Serendipity Science: An Emerging Field and its Methods [1st ed. 2023] 3031335287, 9783031335280

This volume brings together for the first time the diverse threads within the growing field of serendipity research, to

227 20 5MB

English Pages 251 [243] Year 2023

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
Foreword
Preface
Excerpt from a Letter from Horace Walpole to Horace Mann, 28 January 1754
Contents
Editors and Contributors
List of Figures
1 Introduction—A Science of Serendipity?
What does Serendipity Science look like?
Chapter 2—Exploration of “Serendipity” in the Mongolian Language, by Borchuluun Yadamsuren
Chapter 3—Serendipity and Knowledge Organizations, by Toby Burrows and Deb Verhoeven
Chapter 4—Serendipity in Management and Organization Studies, by Miguel Pinha e Cunha and Marco Berti
Chapter 5—Serendipity in Entrepreneurship, Strategy, and Innovation, by Christian Busch and Matthew Grimes
Chapter 6—Serendipity and the History of the Philosophy of Science, by Samantha Copeland
Chapter 7—Serendipity and Ignorance Studies, by Selene Arfini
Chapter 8—Serendipity as a Design Principle of Digital Environments: Theoretical Distinctions, by Urbano Reviglio
Chapter 9—Serendipitous Cognition—The Systematic Consideration of the Accidental Genesis of New Ideas, by Wendy Ross
Chapter 10—Serendipity, Luck and Collective Responsibility in Medical Innovation—The History of Vaccination, by Martin Sand and Luca Chiapperino
Chapter 11—Serendipity Across Contexts: From Offices to Post-conflict Settings, by Michael Soto
Chapter 12: Epilogue, by Sanda Erdelez
References
2 Exploration of “Serendipity” in the Mongolian Language
Adoption and Diffusion of the Word Serendipity in English
The Role of Dictionaries in the Spread of the Word Serendipity
Exploring the Meaning of the Word Serendipity in Mongolian Language
Coining the Term for Serendipity in Mongolian
Serendipity Case Study in Contemporary Mongolia
References
3 Serendipity and Knowledge Organisation
Serendipity and the Physical Library
Serendipity and Information Retrieval
Serendipity and the Information-Seeking Process
Serendipity and the Environment
Serendipity, the Web, and Social Media
Knowledge Organization and the Web
Serendipity and Spatial Representations
Beyond the Knowledge Hierarchy
Humanities Networked Infrastructure (HuNI)
Conclusion
References
4 Serendipity in Management and Organization Studies
The Place of Serendipity in Management Organization Studies
How Organization Counters Surprise and Bulldozes Serendipity
Embracing Serendipity
Exposing the Organization to Serendipitous Emergence
Responding to Serendipitous Opportunities
Future of Serendipity in Organization Studies
Conclusion
References
5 Serendipity in Entrepreneurship, Strategy, and Innovation—A Review and Conceptualisation
Introduction
How Biases of Conventional Decision Making Constrain Serendipity
Step 1: Serendipity Triggers
Step 2: Bisociation
Step 3: Enactment
Discussion
Conclusion
References
6 Serendipity and the History of the Philosophy of Science
The Context of Discovery
Serendipity as a Bridge
From the Hunt to Heuristics
An Epistemology of Serendipitous Discovery
Concluding Comments
References
7 Serendipity and Ignorance Studies
The Serendipity Machine and Why It Could not Work
Not Just a Happy Event
Rational, Willful, and Aching Ignorance
Before and After the Accident
The Embodied and Distributed Aspects of Discovery
The Body Matters: Epistemic Feelings and Discovery
Distributed Serendipity: Interdisciplinarity and Disciplinary Jars
Concluding Thoughts
References
8 Serendipity as a Design Principle of Personalization Systems—Theoretical Distinctions
An Introduction to Digital Serendipity
A Brief History of Digital Serendipity
Digital Serendipity and Its Nuances
Serendipity from a Theoretical Perspective
Serendipity as a Design Principle
Serendipity as a Capability
Two Routes to Design for Digital Serendipity
Hyper-Personalised Serendipity
Pseudo-Personalised Serendipity
Individual and Political Serendipity: Two Sides of the Same Coin
Individual Serendipity
Political Serendipity
Discussion
Conclusions
References
9 Serendipitous Cognition—The Systematic Consideration of the Accidental Genesis of New Ideas
References
10 Serendipity, Luck and Collective Responsibility in Medical Innovation—The History of Vaccination
References
11 Serendipity Across Contexts: From Offices to Post-conflict Settings
References
12 Epilogue
References
Index
Recommend Papers

Serendipity Science: An Emerging Field and its Methods [1st ed. 2023]
 3031335287, 9783031335280

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

Samantha Copeland Wendy Ross Martin Sand   Editors

Serendipity Science An Emerging Field and its Methods

Serendipity Science

Samantha Copeland · Wendy Ross · Martin Sand Editors

Serendipity Science An Emerging Field and its Methods

Editors Samantha Copeland Ethics and Philosophy of Technology Section Delft University of Technology Delft, The Netherlands

Wendy Ross Department of Psychology London Metropolitan University London, UK

Martin Sand Ethics and Philosophy of Technology Section Delft University of Technology Delft, The Netherlands

ISBN 978-3-031-33528-0 ISBN 978-3-031-33529-7 (eBook) https://doi.org/10.1007/978-3-031-33529-7 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

The editors would like to dedicate this book to those who have studied serendipity before them, to those who have been developing serendipity science and to those who seek to know more about serendipity and its study. In particular, we want to mention the members of the Serendipity Society, who have shaped this field and who have found each other unexpectedly seeking the same things.

Foreword

Serendipity is fundamental to science. This quirky and intriguing phenomenon permeates across scientific disciplines, including the medical sciences, psychological sciences, management and organizational sciences, innovation science, philosophy and library and information sciences. Why is it so ubiquitous? Because of what it facilitates and catalyzes: scientific discoveries from velcro to Viagra, innovation of all forms, unexpected encounters of useful information, novel and important ideas, and deep reflection on how we, as individuals, organizations, communities and societies can take leaps forwards by seizing unexpected opportunities and ‘making our own luck.’ Serendipity is therefore a concept that transcends across scientific disciplines and unites them. It can be a powerful ‘stitch in time’ that saves more than nine; it can propel people and organizations forwards in new, exciting and most of all surprising directions. It may even be considered a product of human nature—feeding our curiosity, and our minds through knowledge-building and helping us move boldly forwards into the unknown. This book, Serendipity Science, presents a range of perspectives on serendipity and its importance from across the scientific fields mentioned above. These perspectives are as varied as the fields themselves. They encompass key issues from how to express and communicate serendipity, to how to study it as it happens, to how to cultivate it. These perspectives also incorporate a range of approaches and outputs—theoretical models and taxonomies, methods and genealogies—reflecting the widespread embrace of serendipity across scientific disciplines. I hope reading this book encourages you to embrace serendipity too. Why should you embrace serendipity? Because once you do, it will be forever part of you—encouraging you to make meaningful connections, adopt an open and curious mindset, broaden your interests and help others in doing so—by spreading a serendipity ethos to your colleagues, friends and loved ones to support their growth and nurture their aspirations.

vii

viii

Foreword

This book is itself a potential source of serendipity; I hope reading across disciplinary perspectives on the subject will support you in making new, insightful, useful and possibly even unexpected connections between your scientific interests and the ideas in this book. I hope it will provide new perspectives that drive you forward in your thinking. Welcome to serendipity science. Dr. Stephann Makri Senior Lecturer in Human-Information Interaction, City, University of London and Self-proclaimed ‘Prince of Serendip’

Preface

This book has taken a bit of work, but has been much longer in the making than the relatively short while myself, my fellow editors and the authors of the chapters herein have been actually writing and editing and deliberating over its content. As we mention in the Introduction (Chap. 1), real credit ought to go to Robert Merton, as much as to Horace Walpole himself—while Walpole whimsically coined the term in a letter (see page xi for the passage itself), it was Merton who really delved deep into what it could mean as an explanation for how much of science really happens. The recent formation and growth of the Serendipity Society is but icing on that cake, baked by those who came before—the collectors, the proliferators and the artists of serendipity, who kept the word in circulation and brought out its magic by sharing stories and digging into history to find common elements in the narratives and seeking the traits of those who encounter serendipity’s wonders, so we might better know serendipity itself. This book represents a bit of a different path than has been taken in the past; greats such as Umberto Eco and serendipitists like Pek van Andel have been well known for describing serendipity—but many of those who have joined the Society and who wrote for this volume represent the next step, understanding it in a way that will allow us to create the conditions for it to happen, and as a key to understanding how humans make progress in our complex world. As the co-founder of the Serendipity Society, I have been amazed and intrigued by the expanding circle of researchers and practitioners who want to join our project. The ‘mission’ of the Society has been, from its humble beginnings in 2016, to promote and support rigorous research into the understanding and practice of serendipity. This book comes out of our first Society conference, hosted in London, UK by City University and co-organized by Wendy Ross (co-chair of the Society) and Stephann Makri (who wrote the Foreword, see page vii). At that meeting, only a few of the members of the Society were able to attend, and yet the group was diverse and animated, and the conversations wide-ranging and enthusiastic. What had begun as a mere website and collaboration by researchers who serendipitously met had become a node for a growing network of people who wanted to talk and know about serendipity. Being able to think and discuss it as an important and tangible phenomenon, and to debate its definition without having to justify its importance as ix

x

Preface

a topic of discussion, was a welcome relief to many there—we had found a crowd of our own. The same experience has been expressed by many since, upon finding the Society and participating in our events. But more than a gathering of the like-minded, the Society has allowed many researchers and practitioners to connect to others in unexpected ways; there are collaborations and shared resources that have directly resulted from the Society itself and its activities, and which are producing new work in several disciplines. A canon of texts is forming so fast we cannot keep up with our simple website and volunteer force, and the quality of research is increasingly impressive, especially to those of us who began our work on serendipity by perusing through anecdotes and blog posts, before arriving at the work of key early researchers like, in my case at least, Sanda Erdelez (who wrote the Epilogue for this book, see Chap. 12). Not all who are in the emerging canon are members of the Society now, but many are, and many others who are emerging as sources of key insights into serendipity. Serendipity Science, that is, is one of the most interdisciplinary and mutually respectful fields of research I have encountered in my career. The Society and the fruits of its labour, in turn, have gone beyond anything I could have imagined when it began and has been led and shaped by its members into what it is now. For this reason and others, I would like to extend my personal thanks not only to my co-editors, but we would all three like to thank the members of the Society who have contributed to this effort by reviewing the chapters within, offering insight and recommendations when needed, and whose enthusiasm for the topic has reassured us that the book is timely, important and necessary. As editors, we would like to thank most of all the authors. Each of the chapters in this book is an original contribution, and the book as a whole, thanks to the diverse experience and expertise offered by its authors, offers its reader a broad scope of historical, cultural and disciplinary knowledge. The book was written over the course of the COVID-19 pandemic, and several of us suffered serious hardships during that time—so we would like to extend a further thank you, in particular, to those authors who struggled through to contribute, and who helped each other with reviews and support while writing. We hope you are as proud of the end product as we are. And finally, we would like to thank serendipity researchers who have come before as well as those who will come after—your enthusiasm for the topic and ambitious desire to pursue such a ‘slippery’ subject makes it ever more worthwhile to continue pushing the boundaries of uncertainty and unpredictability in our own, continuing research. Delft, The Netherlands

Samantha Copeland

Excerpt from a Letter from Horace Walpole to Horace Mann, 28 January 17541

… I must tell you a critical discovery of mine à propos: in an old book of Venetian arms, there are two coats of Capello, who from their name bear a hat, on one of them is added a flower-de-luce on a blue ball, which I am persuaded was given to the family by the Great Duke, in consideration of this alliance; the Medicis you know bore such a badge at the top of their own arms; this discovery I made by a talisman, which Mr. Chute calls the sortes Walpolianae, by which I find everything I want à point nommé wherever I dip for it. This discovery indeed is almost of that kind which I call serendipity, a very expressive word, which as I have nothing better to tell you, I shall endeavour to explain to you: you will understand it better for the derivation than by the definition. I once read a silly fairy tale, called The Three Princes of Serendip: as their highnesses travelled, [408] they were always making discoveries, by accidents and sagacity, of things which they were not in quest of: for instance, one of them discovered that a mule blind of the right eye had travelled the same road latterly, because the grass was eaten only on the left side, where it was worse than on the right—now do you understand serendipity? One of the most remarkable instances of this accidental sagacity (for you must observe that no discovery of a thing you are looking for, comes under this description) was of my Lord Shaftesbury, who happening to dine at Lord Chancellor Clarendon’s, found out the marriage for the Duke of York and Mrs. Hyde, by the respect with which her mother treated her at a table. …

1

This passage has been quoted from Horace Walpole’s Correspondence, Yale Edition online. xi

Contents

1

Introduction—A Science of Serendipity? . . . . . . . . . . . . . . . . . . . . . . . . Samantha Copeland, Wendy Ross, and Martin Sand

1

2

Exploration of “Serendipity” in the Mongolian Language . . . . . . . . . Borchuluun Yadamsuren

17

3

Serendipity and Knowledge Organisation . . . . . . . . . . . . . . . . . . . . . . . Toby Burrows and Deb Verhoeven

31

4

Serendipity in Management and Organization Studies . . . . . . . . . . . . Miguel Pina e Cunha and Marco Berti

49

5

Serendipity in Entrepreneurship, Strategy, and Innovation—A Review and Conceptualisation . . . . . . . . . . . . . . . . . . . Christian Busch and Matthew Grimes

69

6

Serendipity and the History of the Philosophy of Science . . . . . . . . . . 101 Samantha Copeland

7

Serendipity and Ignorance Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125 Selene Arfini

8

Serendipity as a Design Principle of Personalization Systems—Theoretical Distinctions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 Urbano Reviglio

9

Serendipitous Cognition—The Systematic Consideration of the Accidental Genesis of New Ideas . . . . . . . . . . . . . . . . . . . . . . . . . . 167 Wendy Ross

10 Serendipity, Luck and Collective Responsibility in Medical Innovation—The History of Vaccination . . . . . . . . . . . . . . . . . . . . . . . . . 187 Martin Sand and Luca Chiapperino

xiii

xiv

Contents

11 Serendipity Across Contexts: From Offices to Post-conflict Settings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205 Michael Soto 12 Epilogue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225 Sanda Erdelez Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231

Editors and Contributors

About the Editors Samantha Copeland is an Assistant Professor in Ethics and Philosophy of Technology at Delft University of Technology. Co-founder and continuing co-chair of the Serendipity Society, Copeland has also recently edited The Art of Serendipity (2022, Palgrave-McMillan) with Ross, and has published in philosophy journals such as Synthese, Perspectives on Philosophy of Science, and in collections on the rationality and epistemology of discoveries made by chance, as well as the impact of contingent environments on the success of potential serendipity. Recent work focusses on the relationships between ethics, serendipity and possibility. Wendy Ross studies the role of material serendipity in higher cognitive processes such as insight problem-solving and creativity. She draws on a range of methods from eye-tracking and experimental psychology to focussed cognitive ethnography. She has recently co-edited the collection on serendipity: The Art of Serendipity (Palgrave). She is Co-Chair of the Serendipity Society and Vice President of the Possibility Studies Network. In 2021, she was awarded the Frank X Barron prize by Division 10 of the APA. Martin Sand is an Assistant Professor of Ethics and Philosophy of Technology at TU Delft. In 2020, he was a member of the NIAS-Lorentz theme group on ‘Accountable and Explainable Medical AI’ at the Netherlands Institute for Advanced Study. Before, he undertook a two-year project on the topic ‘Moral Luck in Science and Innovation’ as a Marie Skłodowska-Curie-Fellow. He is a member of the scientific advisory board of the Journal for Technology Assessment in Theory and Practice and an editorial board member of the journal Philosophy of Management.

xv

xvi

Editors and Contributors

Contributors Selene Arfini Department of Humanities—Philosophy Section, University of Pavia, Pavia, Italy Marco Berti University of Technology Sydney, Sydney, NSW, Australia Toby Burrows University of Oxford, Oxford, UK; University of Western Australia, Perth, WA, Australia Christian Busch Marshall School of Business, University of Southern California, Los Angeles, California, USA Luca Chiapperino Faculty of Social and Political Sciences, Institute of Social Sciences (ISS), UNIL-Mouline, Bâtiment Géopolis, Lausanne, Suiss, Quartier, Switzerland Samantha Copeland Ethics and Philosophy of Technology Section, Department of Values, Technology and Innovation, Faculty of Technology, Policy and Management, Delft, The Netherlands Miguel Pina e Cunha Nova School of Business and Economics, Universidade Nova de Lisboa, Lisbon, Portugal Sanda Erdelez Simmons University, Boston, MA, USA Matthew Grimes Cambridge Judge Business School, University of Cambridge, Cambridge, UK Urbano Reviglio University of Milan, Milan, Italy Wendy Ross Psychology Department, London Metropolitan University, London, UK Martin Sand Ethics and Philosophy of Technology Section, Department of Values, Technology and Innovation, Faculty of Technology, Policy and Management, Delft, The Netherlands Michael Soto University of Minnesota Twin Cities, Minneapolis, MN, USA Deb Verhoeven University of Alberta, Edmonton, AB, Canada Borchuluun Yadamsuren Northwestern University, Evanston, IL, USA

List of Figures

Fig. 1.1

Fig. 2.1 Fig. 2.2 Fig. 2.3 Fig. 2.4 Fig. 5.1 Fig. 9.1 Fig. 9.2

Journal articles and published proceedings mentioning serendipity from 1955 to 2022. Number of publications based on a SCOPUS search conducted on 22/12/2022 with the key word “serendipity” . . . . . . . . . . . . . . . . . . . . . . . . . . . . Vase 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Vase 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Moon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sun . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The process of (cultivation) serendipity . . . . . . . . . . . . . . . . . . . . . . Facets of serendipity in everyday chance encounters (Adapted from Rubin et al. 2011) . . . . . . . . . . . . . . . . . . . . . . . . . . Five volumes of books each consisting of 150 pages. If a bookworm eats its way from the first page of the first volume to the last page of the second volume, how many pages does it eat through? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

5 27 27 28 28 72 171

180

xvii

Chapter 1

Introduction—A Science of Serendipity? Samantha Copeland , Wendy Ross , and Martin Sand

