from blackmailing meat-free diets on terms of convenience, taste, or any other concerns for the sake of legitimacy. Instead, similar to what was as observed in section 9.2, they should frame meat alternatives positively as something that provides continuation and excitement. If it does not conform to the sensory norms of a meat category, this can be transformed into a strength framed in terms of diversity (see section 10.2.2). The recent trend concerning protein may offer opportunities for meat alternatives to be at par with meat by reducing the category to a nutritional element and thus reducing expectations. Labelling meat alternatives as plant-based may provide further promising avenues in setting the products apart from meat. Manufacturers of meat alternatives may continue as usual. But I would like to see even more diversity, and more emphasis on diversity in convenient, easy, and versatile food items. Just some products that don’t look, smell, or taste like meat. Some critical engagement with meat consumption more broadly would also be appreciated–even if it is just a carbon footprint indicator on the packaging, which makes audiences aware of such valuation criterion. Despite recent slow moves by single actors, the meat industry, particularly the British one, should get over their too long held idea that they could fight or sit out the vegetarian trend. Their German colleagues have shown that the established branding of meat manufacturers gives a strategic advantage even when producing meat-free foods–which might be a profitable development anyway, see again section 10.2.5. And their North American colleagues have demonstrated that with the huge revenues and income, investments and stakes in groundbreaking new developments in the food industry can be made, making their corporations future-proof and resilient. While we may not be certain that everyone will become vegetarian, a heavy carbon tax is just

231 10. Discussion a matter of time because the environmental impacts of rearing animals industrially may not stay in the blind spots of politicians and citizens forever. Lastly, I would ask policy-makers to discontinue the subsidies for animal agriculture, to stop financing and eating meat in public canteens or school lunches. One may educate chefs to become skilled with foods that go well beyond the ‘meat (alternative) and two veg’, and instead introduce people to the beauty, taste, and benefits of plant based foods. This may motivate people to engage with other ways of cooking and eating more broadly.

10.7 Conclusion and Future Research Directions

This research project has shed light on the somewhat special relationship between meat alternatives and their model counterpart. While not having immediate practical implications, the findings point towards an increasing normalisation of meat alternatives as another type of ‘meat’. Consumption of meat alternatives has been found affirmative of the social and cultural norms that traditionally only surrounded the consumption of animal flesh and reproduce stigmatisation of other ways of eating. However, meat alternatives are also often utilised as a tool to avoid the material foundations of meat culture, as their consumption is understood as part of a wider contribution to ethical or environmental causes, as well as a contributor to individual health. Notably however, as the combined amount of meat and meat alternatives in British diets is decreasing, dominant meat culture may nonetheless be in decline. This research may be an important guideline for those consumers not only concerned with the immediate implications of their actions, but also the symbolic and cultural power thereof. These findings support previous work that the material sub- version of eating meat can be (and usually is) individually important and relevant, yet that its compliance to the “gastro-ontological assumptions” (Nath & Prideaux, 2011, p. 136) can be criticised on symbolic grounds. While this work has hopefully made a strong contribution to open this debate, future research needs to identify whether the cultural reproduction of meat as highlighted here is actively resembled in people’s attitudes. Previous literature seems to give some tentative support for this hypothesis: In a study using a large Dutch demographic data set, Hoek et al. (2004) found that the explicit lifestyle attitudes–such as interest in cooking, purchasing motives, or ways of shopping–of consumers of meat alternatives range largely in between those of meat consumers and vegetarians; other research has elucidated the mental mechanisms underlying meat consumption, such as belief in the health benefits of meat or a denial of animal suffering (e.g. Rothgerber, 2014). Combining these findings with those made here, future research could map these traits onto consumers of meat alternatives: Do consumers of meat alternatives think higher of the nutritional benefits

232 10.7. Conclusion and Future Research Directions of meat than ‘other’ vegetarians and vegans? Do they think it is more normal or more socially accepted to eat meat? Do consumers of meat alternatives tend to deny animal suffering more than a vegetarian and vegan control group which consumes much less meat alternatives? Due to the nature of these beliefs and their presumed denial in individual consumers, inquiries would need to use implicit measures such as an Im- plicit Association Test, which has previously been used to compare vegetarians and meat eaters (Barnes-Holmes, Murtagh, Barnes-Holmes, & Stewart, 2010; De Houwer & De Bruycker, 2007; see also Shepherd, 2011). Other research could further investigate the relationships between the meat industry and meat alternatives. As argued above, meat alternatives can be understood as a logical next step following the intensification of animal agriculture. Related to the meat industry’s involvement in a meat alternatives category, a longitudinal international comparison of the constitution of the meat alternatives category may highlight different emergence patterns (see also section 4.6). These patterns may be further used to understand the involvement of hegemony in category emergence and negotiation as conceptualised in this thesis. In order to understand the emergence of meat alternatives better (and thus categorisation mechanisms for this category), it might also be helpful to investigate in more detail the similarities and differences throughout time: Such investigation might focus on meat alternatives in Victorian England, on the hype around ‘new protein foods’ in the 1970s that related to single-cell protein (SCP) and textured vegetable protein (TVP; see also section 4.3.1), and the most recent mainstreaming of meat alternatives. A comparison between the emergence and categorisation of in-vitro meat and plant-based meat alternatives may illustrate further the negotiation at work which attacks the integrity of what has been known as ‘meat’ for centuries. For example, Stephens (2010) has described in-vitro meat as an “ ‘as-yet undefined ontological object’ ”; almost a decade on from this paper, it seems that the search for adequate terminology has proceeded but its categorisation is still strongly debated. This debate is for example exemplified in a petition by the US Cattlemen’s Association to the US Department of Agriculture; it suggests forbidding the terms ‘beef’ and ‘meat’ to be used by similar products made from plants or grown in cell cultures (U.S. Cattlemen’s Association, 2018). This petition and its diverse responses from within and beyond the meat industry appear promising to study a hegemonic category and associated boundary work, as well as strategic categorisation. Such investigation could further underline the importance of intermediaries for categorisation (see e.g. Khaire, 2017). While all outlined avenues for future research have the propensity to extend category studies literature using the current case of meat alternatives, it has been established above that through their nature as both a substitution product and an at least somewhat moral choice, clear conclusions for category studies remain open. For

233 10. Discussion example, it needs to be established whether the substitution itself or the values behind the substitution are the main contributor for the findings that emerged in this thesis. This might be achieved by comparing different substitutes for established foods, such as margarine, artificial sweeteners, fat substitutes (all primarily health- motivated), plant-based milk (value- and health-motivated), or liquid meals (presumably convenience-motivated; for a comparison between the substitution stories of a few of these and other foodstuffs, see also Hoek, van Boekel, de Graaf, & Luning, 2010). But other consumption-related fields might also be interesting, such as the normalisation of smoking through e-cigarettes (e.g. Lucherini, Rooke, & Amos, 2018), or an affirmation of individual transport through electric vehicles. Lastly, the impact of a dominant ideology or thought on categorisation could be more firmly established; particularly the position of a hegemonic category as a prototype needs to be strengthened. Other promising aspects for future research include the effects of these hegemonic ideals on nascent categories on the one hand, and the adjustments of hegemony through these slight shifts on the other hand.

234 Epilogue

Already quoted in the introduction, “plant-based products [are said to] . . . ensure that meat remains a staple in the vast majority of the nation’s shopping trolleys” (foodmanufacture.co.uk, 2018). The assumptions and mechanisms outlined in this quote were interrogated and investigated throughout this research project. Whilst meat alternatives used to be part of a vegetarian/vegan subculture, they have become mainstream and fashionable replacements for meat. Not only do meat alternatives take the same shape as meat (i.e. burger, sausage, or roast), they also adapt the socio-cultural embedding and identity of meat. As such, in meat-free products, meat has found its perpetuator more so than its challenger, as it remains the centrepiece of the plate— the ‘staple’ of the nation that spans across time, space, and now also diet. Nath and Prideaux (2011) find that their study participants “can, metaphorically speaking, have their civilised [meat-free] burger and eat it too” by subverting the dominant culture through their material practices. However, as identified in this thesis, the use of meat alternatives may affirm or perpetuate dominant meat culture, rather than challenge it. The only challenge is to the invisible origins of meat and its consequences relating to animal agriculture. In particular, I have shown that cultural practices often remain unchallenged and even reinforced through the use of meat alternatives, as deviant ways of eating are sometimes stigmatised. The range of legitimate ‘meats’ will increase as the notion of what meat could be is extended through less meaty tastes and textures. Thus, meat alternatives are basically just like meat, albeit with more choice and individualised guilt, and without the animals. Driving this to the extreme, a Nestlé official has recently been quoted surrounding the introduction of ‘bleeding’ meat-free burgers to highlight the company’s“increased focus on tasty, authentic plant-based food” (Southey, 2019). I am honestly not sure what is ‘authentic’ in a plant-based food that tries so hard not to be plant-like. If such products ensure that meat remains in the shopping trolley, we should just call them what they really are: Animal-free meat. Not an alternative. Just ‘better meat’.

235

References

Academic References

Adams, C. J. (1990). The sexual politics of meat : A feminist-vegetarian critical theory. New York: Continuum.

Al Abri, M. S. (2016). ”What I ate” videos on YouTube. In Proceedings of the 28th Interna- tional Business Information Management Association Conference - Vision 2020: Innova- tion Management, Development Sustainability, and Competitive Economic Growth (pp. 1314–1316). Seville, Spain.

Alexy, O., & George, G. (2013). Category Divergence, Straddling, and Currency: Open In- novation and the Legitimation of Illegitimate Categories. Journal of Management Studies, 50(2), 173–203.

Altschul, A. M. (1967). Food Proteins: New Sources from Seeds Fortified and processed seed proteins have a role in satisfying human protein needs. Science, 158(3798), 221–226.

Apaiah, R. K., & Hendrix, E. M. T. (2005). Design of a supply chain network for pea-based novel protein foods. Journal of Food Engineering, 70(3), 383–391.

Armstrong, S. J., & Botzler, R. G. (Eds.). (2016). The Animal Ethics Reader. Taylor & Francis.

Asgar, M., Fazilah, A., Huda, N., Bhat, R., & Karim, A. (2010). Nonmeat Protein Alternatives as Meat Extenders and Meat Analogs. Comprehensive Reviews in Food Science and Food Safety, 9(5), 513–529.

Asioli, D., Aschemann-Witzel, J., Caputo, V., Vecchio, R., Annunziata, A., Næs, T., & Varela, P. (2017). Making sense of the ‘clean label’ trends: A review of consumer food choice behavior and discussion of industry implications. Food Research International, 99, 58–71.

Askanius, T., & Uldam, J. (2011). Online social media for radical politics: Climate change activism on YouTube. International Journal of Electronic Governance, 4(1/2), 69.

Ayo, N. (2012). Understanding health promotion in a neoliberal climate and the making of health conscious citizens. Critical Public Health, 22(1), 99–105.

Bajpai, K., & Weber, K. (2017). Privacy in Public: Translating the Category of Privacy to the

237 References

Digital Age. In R. Durand, N. Granqvist, & A. Tyllström (Eds.), Research in the Sociology of Organizations (Vol. 51, pp. 223–258). Emerald Publishing Limited.

Barnes-Holmes, D., Murtagh, L., Barnes-Holmes, Y., & Stewart, I. (2010). Using the Implicit Association Test and the Implicit Relational Assessment Procedure to Measure Attitudes Toward Meat and Vegetables in Vegetarians and Meat-Eaters. The Psychological Record, 60(2), 287–305.

Basáñez, T., Majmundar, A., Cruz, T. B., & Unger, J. B. (2018). Vapingassociated with healthy food words: A content analysis of Twitter. Addictive Behaviors Reports, 8, 147–153.

Battaglia Richi, E., Baumer, B., Conrad, B., Darioli, R., Schmid, A., & Keller, U. (2015). Health Risks Associated with Meat Consumption: A Review of Epidemiological Studies. International Journal for Vitamin and Nutrition Research, 85(1-2), 70–78.

Beardsworth, A., & Bryman, A. (2004). Meat consumption and meat avoidance among young people: An 11‐year longitudinal study. British Food Journal, 106 (4), 313–327.

Beardsworth, A., & Keil, T. (1992). The vegetarian option: Varieties, conversions, motives and careers. The Sociological Review, 40(2), 253–293.

Benford, R. D., & Snow, D. A. (2000). Framing Processes and Social Movements: An Over- view and Assessment. Annual Review of Sociology, 26, 611–639.

Benson, P. (2015). YouTube as text: Spoken interaction analysis and digital discourse. In R. H. Jones, A. Chik, & C. A. Hafner (Eds.), Discourse and Digital Practices: Doing discourse analysis in the digital age (pp. 93–108). Routledge.

Biesalski, H. K. (2005). Meat as a component of a healthy diet – are there any risks or benefits if meat is avoided in the diet? Meat Science, 70(3), 509–524.

Boje, D. M. (2001). Narrative methods for organizational & communication research. Sage.

Bouvard, V., Loomis, D., Guyton, K. Z., Grosse, Y., El Ghissassi, F., Benbrahim-Tallaa, L., . . . Straif, K. (2015). Carcinogenicity of consumption of red and processed meat. The Lancet Oncology, 16 (16), 1599–1600.

Brummette, J., & Fussell Sisco, H. (2018). Holy guacamole! Framing and the Chipotle con- tamination issue. Journal of Communication Management, 22(3), 280–295.

Burgess, J., Green, J., Jenkins, H., & Hartley, J. (2009). YouTube: Online Video and Particip- atory Culture. Oxford, UK: Polity Press.

Buscemi, F. (2014). From killing cows to culturing meat. British Food Journal, 116 (6), 952–964.

Buttlar, B., & Walther, E. (2018). Measuring the meat paradox: How ambivalence towards meat influences moral disengagement. Appetite, 128, 152–158.

238 Academic References

Chang, Y.-c., & Chang, Y.-j. (2018). Identity negotiation in the third space: An analysis of YouTube channels hosted by expatriates in Taiwan. Language and Intercultural Commu- nication, 1–16.

Chiles, R. M. (2013). If they come, we will build it: In vitro meat and the discursive struggle over future agrofood expectations. Agriculture and Human Values, 30(4), 511–523.

Chong, D., & Druckman, J. N. (2007). Framing Theory. Annual Review of Political Science, 10(1), 103–126.

Collingham, L. (2011). The Taste of War: WorldWarTwo and the Battle for Food. London: Allen Lane.

Conrad, C. F. (1978). A Grounded Theory of Academic Change. Sociology of Education, 51(2), 101–112.

Cornelissen, J. P., & Werner, M. D. (2014). Putting Framing in Perspective: A Review of Framing and Frame Analysis across the Management and Organizational Literature. The Academy of Management Annals, 8(1), 181–235.

Counihan, C., & Siniscalchi, V. (2013). Food activism: Agency, democracy and economy. A&C Black.

Cronin, J. K. (2018). Art for Animals: Visual Culture and Animal Advocacy, 1870–1914. Pennsylvania: Penn State University Press.

David, C. C., Atun, J. M., Fille, E., & Monterola, C. (2011). Finding Frames: Comparing Two Methods of Frame Analysis. Communication Methods and Measures, 5(4), 329–351.

Davies, J., & Lightowler, H. (1998). Plant‐based alternatives to meat. Nutrition & Food Science, 98(2), 90–94.

Declercq, J., Tulkens, S., & Van Leuven, S. (2018). The produsing expert consumer: Co- constructing, resisting and accepting health-related claims on social media in response to an infotainment show about food. Health, 1–19.

Deephouse, D. L. (1999). To be different, or to be the same? It’s a question(and theory) of strategic balance. Strategic Management Journal, 20(2), 147–166.

De Houwer, J., & De Bruycker, E. (2007). Implicit attitudes towards meat and vegetables in vegetarians and nonvegetarians. International Journal of Psychology, 42(3), 158–165.

Dibb, S., & Simkin, L. (2001). The marketing casebook: Cases and concepts (2nd ed.). Australia: Thomson.

DiMaggio, P. J., & Powell, W. W. (1983). The iron cage revisited: Institutional isomorphism and collective rationality in organizational fields. American Sociological Review, 48(2), 147–160.

239 References

Dolbec, P.-Y. (2015, November). How Do Mainstream Cultural Categories Emerge: A Multi- Level Analysis of the Creation of Electronic Dance Music (Unpublished Doctoral Disser- tation). York University, Toronto, ON, CA.

Douglas, M., & Nicod, M. (1974). Taking the biscuit: The structure of British meals. New Society, 19, 744–747.

Drost, E. A. (2011). Validity and Reliability in Social Science Research. Education Research and Perspectives, 38(1), 105–124.

Durand, R., Granqvist, N., & Tyllström, A. (2017a). From Categories to Categorization: A Social Perspective on Market Categorization. In R. Durand, N. Granqvist, & A. Tyllström (Eds.), Research in the Sociology of Organizations (Vol. 51, pp. 3–30). Emerald Publishing Limited.

Durand, R., Granqvist, N., & Tyllström, A. (Eds.). (2017b). From categories to categorization: Studies in sociology, organizations and strategy at the .crossroads Emerald Publishing.

Durand, R., & Khaire, M. (2017). Where Do Market Categories Come From and How? Distinguishing Category Creation From Category Emergence. Journal of Management, 43(1), 87–110.

Durand, R., & Paolella, L. (2013). Category Stretching: Reorienting Research on Categories in Strategy, Entrepreneurship, and Organization Theory: Reorienting Research on Cat- egories. Journal of Management Studies, 50(6), 1100–1123.

Durand, R., Rao, H., & Monin, P. (2007). Code and conduct in French cuisine: Impact of code changes on external evaluations. Strategic Management Journal, 28(5), 455–472.

Ehgartner, U. (2019). Environmentally and socially responsible consumption? A study on the discursive construction of values towards sustainability. (Unpublished Doctoral Disserta- tion). University of Manchester, Manchester, UK.

Ehrenberg, A. S. C. (1988). Repeat-buying: Facts, theory and application (New ed.). London: Griffin.

Eisenhardt, K. M. (1989). Building theories from case study research. Academy of Manage- ment Review, 14(4), 532–550.

Ekman, M. (2014). The dark side of online activism: Swedish right-wing extremist video activism on YouTube. MedieKultur: Journal of Media and Communication Research, 30(56), 79–99.

Elo, S., & Kyngäs, H. (2008). The qualitative content analysis process. Journal of Advanced Nursing, 62(1), 107–115.

Elzerman, J. E., Hoek, A. C., van Boekel, M. A. J. S., & Luning, P. A. (2011). Consumer

240 Academic References

acceptance and appropriateness of meat substitutes in a meal context. Food Quality and Preference, 22(3), 233–240.

Elzerman, J. E., Hoek, A. C., van Boekel, M. J. A. S., & Luning, P. A. (2015). Appropriateness, acceptance and sensory preferences based on visual information: A web-based survey on meat substitutes in a meal context. Food Quality and Preference, 42, 56–65.

Elzerman, J. E., van Boekel, M. A. J. S., & Luning, P. A. (2013). Exploring meat substitutes: Consumer experiences and contextual factors. British Food Journal, 115(5), 700–710.

Enax, L., Krajbich, I., & Weber, B. (2016). Salient nutrition labels increase the integration of health attributes in food decision-making. Judgment and Decision Making, 11(5), 460–471.

Entman, R. M. (1993). Framing: Toward Clarification of a Fractured Paradigm. Journal of Communication, 43(4), 51–58.

Fairlie, S. (2010). Meat: A benign extravagance. White River Junction, Vt: Chelsea Green Pub.

Fay, M. (2013). Cyclical patterns in the content of advertisements. European Journal of Mar- keting, 40(1/2), 198–217.

Fiddes, N. (1992). Meat: A natural symbol (Reprinted). London: Routledge.

Filippini, M. (2017). Using Gramsci: A New Approach. (P. J. Barr, Trans.). Pluto Press.

Finnigan, T. J. A. (2011). Mycoprotein: Origins, production and properties. In G. O. Phillips & P. A. Williams (Eds.), Handbook of Food Proteins (pp. 335–352). Cambridge, UK ; Philadelphia: Woodhead Publishing.

Flail, G. J. (2006). The Sexual Politics of Meat Substitutes (Unpublished Doctoral Disserta- tion). Georgia State University; retrieved February 22, 2016, from http://scholarworks.g su.edu/english_diss/9/, Atlanta, GA.

Flick, U. (2018). Triangulation in Data Collection. In The SAGE Handbook of Qualitative Data Collection (pp. 527–544). London, UK: SAGE Publications Ltd.

Forchtner, B., & Tominc, A. (2017). Kalashnikov and Cooking-spoon: Neo-Nazism, Vegan- ism and a Lifestyle Cooking Show on YouTube. Food, Culture & Society, 20(3), 415–441.

Fowles, J. (1996). Advertising and Popular Culture. SAGE.

Fox, J., & Warber, K. M. (2015). Queer Identity Management and Political Self-Expression on Social Networking Sites: A Co-Cultural Approach to the Spiral of Silence. Journal of Communication, 65(1), 79–100.

Fox, N., & Ward, K. J. (2008). You are what you eat? Vegetarianism, health and identity. Social Science & Medicine, 66 (12), 2585–2595.

241 References

Fox, T. (2004). Selling the Symbol: The Vegetarian Society’s Seedling Licence Scheme. In 36th WorldVegetarianCongress. Santinho Beach, Florianópolis, Brazil.

Frank, D., Oytam, Y., & Hughes, J. (2017). Sensory Perceptions and New Consumer Attitudes to Meat. In P. P. Purslow (Ed.), New Aspects of Meat Quality (pp. 667–698). Woodhead Publishing.

Freeland-Graves, J. H., Greninger, S. A., Graves, G. R., & Young, R. K. (1986). Health practices, attitudes, and beliefs of vegetarians and nonvegetarians. Journal of the American Dietetic Association, 86 (7), 913–918.

Fuentes, C., & Fuentes, M. (2017). Making a market for alternatives: Marketing devices and the qualification of a vegan milk substitute. Journal of Marketing Management, 33(7-8), 529–555.

Gibson, M. (2016). YouTube and bereavement vlogging: Emotional exchange between strangers. Journal of Sociology, 52(4), 631–645.

Giglietto, F., Rossi, L., & Bennato, D. (2012). The Open Laboratory: Limits and Possibilities of Using Facebook, Twitter,and YouTubeas a Research Data Source. Journal of Technology in Human Services, 30(3-4), 145–159.

Giorgi, S., Lockwood, C., & Glynn, M. A. (2015). The Many Faces of Culture: Making Sense of 30 Yearsof Research on Culture in Organization Studies. The Academy of Management Annals, 9(1), 1–54.

Glynn, M. A., & Navis, C. (2013). Categories, Identities, and Cultural Classification: Mov- ing Beyond a Model of Categorical Constraint. Journal of Management Studies, 50(6), 1124–1137.

Goffman, E. (1974). Frame analysis: An essay on the organization of experience. Cambridge, MA, US: Harvard University Press.

Goodman, D., Sorj, B., & Wilkinson, J. (1987). From farming to biotechnology: A theory of agro-industrial development. Oxford ; New York, N.Y: Basil Blackwell.

Gramsci, A. (1971). Selections from the Prison Notebooks. (Q. Hoare & G. N. Smith, Eds.) (Later printing edition). New York: International Publishers Co.

Graneheim, U. H., & Lundman, B. (2004). Qualitative content analysis in nursing research: Concepts, procedures and measures to achieve trustworthiness. Nurse Education Today, 24(2), 105–112.

Granqvist, N., Grodal, S., & Woolley, J. L. (2012). Hedging Your Bets: Explaining Executives’ Market Labeling Strategies in Nanotechnology. Organization Science, 24(2), 395–413.

Granqvist, N., & Ritvala, T. (2016). Beyond prototypes: Drivers of market categorization in functional foods and nanotechnology. Journal of Management Studies, 53(2), 210–237.

242 Academic References

Gregory, J. (2007). Of Victorians and Vegetarians: The VegetarianMovement in Nineteenth- Century Britain. London, UK: I.B.Tauris.

Gregory, J. R. T. E. (2002). The Vegetarian Movement in Britain c. 1840-1901. A Study of its Development, Personel and Wider Connections. (Doctoral Dissertation). University of Southampton, Southampton.

Grodal, S., Gotsopoulos, A., & Suarez, F. F. (2015). The coevolution of technologies and categories during industry emergence. Academy of Management Review, 40(3), 423–445.

Grodal, S., & Kahl, S. J. (2017). The Discursive Perspective of Market Categorization: Inter- action, Power, and Context. In R. Durand, N. Granqvist, & A. Tyllström (Eds.), Research in the Sociology of Organizations (Vol. 51, pp. 151–184). Emerald Publishing Limited.

Groß, M. (2016). Mykoprotein als Fleischersatz. Nachrichten Aus Der Chemie, 64(4), 403–405.

Guest, G., Bunce, A., & Johnson, L. (2006). How Many Interviews Are Enough?: An Exper- iment with Data Saturation and Variability. Field Methods, 18(1), 59–82.

Gunter, B. (2016). Food Advertising. Cham: Springer International Publishing.

Gustafsson, N. (2010). This time it’s personal: Social networks, viral politics and identity management. In Emerging Practices in Social Networking and Cybercultures (pp. 35–44). Amsterdam/New york: Rodopi.

Hajer, M. A. (1996). Ecological Modernisation as Cultural Politics. In S. Lash, B. Szerszyn- ski, & B. Wynne (Eds.), Risk, environment and modernity: Towards a new ecology (pp. 246–268). London: Sage.

Hall, M., Gough, B., & Seymour-Smith, S. (2012). ‘I’m METRO, NOT Gay!’: A Discursive Analysis of Men’s Accounts of Makeup Use on YouTube. The Journal of Men’s Studies, 20(3), 209–226.

Halonen, M. (2015). Presenting ‘pissis girls’: Categorisation in a social media video. Discourse, Context & Media, 8, 55–62.

Hannan, M. T., Pólos, L., & Carroll, G. R. (2007). Logics of Organization Theory: Audiences, Codes, and Ecologies. Princeton University Press.

Heffernan, O. (2017). Sustainability: A meaty issue. Nature, 544(7651), S18–S20.

Hoek, A. C., Elzerman, J. E., Hageman, R., Kok, F. J., Luning, P. A., & Graaf, C. de. (2013). Are meat substitutes liked better over time? A repeated in-home use test with meat substitutes or meat in meals. Food Quality and Preference, 28(1), 253–263.

Hoek, A. C., Luning, P. A., Stafleu, A., & de Graaf, C. (2004). Food-related lifestyle and health attitudes of Dutch vegetarians, non-vegetarian consumers of meat substitutes, and meat consumers. Appetite, 42(3), 265–272.

243 References

Hoek, A. C., Luning, P. A., Weijzen, P., Engels, W., Kok, F. J., & de Graaf, C. (2011). Re- placement of meat by meat substitutes. A survey on person- and product-related factors in consumer acceptance. Appetite, 56 (3), 662–673.

Hoek, A. C., van Boekel, M. A. J. S., de Graaf, C., & Luning, P. A. (2010). Food substitutes: A retrospective view on the roles of food technology, governmental policy, and the consumer in development and acceptance over time. In Will Novel Protein Foods beat meat? Con- sumer acceptance of meat substitutes - a multidisciplinary research approach (pp. 25–48). Thesis Wageningen University, The Netherlands.

Hoek, A. C., van Boekel, M. A., Voordouw, J., & Luning, P. A. (2011). Identification of new food alternatives: How do consumers categorize meat and meat substitutes? Food Quality and Preference, 22(4), 371–383.

Hoek, J., & Gendall, P. (2006). Advertising and Obesity: A Behavioral Perspective. Journal of Health Communication, 11(4), 409–423.

House, J. (2018). Insects are not ‘the new sushi’: Theories of practice and the acceptance of novel foods. Social & Cultural Geography, 0(0), 1–22.

Hsieh, H.-F., & Shannon, S. E. (2005). Three Approaches to Qualitative Content Analysis. Qualitative Health Research, 15(9), 1277–1288.

Hsu, G., Roberts, P. W., & Swaminathan, A. (2012). Evaluative Schemas and the Mediating Role of Critics. Organization Science, 23(1), 83–97.

Huang, R. (2016). RQDA: R-based Qualitative Data Analysis. retrieved from http://rqda.r- forge.r-project.org/.

Izmirli, S., & Phillips, C. J. C. (2011). The relationship between student consumption of animal products and attitudes to animals in Europe and Asia. British Food Journal, 113(3), 436–450.

Jamerson, H. (2009). Intoxicators, educators, and gatekeepers: The enactment of symbolic boundaries in Napa Valley wineries. Poetics, 37(4), 383–398.

Janssen, M., & Hamm, U. (2011). Consumer perception of different organic certification schemes in five European countries. Organic Agriculture, 1(1), 31–43.

Jick, T. D. (1979). Mixing Qualitative and Quantitative Methods: Triangulation in Action. Administrative Science Quarterly, 24(4), 602–611.

Johnson, L. A., Myers, D. J., & Burden, D. J. (1992). Soy protein’s history, prospects in food, feed. International News on Fats, Oils and Related Materials, 3(4), 429–444.

Jones, C., Maoret, M., Massa, F. G., & Svejenova, S. (2011). Rebels with a Cause: Formation, Contestation, and Expansion of the De Novo Category ‘Modern Architecture,’ 1870–1975. Organization Science, 23(6), 1523–1545.

244 Academic References

Jönsson, E., Linné, T., & McCrow-Young, A. (2019). Many Meats and Many Milks? The On- tological Politics of a Proposed Post-animal Revolution. Science as Culture, 28(1), 70–97.

Joy, M. (2009). Why WeLove Dogs, Eat Pigs, and WearCows: An Introduction to Carnism. Berkeley, Calif.; Enfield: Conari Press.

Jungherr, A. (2016). Twitter use in election campaigns: A systematic literature review. Journal of Information Technology & Politics, 13(1), 72–91.

Kalamus, W. (2019). Collective Awareness Raising Towards a Plant-Based Diet Through Social Networking Sites. Environmental, Health, and Business Opportunities in the New Meat Alternatives Market, 283–296.

Kanerva, M. (2016). Meat eating as a practice and the acceptance of radical change. In Food futures: Ethics, science and culture (pp. 406–426). Wageningen Academic Publishers.

Kennedy, M. T. (2008). Getting Counted: Markets, Media, and Reality. American Sociolo- gical Review, 73(2), 270–295.

Kennedy, M. T., & Fiss, P. C. (2013). An Ontological Turn in Categories Research: From Standards of Legitimacy to Evidence of Actuality. Journal of Management Studies, 50(6), 1138–1154.

Kennedy, M. T., Lo, J. Y.-C., & Lounsbury, M. (2010). Category currency: The changing value of conformity as a function of ongoing meaning construction. In Categories in Markets: Origins and Evolution (Vol. 31, pp. 369–397). Emerald Group Publishing Limited.

Khaire, M. (2017). The Importance of Being Independent: The Role of Intermediaries in Creating Market Categories. In R. Durand, N. Granqvist, & A. Tyllström (Eds.), Research in the Sociology of Organizations (Vol. 51, pp. 259–293). Emerald Publishing Limited.

Khaire, M., & Wadhwani, R. D. (2010). Changing Landscapes: The Construction of Meaning and Value in a New Market Category–Modern Indian Art. Academy of Management Journal, 53(6), 1281–1304.

Kierans, K. (2017). Common Themes in the VeganYouTube Movement: Community, Health, Ethics, Regret and Education (Unpublished Bachelor Thesis). Thompson Rivers Univer- sity; retrieved December 10, 2018, from https://core.ac.uk/download/pdf/84872074.pdf, Kamloops, BC, CA.

Kolo, C., & Haumer, F. (2018). Social media celebrities as influencers in brand communication: An empirical study on influencer content, its advertising relevance and audience expectations. Social Media Marketing, 6 (3), 273–282.

Kondracki, N. L., Wellman, N. S., & Amundson, D. R. (2002). Content Analysis: Review of Methods and Their Applications in Nutrition Education. Journal of Nutrition Education and Behavior, 34(4), 224–230.

245 References

Kornberger, M. (2017). The Values of Strategy: Valuation Practices, Rivalry and Strategic Agency. Organization Studies, 38(12), 1753–1773.

Kotchen, M. J. (2009). Offsetting green guilt. Stanford Social Innovation Review, 7(2), 26–31.

Kovács, B., & Hannan, M. T. (2010). The consequences of category spanning depend on contrast. In G. Hsu, G. Negro, & Ö. Koçak (Eds.), Categories in Markets: Origins and Evolution (Vol. 31, pp. 175–201). Emerald Group Publishing Limited.

Koylu, C. (2018). Modeling and visualizing semantic and spatio-temporal evolution of topics in interpersonal communication on Twitter. International Journal of Geographical In- formation Science, 1–28.

Krintiras, G. A., Gadea Diaz, J., van der Goot, A. J., Stankiewicz, A. I., & Stefanidis, G. D. (2016). On the use of the Couette Cell technology for large scale production of textured soy-based meat replacers. Journal of Food Engineering, 169, 205–213.

Krippendorff, K. (1980). Content analysis: An introduction to its methodology. Newbury Park, CA, US: Sage Publications.

Krippendorff, K. (2004). Reliability in Content Analysis. Human Communication Research, 30(3), 411–433.

Kubberød, E., Ueland, Ø., Rødbotten, M., Westad, F., & Risvik, E. (2002). Gender specific preferences and attitudes towards meat. Food Quality and Preference, 13(5), 285–294.

Kuilman, J. G., & van Driel, H. (2013). You too, Brutus? Category demise in Rotterdam warehousing, 1871-2011. Industrial and Corporate Change, 22(2), 511–548.

Kvale, S. (2007). Doing interviews. London: SAGE Publications.

Langley, A. (1999). Strategies for theorizing from process data. Academy of Management Review, 24(4), 691–710.

Langley, A., & Abdallah, C. (2011). Templates and Turns in Qualitative Studies of Strategy and Management. In Building Methodological Bridges (Vol. 6, pp. 201–235). Emerald Group Publishing Limited.

Laranjo, L., Arguel, A., Neves, A. L., Gallagher, A. M., Kaplan, R., Mortimer, N., . . . Lau, A. Y. S. (2015). The influence of social networking sites on health behavior change: A systematic review and meta-analysis. Journal of the American Medical Informatics Association, 22(1), 243–256.

Laurier, E. (2015). YouTube: Fragments of a video-tropic atlas. Area, 48(4), 488–495.

Leroy, F., & Praet, I. (2015). Meat traditions. The co-evolution of humans and meat. Appetite, 90, 200–211.

246 Academic References

Levy, D., & Scully, M. (2007). The Institutional Entrepreneur as Modern Prince: The Stra- tegic Face of Power in Contested Fields. Organization Studies, 28(7), 971–991.

Liu, Y., & Lopez, R. A. (2016). The impact of social media conversations on consumer brand choices. Marketing Letters, 27(1), 1–13.

Lorentzen, D. G., & Nolin, J. (2017). Approaching Completeness: Capturing a Hashtagged Twitter Conversation and Its Follow-On Conversation. Social Science Computer Review, 35(2), 277–286.

Lucherini, M., Rooke, C., & Amos, A. (2018). E-cigarettes, vaping and performativity in the context of tobacco denormalisation. Sociology of Health & Illness, 40(6), 1037–1052.

Malav, O., Talukder, S., Gokulakrishnan, P., & Chand, S. (2015). Meat Analog: A Review. Critical Reviews in Food Science and Nutrition, 55(9), 1241–1245.

Mann, A. (2017). Hashtag activism and the right to food in Australia. In Digital Food Activism (pp. 186–202). Routledge.

Matthes, J., & Kohring, M. (2008). The Content Analysis of Media Frames: Toward Improv- ing Reliability and Validity. Journal of Communication, 58(2), 258–279.

Mattick, C. S., Wetmore, J. M., & Allenby, B. R. (2015). An Anticipatory Social Assessment of Factory-Grown Meat. IEEE Technology and Society Magazine, 34(1), 56–64.

McAfee, A. J., McSorley, E. M., Cuskelly, G. J., Moss, B. W., Wallace, J. M. W., Bonham, M. P., & Fearon, A. M. (2010). Red meat consumption: An overview of the risks and benefits. Meat Science, 84(1), 1–13.

McHugh, M. L. (2013). The Chi-square test of independence. Biochemia Medica, 23(2), 143–149.

McIlveen, H., Abraham, C., & Armstrong, G. (1999). Meat avoidance and the role of replacers. Nutrition & Food Science, 99(1), 29–36.

McKendrick, D. G., Jaffee, J., Carroll, G. R., & Khessina, O. M. (2003). In the Bud? Disk Array Producers as a (Possibly) Emergent Organizational Form. Administrative Science Quarterly, 48(1), 60–93.

Michaut, A. M. (2004). Consumer response to innovative products: With application to foods (Unpublished Doctoral Dissertation). Wageningen University; retrieved May 23, 2017, from http://library.wur.nl/WebQuery/wurpubs/fulltext/121564, Wageningen, NL.

Miller, G. A. (1995). WordNet: A lexical database. Communications of the ACM, 38(11), 39–41.

Mitchell, J. (1999). The British main meal in the 1990s: Has it changed its identity? British Food Journal, 101(11), 871–883.

Miyazaki, A. D., & Taylor, K. A. (2008). Researcher Interaction Biases and Business Ethics

247 References

Research: Respondent Reactions to Researcher Characteristics. Journal of Business Ethics, 81(4), 779–795.

Morris, C. (2018). ‘Taking the Politics out of Broccoli’: Debating (De)Meatification in UK National and Regional Newspaper Coverage of the Meat Free Mondays Campaign. Soci- ologia Ruralis, 58(2), 433–452.

Mowlabocus, S. (2018). ‘Let’s get this thing open’: The pleasures of unboxing videos. European Journal of Cultural Studies.

Murcott, A. (1982). On the social significance of the ”cooked dinner” in South Wales. Social Science Information, 21(4-5), 677–696.

Nath, J. (2011). Gendered fare?: A qualitative investigation of alternative food and masculin- ities. Journal of Sociology, 47(3), 261–278.

Nath, J., & Prideaux, D. (2011). The Civilised Burger: Meat Alternatives as a Conversion Aid and Social Instrument for Australian Vegetarians and Vegans. Australian Humanities Review, (51), 135–151.

Navis, C., & Glynn, M. A. (2010). How new market categories emerge: Temporal dynamics of legitimacy, identity, and entrepreneurship in satellite radio, 1990–2005. Administrative Science Quarterly, 55(3), 439–471.

Negro, G., Koçak, Ö., & Hsu, G. (2010). Research on categories in the sociology of organizations. In Categories in Markets: Origins and Evolution (Vol. 31, pp. 3–35). Emerald Group Publishing Limited.

Nettleton, S., & Green, J. (2014). Thinking about changing mobility practices: How a social practice approach can help. Sociology of Health & Illness, 36 (2), 239–251.

Newman, T. P. (2017). Tracking the release of IPCC AR5 on Twitter: Users, comments, and sources following the release of the Working Group I Summary for Policymakers. Public Understanding of Science, 26 (7), 815–825.

Nijdam, D., Rood, T., & Westhoek, H. (2012). The price of protein: Review of land use and carbon footprints from life cycle assessments of animal food products and their substitutes. Food Policy, 37(6), 760–770.

Noe, E., & Alrøe, H. F. (2011). Quality, coherence and co-operation: A framework for studying the mediation of qualities in food networks and collective marketing strategies. Inter- national Journal of Sociology of Agriculture and Food, 18(1), 12–27.

Noonan, R. K. (1995). Women against the state: Political opportunities and collective action frames in chile’s transition to democracy. Sociological Forum, 10(1), 81–111.

Owens, J. (2016). Vegetarian propaganda and the Great War (Unpublished Project Report,

248 Academic References

BA History). Lancaster University; retrieved January 12, 2018, from https://www.academ ia.edu/31737175/Vegetarian_propaganda_and_the_Great_War, Lancaster, UK.

Pace, S. (2008). YouTube: An opportunity for consumer narrative analysis? Qualitative Market Research: An International Journal, 11(2), 213–226.

Pachucki, M. C. (2012). Classifying quality: Cognition, interaction, and status appraisal of art museums. Poetics, 40(1), 67–90.

Paolella, L., & Durand, R. (2015). Category Spanning, Evaluation, and Performance: Revised Theory and Test on the Corporate Law Market. Academy of Management Journal, 59(1), 330–351.

Paolella, L., & Sharkey, A. (2017). Forging Consensus: An Integrated View of How Categor- ies Shape the Perception of Organizational Identity. In R. Durand, N. Granqvist, & A. Tyllström (Eds.), Research in the Sociology of Organizations (Vol. 51, pp. 327–353). Emerald Publishing Limited.

Paquette, H. (2013). Social media as a marketing tool: A literature review.

Patterson, A. N. (2018). YouTube Generated Video Clips as Qualitative Research Data: One Researcher’s Reflections on the Process. Qualitative Inquiry, 24(10), 759–767.

Pearson, K. (1900). X. On the criterion that a given system of deviations from the probable in the case of a correlated system of variables is such that it can be reasonably supposed to have arisen from random sampling. The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science, 50(302), 157–175.

Pedersen, J. S., & Dobbin, F. (1997). The social invention of collective actors: On the rise of the organization. American Behavioral Scientist, 40(4), 431–443.

Pedersen, J. S., & Dobbin, F. (2006). In Search of Identity and Legitimation: Bridging Organizational Culture and Neoinstitutionalism. American Behavioral Scientist, 49(7), 897–907.

Piazza, J., Ruby, M. B., Loughnan, S., Luong, M., Kulik, J., Watkins, H. M., & Seigerman, M. (2015). Rationalizing meat consumption. The 4Ns. Appetite, 91, 114–128.

Pimentel, D., & Pimentel, M. (2003). Sustainability of meat-based and plant-based diets and the environment. The American Journal of Clinical Nutrition, 78(3), 660S–663S.

Piqueras-Fiszman, B., & Spence, C. (2015). Sensory expectations based on product-extrinsic food cues: An interdisciplinary review of the empirical evidence and theoretical accounts. Food Quality and Preference, 40, 165–179.

Plevin, A. (1997). Green Guilt: An Effective Rhetoric or Rhetoric in Transition? Technical Communication Quarterly, 6 (2), 125–139.

249 References

Pollay, R. W. (1985). The Subsiding Sizzle: A Descriptive History of Print Advertising, 1900- 1980. Journal of Marketing, 49(3), 24–37.

Polos, L. (2002). Foundations of a theory of social forms. Industrial and Corporate Change, 11(1), 85–115.

Pontikes, E. G. (2012). Fitting in or Starting New? Invention, Leniency, and Category Emer- gence in the Software Industry, 1990-2002. SSRN Electronic Journal.

Pontikes, E. G., & Barnett, W. P. (2015). The Persistence of Lenient Market Categories. Or- ganization Science, 26 (5), 1415–1431.

Pontikes, E. G., & Kim, R. (2017). Strategic Categorization. In R. Durand, N. Granqvist, & A. Tyllström (Eds.), Research in the Sociology of Organizations (Vol. 51, pp. 71–111). Emerald Publishing Limited.

Porac, J. F., Thomas, H., & Baden-Fuller, C. (1989). Competitive Groups as Cognitive Com- munities: The Case of Scottish Knitwear Manufacturers*. Journal of Management Studies, 26 (4), 397–416.

Porac, J., Rosa, J. A., Spanjol, J., & Saxon, M. (2001). America’s family vehicle: Path creation in the US minivan market. Path Dependence and Creation, 213–242.

Potts, A., & Adams, C. J. (2010). Why feminist-vegan now? Feminism & Psychology, 20(3), 302–317.

Proferes, N. J., & Zimmer, M. (2014). A topology of Twitter research: Disciplines, methods, and ethics. Aslib Journal of Information Management, 66 (3), 250–261.

Quinn, Q. C., & Munir, K. A. (2017). Hybrid Categories as Political Devices: The Case of Impact Investing in Frontier Markets. In R. Durand, N. Granqvist, & A. Tyllström (Eds.), Research in the Sociology of Organizations (Vol. 51, pp. 113–150). Emerald Publishing Limited.

Quist, J. (2008). Does backcasting lead to system innovations to sustainability? The case of meat alternatives and novel protein foods. In Long term policies: Governing socio-ecological change. Berlin.

Quist, J., Thissen, W., & Vergragt, P. J. (2011). The impact and spin-off of participatory backcasting: From vision to niche. Technological Forecasting and Social Change, 78(5), 883–897.

Quoidbach, J., Gilbert, D. T., & Wilson, T. D. (2013). The End of History Illusion. Science, 339(6115), 96–98.

Rao, H., Monin, P., & Durand, R. (2003). Institutional Change in Toque Ville: Nouvelle Cuisine as an Identity Movement in French Gastronomy. American Journal of Sociology, 108(4), 795–843.

250 Academic References

Rao, H., Morrill, C., & Zald, M. N. (2000). Power Plays: How Social Movements and Collective Action Create New Organizational Forms. Research in Organizational Behavior, 22, 237–281.

Rhee, E. Y., Lo, J. Y., Kennedy, M. T., & Fiss, P. C. (2017). Things That Last? Category Creation, Imprinting, and Durability. In R. Durand, N. Granqvist, & A. Tyllström (Eds.), Research in the Sociology of Organizations (Vol. 51, pp. 295–325). Emerald Publishing Limited.

Riffe, D., Lacy, S., & Fico, F. (2014). Analyzing media messages: Using quantitative content analysis in research (Third edition). New York: Routledge/Taylor & Francis Group.

Rindova, V., Dalpiaz, E., & Ravasi, D. (2011). A Cultural Quest: A Study of Organiza- tional Use of New Cultural Resources in Strategy Formation. Organization Science, 22(2), 413–431.

Roberts, C., & Geels, F. W. (2018). Public Storylines in the British Transition from Rail to Road Transport (1896–2000): Discursive Struggles in the Multi-Level Perspective. Science as Culture, 27(4), 513–542.

Rödl, M. B. (2018a). Marketing Meat Alternatives: Meat Myths and Their Replication in Advertising for Plant-Based Meat. In D. Bogueva, D. Marinova, & T. Raphaely (Eds.), Handbook of Research on Social Marketing and Its Influence on Animal Origin Food Product Consumption: (pp. 327–343). Hershey, PA: IGI Global.

Rödl, M. B. (2019a). What’s new? A History of Meat Alternatives in the UK. In D. Bogueva, D. Marinova, T. Raphaely, & K. Schmidinger (Eds.), Environmental, Health, and Business Opportunities in the New Meat Alternatives Market (pp. 202–217). Hershey, PA, USA: IGI Global.

Rosa, J. A., Porac, J. F., Spanjol, J., & Saxon, M. S. (1999). Socio-Cognitive Dynamics in a Product Market.

Rothgerber, H. (2014). Efforts to overcome vegetarian-induced dissonance among meat eaters. Appetite, 79, 32–41.

Rousseau, S. (2012). Food and social media: You are what you tweet. Lanham: Altamira Press/Rowman & Littlefield.

Ruby, M. B. (2012). Vegetarianism. A blossoming field of study. Appetite, 58(1), 141–150.

Ruef, M., & Patterson, K. (2009). Credit and Classification: The Impact of Industry Bound- aries in Nineteenth-Century America. Administrative Science Quarterly, 54(3), 486–520.

Sadler, M. (1988). Quorn Case Study: Enhancing the Meat Alternatives Market by Capitaliz- ing on the Health Trend. Nutrition & Food Science, 88(3), 9–11.

251 References

Sadler, M. J. (2004). Meat alternatives — market developments and health benefits. Trends in Food Science & Technology, 15(5), 250–260.

Saunders, B., Sim, J., Kingstone, T., Baker, S., Waterfield, J., Bartlam, B., . . . Jinks, C. (2018). Saturation in qualitative research: Exploring its conceptualization and operationalization. Quality & Quantity, 52(4), 1893–1907.

Sayer, A. (2000). Realism and Social Science. London, UK: SAGE Publications Ltd.

Schäfer, M., Herde, A., & Kropp, C. (2007). Life events as turning points for sustainable nutrition. Proceedings: SCP Cases in the Field of Food, Mobility and Housing, 115–130.

Schiller-Merkens, S. (2017). Will Green Remain the New Black? Dynamics in the Self- Categorization of Ethical Fashion Designers. Historical Social Research / Historische Sozi- alforschung, 42(1), 211–237.

Schneider, T., Eli, K., Dolan, C., & Ulijaszek, S. (2017). Introduction: Digital food activism– food transparency one byte/bite at a time? In Digital Food Activism (pp. 1–24). Rout- ledge.

Sexton, A. (2016). Alternative Proteins and the (Non)Stuff of ‘Meat’. Gastronomica: The Journal of Critical Food Studies, 16 (3), 66–78.

Sexton, A. E. (2018). Eating for the post-Anthropocene: Alternative proteins and the biopol- itics of edibility. Transactions of the Institute of British Geographers.

Sexton, A. E., Garnett, T., & Lorimer, J. (2019). Framing the future of food: The contested promises of alternative proteins. Environment and Planning E: Nature and Space, 2514848619827009.

Sharpe, M. R. L. (1908). Six hundred recipes for meatless dishes. Cambridge: University Press.

Shepherd, H. (2011). The Cultural Context of Cognition: What the Implicit Association Test Tells Us About How Culture Works. Sociological Forum, 26 (1), 121–143.

Shprintzen, A. D. (2012). Looks Like Meat, Smells Like Meat, Tastes Like Meat: Battle Creek, Protose and the Making of Modern American Vegetarianism. Food, Culture & Society, 15(1), 113–128.

Shurtleff, W., & Aoyagi, A. (1994). Wheat gluten and seitan: Bibliography and sourcebook, A.D. 535 to 1993. Lafayette, CA: Soyfoods Center.

Shurtleff, W., & Aoyagi, A. (2014). History of meat alternatives (965 CE to 2014): Extensively annotated bibliography and sourcebook.

Simões Braga, A. C., & Silva e Meirelles, D. (2014). Collective Identity Construction: A The- oretical Proposal For The Analysis Of New Organizational Forms. Business and Manage- ment Review, 4(3), 89–102.

252 Academic References

Sinnenberg, L., Buttenheim, A. M., Padrez, K., Mancheno, C., Ungar, L., & Merchant, R. M. (2016). Twitter as a Tool for Health Research: A Systematic Review. American Journal of Public Health, 107(1), e1–e8.

Sixsmith, R., & Furnham, A. (2010). A content analysis of British food advertisements aimed at children and adults. Health Promotion International, 25(1), 24–32.

Sjøvaag, H., & Stavelin, E. (2012). Web media and the quantitative content analysis: Method- ological challenges in measuring online news content. Convergence, 18(2), 215–229.

Small, M. L., Harding, D. J., & Lamont, M. (2010). Reconsidering Culture and Poverty. The Annals of the American Academy of Political and Social Science, 629, 6–27.

Smetana, S., Mathys, A., Knoch, A., & Heinz, V. (2015). Meat alternatives: Life cycle assessment of most known meat substitutes. The International Journal of Life Cycle Assessment, 20(9), 1254–1267.

Smil, V. (2002). Eating Meat: Evolution, Patterns, and Consequences. Population and Devel- opment Review, 28(4), 599–639.

Snow, D. A., & Benford, R. D. (1988). Ideology, frame resonance, and participant mobiliza- tion. International Social Movement Research, 1(1), 197–217.

Snow, D. A., & Benford, R. D. (1992). Master frames and cycles of protest. In A. D. Morris & C. M. Mueller (Eds.), Frontiers in social movement theory (pp. 133–155). New Haven, CT, US: Yale University Press.

So, J., Prestin, A., Lee, L., Wang, Y., Yen, J., & Chou, W.-Y. S. (2016). What Do People Like to ‘Share’ About Obesity? A Content Analysis of Frequent Retweets About Obesity on Twitter. Health Communication, 31(2), 193–206.

Stahl. (2017). Making a Mockery of Meat: Troubling Texture and the Failings of the ‘Flesh’. Journal of Asia-Pacific Pop Culture, 2(2), 184.

Steinfeld, H., Gerber, P., Wassenaar, T. D., Castel, V., Rosales M., M., & Haan, C. de. (2006). Livestock’s long shadow: Environmental issues and options. Rome: Food and Agriculture Organization of the United Nations.

Stephens, N. (2010). In Vitro Meat: Zombies on the Menu? SCRIPTed, 7(2), 394–401.

Stephens, N. (2013). Growing Meat in Laboratories: The Promise, Ontology, and Ethical Boundary-Work of Using Muscle Cells to Make Food. Configurations, 21(2), 159–181.

Swedberg, R., & Granovetter, M. S. (1992). Introduction. In M. S. Granovetter & R. Swed- berg (Eds.), The Sociology of economic life (pp. 1–26). Boulder: Westview Press.

Swidler, A. (1986). Culture in Action: Symbols and Strategies. American Sociological Review, 51(2), 273.

253 References

Tainter, J. A. (2011). Energy, complexity, and sustainability: A historical perspective. Environ- mental Innovation and Societal Transitions, 1(1), 89–95.

Thomas, D. R. (2006). A General Inductive Approach for Analyzing Qualitative Evaluation Data. American Journal of Evaluation, 27(2), 237–246.

Thorson, K., Ekdale, B., Borah, P., Namkoong, K., & Shah, C. (2010). Youtube and Proposi- tion 8. Information, Communication & Society, 13(3), 325–349.

Tobler, C., Visschers, V. H. M., & Siegrist, M. (2011). Eating green. Consumers’ willingness to adopt ecological food consumption behaviors. Appetite, 57(3), 674–682.

Trinci, A. P. (1992). Myco-protein: A twenty-year overnight success story. Mycological Re- search, 96 (1), 1–13.

Twine, R. (2018). Materially Constituting a Sustainable Food Transition: The Case of Vegan Eating Practice. Sociology, 52(1), 166–181.

Ugalde, U. O., & Castrillo, J. I. (2002). Single cell proteins from fungi and yeasts. In G. G. Khachatourians & D. K. Arora (Eds.), Applied Mycology and Biotechnology (Vol. 2, pp. 123–149). Elsevier.

Vainio, J., & Holmberg, K. (2017). Highly tweeted science articles: Who tweets them? An analysis of Twitter user profile descriptions. Scientometrics, 112(1), 345–366.

Verbeke, W. (2015). Profiling consumers who are ready to adopt insects as a meat substitute in a Western society. Food Quality and Preference, 39, 147–155.

Verbeke, W., Marcu, A., Rutsaert, P., Gaspar, R., Seibt, B., Fletcher, D., & Barnett, J. (2015). ‘Would you eat cultured meat?’: Consumers’ reactions and attitude formation in Belgium, Portugal and the United Kingdom. Meat Science, 102, 49–58.

Verbeke, W., & Viaene, J. (2000). Ethical Challenges for Livestock Production: Meeting Consumer Concerns about Meat Safety and Animal Welfare. Journal of Agricultural and Environmental Ethics, 12(2), 141–151.

Vergani, M., & Zuev, D. (2011). Analysis of YouTube Videos Used by Activists in the Uyghur Nationalist Movement: Combining quantitative and qualitative methods. Journal of Contemporary China, 20(69), 205–229.

Vergne, J.-P. (2012). Stigmatized Categories and Public Disapproval of Organizations: A Mixed-Methods Study of the Global Arms Industry, 1996–2007. Academy of Manage- ment Journal, 55(5), 1027–1052.

Vergne, J.-P., & Swain, G. (2017). Categorical Anarchy in the UK? The British Media’s Classi- fication of Bitcoin and the Limits of Categorization. In R. Durand, N. Granqvist, & A. Tyllström (Eds.), Research in the Sociology of Organizations (Vol. 51, pp. 185–222). Emerald Publishing Limited.

254 Academic References

Vergne, J.-P., & Wry, T. (2014). Categorizing Categorization Research: Review, Integration, and Future Directions. Journal of Management Studies, 51(1), 56–94.

Vliegenthart, R., & van Zoonen, L. (2011). Power to the frame: Bringing sociology back to frame analysis. European Journal of Communication, 26 (2), 101–115.

Vogt, W. P., Gardner, D. C., & Haeffele, L. M. (2012). When to use what research design. New York: Guilford Press.

Warren, G. C. (Ed.). (1958). The food we eat : A survey of meals. . .. Cassell.

Weber, K. (2005). A toolkit for analyzing corporate cultural toolkits. Poetics, 33(3-4), 227–252.

Weber, K., Heinze, K. L., & DeSoucey, M. (2008). Forage for Thought: Mobilizing Codes in the Movement for Grass-fed Meat and Dairy Products. Administrative Science Quarterly, 53(3), 529–567.

Weber, R. P. (1990). Basic content analysis. Newbury Park, CA, US: Sage Publications.

Welin, S. (2013). Introducing the new meat. Problems and prospects. Etikk I Praksis - Nordic Journal of Applied Ethics, 7(1), 24–37.

Wellesley, L., Froggatt, A., & Happer, C. (2015). Changing climate, changing diets: Pathways to lower meat consumption.

Wild, F., Czerny, M., Janssen, A. M., Kole, A. P., Zunabovic, M., & Domig, K. J. (2014). The evolution of a plant-based alternative to meat. Agro FOOD Industry Hi Tech, 25, 1.

Wry, T., Lounsbury, M., & Jennings, P. D. (2013). Hybrid Vigor: Securing Venture Capital by Spanning Categories in Nanotechnology. Academy of Management Journal, 57(5), 1309–1333.

Yates, L., & Warde, A. (2015). The evolving content of meals in Great Britain. Results of a survey in 2012 in comparison with the 1950s. Appetite, 84, 299–308.

Youmans, W.L., & York, J. C. (2012). Social Media and the Activist Toolkit: User Agreements, Corporate Interests, and the Information Infrastructure of Modern Social Movements. Journal of Communication, 62(2), 315–329.

Zald, D. H. (2009). Orbitofrontal Cortex Contributions to Food Selection and Decision Making. Annals of Behavioral Medicine, 38(suppl_1), s18–s24.

Zhao, E. Y., Ishihara, M., & Lounsbury, M. (2013). Overcoming the Illegitimacy Discount: Cultural Entrepreneurship in the US Feature Film Industry. Organization Studies, 34(12), 1747–1776.

Zittrain, J. (2008). The future of the Internet and how to stop it. New Haven, CT, US: Yale University Press.

Zorpette, G. (2013). Muscling Out Meat. IEEE Spectrum, 50(6), 64–70.

255 References

Zuckerman, E. W.(1999). The Categorical Imperative: Securities Analysts and the Illegitimacy Discount. American Journal of Sociology, 104(5), 1398–1438.

Zuckerman, E. W. (2017). The Categorical Imperative Revisited: Implications of Categoriza- tion as a Theoretical Tool. In R. Durand, N. Granqvist, & A. Tyllström (Eds.), Research in the Sociology of Organizations (Vol. 51, pp. 31–68). Emerald Publishing Limited.

Archival References

Andriukaitis, V. (2016, December 19). Answer to a written question - Food products with misleading names - E-008161/2016. retrieved July 22, 2017, from http://www.europarl.e uropa.eu/sides/getAllAnswers.do?reference=E-2016-008161&language=EN; European Parliament.

Askew, K. (2019, April 16). Nestlé’s withdrawal of Garden Gourmet shows intense competition in UK plant-based. foodnavigator.com. retrieved April 23, 2019, from https: //www.foodnavigator.com/News/Business/Nestle-s-withdrawal-of-Garden-Gourmet- shows-intense-competition-in-UK-plant-based.

Avery, J. (2014). Fighting food fraud (No. 130679REV1) (p. 7). European Parliamentary Research Service; retrieved July 22, 2017, from http://www.europarl.europa.eu/RegData /bibliotheque/briefing/2014/130679/LDM_BRI%282014%29130679_REV1_EN.pdf.

Bamford, V. (2012, March 17). Linda McCartney to offer Britain’s veggies the prawn-free prawn. The Grocer, p. 30. retrieved from the Factiva database.

Bamford, V. (2015, September 26). Hain Daniels gives Linda McCartney’s ’sharper’ new look. The Grocer, p. 44. retrieved from the Factiva database.

Boffey, D. (2019, April 4). ‘Veggie discs’ to replace veggie burgers in EU crackdown on food labels. The Guardian. retrieved April 23, 2019, from https://www.theguardian.com/food /2019/apr/04/eu-to-ban-non-meat-product-labels-veggie-burgers-and-vegan-steaks.

Bramble, B. (2017, December 14). Lab-grown meat could let humanity ignore a serious moral failing. The Conversation. retrieved March 22, 2019, from http://theconversation.com/lab- grown-meat-could-let-humanity-ignore-a-serious-moral-failing-88909.

Brockes, E. (2000, December 1). Cheat meat. The Guardian, pp. A6–A7. London, UK: retrieved from the ProQuest database.

Brooks, B. (2015, May 16). Meat Free Category Report 2015. The Grocer. retrieved July 4, 2017, from https://global.factiva.com/du/article.aspx/?accessionno=GROCR00020150515eb5 g00014.

Brooks, B., & Brown, R. (2014, May 17). Meat-Free Category Report 2014. The Grocer.

256 Archival References

retrieved July 4, 2017, from https://global.factiva.com/du/article.aspx/?accessionno=G ROCR00020140519ea5h0001h.

Buckingham, L. (1999, September 10). United Biscuits hit by GM fears. The Guardian, p. 27. London, UK: retrieved from the ProQuest database.

Bureau van Dijk. (2017, July 30). Fame. Fame. Business Database, retrieved July 30, 2017, from https://fame.bvdinfo.com/.

Carbon Trust. (2015). The Case for Protein Diversity (NGO Report) (p. 21). retrieved March 21, 2019, from https://www.carbontrust.com/media/671648/the-case-for-protein- diversity.pdf.

CB Insights. (2017, November 9). Our Meatless Future: How The $90B Global Meat Market Gets Disrupted. CB Insights Research. retrieved January 7, 2019, from https://www.cbin sights.com/research/future-of-meat-industrial-farming/.

Collins, R. (1975, February 6). Describe a spun protein chop. The Guardian. London, UK: retrieved from the ProQuest database.

Companies House. (2017). ADM NEWPORT PAGNELL LIMITED - Filing history. Gov- ernment, retrieved from https://beta.companieshouse.gov.uk/company/01068279/filing- history?page=6.

Companies House. (n.d.). LAVANT STREET PROPERTIES LIMITED - Filing history. Companies House. Government, retrieved from https://beta.companieshouse.gov.uk/c ompany/00949421/filing-history?page=4.

DEFRA. (2017). Family Food 2015. London: Department for Environment, Food and Rural Affairs. retrieved February 24, 2018, from https://www.gov.uk/government/uploads/sy stem/uploads/attachment_data/file/597667/Family_Food_2015-09mar17.pdf.

Eating Better. (2017). The future of eating is flexitarian: Companies leading theway (NGO Report). retrieved March 24, 2018, from https://www.eating-better.org/uploads/Docum ents/2017/Eating%20Better_The%20future%20of%20eating%20is%20flexitarian.pdf.

Eating Better. (2018). Principles for eating meat and dairy more sustainably: The ‘less and better’ approach (NGO Report). retrieved January 17, 2019, from https://www.eating- better.org/uploads/Documents/2018/better_meat_report_FINAL.pdf.

Euromonitor International. (2017a). Free From in the United Kingdom (Category Briefing). retrieved July 26, 2017, from http://www.portal.euromonitor.com/portal/analysis/tab.

Euromonitor International. (2017b). Plant-based Protein: Assessing Demand for Sustainable Alternatives (Briefing).

Euromonitor International. (2018). Euromonitor Statistics for the Categories: Free From Meat, Meat, Processed Meat and Seafood (Database). Euromonitor International.

257 References

European Commission. (n.d.). GMO legislation - Food Safety - European Commission. Food Safety. retrieved from https://ec.europa.eu/food/plant/gmo/legislation_en.

European Court of Justice. (2017, June 14). Verband Sozialer Wettbewerb e.V. v. To- fuTown.com GmbH. retrieved July 22, 2017, from https://curia.europa.eu/jcms/upl oad/docs/application/pdf/2017-06/cp170063en.pdf.

European VegetarianUnion. (2016, May). Definitions of ‘vegan’ and ‘vegetarian’ in accordance with the EU Food Information Regulation. Position Paper, Berlin, Germany: retrieved July 22, 2017, from http://www.euroveg.eu/wp-content/uploads/2015/06/EVU-PP- Definition-FIC-May2016.pdf.

European Vegetarian Union. (n.d.). Sales Denominations of Vegetarian Meat Alternatives. Position Paper, Berlin, Germany: retrieved July 22, 2017, from http://www.euroveg.eu/w p-content/uploads/2017/03/EVU-PP-Denomination-Feb2017.pdf. foodmanufacture.co.uk. (2018, July 10). Vegetarianism: Threat or Opportunity. foodmanufacture.co.uk. Food Industry News, retrieved July 24, 2018, from https://www.foodmanu facture.co.uk/Products/Vegetarianism-Threat-or-Opportunity.

Food Standards Agency. (2017). Salt Targets. Food Standards Agency. retrieved July 22, 2017, from https://www.food.gov.uk/northern-ireland/nutritionni/salt-ni/salt-targets.

Food Standards Agency. (2015, July 29). Regulation of novel food. Food Standards Agency. Government, retrieved from https://www.food.gov.uk/science/novel/regulation.

Food Standards Agency. (2016, November). Plant products (not fruit/vegetables) & vegetarian products. Trade Information Sheet, retrieved July 22, 2017, from https://www.food.gov.u k/sites/default/files/trade-info11.pdf.

Food Standards Agency. (n.d.). EU FIC regulations on food labelling. Food Standards Agency. retrieved July 22, 2017, from https://www.food.gov.uk/science/allergy-intolerance/label/ labelling-changes.

Frost & Sullivan. (2001). U.S. Soy-Based Meat Alternatives Market (Industry Research Report No. 7881). Frost & Sullivan.

Frost & Sullivan. (2010). Trends and Opportunities in the European Protein Ingredients Market (Industry Research Report No. M5FE). Frost & Sullivan; retrieved February 11, 2016, from http://cds.frost.com/p/52719/#!/nts/c?id=M5FE-01-00-00-00&fq=meat%2 0substitutes.

Frost & Sullivan. (2016). Analysis of the Western European Protein Ingredients Market (Market Insights No. MC4A).

Frost & Sullivan. (2017). Innovations in Plant-based Alternatives for Animal Protein (Tech- Vision Opportunity Engines No. D768/45).

258 Archival References

Gohd, C. (2017, December 12). In the future, the meat you eat won’t come from living organisms. Futurism. retrieved January 7, 2019, from https://futurism.com/future-meat-eat- wont-come-from-living-organisms.

Graham, R. (2019, April 13). Vegan or bust? Why Nestlé’s veggie range didn’t take off. The Grocer, p. 17. retrieved from the Factiva database.

Hartman, L. R. (2018, February 27). A Bright Future is Germinating for Alternative Proteins. Food Processing. retrieved January 7, 2019, from https://www.foodprocessing.com/article s/2018/bright-future-for-alternative-proteins/.

Johnson, W. G. (2018, October 23). Lab-Grown Seafood and Lab-Grown Meat Might End Up With Different Regulations. Thanks, Bureaucracy! Slate Magazine. retrieved April 17, 2019, from https://slate.com/technology/2018/10/lab-grown-meat-seafood-usda-fda- labeling.html.

Just Food. (2000, December 11). Vegetarian and Meat-Free Foods. retrieved April 10, 2018, from https://www.just-food.com/analysis/vegetarian-and-meat-free-foods_id94010.asp x.

Kantar Worldpanel. (2016). The Meat Reducer (Custom Market Research).

Key Note. (2015). Meat & Meat Products – Market Update 2015 (Market Research). retrieved February 11, 2016, from https://www.keynote.co.uk/market-update/food/meat-meat- products-0.

Leatherhead. (1990). Vegetarianfoods in the UK. (Market Research). Leatherhead Food RA, Information Group Services.

Leatherhead. (1993). The UK vegetarian food report (Market Research). Leatherhead: Leath- erhead Food Research Association.

Leatherhead. (1995). The UK vegetarian foods report. Leatherhead: Leatherhead Food Re- search Association.

Leyland, A. (2008, June 14). McCartney family to support revamp of meat-free range. The Grocer, pp. 56–57. retrieved from the Factiva database.

Linda McCartney’s Foods. (2016). Linda McCartney’s – 25 Years (Promotional Video). retrieved July 21, 2017, from http://www.lindamccartneyfoods.co.uk/our-story/our- journey/.

Maurin, J. (2015, April 6). VegetarischeWurst: Veggie-Boombei Fleischfabriken. Die Tageszei- tung. retrieved from the taz.de database.

Michail, N. (2018, June 1). Plant-based meat substitutes are ’useful transition foods’ for a healthier diet. foodnavigator.com. Food Industry News, retrieved February 27, 2019, from

259 References

https://www.foodnavigator.com/Article/2018/06/01/Plant-based-meat-substitutes-are- useful-transition-foods-for-a-healthier-diet.

Mintel. (2000). Vegetarian- UK - June 2000 (Market Research). London, UK: Mintel Group Ltd.

Mintel. (2001). Meat-free Foods - UK - December 2001 (Market Research). London, UK: Mintel Group Ltd.

Mintel. (2004). Meat-free Foods - UK - December 2004 (Market Research). London, UK: Mintel Group Ltd.

Mintel. (2006a). Attitudes Towards Vegetarianism– UK – September 2006 (Market Research). London, UK: Mintel Group Ltd.

Mintel. (2006b). Meat-free Foods - UK - December 2006 (Market Research). London, UK: Mintel Group Ltd.

Mintel. (2008). Meat-free Foods - UK - December 2008 (Market Research). London, UK: Mintel Group Ltd.

Mintel. (2010). Meat Free Foods - UK - December 2010 (Market Research). London, UK: Mintel Group Ltd.

Mintel. (2011). Meat-free and Free-from Foods - UK - September 2011 (Market Research). London, UK: Mintel Group Ltd.

Mintel. (2013). Meat-free and Free-from Foods - UK - September 2013 (Market Research). London, UK: Mintel Group Ltd.

Mintel. (2017). Meat-free Foods - UK - May 2017 (Market Research). London, UK: Mintel Group Ltd.

Mintel. (2018). Meat-free Foods - UK - September 2018 (Market Research). London, UK: Mintel Group Ltd.

Phillips, B. (2012, January 14). Putting the veg in vegetarian. The Grocer, pp. 59–61. retrieved from the Factiva database.

Phillips, B. (2013, July 20). Hain Daniels soya products remain withdrawn from market. The Grocer, p. 6. retrieved from the Factiva database.

Piper, K. (2018, November 15). We could end factory farming this century, thanks to meat alternatives. Vox. retrieved January 7, 2019, from https://www.vox.com/future-perfect/ 2018/11/15/18088776/end-animal-farming-vegetarian-vegan-meat-alternatives-jacy-reese.

Raphael, A. (1969, October 27). Meatless meat–synthetic way to the US heart. The Guardian, p. 3. London, UK: retrieved from the ProQuest database.

260 Archival References

Riley, L. (2010, November 20). Linda McCartney sets out to prove meat-free food is a winner. The Grocer, pp. 24–25. retrieved from the Factiva database.

Rödl, M. B. (2018b, April 30). Why France banned meat names for vegetarian alternatives. The Conversation. Academic Blogging Platform, retrieved May 25, 2018, from http://thec onversation.com/why-france-banned-meat-names-for-vegetarian-alternatives-95549.

Rödl, M. B. (2019b, March 5). Why the meat industry could win big from the switch to veggie lifestyles. The Conversation. Academic Blogging Platform, retrieved March 5, 2019, from http://theconversation.com/why-the-meat-industry-could-win-big-from-the-switch-to- veggie-lifestyles-112714.

Rügenwalder Mühle. (n.d.). Frequently asked question. retrieved February 3, 2019, from https://www.ruegenwalder.de/en/faq.