Abstract In this volume, we bring together for the first time the diverse threads within the field of serendipity research, to reflect both the origins of this emerging field within different disciplines as well as its growing influence as its own field with foundational texts and emerging practices. Many have been drawn to the mystery of serendipity, the wonder of the ‘aha’ moments humans experience when they encounter it. In the present volume we present, in contrast to the storytelling approach that drives common fascination with serendipity’s impact on our lives, chapters by experts who have addressed questions about serendipity from a more scientific approach. They ask difficult questions in the following chapters, such as: what is serendipity, exactly?, what role does it actually play in discovery and how does it influence our methods?, how can we study and measure it?, are its results always or necessarily positive?, and how can we potentially influence it?, each with the intent of scientific rigour in their investigations. This introductory review highlights the impact that the phenomenon of serendipity has on human lives, and thereby its worth as an object of serious study. Keywords Serendipity science · Methodology · Wicked problems · Rigour

In this volume, we bring together for the first time the diverse threads within the field of serendipity research, to reflect both the origins of this emerging field within different disciplines as well as its growing influence as its own field with foundational texts and emerging practices. The phenomenon of serendipity has been described in S. Copeland (B) · M. Sand Ethics and Philosophy of Technology Section, Delft University of Technology, Delft, The Netherlands e-mail: [email protected] M. Sand e-mail: [email protected] W. Ross Psychology Department, London Metropolitan University, London, UK e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 S. Copeland et al. (eds.), Serendipity Science, https://doi.org/10.1007/978-3-031-33529-7_1

1

2

S. Copeland et al.

many ways since Horace Walpole initially coined the term to categorize those discoveries that happen by “both accidents and sagacity” (Walpole 1754, see this volume page xi for the relevant passage). We bring together here a sampling of perspectives from scholars who are together forming what we consider the newly established field of serendipity science, to give the reader an idea of some of the various ways that the study of serendipity has come about and progressed in recent decades, as well as many of the directions serendipity research currently evolves towards. The reader will find within chapters written by both senior experts and early career researchers in the field, from organizational studies, management theory, information science and library studies, psychology, literature, computer science, social science, and the history and philosophy of science. Considerations about the importance and role of serendipity are being raised now across science (both empirical and theoretical) as well as practice (from art and innovation to leadership and governance), with ever more eyes looking closer at its significance in human history and the likelihood it will play a key, while unpredictable, role in forming our future. Thus, not only do we and our fellow authors see serendipity science as an emerging field of study, but with this book we also want to stress that it is equally an important and potentially necessary field of study, if we are to deal well as a society with our complex times and uncertain future. Many have been drawn to the mystery of serendipity, and the wonder of the ‘aha’ moments humans experience when they encounter it; in the present volume we present, in contrast to the storytelling approach that drives common fascination with serendipity’s impact on our lives, chapters by experts who have addressed questions about serendipity from a more scientific approach. They ask difficult questions in the following chapters, such as: what is serendipity, exactly?, what role does it actually play in discovery and how does it influence our methods?, how can we study and measure it?, are its results always or necessarily positive?, and how can we potentially influence it?, each with the intent of scientific rigour in their investigations. This book highlights the impact that the phenomenon of serendipity has on human lives, and thereby its worth and importance as an object of serious study. The concept shows itself to be irreducible, designating a phenomenon sui generis, though not unanalysable despite being unpredictable by definition. Despite its elusive nature—a fate it shares with the concept of luck (Hales 2020)—we and our fellow authors suggest that careful study can lead to better demarcation of the boundaries of the concept, to the identification of the causes of its occurrence; their research can provide insight into potential mechanisms for taming it, and in general contribute valuable knowledge about human endeavours, suitable for a wide scope of purpose. That is, serendipity is not only worthy of study–its study stands to offer us key results in psychology, philosophy, entrepreneurial theory, cultural theory, semantics, information sciences, technology design and ethics. The present book provides an entry ticket to this growing and flourishing interdisciplinary body of literature and community. We admit this book really captures only a part of the scholarly work that has been done on the facets of serendipity over the last several decades. As such, we see it as a first step in the work to be done to trace the genealogy of serendipity science as

1 Introduction—A Science of Serendipity?

3

it is emerging from the increasing connections being made between researchers in diverse fields, as they correlate findings and combine approaches. The growth of this community of scholars prompts us to provide an inventory of the current state of the art of serendipity research. We see what we present in this volume as a collection of key grounding work, from which new research directions could be yet explored, including the continuing study of the antecedents and effects of serendipity. That is, we want to trace this movement we are witnessing, a move beyond the anecdotal and into rigorous scientific approaches to the phenomenon. That a separate emerging field is arising is further evidenced by the quick growth and solidification of the network comprising the Serendipity Society, of which one editor is a co-founder and contributors to this volume are members. Like this book, the Society grew into what it is now as a result of a common interest in the rigorous investigation of serendipity, illustrating a growing curiosity about serendipity but also real work being done around the world toward its understanding and utilization, as well as a growing recognition of the issues that serendipity can raise. The Society itself continues to grow, and the chapters following will show that while the work being done already covers considerable ground, there is much yet to be sought and found (although not necessarily in that order). The wicked problems that we currently face, including the rise of disruptive and autonomous technologies such as AI, the recent impact of the global pandemic on our ideas of governance and morality, and the impact of climate change and more, may need non-monotonic and disruptive strategies to be weathered, and we see understanding the role of serendipity as key to ensuring our chance encounters lead to positive outcomes. Historically, stories of serendipity have been shared mostly for fun—it is enjoyable to read and hear of others’ profound experiences with chance, or the impact that accidents have had on what we normally think of as well-ordered science or wellplanned politics. We can find such examples throughout the historical record, even before the word itself was invented to give us a way to categorise them (Alcock 2010). The infamous apocryphal tale of Archimedes’s shout of Eureka! when he is said to have figured out a problem (how to measure the quantity of gold in a crown without melting it down) by watching the water rise in his bath as he sank beneath the surface, has been entertaining folk since Roman times. Volumes of serendipity stories exist to remind us that much of the history of medical science, neuroscience, ethnography and anthropology, psychology, archaeology, astronomy, chemistry and even physics rely upon accidental discoveries, chance meetings between scientists, errors and outliers—one can find today any number of blogs and reports on the famous accidental discoveries in the history of science. Because of its allure in story form, perhaps, serendipity’s place can seem firmly in the arts, for instance gaining popularity with the Cybernetic Serendipity exhibit capturing the London and American public’s attention in 1968/1969, in the titles of movies and in the names of boutiques and ice cream shops.1 One has to stop and take note at some point, however, 1

Notably, it has been questioned since whether the topic of many of these shared stories should be seen as examples of a singular phenomenon, and whether even the Cybernetic Serendipity exhibit was really about serendipity at all (eg. Sneddon 2022). These questions attest to the problem that

4

S. Copeland et al.

of the prevalence of these stories, especially as they are so frequently mentioned as asides in scientific articles and books of theory. Why do they (admittedly!) occur so often, and yet remain generally dismissed in the sciences as addenda, interference, noise or a fluke? Without the admiration of the father of the sociology of science Robert Merton (Merton 1948; Yaqub 2018) it is likely that serendipity would have not even the small measure of respect it had when the scientific movement this book introduces to the reader began in earnest. Behind the scenes, read by only a lucky few, was the volume written in the 1950s but only released in 2004 as a result of the prodding of fans, The Travels and Adventures of Serendipity by Merton and Barber (2004; see the afterword by Merton). The Merton and Barber volume traces the history of the word ‘serendipity’ and offers a social science approach to tracing its impact as a concept and the scientific community’s understanding of its role in discovery. They track its source back through Walpole to Bacon (see also Silver 2015), and forward through sporadic use in literary papers and dictionaries, onto its gaining popularity at the time of writing (with an update provided by Merton in the foreword just before publication). They note the variations in whether serendipity has been an acceptable explanation for a discovery, and for whom—is it a happy circumstance of human life and productive activity, or a sign of reliance upon luck rather than skill in one’s work, thus tainting the credit one is due? This problem of legitimacy continues to plague particularly junior researchers who are less likely to take a risk by following up on potentially valuable but accidental opportunities in the lab or their practice, and who are taught in schools that good science must follow a method. There are of course degrees of respect given to serendipity, often correlating with which of the two ends of the serendipity spectrum a given account emphasises: Many dictionaries and common parlance equate serendipity with simple luck, whereas the credit given (and accepted by) the heroism in many serendipity narratives draws attention to the ‘sagacity’ or wisdom end of the concept. In this book, it is precisely the middle of this spectrum that captures attention—what serendipity scientists want to know, it seems, is how chance and wisdom interact. As Copeland recently wrote, serendipity is discovery that happens “at the intersection of chance and wisdom” (Copeland 2019). However, while the folk understanding of serendipity would have that it is omnipresent, when research attempts to identify and understand it, it often becomes intangible. As serendipity researcher Stephann Makri has noted, it is a “slippery” concept (Makri and Blandford 2012). Despite this complication, however, there has been a thread of serendipity research of the kind we seek to illustrate in this introduction and volume, since at least the 1950s. Barber and Fox’s (1958) report of two scientists, both of whom noticed the same reaction but only one of whom was able to build on it, is an excellent example of the move to a scientific understanding of the different responses to chance events. Since then, research papers reporting or examining serendipity have risen exponentially (see Fig. 1.1). arises with popular conceptions that conflate serendipity with simple luck, for instance – we talk more on this in the following paragraphs.

1 Introduction—A Science of Serendipity?

5

Artciels with the word serendipity appearing in the title astract or keywords

250

200

150

100

50

0 1955

1965

1975

1985

1995

2005

2015

2025

Year Fig. 1.1 Journal articles and published proceedings mentioning serendipity from 1955 to 2022. Number of publications based on a SCOPUS search conducted on 22/12/2022 with the key word “serendipity”

A Scopus search for papers with serendipity in the title, abstract or keywords in December 2022 yields 212 papers across fields as diverse as computer science, arts and humanities and psychology. Alongside more traditional reports of the role of serendipity in information search (Watson and Bullard 2022) or scientific discovery (Thirugnanasambantham et al. 2022), other papers move to discuss serendipity as a mechanism for online interaction, such as impulsivity in online shopping (Bao and Yang 2022), or as a measure of social media engagement (Olshannikova et al. 2022). Perhaps more salient is that serendipity is also investigated as an outcome rather than a cause, encompassing both the benefits of reopening university campuses (Jansz et al. 2022) and how to design for serendipity in online communities (Qin et al. 2022). Many of the new papers rely on user surveys, but others look at more innovative ways such as experimental design (Smets et al. 2022). The extensive and diverse bibliographies in each of the following chapters further illustrate the intensity and scope of research on serendipity and related topics, especially since the Merton and Barber volume was published. The almost exponential growth in interest in serendipity visible in Fig. 1.1 across a wide range of disciplines has not yet been met by an effort to synthesize and comprehensively display the achievements in the field up to this point, which is what this volume aspires to do. Further, there is still room for a correlated growth in research that directly tackles conceptual analysis and measurement issues. The chapters herein delineate several directions that such research is taking and where it may yet go. A scientific approach to serendipity depends upon conversations around

6

S. Copeland et al.

these questions as well as resources and we hope this volume will encourage this momentum by providing both.

What does Serendipity Science look like? Reflecting the complexity of this concept, the chapters in this volume address serendipity research from various angles: Some are looking to measure its prevalence, others its underlying mechanisms, and yet more are interested in serendipity as a proxy for understanding other aspects of innovation and discovery. On top of this there are those who are less interested in the concept directly and more in the ways in which the surrounding socio-material environment can be constructed to increase incidences of serendipity. Each of these different approaches, however, agree on one thing—the possibility of a systematic research programme moving beyond description and catalogues of incidents. The purpose of this volume and the idea of a “science of serendipity” is to bring together researchers who, while being interested in what serendipity is, its patterns and place in our folk histories, are also proposing more scientific methods for understanding it. A science of serendipity has a difficult challenge. It sets out to measure something which is at best unstable and at its worst slips out of our scientific grasp. The object of interest cannot be seen or directly measured; a researcher can only measure the traces. One issue with serendipity research is that it relies often on retrospective and individual accounts of the past, rather than direct observations of its formation in the present. It has no consistent material reality for arguments or empirical experimentation to uncover. Following Sherlock Holmes (a legitimate move, given that the metaphor of detective work is often used in explaining sagacity), one might consider it as that which remains unresolved after rigorous historical explanation has been taken into account, and both known and familiar causes have been provided. That is, serendipity is recreated anew every time it is examined, and the cause-effect relationship cannot be controlled for. The contextual nature of serendipity is explicitly addressed by Yadamsuren in her reflections on the difficulty of examining the phenomenon in Mongolia, where there is no word that approximates to “serendipity”. Yet, this problem is no less pertinent in the languages in which serendipity exists as a singular word. The border between luck and serendipity is fuzzy in everyday speech and yet, as Copeland makes clear in her chapter, what we often hear about serendipity is that it is distinct and in opposition to accounts of “pure” luck. Indeed, what makes serendipity an interesting and yet complex phenomenon is precisely that: it is not “pure” in any way, but rather is a messily constructed understanding of contingent processes arising from a meshwork of people, objects and cultures. As serendipity researchers, however, we are not discouraged but rather see this as a problem to be tackled. The chapters in this volume demonstrate that a scientific approach to the complexities of situated human activity, with all its meshed, contingent complications, is possible. Indeed, several authors propose methods for

1 Introduction—A Science of Serendipity?

7

its systematic investigation, identify pitfalls that other methods may introduce, and recognize the impact of their results on the world. Thus, it does seem not only that serendipity science can be done, but that common ground may be found. In this volume, the most salient and notable common feature that emerged from its writing is that each author takes people in action as a starting point: The situation is key. For Arfini (Chap. 7), her thought experiment of a Serendipity Machine ensures we are clear about the relational and entangled nature of the production of knowledge, something that Reviglio’s (Chap. 8) study of digital serendipity also makes clear. Burrows and Verhoeven (Chap. 3) discuss the nature of information search in both technological and socio-cultural environments, highlighting the interactions between users and curators of those spaces. Bush and Grimes (Chap. 5) and Cunha and Berti (Chap. 4) similarly take as their starting point the messiness of organisational research “in the real world”, and Copeland (Chap. 6) argues that even the difficult and seemingly mysterious reasoning behind chance discovery can be tackled by philosophical investigation, once we take context into account. Sand and Chiapperino (Chap. 10) are interested in moral luck as it unfolds in real contexts, which mirrors Soto’s (Chap. 11) discussions of power complexities in day-to-day life. Even Ross (Chap. 9), who takes perhaps the most reductionist approach to the study of serendipity is clear about the fundamental need to research people in action. The phenomenon at the heart of serendipity research, the phenomenon of an interaction requires the active embrace of life’s complexities, and it is this that makes it a promising and fertile research area that may spread beyond its initial premise. However, for a research programme to thrive, for there to be science of serendipity, then methods must be selected to understand it even if those methods are pluralistic. For instance, Yadamsuren (Chap. 2) articulates eloquently the need for a word to encompass what is already known and in so doing concretise it. However, too little attention has so far been paid to the way in which the methods used to identify serendipity influence the phenomenon. Throughout serendipity research literature as well as in the contributions in this book, there is a strong reliance on the case study as a method for understanding contingency. This is not without its problems, as Ross argues, but it is worth considering why there is a reliance on this method for understanding the complex nature of our interaction with chance events. What Sune Steffensen has called the probatonic principle (Steffensen 2016), may be particularly relevant here: As a research principle it states that much can be gained from scrutinising single, particular instances in detail. The probatonic principle urges us to study instances, either in their own right or as part of a hypothesis-generating procedure. The principle’s importance lies in the fact that it forces us to attend to small, nonlinear (and at times one-off) phenomena that (also) impact on behaviour. (p. 30)

The focus on the particular helps us to appreciate that generalizations and abstractions come at the cost of reduction and loss of information. Yet, as we can see from reviews such as in Busch and Grimes’ or Reviglio’s chapters, this focus on the particular can also draw our eyes towards patterns and generalities which can be tested as such: generalizations have to start out from an in-depth understanding of the

8

S. Copeland et al.

individual case. Each research programme is required to develop its own tolerance of the boundaries of generalisability and exception; even within mature research programmes these things are subject to change and tension. Serendipity science is particularly affected by this tug and pull because the phenomenon is intimately contingent and yet, as we see here, common across cultures and contexts in a way that shows it to be a generally human and shared experience. Alongside the focus on serendipity, there are other commonalities between the chapters in this volume. In particular, the authors here take serendipity as a lens to cast a critical eye on some of the foundational assumptions of the research field in which they situate themselves. In some ways this means that serendipity functions as a hook (as Napolitano 2013 also argues) to support broader arguments. Serendipity is then used as a foil, with a reflective sheen that highlights the gaps in current research programmes, which in turn rely on outdated views about linear, disembodied and non-emergent processes. Thus, serendipity as explored in this volume is not only a phenomenon of interest in its own right but a call to direct more attention to the non-linear and unpredictable and the role it plays in the making of our world.

Chapter 2—Exploration of “Serendipity” in the Mongolian Language, by Borchuluun Yadamsuren To begin our own travels and adventures with serendipity science, Yadamsuren leads us into her world as a serendipity researcher who works across the boundaries of language, and often in her native Mongolian. Despite its growing popularity in English speaking countries, there are many languages where the term has not yet entered into standard parlance. Yadamsuren surveys various English-Mongolian translations and their semantic differences. Through interviews with various scholars and writers of the Mongolian language, she explores other Mongolian concepts that better capture the semantic nuances, which current translations seem to miss. Despite the lack of a singular word to capture it, she shows that serendipitous experience in Mongolia is “practical and everyday” and “essential to survival as a nomad.” Building upon this, she makes a strong case for going beyond the classic examples of Western serendipitous discovery, to offer equally interesting cases of serendipity and to give credit to serendipities’ collective dimension that is evident in the Mongolian culture. As the opening salvo into serendipity science, Yadamsuren reports to us as a veteran of the study of serendipity in her own field, and an explorer who brings the concept itself into new territory as she adventures there. The kinds of adventures that bring serendipity into everyday life can often be quite mundane—Yadamsuren is known for her work on the role that serendipity plays in everyday news acquisition, a topic that carries extra weight in today’s climate of concern about fake news and tyrannical governments that suppress citizen access to information about ongoing

1 Introduction—A Science of Serendipity?

9

events. This intermingling of chance and purpose is a theme carried through in the following chapter.

Chapter 3—Serendipity and Knowledge Organizations, by Toby Burrows and Deb Verhoeven Burrows and Verhoeven return to Walpole’s original story of the Three Princes of Serendip to remind us that aside from sagacity, it is the various places that the princes travel to that afford serendipity. Consequently, they contemplate the idea that serendipity might be more than a merely passive source of the unsought. The design of physical objects and spaces can either afford values or undermine them. Considering serendipity in the design of environments, whether physical or virtual–their chapter suggests–can help us move away from efficiency in our assumptions about what kinds of systems (libraries, recommender systems) will be more ‘productive’ when it comes to real learning. The value of serendipity itself is brought into question in this chapter, as in others within this volume and elsewhere. What is its value, and why would we hope to cultivate it, if we could? Serendipity is probably most frequently depicted as happening in a library, where physical books might be brought near to each other by organizing systems, whether the Dewey decimal or the librarian themself, so that patrons could find them accidentally and yet profitably. Creating such spaces in which serendipity can happen might be physical, or it might be sociocultural as in the previous chapter, but it can also be an organizational space, as the next chapter considers.

Chapter 4—Serendipity in Management and Organization Studies, by Miguel Pinha e Cunha and Marco Berti In this chapter, Cunha and Berti understand serendipity as a process that requires the right combination of effort and luck. They frame serendipity as “negative capability, i.e. a capacity to pursue a vision that leads to confusion and uncertainty rather than to certainty and clarity.” This chapter draws a picture of effective organizations as organic, acknowledging the inevitability of unpredictability and the uncontrollability of the environment in which the organization has to act. Indirect approaches such as generative doubt and peripheral vision can help. Yadumsuren in the first chapter had to approach serendipity indirectly from another cultural perspective to understand how it operates within that environment; we find throughout this volume, as again here in the chapter by Cunha and Berti, that indirect approaches to serendipity may offer the most potential for success. In the context of management and organizational studies, this success is within the market, achieving positive results from one’s strategies and actions. The next chapter further

10

S. Copeland et al.

develops these ideas about how serendipity can play a positive role in the success of individuals, groups and institutions, and even how its probability, at least, might be purposefully increased overall.

Chapter 5—Serendipity in Entrepreneurship, Strategy, and Innovation, by Christian Busch and Matthew Grimes This chapter offers a review and a conceptualization of serendipity from the perspective of entrepreneurship and innovation studies. Busch and Grimes start out with the basic assumption that—contrary to common notions—success in entrepreneurial enterprises is less a matter of anticipatory planning and intentional decision-making, but rather a result of the complex interaction of various factors, of which many remain unpredictable. Thus, serendipity plays a major role in entrepreneurship; naturally one wonders whether its occurrence can at least to some extent be triggered, fostered and exploited. The authors advance a process-oriented model of serendipity that serves as a basis to elaborate factors that increase the chances for serendipitous encounters, and how to capitalize on them. Amongst those, Busch and Grimes distinguish between individual (including reframing, extrovertedness and perseverance) and organizational factors (including systematic evaluations, iteration and team-based collaboration). This idea that serendipity itself is valuable, that it can be cultivated—and might be exploited—is a theme that comes up within several of the chapters in this volume. For instance, Sand and Chiappareno in Chap. 10 point out that moral valence can be attached to serendipitous discoveries, despite the role of chance. Soto, in the penultimate chapter, points to the implications of leaving change to chance encounters, without attending to the factors that make some of them successful, but not all. But the study of success factors does not always lead to the insight we seek about the nature of serendipity—as Copeland points out in the following chapter, focusing on the outcome can leave the origins of discovery mysterious and underexamined.

Chapter 6—Serendipity and the History of the Philosophy of Science, by Samantha Copeland This chapter continues the exploration of serendipity science from within disciplinary perspectives, taking us from entrepreneurial theory to philosophy of science. Copeland ties the lack of explicit exploration of serendipity, per se, in this field to its history, where elements of discovery that seem contingent, mysterious, or psychological, rather than cognitive, rational and repeatable, have been excised since philosophy of science first drew hard boundaries around what it would study. As with other authors in this volume, Copeland proposes we look to serendipity to instead expand

1 Introduction—A Science of Serendipity?

11

the boundaries of what falls into the realm of topics worthy of study—indeed, given the admitted prevalence of serendipity, and its existence as a category that marks out discoveries made with sagacity from those made out of mere luck, this author argues it may be particularly worthy of study by philosophers. Here, as elsewhere in this book, we find that the author utilizes serendipity to explore boundaries and dichotomies otherwise assumed to be a priori within her discipline of expertise. The emergence of serendipity as a field of study, more generally speaking, has come about in part because of these more independent disciplinary turns toward new approaches that allow unpredictability, chance and accidents into our accounts of otherwise strictly logical realms of reason and organization. As Cunha and Berti point to in Chap. 3, organizations are, like science, often asserted to be about controlling the world around us, and bending it to our preferences and logic. The turn toward serendipity in the understanding of these different disciplines is perhaps a sign of a broader desire to move beyond paradigms that require us to try to control our world by limiting our ignorance. Indeed, this is a theme picked up explicitly by Arfini in the next chapter.

Chapter 7—Serendipity and Ignorance Studies, by Selene Arfini In her chapter, Arfini takes on the long-standing paradox that the movement from ignorance to knowledge presents: How can we find new knowledge when we do not know what are we looking for? This question is a brief version of Meno’s or the Learner’s Paradox, which still manages to be upsetting in contemporary philosophy, despite having been discussed since Plato’s times. Plato’s approach suggests a dichotomy, where knowledge and ignorance are antithetically opposed and mutually inaccessible, but Arfini points to the assumptions this dichotomy relies upon about discovery. Along with Copeland, who describes this situation from a less analytical perspective, Arfini provides suggestions about how we can draw insights from the study of serendipity to reframe ignorance from a cognitive perspective. When we understand discovery as a leap between incommensurable states, it remains mysterious: using a concept such as serendipity which encapsulates both ignorance (as chance) and wisdom (sagacity) can bridge this imagined divide, or can provide insights into what is valuable about ignorance for the production of knowledge. Serendipity, that is, offers a framing that encourages (and allows) us to consider the interaction rather than the distinction between chance and wisdom. The next few chapters take up the issue of framing and its impact as well, considering how we might frame serendipity (Chap. 8), how different understandings of serendipity can impact how we might be able to empirically assess knowledge production (Chap. 9), as well as how we evaluate the discoveries of others (Chap. 10).

12

S. Copeland et al.

Chapter 8—Serendipity as a Design Principle of Digital Environments: Theoretical Distinctions, by Urbano Reviglio In the digital realm, investigations into serendipity have engaged it as a possible principle of design, a way to improve user experience so that our engagements with computers feel authentic and provide us with novelty as well as efficiency when we use recommender systems, for instance. Reviglio points out, however, that the theoretical and ethical analysis of serendipity itself is limited, and so he provides a taxonomy of digital serendipity. His categories of pseudo-personalised and hyperpersonalised serendipity, between individual and political serendipity, and between fake and manipulative serendipity, serve as descriptions to work with when considering what is valuable about digital serendipity. Ethical serendipity would preserve the appropriate aspects, as more designers attempt to build it into their approaches to human–computer interaction. The idea of an ethics of serendipity comes into play more explicitly in the last two chapters; first, the following chapter delves further into another idea raised by Reviglio: that we can understand serendipity as being an individualized experience, or from the perspective of the impact it has on an outcome. In other words, we can study how and when serendipity occurs, or we can focus on the phenomenology of serendipity—depending on which perspective we take, how we study it will vary significantly, as will the idea of serendipity itself that arises from such study.

Chapter 9—Serendipitous Cognition—The Systematic Consideration of the Accidental Genesis of New Ideas, by Wendy Ross As an experimental psychologist, Ross has taken up the problem of how to investigate the ‘slippery’ concept of serendipity as it happens, by way of empirical research. In her chapter, she proposes a method for such investigation, in order to disentangle the occurrence from the experience of serendipity. It is possible, Ross argues, to focus on the micro-level of such incidents in experimental contexts, in order to gain a deeper understanding of the conditions that allow for serendipity to happen. That is, rather than studying the environment, historically or material, that creates more chances for individuals to be serendipitous, we can focus narrowly on the material agency of individuals who notice such opportunities. Meant as a complement to broader, social and cognitive accounts, Ross’ proposed research programme demonstrates that as serendipity itself can happen at multiple levels, so can serendipity science. As Ross comments in her chapter, serendipity is, “an experience, triggered by an event which serves a rhetorical function”—because of this complexity, it requires a plurality of methods to study. The significance of serendipity as a rhetorical category was the topic of our first chapter; in this chapter, Ross discusses the impact

1 Introduction—A Science of Serendipity?

13

of how we use it rhetorically on how we perceive it from a psychological or cognitive perspective. In the following chapters, the authors draw out the importance of this consideration. What we label as serendipity, and how we try to use serendipity, also has ethical importance, something addressed as well by Reviglio in Chap. 8, by Erdelez in the Epilogue Chap. 12, and an increasingly common theme in the science of serendipity.

Chapter 10—Serendipity, Luck and Collective Responsibility in Medical Innovation—The History of Vaccination, by Martin Sand and Luca Chiapperino This chapter takes on the problem of moral luck by using serendipity’s versatility to look closely at how we have valued discoveries in history. Moral luck supposedly occurs when someone receives praise or blame for things beyond control and—given the ubiquity of luck—this seems to be a seriously disquieting aspect of ordinary morality. The rewards and recognition for serendipitous discoveries fall into exactly this category of luck-infested reactions: intentions, actions, attributes of scientists’ characters, are not all that matters for discoveries to obtain, just as cases of moral luck draw our attention to what is beyond morality and yet affects our moral judgments. Looking at Edward Jenner’s discovery of vaccination, Sand and Chiapperino work through the various ways in which credit is awarded, or not, in science and how scientific practice struggles with the continuing presence (even post-discovery) of luck in science. Sand and Chiapperino suggest that it is better to see serendipity in science as emerging from collective action—as Ross suggests in the previous chapter, finding what is common among serendipity events tells us something more than examining each individual narrative. Whereas the merit of Jenner’s discovery comes into question if we look at the influence of luck on the credit he has been given for the impact of vaccines on the world, this does not occur when we see the full contingency of history as playing a role in that success—that is, how we explain the role of serendipity in discovery has ethical implications. The final chapters of this volume follow that by illustrating the ways in which serendipity can be a useful tool, in Chap. 11, or in the Epilogue Chap. 12, how it might be misused, and therefore why we need to understand it better to use it well.

14

S. Copeland et al.

Chapter 11—Serendipity Across Contexts: From Offices to Post-conflict Settings, by Michael Soto We end this volume in a reflective tone, as it was begun. In the final full chapter, Soto provides us with the story of his own investigation of serendipity and its impact in diverse contexts, having introduced Randomised Coffee Trials at Nesta when he worked with that innovation-oriented organization, and then having seen the impact of serendipitous experiences on the ability of individuals in conflict settings to move on in new ways with each other. The chapter thus juxtaposes two distinct contexts, drawing out the significance that random, shared moments can have on the paths people take, and the importance of understanding the role of privilege in allowing for such encounters to happen and for people to be able to follow up on those moments. As with other chapters in this volume, Soto points to how networks arise within contexts and conjoin individuals in unexpected ways. The network alone is insufficient, however, because what has real impact is the meaningfulness of the exchange that occurs between people who do connect. The value of serendipity is raised again as a key issue, and we find once again that its value can also be both individual and contingent, and social and significant for the world.

Chapter 12: Epilogue, by Sanda Erdelez We conclude the volume with an explicit reflection by one of the more important figures in the development of the field which emergence we are attempting to trace. Erdelez organized several of the first international meetings with other researchers in the broader information sciences specifically on the topic of serendipity, and contributes much still to its legitimacy as a subject of empirical study and real-world implications as a senior scholar in the field. Her theories ground much of the work found in this volume and, like other authors herein, she has forged a path within her own discipline with work that, together with the work of others referenced in previous chapters, is beginning to shape the canon of serendipity science. It is from this position that she looks back on her own career and this book, and forward to what serendipity science can mean for future research and how to ensure it meets its potential for doing real good in the world. Erdelez cautions us to remember that serendipity can bring real value, but the accompanying pleasure in experiencing serendipity is both a boon and creates the potential for its exploitation (see also Reviglio, Chap. 8). We ought to attend to the ways that serendipity and surprise are brought into our lives, to ensure we are experiencing serendipity that is truly valuable for us, and not merely or maliciously being guided into an experience much like serendipity, but with value for someone whose intentions we wouldn’t otherwise support. For these reasons, Erdelez, like the other researchers in this volume, feels strongly that there is much yet to be known and to be made known about serendipity and how it can influence the paths we take

1 Introduction—A Science of Serendipity?

15

on the micro and macro scales, as individuals, in the moment, and as societies over time. While this volume, as a consequence of the diversity of approaches it includes, fails to offer a consistent or singular approach or even a central focus for serendipity science, it does illustrate the key threads that this field is beginning to weave together. Each of the ten chapters, as well as this introduction and the epilogue that bracket them, offers a different perspective, disciplinary focus and offers, perhaps most importantly, a collection of resources for those who are keen on studying serendipity themselves. We hope it brings something significant and as yet unsought to each of its readers. Editors: Samantha Copeland, Martin Sand, Wendy Ross

References Alcock, S.E. 2010. The stratigraphy of serendipity. In Serendipity: Fortune and the prepared mind, ed. Iain Morley and Mark de Rond, 11–26. Cambridge University Press. Bao, Z., and J. Yang. 2022. Why online consumers have the urge to buy impulsively: Roles of serendipity, trust and flow experience. Management Decision 60 (12): 3350–3365. https://doi. org/10.1108/MD-07-2021-0900. Barber, B., and R.C. Fox. 1958. The case of the floppy-eared rabbits: An instance of serendipity gained and serendipity lost. American Journal of Sociology 64 (2): 128–136. https://doi.org/10. 1086/222420. Copeland, S. (2019). On serendipity in science: Discovery at the intersection of chance and wisdom. Synthese 196 (6). https://doi.org/10.1007/s11229-017-1544-3 Hales, S.D. 2020. The myth of luck: Philosophy, fate, and fortune. Bloomsbury Academic. Jansz, S.N., M. Mobach, T. van Dijk, E. de Vries, and R. van Hout. 2022. On Serendipitous campus meetings: A user survey. International Journal of Environmental Research and Public Health 19 (21): 14504. https://doi.org/10.3390/ijerph192114504. Makri, S., and A. Blandford. 2012. Coming across information serendipitously—Part 1: A process model. Journal of Documentation 68 (5): 684–705. https://doi.org/10.1108/002204112112 56030. Merton, R. 1948. The bearing of empirical research upon the development of social theory. American Sociological Review 13 (5): 505–515. Merton, R.K., and E. Barber. 2004. The Travels and Adventures of Serendipity: A study in sociological semantics and the sociology of science. Princeton University Publishing. Napolitano, C.M. 2013. More than Just a simple twist of fate: Serendipitous relations in developmental science. Human Development 56: 291–318. https://doi.org/10.1159/000355022. Olshannikova, E., H. Pirkkalainen, T. Olsson, and J. Huhtamäki. 2022. What supports serendipity on Twitter? Online survey on the role of technology characteristics and their use. 25th International Academic Mindtrek Conference, 89–101. https://doi.org/10.1145/3569219.3569346 Qin, C., Y. Liu, X. Ma, J. Chen, and H. Liang. 2022. Designing for serendipity in online knowledge communities: An investigation of tag presentation formats and openness to experience. Journal of the Association for Information Science and Technology 73 (10): 1401–1417. https://doi.org/ 10.1002/asi.24640. Silver, S. 2015. The prehistory of serendipity, from Bacon to Walpole. Isis 106 (2): 235–256. https:/ /doi.org/10.1086/681977.

16

S. Copeland et al.

Smets, A., L. Michiels, T. Bogers, and L. Björneborn. 2022. Serendipity in recommender systems beyond the algorithm: A feature repository and experimental design. Proceedings of the 9th Joint Workshop on Interfaces and Human Decision Making for Recommender Systems (IntRS’22), Seattle, US, September 22, 2022 co-located with 16th ACM Conference on Recommender Systems (RecSys 2022), 46–66. ACM Press. http://ceur-ws.org/Vol-3222/paper4.pdf Sneddon, A. 2022. The pleasure of not knowing and the importance of serendipity in contemporary art practice. In The art of serendipity, ed. W. Ross and S. Copeland, 239–265. Palgrave Macmillan. Steffensen, S. 2016. Cognitive probatonics: Towards an ecological psychology of cognitive particulars. New Ideas in Psychology 42: 29–38. Thirugnanasambantham, P., S. Kovvali, A. Cool, Y. Gao, A. Sabag-Daigle, E.F. Boulanger, M. Mitton-Fry, A.D. Capua, E.J. Behrman, V.H. Wysocki, S. Lindert, B.M.M. Ahmer, and V. Gopalan. 2022. Serendipitous Discovery of a Competitive Inhibitor of FraB, a Salmonella Deglycase and Drug Target. Pathogens 11 (10): 1102. https://doi.org/10.3390/pathogens111 01102. Watson, B.M., and J. Bullard. 2022. I’m really happy when sometimes I end up on a shelf that seems sort of right:” Historians’ Reactions to the Cataloging and Classification of their Own Work. Proceedings of the Association for Information Science and Technology 59 (1): 335–345. https://doi.org/10.1002/pra2.757. Yaqub, O. 2018. Serendipity: Towards a taxonomy and a theory. Research Policy 47 (1): 169–179. https://doi.org/10.1016/j.respol.2017.10.007.