Smithers, R. (2018, June 10). Sainsbury’s to sell vegan ’fake meats’ next to real thing in trial. The Guardian. retrieved June 24, 2018, from http://www.theguardian.com/lifeandstyle /2018/jun/11/sainsburys-to-sell-vegan-fake-meats-next-to-real-thing-in-trial.

Southey, F. (2019, April 2). Nestlé launches ‘cook from raw’ plant-based burger in Europe. foodnavigator.com. Food Industry News, retrieved April 5, 2019, from https://www.food navigator.com/Article/2019/04/02/Nestle-launches-cook-from-raw-plant-based-burger- in-Europe.

Stähle, S. (2018, December 24). Vegane und vegetarische Lebensmittel - Vom Ende der Salami- Taktik. retrieved January 18, 2019, from https://www.deutschlandfunk.de/vegane-und- vegetarische-lebensmittel-vom-ende-der-salami.697.de.html?dram:article_id=436821.

Taifun Tofu. (2014, July 10). Stiftungsgründung: Große Freude bei Taifun. Taifun Tofu. Company, retrieved July 30, 2017, from https://taifun-tofu.de/de/stiftungsgruendung.

The Economist. (2015, March 7). Silicon Valley gets a taste for food. The Economist. retrieved April 28, 2016, from http://www.economist.com/news/technology-quarterly/21645497- tech-startups-are-moving-food-business-make-sustainable-versions-meat.

The Grocer. (1988a, June 11). Last year, Sainsbury introduced a range of savoury pies, made with Quorn mycro-protein derived from a plant, part of the mushroom family. The Grocer, p. 44. retrieved from the Factiva database.

The Grocer. (1988b, November 19). New ready meals from Quorn. The Grocer, p. 37. retrieved from the Factiva database.

The Grocer. (1989a, August 19). Focus on Frozen Foods - Ross Young’s. The Grocer, p. 68. retrieved from the Factiva database.

The Grocer. (1989b, July 8). Quorn basis for new Safeway meals. The Grocer, p. 39. retrieved from the Factiva database.

261 References

The Grocer. (1989c, November 11). Focus on healthy foods - Haldane Foods. The Grocer, p. 104. retrieved from the Factiva database.

The Grocer. (1989d, October 7). Retail ingredients market is next step for Quorn. The Grocer, p. 51. retrieved from the Factiva database.

The Grocer. (1990a, August 25). Haldane Foods backs Vegetable Feasts range with poster advertising. The Grocer, p. 36. retrieved from the Factiva database.

The Grocer. (1990b, February 10). Haldane Acquires Kwality Foods of Gwent. The Grocer, p. 20. retrieved from the Factiva database.

The Grocer. (1990c, May 12). Marlow Foods to launch TV advertising campaign for Quorn. The Grocer, p. 41. retrieved from the Factiva database.

The Grocer. (1990d, October 27). Grocer Focus on Healthy Foods - Haldane Foods Group. The Grocer, p. 128. retrieved from the Factiva database.

The Grocer. (1991a, June 8). The Grocer Focus on Healthy Foods - Dalepak. The Grocer, p. 82. retrieved from the Factiva database.

The Grocer. (1991b, May 11). Ross Young’s Launches Linda Mccartney Range of Frozen Meatless Meals. The Grocer, p. 43. retrieved from the Factiva database.

The Grocer. (1991c, November 16). Promising signs for Quorn sales. The Grocer, p. 37. retrieved from the Factiva database.

The Grocer. (1992, August 15). Food stores take part in national vegetarian week. The Grocer, p. 4. retrieved from the Factiva database.

The Grocer. (1993a, April 17). Grocer Previews Ife 93 - Mince Without the Meat - Haldane Foods. The Grocer, p. 107. retrieved from the Factiva database.

The Grocer. (1993b, April 3). Mince Alternative to Beef up Veggie Sales - Haldane Foods. The Grocer, p. 33. retrieved from the Factiva database.

The Grocer. (1993c, August 21). Mccartney Revamp Targets Bird’s Eye’s Pole Position. The Grocer, p. 35. retrieved from the Factiva database.

The Grocer. (1993d, February 13). Focus on Frozen Foods - Pizza Gets Slice of the Action - Ross Young’s. The Grocer, p. 72. retrieved from the Factiva database.

The Grocer. (1993e, June 12). Focus on Healthy Food - Tv Backs Mccartney Ready Meals - Ross Young’s. The Grocer, p. 82. retrieved from the Factiva database.

The Grocer. (1994a, February 26). Quorn Gets #5m Cash Injection to Make Its Mark Nation- ally. The Grocer, p. 38. retrieved from the Factiva database.

The Grocer. (1994b, July 30). Focus on Vegetarian Foods - Lentil and Nut Roast Image Is History. The Grocer, p. 41. retrieved from the Factiva database.

262 Archival References

The Grocer. (1994c, July 9). Quorn Heads for Mass Market with Burgers Development. The Grocer, p. 36. retrieved from the Factiva database.

The Grocer. (1994d, June 11). Linda Mccartney Sales Heading for #25m. The Grocer, p. 30. retrieved from the Factiva database.

The Grocer. (1995a, April 1). Mccartney Releases Chilled Veggie Range. The Grocer, p. 31. retrieved from the Factiva database.

The Grocer. (1995b, December 23). Premium Cuts for Quorn. The Grocer, p. 22. retrieved from the Factiva database.

The Grocer. (1995c, July 29). Marketing - Marlow Foods. The Grocer, p. 41. retrieved from the Factiva database.

The Grocer. (1995d, March 11). Footballer Promotes Quorn Burgers. The Grocer, p. 35. retrieved from the Factiva database.

The Grocer. (1996, October 5). Mainstream Battle for Meat-Free Brands. The Grocer, p. 32. retrieved from the Factiva database.

The Grocer. (1997a, August 23). McCartney moves to mince ahead of ice cream launch. The Grocer, p. 55. retrieved from the Factiva database.

The Grocer. (1997b, January 18). Ryan Giggs Gets the Chop in Favour of Chef. The Grocer, p. 28. retrieved from the Factiva database.

The Grocer. (1997c, January 4). Altered Images. The Grocer, p. 10. retrieved from the Factiva database.

The Grocer. (1998a, April 25). Linda face stays for now. The Grocer, p. 6. retrieved from the Factiva database.

The Grocer. (1998b, January 10). Tivall gets more bite. The Grocer, p. 12. retrieved from the Factiva database.

The Grocer. (1998c, January 31). Slicing up the Market - Yves Potvin. The Grocer, p. 6. retrieved from the Factiva database.

The Grocer. (1998d, July 11). News – McVitie’s and Haldane take the GMO-free route. The Grocer, p. 8. retrieved from the Factiva database.

The Grocer. (1998e, March 21). Demi-monde - new TV campaign. The Grocer, p. 46. retrieved from the Factiva database.

The Grocer. (1998f, March 7). Meat reducers targeted as Linda revamps and expands. The Grocer, p. 61. retrieved from the Factiva database.

The Grocer. (1999a, August 14). Big buyers queue up for UB division. The Grocer, p. 10. retrieved from the Factiva database.

263 References

The Grocer. (1999b, July 10). Celebrity selling products. The Grocer, p. 52. retrieved from the Factiva database.

The Grocer. (1999c, June 12). Linda GM Soya Belongs to Yesterday. The Grocer, p. 5. retrieved from the Factiva database.

The Grocer. (1999d, March 13). The great pretenders. The Grocer, p. 47. retrieved from the Factiva database.

The Grocer. (1999e, October 30). Heinz buys UB brands. The Grocer, p. 8. retrieved from the Factiva database.

The Grocer. (2000a, February 19). Heinz targets functional and organics. The Grocer, p. 11. retrieved from the Factiva database.

The Grocer. (2000b, February 5). A global Linda? The Grocer, p. 4. retrieved from the Factiva database.

The Grocer. (2000c, November 4). The Grocer focus on vegetarian food. The Grocer. retrieved from the Factiva database.

The Grocer. (2000d, October 7). Linda grows with the Garden. The Grocer, p. 66. retrieved from the Factiva database.

The Grocer. (2001, June 2). Steak-loving Americans will soon have a new word to add to t. The Grocer, p. 38. retrieved from the Factiva database.

The Grocer. (2002a, April 20). Health message wheeled out as Quorn turns up the activity. The Grocer, p. 56. retrieved from the Factiva database.

The Grocer. (2002b, August 17). US group smears Quorn safety record. The Grocer, p. 6. retrieved from the Factiva database.

The Grocer. (2002c, July 13). Patak’s sales growth in US celebrated as FFB hands out its. The Grocer, p. 12. retrieved from the Factiva database.

The Grocer. (2002d, November 13). FDF lines up vegetarian group. The Grocer. retrieved January 21, 2019, from https://www.thegrocer.co.uk/home/topics/fdf-lines-up-vegetari an-group/81630.article.

The Grocer. (2002e, September 7). ASA sides with anti-Quorn set. The Grocer, p. 4. retrieved from the Factiva database.

The Grocer. (2003a, April 12). Basking in mainstream approval. The Grocer, p. 47. retrieved from the Factiva database.

The Grocer. (2003b, July 5). Kellogg’s Longworth to head Quorn marketing. The Grocer, p. 15. retrieved from the Factiva database.

264 Archival References

The Grocer. (2004a, August 14). Meat-free Quorn extends ambition. The Grocer, p. 49. retrieved from the Factiva database.

The Grocer. (2004b, December 18). The exporting of excellence. The Grocer, p. 52. retrieved from the Factiva database.

The Grocer. (2004c, November 20). Quorn is off to a healthy new year. The Grocer, p. 67. retrieved from the Factiva database.

The Grocer. (2005a, June 11). Quorn plans NPD after buyout. The Grocer, p. 66. retrieved from the Factiva database.

The Grocer. (2005b, May 14). Marlow may act against US critic. The Grocer, p. 67. retrieved from the Factiva database.

The Grocer. (2005c, November 5). Cauldron joins Quorn in Premier Foods’ veggie stable. The Grocer, p. 6. retrieved from the Factiva database.

The Grocer. (2006a, February 4). Meat-free not just for vegetarians. The Grocer, p. 47. retrieved from the Factiva database.

The Grocer. (2006b, June 17). In brief. The Grocer, pp. 16–17. retrieved from the Factiva database.

The Grocer. (2007, January 20). Hunks of meat? No thanks. The Grocer. retrieved from the Factiva database.

The Grocer. (2008, February 2). Research Notes. The Grocer, pp. 39–44. retrieved from the Factiva database.

The Grocer. (2009a, August 8). News In Brief. The Grocer, pp. 26–27. retrieved from the Factiva database.

The Grocer. (2009b, February 21). Linda McCartney gets new look and recipes. The Grocer, pp. 30–31. retrieved from the Factiva database.

The Grocer. (2010, January 30). Meat-eaters? Try Veggie. The Grocer, pp. 61–64. retrieved from the Factiva database.

The Grocer. (2011a, January 29). Veggie brands?Feel the pinch. The Grocer, pp. 51–53. retrieved from the Factiva database.

The Grocer. (2011b, July 16). Meat-free pie winner to hit shelves. The Grocer, pp. 28–29. retrieved from the Factiva database.

The Grocer. (2016, May 14). Top meat-free advertisers 2016 ranked by Ebiquity. The Grocer. retrieved from the Factiva database.

The Hain Celestial Group. (2001). 2001 Annual Report. Annual Report, retrieved August

265 References

22, 2017, from http://media.corporate-ir.net/media_files/IROL/87/87078/reports/200 1ar.pdf.

The Hain Celestial Group. (2006). 2006 Annual Report. Annual Report, retrieved August 22, 2017, from http://library.corporate-ir.net/library/87/870/87078/items/219136/hain2 006AR2.pdf.

The Hain Celestial Group. (2007). 2007 Annual Report. Annual Report, retrieved August 22, 2017, from http://library.corporate-ir.net/library/87/870/87078/items/283089/hai n_2007ar.pdf#page=52&zoom=auto,-275,88.

The Hain Celestial Group. (2012, August 22). Hain Celestial Announces Strategic Acquisition In The United Kingdom. Press Release, MELVILLE, N.Y.: retrieved August 22, 2017, from http://ir.hain.com/phoenix.zhtml?c=87078&p=irol-newsArticle&ID=1727796.

The Telegraph. (2006, December 4). Peter Roberts. The Telegraph. retrieved from http: //www.telegraph.co.uk/news/obituaries/1535910/Peter-Roberts.html.

The Vegan Society. (2016, May 17). Find out how many vegans are in Great Britain. The Vegan Society. retrieved March 8, 2018, from https://www.vegansociety.com/whats- new/news/find-out-how-many-vegans-are-great-britain.

Townsend, L., & Wallace, C. (n.d.). Social Media Research: A Guide to Ethics (p. 16). Aberdeen, UK: The University of Aberdeen; retrieved November 13, 2018, from http: //www.dotrural.ac.uk/socialmediaresearchethics.pdf.

United Biscuits. (2017). History. United Biscuits. Company, retrieved from http://www.un itedbiscuits.com/our-business/history/.

U.S. Cattlemen’s Association. (2018, February 9). Petition for the Imposition of Beef and Meat labeling Requirements: To exclude Products not derived directly from Animals raised and slaughtered from the Definition of ”Beef” and ”Meat”. Petition FSIS-2018-0016, retrieved from https://www.fsis.usda.gov/wps/portal/fsis/topics/regulations/petitions.

Vegan Life Mag. (n.d.). Audience. VeganLife Magazine Media Kit. retrieved November 23, 2018, from https://mediakit.veganlifemag.com/audience/.

Vegetarian Society. (2001, January 9). Information Sheet - Statistics - Summary of Realeat Polls. The VegetarianSociety UK. retrieved February 27, 2019, from https://web.archive. org/web/20010109050600/http://www.vegsoc.org:80/info/realeat.html.

Vegetarian Society. (n.d.-a). Advertising in The Vegetarian. Vegetarian Society. retrieved November 23, 2018, from https://www.vegsoc.org/advertising.

Vegetarian Society. (n.d.-b). Vegetarian Society - Approved products. Vegetarian Society. retrieved August 10, 2017, from https://www.vegsoc.org/sslpage.aspx?pid=650.

Waitrose & Partners. (2018). Food and Drink Report 2018–2019 (Market Research) (p. 12).

266 Archival References

retrieved March 5, 2019, from https://www.waitrose.com/content/dam/waitrose/Inspir ation/Waitrose%20&%20Partners%20Food%20and%20Drink%20Report%202018.pdf.

Watson, E. (2011a, December 7). Quorn: ‘CSPI has had a long running grudge against us’. Food Navigator USA. retrieved from http://www.foodnavigator-usa.com/Regulation /Quorn-CSPI-has-had-a-long-running-grudge-against-us.

Watson, E. (2011b, January 24). Premier sells Quorn and Cauldron in £205m deal. Food- Manufacture.co.uk. retrieved from http://www.foodmanufacture.co.uk/Business- News/Premier-sells-Quorn-and-Cauldron-in-205m-deal.

Zuke, E. (2012, August 25). Hartleys, Gales and Sunpat among the jewels snapped up by Hain Celestial’s £200m acquisition. The Grocer, p. 5. retrieved from the Factiva database.

267

Appendix A

Further Details of the Contextual Data

A.1 Detailed Overview of the Sampling Procedure

In order to achieve a good overview of the market for meat alternatives and its history, I determined a number of data sources on which this contextualisation should build up; this selection was inspired by previous literature, recommendations by colleagues, and library and online searches. The sampling was purposive as it should contribute to a narration of the history of meat alternatives, as well as provide an overview of the current market. The topics for this overview were inspired by the field of economic sociology (cf. Swedberg & Granovetter, 1992) in order to provide an account of the socio-economic arrangements within the market, such as supply chains, actors, finance, legislation, competition, marketing, and the consumership. The data collection followed an approach of ‘data saturation’, whereby I sampled data until no new information could be obtained from it (Saunders et al., 2018). Apart from academic research three other types of data were chosen ex ante: Firstly, an extensively annotated bibliography on meat alternatives (Shurtleff & Aoy- agi, 2014), which contains summaries of interviews related to meat alternative, books, scholarly work, and product introductions1. Secondly, historical and more recent market research reports on meat alternatives and vegetarian foods were sampled; not only their content but also their titles and existence was used to infer on the state of the market. Lastly, industry news were chosen as an ongoing reports on developments in the market. Both, the choice of market research reports and the industry newspaper were largely determined by availability and accessibility via the university’s digital subscriptions. Accordingly, the main source of market research is Mintel Academic, where nine relevant market research reports about UK meat alternatives or related topics starting in 2000 could be accessed2. The source for industry news was chosen to be

1UK-related entries were identified by being tagged with category 483 in the index: “Europe, Western– United Kingdom of Great Britain and Northern Ireland.” Further entries were sampled when specific companies were mentioned. 2These reports include two reports on vegetarianism deals with ‘vegetarian’ foods (2000), while all following reports target ‘meat-free foods’ (2001, 2004, 2006b, 2008, 2010, 2017), or additionally ‘free-

269 A. Further Details of the Contextual Data

The Grocer, available via the newspaper database Factiva3; recommended as a good source by some colleagues, the newspaper was founded in 1985 and targets a food industry audience. Via a systematic search of market research reports in the British Library, three reports from the Leatherhead Food Research Association published in the first half of the 1990s were accessed4. Further, a custom-made market research report by Kantar Worldpanel was kindly provided to me for reference by one of my interviewees. Data on the development of the market for ‘free-from meat’ products from 2003 onward was obtained from Euromonitor International. This data was supplemented by targeted searches for older books on meat alternatives, new or novel protein foods, and similar terms, as well as references from the annotated bibliography. Further information on businesses was obtained from Com- panies House, which is the registrar for companies in the UK, and the FAME companies database, which combines various publicly available data about companies in the UK and abroad. In case of uncertainties or gaps in the previously collected data, supplementary news articles and press releases were obtained from Factiva and internet searches, or from ‘the archive of the internet’ on archive.org.

A.2 Key Companies and their Histories

During the process of creating a narrative for the category (see section 4.3), I came across a few companies, which seem rather important for the emergence of a UK meat alternatives category. Supplementing the insights presented before, in the following sections, detailed narratives of these few companies are presented. These cover:

Direct Foods was pioneer in soy and TVP applications for consumers as most were only sold to food businesses. Starting in 1969, it was bought by British Arkady in 1985 and became part of the Haldane Foods Group later. Haldane Foods is another health foods pioneer which started business in the early 1980s, expanded and gained interest from British Arkady. It became namesake

from foods’ (2011, 2013). The most recent report (2017) further included a spreadsheet with processed data of their consumer surveys. As I found out later, Mintel is also used by a number of the smaller organisations in my sample, while bigger organisations prefer other market research organisations. 3The systematic search for articles in The Grocer covered standard terms such as meat-free or meat alternatives described elsewhere (e.g. by Sadler, 2004) and was iteratively amended with terms emerging for the investigation (search terms: ‘meat alternatives’, ‘meat substitutes’, ‘meat analogs’, ‘meat analogues’, ‘mock meat’, ‘fake meat’, ‘vegetarian meat’, ‘vegan meat’, and ‘meat free’). Additional searches were undertaken for prominent brands identified from the emerging narrative (search terms: ‘Quorn’, ‘Linda McCartney’, ‘Redwood’, ‘VBites’, ‘Haldane’, and ‘Goodlife’). 4The Leatherhead reports are titled ‘Vegetarian Foods in the UK’ (1990) and ‘The Vegetarian UK Food[s] Report’ (1993, 1995). — At this point I would like to thank the librarians of the British Library for their efforts to find these reports that had been misplaced and were thus unavailable for a while.

270 A.2. Key Companies and their Histories

of the Haldane Foods Group in 1988, which became one of the most influential early meat alternatives businesses. Currently, it is owned by Hain Celestial and only few of its brands are still alive. Quorn Foods originated from a research project in the 1960s to maintain human protein supply. Equipped with food safety approval in 1984, the first products were released in 1985 in own-label products. In the early 1990s it developed to become the strongest brand in the UK meat alternatives market, and is now world market leader. Linda McCartney Foods was founded in 1991 as a collaboration between frozen food manufacturer Ross Young’s, and celebrity and vegetarian cookbook author Linda McCartney. The brand is produced in license by Hain Celestial, and is a long-time runner-up on meat alternatives. VBites Foods was founded as Redwood Wholefoods in 1994 as a healthy and environmentally-conscious meat alternatives manufacturer and is popular among vegan and health-food consumers.

A.2.1 Direct Foods Peter and Anna Roberts, among others, owned a dairy farm in Hampshire, set up a trust against animal cruelty, became vegans in the early 1960s, and started a health food in Petersfield in 1979 (Shurtleff & Aoyagi, 2014, n. 2385; The Telegraph, 2006). On 7 March 1969 they founded Direct Foods Ltd (company number 00949421, Companies House, n.d.). British Arkady (formerly British Soy Mills), were manufacturing TVP, licensed by Archer Daniels Midland from the late 1960s, and were acquired by them in 1974; this acquisition was related to the booming TVP market in Europe (Shurtleff & Aoyagi, 2014, nn. 1981, 2385)5. In 1969, Peter Roberts approached Arkady about exclusive rights to sell TVP to consumers, but Arkady declined as they were only interested in producing for food manufacturers. The Roberts nonetheless trialled their idea by ordering TVP as “beef chunks or mince,” repackaged it, added a recipe leaflet and created a form for mail order; their first distribution was to acquaintances from animal activist groups, such as their own trust (Shurtleff & Aoyagi, 2014, n. 533). They were marketed, for example, as “beef-style” as the labels stated that the products were vegetarian and included no meat (Shurtleff & Aoyagi, 2014, n. 533). The business

5Note 504 and 506 state that ADM had only started producing TVP from 1967, so 1965 is probably unlikely. However, note 533 about a product launched in 1969 reads “This was Direct Foods’ first product, based on ADM’s TVP purchased from British Soya Mills (British Arkady).” Two options remain, none of which I can confirm or reject: (1) TVP was imported by British Arkady and redistributed, which happened likely in note 504; (2) the product was originally based on some other product, such as the ‘Sosmix’ and the ‘Burger Mix’ in note 2385.

271 A. Further Details of the Contextual Data then started off as they bought more TVP in bulk, repackaged and redistributed them (Shurtleff & Aoyagi, 2014, n. 2686). According to their own knowledge, their company was one of the first in the UK to market dried soy products, such as TVP, directly to consumers and the first to sell products such as dry burger mixes made of these (Shurtleff & Aoyagi, 2014, n. 2345). Direct Foods also introduced a sausage mix ‘Sosmix’ in around 1970, which was originally made by Arkady but Direct Foods asked them to replace the pig fat with hardened vegetable fat; the product became a success, and British Arkady wanted Direct Foods to market a burger mix (Shurtleff & Aoyagi, 2014, n. 2385). In the early 1970s, also catering packs were introduced (Shurtleff & Aoyagi, 2014, n. 595). Within the first year over 20 products in two product ranges were developed (Shurtleff & Aoyagi, 2014, n. 2385): the Protoveg range consisted of meat alternatives such as for burgers or sausages; the Ranch House brand comprised of more convenient products such as bolognese, goulash, or curry (Shurtleff & Aoyagi, 2014, n. 595). In 1972, Brewhurst Health Food Supplies, a major health food distributor, asked to distribute Direct Foods; sales increased thereby, and other distributors and export- ers (to Jamaica, Malta, and Greece) were appointed (Shurtleff & Aoyagi, 2014, n. 2385). Until then a business run from their home, the Roberts rented a warehouse in Petersfield in around 1974, continuing to be a product developer, marketer and mail order house (Shurtleff & Aoyagi, 2014, n. 2385). The founder of Haldane Foods, Brian Welsby, used to mix products under contract for Direct Foods for a short while in the early 1970s (Shurtleff & Aoyagi, 2014, n. 2345)6. Operations expanded through the participation in the Ideal Home Exhibition in London, which attracted many people due to low prices (half of the price of real sausages), easy preparation, health benefits, and the smell (Shurtleff & Aoyagi, 2014, n. 2385). In the late 1970s, first competitors arose, such as Itona in 1978, or Real Foods (date unclear), which all sold TVP to the consumer (Shurtleff & Aoyagi, 2014, n. 2385). A few issues with authorities arose, and were fought at court. Attempts to out- law the names ‘Sosmix’—which was said to sound too much like sausage–or ‘Veget- able Goulash’—which was claimed to be contradictory labelling–, were unsuccessful (Shurtleff & Aoyagi, 2014, n. 2385). As Direct Foods was a large consumer of British Arkady’s TVP, they offered to buy a large share or all of Direct Foods, which would allow the company to make extra margins and to get hold on the consumer market (Shurtleff & Aoyagi, 2014, nn. 2686, 2385); as Arkady was the only supplier to Direct Foods, in case of a negative response,

6Note 2385 claims that Haldane Foods was mixing the products under contract for Direct Foods, placed in a chronological account that must be between 1976; according to note 2345, Haldane Foods was only founded in 1983, and Brian Welsby is said to be active in the business before 1980 in his function to mix products.

272 A.2. Key Companies and their Histories they could have damaged Direct Foods business and made business survival critical (Shurtleff & Aoyagi, 2014, n. 2385). At the point of sale, Direct Foods bought 18 tons of TVP weekly, and had 18 employees, some of which family members (Shurtleff & Aoyagi, 2014, n. 2385). The owners accepted the offer to spend more time on their animal welfare work and health food shop, and in March 1985, Direct Foods became the first part of what Arkady later built into the Haldane Foods Group (Shurtleff & Aoyagi, 2014, nn. 2385, 2686). Some of the original product names such as the ‘Protoveg Sosmix’ still exist. After the assets and brand(s) of Direct Foods had been incorporated into Arkady, the Roberts renamed their company Lavant Street Property Ltd on 25 March 1985 (Companies House, n.d.).

A.2.2 Haldane Foods Haldane Foods Ltd was founded in 1972 in London by George Albert Corderoy (company number 01068279, Companies House, 2017). In the early 1980s, the company changed owners to Blanka and Brian Welsby (Companies House, 2017)7. Haldane was a ‘specialist health foods’ company. Buying up other companies with similar product ranges, Haldane Foods acquired T.J. Bilson & Co. Ltd and Raven Leisure Products Ltd; in March 1985, a majority of the shares of The Regular Tofu Company Ltd were acquired (Companies House, 2017). From 1984 to 1987, Haldane Foods’s assets grew by almost 50%, and the company’s turnover exceeded £2m by 1986, although a small loss was made in 1986 and 1987 (Companies House, 2017). With the acquisition of Direct Foods by British Arkady in 1985, the latter realised the opportunities in adding value to their TVP products as they continued to buy more companies in the sector; and in early 1988 Haldane Foods was bought by British Arkady (Shurtleff & Aoyagi, 2014, n. 2686). According to the annual director’sreport this acquisition for Haldane had been motivated by the search for “outside sources of finance,” as the company had been making losses recently. In an agreement signed on 13 January 1988, British Arkady acquired 75% of Haldane Foods Ltd, and all share capital of Direct Foods Ltd and Vegetarian Feasts Ltd8 was transferred to Haldane. Haldane then became part of Arkady’s Health Foods Division, and new directors from Arkady were appointed alongside the previous directors (Companies House,

7When exactly this happened cannot be traced. Shurtleff and Aoyagi (2014, n. 2345) claim that Brian Welsby had founded Haldane in 1983, so it is possible that it changed owners then. Brian was mixing ingredients for Direct Foods before he took ownership of Haldane Foods (Shurtleff & Aoyagi, 2014, n. 2345). 8And possibly also VegetarianCuisine Ltd, as this company is part of the 1988 returns sheet, but not marked as newly acquired.

273 A. Further Details of the Contextual Data

2017). Arkady officials thought it had the “nicest sounding name,” was considerably large, and the the ‘best’ factory and offices among the acquired companies; thus, the Haldane Foods Group Ltd was formed (Shurtleff & Aoyagi, 2014, n. 2686)9. With more economic power, the company and its subsidiaries were reorganised, two factories were closed, and Realeat Foods was acquired as the company aimed to extend its range towards making vegetarian burgers (Shurtleff & Aoyagi, 2014, n. 2686). Turnovers dropped until 1989, and the Welsby’s quit as directors in May 1989, selling their remaining shares to British Arkady, as well. Saucemasters (75% in 1989, up to 87.5% in 1990) and Genice Foods (76%) were acquired in 1989, as well (Companies House, 2017). Saucemasters had been called Kwality Foods until that acquisition, and manufactured sauces largely for retailer’s own-label, but also produced two products for Haldane before (The Grocer, 1990a). As a result of the restructuring, new products introductions, sales programmes, and turnovers increased dramatically, rising from £1.6m in 1989 to £4.6m in 1990, and £6m in 1991, and grew to become the largest health foods company in the UK. For the first time since 1985, the company made profits again in 1990 (Companies House, 2017), and claimed to be “Europe’s leading healthier food group” (Shurtleff & Aoyagi, 2014, n. 2311). They could benefit by the increasing number of vegetarians and those interested in a healthier and more varied diet; for example, more than “60 million servings of VegeBurgers” were sold in 1990, and manufacturing for private label foods was increasing (Shurtleff & Aoyagi, 2014, n. 2365). The company’s products could then be found in all major grocery multiples and retailers (The Grocer, 1989a). In 1990, Quorn products were started to be used in the VegetableFeasts ready-meals, and in vegetable cutlets, which were both advertised by Haldane (The Grocer, 1990b, 1990d). In the annual reports from 1992 to 1994, the directors report ‘difficult trading conditions’, so that the company could not grow as much as in previous years (Companies House, 2017). In an interview in 1994, then co-director John Mahlich (also director of Arkady and ADM in the UK), claimed, that the powerful competition (e.g. then- Unilever’s Birds Eye) just waited until Haldane Foods had created a market, before launching some products under their own powerful and well-known brands with high advertising budget and much higher negotiation power with retailers. Haldane had only a small advertising budget and owned no well-known brands; at that time, alongside continuation in brand-building, further growth was sought in an expansion into continental Europe using Arkady’sfactories on the continent, as well as increased production of private label for supermarkets. Further, a patent (GB000002294192A)

9At the Companies House no registration could be found for such company; this is possibly because the company had been dissolved before a certain date.

274 A.2. Key Companies and their Histories for the production of mince meat10 was intended to be licensed to other companies, or contracted production to be undertaken (Shurtleff & Aoyagi, 2014, nn. 2687, 2688). Aims to expand even more were limited by Haldane having bought up most of the small food businesses, founded by “devotees of healthier foods or vegetarians” who struggled in the high capital investment and necessary business skills in the food manufacturing business. Most of Haldane’s products were vegan and often classified as health foods; consisting only of a–considerably large–number of health foods shops in 1985, by the mid 1990s, most supermarkets had separate health foods sections (Shurtleff & Aoyagi, 2014, n. 2687). In 1995, Snackmasters and Granose were acquired, and in 1996, some of Genice Foods was disposed of (Companies House, 2017; Shurtleff & Aoyagi, 2014, n. 2451); Granose was a well-established health foods producer, originally run by Seventh Day Adventists (Shurtleff & Aoyagi, 2014, nn. 2288, 2451). In 1998, Haldane Foods Ltd became an independent business unit and was transferred from British Arkady to the ADM Investment (UK) Ltd (Companies House, 2017). At that time, around a third of Haldane’s sales came from third party manufacturing for supermarkets or other companies (Shurtleff & Aoyagi, 2014, n. 3106). From 1995 to 1996, turnover almost doubled to over £16m, but from then stagnated for two years. In 1998, Haldane developed a range of non-dairy desserts, licensing the Linda McCartney brand name, and featured the in national TV ads and in print (Shurtleff & Aoyagi, 2014, n. 3106). In spring 1998, a new EU-legislation for- bade the use of dairy or dairy-like names in non-dairy products, such as “soya milk,” or “yoghurt,” which required a lot of re-labelling. The business was further heavily affected from 1998–2000 by EU-legislation requiring the labelling of ingredients stemming from genetically-modified organisms; simultaneously, wholesalers had announced a boycott of Haldane’s and Linda McCartney’s products if they were not entirely GM-free. Haldane was tentative but in the end announced to only use GM- free products; consumer confidence in the products, which were largely based on soy, was heavily affected (Companies House, 2017; Shurtleff & Aoyagi, 2014, n. 3106; The Grocer, 1998d). The company diversified and reduced costs, but could only get back into profits by 2001 (Companies House, 2017). Especially sales of soy milk proved beneficial in the following years, but price pressure from retailers impaired the business; in 2005 over £15m turnover were achieved, getting close to the sales of 1996 (Compan- ies House, 2017). In late 2006 some assets of Haldane Foods Ltd—the “portfolio of VegeMince, Sosmix and Burgamix ranges under the Granose, Direct Foods, Realeat,

10The mincemeat was “based on wheat protein, free range egg white, seasoning and malt extract for colour” and was said to be the first vegetarian product (in the UK) resembling beef mince (The Grocer, 1993a, 1993b, 1994b).

275 A. Further Details of the Contextual Data

Granose and White Wave brands”—were sold to The Hain Celestial Group, Inc., which also acquired Linda McCartney Foods earlier that year (The Grocer, 2007). Haldane Foods Limited changed its name to ADM Newport Pagnell Limited and stayed at ADM (Companies House, 2017); the food group as such ceased to exist at that time.

A.2.3 Quorn Foods With 10.7% of the world market and 44.4% of the UK market in 2016 (Euromonitor International, 2018), Quorn is national and global market leader for meat alternatives. It’s ingredient ‘myco-protein’, evolved from a planned research project concerned with food security: in the late 1950s, concerns grew over a possible world famine, general protein deficit, and increasing meat prices by the 1980s (Finnigan, 2011; Trinci, 1992). Many projects were concerned with finding bacteria to convert byproducts of the petroleum industry into protein to supplement animal feed, but none were successful on the long term (Trinci, 1992); the British cereal producer Rank Hovis McDougall (RHM) launched a project in 1964 to convert the abundant by-product starches from their production into human-edible protein (Finnigan, 2011; Trinci, 1992). They intended to cultivate a fungus that grew on a wheat starch medium; growing protein through its metabolism, it then could be ground up, for example like soy flour, for further processing (Trinci, 1992). A fungus had multiple advantages over bacteria: it was expected to face less consumer resistance than bacteria through the long history of human usage, for example as yeast; fungus fermentation was a well known process, and edible fungi had been cultivated for example in Germany in WW2; it was expected that such fungus can provide most needed protein and many vitamins; lastly, a fungus was considered to have advantageous organoleptic properties, such as taste, or smell (Trinci, 1992). After a first found fungus could not be developed satisfactorily, in 1968 RHM spent £1m to identify 3000 potential fungi from soil samples around the world; of these, 20 were identified suitable and one Fusarium venenatum (PTA 2684) was chosen to develop further (Trinci, 1992)11. It converts 1 kg of carbohydrates into 136 g of protein12 of highly favourable amino acid profile (Sadler, 1988), and its “filamentous nature . . . gives . . . products a chewiness and succulence akin to meat.” It further retains all colourings and flavourings during cooking (Trinci, 1992); the product is said to resemble white meat more than red meat (The Grocer, 1993b) During the next few years, test-scale production was undertaken and the product

11The fungus was misidentified as Fusarium graminearum until 1998 (Finnigan, 2011). 12Trinci (1992) lists as comparison chicken with 49 g of protein per 1 kg of carbohydrates, pig with 41 g / kg, and cattle with 14 g / kg.