Samantha Copeland is an Assistant Professor in Ethics and Philosophy of Technology at Delft University of Technology. Co-founder and continuing co-chair of the Serendipity Society, Copeland has also recently edited The Art of Serendipity (2022, Palgrave-McMillan) with Ross, and has published in philosophy journals such as Synthese, Perspectives on Philosophy of Science, and in collections, on the rationality and epistemology of discoveries made by chance, as well as the impact of contingent environments on the success of potential serendipity. Her recent work has focussed on the relationships between ethics, serendipity and possibility. Wendy Ross studies the role of material serendipity in higher cognitive processes such as insight problem solving and creativity. She draws on a range of methods from eye-tracking and experimental psychology to focused cognitive ethnography. She has recently co-edited the collection on serendipity: The Art of Serendipity (Palgrave). She is Co-Chair of the Serendipity Society and Vice President of the Possibility Studies Network. In 2021 she was awarded the Frank X Barron prize by Division 10 of the APA. Martin Sand is an Assistant Professor of Ethics and Philosophy of Technology at TU Delft. In 2020, he was a member of the NIAS-Lorentz theme group on “Accountable and Explainable Medical AI” at the Netherlands Institute for Advanced Study. Before, he undertook a twoyear project on the topic “Moral Luck in Science and Innovation” as a Marie Skłodowska-CurieFellow. He is a member of the scientific advisory board of the Journal for Technology Assessment in Theory and Practice and an editorial board member of the journal Philosophy of Management.

Chapter 2

Exploration of “Serendipity” in the Mongolian Language Borchuluun Yadamsuren

Abstract Borchuluun Yadamsuren explores serendipity across linguistic boundaries with a focus on the Mongolian language. Despite its growing popularity in English speaking countries, there are many languages where the term has not yet entered into standard parlance. Mongolian is amongst the languages for which English dictionaries are rare and have been largely inaccessible to the public until some 20 years ago. How does the inexistence (or lack of knowledge) of a concept in a particular language affect speakers’ perception of it? Yadamsuren surveys various English-Mongolian translations and their slight semantic differences. Through interviews with various scholars and writers in the Mongolian language, she explores similar Mongolian concepts that may capture semantic nuances that current translations seem to miss. Yadamsuren easily repudiates the idea that a lack of linguistic conceptualization means there is no perception of serendipity. Instead, she shows that serendipitous experience in Mongolia is “practical and everyday” and “essential to survival as a nomad.” Mongolian fairytales and storytelling reveals equally interesting cases of serendipity; serendipities’ collective dimension is evident in how it is perceived within Mongolian culture. Building upon this, she makes a strong case for going beyond the classic examples of Western serendipitous discovery. Keywords Serendipity · Culture · Mongolia · Linguistics · Archaeology

When my Mongolian friends ask about my research, I have a hard time explaining what “serendipity” means because this word does not exist in my native tongue. Since I started exploring serendipitous news discovery many years ago, I have met only one Mongolian scholar who recognized the word “serendipity” without further explanation, recognizing the word from a training session on innovation organized by a Finnish scholar. In my Mongolian writing, I have used two imperfect approaches to explain the word serendipity. The first is transliteration, in which the English word is expressed B. Yadamsuren (B) Northwestern University, Evanston, IL, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 S. Copeland et al. (eds.), Serendipity Science, https://doi.org/10.1007/978-3-031-33529-7_2

17

18

B. Yadamsuren

in the corresponding characters of the Mongolian alphabet (“cependipiti”). With this approach, I found readers didn’t understand the meaning of this new word. Second, upon requests for me to find the appropriate Mongolian word, I used the dictionary definition of serendipity based on an English-Mongolian dictionary. But that translation meant serendipity needed to be written as an expression, not a single word, and was therefore too long and complex to use as I wanted. Merton and Barber (2004) examined the translations of serendipity in foreign languages. They found three general patterns: 1. As a foreign loanword (the Italian “serendipità”). 2. As a transliteration in which the word is expressed in the corresponding characters of another alphabet (the Russian “cependipnoctb”). 3. The approximate definitional expressions (such as in Arabic, Chinese, Hebrew, Japanese, Persian, Tamil, and Urdu). Of these three patterns, I had applied the second and third approach in my attempt to introduce the word serendipity in Mongolian language. However, in my ongoing research on serendipity in Mongolia, the question still remains: how does one study this complex phenomenon when there is no designated word in the given language? How else can one capture serendipity incidents, important scientific discoveries, and other important events associated with serendipity? These questions drove me to explore the root of the English word “serendipity” itself, specifically its adoption into the English language from Walpole’s invented term, based on a Persian fairy tale. I attempted to find a way to coin a simple term in Mongolian that captures the nuances of this complex phenomenon. This chapter presents my exploration based on my linguistic analysis of the Mongolian bilingual dictionaries and interviews with Mongolian writers, poets, linguists, and scholars.

Adoption and Diffusion of the Word Serendipity in English It is interesting to see how the word “serendipity” was introduced first to the English language and later to other languages. Merton and Barber (2004) stated that the cultural diffusion of the word serendipity must be traced through social channels, where the diffusion took place. They examined how the meaning of this word changed and the nature of these changes as they occurred in different segments of the intellectual world. Serendipity was coined by Horace Walpole, when he described this word in his letter to Horace Mann in 1754. He wrote how he made a critical discovery about the Capello family arms in an old book of Venetian arms. The most interesting part of his description was how this discovery was “almost of that kind” which he calls Serendipity, “a very expressive word.” Walpole further explained that Mann would understand this word better by the derivation than by the definition. He described “a silly fairy tale” called the three Princes of Serendip: as their highnesses travelled, they were always making discoveries by accidents and sagacity. For 75 years after it was

2 Exploration of “Serendipity” in the Mongolian Language

19

described by Walpole in his letter, the word serendipity remained dormant. Although his letter was published in Lord Dover’s edition of the Mann correspondence in 1833 and later in Peter Cunningham’s edition in ‘1857, the sheer existence of the word serendipity in print was not enough to give it the momentum necessary for diffusion. Only when Edward Solly and Andrew Lang were intrigued by the word serendipity did the word enter into the English dictionary landscape. They introduced the word to the readers of Notes and Queries in 1878. Merton and Barber (2004) explained that the word serendipity fit rather easily into the literary world by providing a “nice tidbit of esoteric information,” and that “its meaning aptly described many of their adventures in the search of such esoteric morsels” (p. 62). Thereafter, the use of this word serendipity was restricted to a small group of “dilettantes” until about 1900. Wilfrid Meynell extended the use of serendipity beyond the small circle of literary erudites between 1900 and 1935 (Merton and Barber 2004). Makers of dictionaries, popular essayists, a bookseller, a writer of detective fiction, and others accepted the word. The spread and diffusion of the word serendipity was accelerated immensely by its adoption in the scientific community in the 1930s. Walter B. Cannon, a professor of psychology at the Harvard Medical School, used the word serendipity frequently not only for the phenomenon of accidental discovery in science, but also to express a whole philosophy of scientific research. Its popularity grew in the scientific world and among journalists reporting on science to the general public.

The Role of Dictionaries in the Spread of the Word Serendipity According to Merton and Barber (2004), the authority of the dictionary is considered sufficient to establish claims to the proper usage of new words. However, if the standard dictionaries fail to list a word, knowledge of its history and etymology may be a substitute vindication of its usage (p. 71). The dictionary legitimizes by sheer inclusion the use of a word as part of standard literary language; it rules on legitimate meanings that may be attributed to such a word; most dictionaries may be expected to have at least some accurate information about the history and etymological derivation of a word that is included; and the dictionary rules on permissible spelling and pronunciation (p. 104). Until 1912, the word serendipity was not found in any dictionary. The first dictionary that included the word serendipity was the earliest version of The Century Dictionary and Cyclopedia, which was published over 1889–1891 in six volumes. This dictionary has been revised many times since then. A lengthy description of the word serendipity appeared in the two supplemental volumes of this dictionary. This description provided the most elaborate etymology of Serendipity as: “a humorous formulation with an allusion to dip, from Serendip, a form of Serendib, a former name of Ceylon…The happy faculty or luck, of finding by ‘accidental sagacity’

20

B. Yadamsuren

interesting items of information or unexpected proofs of one’s theories; discovery of thing unsought: a factitious word humorously invented by Horace Walpole.” (Merton and Barber 2004, p. 110). Prior to introduction of the word serendipity into common parlance, however, it was difficult to figure out the meaning of this exotic word— unless the discoverer was sufficiently familiar with the resources of the literary world to insert a query in Notes and Queries, in which case he could obtain full and accurate information about this new word in English. According to de Rond and Morley (2010), even though “serendipity” is one of the most queried words in the English dictionary, it is ironically one of the most difficult to translate. In typical use, it is interchanged with “luck,” “chance,” and “coincidence” (p. 1). Merton and Barber (2004) wrote that many people are simultaneously puzzled and intrigued by the nature of the word “serendipity,” which is “unfamiliar, etymologically puzzling, and seemingly expressive” (p.88). Other adjectives for this coined term include curious, strange-looking, extraordinary, and silly.

Exploring the Meaning of the Word Serendipity in Mongolian Language My exploration of the definition of serendipity and its meaning in Mongolian language started with dictionaries, particularly translations in multilingual dictionaries. Mongolian is one of the Altaic family languages. According to Hansakunbuntheung et al. (2011), it is spoken by nearly 8 million speakers living mainly in Mongolia, parts of China and Russia, and neighboring countries. Due to its historical and geographical background, Mongolian has several dialectal variations in its linguistic, phonetic, and graphic expressions, such as Khalkha, Oirat, Buryat, Khalmyc, and Inner Mongolian. In Mongolian, two different alphabets are used, Cyrillic and Mongolian. While the Cyrillic alphabet is mainly used in Mongolia, the Mongolian alphabet is mainly used in the Inner Mongolian Autonomous Region of China. Depending on the alphabet used, the writing system is also different in Mongolian (Chimeddorj and Fuji 2012). For this chapter, my exploration of the word serendipity has been limited to Khalkha dialect with Cyrillic alphabet-based writing. According to Kara and Krueger (2005), the earliest modern multilingual and bilingual dictionaries in the Mongolian language trace back to the Manchu period in the eighteenth century. Then, the dominant language featured alongside Mongolian was Tibetan, though bilingual dictionaries also included the Manchu language. TibetanMongolian and Manchu-Mongolian dictionaries were topical dictionaries, in which words were arranged in semantic groups, polyglot dictionaries of Buddhist terms, like the Pentaglot (Sanskrit, Tibetan, Mongolian, Manchu, and Chinese), or the “FiveLanguage Mirror of the Manchu Language” (in Manchu, Tibetan, Mongolian, Turki,

2 Exploration of “Serendipity” in the Mongolian Language

21

and Chinese) (p. 283). Clark and Walravens (2006) put together the most comprehensive list of Mongolian, Manchu-Tungus, and Tibetan dictionaries. Beyond these records, I was not able to find more in-depth research on the history of multilingual or bilingual dictionaries in Mongolia, especially with Western language combinations. In the twentieth century, Mongolia was a satellite nation of the former Soviet Union. Russian language and culture were the dominant and only accepted foreign influences in Mongolia, so much so that the Russian Cyrillic alphabet officially replaced Mongolian Uyghur script in 1941 (Dovchin 2020). The Cyrillic script remained the official orthographic system of Mongolia even post-socialism, into the present day. The English language and Western culture were considered to be “capitalists’ commodities” (Dovchin 2020). After 70 years of communist rule, Mongolia’s socioeconomics, politics, culture, and language experienced dramatic changes, coinciding with and often driven by the shift to a free market economy. French, German, Korean, Japanese, Chinese, and—most of all—English competed with the oncedominant Russian as foreign sociolinguistic influences in Mongolia. Despite these transformations, Russian influence remains strong. My own experience of learning English was based on English-Russian dictionaries because English-Mongolian and Mongolian-English dictionaries, to the extent they existed at all, were not available to the general public. The most popular English Mongolian dictionary after the collapse of the Soviet Union was published in 1998 (Altangerel et al. 1998). Today, the Library of Congress catalog lists 16 dictionaries in response to a query of “English Mongolian dictionary” and 14 dictionaries in response to “Mongolian English dictionary.” However, these lists contain all sorts of dictionaries, including specialty dictionaries for certain professions and fields, such as health management, geography, chemistry, cement and concrete technology, hotel management, and oil and oil product processing and utilization. While the catalog of the National Library of Mongolia shows several multilingual dictionaries, most of these dictionaries are rare books and not available for public. For this limited research, I checked the word serendipity in the following dictionaries: Troxel (1953), Altangerel et al. (1998), Amarsanaa (2006), in addition to a few further dictionaries used by an interviewee. The only dictionary that contains the translation of the word “serendipity” was an online bilingual dictionary Bolor toli, which means Crystal dictionary. Unfortunately, there is no research on how this dictionary was created, despite its popularity. According to Mongolian linguist Dr. Saruul-Erdene Myagmar (personal correspondence 2021), the Bolor dictionary was likely created based mainly on two other dictionaries: Lessing and Haltod (1960) and Hangin (1970). According to the Bolor online dictionary, the word “serendipity” has four different meanings: • Canamcapgyu aztau toxioldloop xiuc϶n n϶϶lt [focused on discovery]: a literal, word-by-word translation yields “a discovery made by an accidental, lucky occurrence,” or, paraphrased, “a discovery by happy accident.” This definition embeds “luck” and “chance” into the meaning of the word. Canamcapgyᴎ

22

B. Yadamsuren

[adjective: lots of meanings, including “accidental”] aztaᴎ [adjective: “luck”] toxioldloop [“by chance”] xiᴎc϶n n϶϶lt [“discovery”] aztaᴎ toxioldloop [“by lucky chance”]. • canamcapgyu n϶϶lt xiux xault [focused on search-process]. Translation: “search for accidental discovery.” • n϶϶lt xiuc϶n aztau avbЯac [focused on personal quality for serendipitous discovery]. Translation: “lucky talent for discovery,” or “lucky skill to make discovery.” • aztaugaap xiuc϶n n϶϶lt—“discovery that happened by luck.” Four of these translations contain the word “n϶϶lt,” or “discovery.” “H϶϶lt” can also be translated as breakthrough, development, opener, and revelation. Two of these definitions include the word “canamcapgyᴎ,” which means “accidental.” According to the Mongolian-English version of Bolor dictionary, the word “canamcapgyᴎ” has several meanings: unpredictable, unmediated, accidental, fortuitous, adventitious, precarious, occasional, incidental, mechanical, ineffective, fluky, facultative, unrehearsed, unintentional, unhoped-for, unexpectant, undeliberate, unadvisable, sudden, promiscuous, casual, without realizing. Another important word is “aztaᴎ,” an adjective that means “lucky.” Other translations include happy, fortunately, providential, white, weirdly, successful, satisfactory, chancy, fluky, felicitous, clever, merciful, lucky, successfully, thrifty, and joyfully. In the Mongolian language, the combination of two words sometimes has a different meaning than the individual words. The combination of “aztaᴎ toxioldloop”: fortunately, blessed, blessed with good luck, well-off, successfully, providential, happy, happily, well-off, white, weirdly, turn-up, thrifty, successful, satisfactory, merciful, lucky, joyful, hammy, go-getter, fortunate, fluky, felicitous, clever, and chancy. While the combination of the two words, “aztaᴎ toxioldol” yields the following: stroke of luck, fortune, fluke, chance, break, and “happy end.” The Mongolian definitions capture that the English word serendipity is often used synonymously with “luck,” “chance,” and “coincidence.” In addition, the Mongolian translations also capture that a serendipitous experience is a happy one, yielding positive results, such as what Merton and Barber (2004) described as “interesting items of information or unexpected proofs of one’s theories” (p. 110). However, Merton and Barber also describe serendipity as “discovery of thing unsought” (p. 110). By contrast, none of the Mongolian definitions suggest the happy or accidental discovery to be “unsought.” In fact, the third definition, canamcapgyu n϶϶lt xiux xault, focuses on the personal ability, skill, or talent to make such discoveries, or indeed to encounter serendipity. While Mongolian definitions overlap with English ones in that serendipitous discoveries are made by accident or luck, they may also suggest a lifestyle or culture that fosters a serendipity “mindset,” or a general willingness and desire to make such discoveries, “accidental” though they may be. To further explore how Mongolians experience serendipity, even if they lack the word for it, I interviewed scholars, linguists, and writers.

2 Exploration of “Serendipity” in the Mongolian Language

23

Coining the Term for Serendipity in Mongolian Mongolians have a strong oral tradition, with oral histories passed down from generation to generation. Unlike other research methodologies recorded in the existing literature on serendipity, my methodology was one centered on storytelling. I spoke with Mongolian scholars and writers using the translated short version of the original story about the Princes of Serendip as told in the Travels and Adventures of Serendipity (Merton and Barber, 2004), as well as the stories of the discovery of Fleming’s accidental discovery of penicillin and Newton’s apple, falling on his head to inspire the idea of gravity. These short stories helped me to explain the meaning of the word “serendipity.” The word serendipity was initially unknown to the interviewees, and all of them were intrigued by the sound and meaning of this word. But true recognition of serendipity occurred through the story about the genesis of this word and connection to the original fairy tale in Persian. Because Mongolian culture is deeply embedded with oral history, fairy tales are very popular in our culture. Upon hearing a story or two, interviewees were eager to learn this new word, thought of similar Mongolian fairy tales and other stories of serendipitous discovery, and even came up with new words to describe serendipity in Mongolian. Knowing the story of how Walpole introduced the word serendipity into the English language, it was interesting to think of how one of the new words for serendipity in Mongolian might be injected into the Mongolian lexicon. Several of my interviewees seemed intuitively to have compatibility with the Mongolian language in mind when they constructed words that might be appropriate for the word serendipity in Mongolian. For example, when linguist Dr. Saruul-Erdene Myagmar created the word “serenget” (c϶p϶ng϶t), his use of “ser” as the first syllable simultaneously fulfilled two purposes. First, it maintained the similarity of the new word to the English “serendipity.” Second, it represented that serendipity was about discovery, because “ser” (c϶p) means to wake up or to be awake in Mongolian. While discussing the meaning of the word serendipity over the course of our interview, we discussed whether serendipity was about process or quality. Dr. Saruul-Erdene Myagmar thought of the word “serenget” as representing the quality of serendipity, as an adjective. Noun and verb formulations were also possible from the word, such as “c϶p϶ng϶tiv” (serengetiv), with the same “c϶p϶ng϶t” root. By contrast, the word that the famous Mongolian writer Tsendjav Dolgor came up with during our interview focused on the process of serendipity. His word, “serdevtiim” (c϶pd϶vtiᴎm) interestingly preserved “ser” as the first syllable of the word, while “dev” means progress or development, and “tiim” means yes. Put together, the three syllables signify a process, rather than quality, of awakening to progress.1 1

It is interesting to note that while both scholars created these words for serendipity during our interviews from existing Mongolian words and with usability in Mongolian in mind, Google Translate is able to give a translation for the newly created words. Dr. Saruul-Erdene’s “c϶p϶ng϶t” is translated as “awake,” while Dr. Tsendjav’s “c϶pd϶vtiᴎm” is translated as “I was shocked.” Inputting “serendipity” for an English to Mongolian translation, however, spits out “dapyy baᴎdal,” which translated back into Mongolian is “humility.”.