276 A.2. Key Companies and their Histories was rigorously tested to be permitted as a food item: after many microbiological tests, in vitro tests, long-term feeding of animals, and chronic trials with humans it was permitted in 1984 (Trinci, 1992). As RHM’s production could not be upscaled sufficiently, they founded a joint venture called Marlow Foods with Imperial Chemical Industries (ICI), who provided experience in large-scale fermentation processes. The product was trademarked as Quorn, and the Food Standards Committee coined the term ‘myco-protein’ as a generic descriptor for such food (Trinci, 1992). As worries about food security had not come into place, it became more of a health food (Trinci, 1992), but with increasing concern about the global environment, it is said to be a promising alternative to meat (Finnigan, 2011). The first product for public consumption was a savoury pie, a traditional food in order to foster interest by consumers, sold at Sainsbury’s in January 1985, driven by the interest and support by Sir John Sainsbury (Finnigan, 2011; Trinci, 1992); to show consumers that it was a vegetarian food item despite its meaty taste and texture, products carried a “Approved by the Vegetarian Society” logo (The Grocer, 1988a; Trinci, 1992)13. In 1988, products were introduced in Tesco and Waitrose; in 1989 formed the basis for some frozen ready meals for Safeway as well as for Ross Young’s who created Linda McCartney in 1991 (The Grocer, 1988b, 1989b, 1989c). Despite the growing market of manufacturers, catering was responsible for many sales, as well (The Grocer, 1991c). By 1989, 35 products were on the market, including not only main meals, but also entrees and desserts, as well as an independent cooking ingredient (The Grocer, 1989d; Trinci, 1992). Quorn invested £1m in press advertising to promote the brand, on top of bene- fitting from mutual engagement with retailers; the brand saw itself in a response to green consumers and calls for healthier food (The Grocer, 1989d). Brand value increased to over £10m in 1989 (The Grocer, 1989d), and in 1990 its first TV advertising campaign was launched (The Grocer, 1990c). By now widely recognised as ‘an alternative to meat’, its marketing targeted skilled and educated females aged 25–4514(The Grocer, 1990c; Trinci, 1992). In 1990, Quorn became ingredient to some products of the VegetableFeasts range from Haldane Foods, and in 1991 part of a crispbake of competitor Dalepak(The Grocer, 1990d, 1991a). Further plans to grow were manifested by exports to Belgium in late 1991, and the construction of a new factory (The Grocer, 1991c). According to Wikipedia, RHM sold its shares in 1990

13At some point after 1985, the vegetarian society may have either withdrawn its approval of Quorn products, or new products could not be approved anymore because of the usage of battery eggs. Marlow Foods worked with the vegetarian society, and in 2000 had around a third of their range approved, having set up free-range farms (Fox, 2004). By spring 2005, the company’s whole range was approved (according to an advert found in The Vegetarian, Spring 2005, p.2). 14Quorn’s marketing targeted A, B, C1 and C2 females aged 25–45 which report to cut down on meat. This group was said to cover around 50% of all women (Trinci, 1992).

277 A. Further Details of the Contextual Data to ICI, which separated Marlow Foods and other divisions into the Zeneca Group in 1993 (later: AstraZeneca). The factory, opened in 1994, increased production from 1,000 tons annually to 7,000 in 1994 and 14,000 in 1995. With it, the sales that targeted the south-east and London were expanded nationally; the move included a £5m advertising campaign, including national TV spots which were supported by famous sportspeople (The Grocer, 1994c, 1995d). The product was sold in Belgium, the Netherlands, Germany and Ireland (The Grocer, 1994c)15. It’s target group changed to a general audience of healthy eaters, including “slimmers, vegetarians, and those interested in low cholesterol foods,” which were estimated as 25% of the population (The Grocer, 1994a). Quorn’s first branded frozen products, burgers, launched in 1994 intended to more households wanting “added product benefits;” at the same time, the company reformulated some of its recipes to contribute more to healthy eating (The Grocer, 1994a). Within a year, the burger became second most bought product in the UK burger market. Not only through this development, the companies dependence on manufacturers for sales almost disappeared with more than 80% of sales coming from branded products (The Grocer, 1995c); to further boost sales and consumer awareness, the brand distributed 12 million free product vouchers and 3 million on-pack vouchers (The Grocer, 1995b). The next years, Quorn expanded its product range and export, with sales reaching £36m in 1996, which was in part supported by the BSE crisis. Increasing product development and marketing investment continued the brand’s growth; support by sportspeople was dropped to engage a high profile chef in its marketing (The Grocer, 1997b). Hot product sampling reached around a million consumers across the country (The Grocer, 1998e). In 1999, Quorn relaunched some of its products to advocate a more meaty taste in burgers, fillets, and sausages, which were described as “entry level products aimed at mainstream first timers.” At that time, only around 20-30% of Quorn consumers were vegetarian, while the brand aimed to engage more meat reducers and health conscious consumers; Quorn identified 95% of purchasers to be female, with three quarters serving the food to their families (The Grocer, 1999d). In the meat-free market, the brand claimed to be market leader, with 18.3% of the £361.3m market (The Grocer, 1999d). In 2000, sales were around £100m, 80% of which from the UK and 20% from Central and WesternEurope (The Grocer, 2000c). With a relaunch in the UK in 2002, the company tried to emphasise its health promises (The Grocer, 2002a). Having

15These products have either been retracted later on or were only available on small scales. Finnigan (2011) reports that Quorn was only launched in Ireland in 2002; in Germany, Quorn was (re)launched in 2012 (Groß, 2016).

278 A.2. Key Companies and their Histories gained FDA approval in 2000, Quorn entered US markets in 2002, which was seen as a massive growth opportunity (The Grocer, 2000c, 2001, 2002c); some US food campaigners attacked the brand for food poisoning, among others, and criticised the brand for labelling its product as ‘mushroom’ not as ‘fungus’. The campaign also placed adverts in major British newspapers, inviting consumers to put forward their complaints (The Grocer, 2002b). The UK’s Advertising Standards Authority banned the company from using mushroom protein on its packages without further, equally prominent explanation (The Grocer, 2002e). Concerning food safety, the brand was continued to be backed by authorities (The Grocer, 2003a, 2005b) whilst the campaign continued for many years but without success (Watson, 2011a). In May 2003, Quorn was sold to Montagu Private Equity (The Grocer, 2003b). The company’s sales dropped to £84m (The Grocer, 2004a), with 52% of sales to non-vegetarians (The Grocer, 2004c). Exports increased their share to 27% of total turnover (The Grocer, 2004b). Valued at £95m, the brand was sold to Premier Foods plc; the June 2005 deal was worth £172m (The Grocer, 2005a). In November 2005, Premier Foods acquired tofu manufacturer and competitor Cauldron Foods for £27m and integrated it into Quorn producer Marlow Foods (The Grocer, 2005c). With 20% market share, this made Premier the largest player in the UK meat free market valued at £657m in 2005 (The Grocer, 2006a). Quorn was repositioned as healthy more than as vegetarian, with the latter term disappearing from its marketing (Mintel, 2006b; The Grocer, 2006a). In 2007, the brand advertised cooking ingredients strongly, reflecting a growing interest in cooking from scratch (Mintel, 2008); Quorn was then relaunched in 2009, attempting to reach more consumers beyond its vegetarian niche through a health focus, highlighting low fat and high protein in TV and press (Mintel, 2010). In March 2011, Marlow Foods was sold for £205m to Exponent Private Equity and Interme- diate Capital Group (Watson, 2011b). It reverted to targeting vegetarians as its traditional audience, but also dieters as part of the mainstream. As part of such, the brand was recommended by other outlets or brands focusing on weight and dieting, such as Weight Watchers (Mintel, 2011). In 2012, Quorn launched a premium sub-brand called Chef’s Selection and trial-size packaging, while recipes were revised to accom- modate consumer complaints of being bland or lacking flavour (Mintel, 2013). More recently, vegan products became central to new product development, and protein was emphasised strongly. Supported by a new drive in advertising spending through its acquisition by Monde Nissin Corp in October 2015, Quorn reacts to desires of meat reduction and encourages consumers to do so; accordingly, a 2017 campaign relates to desires for a healthier and more sustainable diet (Mintel, 2017). In 2016, Quorn had worldwide sales of USD 273.1m, from which £134.8m were earned in the United Kingdom. In both, the company is market leader; Quorn

279 A. Further Details of the Contextual Data products can be bought in 19 different countries (Euromonitor International, 2017a). Quorn’s consumers are vegetarians, meat eaters, and flexitarians alike; the product became an almost independent product of cuisine, for many consumers existing next to ‘chicken’ or ‘beef’ (Mintel, 2017).

A.2.4 Linda McCartney Foods Being vegetarian herself, Linda McCartney had authored a number of popular vegetarian cookbooks in the 1980s (Linda McCartney’s Foods, 2016). They involved many home-made and pre-mixed meat alternatives. In 1991, she was approached by frozen food company Ross Young’s that recognized a gap in meat-like ready meals; the inten- tion was to create a range of ready meals and meat-style products from her recipes; the meat substitute was based on textured vegetable protein, its recipe refined from another product launched a few years before (Linda McCartney’s Foods, 2016; The Grocer, 1991b, 1993d). The brand sells a large range of ready meals and a number of meat alternatives. Ross Young’s was a frozen food specialist and acquired by United Buscuits (UK) Ltd in 1988 (United Biscuits, 2017). The Linda McCartney brand was set up so that parts of the revenue are paid as royalties by the producer to Linda McCartney Foods to continue develop the range and to help fund her animal charity (The Grocer, 1991b). Since Linda’s death in 1998 (The Grocer, 1998a), her husband Paul McCartney is the sole owner of Linda McCartney Foods Ltd and Linda Enterprises Ltd, which manages the brand royalties (Bureau van Dijk, 2017). Two years after its inception, it was estimated £13m with 14% market share in the non-meat ready meals market (The Grocer, 1993d); the company was said to take market share from all its competitors, such as Heinz, Findus, or Dalepak (The Grocer, 1993c). By that time, the range included, pies, pasties, lasagne, and (since Autumn 1992) even a Quorn-based dish (The Grocer, 1993e). After a first TV campaign in April 1993 (The Grocer, 1993d), the range was relaunched in June 1993 in order to challenge the then-market leader Birds Eye, owned by Unilever at that time: this change involved changing the base ingredient from soy protein to wheat protein in order to be more meat-like, more neutral and “allowing the recipes to incorporate subtle flavours” (The Grocer, 1993c). In its marketing, Linda McCartney was targeting the estimated 40% of the population who are or want to reduce meat intake, and not necessarily the 4% of vegetarians (The Grocer, 1993c). In 1994, the brand expanded to North America and to independent retailers in the UK (The Grocer, 1994d); the company, possibly aided by multiple TV campaigns (The Grocer, 1994d, 1995a), became market leader for frozen ready-made meals in 1994. It had a market share of 25% (Birds Eye: 22%) of the £100m market; the sales of

280 A.2. Key Companies and their Histories sausages, or burgers were continuously growing in size (The Grocer, 1995a). In 1995, chilled foods were added to the range of frozen foods (The Grocer, 1995a). In 1996, the brand is reorganised and becomes part of United Biscuits’ McVitie’s Prepared Foods division (The Grocer, 1996). The brand size now growing proportionally with the market, in 1997 it introduced vegetarian mince aiming to replicate red meat without however including much fat (The Grocer, 1997a); later that year, Haldane Foods produced dairy-free products under licence for the brand. It further uses other brands’ meat-free products, among others from Yves’ Veggie Cuisine from the US (The Grocer, 1998c). In 1998, it introduced slices, as well as a range of internationally inspired food items named Linda McCartney On Tour (The Grocer, 1998f). Further along that year, Linda McCartney and Haldane were facing a boycott by the wholefood trade, in case they were not switching GM-free soy; Linda McCartney subsequently announced a full rejection of GM produce (The Grocer, 1998d). How- ever, in Feburary 1999 a trace of GM soy was found in their products, so that in June, £3m were invested to steam clean the factory in order to get rid of every trace of GM soy; the range was changed to be based on wheat instead of soy protein (Buckingham, 1999; The Grocer, 1999b, 1999c). In July 1999, United Biscuits proclaimed to sell its frozen and chilled foods division (£243m annual turnover), including the Linda McCartney brand, asking for £250m (The Grocer, 1999a). The division (excluding the chilled fish trade) was bought for £190m in October the same year by Heinz, who aimed to continue building the brands (The Grocer, 1999e). In September that year, a subsidiary of Heinz had also acquired a 19.5% share of Hain Celestial (which later bought Linda McCartney off Heinz) in an attempt to push the market for functional and organic foods (The Grocer, 2000a); the deal was part of a “global strategic alliance . . . related to the production and distribution of natural products domestically and internationally;” Hain had bought the trademark Earth’s Best (a baby food line) with capital and shares (The Hain Celestial Group, 2001). Short after, the brand investigated expansion into the US, Australia, and New Zealand (The Grocer, 2000b)16, the expansion was, however, never executed. In late 2000, Linda McCartney launched the Kitchen Garden range of organic and low-fat meat substitute frozen meals; the launch was accompanied by a TV campaign, in which the company targeted its core target group women between 25 and 45 years (The Grocer, 2000d). While Heinz had sold all its shares of the Hain Celestial Group by the end of 2005, it was active before and afterwards in buying brands off Heinz (The Hain Celestial Group, 2006). In June 2006, parallel to a decline in sales of frozen ready meals, the

16It is unclear when the distribution in the US mentioned earlier was stopped again.

281 A. Further Details of the Contextual Data

Hain Celestial Group acquired Linda McCartney from Heinz for around £6.6m; the deal encompassed the brand licensing and its manufacturing facility in Falkenham (The Grocer, 2006b; The Hain Celestial Group, 2007). Hain had also acquired meat alternatives manufacturer Haldane Foods later that year (The Grocer, 2007). Soon after the acquisition, the brand was relaunched and innovation activity was increased again (The Grocer, 2007). Albeit more investment, the brand’smarket share declined from second after Quorn to the fourth position, among others due to several delistings in supermarkets (The Grocer, 2008); sales dropped from £40m in its best times to £10m (Leyland, 2008). The brand counteracted with more involvement of the McCartney family, and introduced more ingredients as opposed to ready meals (Leyland, 2008). Yet another redesign and more products were done in the following year (The Grocer, 2009b); which sometimes involved exclusive or selective supply to supermarkets (The Grocer, 2009a, 2011b), this time successful despite overall decline in market size (Riley, 2010). This was followed by yet another consumer-engaging campaign in 2010 which targeted environmentally and health-conscious consumers in a recipe context; the core consumer base of the brand remained vegetarian (Riley, 2010; The Grocer, 2011a). In 2012, Linda McCartney launched a “prawn-free prawn” (Bamford, 2012). After Hain Celestial bought UK food producer Daniels Group in 2011, and renamed the UK business into Hain Daniels Group, it acquired more UK food producers in 2012 from Premier Foods who had owned Quorn and Cauldron until a year before; this investment was said to strengthen negotiation power especially regarding major retailers; (The Hain Celestial Group, 2012; Zuke, 2012). In 2013, a routine check of the raw soybeans, the company’s meat-free mince revealed some potential health threats in the products; they were sold as frozen own-label mince in all major retailers and via the Linda McCartney brand, while being supplied by another organisation (Phillips, 2013). The products remained withdrawn for six months, in which both the overall market and the Linda McCartney brand dropped in sales during that time; supermarkets Tesco and ASDA ordered their supply of frozen meat-free mince from Meet the Alternative instead (Brooks & Brown, 2014). The brand was polished again in 2016 (Bamford, 2015). While other competitors to Quorn declined in market share, Linda McCartney remained around or below £10m sales in the last few years, becoming runner-up in market leadership albeit with a huge distance (Brooks, 2015; Euromonitor International, 2018). The celebrity support of the brand may be off-putting for some, and media especially gives more negative attention in case of scandals or uncertainties; however, it was frequently named important as Linda’s ethics and values are well known and drive identification especially among the target group of younger and middle-aged women (The Grocer, 1996, 1997c, 1999b).

282 Appendix B

Further Details of the Advertisements

B.1 Detailed Overview of the Sampling Procedure

Since the mid 2000s, the advertising market for meat alternatives is very concentrated, as most of the advertising expenditure in the meat alternatives market came from market leader Quorn and spent on their TV advertising campaigns (Mintel, 2013, 2017). However, print advertising is a popular advertising channel for many companies as it is rather inexpensive and allows to choose a target audience; in 2004 a Mintel market report noted that medium or larger brands published advertisements in broader magazines, such as the “women’s or mainstream food press,” whereas often enough less budget is spent on advertising in specialist magazines, for example those focusing on health or vegetarianism (Mintel, 2004). Accordingly I performed an initial sampling of two magazines, one focusing on ethical consumption, the other one on vegetarianism: Browsing through all issues of the ethical product review magazine the ethical consumer available at the library of Manchester Metropolitan University in October 2016, I could identify two large product comparisons of meat alternatives since 1989 but no advertisement for such products. This discrepancy may be indicative that parts of the audience are likely concerned with replacing meat, but advertisers might be wary of advertising in an ‘ethical’ magazine. Browsing through a few years of issues of The Vegetarian in the library of the Vegetarian Society in Altrincham near Manchester in December 2016 I identified numerous advertisements. This magazine is the member’s magazine of the Vegetarian Society of England with a readership consisting largely of female vegetarians; the magazine focuses on a broad range of issues surrounding vegetarianism, such as cooking, lifestyle, debates, and more. This initial sampling led me to review food magazines more systematically. To identify suitable magazines with a high frequency and a large diversity of meat alternatives advertisements, I sampled popular magazines in the food category of four UK magazine subscription in October and November 2017. I browsed through 2 to 4 issues of each magazine which did focus on food (excluding wine and baking) and counted advertisements for meat alternatives. Accordingly I decided to systematically

283 B. Further Details of the Advertisements sample magazines on the basis of (a) a high ratio of meat alternatives adverts per issue, (b) a considerable number of issues, and (c) the availability in an archive in order to sample advertisements. Detailed information on the websites, the sampled magazines, and the found adverts is available in Appendix B.2. As most magazines are fairly new, many only established in 2016 or 2017, criterion (b) and (c) were the limiting factors; even though not sold on any of these websites as it is a member’s magazine, The Ve- getarian was kept in the sample in particular also because of its long tradition. Three magazines1 were ultimately sampled in February and November 2017 at the British Library and the library of the Vegetarian Society:

• The Vegetarian2: from January/February 1985 to Winter 2017; 152 issues, 76 with at least one sampled advert • Cook Vegetarian! : from November 2008 to June 2015; 78 issues (1–69, 71–79), 77 with at least one sampled advert • Veggie (successor of Cook Vegetarian! ): July 2015 to October 2017; 28 issues (80–108), all with at least one sampled advert • VeganLife: January 2015 to November 2017: 27 issues (5–11, 13–32), 21 with at least one sampled advert

In all sampled issues, I took high-resolution pictures of all advertisements and promotional content featuring meat-free products, for which there is an animal-based counterpart, including sausages, pies, or even vegan pizzas, as well as brand advertising for those brands. As there is a wide variety of more or less meat-like and meat-similar products advertised, a delineation of what was considered a meat alternative advert worthwhile analysing for the purpose of this study was established. Not included were in this study:

• adverts covering less than a quarter page (for convenience purposes and meaningfulness of derived data) • meat-free pet food • plain or marinated tofu (however tofu mince or tofu nuggets due to their naming were included in the sample) • pies, crisp bakes, and pates • meat-free ready meals that did not include meat alternatives (i.e. ‘vegan pizza’ was excluded, but ‘vegan pizza with chorizo-style slices’ was not)

1The magazine Vegan Food and Living had a similarly high advertising content, but out of 23 issues only four were available to the author, and thus this magazine was excluded. 2The Vegetarian carried different names throughout this time period, such as The VegetarianQuarterly (in 1993), and had issue frequencies of four, six, or ten issues annually. During the sampling period, issues were only numbered between the autumn issue of 1994 (V3N2) to Winter 1997 (V6N4).

284 B.1. Detailed Overview of the Sampling Procedure

• generic advertising from 3rd parties, such as (online) supermarkets or retailers (however these were included if the brand or the advertised products including own-label were clearly visible) • generic products where no clear meat alternative relation could be seen (e.g. ‘party pieces’ vs ‘meat-free party sausages’) • advertising for product ranges unless more than 33 % of displayed products are meat alternatives or visibly made with meat alternatives

Brand advertising was included when it was clear from the advertising content that the brand sells meat alternatives, or when the reader of the respective magazine is likely to know that the brand sells meat alternatives, for example through editorial content or previous advertisements; this mostly accounted for the brands Fry’s, Linda McCartney’s, VBites, Quorn, and Vegusto. The final sample consisted of 557 printed advertisements and considering duplicates 319 unique advertisements. To ensure that the breadth of advertisements and advertising strategies is covered in the sample, those previously identified adverts in the 37 screened magazines screened to identify suitable magazines (all published in 2016 and 2017) were reviewed again as part of a spot check. Only two out of in total ten adverts found outside of the sample did not have an exact duplicate in the sample. One was a bigger, slightly re- designed version of an advert found in the sample, while the other one was a very similar to others in the sample, with a different picture and picture caption, but the same content. It may thus be assumed that while there might be specific adverts without a duplicate in the sample, genuinely different advertisements are unlikely to have been printed in the sources. Particularly for the qualitative analysis, this would have been a relevant criticism; for the quantitative analysis, the ten extra advertisements would make up around 10% in the last time period, and since they are duplicates they will not have altered frequencies too much. Therefore can be assumed that at least for the period covered in this spot check, the advertising sample is representative of the diversity of print adverts in UK food magazines. Possibly a shortcoming of the advertising data is that it only originates from vegetarian and vegan magazines, which furthermore are more concentrated in recent years. While the concentration somewhat reflects the trends in the market of increasing interest in meat-free foods and diets (see chapter 4), and is very much bound by practical concerns of data availability. Furthermore, although more recently the amount of meat reducers consuming meat alternatives has increased and is roughly at par with the consumptions of meat avoiders (see Kantar Worldpanel, 2016), the larger share throughout its history has been consumed by vegetarians. Although some adverts appear as if they aimed at meat reducers, it needs to be assumed that a majority of

285 B. Further Details of the Advertisements the audience in the magazines are actually vegetarians3, and this concern might not be too big an issue. While also other magazines for different audiences had a few advertisements for meat alternatives printed, a spot check suggested that advertisements substantially different from the sampled ones were not to be expected.

B.2 Magazine Selection

The food and drink/cooking and food categories of the following websites were sampled: newsstand.co.uk, magazine.co.uk, whsmith.co.uk, and isubscribe.co.uk Magazines being sold by three or more of these websites (as an indicator for circulation) and not dealing with baking or wine were sampled. From the remaining 15 magazines, two to four recent issues were screened for meat alternatives advertisements. Exceptions are BBC Good Food for which no issues were available to me, and PlantBased which was a new magazine at the time of sampling so that only one issue could be found. Magazines were selected on the basis of (a) a high ratio of adverts per issue, (b) a considerable number of issues, and (c) the availability in an archive in order to sample advertisements.

Table B.1: Overview of the food magazines which were screened initially, and how many ads were found per screened issue.

Magazine Since Issued # Ads # Issues Ratio

BBC Good Food 1989 N/A 0 0 delicious. 2003 monthly 0 4 0.00 food to love 2016 monthly 0 4 0.00 free-from heaven 2014 monthly 1 4 0.25 gluten-free heaven ? 1 4 0.25 good things 2013 bimonthly 0 2 0.00 Great British Food 2007 annual 0 4 0.00 Healthy Food Guide 2011 monthly 0 2 0.00 olive 2003 monthly 0 2 0.00 PlantBased 2017 1 1 1.00 vegan food & living 2016 bimonthly 6 3 *2.00

3The readership of The Vegetarian consists of 92% vegetarians or vegans, and 75% females (Vegetarian Society, n.d.-a); the readership of VeganLife consists of 64% vegans, and another 18% vegetarians, while 87% are females (Vegan Life Mag, n.d.); for the readership of Veggie even through personal contact I could not find out any data on vegetarian readership, but their readership consists of 80% females (personal communication, 3 May 2018). The ratio of vegetarian readers is likely to be similar to the other two magazines. By the nature of the magazines’ contents it is furthermore to be expected that the remaining part of the audience is at least interested in vegetarian or vegan food and/or reducing their meat intake.

286 B.3. Software used for Coding

Magazine Since Issued # Ads # Issues Ratio

vegan life 2014 bimonthly 3 4 *0.75 vegetarian living 2010 monthly 1 4 0.25 veggie4 2008 monthly 15 4 *3.75 woman & home feel good food 2008 quarterly 0 3 0.00

B.3 Software used for Coding

A custom coding software for the adverts was employed for the following reasons: Firstly, many of the existing applications to code visual data were unavailable for my operating system; secondly, I desired features in my analysis that allowed me to use a digital coding schedule. For this, a coding schedule could be defined externally and loaded into the software, so that during the exploratory coding phase, the software prompts the coder to go through each of the categories and code the advert corres- pondingly; it also reminds the coder of the purpose, scope, and background to the code if defined. As the coding schedule was populated with defined codes in the structured coding phase, these codes and their descriptions were suggested to the coder in a drop-down menu (Figure B.1c; for a full view of a single coded advert within the programme, see Figure B.1a). The integration of the coding schedule increased structure and consistency which in its function resembles the use of paper-based coding schedules. Thirdly, I wanted to merge duplicate images during the analysis but still be able to analyse them separately. The software was constructed so that data points could be merged to a set and only one item of a set would need to be coded while all obtained the codes given. This process reduced the amount of data coded and thus the time to code by more than a third. Lastly, I wanted full control over the coded data and the following analysis which was achieved by using a standard SQLite-database that could be read in most external programming languages, such as the statistical programming language R. A feature which other software also has (but is not possible for example when choosing a spreadsheet) is that codes could be attributed to specific parts of the advertisement. As then all marked headlines, texts, or images coded could be viewed alongside each other, this allowed an increased productivity in the exploration of the codes and their interpretation (Figure B.1b).

4Before 2015, Veggie was called cook vegetarian.

287 B. Further Details of the Advertisements

b) c)

Figure B.1: Exemplary view of the coding software: (a) Coded advertisement with its attributes; (b) overview of codings for a specific code; (c) use of the coding schedule.

B.4 Statistical Considerations

The χ2-test allows to identify whether an observed variance in distribution across categories (i.e. time periods and coded/not coded) significantly differs from an expected distribution, or whether the distribution might also occur by chance (Pearson, 1900). The choice for a χ2-test was motivated by the fact that it would be able to take into account the multiple time periods identified, and that no hypothesis was formulated in advance; furthermore, as the χ2-test is a non-parametric test, the distribution is irrelevant for its results. However, among others the χ2-test assumes the data across the time periods to be independent, and that “the value of the cell expecteds should be 5 or more in at least 80% of the cells, and no cell should have an expected of less than one” (McHugh, 2013, p. 144). The requirement of independence can be met if for simplicity it is assumed that brands have no continuous advertising strategy, but decide separately on new adverts in every year or time period. In the specific case of this study, the latter requirement is met when there are 31 or more observations for a

288 B.5. Bibilographic Details of the Collected Advertisements code, which requires codes with less observations to be aggregated to bigger groups or otherwise treated with caution5. Other statistical evaluations considered were Fisher’s exact test and a logistic regression. Fisher’s exact test however can only deal with 2x2 matrices and would thus only allow for two time periods; this would lead to the test possibly overlooking U- shaped or inverted-U distributions of codes. A logistic regression would have been able to statistically assert directionality of the findings or in case of polynomial fits even identify U-shaped patterns; it would furthermore be able to account for the impact of brands on the developments, but initial tests revealed that the developments were over-fitted and thus only a regression based on coding and year would be meaningful.

B.5 Bibilographic Details of the Collected Advertisements

In the following, all sampled adverts are listed in chronological order; the number and references of reprints are indicated after the earliest print of an advert.

1985-01 Direct Foods. (1985, May/June). “Look Out for Savoury Bakes”. The Veget- arian, p. 19. 1985-02 Cauldron. (1985, July/August). [Advertisement for Tofu Burgers]. The Vegetarian, p. 22. 1986-01 Direct Foods. (1986, September/October). [Advertisement for a Nut & Grain Bake and a Vegetable Pastie Mix]. The Vegetarian, p. 23. 1986-02 Realeat. (1986, November/December). “Vege Burger — VegeBanger — The Natural Choice”. The Vegetarian, p. 24. 1987-01 Realeat. (1987, July/August). “Leaves meat standing.”. The Vegetarian, p. 29. 1989-01 The Dietburger. (1989, May/June). “Still adding an Egg?”. The Vegetarian, p. 19. 1989-02 The Dietburger. (1989, July/August). “Recipe of the Month”. The Veget- arian, p. 19.

5 An expected value is calculated by E = MC × MR/N, with MC and MR here representing the total counts per time period and per code respectively, and N the total number of codes. The two time periods with the smallest number of adverts are the first (MC1 = 47) and the last (MC4 = 91). 80% of the eight cells are 1.6 cells, and as MC1 is more than half of MC4, the second assumption of 80% of cells having expected values of more than 5 needs to hold for reliable results. Thus, E = 5 for MC4 = 91 leads to MR = E × N/MC4 ≈ 30.6. Conversely there should also be a maximum of 557 − 31 = 526 observations for each particular code, but this is not the case for any observations.

289 B. Further Details of the Advertisements

1989-03 Dalepak. (1989, September/October). “Take a fresh look at vegetables”. The Vegetarian, p. 30. 1989-04 The Dietburger. (1989, September/October). “Vive la Burger”. The Veget- arian, p. 63. 1989-05 Granose. (1989, November/December). “As always, a new roast from Granose is something rather special.”. The Vegetarian, p. 28. 1 reprint: The Vegetarian, Jan/Feb 1990, p. 68. 1989-06 The Dietburger. (1989, November/December). “Dietburger Mix before. . .”. The Vegetarian, p. 31. 1990-01 The Dietburger. (1990, Jan/Feb). “New Decade Diet?”. The Vegetarian, p. 19. 2 reprints: The Vegetarian, Mar/Apr 1990, p. 55.; The Vegetarian, May/June 1990, p. 20. 1990-02 The Dietburger. (1990, July/Aug). “Self catering holiday?”. The Vegetarian, p. 39. 1990-03 The Dietburger. (1990, Sep/Oct). “Dip into a Burger”. The Vegetarian, p. 34. 1991-01 The Dietburger. (1991, Jan/Feb). “New Year Diet?”. The Vegetarian, p. 43. 1991-02 No Bull. (1991, Mar/Apr). “So Tasty, even your Burger experts won’t have any beef”. The Vegetarian, p. 19. 1991-03 The Dietburger. (1991, May/June). “Two new recipe ideas for Dietburger Mix”. The Vegetarian, p. 36. 2 reprints: The Vegetarian, July/Aug 1991, p. 41.; The Vegetarian, Sep/Oct 1991, p. 38. 1991-04 Tivall. (1991, July/Aug). “No Preservatives, No Artificial Colourings, No Meat. . . No wonder they taste so good!”. The Vegetarian, p. 68. 2 reprints: The Vegetarian, Apr 1992, p. 84.; The Vegetarian, Sep 1992, p. 84. 1991-05 Country Club Cuisine. (1991, Sep/Oct). “A new experience in vegetables”. The Vegetarian, pp. 44–45. 1991-06 Batchelors. (1991, Nov/Dec). “Healthy eating for vegetarians”. The Veget- arian, pp. 24–25. 1992-01 Linda McCartney’s. (1992, Mar). “Try this. . . another good reason not to eat meat”. The Vegetarian, p. 12. 1992-02 Direct Foods. (1992, Apr). “Bags of Choice”. The Vegetarian, p. 2. 1 reprint: The Vegetarian, Sep 1992, p. 56. 1992-03 Granose. (1992, Oct). “If any tins can satisfy, Granose cans can!”. The Vegetarian, p. 19. 1993-01 Realeat. (1993, Feb). “Make meal times sparkle. Try our Bangers.”. The Vegetarian, p. 13.

290 B.5. Bibilographic Details of the Collected Advertisements

1993-02 Granose. (1993, Feb). “Our latest offering!”. The Vegetarian, p. 2. 1993-03 Vegetable Feasts. (1993, Spring). “Try our new Toad in the Hole. You’d have to be hopping mad not to.”. The Vegetarian, p. 2. 1999-01 Linda McCartney’s. (1999, Summer). “Linda McCartney — Cooking with kindness”. The Vegetarian, p. 64. 1999-02 Quorn. (1999, Winter). “NOW serving . . .mince, pieces, Deli slices and Deli rashers”. The Vegetarian, p. 2. 1 reprint: The Vegetarian, Summer 2000, p. 46. 2000-01 Dalepak. (2000, Summer). “How to please everybody at meal times with Dalepak’s delicious vegetarian range”. The Vegetarian, p. 37. 2000-02 Redwood. (2000, Autumn). “Reader Offer”. The Vegetarian, p. 19. 2000-03 Dalepak. (2000, Autumn). “For the discerning”. The Vegetarian, p. 54. 2000-04 Quorn. (2000, Winter). “christmas cooking with Quorn”. The Vegetarian, pp. 26–27. 2000-05 Dalepak. (2000, Winter). “Good food, no faff.”. The Vegetarian, p. 63. 2001-01 Morningstar Farms. (2001, Spring). “The great taste of bacon without the pig”. The Vegetarian, p. 13. 2001-02 Dalepak. (2001, Spring). “Vegtastic Vegetarian Ideas”. The Vegetarian, p. 54. 1 reprint: The Vegetarian, Summer 2001, p. 22. 2001-03 Quorn. (2001, Winter). “Quorn: a healthy alternative”. The Vegetarian, pp. 34–35. 2001-04 Redwood. (2001, Winter). “NEW”. The Vegetarian, p. 64. 2002-01 Goodlife. (2002, Spring). “Brighten up your Life!”. The Vegetarian, p. 2. 2002-02 Fry’s. (2002, Spring). “Fry’s Special Vegetarian Foods”. The Vegetarian, p. 20. 2002-03 Fry’s. (2002, Spring). “Are you Frying tonight?”. The Vegetarian, p. 21. 2002-04 Goodlife. (2002, Summer). “Brighten up your Life!”. The Vegetarian, p. 19. 2002-05 Dalepak. (2002, Summer). “The ‘Penny’s latest boyfriend doesn’t eat meat’ pak”. The Vegetarian, p. 63. 2 reprints: The Vegetarian, Autumn 2002, p. 63.; The Vegetarian, Winter 2002, p. 63. 2002-06 Redwood. (2002, Summer). “All the taste of the ocean. But none of the fish.”. The Vegetarian, p. 64. 1 reprint: The Vegetarian, Autumn 2002, p. 64. 2002-07 Fry’s. (2002, Winter). “Now we want your approval”. The Vegetarian, p. 24. 1 reprint: The Vegetarian, Spring 2003, p. 52. 2002-08 Tofurky. (2002, Winter). “Of all the Surprises under the tree that year, none brought more Amazement than the Tofurky”. The Vegetarian, p. 28.