24

B. Yadamsuren

With both words, interesting questions remain: how and who can introduce these words into the Mongolian language? Which, if either, might the public easily understand and accept? When Merton first encountered the word serendipity in the Oxford English Dictionary, he noted that the word was strange-looking and melodioussounding. Ought a Mongolian version of the term capture the melodious nature of the English word? Mongolian interviewees seemed immediately intrigued by the novelty and sound of the word serendipity—is this melodious characteristic of the English word unique and universal? Though it took three quarters of a century for the English word serendipity to make its way from its initial use in print to more general, diffused use, the ways in which Mongolian scholars and writers had experienced serendipity, even without a Mongolian term for the word, suggest that a properly coined term may spread more quickly in Mongolia. As compared to the western experience with serendipity, in which experiences and stories of serendipity were described as silly (Walpole) and the word itself curious, strange-looking, extraordinary, and, again, silly, Mongolians were quick to think of personal and historical examples of serendipity. In short, in Mongolian, regardless of the word used to describe serendipity, such experiences and the term itself were not regarded as silly or extraordinary, but in fact practical and everyday. To be observant of one’s surroundings was and remains essential to survival as a nomad. Noticing unexpected items or events, quickly adapting to them, and spreading the news to others is a part of Mongolian culture, and features prominently in storytelling. For example, some interviewees reexamined fairy tales that featured “sohor az” (blind luck) to discern whether serendipity, had such a word existed, might have been a better fit. These fairy tales described varying degrees of chance encounter, sagacity, and coincidence. Merton and Barber (2004) emphasized that careful word choice is an occupational necessity for professional scholars and writers (p.78). Yet Mongolian authors throughout history have lacked a word that precisely describes serendipity. Thus, perhaps only because the word does not exist in Mongolian, what might otherwise be famous examples of serendipity have gone thus far overlooked.

Serendipity Case Study in Contemporary Mongolia From my interviews, one such example comes immediately to mind. Dr. Bayarsaikhan Jamsranjav is a preeminent Mongolian archaeologist, with whom I became acquainted through a serendipitous meeting at a conference. He worked on the Smithsonian Institution’s multi-disciplinary American-Mongolian Deer Stone Project team2 to study archaeological, cultural heritage, and ecological resources across Mongolia. A major component of Dr. Bayarsaikhan’s research is focused on deer 2

William W. Fitzhugh and George Michael Leader, “Mongolia Deer Stone Project: Field Report 2005” (2006), https://repository.si.edu/bitstream/handle/10088/30976/ASC_NL_21_2014%20F INAL_April-p14-17.pdf.

2 Exploration of “Serendipity” in the Mongolian Language

25

stones, ancient Mongolian monuments of carved stone. According to Fitzhugh and Leader (2006), the plinths are from the Bronze and Early Iron Ages, stand between 20 cm and four meters in height, and mark burial sites. Scattered through central, northern, western, and northwestern Mongolia and beyond its borders, into Buryatia, Tuva, Mountain Altai of Russia, Eastern Kazakhstan, and Xinjiang of China and numbering in about 1600 in total, these stones are conspicuous, with figures of deer and other animals, such as horse, boar, ibex, and feline, as well as earrings, necklaces, a belt, mirror, and set of weaponry carved into the stone (Bayarsaikhan 2017). Despite, or perhaps because of, their widespread presence, the deer stones are under-researched, representing an untapped wealth of archeological knowledge. In 2017, Dr. Bayarsaikhan continued his ongoing joint research with the University of Pittsburg’s “Northern Mongolia” project on a trip to the Ulaan-Uul soum of Huvsgul aimag. He went out of his way to detail in documentation and photograph the deer stones, which he had wanted but had been unable to do on a previous trip. Next to the deer stones, he found significant looting, which frustrated and worried him. Dr. Bayarsaikhan had been working with both local herders and an international team to combat looting in the region, which, among other issues, leaves previously frozen remains and artifacts exposed to melting and degradation.3 That particular day, in the countryside of the least densely populated country in the world, there was no one to ask about the looting. So Dr. Bayarsaikhan ventured to nomadic gers (Mongolian yurts) in the distance. Knowledge sharing in Mongolia, especially among nomads, occurs primarily through oral traditions, and Dr. Bayarsaikhan was lucky to find Mrs. Battur, an old woman in one of the gers who was willing to talk to him. In conversation about the looting Dr. Bayarsaikhan had seen next to the deer stones, he volunteered his affiliation with the Natural Museum of Mongolia. For the nomadic woman, this information served as a trigger element for her own serendipitous experience, in which she knew of looting at a nearby mountain that needed investigating but had not known to whom to pass on that information. A marker of a serendipitous experience is that it compels the discoverer to want to share their finding with someone, but the woman had thus far been unable to share what she knew, either from lack of foot traffic or likely from fear of contributing to the looting by allowing what she knew to reach the wrong ears. When this woman happened to find out that the man standing before her was an archaeologist, she volunteered that there was a nearby mountain, Khorig Uul, which had suffered from recent looting and from which had surfaced burial remains, silk, and cloth. Children and livestock had been fearful of venturing near the mountain ever since, she told him. Because Mongolians believe that mountains, rivers, and other natural elements have religious significance, she worried that there could be consequences for the unearthed burial site. For both Dr. Bayarsaikhan and the woman, conversation with each other triggered events of serendipity—without Dr. Bayarsaikhan’s curiosity about the looting by the deer stones, the woman would not have found an appropriate scholar to whom to 3

https://www.shh.mpg.de/1439414/mongolia-vessels-and-silks-ventresca-miller.

26

B. Yadamsuren

pass her knowledge of Khorig Uul, and without her knowledge and willingness to share, Dr. Bayarsaikhan would never have made his way to Khorig Uul. As Dr. Bayarsaikhan left the woman’s ger and went to investigate the site she described, he realized that the name of the mountain itself, Khorig Uul, was significant—“khorig” directly translates to “forbidden.” In fact, the ancient history of Mongolia reported that “khorig” and “ikh khorig” meant the sacred, hidden, and royal burial sites (Rash¯ıd al-D¯ın Fad.lull¯ah Hamad¯an¯ı 1383). The names that Mongolian nomads have given physical landmarks often mark a point of significance—great natural resources, for instance, abound under mountains aptly named “Golden Hill” or “Turquoise Head.” In this case, Khorig Uul, or Forbidden Hill, is thought to have been named as such since the time of Chinggis Khaan, likely because of its role as a significant burial site. Immediately after recognizing the historic burial by his serendipitous discovery, Dr. Bayarsaikhan collected the remaining exposed items from looted burial area. The challenge was to protect the area from further looting while he awaited special permission to dig further. From the initial burial site, they found that the human remains were those of a “baatar,” or warrior, from 700–800 years ago, and that many items had accompanied him in the same burial. Once they had the permits in hand, in 2018 and 2019, Dr. Bayarsaikhan’s team expanded their search area and found 67 other looted burials of aristocratic families from the Mongolian Great Empire from 13th to fourteenth centuries. In 2019, they found four giant urns from several different sites with still-frozen clotted cream and yellow butter,4 which Mongolians continue to use today for the Buddhist tradition of lighting candles for the dying and deceased (Bayarsaikhan and Clark 2020). The first urn had a dragon design, and held clotted cream; the next was a black urn with yellow butter; another was brown enamel; the remaining one was a simple mud jar (Figs. 2.1 and 2.2). Accompanying the urns were burnt remainders of thick incense cores, made of organic matter, indicating a ceremony of lighting the incense within the urn at the time of burial.5 Valuable artifacts with historical significance to the Mongols also emerged, made of gold, silver, iron, silk, bones, hide, and cork.6 In 2019, Dr. Bayarsaikhan and his team also found several golden and silver relics across three burial sites that depicted the sun and moon together (Figs. 2.3 and 2.4), which had never before been discovered from that era. That symbol, found on the Mongolian flag today and known to be used by the Huns, had until then never been found physically present along with the remains of a Mongol royal family (Bayarsaikhan and Egiimaa 2008; Miller et al. 2008; Bayarsaikhan 2021). These findings most significantly provided an important connection between the Mongols with the Huns,7 similar in their worship of the sky and other religious practice. For example, the Huns are said to have worshipped the 4

https://www.montsame.mn/en/read/197987. The Huns had a similar burial tradition, according to historic sources. See footnote 7 for discussion of the use of “Huns.”. 6 https://akipress.com/news:623822:800-year-old_vases_containing_frozen_clotted_cream_and_ yellow_butter_found_from_glacial_in_Mongolia/. 7 Though I use the term “Huns” throughout for internal consistency, there is heated debate among historians, linguists, and archaeologists about the accuracy of, preference for, and differences 5

2 Exploration of “Serendipity” in the Mongolian Language

27

Fig. 2.1 Vase 1

Fig. 2.2 Vase 2

sky in the morning and the moon at night, and Huns of higher rank were found to be buried with the sun and crescent moon made of gold and metal, while those of lower rank had the same symbols depicted in only metal. The most cited historical document on the role of sun and moon in the worship of the Huns is Sina Qian (1959). The initial happenstance of finding the looting next to the deer stones became the trigger in a series of serendipitous events that led Dr. Bayarsaikhan to unearthed between various terms for the pre-Mongol nomadic people discussed here. According to Prof. Atwood (2012), “Hun” is the Greek and Latin term for the nomadic group in East and Central Europe, and “Xiongnu” (匈奴) is the modern Chinese pronunciation of the ancient Chinese name for nomads in Mongolia. Though scholars seem to agree that the names are related, the people each name represents are not necessarily the same or related. Mongolian scholars, including Dr. Bayarsaikhan (2021), use the term “Hunnu” (“Xynny”). Other (possibly) related terms include the Sanskrit “Huna,” the Greek “Khonai,” “Ounna,” and “Ounnoi,” and the Latin “Hunni.” (Atwood 2015, 2012).

28

B. Yadamsuren

Fig. 2.3 Moon

Fig. 2.4 Sun

remains at Khorig Uul, from which he discovered the first artifact that connects the Huns to the Mongol Empire. Though Mongolian tradition claims the Huns as ancient ancestors of the Mongols, no physical object had previously proven that connection. Discovery of the link between these two empires constituted a collective serendipitous experience, possible only through collective information encountering. In Mongolian culture, nomads are especially attuned to their environments, from natural weather conditions to foreign disturbances of the land. That animals, who Mongolians consider to be especially sensitive to the land, were suddenly unwilling to approach the Khorig Uul cemetery area was an important trigger element that a nomad had noticed and passed on to other nomads. That information eventually made its way to the nomadic woman with whom Dr. Bayarsaikhan happened to converse about a different site of looting. From a large store of inputs about the land, collectively gathered by the many nomads in the region, the woman was able to match her trigger of meeting Dr. Bayarsaikhan with the particular information a different nomad had encountered about Khorig Uul. The case study offers several important realizations about serendipity in the Mongolian context: that the 800-year-old sun

2 Exploration of “Serendipity” in the Mongolian Language

29

and moon artifact that links the Huns that existed 2000 years ago to the Mongols was only possible through a collective serendipitous discovery; that Mongolian culture may be particularly conducive to collective serendipitous discovery; and that Mongolian nomads may themselves live lifestyles of serendipity, encountering and making decisions based on small serendipitous discoveries on a daily basis.

References Altangerel, D., Z. Khaffmann, and P. Chimg˙ee˙ . 1998. Angli-Mongol tol'. New. Ulaanbaatar: Interpress Pub. & Print. Co. Amarsanaa, L. 2006. Oxford-Monsudar English-Mongolian dictionary = Oksford-Monsudar anglimongol tol', 1st ed. Ulaanbaatar London: Monsudar Pub.; Oxford University Press. Atwood, C.P. 2015. The Qai, the Khongai, and the names of the Xi¯ongn´u,” Int J Eurasian Stud 35. Atwood, C.P. 2012. Huns and Xi¯ongn´u: New thoughts on an old problem in Dubitando: Studies in history and culture in honor of Donald Ostriwski 27. https://repository.upenn.edu/ealc/19/ Bayarsaikhan, J., and J. Egiimaa. 2008. Taxiltyn xotgopyn 64-p bylxny maltlaga cydalgaany yp dyn’, Talyn ix ϶z϶nt gyp϶n- Xynny. International symposium in celebration of the 10th anniversary of “MON-SOL” project. Bayarsaikhan, J. 2017. Deer Stones in the Northern Mongolia. InterPress printing. Bayarsaikhan, J., and J.K. Clark. 2020. Salvage archaeology in Northern Mongolia: Surprising finds from the Mongol empire. Paper presented at the XIV Mongolian Studies Conference “Art and Culture among the Mongols,” Washington DC Bayarsaikhan, J. 2021. Golden and silver images of the sun and Crescent Moon from the Tombs of Mongol Aristocrats on Khorig Mountain and the belief and ritual continuity of the Huns and Mongols. Magazine of Chinggis Khaan Museum, I (Fasc. I-XII): 5–19. Bolor Toli dictionary: http://www.bolor-toli.com/dictionary/word?authenticity_token=dtFxEu QoMVW6IIqI77WU%2BldLwfBJqxWWyPbmkAEvus8%3D&selected_lang=4-1&search= serendipity) [cite properly]. Chimeddorj O., and A. Fujii. 2012. Enhancing lemmatization for Mongolian and its application to statistical machine translation. Paper presented at the 10th Workshop on Asian Language Resources, Mumbai, India. Clark, L.V., and H. Walravens. 2006. Bibliographies of Mongolian, Manchu-Tungus, and Tibetan dictionaries. Wiesbaden: Harrassowitz. Dovchin, S. 2020. Translingual English, Facebook, and gay identities. World Englishes 39 (1): 54–66. https://doi.org/10.1111/weng.12445. Fitzhugh, W., and G.M. Leader. 2006. Mongolia Deer Stone Project: Field Report 2005. Arctic Studies Center. Hangin, J.G. 1970. A concise English-Mongolian dictionary. Bloomington: Indiana University. Hansakunbuntheung, C., A. Thangthai, N. Thatphithakkul, and A. Chagnaa. 2011. Mongolian speech corpus for text-to-speech development. Paper presented at the 2011 International Conference on Speech Database and Assessments (Oriental COCOSDA), , 26–28 Oct. Kara, G., and J.R. Krueger. 2005. Books of the Mongolian nomads: More than eight centuries of writing Mongolian (1st English ed.). Bloomington, IN: Indiana University, Research Institute for Inner Asian Studies. Lessing, F., and M. Haltod. 1960. Mongolian-English dictionary. London and New York: Routledge. Merton, R.K., and E. Barber. 2004. The travels and adventures of serendipity: A study in sociological semantics and the sociology of science. Princeton, NJ: Princeton University Press. Miller, B.K., J. Bayarsaikhan, T. Egiimaa, and C. Lee. 2008. Xiongnu Elite Tomb Complexes in the Mongolian Altai: Results of the Mongol-American Hovd Archaeology Project, 2007. The Silk Road 5 (2): 27–36

30

B. Yadamsuren

Rash¯ıd al-D¯ın Fad.lull¯ah Hamad¯an¯ı. 1383. J¯amic al-Taw¯ar¯ıkh, J. II. Tehran. Rond, M.D., and I. Morley (eds.). 2010. Serendipity: Fortune and the prepared mind: Cambridge University Press. Troxel, D. 1953. Mongolian vocabulary (Modern Khalkha Language), Mongolian-English, EnglishMongolian. Washington, DC: Govt. Printing Office.

Borchuluun Yadamsuren is a research facilitator at Northwestern University, advising up to 20 high level industry-supplied data science projects with students in the Master of Science in Analytics program. Her current research focusses on data analytics, natural language processing, machine learning, serendipitous information discovery, and social media. Previously, she worked for the Information Experience Laboratory of the University of Missouri as user experience researcher. She has published seminal work on incidental exposure to online news.

Chapter 3

Serendipity and Knowledge Organisation Toby Burrows

and Deb Verhoeven

Abstract Toby Burrows and Deb Verhoeven return to Walpole’s original story of the Three Princes of Serendip to remind us that aside from sagacity, it is the various places that the princes travel to that afford serendipity. Consequently, they contemplate the idea that serendipity might be more than a mere passive source for finding the unsought. The design of physical objects and spaces can either afford values or undermine them. Burrows and Verhoeven investigate whether and how the value of serendipity has been realized –consciously or unconsciously—in the structures and designs of libraries and other physical collections. The ways in which such physical spaces afford serendipity is highly diverse. Libraries’ fostering of serendipity is at least commonly assumed to be more pronounced than in virtual spaces, where the guiding design principle seems to be efficiency, or quickly finding what one seeks. This serves Burrows and Verhoeven as a backdrop to investigate whether virtual spaces can possibly foster serendipity to the same degree as physical spaces. In the end, Burrows and Verhoeven plea for a diminished focus on efficiency and a return to the foundational story of serendipity to guide the design of virtual search engines in the future.

Organising and displaying knowledge has been a key purpose of the Western tradition of libraries since the era of the Ancient Greeks. With the emergence of museums in early modern Europe, this also became a central principle in the collection and display of scientific and historical objects and artefacts. But there is long-standing evidence for unexpected discoveries being made within these environments, especially of T. Burrows University of Oxford, Oxford, UK University of Western Australia, Perth, WA, Australia T. Burrows e-mail: [email protected] D. Verhoeven (B) University of Alberta, Edmonton, AB, Canada e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 S. Copeland et al. (eds.), Serendipity Science, https://doi.org/10.1007/978-3-031-33529-7_3

31

32

T. Burrows and D. Verhoeven

material which was unsought and unknown but proved to be relevant or stimulating for researchers. In this chapter we examine the interplay between these traditional physical environments and more recent digital environments, and consider the ways in which the digital world might encourage this kind of serendipitous encounter. At the heart of this is the notion of browsing or “information meandering”—the ability to traverse these collections in ways which cut across the grain of their physical and conceptual structures. In our view, such meaningful structures are an important element of serendipitous encounters. Serendipity can occur at many points on the continuum from completely random to completely structured knowledge environments. But our hypothesis is that it is most likely to be encountered where there is enough structure to encourage relevance, but not so much as to suppress the unexpected and unpredicted. Another crucial vector is the existence of multiple perspectives on the concepts and relationships embedded in a body of research materials, whether these are physical or digital. Serendipity will also arise from the ability to meander among these different understandings and conceptualisations of the world and knowledge. Our own personal serendipitous encounters began in 2012 when we crossed paths during the process which led to the development of the Humanities Networked Infrastructure, or HuNI. HuNI encapsulates many of these ideas about equilibrium of structure, randomness, and multiple perspectives that inform our approach to serendipity in digital information platforms. It is designed to encourage information meandering and to provide a digital environment hospitable to the kind of serendipitous encounters which lie at the heart of humanities research.