291 B. Further Details of the Advertisements

2002-09 Quorn. (2002, Winter). “Quorn. . . for a sophisticated vegetarian Christ- mas”. The Vegetarian, pp. 34–35. 2002-10 Redwood. (2002, Winter). “great news for turkeys.”. The Vegetarian, p. 64. 2003-01 Tofurky. (2003, Spring). “America’s Famous Tofurky — Now Available in UK Packaging!”. The Vegetarian, p. 24. 2003-02 Dalepak. (2003, Spring). “Make a Worldof Difference to Meal Times”. The Vegetarian, p. 63. 6 reprints: The Vegetarian, Summer 2003, p. 63.; The Vegetarian, Autumn 2003, p. 63.; The Vegetarian, Winter 2003, p. 63.; The Vegetarian, Spring 2004, p. 63.; The Vegetarian, Summer 2004, p. 63.; The Vegetarian, Autumn 2004, p. 63. 2003-03 Redwood. (2003, Spring). “Extinct? — It Need Not Be This Way”. The Vegetarian, p. 64. 2003-04 Marks & Spencer. (2003, Summer). “A Vegetarian Feast at Marks & Spen- cer”. The Vegetarian, p. 13. 2003-05 Safeway. (2003, Summer). “Simply Vegetarian! Simply Irresistible!”. The Vegetarian, p. 26. 2003-06 Fry’s. (2003, Summer). “Special vegetarian food”. The Vegetarian, p. 40. 3 reprints: The Vegetarian, Autumn 2003, p. 19.; The Vegetarian, Winter 2003, p. 54.; The Vegetarian, Spring 2004, p. 26. 2003-07 Redwood. (2003, Summer). “redwood”. The Vegetarian, p. 64. 2003-08 VeggieMaster. (2003, Autumn). “100% Natural and Tasty Meat-free Meal”. The Vegetarian, p. 13. 3 reprints: The Vegetarian, Winter 2003, p. 13.; The Vegetarian, Spring 2004, p. 13.; The Vegetarian, Summer 2004, p. 15. 2003-09 Redwood. (2003, Autumn). “Special September Offer”. The Vegetarian, p. 64. 2003-10 Quorn. (2003, Winter). “Winter warmers”. The Vegetarian, pp. 36–37. 2003-11 Redwood. (2003, Winter). [Advertisement for a roast]. The Vegetarian, p. 64. 2004-01 Redwood. (2004, Spring). “March special offers”. The Vegetarian, p. 64. 2004-02 Fry’s. (2004, Summer). “Special vegetarian food”. The Vegetarian, p. 22. 5 reprints: The Vegetarian, Autumn 2004, p. 55.; The Vegetarian, Winter 2004, p. 55.; The Vegetarian, Spring 2005, p. 30.; The Vegetarian, Summer 2005, p. 28.; The Vegetarian, Autumn 2005, p. 50. 2004-03 Marks & Spencer. (2004, Summer). “Food Love at Marks & Spencer”. The Vegetarian, p. 24. 2004-04 VeggieMaster. (2004, Autumn). “100% Natural and TastyMeat-free Meal”. The Vegetarian, p. 43. 5 reprints: The Vegetarian, Winter 2004, p. 15.; The Vegetarian, Spring 2005,

292 B.5. Bibilographic Details of the Collected Advertisements

p. 16.; The Vegetarian, Summer 2005, p. 25.; The Vegetarian, Autumn 2005, p. 21.; The Vegetarian, Winter 2005, p. 21. 2004-05 Dalepak. (2004, Winter). “Freeze your Imagination”. The Vegetarian, p. 63. 5 reprints: The Vegetarian, Spring 2005, p. 63.; The Vegetarian, Summer 2005, p. 63.; The Vegetarian, Autumn 2005, p. 17.; The Vegetarian, Winter 2005, p. 25.; The Vegetarian, Spring 2006, p. 29. 2004-06 Redwood. (2004, Winter). “All the Taste of the Ocean . . . — . . .but none of the Fish!”. The Vegetarian, p. 64. 2005-01 Quorn. (2005, Spring). “Aneggciting announcement”. The Vegetarian, p. 2. 2005-02 Quorn. (2005, Summer). “So easy, so tasty, so spicy! Quorn Fajita Strips”. The Vegetarian, p. 39. 2005-03 Redwood. (2005, Summer). “NEW from the Redwood Vegideli — Veget- arian and vegan — naturally”. The Vegetarian, p. 64. 2 reprints: The Vegetarian, Autumn 2005, p. 64.; The Vegetarian, Winter 2005, p. 64. 2005-04 Quorn. (2005, Autumn). “So tasty, they make eating in like dining out”. The Vegetarian, p. 11. 2005-05 Fry’s. (2005, Winter). “100% Vegan”. The Vegetarian, p. 30. 2005-06 Thank Goodness. (2005, Winter). “Introducing Bernadette McCarthy and her range of delicious and convenient nutroasts”. The Vegetarian, p. 30. 2005-07 Quorn. (2005, Winter). “’tis the season to be tasty”. The Vegetarian, p. 9. 2006-01 Fry’s. (2006, Spring). “100% Vegan”. The Vegetarian, p. 41. 2006-02 Redwood. (2006, Spring). “WINNER —”Best Vegan Food Product””. The Vegetarian, p. 64. 2006-03 Quorn. (2006, Summer). “Lunchtime’s now much more deli-cious with Quorn Deli Slices”. The Vegetarian, p. 15. 2006-04 Fry’s. (2006, Summer). “100% Vegan”. The Vegetarian, p. 52. 1 reprint: The Vegetarian, Autumn 2006, p. 28. 2006-05 Redwood. (2006, Summer). “NEW from the Redwood Vegideli — Veget- arian and vegan — naturally”. The Vegetarian, p. 64. 1 reprint: The Vegetarian, Autumn 2006, p. 64. 2006-06 Cauldron. (2006, Winter). “Introducing Our New Frozen Range”. The Vegetarian, p. 63. 2006-07 Redwood. (2006, Winter). “WINNER —”Foods For Life Award 2006””. The Vegetarian, p. 64. 2007-01 Redwood. (2007, Spring). “AtRedwood there are times when less = more”. The Vegetarian, p. 64. 2007-02 Redwood. (2007, Summer). “Natural healthy vegan foods”. The Veget- arian, p. 64.

293 B. Further Details of the Advertisements

2 reprints: The Vegetarian, Winter 2007, p. 64.; The Vegetarian, Summer 2008, p. 64. 2007-03 alpro soya. (2007, Winter). “Tasty Tofu!”. The Vegetarian, p. 17. 2007-04 Yagga. (2007, Winter). “100% organic, 100% vegan and less than 1% fat!”. The Vegetarian, p. 63. 2008-01 Yagga. (2008, Spring). “Yagga — VeggieStuff’s most popular product!”. The Vegetarian, p. 33. 2008-02 Fry’s. (2008, Spring). “New!”Veg Express” are new vegan meal solutions from Fry’s”. The Vegetarian, p. 39. 2008-03 Redwood. (2008, Spring). “Naturally Versatile Delicious Chorizo Style Chunks”. The Vegetarian, p. 64. 2008-04 Grassington’s. (2008, Autumn). “New look — New range — New recipes — Same passion for Healthier eating”. The Vegetarian, p. 37. 2008-05 Redwood. (2008, Autumn). “foods for body and soul”. The Vegetarian, p. 64. 2008-06 Grassington’s. (2008, November). “Be one of the first to try a NEW meat substitute which tastes as good as it looks!”. Cook Vegetarian!, p. 108. 3 reprints: The Vegetarian, Summer 2009, p. 15.; The Vegetarian, Summer 2009, p. 19.; The Vegetarian, Autumn 2009, p. 15. 2008-07 Yagga. (2008, November). “100% Organic & Vegan”. Cook Vegetarian!, p. 17. 2008-08 Fry’s. (2008, November). “New vegan meal solutions from FRY’s.. .”. Cook Vegetarian!, p. 20. 2008-09 Goodlife. (2008, November). “New!”. Cook Vegetarian!, p. 42. 2008-10 Wicken Fen. (2008, November). “Chilled and Frozen”. Cook Vegetarian!, p. 60. 2 reprints: Cook Vegetarian!, February 2009, p. 78.; Cook Vegetarian!, Dec/Jan 2009, p. 14. 2008-11 Vegetarian’s Choice. (2008, November). “Vegetarians Choice”. Cook Veget- arian!, p. 91. 2008-12 Grassington’s. (2008, Winter). “Be one of the first to try a NEW meat substitute which tastes as good as it looks!”. The Vegetarian, p. 23. 11 reprints: Cook Vegetarian!, February 2009, p. 108.; Cook Vegetarian!, March 2009, p. 108.; Cook Vegetarian!, April 2009, p. 108.; Cook Vegetarian!, May 2009, p. 108.; Cook Vegetarian!, June 2009, p. 108.; Cook Vegetarian!, July 2009, p. 108.; Cook Vegetarian!, August 2009, p. 108.; Cook Vegetarian!, September 2009, p. 108.; Cook Vegetarian!, November 2009, p. 18.; Cook Vegetarian!, November 2009, p. 108.; Cook Vegetarian!, Dec/Jan 2009, p. 108.

294 B.5. Bibilographic Details of the Collected Advertisements

2008-13 Redwood. (2008, Winter). “Carved with compassion. . . served with love!”. The Vegetarian, p. 64. 2009-01 Goodlife. (2009, February). “Nut Cutlet”. Cook Vegetarian!, p. 72. 2009-02 Veggie World. (2009, February). [Advertisement for their online shop]. Cook Vegetarian!, p. 76. 1 reprint: Cook Vegetarian!, Dec/Jan 2009, p. 14. 2009-03 Tivall. (2009, February). “The Healthier Alternative To Meat”. Cook Ve- getarian!, p. 96. 4 reprints: Cook Vegetarian!, August 2009, p. 2.; Cook Vegetarian!, March 2010, p. 54.; Cook Vegetarian!, April 2010, p. 51.; Cook Vegetarian!, October 2010, p. 44. 2009-04 Goodlife. (2009, March). “2 wild mushroom & mature Cheddar quarterp- ounders”. Cook Vegetarian!, p. 71. 1 reprint: Cook Vegetarian!, May 2009, p. 40. 2009-05 granose. (2009, April). “granose — The natural choice for vegetarians for over 100 years”. Cook Vegetarian!, p. 10. 5 reprints: Cook Vegetarian!, December 2009, p. 7.; Cook Vegetarian!, January 2010, p. 5.; Cook Vegetarian!, February 2010, p. 64.; Cook Vegetarian!, March 2010, p. 50.; Cook Vegetarian!, June 2010, p. 41. 2009-06 Provamel. (2009, May). “Cordon Bleu that’s a rather refreshing shade of Green.”. Cook Vegetarian!, p. 6. 2009-07 Tivall. (2009, May). “Soya Good!”. Cook Vegetarian!, p. 73. 2 reprints: Cook Vegetarian!, December 2009, p. 48.; Cook Vegetarian!, August 2010, p. 14. 2009-08 Veggie World. (2009, May). [Advertisement for their online shop]. Cook Vegetarian!, p. 84. 2009-09 Goodlife. (2009, June). “from our kitchen straight to yours”. Cook Veget- arian!, p. 63. 1 reprint: Cook Vegetarian!, July 2009, p. 30. 2009-10 Redwood. (2009, Summer). “Probably the Best Vegan Quarter Pounder Money can Buy”. The Vegetarian, p. 64. 3 reprints: The Vegetarian, Autumn 2009, p. 64.; The Vegetarian, Winter 2009, p. 64.; The Vegetarian, Spring 2010, p. 64. 2009-11 Redwood. (2009, Summer). “UK’s highest rated ethical producer of vegetarian foods 2009”. The Vegetarian, p. 64. 2009-12 Provamel. (2009, Summer). “Organically grown. Ethically sourced. Utterly delicious.”. The Vegetarian, p. 9. 2009-13 Vegetarian’s Choice. (2009, July). “The choice for Vegetarians!!!”. Cook Vegetarian!, p. 54. 1 reprint: Cook Vegetarian!, August 2009, p. 34.

295 B. Further Details of the Advertisements

2009-14 Tivall. (2009, July). “Have a feast Alfresco”. Cook Vegetarian!, p. 58. 2009-15 Vegetarian’s Choice. (2009, October). “The choice for Vegetarians!!!”. Cook Vegetarian!, p. 103. 1 reprint: Cook Vegetarian!, November 2009, p. 76. 2009-16 Fry’s. (2009, October). “New!”. Cook Vegetarian!, p. 82. 2009-17 Redwood. (2009, December). “The Natural Alternative”. Cook Veget- arian!, p. 37. 1 reprint: Cook Vegetarian!, January 2010, p. 7. 2009-18 Goodlife. (2009, Dec/Jan). [Advertisement for a roast and a bake]. Cook Vegetarian!, p. 46. 2009-19 Vegetarian’s Choice. (2009, Dec/Jan). “Vegetarians Choice”. Cook Veget- arian!, p. 66. 2010-01 Fry’s. (2010, January). [Advertisement for their range]. Cook Vegetarian!, p. 30. 2010-02 Vegetarian’s Choice. (2010, January). “The choice for Vegetarians!!!”. Cook Vegetarian!, p. 30. 2010-03 granose. (2010, February). “Store Cupboard Essentials”. Cook Vegetarian!, p. 108. 2010-04 Fry’s. (2010, March). [Advertisement for their range]. Cook Vegetarian!, p. 5. 1 reprint: Cook Vegetarian!, April 2010, p. 5. 2010-05 Goodlife. (2010, June). “Goodlife”. Cook Vegetarian!, p. 23. 1 reprint: Cook Vegetarian!, July 2010, p. 64. 2010-06 Cauldron. (2010, June). “Cauldron”. Cook Vegetarian!, p. 34. 2010-07 Goodlife. (2010, June). “Goodlife”. Cook Vegetarian!, p. 60. 3 reprints: Cook Vegetarian!, August 2010, p. 8.; Cook Vegetarian!, October 2010, p. 24.; Cook Vegetarian!, December 2010, p. 20. 2010-08 Redwood. (2010, Summer). “foods for body and soul”. The Vegetarian, p. 64. 2010-09 Fry’s. (2010, August). [Advertisement for their range]. Cook Vegetarian!, p. 72. 2010-10 Goodlife. (2010, September). “Goodlife — discover BBQ meatfree”. Cook Vegetarian!, p. 68. 2010-11 Fry’s. (2010, September). [Advertisement for their range]. Cook Vegetarian!, p. 82. 2010-12 Tivall. (2010, September). “King of Convenience”. Cook Vegetarian!, p. 97. 2010-13 Quorn. (2010, Autumn). “It’s not a novel, but its recipes are.”. The Veget- arian, p. 2. 1 reprint: Cook Vegetarian!, November 2010, p. 5.

296 B.5. Bibilographic Details of the Collected Advertisements

2010-14 Quorn. (2010, October). “For healthier eating, take a leaf out of our book.”. Cook Vegetarian!, p. 5. 2010-15 Garden Gourmet. (2010, November). “100% vegetarian, 200% taste”. Cook Vegetarian!, p. 30. 2 reprints: Cook Vegetarian!, December 2010, p. 26.; Cook Vegetarian!, January 2011, p. 15. 2010-16 Goodlife. (2010, November). “Goodlife . . .Naturally delicious”. Cook Veget- arian!, p. 54. 2010-17 Fry’s. (2010, December). [Advertisement for their range]. Cook Vegetarian!, p. 88. 2010-18 Redwood. (2010, December). “the 100% natural vegetarian alternative”. Cook Vegetarian!, p. 99. 2011-01 Goodlife. (2011, January). “Goodlife — NEW!”. Cook Vegetarian!, p. 23. 1 reprint: Cook Vegetarian!, February 2011, p. 108. 2011-02 Redwood. (2011, January). “the 100% natural vegetarian alternative”. Cook Vegetarian!, p. 43. 2011-03 Fry’s. (2011, January). [Advertisement for their range]. Cook Vegetarian!, p. 72. 2011-04 Garden Gourmet. (2011, February). “100% vegetarian, 200% taste”. Cook Vegetarian!, p. 107. 4 reprints: Cook Vegetarian!, March 2011, p. 5.; Cook Vegetarian!, April 2011, p. 107.; Cook Vegetarian!, May 2011, p. 115.; Cook Vegetarian!, June 2011, p. 107. 2011-05 Redwood. (2011, February). “new from Redwood!”. Cook Vegetarian!, p. 2. 2011-06 Redwood. (2011, February). “New tricks”. Cook Vegetarian!, p. 32. 2011-07 Fry’s. (2011, February). [Advertisement for their range]. Cook Vegetarian!, p. 38. 3 reprints: Cook Vegetarian!, April 2011, p. 48.; Cook Vegetarian!, May 2011, p. 59.; Cook Vegetarian!, June 2011, p. 64. 2011-08 Goodlife. (2011, March). “Goodlife”. Cook Vegetarian!, p. 108. 4 reprints: Cook Vegetarian!, April 2011, p. 19.; Cook Vegetarian!, May 2011, p. 116.; Cook Vegetarian!, June 2011, p. 108.; Cook Vegetarian!, July 2011, p. 36. 2011-09 Quorn. (2011, June). “Catch of the day.”. Cook Vegetarian!, p. 21. 8 reprints: The Vegetarian, Summer 2011, p. 30.; Cook Vegetarian!, July 2011, p. 27.; Cook Vegetarian!, August 2011, p. 7.; The Vegetarian, Autumn 2011, p. 64.; Cook Vegetarian!, October 2011, p. 15.; Cook Vegetarian!, November 2011, p. 19.; Cook Vegetarian!, January 2012, p. 99.; Cook Vegetarian!, February 2012, p. 17. 2011-10 Quorn. (2011, June). “Fish without the Fishing”. Cook Vegetarian!, p. 7. 2011-11 Goodlife. (2011, July). “Goodlife — Naturall delicious, naturally meat & dairy free”. Cook Vegetarian!, p. 100.

297 B. Further Details of the Advertisements

5 reprints: Cook Vegetarian!, August 2011, p. 108.; Cook Vegetarian!, September 2011, p. 108.; Cook Vegetarian!, October 2011, p. 108.; Cook Vegetarian!, Novem- ber 2011, p. 108.; Cook Vegetarian!, November 2012, p. 72. 2011-12 Quorn. (2011, July). “Best Ever Banger”. Cook Vegetarian!, p. 7. 2011-13 Quorn. (2011, August). “Fish without the fish-in”. Cook Vegetarian!, p. 81. 2011-14 Fry’s. (2011, August). [Advertisement focusing on Awards]. Cook Veget- arian!, p. 90. 1 reprint: Cook Vegetarian!, September 2011, p. 42. 2011-15 Dee’s. (2011, September). “Have you tried my DEElicious Organic Whole- food Burgers?”. Cook Vegetarian!, p. 46. 2011-16 Dee’s. (2011, October). “Easy to cook and taste great!”. Cook Vegetarian!, p. 107. 2011-17 Cauldron. (2011, October). “Takea pack of. . . tofu mince”. Cook Vegetarian!, pp. 70–71. 2011-18 Cauldron. (2011, October). “Cauldron — Organic Tofu Mince”. Cook Veget- arian!, p. 72. 1 reprint: Cook Vegetarian!, January 2012, p. 97. 2011-19 Dee’s. (2011, November). “Easy to cook and taste great!”. Cook Vegetarian!, p. 107. 2011-20 Fry’s. (2011, November). [Advertisement for their range]. Cook Vegetarian!, p. 72. 2011-21 Goodlife. (2011, December). “Goodlife — Naturally delicious, naturally meat & dairy free & now fairtrade!”. Cook Vegetarian!, p. 108. 2011-22 Fry’s. (2011, December). “New products being launched in January 2012. . .”. Cook Vegetarian!, p. 47. 2012-01 Goodlife. (2012, January). “Goodlife . . .New & naturally delicious”. Cook Vegetarian!, p. 108. 1 reprint: Cook Vegetarian!, April 2012, p. 72. 2012-02 Fry’s. (2012, January). “Fry’s launch 4 new products in January 2012.”. Cook Vegetarian!, p. 14. 2012-03 Redwood. (2012, January). “Delicious Vegan Food From Our Vegideli Range”. Cook Vegetarian!, p. 2. 2 reprints: Cook Vegetarian!, February 2012, p. 2.; Cook Vegetarian!, March 2012, p. 2. 2012-04 Fry’s. (2012, February). “FRY’s launch 4 new products in January 2012.”. Cook Vegetarian!, p. 56. 2012-05 Fry’s. (2012, March). “NEW exciting additions to our frozen food range. . .”. Cook Vegetarian!, p. 80. 5 reprints: The Vegetarian, Spring 2012, p. 17.; Cook Vegetarian!, June 2012,

298 B.5. Bibilographic Details of the Collected Advertisements

p. 44.; The Vegetarian, Summer 2012, p. 40.; Cook Vegetarian!, September 2012, p. 78.; Cook Vegetarian!, October 2012, p. 84. 2012-06 Redwood. (2012, April). “why not have a bite of our delicious vegan food?”. Cook Vegetarian!, p. 2. 2012-07 Fry’s. (2012, April). “NEW exciting additions to our frozen range. . .”. Cook Vegetarian!, p. 99. 1 reprint: Cook Vegetarian!, May 2012, p. 66. 2012-08 Redwood. (2012, May). “All the taste of the ocean but with none of the fish!”. Cook Vegetarian!, p. 2. 2012-09 Goodlife. (2012, June). “70 Days of Summer”. Cook Vegetarian!, p. 100. 2012-10 Redwood. (2012, June). “All the taste of the ocean but with none of the fish!”. Cook Vegetarian!, p. 2. 2012-11 Redwood. (2012, July). “Menu Ideas From Redwood”. Cook Vegetarian!, p. 2. 2012-12 Fry’s. (2012, July). “Perfect for barbecues. . .”. Cook Vegetarian!, p. 56. 1 reprint: Cook Vegetarian!, August 2012, p. 7. 2012-13 Redwood. (2012, August). “Menu Ideas From Redwood”. Cook Vegetarian!, p. 2. 2012-14 Redwood. (2012, September). “Helping you make delicious vegan meals — free from meat, egg & dairy”. Cook Vegetarian!, p. 2. 2012-15 Redwood. (2012, October). “Gourmet Foods — for discerning vegetarians and vegans”. Cook Vegetarian!, p. 2. 2012-16 Sheese. (2012, October). “New and Exciting! — And now. . . Deli Fry’s”. Cook Vegetarian!, p. 76. 3 reprints: Cook Vegetarian!, November 2012, p. 78.; Cook Vegetarian!, February 2013, p. 61.; Cook Vegetarian!, March 2013, p. 17. 2012-17 Redwood. (2012, November). “Our New Festive Vegetarian and Vegan Range”. Cook Vegetarian!, p. 2. 2012-18 Redwood. (2012, December). “Time for our Seasonal Roasts”. Cook Veget- arian!, p. 2. 1 reprint: Cook Vegetarian!, December 2012, p. 35. 2012-19 Tesco. (2012, December). “Win! Your Christmas food shopping. . . on us!”. Cook Vegetarian!, p. 34. 2012-20 Fry’s. (2012, December). “Winter Warmers from. . .”. Cook Vegetarian!, p. 99. 2013-01 Redwood. (2013, January). “Pick up a Pasty”. Cook Vegetarian!, p. 2. 2013-02 Fry’s. (2013, January). “Winter Warmers from. . . Fry’s”. Cook Vegetarian!, p. 99. 1 reprint: Cook Vegetarian!, February 2013, p. 99.

299 B. Further Details of the Advertisements

2013-03 Redwood. (2013, February). “Treat your heart this valentine with Redwood Foods”. Cook Vegetarian!, p. 2. 2013-04 GranoVita. (2013, March). “Tasty, Dairy Free Treats”. Cook Vegetarian!, p. 11. 2013-05 Redwood. (2013, March). “It’s Party Time at The Redwood Co”. Cook Vegetarian!, p. 2. 1 reprint: Cook Vegetarian!, March 2013, p. 2. 2013-06 Fry’s. (2013, Spring). “Spice Up Your Meal Times”. The Vegetarian, p. 42. 2013-07 Redwood. (2013, April). “Enjoy the taste of SPRING with Redwood Foods”. Cook Vegetarian!, p. 2. 2 reprints: Cook Vegetarian!, May 2013, p. 2.; Cook Vegetarian!, May 2013, p. 2. 2013-08 GranoVita. (2013, May). “Veggie Guide — Essential Pantry Products”. Cook Vegetarian!, p. 34. 2013-09 Secret Sausages. (2013, May). “Secret Sausages”. Cook Vegetarian!, p. 78. 1 reprint: Cook Vegetarian!, June 2013, p. 96. 2013-10 Fry’s. (2013, May). “NEW spicy addition to our range. . .”. Cook Vegetarian!, p. 99. 1 reprint: Cook Vegetarian!, June 2013, p. 99. 2013-11 Fry’s. (2013, June). “Brilliant Barbecues”. Cook Vegetarian!, p. 15. 2013-12 Vegusto. (2013, June). “World class Swiss dairy-free cheeses and meat alternatives”. Cook Vegetarian!, p. 16. 4 reprints: Cook Vegetarian!, June 2013, p. 52.; Cook Vegetarian!, August 2013, p. 44.; Cook Vegetarian!, September 2013, p. 52.; Cook Vegetarian!, October 2013, p. 46. 2013-13 Redwood. (2013, June). “Start your Summer with a BBQ from Redwood Foods”. Cook Vegetarian!, p. 2. 2013-14 VBites. (2013, June). “The Redwood Co are changing their name to VBites”. Cook Vegetarian!, p. 2. 1 reprint: Cook Vegetarian!, July 2013, p. 2. 2013-15 Fry’s. (2013, Summer). “PERFECT for barbecues. . .”. The Vegetarian, p. 42. 2013-16 Vegusto. (2013, July). “Exciting news!”. Cook Vegetarian!, p. 14. 4 reprints: Cook Vegetarian!, November 2013, p. 52.; Cook Vegetarian!, January 2014, p. 76.; Cook Vegetarian!, February 2014, p. 42.; Cook Vegetarian!, March 2014, p. 82. 2013-17 VBites. (2013, July). “VBites Festive Range”. Cook Vegetarian!, p. 2. 1 reprint: Cook Vegetarian!, December 2013, p. 2. 2013-18 Sheese. (2013, July). “Vegandelimeat-free continental slicing sausages”. Cook Vegetarian!, p. 68. 2013-19 Fry’s. (2013, July). “The taste of summer. . .”. Cook Vegetarian!, p. 99.

300 B.5. Bibilographic Details of the Collected Advertisements

2 reprints: Cook Vegetarian!, August 2013, p. 99.; Cook Vegetarian!, September 2013, p. 99. 2013-20 VBites. (2013, August). “Back to school lunches”. Cook Vegetarian!, p. 2. 2013-21 Secret Sausages. (2013, September). “Secret Sausages”. Cook Vegetarian!, p. 52. 2013-22 VBites. (2013, October). “Great Recipes from VBites”. Cook Vegetarian!, p. 2. 2013-23 Fry’s. (2013, October). “Spice up mealtimes this autumn”. Cook Vegetarian!, p. 99. 2013-24 Secret Sausages. (2013, November). “Secret Sausages”. Cook Vegetarian!, p. 7. 2 reprints: Cook Vegetarian!, March 2014, p. 7.; Cook Vegetarian!, April 2014, p. 7. 2013-25 Secret Sausages. (2013, November). “Food’s best-kept secret”. Cook Veget- arian!, p. 84. 2013-26 Fry’s. (2013, November). “Whether you’re eating in or out”. Cook Veget- arian!, p. 99. 2013-27 Fry’s. (2013, December). “Have a very Veggie Xmas”. Cook Vegetarian!, p. 91. 2013-28 Fry’s. (2013, December). “A world of ideas for festive food and party treats”. Cook Vegetarian!, p. 99. 2014-01 Fry’s. (2014, January). “Easy meat free cooking with Fry’s”. Cook Vegetarian!, p. 62. 1 reprint: Cook Vegetarian!, February 2014, p. 46. 2014-02 Secret Sausages. (2014, January). “Secret Sausages”. Cook Vegetarian!, p. 9. 2014-03 GranoVita. (2014, February). “Why not try some of our other delicious meat free products. . .”. Cook Vegetarian!, p. 14. 2014-04 VBites. (2014, February). “Healthy New You”. Cook Vegetarian!, p. 2. 2014-05 VBites. (2014, March). “Healthy New You”. Cook Vegetarian!, p. 2. 2014-06 Fry’s. (2014, March). “Easy meat free cooking with Fry’s”. Cook Vegetarian!, p. 99. 2014-07 Quorn. (2014, Spring). “Make Quorn part of your programme”. The Vegetarian, p. 15. 2014-08 Sheese. (2014, April). “100% Vegan — Sheese”. Cook Vegetarian!, p. 34. 2 reprints: Cook Vegetarian!, June 2014, p. 19.; Cook Vegetarian!, July 2014, p. 19. 2014-09 Secret Sausages. (2014, April). “5 Ways to Boost your Health Today!”. Cook Vegetarian!, p. 49. 2014-10 Fry’s. (2014, April). “A World of Inspiration from Fry’s”. Cook Vegetarian!, p. 64. 1 reprint: Cook Vegetarian!, May 2014, p. 72.

301 B. Further Details of the Advertisements

2014-11 VBites. (2014, May). “Just Launched — 3 new Products”. Cook Vegetarian!, p. 2. 2014-12 Fry’s. (2014, June). “Ideal for picnics, lunchboxes and outdoor eating”. Cook Vegetarian!, p. 100. 2014-13 VBites. (2014, June). “Making the perfect Sandwich”. Cook Vegetarian!, p. 2. 2014-14 VBites. (2014, June). “Making the perfect Sandwich”. Cook Vegetarian!, p. 2. 2014-15 Secret Sausages. (2014, June). “Secret Sausages”. Cook Vegetarian!, p. 29. 2014-16 VBites. (2014, June). “Rise and Shine”. Cook Vegetarian!, p. 82. 2014-17 Secret Sausages. (2014, Summer). “Secret Sausages”. The Vegetarian, p. 15. 2014-18 VBites. (2014, July). “Life with VBites is always a Picnic!”. Cook Vegetarian!, p. 2. 1 reprint: Cook Vegetarian!, August 2014, p. 2. 2014-19 VBites. (2014, July). “Best of British Barbecue”. Cook Vegetarian!, p. 2. 2014-20 Sheese. (2014, July). [Advertisement for their range]. Cook Vegetarian!, p. 45. 2014-21 Fry’s. (2014, July). “Ideal for picnics, lunchboxes & outdoor eating”. Cook Vegetarian!, p. 50. 2 reprints: Cook Vegetarian!, August 2014, p. 72.; Cook Vegetarian!, October 2014, p. 79. 2014-22 Secret Sausages. (2014, July). “Secret Sausages”. Cook Vegetarian!, p. 8. 2014-23 VBites. (2014, October). “Back to School”. Cook Vegetarian!, p. 2. 2014-24 VBites. (2014, November). “VBites Festive Season — New Festive Range 2014”. Cook Vegetarian!, p. 2. 4 reprints: Cook Vegetarian!, November 2014, p. 2.; Cook Vegetarian!, December 2014, p. 2.; Cook Vegetarian!, December 2014, p. 2.; Cook Vegetarian!, January 2015, p. 2. 2014-25 just wholefoods. (2014, November). “Everyday food, fit for every occasion. . .”. Cook Vegetarian!, p. 80. 7 reprints: Cook Vegetarian!, December 2014, p. 70.; Cook Vegetarian!, January 2015, p. 76.; Cook Vegetarian!, April 2015, p. 97.; Vegan Life, May 2015, p. 37.; Veggie, October 2015, p. 97.; Veggie, June 2016, p. 46.; Veggie, December 2016, p. 68. 2014-26 meat the alternative. (2014, December). “A great new range of tasty meat free products”. Cook Vegetarian!, p. 100. 2 reprints: Cook Vegetarian!, January 2015, p. 100.; Cook Vegetarian!, February 2015, p. 100. 2014-27 Goodlife. (2014, December). “We’ve got a Vegtastic New Range!”. Cook Vegetarian!, p. 19. 1 reprint: Cook Vegetarian!, February 2015, p. 11.

302 B.5. Bibilographic Details of the Collected Advertisements

2014-28 Vegusto. (2014, Winter). “Festive Roasts for Luxury Christmas Dining”. The Vegetarian, p. 63. 2015-01 Wicken Fen. (2015, January). “Delicious Sausages and Falafel”. Cook Veget- arian!, p. 49. 9 reprints: Cook Vegetarian!, February 2015, p. 71.; Cook Vegetarian!, June 2015, p. 86.; Veggie, July 2015, p. 54.; Veggie, July 2016, p. 48.; Veggie, August 2016, p. 86.; Veggie, November 2016, p. 61.; Veggie, December 2016, p. 49.; Veggie, May 2017, p. 68.; Veggie, August 2017, p. 88. 2015-02 Vegusto. (2015, January). “Festive Roasts”. Cook Vegetarian!, p. 72. 1 reprint: VeganLife, December 2015, p. 9. 2015-03 VegiNugs. (2015, January). “VegiNugs”. Cook Vegetarian!, p. 76. 2015-04 Fry’s. (2015, January). “Food Made Good”. Cook Vegetarian!, p. 99. 3 reprints: Cook Vegetarian!, April 2015, p. 100.; Cook Vegetarian!, May 2015, p. 46.; Cook Vegetarian!, June 2015, p. 100. 2015-05 VBites. (2015, February). “New Year, NEW YOU”. Cook Vegetarian!, p. 2. 2015-06 Goodlife. (2015, February). “Leave it to the Vegenistas!”. Cook Vegetarian!, p. 55. 2015-07 Vegusto. (2015, February). “New customers special offer”. Cook Vegetarian!, p. 65. 1 reprint: VeganLife, November 2015, p. 9. 2015-08 VBites. (2015, February). “5 Ways to Go Vegan”. Cook Vegetarian!, p. 67. 2015-09 VBites. (2015, March). “Roast Dinners are not just for Sundays.. they’re for Everyday!”. Cook Vegetarian!, p. 2. 2015-10 OMFG. (2015, March). “OMFG. . . — . . .these taste incredible”. Cook Veget- arian!, p. 42. 3 reprints: Cook Vegetarian!, April 2015, p. 51.; Cook Vegetarian!, May 2015, p. 40.; Cook Vegetarian!, June 2015, p. 86. 2015-11 Dee’s. (2015, May). “Winner of the Best Vegan Sausage in the UK!”. Vegan Life, p. 23. 3 reprints: VeganLife, June 2015, p. 21.; VeganLife, July/August2015, p. 9.; Vegan Life, September 2015, p. 9. 2015-12 VBites. (2015, June). “What will you find in the woods today?”. Cook Veget- arian!, p. 2. 2015-13 More than Meat. (2015, June). “The Best Burger Quest is over!”. VeganLife, p. 17. 2 reprints: Vegan Life, July/August 2015, p. 35.; Vegan Life, September 2015, p. 67. 2015-14 Fry’s. (2015, July). “Give your family 100% tasties with 0% nasties”. Veggie, p. 100.

303 B. Further Details of the Advertisements

3 reprints: Veggie, August 2015, p. 100.; Veggie, September 2015, p. 100.; Veggie, November 2015, p. 132. 2015-15 VBites. (2015, July). “Food for Sharing”. Veggie, p. 2. 2015-16 Fry’s. (2015, July). “Give your family 100% tasties with 0% nasties”. Veggie, p. 35. 1 reprint: Veggie, August 2015, p. 35. 2015-17 VBites. (2015, August). “what will you use in your sandwich?”. Veggie, p. 2. 2015-18 Goodlife. (2015, August). “Vegtastic BBQ Bangers”. Veggie, p. 32. 1 reprint: Veggie, September 2015, p. 17. 2015-19 More than Meat. (2015, August). “The Best Burger Quest is over!”. Veggie, p. 47. 1 reprint: Veggie, November 2015, p. 68. 2015-20 Fry’s. (2015, August). “Ultimate fuss-free veggie barbecue”. Veggie, p. 61. 2015-21 VBites. (2015, September). “Try some of our Top Selling Products”. Veggie, p. 2. 2015-22 VBites. (2015, September). “Coming Soon!”. Veggie, p. 85. 2015-23 More than Meat. (2015, October). “Meet the new King of Mock ‘n’ Roll. . .”. VeganLife, p. 51. 2 reprints: VeganLife, November 2015, p. 49.; VeganLife, December 2015, p. 83. 2015-24 VBites. (2015, October). “You won’t believe what you can do on the BBQ!”. Veggie, p. 2. 2015-25 Great Food. (2015, October). “One fresh Brand”. Veggie, p. 40. 1 reprint: Veggie, November 2015, p. 40. 2015-26 Fry’s. (2015, October). “Welcoming New Members to the Tasty Team”. Veg- gie, p. 70. 2015-27 Great Food. (2015, October). “Braver with Flavour”. Veggie, p. 89. 2015-28 Fry’s. (2015, October). “Soy & Quinoa Country Roast”. Veggie, p. 99. 3 reprints: Veggie, November 2015, p. 108.; Veggie, December 2015, p. 53.; Veggie, Christmas 2015, p. 108. 2015-29 VBites. (2015, November). “Party Mood — Good Food”. Veggie, p. 2. 2015-30 Goodlife. (2015, November). “Live the Goodlife”. Veggie, p. 55. 2015-31 Goodlife. (2015, November). “Vegtastic!”. Veggie, p. 6. 2015-32 VBites. (2015, December). [Advertisement for snack foods]. Veggie, p. 2. 1 reprint: Veggie, Christmas 2015, p. 2. 2015-33 Goodlife. (2015, December). “We’ve got some Vegtastic New Products!”. Veggie, p. 6. 1 reprint: Veggie, May 2016, p. 35. 2015-34 Linda McCartney’s. (2015, Christmas). “Here’s the festive Solution”. Veggie, p. 53.