Serendipity and the Physical Library One of the most interesting critiques of how knowledge is encountered in the digital world centers on a melancholy memory of the traditional, physical library. Often expressed in terms of a “loss of serendipity”, this despondency has become a recurring trope about the decline of the physical library and the migration to electronic books and journals. This nostalgia for the serendipitous effects of the traditional library is usually accompanied by warnings about what has been lost in the digital world. In an op-ed piece in 2006, journalism professor William McKeen lamented that “the concept of serendipity is endangered” because of our reliance on search engines. He contrasted this with browsing the shelves in a library (McKeen 2006): Think about the library. Do people browse anymore? We have become such a directed people. We can target what we want, thanks to the Internet. Put a couple of key words into a search engine and you find – with an irritating hit or miss here and there – exactly what you’re looking for. It’s efficient, but dull. You miss the time-consuming but enriching act of looking through shelves, of pulling down a book because the title interests you, or the binding. Inside, the book might be a loser, a waste of the effort and calories it took to remove it from its place and then return. Or it might be a dark chest of wonders, a life-changing first step into another world, something to lead your life down a path you didn’t know was there. Same thing goes with bookstores.

3 Serendipity and Knowledge Organisation

33

There are many similar observations about libraries and their innate serendipity— enough to suggest that this is something more than just a nostalgic cliché. There is a fascinating echo here of the approach to building and office design espoused by Steve Jobs and others in the world of corporate innovation (Silverman 2013). Jobs and his imitators were aiming, deliberately, for shared physical spaces which promoted serendipitous encounters between people, with a view to encouraging new ideas (the “water cooler” effect). It is interesting to think that libraries might have anticipated such manoeuvres by many decades, if not centuries. In one sense, their impact directly paralleled the effect for which Jobs and others were striving. Libraries often used to be good places to run into other scholars and students with a shared thirst for knowledge, and even today the reading rooms for rare books, manuscripts and special materials tend to encourage these kinds of encounters.1 The biggest research libraries often encourage readers’ groups in these settings, like the National Library of Australia’s Petherick Readers group, which brings together users of its Rare Books Room. The University of Western Australia Library had its Scholars’ Centre, which was specifically intended to encourage serendipitous social encounters between humanities researchers (Burrows 1994). The lament of William McKeen and others like him is not primarily for the social environment of libraries, however. It has more to do with the ways in which libraries are thought to have promoted unexpected connections between the reader and various different expressions of knowledge—primarily in the form of books. What is there about the physical layout of library collections that might produce such an effect? For a start, these stories must refer to libraries which allow users to have direct access to their shelves, and not to those libraries where most of the stock has to be retrieved from closed stacks on request. They also appear to be talking about libraries which arrange their shelves in a meaningful way. The serendipity lament generally presupposes some form of coherent and non-random arrangement of the library collection, usually based primarily on the content of the materials. Arranging the collection by other factors—like size or date of acquisition—may still produce serendipitous encounters, but they are likely to have a higher level of randomness and to lack the meaningful relationships given by a content-based structure. Arranging books by their content ensures the proximity of items which are sufficiently similar in focus to be worth encountering, even if they are not what is being specifically sought. For much of the twentieth century, the most common method of arranging libraries by subject was through the use of the Dewey Decimal Classification (DDC). The DDC’s weaknesses and biases are well-known (Berman 1980). Its origin in the late nineteenth century means that its ability to cope with manifold, or new and changing fields of knowledge has become increasingly limited, despite regular revisions. It gives a disproportionate emphasis to particular histories such as the United States or the Christian religion for instance.

1

It is worth noting, however, that one person’s happy encounter in the library stacks may equally be another person’s harassment or assault (von Stackelberg 2018).

34

T. Burrows and D. Verhoeven

Many culturally specific alternatives to the DDC have emerged to address these limitations from as far afield as the Xwi7xwa Library (pronounced whei-wha) at the University of British Columbia in Vancouver in Canada to the community library of Galiwin’ku in East Arnhem Land in Australia. These libraries have rearranged their shelves so they can be more meaningfully navigated by Indigenous community users. Research on the serendipitous capability of these arrangements for Indigenous library users has not yet been reported. The conceptual and political failings of the DDC demonstrate its limitations as a tool for organising human knowledge. But these very weaknesses and problems do not necessarily work against serendipity; in some ways, organisational inexactness and ambiguity may even help to encourage information meandering and the possibility of unexpected encounters and connections. In parallel with browsing the open shelves of the physical library, arranged by subject, readers were usually able to browse through library catalogues in the form of card collections arranged by classification number. These “shelf lists” presented a surrogate walk-through of the shelves—a proxy representation, in abstract form, of the hierarchical subject groupings embodied in the physical collection. At their most sophisticated, they could provide a walk-through in which a book could appear in more than one place in the conceptual framework—achieved by having multiple cards with different classification numbers for the same book and sometimes an additional set of “cross reference” cards. Subject-based arrangements of library collections already had a long history before Dewey published his classification scheme in 1876. Many libraries—even before the invention of printing—were arranged into subject categories, however loose and rough these may have been. Medieval college libraries were often arranged to correspond with the four university faculties: theology, law, medicine and arts. The Syon Abbey library catalogue of 1526 grouped books under subject classes: grammar and classics, medicine and astrology, philosophy, and so on (Jayne 1956:33). Trinity College Dublin, in the seventeenth century, employed a classification system which divided books into two main categories—Theology and History—with sub-classes for each of these (Fox 2014).

Serendipity and Information Retrieval This venerable history of arranging libraries so they could be browsed by subject was increasingly seen as ineffective and obsolescent by information retrieval researchers and specialists from the 1960s onwards. In a textbook published in 1987, Jennifer E. Rowley describes it, with a degree of condescension, as “a very unsophisticated information retrieval device” (Rowley 1987:419–420). She wonders why it is so popular and widely used, given that it only allows a book to be placed in one location, where it can only be approached through one set of characteristics. Her only explanation, apart from “the simple pleasure of browsing”, is the ability it confers to examine

3 Serendipity and Knowledge Organisation

35

items for their distinctive physical features. She does not mention serendipity or unexpected discovery. In her discussion of information retrieval by subject, Rowley focuses on keyword searching and subject indexing. The goal is “effectiveness in retrieval”, which can be measured through precision and recall (Rowley 1987:167–183). This efficiency driven approach gradually rose to a dominant position with the diffusion of computerbased retrieval systems, the addition of full-text searching, and its application to the Web. The ever-growing sophistication of search algorithms, together with relevance ranking for results, has resulted in the dominance of search and the marginalisation of other approaches to knowledge discovery. This has reached the point where the keyword search process across a large corpus of words, typified by Google’s ubiquitous search engine, is assumed to be the definitive answer to information retrieval. Despite evidence to the contrary (Poulter 1999), browsing by subject has been generally dismissed as a far less efficient means of locating relevant material. This is where the other side of the nostalgic story about the serendipity of the browsable physical library comes into play: the perceived loss of serendipity in digital environments. The success of Google’s search box has had a profound effect on the design of other knowledge discovery systems. This is especially so in the world of library catalogues and databases, which have almost universally adopted the “single search box” as their gold standard. A typical example is the British Library’s “Explore Further” search box, which is featured on the Library’s Web site. This is a Googlelike search across a whole range of external catalogues and indexing databases. The Primo software which powers the British Library’s search box is widely used in the world of academic and research libraries, some of which merge the external search with a search of their own catalogue. In its purest form, the “single search box” assumes that the only starting-point for a search is a word, term, or named entity. It is entirely reliant on a linguistic framework and on keyword indexing, and no other avenues for the initial exploration of the underlying body of knowledge are offered. Underpinning the initial search box is a range of techniques for working with the results of this search, including sophisticated algorithms for ranking the results and determining their degree of relevance, and functions for filtering the results by various facets and for sorting by date, title, author and other similar criteria. It is certainly possible to make unexpected discoveries from a list of search results in Google, Primo or other similar search-based systems, and to use a document or page returned by these services as a jumping-off point for serendipitous exploration. Tammera Race evaluated four leading library “resource discovery” systems against a number of “features that support serendipity”, including faceted browsing, hypertext links, suggestions or recommendations, and changing search strategies (Race 2012:145). But this differs from the perceived serendipity of the physical library in a fundamental way: some kind of keyword search needs to be carried out first, and any accidental resemblances are primarily linguistic rather than subject-based or classificatory. Keyword searching systems like Google and Primo systematically elide the distinction between the linguistic and the conceptual, and are built on the premise that strings of characters are reliable analogues for concepts. This ignores

36

T. Burrows and D. Verhoeven

a vast body of philosophical discussion about the nature of language and meaning, including critiques of computational linguistics (Harris 1987). The Serendip-o-matic tool (http://serendipomatic.org/about/) is an implementation of this approach. Described as a “serendipity engine”, it takes a sample of text input by the user, parses it to extract key terms, and uses these to create keyword searches of a number of target databases, including the Digital Public Library of America, Europeana and Trove. Results are displayed as tiled images, though it is not clear what kind of ranking algorithm is used to arrange them. The developers comment: “Because the tool is designed mostly for inspiration, search results aren’t meant to be exhaustive, but rather suggestive, pointing you to materials you might not have discovered.” Another similar example is the “Never Been Seen” service (https://thesciencemu seum.github.io/never-been-seen/index.html) of the UK Science Museum in which an object from the digital collection with no Web page views is algorithmically selected and presented online as if the viewer is the very first person to encounter it. The idea that this is a serendipitous service rests on a deeply problematic, colonial narrative about discovery as the outcome of a “first contact” and in which the algorithmic production of “chance” conveniently renders invisible the human labour and labourers behind and therefore already familiar with these collections. A more elaborate application of a similar idea can be found in “SAGE Recommends,” launched by SAGE Publishing in 2016 (Maloney and Conrad 2016). Based on an investigation into “Serendipity, discovery, and the scholarly research process,” this functionality is intended to address “unexpressed, unarticulated information needs.” It produces recommendations by a combination of “string-based similarity” and matching a user’s input text against a large social science vocabulary. A different approach to recommendation systems is used by services like Pandora and Netflix, where music and movies are matched against an overall classification scheme and an ontology of attributes. But the best-known recommendation service is quite different: Amazon recommends books and other products based on a comparison with purchases by other customers who bought the same product as you did. Aleks Krotoski (2011) warns against reducing serendipity to an output of algorithms and computer code. In her view, services like Google and Amazon only aim “to serve up a result that will appear novel and unexpected, but is in fact tightly algorithmically defined, based on a user’s profile and derived using the data collected on him or her.” She argues, in effect, that serendipity is in the eye of the (human) beholder and cannot be manufactured in this way. There is a good deal of truth to her argument. While serendipitous encounters can certainly occur in the output of recommendation services, the whole purpose of such services is to minimise randomness and encourage predictability in the results (see also Reviglio, Chap. 8 and Erdelez, Chap. 12 of this volume). Serendipity would seem to require a different balance between these two extremes.

3 Serendipity and Knowledge Organisation

37

Serendipity and the Information-Seeking Process A different approach to the question of serendipity in digital environments can be derived by starting with the user, and attempting to model the process as a form of information retrieval. Foster and Ford (2003) focus on the behaviour of the information seeker, and the personal characteristics of people who appear to be good at serendipitous discoveries. They cite earlier work by Erdelez (1996) on “information encountering”, in which she identifies people who seem to be “super-encounterers” (later updated in Erdelez 2004, see also Erdelez, Chap. 12 of this volume). For Foster and Ford the aspect of serendipity related to discovering hidden connections, and its relationship to interdisciplinary research is an undertheorised dimension of the concept. A research group at the University of Western Ontario—including Victoria L. Rubin, Anabel Quan-Haase and Kim Martin—developed a model of serendipitous discovery which centers around the idea of “the find”: the object or idea discovered. Their data were derived from blog posts in which people described serendipitous moments from a variety of contexts, not just libraries (Rubin et al. 2011). They also tested a mobile app designed to connect digital searches with the physical library; STAK (Serendipitous Tool for Augmenting Knowledge) uses search results as a prompt for recommending nearby items while browsing library shelves (Martin et al. 2014). A somewhat different approach has been taken by Makri et al. (2014). Based on semi-structured interviews with researchers, they identified a set of human behaviours which are likely to increase serendipitous encounters with knowledge systems. These behaviours might serve as design principles for information systems at a conceptual level, guiding the ways in which people ought to be able to behave in a system, which can be summarized as: 1. Vary your routine, to provide more opportunities for making connections; 2. Be observant, to draw attention to unexpected aspects of the information environment; 3. Make mental space, reducing cognitive load to create contingencies for unexpected discovery; 4. Relax your boundaries, to avoid excluding valuable information entities due to overly narrow constraints; 5. Draw on previous experience, as a way of making and projecting the value of connections; 6. Look for patterns, in order to recognise unexpected alignments and extract meaning from them; and, 7. Seize opportunities, to exploit value from connections made.

38

T. Burrows and D. Verhoeven

Serendipity and the Environment While these approaches help to build a more detailed understanding of the process that people are following when they encounter something they label “serendipity”, it tends to focus on the user at the expense of the environment in which they are searching or browsing. Lennart Björneborn’s research into the browsing process in the physical library concentrates on the environment itself. He identifies no less than ten dimensions through which the design and operation of physical libraries may be thought to encourage serendipitous browsing (Björneborn 2008): Unhampered access: direct, unimpeded access to information resources Diversity: a variety of topics, genres, resources, activities, and sections Display: drawing attention to specific items to stimulate curiosity Contrasts: differentiation between types of physical zones Pointers: signage, maps, and markers Imperfection: imperfect ‘cracks’ and ‘loopholes’ in library processes (such as books left on tables) Cross contacts: contact across different topics, genres, and resources Multi-reachability: different access routes through the library Explorability: users are invited to move, explore and browse Stopability: users are invited to stop, touch and assess the items they find

These affordances in the physical library may have been intended by the designers of the library, or they may arise in an unintended way. The library’s aim should be to find a suitable balance between the need for order and the encouragement of serendipitous potential. Björneborn speculates that these affordances may be replicable in digital systems, but notes that physical libraries have an important advantage: “only in physical library spaces is it possible for users to have … direct, tangible access to physical resources.” In the digital world, all access to resources is mediated through the digital interface. Björneborn’s affordances could serve as valuable design criteria for digital systems as well as physical ones. But he leaves out—or takes for granted—the very foundation on which these affordances are built: the way in which the physical library’s collections are organised in a coherent sequence derived from Dewey or something similar. Serendipity appears to emerge from the variety of approaches through which users gain access to this organised structure. In the digital environment, this seems to suggest, serendipity should also require the establishment of a coherent, structured framework which can be approached through a variety of channels. The crucial question is whether knowledge organization systems in the digital environment can also be organised and presented to encourage serendipity in the way that physical libraries do.

3 Serendipity and Knowledge Organisation

39

Serendipity, the Web, and Social Media The Web itself, in the view of some commentators, is inherently serendipitous. Steven Johnson (2006), for example, describes it as “the greatest serendipity engine in the history of culture”. He acknowledges that the architecture of the Web “was not explicitly designed for accidental discoveries (quite the opposite, in fact)”, but observes that it “turns out to have an additional, unplanned serendipity-enhancing effect”. As a result, he says, the Web is “a machine for serendipity”. This kind of generalization glosses over the fact that “the Web” is not a unitary thing. It is actually a conglomeration of different types of services, glued together by the HTTP protocol—and perhaps given a superficial unity by Google’s single search box. But Johnson, writing in 2006, based his argument for the Web’s serendipity not on Google but on hyperlinks. For him, “the Web is far better at … following a trail of associations from some original starting point. Pick a given entry in the print Britannica and a given entry in Wikipedia, and try to find ten interesting loosely associated articles published in other venues around the world. It’s a thousand times easier to do this on the Web, obviously.” His starting-point has been criticised by Krotoski (2011), who argues that this is not serendipitous discovery at all. Instead, “it’s discovering something … via a path laid out by the people who edit the articles. It’s not making conclusions, drawing parallels or creating something new.” Johnson’s second example of the serendipity of the Web is blogging—and more specifically daily blog digests like Boingboing. This also reflects the situation in 2006, since a stronger case can be made from the subsequent dramatic rise of the social media services, which can filter and stream Web content in various ways. It is clear that they have their own contribution to make to the idea of serendipity. Social media streams are, in many ways, analogous to loosely-structured conversations. They are a mixture of the kinds of conversations you have with friends and family, on the one hand, and the coffee-machine and water-cooler conversations typical of the office environment, on the other. As long ago as 2009, Twitter was also described as “a serendipity engine” (Brogan 2009). As a user, you can choose the number and type of accounts you follow— personal, professional, organizational—or use the rough framework for classification by topic and subject which is provided by hashtags. It’s a kind of online equivalent to the “getting out and about” and “filling your mental filing cabinet with provocative stimulus” recommended by Matt Kingdon (2012) and other innovation consultants. Following a narrowly defined group of people with very similar interests is unlikely to produce much in the way of serendipitous encounters. Following a large number of people with very disparate interests, on the other hand, is likely to overwhelm you with randomness and chaos. Curating your feed to find the right balance between these extremes is essential if serendipity is the goal. Twitter has even been studied as a channel for talking about serendipity (Bogers and Björneborn 2013). It was interesting, in this context, to observe the hugely negative reaction to the suggestion in February 2016 that Twitter would start using an algorithm—rather than

40

T. Burrows and D. Verhoeven

a simple chronological sequence—to arrange users’ feeds in some kind of “meaningful” ranking by relevance or popularity. One of the most common reasons for opposing this change seemed to be loss of individual control over what appeared in a user’s feed and in what order. Many users complained that Twitter was in danger of becoming like Facebook—presumably because of Facebook’s default to “Top Stories” in ordering and promoting posts, though Facebook does also allow ordering by “Most Recent.” Curating for serendipity is possible in Facebook too, by finding a suitable balance between variety of friends and number of friends, and choosing to hide posts by selected friends from your timeline. The co-founder of Facebook Mark Zuckerberg is an enthusiast for the idea of serendipity. When the timeline feature was introduced to Facebook in 2011, he described it as “real-time serendipity.” In an interview for Time magazine the previous year (Grossman 2010), he explained his approach in more detail, beginning with a definition of serendipity as “a lucky coincidence”: It’s like you go to a restaurant and you bump into a friend that you haven’t seen for a while. That’s awesome. That’s serendipitous. And a lot of the reason why that seems so magical is because it doesn’t happen often. But I think the reality is that those circumstances aren’t actually rare. It’s just that we probably miss like 99% of it. How much of the time do you think you’re actually at the same restaurant as that person but you’re at opposite sides so you don’t see them, or you missed each other by 10 minutes, or they’re in the next restaurant over? When you have this kind of context of what’s going on, it’s just going to make people’s lives richer, because instead of missing 99% of them, maybe now you’ll start seeing a lot more of them.

But the serendipity encountered in Twitter and Facebook is very much of a social and news-oriented kind. What is lacking is the notion of serendipitous access to organised knowledge, as found in the physical library collection. Is this even possible (or desirable) in the digital environment?

Knowledge Organization and the Web In the early days of the Web, finding information was not primarily a matter of using search engines. Following hyperlinks between pages was certainly novel and important, but there were also organizational frameworks based on subject-matter. One of the best known was the Internet Public Library, established in March 1995, which included a Reference Center with listings of Web resources arranged under ten basic subject classes, each with their own sub-classes. The Yahoo! company made its name in 1994 with its Directory, which organised the Web by listing sites under nineteen subject categories. The Yahoo Directory was taken down at the end of 2014, while the Internet Public Library was closed and frozen during the following year. There were numerous other sites which classified Web resources within specific areas of knowledge. A long-running and highly respected attempt to categorise “humanities research on the Web” was provided by Alan Liu’s The Voice of the Shuttle (http://humanitas.ucsb.edu). This listed Web resources under subject or disciplinary

3 Serendipity and Knowledge Organisation

41

categories, arranged alphabetically from Anthropology to Women’s Studies, Gender Studies and Queer Theory. There were also supplementary lists of different types of resources. This approach was gradually abandoned, partly because the Web grew far too big and diffuse, and partly because of the increasing success of search engines (culminating in the triumph of Google). These early Web subject guides were heavily influenced by the predecessor of the Web: Gopher. Between the middle of 1991 and the end of 1993, Gopher grew rapidly, until it was superseded by the first Web browsers (especially Mosaic and Netscape) and by the ability of Web servers to subsume the Gopher protocol. Gopher servers worked from a structure based on folders and files; a user navigated from an initial menu through sub-menus to specific files, analogous to browsing through a file directory. The rapid obsolescence of Gopher in the face of the Web was partly the result of a decision by its developers, the University of Minnesota, to demand license fees (in contrast to the free distribution of Web server software). Gopher’s limited support for non-text media, and increasing modem speeds, played their part too (Anderson 2009). But Gopher’s directory structure and hierarchical navigation were also part of the equation. Gopher offered a rigid approach to browsing, in contrast to the hyperlinks of the Web. Interestingly, Jennifer Rowley in her 1987 textbook appears to have anticipated the emergence of Gopher and the Web with her comments on the recent rise of menu-based information retrieval systems like PRESTEL (Rowley 1987:192). These systems “have found favour because of their apparent simplicity” and are designed for the general public. Specialist information retrieval systems which use commandbased keyword searching are, by implication, far more sophisticated. And yet, she observes, menu-based systems “may overcome many of the limitations of enumerative classification schemes [like the DDC] in their traditional applications,” since they have a strong similarity to searching through the hierarchical structure of a classification scheme. She also suggests, somewhat perspicaciously as it turns out, that this process could be viewed as involving “not hierarchies, but networks.”

Serendipity and Spatial Representations If menu-based approaches to the Web have run their course, then what digital alternatives are there for organising or arranging knowledge to encourage serendipitous discoveries? One argument is to design the interface to resemble a physical library, as various projects and services—including the Digital Public Library of America— have attempted to do (Cohen 2014). This seems far too literal in its assumptions about both the format of digital resources (as surrogates for specific types of physical objects) and the ways in which users might want to navigate through them—though there does not appear to have been any research to test its effectiveness. A less literal analogy from the physical library might suggest that a combination of the following elements would be required: a framework for representing knowledge (regardless of

42

T. Burrows and D. Verhoeven

how it is conveyed), together with some form of spatial orientation and a variety of methods for self-navigation by users. Alan Liu’s comments about the erosion of the traditional “knowledge space” of the humanities disciplines are pertinent here (Liu 2015). He refers to the “tacit orientation provided by place-based collections” and the way in which “spatially organized structures” like the book and “associated finding aids” provided an interface to knowledge. He also mentions the “spatial-juridical architecture of archives.” This suggests that spatial browsing facilitates access to knowledge across a whole range of different settings, and encourages us to focus on the role of spatial organization in other types of cultural heritage institutions. It is worth looking at some of these other physical settings, in the light of Liu’s comments. Archives—at least in the institutional sense—have a strongly contextual structure, usually based around the various administrative functions from which archival records derive. While archival collections themselves are not normally directly browsable, finding aids and descriptive databases often reflect this contextual structure and enable users to walk through a digital or documentary surrogate which sets out what Liu calls the “spatial-juridical architecture”. In this sense, at least, there is a spatial, conceptual and contextual framework to orient the user. Private papers now in public institutions are usually treated similarly, in that they are normally arranged either in the order in which their creator kept them or in some kind of logical or thematic order imposed by the institution. This may involve difficult choices: should correspondence be arranged primarily by date or by correspondent, for example? Interestingly, Horace Walpole’s letters (including the one in which he coined “serendipity”) are arranged by correspondent, not by date, in the published edition, which has annoyed some readers (Wheatcroft 2016). Either way, however, private papers can be browsed within a meaningful framework, providing the kind of spatial orientation to which Liu refers. Manuscript collections, on the other hand, are much more likely to be arranged in the order in which individual items were acquired by the institution or collector. They are too precious or vulnerable to be browsed directly, for the most part, so readers must rely on catalogues or finding aids. But these generally follow the same order as the shelves, and are therefore devoid of any spatial orientation in Liu’s sense. Private manuscript collections which have been incorporated into public collections are often arranged by the original collector’s shelf-mark, which probably reflects their order of acquisition. The situation is different again for the modern museum or art gallery. Only a small selection of items is permanently displayed, in an arrangement dictated by curators. Conventionally the aim has been to illustrate the different types of object, with a selection of each, as well as to provide historical and thematic groupings or narratives. These exhibits are expected to be browsed by the visitor in the way recommended by the curator, reducing the likelihood of unexpected encounters. It is also possible, of course, to browse against the grain, e.g., by deliberately walking through rooms in the “wrong” order. The latter seems far more likely to produce unexpected connections and discoveries. But the majority of the contemporary museum’s items are in storage,

3 Serendipity and Knowledge Organisation

43

where they are not usually browsable. A recent trend has seen a few of these stores— such as that of English Heritage—open to the public for browsing. But the stores are normally arranged to maximise the efficient use of space rather than to encourage browsing, though similar materials may be grouped together nevertheless. There are exceptions to the usual museum layout, such as that of the Pitt Rivers Museum in Oxford. The original collector, Augustus Pitt Rivers (1827–1900), insisted that large numbers of items be displayed, and that they be arranged for browsing according to the function of the object, which takes precedence over a cultural or chronological arrangement (Chapman 1985). Another exception is the Isabella Stewart Gardner Museum in Boston, which also reflects an unalterable idiosyncratic personal arrangement devised by the original owner. Her approach was primarily intended to produce an artistic, decorative and emotional effect, but also has room for thematic and chronological sections, such as her Renaissance Room (Goldfarb 1995). Most of the objects are displayed for browsing, and they are very varied in type, including documents in cases. Sir John Soane’s Museum in London has something of a similar approach. Another exception is the Museum of New Art (MONA) in Hobart. Its founder, David Walsh made his fortune in online gambling and, in keeping with a life that benefited from massaging algorithms to increase chance, the museum has no obvious sense of an ordered progression or coherent curatorial agenda. Visitors are welcome to meander exhibits in any order without recourse to familiar knowledge aids like wall signage. Museums like these reflect a much older and more personal tradition, which goes back beyond the emergence of the new institutional museums of the nineteenth century (Pearce 1995). Before this, individual collectors formed and arranged collections which covered many different types of material, across many different fields of knowledge, arranged them in some kind of meaningful order(s), and made them accessible and browsable, even if only by friends and acquaintances of the collector. Two good examples are the enormous collections of Sir Hans Sloane and Roger de Gaignières, formed in the late 17th and early eighteenth centuries, in England and France respectively (Delbourgo 2017; Ritz-Guilbert 2012). Their collections were very broad in scope, were intended to be seen and browsed, and were arranged in deliberate, systematic ways. Both were subsequently dispersed and are now housed in multiple institutions, where they are not browsable, and where cataloguing and description are inconsistent and incomplete. These large, eclectic personal collections built on an even earlier tradition: that of the Wunderkammer and the Cabinet of Curiosities. Gaignières was specifically described as “amassing an incomparable cabinet” (de Grandmaison 1891:201–2).

44

T. Burrows and D. Verhoeven

Beyond the Knowledge Hierarchy Do these earlier approaches give us some inspiration for the digital environment? The Dewey approach in libraries relies ultimately on the idea that knowledge can be categorised into a hierarchy of disciplines and topics. Though it lingers on in the physical library, it has foundered more generally in the face of post-modernist scepticism about the master narratives of the Enlightenment. The DDC owes its origins to efforts by thinkers like Diderot and D’Alembert, in their vast Encyclopédie, to sum up and organise the new learning of the seventeenth and eighteenth centuries. In this sense, the classified libraries arranged by DDC are a physical embodiment of the tree diagrams used by Diderot and D’Alembert and their predecessors, including Francis Bacon and René Descartes. As Manual Lima observes, in his definitive book on the subject, “the recursive branching structure of trees … provides a compelling metaphor for organizing knowledge” (Lima 2014:7). Today, curators and audiences are far more likely to look for multiple perspectives and multiple categorizations, and to prefer pluralist and personalised approaches. The Cabinet of Curiosities and the Wunderkammer were less systematic and less organised, as well as more personal and more eclectic. And yet they could be large and diverse, and were intended to stimulate curiosity through browsing. Could we reconstruct similar spaces in a digital environment? Could researchers construct such a space for themselves by rearranging the data provided by the custodial institution? One danger of this approach would be such a fragmentation of perspectives that no connections can be made between them. To enable dialogue between researchers, there also need to be digital environments where interpretations and points of view can be shared. Alan Liu’s observations about the need to reconstitute and re-orient the “humanities knowledge space” in new ways focus on “maps, networks, and provenance structures serving as way-finding aids” (Liu 2015) echoing Rowley’s earlier speculations. Liu’s comments suggest that we focus, not on organising knowledge as such, but on designing the paths by which different pieces of knowledge can be connected. These paths—his “maps, networks, and provenance structures”—can provide a basis for browsing in a pluralist, personalised, eclectic knowledge space. They also point to the centrality of three elements in this process. Visualisation is critical, as a means of designing digital interfaces suitable for browsing. We also need techniques for presenting knowledge in the form of building-blocks which can be combined, connected, and linked; network graphs are perhaps the most obvious example of this. Finally, we need interfaces which can allow for multiple perspectives and interpretations, while recording their origins and joining them together to be seen in a holistic way. In this kind of environment, browsing—and serendipity—can flourish.

3 Serendipity and Knowledge Organisation

45

Humanities Networked Infrastructure (HuNI) An Australian information initiative that anticipated Liu’s aspirational vision is the Humanities Networked Infrastructure (HuNI) project (http://huni.net.au, 2012) (Burrows and Verhoeven 2016). HuNI was designed by a multi-disciplinary and multi-institutional coalition of stakeholders (universities, digital infrastructure agencies, museums, libraries, archives) interested in determining how to interoperate digital cultural collections without compromising the underlying precepts of humanities research—curation, complexity, co-existence, contestation, and connection (including non-logical connections in the form of serendipity, for example) (Verhoeven 2016). The result is a virtual laboratory that enables users to propose, describe (in their own words) and visualise the relationships between information entities. HuNI is a graphic network of interconnected information in which the organising ontology is built by researchers themselves. New interlocutors can build on the network either by extending a trail of relationships, or by challenging an existing connection and adding an alternative. The experience of using HuNI is very close to the kinds of descriptions of serendipity in the physical library. Visitors can meander along information paths, following conceptual connections proposed by many different researchers, or in some cases pre-proposed by source data collection providers. As you skip along from one record to the next you are led to records you may not have expected to find but which, through proximity in the graph, enjoy a meaningful connection. In this way HuNI acknowledges the “spatial” sense of serendipity that is often lost in debates in the digital information sciences where the focus instead is on knowledge organisation and management.

Conclusion In focusing so intently on information retrieval and efficiency, the developers of digital information platforms have privileged a very specific definition of serendipity. Theirs is an online world of proficient information accidents, of algorithmically intended happenstance. To date, these developers of digital serendipity engines have focused on what are perceived to be the definitional requirements for unexpected information encounters which are typically interpreted to mean chance or left-field encounters. This gives rise to the complaint that there is a circular contradiction in such endeavours—that it is not conceptually possible to plan for the unplanned, to engineer for genuine chance (see also Arfini, Chap. 7 of this volume). In a sense this tendency to imagine digital serendipity systems as an almost impossible combination of chance and logic can be traced back to Horace Walpole’s initial definition of the term in his interpretation of the Three Princes of Serendip: that serendipity is a combination of accidents and sagacity. But Walpole (like many later commentators and information engineers) misses a key aspect of the foundational

46

T. Burrows and D. Verhoeven

story: that it is at its heart a chronicle of the value of attentive meandering. As the princes wander along and move through an unfamiliar environment they glean pieces of information that unexpectedly “add up” and that—with hindsight—become useful later in the story. Serendipity, then, is the ability to recognise a relationship that initially seems “out of place” and yet is retrospectively useful. It’s a dual move—serendipity contemplates not just how we locate knowledge (in space and time) but how knowledgelocates us. Serendipity always implicates us, the researchers, in the story of “discovery” as historically problematic as the colonial terms of that encounter may be. Which means it is always going to involve more than just the one-way transaction of search boxes or other one-dimensional information retrieval technologies. In view of this manifold definition, developers of information platforms seeking to encourage serendipity would do well to consider an expanded reading of the foundational story—one that acknowledges the provisional spatial and temporal (and political) dimensions that enable serendipitous encounters. One that values the meandrous, the human and the meaningful over the efficient, the economic and the effective (see also Soto, Chap. 11 of this volume).

References Anderson, N. (2009). The Web may have won, but Gopher tunnels on. Ars Technica. Retrieved June 1, 2023, from https://arstechnica.com/tech-policy/2009/11/the-web-may-have-won-but-gophertunnels-on/ Berman, S. 1980. DDC 19: An indictment. Library Journal 105 (5): 585–589. Björneborn, L. 2008. Serendipity dimensions and users’ information behaviour in the physical library interface. Information Research 13 (4). http://www.informationr.net/ir/13-4/paper370. html. Bogers, T., and L. Björneborn. 2013. Micro-serendipity: Meaningful coincidences in everyday life shared on Twitter. In iConference 2013 Proceedings (196–208). https://doi.org/10.9776/13175. Brogan, C. 2009. The beauty of collaboration, September 13. Retrieved November 4, 2021, from http://chrisbrogan.com/the-beauty-of-collaboration/. Burrows, T. 1994. Integrating electronic services into the academic library: The Scholars’ Centre at the University of Western Australia. Australian Academic and Research Libraries 25: 213–220. Burrows, T., and D. Verhoeven. 2016. Aggregating data for social linking in the humanities and creative arts: The Humanities Networked Infrastructure (HuNI). Signa: Revista de la Asociacion Espanola de Semiotica 25: 109–119. Chapman, W.R. 1985. Arranging ethnology: A. H. L. F. Pitt Rivers and the typological tradition. In Objects and others: Essays on museums and material culture, ed. G.W. Stocking (15–48). University of Wisconsin Press. Cohen, D. 2014. Planning for serendipity, February 7. Retrieved November 4, 2021, from http://dp. la/info/2014/02/07/planning-for-serendipity/. de Grandmaison, C. 1891. Gaignières, ses correspondants et ses collections de portraits (suite). Bibliothèque de l’école des chartes, 52: 181–219. Delbourgo, J. 2017. Collecting the world: The life and curiosity of Hans Sloane. Allen Lane. Erdelez, S. 1996. Information encountering: A conceptual framework for accidental information discovery. In Information seeking in context: proceedings of an international conference on

3 Serendipity and Knowledge Organisation

47

research in information needs, seeking, and use in different contexts, Tampere, Finland, ed. P. Vakkari, R. Savolainen, and B. Dervin (412–421). Taylor Graham. Erdelez, S. 2004. Investigation of information encountering in the controlled research environment. Information Processing and Management 40 (6): 1013–1024. Foster, A., and N. Ford. 2003. Serendipity and information seeking: An empirical study. Journal of Documentation 59 (3): 321–340. Fox, P. 2014. Trinity College Library Dublin: A history. Cambridge University Press. Goldfarb, H.T. 1995. The Isabella Stewart Gardner Museum: A companion guide and history. Yale University Press. Grossman, L. 2010. Mark Zuckerberg. Time, December 15. Retrieved November 4, 2021, from http://content.time.com/time/specials/packages/printout/0,29239,2036683_2037183_203 7185,00.html. Harris, R. 1987. The language machine. Duckworth. Jayne, S. 1956. Library catalogues of the English renaissance. University of California Press. Johnson, S. 2006. Everything bad is good for you: How today’s popular culture is actually making us smarter. Riverhead. Kingdon, M. 2012. The science of serendipity: How to unlock the promise of innovation in large organisations. Wiley. Krotoski, A. 2011. Digital serendipity: be careful what you don’t wish for. The Guardian, August 21. Retrieved November 4, 2021, from http://www.theguardian.com/technology/2011/aug/21/ google-serendipity-profiling-aleks-krotoski. Lima, M. 2014. The book of trees: Visualizing branches of knowledge. Princeton Architectural Press. Liu, A. 2015. Digital humanities and the reorientation of the humanities knowledge space. Expert Meeting on Spatial Discovery. University of California Santa Barbara, June 18. Retrieved November 4, 2021, from http://liu.english.ucsb.edu/digital-humanities-and-the-reorientationof-the-humanities-knowledge-space-ucsb/. Makri, S., A. Blandford, M. Woods, S. Sharples, and D. Maxwell. 2014. ‘Making my own luck’: Serendipity strategies and how to support them in digital information environments. Journal of the Association for Information Science and Technology 65 (11): 2179–2194. Maloney, A., and L.Y. Conrad. 2016. Expecting the unexpected: Serendipity, discovery, and the scholarly research process. SAGE Publishing. https://doi.org/10.4135/wp160129.1. Martin, K., B. Greenspan, and A. Quan-Haase. 2014. STAK—serendipitous tool for augmenting knowledge: Bridging gaps between digital and physical resources. Paper presented at Digital Humanities 2014, Lausanne, Switzerland, July. Retrieved November 4, 2021, from http://dha rchive.org/paper/DH2014/Paper-314.xml. McKeen, W. 2006. The endangered joy of serendipity. St. Petersburg Times, March 26. https://www. tampabay.com/archive/2006/03/26/the-endangered-joy-of-serendipity/. Poulter, A. 1999. Browsing the virtual library. In Encyclopedia of library and information science, ed. A. Kent (vol. 62, suppl. 5, 389–396). Marcel Dekker. Pearce, S.M. 1995. On collecting: An investigation into collecting in the European tradition. Routledge. Race, T.M. 2012. Resource discovery tools: Supporting serendipity. In Planning and implementing resource discovery tools in academic libraries, ed. M.P. Popp and D. Dallis, 139–152. Information Science Reference. Ritz-Guilbert, A. 2012. La collection de François-Roger de Gaignières (xviie siècle). Annuaire de l’École pratique des hautes études (EPHE). Section Des Sciences Historiques Et Philologiques 143: 220–221. Rowley, J.E. 1987. Organising knowledge: An introduction to information retrieval. Gower. Rubin, V.L., J. Burkell, and A. Quan-Haase. 2011. Facets of serendipity in everyday chance encounters: A grounded theory approach to blog analysis. Information Research 16 (3). http://www.inf ormationr.net/ir/16-3/paper488.html. Silverman, R.E. 2013. The science of serendipity in the work place. The Wall Street Journal, April 30. https://www.wsj.com/articles/SB10001424127887323798104578455081218505870.