304 B.5. Bibilographic Details of the Collected Advertisements

2015-35 Vegusto. (2015, Christmas). “Festive Roests for Luxury Christmas Dining”. Veggie, p. 54. 2016-01 Fry’s. (2016, January). “Nature’s Plant Proteins”. Veggie, p. 108. 5 reprints: Veggie, February 2016, p. 24.; Veggie, March 2016, p. 108.; Veggie, April 2016, p. 108.; Veggie, May 2016, p. 14.; Veggie, September 2016, p. 76. 2016-02 Linda McCartney’s. (2016, January). “Linda McCartney’s — Deliciously Vegetarian”. Veggie, p. 11. 2016-03 VBites. (2016, January). “Spicy Sausage and Chorizo Style Casserole”. Veggie, p. 2. 2016-04 VBites. (2016, February). “Beat those January Blues and Spice Up your February with our Healthy Chilli Non Carne using VBites Meat-Free Mince”. Veggie, p. 2. 2016-05 Fry’s. (2016, March). [Advertisement for a Falafel Burger and other products]. VeganLife, p. 101. 2016-06 Linda McCartney’s. (2016, March). “Linda McCartney’s — Deliciously Vegetarian”. Veggie, p. 11. 1 reprint: Veggie, April 2016, p. 80. 2016-07 VBites. (2016, March). “Put a Spring into your Step and eat the Healthy Way!”. Veggie, p. 2. 2016-08 More than Meat. (2016, March). “Introducing our NEW More Than Jerk Burgers”. Veggie, p. 83. 2 reprints: Veggie, April 2016, p. 70.; Veggie, July 2016, p. 96. 2016-09 VBites. (2016, April). “Fish Style Fingers Sandwich”. Veggie, p. 2. 2016-10 Fry’s. (2016, May). [Advertisement for a Falafel Burger]. VeganLife, p. 77. 2016-11 Linda McCartney’s. (2016, May). “Linda McCartney’s — Deliciously Veget- arian”. Veggie, pp. 4–5. 2016-12 VBites. (2016, May). “Host the ultimate meat-free alfresco”. Veggie, p. 57. 2016-13 Goodlife. (2016, May). “Simple Summer Sensation”. Veggie, p. 80. 2016-14 Dee’s. (2016, June). “Winner of the best Vegan Sausage in the UK!”. Vegan Life, p. 79. 2016-15 VBites. (2016, June). “Producing Delicious, Tasty, Creative Alternatives to Meat, Fish, Dairy & Egg”. Veggie, p. 2. 2016-16 Fry’s. (2016, June). “Set your Plants on Fire!”. Veggie, p. 49. 2016-17 Fry’s. (2016, June). [Advertisement for barbecue food]. Veggie, p. 92. 3 reprints: Veggie, July 2016, p. 108.; Veggie, August 2016, p. 108.; Veggie, Septem- ber 2016, p. 108. 2016-18 Fry’s. (2016, July). “Plants on Fire”. VeganLife, p. 116. 2016-19 Dee’s. (2016, July). “Winner of the best Vegan Sausage in the UK!”. Vegan Life, p. 61.

305 B. Further Details of the Advertisements

2016-20 Gosh!. (2016, July). “Nature’s New Delivery!”. VeganLife, p. 61. 2016-21 Linda McCartney’s. (2016, July). “Linda McCartney’s — celebrating 25 Years of a Food Pioneer”. Veggie, p. 11. 1 reprint: VeganLife, September 2016, p. 7. 2016-22 VBites. (2016, July). “Happiness is . . .an absurdly delicious vegan sandwich”. Veggie, p. 2. 2016-23 Linda McCartney’s. (2016, August). “Linda McCartney’s — Deliciously Vegetarian”. Veggie, p. 11. 1 reprint: Veggie, January 2017, p. 27. 2016-24 VBites. (2016, August). “Please join us for a Barbecue”. Veggie, p. 2. 1 reprint: Veggie, September 2016, p. 72. 2016-25 More than Meat. (2016, August). “More Than Jerk Burgers”. Veggie, p. 52. 2016-26 Gosh!. (2016, September). “Nature’s New Delivery!”. VeganLife, p. 93. 2016-27 Fry’s. (2016, October). [Advertisement for a Falafel Burger]. Veggie, p. 108. 2 reprints: Veggie, November 2016, p. 108.; Veggie, March 2017, p. 10. 2016-28 VBites. (2016, October). “Back to School”. Veggie, p. 2. 2016-29 Linda McCartney’s. (2016, December). “Linda McCartney’s — celebrating 25 Years of a Food Pioneer”. VeganLife, p. 116. 2016-30 Fry’s. (2016, December). “A festive meat-free Feast”. Veggie, p. 108. 2016-31 Fry’s. (2016, December). “Get set for a tasty vegan Christmas”. Veggie, p. 124. 1 reprint: VeganLife, December 2016, p. 69. 2016-32 Linda McCartney’s. (2016, December). “Linda McCartney’s — celebrating 25 Years of a Food Pioneer”. Veggie, p. 6. 2017-01 Linda McCartney’s. (2017, January). “Linda McCartney’s — Deliciously Vegetarian”. VeganLife, p. 116. 2 reprints: VeganLife, March 2017, p. 35.; VeganLife, May 2017, p. 63. 2017-02 Gosh!. (2017, January). “Good Food Swoops in!”. VeganLife, p. 31. 2017-03 Fry’s. (2017, January). [Advertisement for Soy & Flaxseed Schnitzels]. Veg- gie, p. 100. 1 reprint: Veggie, February 2017, p. 100. 2017-04 VBites. (2017, January). “10,000 People are trying Vegan this January — Dare You?”. Veggie, p. 2. 2017-05 VBites. (2017, February). “4 new products for 2017”. Veggie, p. 2. 2017-06 Gosh!. (2017, March). “Gosh! I’m Good. (Try me)”. Veggie, p. 35. 2017-07 Linda McCartney’s. (2017, April). “Linda McCartney’s — Deliciously Ve- getarian”. VeganLife, p. 77. 2017-08 VBites. (2017, April). “Chicken Style Chowder”. Veggie, p. 2. 2017-09 Gosh!. (2017, April). “Gosh! I’m Good. (Try me)”. Veggie, p. 50. 2017-10 VBites. (2017, May). “Chilli Non Carne”. Veggie, p. 2.

306 B.6. Coding Schedule

2017-11 Linda McCartney’s. (2017, May). “Linda McCartney’s — Deliciously Veget- arian”. Veggie, p. 23. 2017-12 Linda McCartney’s. (2017, June). “Linda McCartney’s — Deliciously Veget- arian”. VeganLife, p. 43. 5 reprints: Veggie, July 2017, p. 82.; VeganLife, August 2017, p. 27.; Veggie, Au- gust 2017, p. 31.; Veggie, September 2017, p. 100.; Veggie, October 2017, p. 39. 2017-13 Moodley Manor. (2017, June). “Barbecue with Boss Burgers this Summer”. VeganLife, p. 49. 2017-14 Fry’s. (2017, July). [Advertisement for a Burger]. Veggie, p. 100. 4 reprints: Veggie, August 2017, p. 100.; Vegan Life, September 2017, p. 37.; Veggie, September 2017, p. 60.; Veggie, October 2017, p. 100. 2017-15 VBites. (2017, July). “Backyard BBQ — Celebrate Summer with Awesome Food!”. Veggie, p. 2. 2017-16 VBites. (2017, July). “Lemon & Black Pepper Scampi Style Bites — . . . just great tasting food”. Veggie, p. 88. 2017-17 Fry’s. (2017, August). [Advertisement for barbecue food]. Veggie, p. 12. 2017-18 Gosh!. (2017, September). “My Gosh! A new take on a classic”. VeganLife, p. 69. 2017-19 VBites. (2017, September). “Make it a fish free Friday!”. Veggie, p. 2. 2017-20 Linda McCartney’s. (2017, October). “Linda McCartney’s — Deliciously Vegetarian”. VeganLife, p. 57. 2017-21 VBites. (2017, October). “Happy Halloween”. Veggie, p. 2.

B.6 Coding Schedule

General instructions: Logos and parts of adverts clearly dealing with products which are not meat alternatives are ignored for the coding, as well as product packaging, recipes and “can be bought in” or “available in.” The following metadata was assessed for each advert: brand, magazine, year, month, page, issue number (if available), size, title/headline. The following guidelines for coding were used. The first five coding categories further show the aggregation used for quantitative analysis; this was only done after the coding due to potential analytical insights and practical constraints of group sizes for the χ2-test.

Label or headline of the advertisement. Up to two different labels which define the advertised product(s), judged by the main texts (by size, prominence, flow of reading) found in the advert. (Qa1)

307 B. Further Details of the Advertisements

x-free : product labelled or signposted as meat-free, fish-free, etc alternative : product labelled or signposted as meat alternative, or alternative to meat substitute : product labelled or signposted as meat substitute x-style : product labelled or signposted as chicken-style, fish-style, etc ingredients : product labelled or signposted by its ingredients, such as crunchy vegetables, “soy and quinoa country roast” (2015-28) plant-based : product labelled or signposted as ‘plant-based’ protein : product labelled or signposted as a protein product, such as “nature’s plant proteins” (2016-01) dietary : product labelled or signposted as vegetarian or vegan purposive : product labelled or signposted as purposive, situational, or contextual, such as food for celebrations, summer, or barbecues shape : product labelled or signposted as sausage, burger, roast, etc category : product labelled or signposted as a food category, such as ‘chilled sliced deli range’ temperature : product labelled or signposted by its storage temperature, such as frozen or chilled generic : product labelled or signposted as generic ‘food’, products, etc fish : product labelled or signposted as fish which is not necessarily marked as a fish alternative brand : product not labelled but instead uses a prominent display of the brand name visual : product not labelled but instead there is a depiction of the advertised or a product Description of the product. Up to four different characterisations, which may be part of the product label already. Consider those most prominent (size, position, colour). (Qa1, Qb1) taste : product described by its taste or flavour, such as tasty or delicious texture : product described with reference to its texture, such as juicy or succulent appearance : mentions the appearance or the looks of the product ingredients : lists ingredients as further description convenience : mentions how convenient, easy, or simple the product is naturalness : suggests that the product is ‘natural’ or free from additives shape : describes the product by mentioning its shape, such as burger, or sausage exclusion : mentions what a product is not, or what is not in the product, such as ‘low cholesterol’, or ‘no fat’

308 B.6. Coding Schedule

quality : describes the product as high quality, possibly in comparison, such as ‘best ever’ vegetarian/vegan : product is described as vegan or vegetarian tradition : product can be used in traditional contexts authenticity : the product or its raw materials are linked to a person and/or their efforts, such as ‘local farmers’ health : product is described as healthy or related to good health temperature : the storage temperature of the product is mentioned, such as frozen, or chilled sustainability : the product is described as sustainability enhancing, sustainable, or generally good for environment or planet animal welfare : the product is mentioned in context of animals which may be saved or rescued through the product protein : protein is mentioned in relation to the product, such as ‘high in protein’ or ‘high-quality protein’ price : price is mentioned regional : the product’s origin is described concretely, such as manufactured in Britain; undefined claims such as ‘supported by local farmers’ would be authenticity visual : description is supported by a prominent picture with something meat like Product distinction or comparison. The distinction and thus evaluative setting drawn on or evoked in the advert. (max. 2) (Qa2) like meat : product is suggested to be like meat different from meat : product is suggested to be different from meat alternative to meat : product is presented as an alternative to meat or a substitute; this tends to be neither similar nor dissimilar, but may also be both like vegetables : product is suggested to be like vegetables different from vegetables : product is suggested to be different from vegetables and unprocessed plant-based foods different from vegetarian food : product is promoted as being different from vegetarian food (not meat alternatives) different from meat alternatives : product is suggested to be different from other meat alternatives different from competition : product is suggested to be different from the competition or in its category, such as ‘best vegan burger’ better than before : product is suggested to be better than before, having an improved recipe or being different from how products used to be

309 B. Further Details of the Advertisements

food culture : product is different from other food products around, for example by allowing a unique allergen-free or wholefoods experience absolute : the advertisement makes value statements about the product, but this is presented to be absolute, such as “these taste incredible” (2015-10), or “the best burger quest is over” (2015-19) none : if the adverts does not make any value statements and draws no comparisons Presented novelty or achievement. What is the novelty or achievement presented that makes the product relevant for the reader? (Qa3) helping animals : if animals can be rescued by switching to different products social acceptance : if the social acceptance of the own diet (or of vegetarian products for others) can be increased, such as “how to please everybody” (2000- 01), or “now . . . veggies can consider themselves well catered for” (2009-14) societal change : if societal change can be induced or supported through buying this product, such as fair trade products, the saving of ecosystems, or a change in the food system health : if there is a promise or suggestions consumers might become healthier through eating the product or through replacing other foodstuffs with the product naturalness : if the naturalness of the product is presented as unprecedented or unusual in the market taste : if a novel recipe or taste is suggested to be a major achievement or change for the consumers online shop : if a new online shop is introduced or referred to new product : if a product is advertised as being new none : nothing appears like an achievement Theme of the advert. The dominant theme of the advert which suggests meat alternatives as a problem or solution, or motivation to eating meat alternatives. The main context in which the advert/product suggests to be evaluated in. (Qa3, used as indicative of the ‘dominant meaning’ to illustrate Qc1) food safety : a concern for food safety of other products, usually meat health : a concern for general health of the audiences taste : focused on taste of the product convenience : focused on convenience of the product naturalness : focus on the ‘naturalness’ of the product animals : focus on animals, which may for example be saved or rescued by switching to the advertised product self : focused on the individual audience, such as concerning dieting, mindful- ness or as ‘good for you’

310 B.6. Coding Schedule

family : focus on the family and family gatherings or shared meals of the audience friends : focused on the friends or partners of the audience school : a product for school lunches, usually as part of ‘back to school’ advertisements everyday : focus on everyday things like breakfast bbq : focus on barbecues picnic : focus on picnics xmas : focus on christmas and the festive period celebration : focus on other celebrations or general parties other seasonal : other seasonal focus, e.g. summer, autumn, spring (for winter, see xmas) recipe : focus on a recipe using the advertised product award : award won by product or brand dietary : focus on the dietary credentials of the product, such as whether it is vegan or vegetarian online shop : advertised product in context of an online shop global stewardship : taking care of other people, e.g. through buying fair trade products or paying good prices to farmers offer : product contextualised as part of an offer dining : product proposed as suitable or replacement for fine dining other : other contexts not mentioned here none : no obvious context

Product evaluation. What is the main evaluative reference point? (Qb2) meat : product should be evaluated in comparison with meatstuffs format : product should be evaluated in reference to other burgers, sausages, etc vegetables : product should be evaluated as a vegetable-based or a wholefood product generic food : product should be evaluated as a food and by its goodness, but not in reference to other foods vegan food : product should be evaluated in reference to vegan foods contextual : product should be evaluated in a specific context or with a specific purpose in mind brand : product should be understood as part of a brand, not as a product in itself.

311 B. Further Details of the Advertisements

Categorisation put forward in the advertisement. What is the category that the foodstuff suggests? (Qb3) product format : food should be understood as a burger, sausage, roast, or similar food for a specific diet : food is ‘vegetarian’ or ‘vegan’ purposive food : food is meant for a specific purpose or specific occasions meat-free : food is free from meat or animals protein food : food should be understood as constituted by ‘protein’ generic foodstuff : food should not be understood in any of these categories, but simply as food

Frame for meat. Context in which meat is part of the advert for meat alternatives. (Qc2) tradition : food can revive past experiences made with meat; food that has ‘proven’ to be good or suitable in certain contexts; food which is used in a recipe or a dish that is contextualised as being based on meat taste : food which is tasty or enjoyable because it is or similar to meat unlike meat : implies meat is something harmful or dangerous, and is different from meat visual : only a visual reference is made none : no reference to meat can be identified Relation of the advertised product to meat. If there is a frame for meat (see above), how is the relation between this product and meat? (Qc2) similarly established : proposes that the product is equivalent to meat, and can be considered as established and normal as meat alternative to meat : suggests the advertised product as an alternative to meat, as it is not quite the same but similar enough to be easily replaced or similarly successful reinvented meat : highlights that the advertised product is substantially different from meat none : no relation to meat explicit

B.7 Statistical Evaluations

In the following, there are the tables with statistical evaluations for the analysis of emergence as covered in sections 5.2 and 9.2. All tests compare the adverts coded with those not coded with a specific code. To make the amount of data a bit more

312 B.7. Statistical Evaluations homogeneous, the time periods for 1985–1993 and 1994–2001 were combined, resulting in df = 3 for all analyses. As elaborated in Appendix B.4, the assumption for the expected-values in the utilised χ2-test suggests that 31 or more codes are required for a correct approximation; this makes statistical results with N ≪ 31 likely invalid and reduces the reliability of the statistical test for codes with N < 31. The tables show the collapsed categories as well as the codes included in it; sums do not match when one advert included multiple codes of one category.

Table B.2: Results of the χ2-tests for the labels/headlines of advertisements (Qa1).

Labelled as n p χ2

purpose 128 0.0000000 41.199472 others 124 0.0000001 36.063105 – generic 97 0.0009381 16.401369 – temperature 33 0.0014811 15.433691 format 164 0.0000003 33.450869 – shape 151 0.0000041 27.757672 – category 14 0.1208617 5.816988 dietary 125 0.0000020 29.187500 meaty 159 0.0000068 26.710862 – [e.g. meat]-free 98 0.0002760 18.980404 – alternative 33 0.3677213 3.159225 – [e.g. beef]-style 24 0.3427384 3.335455 – substitute 19 0.0008088 16.715068 – fish 13 0.0374899 8.454838 brand 56 0.0016551 15.197972 visual 40 0.0044299 13.097755 contents 81 0.0131884 10.744897 – ingredients 59 0.3585177 3.222905 – protein 11 0.0000000 57.464517 – plant-based 11 0.0012644 15.769325

Table B.3: Results of the χ2-tests for the descriptions of advertisements (Qa1).

Description n p χ2

focus on the format 295 0.0000000 37.8164744 – shape 208 0.0000000 41.4482893 – visual 128 0.0000045 27.5790760 – appearance 17 0.0000787 21.6085491

313 B. Further Details of the Advertisements

Description n p χ2

ingredients and health 356 0.0000141 25.1846413 – exclusion 162 0.0737097 6.9438324 – naturalness 149 0.0000102 25.8637900 – ingredients 110 0.6104447 1.8206819 – health 82 0.9458917 0.3723574 – protein 44 0.0000000 103.3910738 – quality 34 0.0599234 7.4097449 – plant-based 11 0.0012644 15.7693254 use and function 170 0.0002887 18.8855054 – convenience 88 0.0099761 11.3500487 – temperature 53 0.0000001 36.7540754 – tradition 34 0.4159598 2.8461454 – price 12 0.0021911 14.6013338 vegetarian / vegan food 232 0.0705406 7.0429734 worldly context 61 0.1553635 5.2351909 – sustainability 27 0.0292785 9.0009327 – animal welfare 19 0.0152439 10.4299245 – authenticity 17 0.3813145 3.0676334 – regional 6 0.0001548 20.1924107 focus on sensory aspects 256 0.1988466 4.6553308 – taste 255 0.1977633 4.6682683 – texture 9 0.2063887 4.5670302

Table B.4: Results of the χ2-tests for the comparisons drawn in the advertisements (Qa2).

Distinction n p χ2

meat alternatives 70 0.0000001 34.9786793 – different from competition 41 0.0234546 9.4884427 – different from meat alternatives 29 0.0000040 27.8073809 absolute 154 0.0000082 26.3225039 vegetarian food and vegetables 45 0.0113490 11.0708226 – different from vegetarian food 39 0.0426792 8.1672238 – different from vegetables 3 0.0031632 13.8183142 – like vegetables 3 0.7303707 1.2947895 others 45 0.2532533 4.0771468 – better than before 25 0.0791778 6.7820772

314 B.7. Statistical Evaluations

Distinction n p χ2

– food culture 20 0.9403608 0.3994271 meat 137 0.6745468 1.5335595 – alternative to meat 57 0.0762944 6.8659792 – like meat 51 0.6279803 1.7404422 – different from meat 38 0.2030751 4.6054491 none 106 0.7223651 1.3285497

Table B.5: Results of the χ2-tests for the themes of the advertisements (Qa3).

Theme n p χ2

characteristics 173 0.0000011 30.4421462 – taste 94 0.0014816 15.4329889 – naturalness 42 0.0010546 16.1536012 – health 28 0.8105500 0.9615687 – dietary 9 0.0254830 9.3063709 people 28 0.0000182 24.6632837 – self 15 0.0028734 14.0234582 – family 8 0.0003248 18.6380294 – friends 5 0.6161867 1.7942515 special occasions 70 0.0000376 23.1476986 – bbq 30 0.0000001 34.6713411 – xmas 16 0.2588442 4.0243740 – celebration 15 0.2075399 4.5538172 – picnic 7 0.0047356 12.9546839 – dining 2 0.8800921 0.6706181 values 46 0.0064079 12.3050887 – animals 26 0.0310162 8.8738356 – food safety 11 0.0305094 8.9101697 – global stewardship 9 0.0031843 13.8041280 regular use 90 0.0114793 11.0460790 – other seasonal 30 0.0353950 8.5821253 – recipe 21 0.3172466 3.5272377 – convenience 18 0.1363594 5.5384066 – everyday 18 0.3683827 3.1547016 – school 3 0.4193370 2.8253921 commercial 125 0.0157679 10.3563241 – new products 82 0.3533746 3.2591088

315 B. Further Details of the Advertisements

Theme n p χ2

– award 21 0.0226718 9.5628786 – onlineshop 13 0.0132370 10.7369152 – offer 9 0.1855081 4.8194167 other 25 0.2322591 4.2853020 – other 17 0.5245280 2.2378686 – none 8 0.0065802 12.2479857

Table B.6: Results of the χ2-tests for the novelty with which products are advertised (Qa3).

Reported Novelty n p χ2

others 113 0.0000302 23.61 – taste 33 0.3178231 3.52 – societal change 32 0.0164077 10.27 – social acceptance 18 0.0000214 24.32 – online shop 9 0.0090037 11.57 – naturalness 8 0.4796192 2.48 – health 7 0.1122905 5.99 – helping animals 6 0.0000045 27.57 none 279 0.0001654 20.05 new product 165 0.5259835 2.23

316 Appendix C

Further Details of the Interviews

C.1 Detailed Overview of the Sampling Procedure

Initially, I intended to speak to a larger number of employees of a few key companies in the market; however, experiencing low response rates and willingness to participate, I extended my sample. I particularly contacted UK-based or UK-selling manufacturers of meat alternatives and a number of food-related NGOs. These companies were identified on the basis of the investigations of listed brands in UK supermarkets (see section 4.1), as well as contacts they suggested to speak to. To get a broad yet targeted and purposeful sample of interviewees, four types of target organisations were approached:

a. meat alternatives manufacturers based and selling in the UK; b. meat alternatives manufacturers selling in the UK; c. UK consultancies specialising in plant-based foods; d. UK civil society organisations relating to reducing meat and/or promoting alternatives.

In total, I approached 29 manufacturers and 10 other organisations for interviews. Of the other attempts, three manufacturers offered to reply to some questions in writing but never did so, six declined, with four I never managed to arrange an interview despite initial contact, and six never replied or were not available. From the remaining attempts, two organisations never replied, one declined, and for one I could not manage to arrange an interview despite initial contact; the latter, just as before, was due to initial contacts forwarding or involving third people who I could never get hold of. Appropriate people in each organisation were identified via the web, for example on LinkedIn, Companies House, or the companies’ websites; email addresses were reverse engineered from their names and the company domains. For smaller companies I generally used the contact form or an email address provided on the website. After three unanswered emails, I attempted to reach companies via phone; unless they had agreed to an interview by that time, no further attempts to contact organisations were made. While in some cases I spoke to the person I had approached, some organisa-

317 C. Further Details of the Interviews tions willing to participate suggested an interview partner related to the focus of my questions. Thus, while still being very diverse, the sample represents a convenience sample of respondents willing to participate, possibly also because no compensation was offered to the participants. All interviewees were provided with a participant information sheet approved by the Ethics Review Panel of the Manchester Business School in advance and were required to give their consent via an approved consent form. The interviews were conducted between November 2017 and May 2018 via voice calls on the phone, via video chats on the internet, or in two cases in person in a cafe. All interviews were recorded, transcribed verbatim, and sent to the interviewees to give them a chance for further remarks, clarifications, or to exclude or rephrase passages if required. Of the 14 interviews conducted and transcribed, three are not included in this work as one person was worried about their reputation as an expert due to my verbatim transcription; a second interview was too bad in recording and connection quality, and we never got around to conducting a second one; for a third interview, the interviewee had moved their organisation by the time I had sent the transcript, and due to reasons of data protection I was not given any new contact details. All interviews were coded using the software RQDA (Huang, 2016). The data is limited strongly by its arguably limited coverage of different brands; particularly so, as Quorn is the only major brand on the UK market which is represented it the sample (for example of those mostly listed in UK supermarkets, see again Figure 4.1). While the two international brands Vivera and Tofurkyhave strong global representation, their standing on the UK market is yet limited. However, the sample represents five brands which are available nationwide, two still young companies, and one whose production has been discontinued. Although the sample could have been bigger, this chapter and the research project at large aim to investigate diversity rather than to provide an accurate representation every single actor. Therefore, these shortcomings should not impact the significance of the results too much as a diversity of different views is covered amongst the wide range of interviewees.

C.2 Overview of Interviewees

Table C.1: Overview of the interviewees who kindly gave their time to participate in this study.

Organisation Name1 Position Duration

Quorn Louise Sustainability Manager 0:32 Quorn Eddy Insight Manager 0:51

318 C.3. Interview Guide

Organisation Name Position Duration

Vivera Geert-Jan Managing Director 0:41 Meat the Alternative Tony CEO and Founder 2:02 Vegetables in Disguise Rachel CEO and Founder 0:53 Tofurky Seth Founder 1:01 Sgaia Foods Hilary Co-Founder and CEO 0:33 LoveSeitan Steve Co-Founder and CEO 1:10 More than Meat Daniel Founder 0:42 eating better alliance Sue Executive Director 0:36 PETA Dawn Director of Vegan Corporate Projects 0:16

C.3 Interview Guide

On the following pages, the interview guide used for the semi-structured interviews with manufacturers is printed. The first level bulled points represent the trigger question, whereas indented questions and notes are for potential follow up questions in case the interviewees did not cover a particular topic.

Introduction

• Is it okay to record the interview? • So I’d like to do a semi-structured interview with you. That means, I will ask a few guiding questions, but mostly I will listen to what you have to say. I might pick up on points you have made or will ask further questions. Ideally, you would answer these questions from a corporate and strategic point of view. And please don’t hesitate to touch upon important issues that are related to my questions! • How did you get into this job? What was tempting about it?

History

• Can you tell me a bit about the history of X?

– What market gap did X try to fill? / What was your motivation to start X? – How did you enter the market?

1Some interviewees disagreed to have their names used, which is why I changed them. While I cannot prevent you to find out their real names (which was something my interviewees were all made aware of in the consent form they signed), I will not indicate which ones I altered.

319 C. Further Details of the Interviews

– Was this idea shared by your consumers? — Did consumers change since then? – Could you tell me about the more difficult periods? – What other people or organisations were important for X? – In which ways did X impact the market and its development?

Vision

• Could you elaborate on the vision of X?

– Does that vision resonate with consumers? – How permanent is this vision? Have you changed it in the last decade? Do you think it needs adjusting in a few years? — What caused the adjustment? – Can you describe how X tries to achieve this vision? – How effective are these attempts? – What does X mean to its consumers? – Does X mean similar things to stakeholders? – How is the vision reflected in the product range?

Products

• When do you decide to create new products? Could you elaborate a little bit about what happens in that process?

– What do you look out for when developing? And why are these things important? — What are consumers demanding? – Are these focal points communicated in marketing? If yes, how so? – What attributes would you use to describe your products? / Is your product ‘natural’? – Could you elaborate about the similarities between your products and meat? — Do you tell customers about it? – What are the most important differences between your products and meat? — Is there a difference in communication? – Where do you sell, and why there and not elsewhere? – Topics that should be covered (as emerged throughout the interviews): relationship to retailers, protein, authenticity, naturalness, naming

Market

• Can you tell me a bit about the market that X targets?

320 C.3. Interview Guide

– How did the market and the competition change since X began? — What are X’s main competitors? – Does any competition seem like an obstacle to your business growth? – How do you distinguish yourself from your competitors? And what do you have in common? – Could you tell me about the relationships of companies within the market? — Do you have co-operations with other organisations or manufacturers, or maybe even with civil society groups? — Are there any organisations you are less positive about? – What are you doing to shape the market to your favour? – How does retail affect your business and the market? How would the ideal supermarket look like? – At the moment the market is said to be booming. Is X affected by this? [Are new investors coming up?] – Can you tell me a bit about all the regulation that you need to consider? — [I’m asking because there are cases where companies could not give meat-like names to their products any more, for example, or others where large amounts of mineral oils were found in meat alternatives.] — Are there any regulations upcoming, that are important to X? — Did you attempt to influence regulation? – How does X look at future of the market? Will novel technologies affect you?

End

• Last big question: maybe you could tell me a bit about the main strengths and weaknesses of X over the last years? How did these develop, and what impact will they have on the future? • So, I have a few more questions about details that you have mentioned: When was Y? — What was the name of that person who . . . again? — Can you say a few words about Z? • Maybe lastly, can you think of anyone who I’d benefit talking to? [Some colleagues or business partners maybe?] • Thank you so much for you time. I really appreciate it! So I will send you a transcript of the interview for reference; please let me know if anything has gone wrong in the process, or you would like to withdraw any of your statements. Otherwise, I am looking forward to sending you a copy of my research when it’s done.

321

Appendix D

Further Details of the Taste Test Videos

D.1 Detailed Overview of the Sampling Procedure

Having chosen online engagement at large as a source for audience-based data, to address the needs of the research project I searched for instances where normal people described meat alternatives and meat-free eating in order to assess the data of interest through these descriptions. Blogs were not frequently discussing meat alternatives and tended to be sponsored or triggered by free samples from a company, while the ubiquitous question-answer-forums on the internet did not engage in descriptions. Thus, I decided to collect data from social media platforms. After scoping out the quality and quantity of data, I decided to analyse videos from YouTube; particularly those in which users reviewed and in most cases tested meat alternatives in front of their camera, which are usually termed ‘taste test’ or ‘testing’ videos. In the process of identifying taste test videos as a promising data source for this research project, I first scoped out and made myself familiar with generic user-created content on meat alternatives. I soon recognised that taste test videos were an interesting genre of videos: less popular than often studied ‘unboxing’ (Mowlabocus, 2018), or ‘what I ate’ (Al Abri, 2016) videos, and maybe similarly prominent as ‘food haul’ (i.e. shopping review) videos. Taste test videos could contribute meaningfully to the research project by providing a consumer voice to categorisation as people speak about, prepare, eat, and evaluate products of their choice. However, taste test videos had not been studied academically before the time of writing, and thus by watching a number of them I established that these videos were suitable for an analysis at large, and meaningful for the research questions in specific. Staying within the limits of this research project, the range of videos sampled was limited to videos that originating from the UK. Through further scoping I identified search terms with which the videos could be sampled or found on YouTube. To alleviate potential shortcomings in reproducability of my original queries, systematic sampling was conducted: I entered any of the twelve combinations of vegan/vegetarian, meat/burger/sausage, taste test/tasting, and UK in the YouTube search mask on 17 May 2018; I deleted browsing data such as cookies after each search so as not to bias

323 D. Further Details of the Taste Test Videos the output by previous searches and clicks1. From the result list presented in ‘order by relevance’2, I took the first five UK-based videos. If not evident from the description, I used my knowledge of what was available in UK supermarkets (most videos were from the USA, a lot from Australia3), together with distinguishing videos by the accents of the YouTubers and the autobiographies offered on their linked websites. I further excluded videos from buzzfeed and other news agencies, and recipe or cooking videos from my sample. The videos needed to be specifically about meat alternatives, excluding those videos where meat alternatives were part of ready meals, only one part of a set of product reviews, or part of a so-called ‘food haul’, where people buy a variety of products from a specific store and display them. For some searches, I had to scroll through more than 50 videos to find five videos matching these criteria. From the resulting set of 60 videos (twelve search queries times five videos), I removed duplicates which resulted in a list of twelve unique taste test videos. Further videos had been identified during scoping, for example via other search terms, more scrolling, or ‘see also’ recommendations; these videos were not included in the sample as the systematic sampling did not point at these videos. For the selected videos I retrieved the metadata from the YouTubeAPI and stored the video. I then transcribed the videos in verbatim, including minor remarks on actions in case they are relevant for the content, and assigned random names to the YouTubers; all following analyses are based on the audio content of the video. Analyses have been conducted with the qualitative data analysis tool RQDA (Huang, 2016). The availability of all videos was revised six months after the initial sampling, and only eleven out of overall twelve videos could still be obtained. However, following the advise of Patterson (2018) who describes their engagement with a loss-of-subject scenario in considerably more sensitive YouTube research, all videos were kept in the sample. In the specific context, the video was likely deleted with the a lot of other videos of the user, as their whole YouTube channel at the time of revision only contained small number of videos exclusively posted in the previous few months. The particular video had been online for more than 18 months at the time of sampling.

1It would have still been possible for YouTubeto identify me via my IP address, my user-agent indicator, my screen size, and many more. However, the search should also somewhat represent a ‘normal search’ on YouTube which through the privacy settings of an average internet user is hardly ever going to succeed without the platform knowing something about the user. 2Ordering items by relevance was the only setting in which the found videos did not need to strictly include the entered search terms, increasing search results from none or one video per query to a few dozen. Thus, the search was based on the videos that the platform suggested as relevant to my search query on the basis of algorithms unknown to me. While for systematic sampling ordering the videos by country, date, or view counts would be preferrable, the default setting to order by relevance is likely the mechanism which an audience would use to find an initial taste test. 3While most products sampled in the videos were only available in UK supermarkets, Linda McCart- ney’s is also sold in Australia, which made distinction on such product rely stronger on other contextual information.

324 D.2. Overview of Taste Test Videos

The taste test videos sampled have diverse reach, with Lucy’s channel having over 450,000 subscribers and their video had almost 100,000 views at the time of sampling, while Robert’s channel has less than 25 subscribers and 100 views. The view counts may be further skewed by the variance in publication dates, as some were published three years before sampling, and others just a few days before (see section 7.1). It may thus appear unjustified to give all videos the same representation in this analysis. How- ever, in analogy to the adverts for which no correction for the magazine circulation was undertaken, and similar to how consumer interviews are conducted no matter whether people have ten or a thousand friends, I believe this is not a major issue for this analysis. In fact, treating the videos as a set of diverse ideas and opinions which may or may not converge, the argument of views further falls out of place. The choice of videos may be further criticised on the basis of a selection bias, as those people who are doing, filming, and putting online a taste test are open to the consumption of meat alternatives. This would possibly rule out some purist vegans and those who do not like meat. But in fact, the sample includes one video of a person admitting they do not actually like meat (anymore), and another video of a person who saw refer to an example in their online community as a motivation to sample “vegetarian food” they are quite sceptical about. While these examples can not rule out the general concern, the sample accordingly represents people engaged in shaping and negotiating what a category of meat alternatives is and what it means: An audience that buys the products and engages in its valuation. People who do not eat meat alternatives are of course still participants in the construction of the category; but unless explicitly involved through discursive acts, their agency remains in not stimulating market demand as a proxy of categorisation.

D.2 Overview of Taste Test Videos

Because of research ethics, I do not mention title or YouTube video ID alongside their description; the names of the YouTubers are made up for referencing purposes. If you would like know in detail which videos I sampled, please contact me directly.

Table D.1: Overview of the YouTube videos used for analysis.

diet shop product motivation

Alice considers — various meat-free products their vegetarian mum (teenager) vegetarian- such as sausages or Souther bought the products ism Fried VBites

325 D. Further Details of the Taste Test Videos

diet shop product motivation

Charlotte vegan, subscrip- Texas barbecue meatless part of subscription (student) meat tion vegan soy (and other vegan box, tasting vegan stuff eating box foodstuffs) with friends friends Elliot vegan Iceland Iceland No Bull Burger hype around the (adult) product Emily, vegan Iceland Iceland No Bull Burger unclear Finlay (couple?) Georgia something — Linda McCartney (adult) meat-free vegetarian Chorizo and Red Pepper Sausage Jessica meat eater — Linda McCartney’s other people’s (teenager) vegetarian sausages vegetarian food reviews Lucy new vegan online various vegan products, sponsored video, recent (student) such as burgers or pizza dietary transition Lydia, vegetarian, Tesco, various vegetarian and product comparison Robert meat eater Asda, vegan sausages (parents) Aldi, Lidl Mia, Ollie vegan Iceland Iceland No Bull Burger online reviews from (couple) others Phoebe, vegan The various vegan bacons product comparison Steven Vegan (couple) Kind Pippa new pes- Sains- Linda McCartney’s offer at the shop, recent (parent) cetarian bury’s vegetarian sausages (tried dietary transition before) Sophia, vegan Iceland Iceland No Bull Burger hype on social media Zach (couple)

Tags associated with more than one video are: Vegan (7). Taste test (5). Linda mccartney, no bull burger, vegetarian (4). Food, health, iceland vegan burger, uk, veganism, vlog (3). Bleeding vegan burger, burger, challenge, cooking, fitness, food review, grocery haul, haul, how to, iceland, life, plant based, prep, quorn, testing, vegan burger, vegan food, vegan meat, what the health, workout (2).

326 D.3. Structure of Taste Test Videos

D.3 Structure of Taste Test Videos

At large, I found the videos to be divided into four major parts: (1) an introduction to the video, (2) a description of the product, anticipations, and preparations, (3) eating and evaluating the foods, and (4) a conclusion with outlook. In the introduction, people introduce themselves and or the purpose or scope of the video. The extent of this seems to depend on the professionality of the respective YouTuber, and might cover updates of the people’s lives, or just be a recap of a conversation they have had. This usually covers what motivated them to do a specific video, may update the audience on previous episodes or conversations. The introduction could further cover scenes from shop, till, or in case of online shops the unwrapping of the contents somewhat similar to the genre of ‘unboxing’ videos which some social science research has been concerned with previously (see e.g. Mowlabocus, 2018). The following elements of description/preparation, and eating/evaluation can be found in various constellations, especially so when more than one product is sampled. The creators describe the products by taking them out of their packaging, by reading from the packaging, and by commenting on the product’s appearance (or the packaging). Then they tend to proceed to stating some general preferences about foodstuffs they choose or do not choose, by embedding a particular product into their individual narrative, by mentioning where they had heard of a particular product or what motivated them to buy it. They finally tend to voice expectations about the food to be tasted. If product preparation is covered in the video (some simply present the prepared product to the audience), this is usually done after voicing some initial expectations. The preparation could cover frying, grilling, baking, or generally assem- bling a dish, for example by putting a patty with salad, sauces, and tomatoes into a bun. What usually follows is the tasting of the product. For this, usually the product— assembled or not–is presented in front of the camera. In fact, some videos feature tastings of a plain products, such as a patty itself, as well as a product in its ‘established’ context, such as a patty as part of an assembled burger. After presenting the product, the look and smell of the prepared food are described and/or evaluated again. They tend to eat one or two bites, and comment on the experience; usually after this they would cut the video and skip to a part where they appear to be finished with eating and provide a more general product evaluation. Here, the YouTubers would reflect on taste and texture, and might contrast this with the smell or expectations they had. Sometimes, products would be compared with other products they had tasted before (usually meat-free but also meat products). Many participants further rated the product on scales from 1 to 5 or 1 to 10. Finally, they might cover some more general thoughts or reflections, and thank

327 D. Further Details of the Taste Test Videos their audiences for their attention. This includes for example suggestions to the audience to try out the sampled product, or to reduce their meat consumption. They then would ask their audience whether they had liked the video, would urge them to comment, and to ask them if they had any requests for other or similar videos, or whether they wanted any products sampled. Videos tended to conclude by asking for likes or subscriptions. Many YouTubers closed by suggesting to see them again in a next episode. As any attempt to describe a general structure, this is incomplete and subject to exceptions. For example, Lydia and Robert elaborately document their browsing through supermarket shelves and hardly have any cuts during the food preparation and sampling stage, which extends the length of their video to over 30 minutes. Or Pippa, whose face apart from an introductory picture is not visible in the video at all and who does not eat the sampled sausages in front of the camera, but instead chooses to have an ongoing narration throughout the video and its cuts.

328 Appendix E

Further Details of the Tweets

E.1 Detailed Overview of the Sampling Procedure

Tweets in this research project are used to further triangulate the data, particularly providing a counterpoint to the few videos which are also made by consumers. The tweets allow to draw much larger samples albeit providing lower information density per data point. Sampling aimed at collecting a targeted, concentrated data set of tweets that are likely to cover meat alternatives. In the lack of adequate hashtags despite brand names (which would be partial), I decided to track two different events. The first one was the National Vegetarian Week (#NationalVegetarianWeek) organised by the Vegetarian Society of England. It was first held as an offline event with the support of retailers and manufacturers in 1992 (The Grocer, 1992), but has in recent years increasingly been accompanied by online activities. The other event was the WorldMeat Free Week(#WorldMeatFreeWeek), which was first held in 2015 as a World Meat Free Day and extended to a whole week in 2018. Their website is run from the United States, and a US-based PR agency listed on the contact form, but the event is also sponsored by Quorn, which made me believe that there might be some UK involvement in the event, as well. As both of these were top-down facilitated events rather than emergent properties of a conversation or events in world history and as the amount of data was considered low compared to the overall twitter traffic, sampling could be restricted to these two hashtags and did not need to be expanded (cf. Lorentzen & Nolin, 2017). Using the Python programming language and the ‘tweepy’ package, I gathered the data via the TwitterStreaming API and where I could not collect the data live from the Rest API1. For the hashtag #NationalVegetarianWeek data was collected from starting Sunday 13 May 2018 to ending Monday 21 May 2018, and from Sunday 10 June 2018 to

1The use of the so called Rest API for obtaining data in retrospect is only possible for a few days afterwards. This is why I generally preferred the Streaming API to collect data. However, due to the university’s standard computer settings, which I thought I had disabled with the IT service, but apparently I did not, the computer turned off on the first week’s Saturday morning. Further, I initially had only started recording a few hours after the first week started, leaving me to download data from these two time periods in this alternative manner. Any overlap in the data was reduced, and within this overlap it turned out that two tweets had been deleted soon after posting, and I removed them

329 E. Further Details of the Tweets

Monday 18 June 2018 for the hashtag #WorldMeatFreeWeek. As the Streaming API sends relevant tweets in real time to the subscribed device, I collected the data with at least one day padding to either end of the sampling period in order to not have to be on my computer by midnight sharp. All data was collected and analysed with timestamps referring to British Summer Time, BST. All tweets were stored at the time of posting with the Streaming API, which renders various popularity metrics (e.g. likes, favourite counts, etc) useless, since they build up over time. Retweets were still recorded as they covered the same hashtag, and replies and quotes of a tweets with a particular hashtag need not include that hashtag themselves. Again using the programming language Python2, the tweets were screened and preprocessed, for example to obtain the date in BST, or to transform the output format of the API (json) into tabular format3. The gathered data was then filtered to cover only English-language tweets; additional filtering for user-chosen time zone or geographic location was considered but due to the limited spread of these settings not done4. Therefore, any English language tweet is part of the data, although likely most tweets originate in the UK5. That the UK-origin of the tweets cannot be shown for certain is a considerable shortcoming of this data set. To explore the relation of the data to food, the WordNet database was used; this is a lexicographic database for semantic word relationships (Miller, 1995). For each hyponym (i.e. subordinate word) of the WordNet synset (definition) ‘food.n.02’ (“any solid substance (as opposed to liquid) that is used as a source of nourish-

from the data set. Deleted posts were not corrected for in the overall streamed sample as this would have caused a much bigger logistic effort. 2Packages beyond the standard library include ‘pandas’ to operate with large data frames, ‘numpy’ for calculations and support for larger numbers, and ‘altair’ for plotting; later, the ‘nltk’ package for language processing was used for its link to the WordNet database mentioned above. For data storage, the column-based binary feather format was chosen for its speed and smaller file size compared to commonly used plain text files with comma-separated values (csv). 3In fact, as json is an object structure, not a tabular structure, the data was converted into three tables with relational connections amongst them. The three tables for tweets, users, and so called entities such as hashtags, mentions, images, or links, was modelled on the structure of the json-object. 4In particular I looked at the fields for ‘geo’, ‘coordinates’, ‘place’ which can be associated to individual tweets, and ‘utc_offset’, ‘time_zone’, ‘location’ associated to the user; however, users can decide whether or not they wish to set these settings and or make them public. — Filtering only accounts with these settings would have resulted in too little data; conversely, to successively remove users where a respective field indicated they were not UK users would have resulted in a similar amount of uncertainty about what the data set represented. 5There are reasons to believe that most of the tweets originated from the UK. Firstly, National Veget- arian Week is theoretically a UK-focused event. Secondly, in a large majority of the user settings (if available) UK places or counties are mentioned. And thirdly, the distribution of tweets throughout the day which very much resembled a British working day, although there is more ‘international’ noise in the World Meat-Free Week data (whereby all times were converted to represent BST): Activity started from around 7am, with a slight morning peak around 9am, and much higher activity during the afternoon hours, a daily peak around dinner time, and tweets then gradually ceding towards midnight, with the last activity around 3am (see also section 8.1 and particularly Figure 8.1).

330 E.2. Food Words Searched in the Data Set ment”) was written into a list, assorted to the synsets first-level hyponym. These 47 groups covered foodstuffs such as vegetable, edible fruit, or meat, but also concoc- tion, starches, seafood, or cheese. In each original tweet (i.e. excluding retweets) the amount of words in each group of food words was counted. Following this, only those tweets were filtered which included terms that resemble meat or types of meat based on synset meat.n.01: “the flesh of animals (including fishes and birds and snails) used as food.” Words in the filter covered among others ‘beef’, ‘hamburger’, ‘roast’, ‘chop’, ‘venison’ or more specific terms like ‘guinea hen’, or ‘liver’; Notably, the sample did not include the word ‘burger’ as this is categorised in the WordNet database as a ‘sandwich’ and in higher order as a ‘dish’. a full list of these is reproduced in Appendix E.2. From the resulting 1,675 tweets, those relating to meat alternatives were manually identified. The final set of 225 tweets was then qualitatively analysed using a LibreOffice spreadsheet.

E.2 Food Words Searched in the Data Set

In the following, all words with which the twitter data set was filtered for meat-related words are listed in alphabetical order. All 267 words originate from the WordNet database (Miller, 1995), and are direct or indirect hyponyms (i.e. specific words covered under an umbrella term) of ‘meat’ defined as “the flesh of animals (including fishes and birds and snails) used as food” (synset meat.n.01). Aitchbone. Bacon, bacon strip, banger, beef, beef jerky, beef loin, beef neck, beef patty, beef roast, beef tenderloin, beef tongue, beefsteak, biltong, bird, black pudding, blade, blade roast, blood pudding, blood sausage, boeuf, bologna, Bologna sausage, bottom round, brain, brat, bratwurst, breast of lamb, breast of veal, brisket, broiler, buffalo, bully beef. Calf’s brain, calf’s liver, calf’s tongue, calves’ feet, calves’ liver, Canadian bacon, capon, carbonado, Chateaubriand, chicken, chicken liver, chine, chipolata, chitlings, chitlins, chitterlings, chop, chop steak, chopped steak, chopsteak, chorizo, chuck, chuck short ribs, club steak, cochon de lait, cold cuts, confit, corn beef, corned beef, crown roast, cut, cut of beef, cut of lamb, cut of meat, cut of mutton, cut of pork, cut of veal. Dark meat, Delmonico steak, dog, dove, duck, duckling. Entrecote, escargot. Fatback, filet, filet mignon, fillet, fish steak, flank, flank steak, flat bone, flitch, forequarter, foreshank, fowl, frank, frankfurter, fricandeau, frier, fryer.

331 E. Further Details of the Tweets

Game, gammon, giblet, giblets, gigot, gobbet, goose, goose liver, ground beef, grouse, guinea hen. Halal, ham, ham hock, hamburger, hamburger steak, hare, haslet, headcheese, heart, hen, hindquarter, hindshank, honeycomb tripe, horseflesh, horsemeat, hot dog, hotdog. Jambon, jerk, jerked meat, jerky, joint. Knackwurst, knockwurst. Lamb, lamb chop, lamb roast, lamb-chop, lambchop, leg, leg of lamb, linguica, liver, liver pudding, liver sausage, liverwurst, loin, loin of lamb, lunch meat, luncheon meat. Meat, minute steak, mouton, mutton, mutton chop. Neck, neck sweetbread, New York strip. Offal, organs, Oven Stuffer, Oven Stuffer Roaster. Partridge, pastrami, pemican, pemmican, pepperoni, pheasant, picnic ham, picnic shoulder, pigs’ feet, pigs’ knuckles, pin bone, plate, poitrine d’agneau, polony, porc, pork, pork belly, pork loin, pork roast, pork sausage, pork tenderloin, porkchop, porterhouse, porterhouse steak, pot roast, poulet, poultry, prosciutto, pullet. Quail. Rabbit, rack, rack of lamb, raw meat, red meat, rib, rib roast, roast, roast beef, roast lamb, roast pork, roast veal, roaster, Rock Cornish hen, round, round steak, rump, rump roast, rump steak. Saddle, saddle of lamb, salami, salt pork, sausage, sausage meat, saveloy, scrag, scrag end, shank, shin, shin bone, short ribs, shoulder, side, side of bacon, side of beef, side of meat, side of pork, sirloin, sirloin steak, sirloin tip, snail, souse, sowbelly, sparerib, spareribs, spatchcock, squab, standing rib roast, steak, stew meat, stomach sweetbread, strip steak, suckling pig, sweetbread, sweetbreads. T-bone steak, tenderloin, throat sweetbread, tomalley, tongue, top round, tourne- dos, tripe, turkey. Undercut. Varietymeat, veal, veal roast, veau, venison, Vienna sausage, Virginia ham, volaille. Wedge bone, weenie, wiener, wienerwurst, wildfowl.

E.3 Foods Mentioned in the Tweets

Table E.1 summarises the amount of food words found in tweets during National Vegetarian Week (NVW) and World Meat Free Week (WMFW). The total number of tweets with at least one of these food words (n), and the total number of matched

332 E.4. Reprint of Tweets Used for the Qualitative Analysis

words in that food words category (Σ) is presented. The last column lists Cramér’s V for that respective category; based on a χ2-test, it indicates the degree of difference across categories and is often used for large sample sizes, as p-values tend to be very small. In the way used here, results of 0 mean that the food categories are completely evenly distributed across the theme weeks, with any results below 0.3 indicate weak differences. Given the two-by-two matrix used here to calculate χ2, Cramér’s V is calculated as V = χ2/N with N = 12429 and df = 1 (see also McHugh, 2013). � Table E.1: Mentions of various food categories in the tweets.

NVW (n) NVW (Σ) WMFW (n) WMFW (Σ) Cramér’s V

Total 10333 — 2096 — meat 1090 1244 585 630 0.190 vegetable 3394 4708 251 320 0.171 ingredient 1033 1341 82 104 0.079 edible fruit 1255 1357 117 123 0.078 dish 1669 1938 196 226 0.071 dairy product 480 521 26 28 0.064 cheese 384 405 16 16 0.062 baked goods 478 517 42 44 0.049 None 4312 — 1072 — 0.044 dainty 383 408 34 36 0.043 Other 869 901 118 120 0.038 pasta 245 259 25 26 0.030 starches 374 419 50 54 0.025 meal 961 1284 223 367 0.017 course 281 306 51 54 0.006

E.4 Reprint of Tweets Used for the Qualitative Analysis

This section lists the tweets used for the qualitative analysis. They are sorted by theme week and afterwards displayed in random order; dates, authors, and links are omitted to preserve the authors’ privacy. Mentions have been anonymised unless they represent a company or organisation of interest to this research project. All non-standard characters (e.g. emoticons) as well as empty lines have been removed to ensure print- ability.

333 E. Further Details of the Tweets

E.4.1 During #NationalVegetarianWeek 1. been living a good food life recently (l-r: pulled “chicken” hoagie, nachos with homemade mayo and curly fries, burger with more homemade mayo and more curly fries) #vegan #nationalvegetarianweek (link)

2. southern fried bites are god level meat substitutes (link)

3. It’s #nationalvegetarianweek, our menus always include a tasty vegetarian option with every lunch. Peri Peri Vegetarian Sausage Rolls, Mini Deep Pan Margherita Pizzas and Boc- concini & Tomato Caprese Brochettes are just a few our homemade vegetarian finger foods available. (link)

4. Nut Roast anyone #NationalVegetarianWeek

5. Granovita Mock Duck, now just 2.25. PLUS more fantastic meat alternatives available at great prices! #NationalVegetarianWeek (link)

6. How hard is it to give up meat in the UK? Easier than you think - (link) To celebrate #nationalvegetarianweek check your GreenJinn app for #vegetarian #coupons on @LindaFoods Sausages and @QuornFoods Mince at @Tesco, @sainsburys & @waitrose! (link)

7. Since it’s #NationalVegetarianWeek, we thought we’d take a moment to drool over all our tasty ‘veggie’ tapas dishes. From our Tartar de Vegetal to our Ens- alada de Burrata - we’re sure the meat eaters will even these! We also have a full VEGAN menu. (link)

8. It’s #NationalVegetarianWeek! We’ve created some tasty #vegetarian recipes for you to try: Buffalo Cauliflower Wings with our Carrot & Parsnip Fries a perfect party nibble with a bit of spice! Full recipe: (link) (link)

9. Tesco is launching a totally plant-based #vegan steak next week (link) #nationalvegetarianweek

10. #NationalVegetarianWeek! I do my best @QuornFoods chicken sweet and sour rice and salad, Very healthy and very tasty. (link)

11. Try these crispy mock duck pancakes this #NationalVegetarianWeek for a #vegan twist on a takeaway favourite #mockmeat (link) (link)

334 E.4. Reprint of Tweets Used for the Qualitative Analysis

12. It’s @vegsoc’s #NationalVegetarianWeek and at The White Bear there’s always something delicious and plant powered on the menu. How do Mediterranean vegetable ragu or roast vegetable lasagne with garlic bread and salad sound? (link) #Surreypub (link)

13. Plans to watch a game? Bring on the buffalo without the wing with a big bowl of tempting cauliflower bites. (link) #NationalVegetarianWeek

14. #NationalVegetarianWeek 14-20th May. So many gorgeous recipes on #Veget- arianSociety’s website. It’s easy to swap meat for #chunky vegetables or #pulses, #legumes or #soya protein. (link) (link)

15. As a veggie from planet Earth, I don’t feel a National Vegetarian, but it’s nice to see there a #NationalVegetarianWeek somewhere. What nation? There’s not much reason to eat meat given today’s tasty and healthy food options. P.S. I hate veggie products that look like meat.

16. Pea panna cotta with roast butternut squash and mint salsa #NationalVegetarianWeek #ToastedBarley Enjoy (link)

17. Happy National Vegetarian Week! This week we’re all about our chickpea, coriander & parsley bean burger with Cheddar, tomato mayo, rocket & skinny fries! Meat eaters you’ll love it too. . . be bold, give it a try #nationalvegetarianweek #vegieburger #foodporn (link)

18. Nut Roast (vegan & gluten-free) make ahead and enjoy hot or cold! #Nation- alVegetarianWeek @vegsoc #ad (link) (link)

19. For Day 4 of my #NationalVegetarianWeek recipes, this is one of my substitution recipes. I enjoy trying meat/fish dishes with the meat/fish swapped for meat alternatives. This one is great with quorn chicken pieces. #food #recipes #vegetarian #cooking (link)

20. It’s#NationalVegetarianWeekthis week, so we wanted to share a delicious cheap recipe for you to try! You can pick up a tin of kidney beans for around 50p, and if you use this as a meat substitute it can save you some cash! #foodie #recipe. . . (link)

21. Going #vegetarian doesn’t mean a big lifstyle change. Eat the same as before only without the meat. Sunday lunch? Extra veg instead of the dead animal. #NationalVegetarianWeek

335 E. Further Details of the Tweets

22. Veggie sausage cassoulet and stir fry veggie noodles on the menu today #nationalvegetarianweek #binksbrasserie

23. Just a few of the delicious #plantbased dishes @(hidden) Dirty #Vegan Pop Up last night + perfectly timed for #NationalVegetarianWeek Cauliflower “Chicken” + Waffles Mac + Cashew “Cheese” Buffalo Aubergine Wings Gem Lettuce Salad (link)

24. Our cauliflower steak with caramelised pure, deep fried capers and lemon is one tasty looking (and tasting) cauliflower! #NationalVegetarianWeek (link)

25. NATIONAL VEGETARIAN WEEK Have you tried our vegetarian spring rolls yet? A great starter and we often have meat eaters go for these #nationalvegetarianweek #altrincham (link)

26. Vegetarian Sausage Rolls - perfect for a quick lunch, a hearty dinner or a relaxed picnic! Recipe (link) #vegetarianrecipe #MakeLunchCount #NationalVegetarianWeek #MeatFree #picnicfood #UKFoodHour (link)

27. Celebrate #nationalvegetarianweek with us at @(hidden) Vegetarian hot dog anyone? #vegetarian #veggie #TheDirtyDog #MondayMood (link)

28. It’s #NationalVegetarianWeek - celebrate with vegan Badass Bacon. Click the link below for more details #veggie #vegan #vegetarian #plantbased #plant- strong (link) (link)

29. The fake burger test: Could meat made of plants ever fool you? #NationalVe- getarianWeek (link) (link)

30. Happy #nationalvegetarianweek everyone! It’s a perfect occasion to try our new vegetarian option hot dog - The Dirty Dog #ChickenShop #MondayMo- tivation @(hidden) (link)

31. #NationalVegetarianWeek Veg burgers, fries, and fruit smoothies last night. Tonight: my vegetarian nachos and cantaloupe slices. What’s on your plate?

32. Are you veggies looking for a #meatfree roast? Check out this #vegan lentil, chestnut & mushroom loaf! #NationalVegetarianWeek #veganfood (link)

336 E.4. Reprint of Tweets Used for the Qualitative Analysis

33. In aid of #NationalVegetarianWeek, why not swap out your meat with the help of this menu packed full of delicious veggies? (link) (link)

34. Looks like it’s #NationalVegetarianWeek - what’s your favourite meat replacement item? We like a cheeky squeaky bit of Halloumi (link)

35. Vegetarian Sausage Mushroom Sweet Potato Pie - (link) via @(hidden) #nationalvegetarianweek #everydayhealthyrecipes #familydinner #easyrecipe #easy- meal #meatless #mushroom #sweetpotato #lowcalorie #healthyrecipe #pie #shepherdspie #cottagepie #onepot

36. So I’ve been meat-free for over 5 years now - how have I not succumbed to chicken wings while drunk?! Answer: With great difficulty, but pro-tip, keep some @QuornFoods chicken nuggets in your freezer for any drunken, hungry times #NationalVegetarian- Week

37. If you’re worried about going veggie, trust me you can do it! I’ve done it for almost 2 years and I can honestly say I don’t miss meat, there are so many yummy alternatives out there and I’m definitely healthier now I think more about what I’m cooking! #NationalVegetarianWeek

38. Go light on meat, and discover how satisfying peanut butter can be in its place. Spinach and sweet potatoes in satay sauce hits the spot - thanks @(hidden)! #NationalVegetarianWeek (link)

39. Quorn ‘Cowboy’ on the menu today, one of the children’s favourites. Great replacement for chicken. #NationalVegetarianWeek #quorn (link)

40. @frysfamilyUK thick cut chunky strips delicious in stir fry! #NationalVegetari- anWeek (link)

41. From Chicken Green Curry to King Prawn Jungle Curry, we can guarantee we have your Thai favourite! All can be made Vegetarian,so you can celebrant #Na- tionalVegetarianWeek in style! [phone number] to order! (link) #ThaiNow #ThaiFood #LoveThaiFood #rdguk (link)

42. Going to try some new food for #NationalVegetarianWeek? Check out these meat substitutes that are definitely worth a try! (link)

43. #NationalVegetarianWeek Have you tried our amazing Beetroot & Feta Bur- ger? It’s popular even with meat lovers! #veggieburger #bristol (link)

337 E. Further Details of the Tweets

44. @(hidden) Thank you @(hidden) but sorry, this is not a hotdog, this is a falafel log. #Vegan #vegansnacks #veganism #plantbased #vegan #Nation- alVegetarianWeek #Food #Foodie @TheVeganSociety @LindaFoods @Quorn- Foods @Hooba_Foods @VBitesFoods @DeesWholefoods @Vegindisguise @vegusto_uk (link)

45. It’s #NationalVegetarianWeek this week Reducing meat consumption is good for you, animals, and the environment Shout out @LindaFoods for your Veggie sausages!

46. In occasion of the #NationalVegetarianWeek , two special coupons are waiting for you on your GreenJinn app: Linda Mccartney vegetarian sausages at Sains- bury’s and Quorn Mince (meat free) at Tesco and Waitrose Buy them this week and save some money only with GreenJinn!

47. #Vegan jackfruit ‘duck’ pancakes #recipe courtesy of @(hidden) (link) #Nation- alVegetarianWeek (link)

48. They said to eat your fruit and veggies. . .well, for #NationalVegetarianWeek, we present you with the TootiFrooti Chicken Burger! Stack your Fry’s Chicken Burger pineapple, pomegranate seeds and mango Be bold. . . give it a try! Re- cipe below (link) (link)

49. Mmmm, bet you didn’t think you could get 6 tins of @SumaWholefoods Baked Beans and Sausage for just 8.50 did you? #vegan #veganfood #veganism #NationalVegetarianWeek (link)

50. When I was young, being #vegetarian was a bit of a bore. But now there are a lot of decent, tasty, easy options in supermarkets. Why not give #NationalVe- getarianWeek a try? Cutting down on meat can be healthy and good for the environment, even if you don’t go the whole-hog (pun!)

51. #NationalVegetarianWeek! @QuornFoods Rainbow Salad with your Chicken Pieces xx (link)

52. Amongst all the work tweets. . . it’s #NationalVegetarianWeek this week. Being 5 years meat free has radically changed the way I think about consumption. I’d encourage anyone to try it. It’s easier than you think. Plus veggie haggis is infinitely better than real haggis

53. It’s #NationalVegetarianWeek and while we are all about safety, we are also all about HEALTH!Just a few swaps of meat products and adding more fruit and

338 E.4. Reprint of Tweets Used for the Qualitative Analysis

veg will not only be good for you but it’s good for the planet too! #Monday- Motivation (link)

54. Don’t risk frying bacon when you can enjoy a radish? Why #BBQ those chicken drummers when you can gobble down a raw carrot. . . If these measures seem extreme at least read our vital safety tips (link) @(hidden) #NationalVegetari- anWeek #MondayMotivation v (link)

55. Swap out meat for veggies for one week and I guarantee you will feel better, save tonnes of water, crops, and energy #NationalVegetarianWeek

56. @(hidden) @(hidden) @(hidden) @(hidden) I don’t think that human stom- achs can digest grass easily but there are loads of alternatives to meat that we can eat. Check out some of the meal ideas posted under #NationalVegetarian- Week if you are interested.

57. Planning a hen party at home? To be the hostess with the mostest you need to cater for the vegetarians too. Why not wow your guests with these amazing home made veggie alternatives (link) #nationalvegetarianweek #HenPartyHero (link)

58. It’s #NationalVegetarianWeek this week . Why not give it a try? Reducing meat consumption is good for you, animals, and the environment . Also there are some amazing alternatives so it couldn’t be easier! (Shout out @LindaFoods )

59. It’s #NationalVegetarianWeek - make the switch with Cauldron! These delicious veggie sausage & beetroot burgers with horseradish cream go great in a brioche roll with skinny sweet potato fries and a crisp salad: (link) @vegsoc (link)

60. Still unsure if you want to be a #vegetarian? Here’s some a few reasons: More balanced diet Eat more vitamins due to meat alternatives Eco-friendly Helps weight loss Saves money Encourages positive lifestyle choices #NationalVegetarianWeek #ItsAllGood (link)

61. Study: Nut, seed proteins more heart healthy than meat (link) via @(hidden) #nationalvegetarianweek

339 E. Further Details of the Tweets

62. We feel like keeping it healthy this weekend so why not join us and tuck into this amazing tofu steak with Snow Pea, Rocket and Radish salad with a Wasabi dressing it definitely packs a punch! (link) #NationalVegetarianWeek #tofu #wasabi #welovefood (link)

63. It’s #NationalVegetarianWeek everything in our restaurants is Vegetarian, even our Chicken style burgers and Pepperoni Pizzas!! Give it a try this week! (link)

64. My apprentice once told me he eats all meats accept vegetarian sausage #Na- tionalVegetarianWeek (link)

65. Get inspired for #NationalVegetarianWeek with this tofu minute steak recipe! Our organic tofu is grilled for a delicious smoky flavour and served with a creamy peppercorn sauce, french fries and crusty bread: (link) @vegsoc (link)

66. Happy #NationalVegetarianWeek! Did you know we have a range of delicious vegetarian food in our Bistro #BlackburnHouse, and @(hidden) Meeting House Cafe is solely vegetarian? Even our staff can’t get enough of the exquisite nut roast (warning, it’s very moreish!) @(hidden) (link)

67. Trust me: it’s much easier to give up meat than you might think. There are so many alternatives and substitutes, you won’t feel like you’re missing out at all - AND NOBODY HAS TO DIE #NationalVegetarianWeek

68. Sir Richard Branson says, ”I believe that in 30 years or so, we will no longer need to kill any animals and that all meat will either be clean or plant-based, taste the same and also be much healthier for everyone’ #NationalVegetarianWeek

69. Have you been to this BYO vegan burger joint? @(hidden) #NationalVegetari- anWeek (link)

70. It’s #NationalVegetarianWeek, and as always, we’re slinging seitan Side Chick burgers, smoky vegan House Dogs topped with caramelised onions, Hanoi Hangover Fries laden with Sriracha/hoisin sauce and crushed peanuts, and our saucy Kentucky Fried Cauliflower (link)

71. @(hidden) Hi, have you tried our veggie sausage and butternut squash traybake recipe? Let me know what you think? - Sandeep :) #NationalVegetarianWeek (link)

72. Try our tasty #vegetarian recipes this #NationalVegetarianWeek! Our Buffalo Cauliflower Wings with our Carrot & Parsnip Fries are a perfect party nibble with a bit of spice! Full recipe: (link) (link)

340 E.4. Reprint of Tweets Used for the Qualitative Analysis

73. In honour of #nationalvegetarianweek chefs at White Gables have cooked a selection of treats for all Residents, vegetarian or not to sample. Vegetarian options are always offered in our homes. On offer to try was a nut roast, chilli apricot and feta tart, veggi curry and more! (link)

74. Meat free sausage and leek hash for tea. Shhh, the boyfriend doesn’t know it’s @QuornFoods! #NationalVegetarianWeek #meatfree @(hidden) (link)

75. Make sure everyone can indulge in a hearty cooked breakfast with Linda Mc- Cartney’s Meat Free Sausages. On offer in this month’s Stir it up! #VegetarianWeek #nationalvegetarianweek #veggie (link)

76. Do you fancy a steak but vowed a week of veggies for #NationalVegetarian- Week? How about Cauliflower Steaks with Olive and Herb Salsa? Nutritious & delicious (link) #wine #foodie #veggie (link)

77. #NationalVegetarianWeek calls for one thing; Mixed Mushroom Stroganoff, with #JustASplash of brandy; who needs meat on a #Monday? (link) Pop our handy little pouches in your basket at @Ocado #MeatlessMonday #Ocado

78. Not vegetarian in our house, however I am happy to cook meals without meat. I think you become a better cook by getting flavour. So here’s some of my Aubergine ‘meat’ balls ready to cook, people are very surprised to find out there is no meat in them. #NationalVegetarianWeek (link)

79. #NationalVegetarianWeek Tesco To Launch ‘Breakthrough’ #Vegan Steak In 400 Stores Nationwide. #MeatlessMonday (link)

80. My top Vegetarian Foods/Meat free alternatives | #NationalVegetarianWeek (link)

81. Today marks the beginning of #NationalVegetarianWeek so we’re sharing a @Tesco favourite #vegetarian recipe for you to try out! Pick up all the ingredients at Tescohere at The Marlowes for this delicious veggie sausage & butternut squash traybake recipe (link) (link)

82. #NationalVegetarianWeek. Our TruTex wheat protein is Halal and Kosher certified as well as being GMO free, it can be used as a #meatreplacement in

341 E. Further Details of the Tweets

traditional dishes such as spaghetti bolognaise or lasagne, fast food such as burgers and sausages (link) #veggie (link)

83. This lemon and pepper chicken is perfect for this weather Swap the chicken for tofu for a vegetarian treat instead #NationalVegetarianWeek Going to enjoy this for dinner tonight! (link) (link)

84. Get veggie inspired with @(hidden) and the #BOSHbook. It’s packed with #plantbased recipes. A good way to find creative and tasty alternatives to meat and fish dishes this #nationalvegetarianweek. Get Bosh! in the Bookshop. (link)

85. For #NationalVegetarianWeek we will be posting lots of delicious recipes, ideas for meat substitutes and general vegetable appreciation. Here at CSF, we cannot get enough of them!

86. Here’s a dish that ALL will love this #NationalVegetarianWeek The ultimate Veggie Pulled Pork Fajitas (link)

87. It’s #NationalVegetarianWeek or for someone like myself who has been Veget- arian since the age of 13 it’s just another week but definitely go give it a try (I managed to win an anti -quorn eater round with my vegetarian chilli last week ) (link)

88. #NationalVegetarianWeek Eat your greens! Try @QuornFoods @frysfamilyUK or the fab @LindaFoods instead of meat this week. Look at @(hidden) for some #plantbased inspo. Give it a go, it’s really easy and you will feel better! (link)

89. We provide you with protein without the meat #NationalVegetarianWeek @vegsoc (link)

90. Apparently it’s national #NationalVegetarianWeek I have been a strict Vege for 24years. Reflecting back there was hardly any vege products and had to have herb water as an alternative to gravy when I was little. Now there’s so much delicious meat alternatives. Thank goodness

91. Sorry @(hidden), no talk of Manchester’sbest veggie dishes is complete without the @(hidden) Cauliflower Buffalo Wings. #IfYouKnowYouKnow (link)

92. Just because we’re a chicken shop don’t think we’re not all over #nationalvegetarianweek All our sandwiches come with our awesome veggie patty option.

342 E.4. Reprint of Tweets Used for the Qualitative Analysis

Throw in our banging sides like loaded fries and bang bang cucumber and we might just be the best not chicken shop around (link)

93. It looks like another weekend of sun across the UK, which means one and only thing. . . barbecues! Make sure you pick up some of our meat free burgers. You can find out more about them on our website. (link) #NationalVegetarianWeek (link)

94. It’s #NationalVegetarianWeek so it’s a great time to try to explore meat substitutes (like @QuornFoods ) and look for some yummy plant-based recipes! If you try it for a week there’s a good chance you’ll be pleasantly surprised at how easy it is to go without meat!

95. The fake burger test: Could meat made of plants ever fool you? #NationalVe- getarianWeek (link) (link)

96. More and more people are choosing vegetarian alternatives to meat. Dutch companies @Vegebutcher and @vivavivera are making that choice easier in the UK now too @(hidden) @(hidden) @(hidden) #vegetarian #NationalVegetari- anWeek (link)

97. Oh! It’s also National Vegetarian Week! I have been vegetarian for almost 5 years and honestly it’s made me feel so good! No gelatine or meat or leather or anything tested on animals! Thanks @QuornFoods and @LindaFoods for making my meals simpler! #NationalVegetarianWeek

98. It’s #NationalVegetarianWeek, we have 21 Vegetarian Main Meal options available on our seasonal menu cycle, why not try something new this week like Nut Roast & Creamy Garlic Sauce, Mediterranean Stuffed Pepper or Quorn Swedish Meatballs & Tomato Salsa @(hidden) @(hidden) @(hidden)

99. Vegan protein from nuts and seeds better for health than meat, study finds. (link) #Nationalvegetarianweek (link)

100. @IcelandFoods Firstly, No Thanks, why would I want a Burger that looks like it’s Bleeding. And how can you talk about #NationalVegetarianWeek when you have again discontinues the Veggy Lasagna and your Veggy ranger is half a freezer and most of it is FAKE Meat. You might want to skip this 1

101. Check out Caribbean Jerk @QuornFoods Fillets Recipe with Rice at the #Na- tionalVegetarianWeek website via this link (link) (link)

343 E. Further Details of the Tweets

102. We’ve got you covered this #NationalVegetarianWeek with a rundown of the UK’s most delicious veggie dogs. Check out where you can get that hotdog hit delivered to your door (link)

103. Tip for #NationalVegetarianWeek when you’re cooking a sauce with veggie alternatives to meat/protein sources - Quorn mince, green lentils, etc. - add tamari when cooking to give it a slightly ‘beefy’ flavour

104. This week is #NationalVegetarianWeek - a great reason to see if you can go meat-free for 7 days. Every day we will be sharing with you the benefits of swapping out meat in your diet for some tasty alternatives. Why not give it a try? #EatRealFood #ItsAllGood #MondayMotivation (link)

105. #NationalVegetarianWeek quorn is a god send and for everyone who says its disgusting or say shit like “what about bacon” are ignorant

106. #NationalVegetarianWeek My “No Meat” burger, avocado on toast and chocolate strawberry enamel pins are all for sale ..just saying (link) (link) (link)

107. Our homemade vegetable lasagne is set to up your ‘veggie option’ game this #NationalVegetarianWeek! It is served alongside a fresh dressed salad garnish, garlic bread or homemade chips on our Evening Menu. (link)

108. Vegetariansausage and butter nut squash pasties - Craft with Cartwright - (link) via @(hidden) #NationalVegetarianWeek

109. This #NationalVegetarianWeek try these flavoursome veggie sausage and beetroot burgers - perfect in a brioche roll with horseradish cream, skinny sweet potato fries and a crisp salad: (link) @vegsoc (link)

110. Well well what a nice surprise. For #nationalvegetarianweek the #Dutch government shows pride in its innovators that are producing #plantbased #vegetarian alternatives to #meat that can also be found in the #UK. Can we have a #EuropeanVegetarianWeek soon, @(hidden)? (link)

111. Kicking off #NationalVegetarianWeek with some homemade burritos. Tortilla, brown rice, beans, salsa, quorn chicken, sweet corn, chopped tomatoes, peppers, onion (link)

344 E.4. Reprint of Tweets Used for the Qualitative Analysis

112. It’s #MeatFreeMonday and the start of #NationalVegetarianWeek so why not be a little creative in the kitchen! Pick up a pack of tofu from Taifun Tofu and incorporate this tasty, meat alternative into a variety of salads, stir-fry’s and pastas. Get cooking! (link)

113. If ur Tweeting about #nationalvegetarianweek it maybe a good idea to check how u may have offended a Vegetarian first. IE, a wrap on a Train with little in it. Your range is poor & full of Fake meat, Maybe a cookery program that put Parmesan Cheese on A Veggy Lasagna. #justsayin

114. Can you believe this burger is vegan, its even fooled a few meat eaters. Try it for yourself at @(hidden) in Bingham #vegan #veganburger #veganfoodporn #nationalvegetarianweek nottingham (link)

115. Day 1 of #nationalvegetarianweek - Veggie Sausage Sandwiches for me and the kids for lunch, the boy wasn’t impressed, Princess ate half (she’s not very well today). #newbievegetarian #doingitasafamily #mentalhealthawarenessweek2018 #nu- tritionalchanges . (link) (link)

116. Dish nmero tres for #NationalVegetarianWeek is @(hidden)‘s Hot Nachos Topped with Al Pastor ’pork’, charred pineapple salsa, pickled jalapeos, black beans, sour cream, arbol hot sauce. Get them at #Dinerama in Shoreditch every Thu/Fri from 5pm and Sat/Sun from midday. (link)

117. These 30+ #veganrecipes all provide the authenticity of #bacon - minus the #animalcruelty! (link) #NationalVegetarianWeek @(hidden) @(hidden) @(hidden) (link)

118. Tempura veggie hot dog, anyone? Our freshest summer recipes for #Nation- alVegetarianWeek: (link) (link)

119. Happy #NationalVegetarianWeek! It’s really not as hard as you think to swap the steak for a salad. . . #MondayMotivation #VeggieWeek #SearcysHealthy- Notes (link)

120. A veg-version of our latest special, Moroccan the Boat. How so, Danielsan? Easy. Just swap the beef on this burger (or *any* burger) to our veggie patty. A menu hack, in time for #NationalVegetarianWeek. Go frolic. (link)

121. Cauliflower steak - This Veg Can idea for @(hidden) #NationalVegetarianWeek #creative #advertising (link)

345 E. Further Details of the Tweets

122. It’s #nationalvegetarianweek and that could only mean one thing: we’re getting pumped for the ROSA’S VEGGIE COOKBOOK! It features dishes like this awesome butternut cashew stir-fry inspired by the Thai classic chicken cashew nuts. Pre-order now available (link) (link)

123. For @vegsoc #NationalVegetarianWeek please feel free to ask if you want further information about the ingredients of our dishes. If you have a favourite dish, please ask and we may be able to prepare it specially for you or replace the meat with a vegetable substitute #NVW (link)

124. #NationalVegetarianWeekalso there’s no reason why u can’t buy veggie burgers or quorn nuggets etc if ur not vegetarian, u don’t have to fully cut out meat to make a difference with little changes like that. try it out n u might be surprised to find you like it, i know i was

125. Another tasty idea for #NationalVegetarianWeek - check out our food development team’s #vegan jerky recipe. A great #healthy and #nutritious snack which can be made from loads of different fruits and vegetables (link) (link)

126. Subway Veggie Patty is a delicious patty made of carrots, corn, red peppers and soy. Try this tasty meat alternative today #NationalVegetarianWeek #Cast- lesSubway #Subs @(hidden) (link)

127. .@Tesco to launch ‘breakthrough’ vegan steak In 400 stores nationwide (link) #NationalVegetarianWeek (link)

128. Kicking off #NationalVegetarianWeek with Jerk Quorn Fillets - if you’re short on time, you could sub this recipe for your favourite premade jerk sauce. (link)

129. Happy #NationalVegetarianWeek! Do you want to try and cut more meat out of your #diet? Lentils make for a great alternative to meats like beef, pork and chicken. Give them a go this week. (link)

130. Crispy Aromatic Seitan Duck Who said Vegetarians are missing out?? #NationalVegetarianWeek #AniseVeganCreations #Anise (link) (link)

131. On the seventh day of #NationalVegetarianWeek , vegetarians practise rituals to honour their saviour. Similar to Thanksgiving, vegetarians fill bathtubs with

346 E.4. Reprint of Tweets Used for the Qualitative Analysis

Quorn chicken nuggets, & give thanks to those whom acknowledged we don’t want to eat some ugly ass vegetables. (link)

132. You can still enjoy “steak” in a meat-free diet by learning how to make seitan. I’ve been vegan nearly a year and not had a boring meal so far! #NationalVeget- arianWeek #Vegan (link)

133. Sink your teeth into these meat free burgers on #NationalVegetarianWeek(link)

134. Jackfruit, now just 1.80 each and a great alternative to meat. (link) #NationalVe- getarianWeek (link)

135. People ask me why I’m proud to now be Dutch. One good answer is that the Netherlands is leading the way in the production of plant-based meat alternatives!! (link)

136. Looking for veggie recipes so delicious that you won’t miss the meat? We’ve got it covered! See (link) and @vegsoc for more info #NationalVegetarianWeek #meals #RecipeOfTheDay (link)

137. @(hidden) @Vegebutcher @VegaSlager @vivavivera @waitrose @Tesco Yeah the NL will become famous because of its super delicious plant based food! Even the most fervent meat eaters won’t taste the difference #NationalVegetari- anWeek

138. For the last day of #NationalVegetarianWeek,why not swap your Sunday Roast for a veggie alternative? Find the recipe for this flavoursome Caramelized But- ternut Squash with Whipped Feta and Zhoug from MasterChef finalist @(hidden) on our website (link) (link)

139. Celebrate the start of #NationalVegetarianWeek with Ardo’s delicious Cauli- flower Cheese Burger and a side of Honey Glazed Parsnips. (link)

140. Been at this veggie thing for a year. Everyone says that you lose weight + get trim with no meat. They forget to mention that often you substitute chicken for bread. Pork for bagels. Beef for croissants and ham for EVERY TYPE OF CHEESE TESCOS SELL. #NationalVegetarianWeek

141. Out with the beef burger, in with a veggie-friendly alternative #MeatFree- Monday #NationalVegetarianWeek (link)

142. - #NationalVegetarianWeek - Best Veggie Dishes in #Manchester It’s a toss up between the BLUEBERRY HILL (Feta, Red Onion, Blueberry, Watercress, Balsamic) or the THAI QUORN DOUGH (Quorn Chicken

347 E. Further Details of the Tweets

pieces, chillies, Thai Green Curry, Coconut) at @(hidden) My advice? HAVEBOTH! (link)

143. With #NationalVegetarianWeek well under way, we have been looking at the 100% plant-based steak going on sale in @Tesco from next week #vegetarian #supermarket #retailer (link)

144. Who needs #meat when you’ve got sweet potato falafels!? An awesome lunch cooked up by @(hidden) today @(hidden) HQ! Smashing through the #Na- tionalVegetarianWeek @(hidden) up on Friday! (link)

145. #NationalVegetarianWeek! @QuornFoods my Veggie salad with bulgar wheat/spelt lettuce/quinoa and your chicken pieces xx (link)

146. Are you meal planning for #NationalVegetarianWeek? Check out this Buffalo Cauliflower Wings with Carrot + Parsnip Fries recipe by @(hidden) ! It’s easy + delicious #vegetarian (link)

147. It’s #NationalVegetarianWeek and here at Groovy HQ there’s nothing better than a veggie-powered meal! Channel the power of veggies this weekend & throw a #GroovySwitch by swapping out meat for a #plantbased meal. Head to our website for some veg action! (link) (link)

148. It’s #NationalVegetarianWeek! Did you know we stock @(hidden) Meat Free Sausage Rolls? Grab yours today. . . visit (link) or call [phone] to place an order. (link)

149. Not a vegetarian as I do enjoy eating meat. However I buy Quorn nuggets instead of chicken nuggets as they actually taste great. If you cooked them right then did a blind tasting to friends & family they’d never know they where vegetarian. #NationalVegetarianWeek

150. Is this the best plant-based #fastfood joint? Head to London’s @(hidden) for burgers and fried ‘chicken’. Read about more veggie restaurants in the UK here: (link) #NationalVegetarianWeek (link)

151. @(hidden) Great idea! Every week is #NationalVegetarianWeek for me too. It’s a great time for would be veggies to give it a try, the meat substitutes on the market are better than ever, delicious, healthy, ethical & sustainable

152. Now doesn’t that look delicious?! Enjoy our homemade #vegetarian buffalo wings served with green onions and sesame as part of our #NationalVegetarian- Week menu. (link) #VegetarianWeek #Ripon (link)

348 E.4. Reprint of Tweets Used for the Qualitative Analysis

153. Spicy and meat free! Vegetarian curries are where it’s at. Try my Quorn Keema for #NationalVegetarianWeek (link) (link)

154. Great #food doesn’t need meat. . . said nobody ever until they saw this #burger #NationalVegetarianWeek *roasted mushroom with halloumi *sauteed #kale *Smashed #avocado *orange *basil *sweet peppers (link) @vegsoc (link)

155. Since it is #NationalVegetarianWeek, perhaps I could point out that beetroot offers a very acceptable alternative to #gammon?

156. Yuuuuummm meat free lasagne for #NationalVegetarianWeek (link)

157. Happy #NationalVegetarianWeek! When I stopped eating meat a few years ago I genuinely didn’t think I’d get to keep eating curries, burgers, fried “chicken” and pies like this, but I can and it’s effing AMAZING. (link)

158. Flavouring meat alternative products can be challenging. Find out how our specialised meat-free solutions can help: (link) #NationalVegetarianWeek(link)

159. Just as mental health is important, so is diet. Having a good diet can actually impact your mental health, and that starts with healthy eating. Choose veggies over meat for a week and see for yourself! #NationalVegetarianWeek

160. #NationalVegetarianWeek day 2 - grilled portobello mushroom with balsamic, chilli and garlic with @QuornFoods #vegan chicken, grilled tomatoes, rocket and fresh vegan pesto (link)

161. Chicken fried seitan and cous cous and potato salad. Eaten outside, which was nice. #vegan #seitan #nationalvegetarianweek #govegan (link)

162. @(hidden) In NZ, went to a fast food restaurant and was told a veggie burger was a regular burger with no meat in it. So. . .. two bits of bread and some soggy lettuce? #NationalVegetarianWeek (link)

163. even if you just stop eating AS MUCH meat instead of completely going veggie it helps the animals and the environment and your general health!. . . also quorn chicken is fucking lit #NationalVegetarianWeek

349 E. Further Details of the Tweets

164. It’s #NationalVegetarianWeek! We have several herbivores in our midst, but the real question is, which is the best meat substitute?

165. Vegan BBQ pulled pork burger with smoked tofu and crispy onions ! #veganfood #Vegan #burger #NationalVegetarianWeek #hireme (link)

166. DAY 7 #NationalVegetarianWeek One Pot Spanish #Quorn Chicken & Rice (link) #vegan #flexitarian #recipes @vegsoc (link)

167. #NationalVegetarianWeek Day 1, using up leftovers/kinda Nicoise salad: egg, beans, new potatoes, Linda M. shredded chicken, olives, etc. (link)

168. I find that Quorn is a very simple and easy replacement for meat. It’s great for beginners who want to have their favourite meals, just minus the meat. #NationalVegetarianWeek @QuornFoods

E.4.2 During #WorldMeatFreeWeek 1. @Meat_Free_Week @(hidden) Tangy BBQ Cauliflower Wings w/ side baby spinach salad, sliced chunks of vegetarian, gluten free beet burger with honey. #WorldMeatFreeWeek (link)

2. @(hidden) @(hidden) @Meat_Free_Week @(hidden) @(hidden) I am sorry I don’t mind substituting fish for meat but I really dislike most of the veggie substitutes. I am with you in spirit if that helps. #WorldMeatFreeWeek

3. Tocelebrate #WorldMeatFreeWeekhere’sa delicious recipe for Japanese Gyozas. The filling showcases meaty shiitake mushrooms instead of pork. Little changes like this can make a big difference on your way to eating less meat. Hope you enjoy. (link) (link)

4. Tryseitan this #WorldMeatFreeWeek- a delicious ethical and healthy alternative to meat. Our Seitan Bites contain only natural ingredients, no palm oil, are low in saturated fat and very high in protein! In two flavours: Beef-Style and Chicken Satay-Style @Meat_Free_Week #vegan (link)

5. So far during #WorldMeatFreeWeek the week I had 14 meat free meals and cooked 2 new vegetarian meals which are Sweet Potato chips with baked beans and a Sweet Potato Fajita

6. Some ideas for #WorldMeatFreeWeek Shiitake bacon, carrot salmon and soya calamari: vegan meat taste test (link)

350 E.4. Reprint of Tweets Used for the Qualitative Analysis

7. BBQ 51/18: Vegan Paella This from #Bosh. Great flavour, my youngest (who is a meat eater) said “does everything taste better on the BBQ” #WorldMeatFreeWeek #bbq #UKBBQ #ukbbqweek (link)

8. Calm down, carnivores: fake meat with real flavour is good for all of us | Aine Carlin | Opinion | The Guardian #vegan #Vegetarian #MeatFree #WorldMeatFreeWeek (link)

9. BREAKING NEWS: Our Quorn food technologists (aka foodie boffins) have created a Quorn Doner Kebab . We’ve unleashed it in Manchester, with the help of Hunters Kebab Shop. Watch and find out what our hungry meat eating customers thought. . . #QuornKebab #WorldMeatFreeWeek (link)

10. Swap 125g regular beef mince for mix of 75g lean beef mince,75g mushrooms & 50g green lentils and you’ll have 60% more on ur plate, will reduce ur calories intake by a 3rd & ur total fat intake by over two-thirds, plus you’ll boost your fibre intake threefold.#WorldMeatFreeWeek

11. So thanks to @QuornFoods there’s finally a vegan vegetarian #meatfree #plant- based alternative to the ubiquitous Hog Roast! (link)

12. Finally got round to making veggie meatballs with caponata Quite pleased how they turned out! will definitely make again! Thanks @(hidden) +Chef Glen for the cooking demonstration; creative + tasty use of @LindaFoods products Have really enjoyed #WorldMeatFreeWeek (link)

13. It’s #FathersDay and of course Dad deserves his favourite meal tonight. Swop out the meat in his curry for a veggie version, and his normal rice for brown. . . We bet he’ll love it just as much. #MensHealthWeek #WorldMeatFreeWeek (link)

14. Only two meats I’ve ever missed: (1) kebab, (2) corned beef. If this quorn kebab meat is even half-decent. . . (link)

15. Sriracha & Mango Salad recipe from Quorn Great serving suggestion for our #Quorn chicken style pieces (Product: 7034) #HealthyEatingWeek #MeatFreeMeals #WorldMeatFreeWeek (link)

16. Who needs meat when you can have Buffalo Cauliflower wings? Check out our tasty recipe: (link) #WorldMeatFreeWeek #WorldMeatFreeDay #Vegan #veganrecipes #instavegan (link)

351 E. Further Details of the Tweets

17. What’sstopping you from having a meat free peperami? Sound good to anyone else? #WorldMeatFreeWeek (link)

18. BRING BACK MY LAMB STYLE GRILLS! Also, 6 years ago I discovered sticky BBQ fillets and within a year they had disappeared too (link)

19. You do not want to miss this- The Dirty Vegan Available at @(hidden) shacks and Chicken Shop and Dirty Burger restaurants until Sunday the 17th @(hidden) #WorldMeatFreeWeek #veganburger #vegan (link)

20. #WorldMeatFreeWeek I think won’t feel meat free with these bad boys (link)

21. This #WorldMeatFreeDay we think it’s important to make something extremely clear. Our B12 Burger sizzles like beef and bleeds like beef but what matters most is that it’s ALL leaf. What you see is what you get. #MoveOver- Meat #BiteIntoTheFuture #MeatFreeMission (link)

22. There are some legitimate reasons that plant-based protein sources like beans are a healthier alternative to bacon.” - @(hidden) Read More: (link) #World- MeatFreeWeek

23. Sainsbury’s launch a new vegan-friendly range including a plant based burger in an attempt to be at the bleeding edge of cultured meat. @Meat_Free_Week #WorldMeatFreeWeek #Sustainability #BleedingEdge #Sainsburys #supermarket #vegan #plantbased (link)

24. It’s day 3 of #WorldMeatFreeWeek. Have you had a meat free meal this week? If not, Tiago @(hidden) is here to help with this Lentil and Sweet Potato stew made with @quornfoods! All of our great range of our #MeatFree recipes can be found here (link) (link)

25. @(hidden) @Meat_Free_Week @(hidden) Corny joke? Wouldn’t dream of it I enjoyed trying out new recipes & products for #WorldMeatFreeWeek Swapping out meat for a veggie option for just one extra meal a week long term will be of benefit for Planet Earth we owe it to the next generation

26. Step 1: Invent a brand new Quorn Kebab Step 2: Install it in a Manchester Kebab Shop Step 3: Serve it to hungry meat eating customers Watch and find out what happened! #QuornKebab #WorldMeatFreeWeek (link)

352 E.4. Reprint of Tweets Used for the Qualitative Analysis

27. To celebrate #WorldMeatFreeWeek, @LindaFoods #MeatFreeMonday today, and available all week! CIO seminar in Ham Yard Hotel in London. (link)

28. @(hidden) Had a fruity breakfast and am now preparing cauliflower buffalo wings for later #ProtectorsOfThePlanet #WorldMeatFreeWeek

29. Waffles and Savory, Protien-Powered Tempeh Bacon by @Tofurky . I like my waffles with Peanut Butter! #WorldMeatFreeDay #WorldMeatFreeWeek #food #vegan #vegetarian #dairy- free #keto #paleo (link)

30. Did you know it’s World Meat Free Day today? You can grab your delicious vegan burger from Dirty Burger till 17th June. . . #worldmeatfreeday #healthyeatingweek (link)

31. Today’s team lunch was a bit different, consisting of meat free donner kebabs! Not even the most hardened meat eaters could tell the difference and we saved the carbon equivalent of boiling 2330 kettles. Not a bad days work. @Quorn- Foods @Meat_Free_Week #worldmeatfreeweek (link)

32. @(hidden) @(hidden) Had the same discussion with my mum last weekend. Never been a big meat eater and I find I go days meatfree but I’ve made a little more effort in #WorldMeatFreeWeek I’ve been a little more adventurous & found some great alternatives

33. With #WorldMeatFreeWeek upon us, we have some delicious vegetarian alternatives to our meat based canaps, finger food and bowl food! Give us a call on [phone] for more information or email @(hidden) for a be- spoke #MeatFreeWeek menu selector! (link)

34. This #WorldMeatFreeWeek we’re bringing you the “Dirty Vegan”, featuring a @LindaFoods burger available at @(hidden) shacks and Chicken Shop @(hidden) and Dirty burger restaurants #veganburger (link)

35. Making meat alternatives more accessible is great way to get people to try them and realise that they can reduce and or stop eating meat. Would love to see this locally (link)

36. It’s still #WorldMeatFreeWeek and we have lots of meat free food this lunch time, why not try the veggie Shepherd’s pie in Foragers @(hidden) @Meat_Free_Week (link)

353 E. Further Details of the Tweets

37. This weekend why not celebrate #WorldMeatFreeWeek with a sunny (hopefully!) BBQ and cook up the Jamaican-Me-Crazy Not Dog that we created exclusively for the campaign! The recipe’s here: (link) (link)

38. @sainsburys says sales of meat alternative products is increasing by 20% each week and they will start stocking meat alternatives next to meat in 400 stores #WorldMeatFreeWeek (link)

39. Happy World Meat Free Day! Here’s a great recipe that even meat eaters will love! Chickpea Potato Patties - VeganRecipe (link) via @YouTube#WorldMeat- FreeDay #foodblogger #canadianfoodblogger #blackcanadianfoodblogger

40. Its World Meat Free Week, head over to our website for 100’s of meat-free recipes. Including this amazing Cheat day kebab: (link) #meatfreeweek #worldmeatfreeweek #meatfreemonday #vegan #plantbased #vegansofig #vegandiet #veganmeals # (link) (link)

41. @(hidden) Taking part @(hidden) with meat free menus @(hidden) today and delighted to welcome @QuornFoods for a tasting session #plantbased #meat- free (link)

42. Getting a kebab after a night out is sewn into UK culture. Why not feed the drunks meat free alternatives that help the animals and the planet? Everything just tastes like garlic sauce anyway, so surely eating vegetables is better than chicken lips and pigs feet Thoughts? (link)

43. You don’t like meat but make it look like meat and use meat sounding names? (link)

44. Try easing into meatless meals. Consider going meatless one day a week.Start with a couple of meatless dinners each week.Or try substituting the following protein-rich foods for meat in your favorite recipes: #WorldMeatFreeWeek #diet (link)

45. Will you be having a #meatfree weekend? Skip the sausages , bin the burgers and put the kibosh on the kebabs (unless they are meat free of course!). It’s better for you and it’s better for our planet #WorldMeatFreeWeek (link)

46. Tried something new tonight vegetarian sausage pasta using @LindaFoods vegetarian sausages #meetfree #WorldMeatFreeWeek (link)

354 E.4. Reprint of Tweets Used for the Qualitative Analysis

47. It’s #WorldMeatFreeWeek & perhaps you’d like to start your day with @(hidden) delicious vegan sausage rolls A tasty treat for brunch or lunch too (link)

48. @(hidden) Id rather have the version that isnt carcinogenic, requires no suffering, and isnt really bad for the environment. #Vegan Jerky all the way! #worldmeatfreeweek

49. Jackfruit is a great alternative to meat it’s a rich source of protein, fibre and vitamins. We whipped up some tasty jackfruit pancakes with help from our friends Vegan Apron. They’re perfect for #WorldMeatFreeWeek. (link)

50. Switch your daily dose of tough to cook protein to green and crunchy vegetables this WorldMeat free Day with Elica India products. Even swapping one meal to meat free makes a difference! #elicaindia #worldmeatfreeday #meatfree (link)

51. Our planet needs less meat and more plants! Today, the first day of #World- MeatFreeWeek volunteers from @(hidden) were out to inspire Copenhagen to eat less meat and more green. We distributed samples of a tasty plant based Bo- lognese - and it was a success! #GPNvol #lessismore (link)

52. I rarely mention brands - I don’t like to give them free PR and I am so hard to please, few products are worth a mention :) But since it is #WorldMeatFreeWeek apparently. . . Personally I never needed to have substitutes, but if you want to try something low-risk: Quorn Roast!

53. Happy #WorldMeatFreeWeek 2018! From now until the 15th, we urge all the meat eaters out there to swap at least one meal to #plantbased foods. Not sure where to start? Look no further than our PBFA members, who provide delicious plant-based food options. (link) (link)

54. @(hidden) @Meat_Free_Week 10-ingredient vegan pepperoni made with tofu and a blend of spices! Perfect for Cardboard Pizza #WorldMeatFreeWeek #MeatFreeWeek

55. All of our kits are #vegetarian and #vegan friendly, just add in your favourite veggies or meat substitutes. (link) #WorldMeatFreeWeek (link)

56. Hello, is it meat you’re looking for? Read our @(hidden) article into the merits of meat free burgers. #WorldMeatFreeWeek (link)

355 E. Further Details of the Tweets

E.5 Coding Schedule

The following coding categories were defined upfront from the research questions:

Metadata

Subject of the tweet, what the tweet does. reporting about oneself : reporting about one’s own food experiences or cooking, such as about an eaten meal, ones decisions, life as a vegetarian or meat eater; e.g. “now preparing cauliflower buffalo wings for later” (W28) promoting : promoting something or making a case for something, such as a restaurant, a product, a recipe, veganism, discounts, cook books, websites and other media, or participation in the theme week; e.g. “Allour sandwiches come with our awesome veggie patty option.” (N92) valuing : Giving a personal judgement on a situation, development, context, product, product substitution, or feeling; e.g. “Shout out @LindaFoods for your Veggie sausages!” (N45) sharing information : making a statement about something, conveying information, or similar by invoking a third party, such as about nutrition, health, environmental aspects, taste, or stockings of supermarkets; e.g. “Swap out meat for veggies for one week and I guarantee you will feel better, save tonnes of water, crops, and energy” (N55) other : making conversation with audiences by asking them questions, or crack- ing a joke or a pun. What is the (meat alternative) foodstuff discussed? Coding as applicable, multiple values separated by commas.

What are the characteristics mentioned for the meat alternative foodstuff? (zero or more per tweet; grammatically usually attributes; excludes other products that are not meat alternatives as well as the general tweet context) pleasure : when a food is described as tasty, nice, or otherwise pleasurable. taste description : when a taste is described without imparting a judgement, such as beefy or spicy. texture description : when the texture of the foodstuff is sought to be described, such as crispy or chunky. health : when health benefits are claimed, nutrients are mentioned, or health- related topics are discussed. preparation : when products are established to be convenient or cumbersome, simple, easy, difficult, or versatile.

356 E.5. Coding Schedule

environmental aspects : when environmental aspects become a characteristic related to the foodstuff. animals : when animal welfare or animal ethics are associated with the product. others. What are the categorising labels used for the product? (one or more per tweet) related to meat : related to being like meat, such as meat replacements, meat alternatives, or mock-meat. related to other labelled diets : such as veganism vegetarianism, but also carnism. on the basis of contents : such as nutrients (e.g. protein) but also plant-based, or vegetable-based. by proxy of brand : when a brand is used to qualify or label the product, such as ‘Quorn chicken’. as (part of) a dish : when the meat alternative is identified as part of a dish or when a meat-based dish is positioned as meatless. related to the food format : such as burger, sausage, roast, steak. others : such as as generic food, or food with a purpose.

How are the tweets framed? What view of the world is put forward? (exactly one per tweet) normal : when meat alternatives are not questioned or debated, but instead treated as a normal, established foodstuff. choice : when a dish, recipe, or menu with meat alternatives are presented as an equal choice to dishes with meat, or when generally a choice is offered, for example marked with “option,” or “try.” swapping meat : when an option between meat and meat alternatives is given or a swap is suggested, but they are presented as unequal or different; this is usually described on the basis of individual meals or short time spans, often through words like “switch” or “swap.” reducing meat consumption : meat is presented as something to be reduced or reducable, usually on the long-term whereby meat alternatives can aid. as something with external impacts : when meat is not only (to be) reduced, but external impacts and changes as a consequence of this reduction are mentioned; usually these cover health, environment, and animals. others : anything else.

357

Appendix F

Data Summaries Across Data Sets

In this section, the data for questions Qb1, Qb2, and Qb3, as well as for Qc1 and Qc2 are collated across all four data sets.

F.1 Category Identity

Table F.1: Summary of the different evaluations found across the data sets (Qb2).

Data Set Evaluation of Meat Alternatives

Advertisements as (similar to) meat; as meat-based food format; as generic food; as vegan food; in context Interviews as meat-based food format; as dissociated from animals or meaty tastes; in relation to vegetarian food; relative to competitors Taste Test Videos as compared with meat and meatstuffs; as compared with other meat alternatives (although the reference remains unchanged); as compared to other products or untasted products; evaluations based on comparison, subjective appeal, description, or simply mentioned; as good in itself Tweets (not systematically evaluated, see again the research design)

Table F.2: Summary of the different labels and categorisations found across the data sets (Qb3).

Data Set Categorisation

Advertisements by format; food for a specific diet; food for a specific purpose; meat-free food; protein food; generic foodstuff Interviews meat-free category; vegetarian category; independent seitan category; meat category; protein category Taste Test Videos (not systematically evaluated) Tweets by format; product related to meat; food for a specific diet; food known from a specific brand; food denoted by its contents; foodstuffs in a specific dish 359 F. Data Summaries Across Data Sets

Data Set Categorisation

F.2 Frames

Across all data sets, meat alternatives are largely described in context of five different groups: Sensory characteristics, practices, health, values, and product-related characteristics. Apart from product-related characteristics, which are largely of practical concern, the characteristics can be found across the frames identified in the individual data sets in chapters 5 to 8. Looking at health-related frames, differences can be observed between frames advocating public health (e.g. “this is about really taking control of your health”, interviewee Daniel), health of the audience (e.g. “eat the healthy way”, advertisement 2016-07), or health credentials of the product (e.g. “designed to be delicious and healthy”, advertisement 2016-08). They broadly concern either systemic issues about the unhealthy ways in which individuals and society eats, or the properties of the product advocated. Thus, they can be merged with the values related to the diets afforded by meat alternatives as they point towards systemic issues of established diets on the one hand, and with the characteristics related to eating a specific product on the other hand.

PRACTICES established formats, ‘normal’, convenient, known preparation V functional substitute I convenient alternative T legitimate alternative DIET responsible for animal cruelty, T I identical to meat environmental and public health issues A as established as meat V anchor in change I A reinvented meat* V tasty alternative T aid to replace meat A alternative to meat T legitimate choice

CHARACTERISTICS pleasurable, tasty, healthy V tasty nostalgia I nterviews A dvertisements * Note that sensory characteristics here are V ideos highlighted as being different/unrelated to meat. T weets Figure F.1: Frames used for meat alternatives across the data sets. Building on the characteristics identified, the frames for meat alternatives within single texts have been mapped into this scheme. The letters in circles denote the data set this frame occurred in; the elliptic lines identify which items represent which of the three frames.

All previously identified frames across data sets (Qc1) can now be mapped onto the major groups of characteristics: What sensory and health characteristics are attributed

360 F.2. Frames normality, continuation of practices, easy and versatile offerings, choice vegan/vegetarian/veggie, price, established practices, convenience, normality, continuation of practices, product preparation healthier diet, proteinanimals healthier diet product packaging, price, online shopping, word of mouth and social media, meat-free offerings by retailers, animals, branding environment vegan/vegetarian, ingredients, format-focus, versatility, convenience, normality, identity, occasions clean labels, naturalness/processing, additives, nutritiousness, protein, allergens animals, environment, global issues, case supports values, product origin listings, product packaging, business experience values justify business case, business vegan/vegetarian, price, retail Summary of the most prominent characteristics by data source (Qb1). Table F.3: taste, appearanceformat-focus, versatility, convenience, occasions, traditionality/normality generic health claims, taste, clean texturelabels, naturalness, allergens, nutrition, protein, plant-based animals, spirituality, product origin taste, appearance, smell, texturebusiness experience, availability ‘pleasurable’, taste vegan/vegetarian, ingredients, Advertisements Interviews Taste Test Videos Tweets sensory practices health values products

361 F. Data Summaries Across Data Sets to a product, what practices does it afford, and how can this facilitate a different diet. Figure F.1 presents an overview of these frames while assorting the different frames for meat alternatives (in particular the ‘Meat alternatives as . . .’ columns in the respective summary tables) found in the data sets into this schema. Notably, in most cases two or more of these are combined; and in line with the predictions made in the conceptual framework, it appears as if all understandings of meat alternatives at least implicitly rely on all three elements. A summary of the then allocated frames can be found in Table F.4.

Table F.4: Summary of the three overarching frames for meat alternatives identified, and their breakdown into diagnostic, prognostic and motivational components; these frames do hardly occur independently, but are usually used in combination (Qc1).

Frame Component Examples

Characteristics Diagnostic too meat-like or not suffciently meat-like meat Frames alternatives; unhealthy products or diet Prognostic finding (consumers) or offering (producers) meat alternatives filling this perceived gap Motivational maintaining or recovering pleasurable experiences or nostalgia; concerns about health Practice Diagnostic meat is normal, established, and known, so that normality Frames is dependent on meat Prognostic meat alternatives may be used for the same purposes and in the same contexts Motivational ‘normality’ such as established traditions or everyday routines can be upheld or joined Diet Frames Diagnostic established diets are harmful for environment, animals, and public health Prognostic meat alternatives allow are an easy support mechanism to reduce or eliminate meat consumption Motivational saving animals or the planet; doing good

Building on the understandings of meat as described throughout this thesis (Qc2; as listed in the “Meat as . . .” columns in the respective summary tables), similar patterns can be observed here as above. The main difference is that these frames focus on meat itself instead of the role of meat alternatives as a replacement. Table F.5 summarises the manifestation of these frames across the data sets. Particularly, the frames identified related to meat as food with specific sensory properties which can be liked or disliked; secondly, meat can be seen as part of life and embedded in individual

362 F.2. Frames or collective cultural understandings, histories, identities, or experiences, and has a symbolic importance in everyday life and at specific occasions; and thirdly, meat is understood as part of an animal, which means that meat is not only responsible for the death of this animal but also for environmental destruction through the rearing of such animals, and for public health issues through agricultural intensification and societal over-consumption.

Table F.5: Frames used for meat across the data sets. These are from the different texts, whereby specific frame might be split into its components to match the three frames identified (Qc2).

Meat as. . . Food Life Animal

Adverts tasty and entrenched harmful but replaceable pleasurable Interviews tasty normality; convenience inappropriate and and culture harmful, but replaceable Videos tasty entrenched; context cruel; harmful but replaceable Tweets foodstuff normality countable and reducible; harmful

363