48

T. Burrows and D. Verhoeven

von Stackelberg, A. 2018. Sexual harassment by library patrons: #TimesUp”. BCLA Perspectives 10 (1). https://bclaconnect.ca/perspectives/2018/03/02/sexual-harassment-by-library-pat rons-timesup/. Verhoeven, D. 2016. As luck would have it: Serendipity and solace in digital research infrastructure. Feminist Media Histories 2 (1): 7–28. Wheatcroft, G. 2016. Walpole: The house and the letters. New York Review of Books 63 (5): 33–35.

Toby Burrows is a Digital Humanities researcher at the University of Oxford and the University of Western Australia. His research focuses on cultural heritage data and knowledge graphs, and he has a particular interest in the history of cultural heritage collections, especially medieval and Renaissance manuscripts. His recent projects include HuNI (the Australian Humanities Networked Infrastructure), Collecting the West, and Mapping Manuscript Migrations. Deb Verhoeven is the Canada 150 Research Chair in Gender and Cultural Informatics at the University of Alberta. Previously she was Associate Dean of Engagement and Innovation at the University of Technology Sydney, and before this she was Professor of Media and Communication at Deakin University. She was recently awarded the Social Sciences and Humanities Research Council (SSHRC) grant to fund the Gender Equity Policies (GEP) Analysis project alongside international collaborators, and is renowned for her work on gendered networks in the Australian film industry, as well as in digital humanities broadly speaking.

Chapter 4

Serendipity in Management and Organization Studies Miguel Pina e Cunha

and Marco Berti

Abstract Miguel Pina e Cunha and Marco Berti understand serendipity as a process that requires the right combination of effort and luck. In this chapter, they suggest that organizations are better off embracing or “cultivating” serendipity as learning opportunities. Their approach is to frame serendipity as a “negative capability, i.e. a capacity to pursue a vision that leads to confusion and uncertainty rather than to certainty and clarity.” Their argument sets out by making a distinction between the mechanistic view and the organismic view of organization. In contrast to mechanistic views, organismic views acknowledge the unpredictability and uncontrollability of the environment external to the organization and, thus, its serendipitous potential. Cunha and Berti lament that organizational cultures rather engage in a paradoxical strategy, of trying to predict what their competitors will do, while at the same time trying to remain unpredictable to them. They suggest several ways in which effort towards serendipity can be realized, including generative doubt—the fostering or embracing of a state of not-knowing, and peripheral vision—paying more attention to areas outside the organization’s attention. Conclusively, Cunha and Berti argue that while not being controlled, serendipities’ probability can be increased.

The concept of ‘serendipity’ is a newcomer and still a rare observation in the conceptual landscape of management and organization studies (MOS). There is one easy explanation for this absence: Organization is often taken to be the removal of that which is accidental, unpredictable, unnecessary, inefficient; serendipity ticks all of those boxes. Even more: if the role of management consists in eliminating unpredictability and uncertainty, serendipity is a process/outcome to be carefully avoided. Organization, historically, thus consisted in the adoption of practices that increase certainty: routines (Parmigiani and Howard-Grenville 2011), audits (Power 1997), M. P. Cunha (B) Nova School of Business and Economics, Universidade Nova de Lisboa, Lisbon, Portugal e-mail: [email protected] M. Berti University of Technology Sydney, Sydney, NSW, Australia e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 S. Copeland et al. (eds.), Serendipity Science, https://doi.org/10.1007/978-3-031-33529-7_4

49

50

M. P. Cunha and M. Berti

standards (Brunsson and Jacobsson 2010), reliability (Roberts 1990). These practices have certainly contributed to increase the organization of organizations. However, there is a problem with this view of organization: founded upon mechanistic views and imagined as an outcome of engineering, classical organization studies discounted the importance of process and change, as well as of the “small elements of randomness” that can significantly affect the evolution of institutions (Acemoglu and Robinson 2012) and society (Jones and Olken 2009). For example, during the clinical trial of Covid-19 vaccines, participants in an experiment conducted by AstraZeneca mistakenly received half a dose, which allowed researchers to ‘stumble’ onto a more promising dosing regime. Mendelas Pangalos, the company’s executive in charge of research and development described the process as a “quite useful mistake” (Robbins and Mueller 2020). Classic mechanistic1 organizations and functions are now being reinvented under the auspices of AI and algorithms (Capelli 2020; Volberda et al. 2021), but this automation of work is socially bounded (Fleming 2019), and individuals adapt and resist in unpredictable ways to ‘algorithmic management’ (Walker et al. 2021). In relentlessly shifting environments, functional predictability may become less attractive than the capacity to analyze and respond to the environment. In a sense, planning may have become less important and improvisation may have become more important, open as it is to a logic of opportunity (Roberts and Eisenhardt 2003) attentive to fleeting, serendipitous discoveries. Moreover, improvisation and capacity to adapt are not anymore exclusively human prerogatives, thanks to Artificial Intelligence and machine learning (Giles and Walkowicz 2019). In this chapter we frame serendipity as a negative capability, i.e. a capacity to pursue a vision that leads to confusion and uncertainty rather than to certainty and clarity. The concept of negative capability was originally introduced by the Romantic poet John Keats, who defined it as a capacity (which he ascribed to Shakespeare) “of being in uncertainties, mysteries, doubts, without any irritable reaching after fact & reason” (Keats 1817 [1958], cited in Ou 2011, p. 1). The idea has been borrowed by the Brazilian philosopher Roberto Unger (2004), to illustrate the need to oppose enforced social constraints that are based on fixed, naturalized distinctions and positions. In this view, negative capabilities can be associated with the possession of a paradoxical mindset, the capacity to accept and feel energized by tensions and ambiguities (Miron-Spektor et al. 2018). Serendipity cannot be controlled in the traditional sense of the word, but organizations can learn to “maneuver the journey” (Coyne and Van de Ven 2021), meaning that it can be provoked (Lane et al. 2020), not in concrete but in probabilistic terms. The rise of machine learning and its uncovering of patterns ignored by humans, may offer unprecedented opportunities for serendipitous discovery (Leavitt et al. 2020). Most of these serendipitous strategies will probably not succeed—even though the data on this would be hard to gather because of retrospective narration of the serendipitous process (Ross and Vallée-Tourangeau 2020), which erases failure. Deviations, anomalies and surprises are common in everyday organizational life, which confronts 1

Here we use ‘mechanistic’, after the classic definition by Burns & Stalker (1961).

4 Serendipity in Management and Organization Studies

51

Table 4.1 Serendipity in mechanistic and organismic organizations Mechanistic organizations

Organismic organizations

Organization

A system of well-articulated routine

A flexible, adaptive system

Root metaphor

Machine

Living organism

Core outcomes

Reliability, efficiency

Agility

Core competences

Improvement Efficient use of resources Paradox as absurdity Error elimination Uncertainty removal

Improvisation Bricolage and new affordances Paradox management Learning through errors Uncertainty as normal

Serendipity

A factor to be removed, an intrusion of uncertainty A distraction from core routines

A process to be cultivated, a negative capability A dynamic capability

Attention

Focused—on core activities

Mindful—of the periphery

Prevailing attitude

Certainty

Doubt

Problems with serendipity

As a source of deviation, drift

As a cause of zemblanity, drift

organizations with decisions regarding how risk-tolerant they aim to be. As organizations move from mechanistic to organic designs, as they are doing under the so-called agile paradigm (Birkinshaw et al. 2021), they aim to become nimble enough to change with the unanticipated, lowering the cost of change. Organismic organizations frame the nurturing of serendipity as a dynamic capability (Busch and Grimes, Chap. 5 this volume). But organismic organizations are also exposing themselves to drift, chasing one opportunity after another or tackling threats reactively rather than systematically. The change from mechanistic to organic presents a number of implications, summarized in Table 4.1, that organize the core concepts used in this chapter. We thus use the notion of serendipity as meaning that luck tends to favour those who remain flexible and alert, even though acknowledging the existence of other definitions (Liu and De Rond 2016). Inspired by the works of Chia and Morgan (1996) and of Hahn and Knight (2021) we consider that one cannot provoke a given type of discovery but that some approaches may make it more likely for some discoveries to be made, by refusing to take linear answers as the right or necessary responses to problems. “Negative” in negative capability implies the refusal to scrutinize every piece of reality through the lens of reason or to grasp, in the immediate moment, a given process or explanation. Negative capabilities, like the others, are effortful: as we have noted in the definition, there are elements of agency and effort in serendipity. As Krumboltz (1998) ironically observed, serendipity is not serendipitous: it flourishes when the conditions are adequate. When the context is not propitious, such as when people work from home during a pandemic, serendipity is affected (Gratton 2020). Therefore, organizations cannot systematize the serendipity process, but they might create contexts that will make serendipity more likely, in a search for planned luck (Busch and Barkema 2020).

52

M. P. Cunha and M. Berti

This will demand purposeful attempts at exploration combined with a culture that establishes doubt, ignorance, and luck as valid management ingredients, rather than as extraneous factors that should be eliminated or ignored, for example by retrospective justifications that transform luck into some expression of managerial prescience. To discuss serendipity in organizations we structured the chapter in the following way. First, we start with a note on the place of serendipity in management and organization studies. Next we discuss how mechanistic organization theory and practice often exclude serendipity from attention and why organismic organization theory is re-appreciating its role. A habitual emphasis on certainty and predictability turn serendipity into a dysfunction. We then discuss how organizations may instead embrace serendipity, to conclude with a note on the possible future of serendipity in organization theory.

The Place of Serendipity in Management Organization Studies As Yaqub (2018) has explained, serendipity is a broad and multifaceted phenomenon, which manifests as different expressions, namely in the organizational context. Serendipity happens as people in organizations make discoveries that were not programmed. We define serendipity as an effortful process leading to an accidental discovery. In this definition serendipity is both process and outcome. Taking serendipity as process means that it should not be reduced to some sudden event or episodic occurrence. The process is effortful, in the sense that it requires investment of resources in being alert to the possibility of encountering opportunities, in assessing such opportunities when they are fortuitously encountered, and in testing their integration in existing practices. Thus, effort and agency sometimes leads to discoveries that were not programmed, meaning that someone discovers something that they were not trying to find—even if they were trying to discover something else. The concept still constitutes only a minor presence in MOS. This is partly due to the fact that organization theory started under the influence of mechanical engineering and of the machine metaphor (Shenhav 2002). The notion of the organization as a machine and its implications for organizational theory and practice molded the discipline for decades. Core management concepts are machine-oriented and antiserendipity; they emphasize certainty and avoid the role of chance: focus, goal setting, segmentation-targeting-positioning, and so forth. These concepts leave little room for chance and unexpectedness. In this section we consider two views of organization: a mechanistic view, in which the organization is assembled, and a process view, in which organization is cultivated or nurtured (Brown and Eisenhardt 1998). Serendipity in mechanistic views of organization theory. Machine-like organization theory (Morgan 1986), aimed to create organizations in the image of artificial mechanisms. They thus departed from a Newtonian desire to explore organization

4 Serendipity in Management and Organization Studies

53

as a form of social physics (Bowles 1990), epitomized by Taylor’s (1911) scientific approach to management that aimed to eradicate waste. In this perspective, organizing is the effort of introducing predictability by removing uncertainty. It is a vision of organization based on engineering processes (Shenhav 2002) that allowed managers to perfect organizational routines—sequences of repeated activities to be executed as prescribed. Tsoukas (2005) qualified uncertainty as the nemesis of traditional organizational thinking. In this perspective, organization consists in the elimination of waste and uncertainty. Good management, in other words, consisted in the creation of organizations immune to external interference. Well-designed organizations have their technical core buffered from the influences of the environment (Thompson 1967) and are designed with an implicit assumption of designer “omniscience” (Kendall and Kendall 1993, p.153). The organization was thus represented as a closed system, imagined as a clockwork in which a “one best way” served as the organizing principle. Once such best practice was identified, organization would consist in repeating operations relentlessly, reliably. Yet, even within a mechanistic frame of reference, it is impossible to eliminate the complexity and unpredictability that emerges from the interaction of the components of a system. Even a physical system made of just two simple components, such as a double pendulum (a pendulum attached to another pendulum) can produce chaotic results (Levien and Tan 1993). Similarly, the application of simple schemata can produce complex results: the interaction of unsophisticated agents operating on the basis of simple rules (for example, ants in an anthill) generates complexity by means of self-organizing (Anderson 1999). Serendipity in organismic views of organization theory. In spite of assumptions of rationality and mechanistic certainty, organizations have become more organismic (Birkinshaw et al. 2021). They still need to be as efficient as possible but they need to become agile. Management, in other words, is gaining awareness about the paradoxical challenges involved in agile organizations (Cunha et al. 2021). This is because as they operate, organizations transform the very rules and routines that are supposed to give them stability (Feldman and Pentland 2003; Tsoukas 2017) into expressions of irrationality, such as rules that must be disobeyed to be obeyed (Berti and Simpson 2021). A process view of organizations brings a new role for serendipity, no longer seen as a dimension to factor out of design, but as a condition of organizing. As Mackay et al. (2020) pointed out, unpredictability, equivocality, and serendipity are integral to organizing as process. Serendipity, together with other non-cognitive processes can also play an important role in the constitution of organizational dynamic capabilities, enabling firms to extract new value from existing knowledge and technologies (Nayak et al. 2020). Indeed, the very notion of ‘sensing’ opportunities and threats, which is central to dynamic capabilities (Teece 2007), is much closer to serendipity than it is to forms of systematic survey. In this view, management may pay attention to the unfolding of organizing efforts in a world that constantly changes. Organization, in a process perspective, refers to becoming (Tsoukas and Chia 2002), the nature of change in a changing world.

54

M. P. Cunha and M. Berti

Serendipity and surprise are integral to this view (e.g. Ansoff 1975) and openness to surprise and anomaly (Von Krogh 2020) can be used to mitigate the threat of self-referentiality (Schoeneborn 2011), which might reduce attention to novel possibilities. Previous work has started to integrate serendipity in management and organization studies (Cunha et al. 2010; Cunha et al. 2015) and represent serendipity as more than luck, rather as integral to organizing in a complex, open-ended world. When systems are organized around a purpose rather than a plan, their efforts may produce the desired but also unintended consequences (Merton 1936). Some of these consequences are negative, others may be positive, i.e., serendipitous. New ways of seeing organizations, such as those inspired by a jazz metaphor, have integrated structure and serendipity as dual dimensions of organizing (Fisher et al. 2021). Organization is no longer viewed as predictable machinery. For the above reasons, organizations may be more or less open to serendipitous emergence, depending on how they represent themselves and their environments. Serendipity and luck may be more acceptable or less depending on the organization’s culture. Some cultures are tolerant or even supportive of happy accidents, whereas for others deviations from a script can be perceived negatively and hard to tolerate, especially when deviation is equated with insubordination (Mainemelis 2010). Given tensions between exploring and exploiting (March 1991), nurturing exploration implies that a measure of experimentation is necessary for learning to take place, namely through serendipity. Because not all forms of learning are equal, organizations that learn what they expected to learn (via exploitation) will miss opportunities to learn what they were not striving to learn, for example, serendipitously. A process-organic view of organizations, therefore, portrays serendipity as integral to organizing. Organizations may cultivate the skills that counseling psychologists called planned happenstance (Kim et al. 2014). Yet, traditionally management is not oriented towards planned happenstance. This is the theme that we discuss next.

How Organization Counters Surprise and Bulldozes Serendipity Serendipity plays a role in a process view of organization theory in which the openendedness of the world necessarily confronts organizations with surprise and the unexpected. In this perspective, reality is rich in surprise and organizations need to accommodate the unexpected in their action (Cunha et al. 2012). The world being open-ended, it cannot be “predicted”, and organizations cannot reliably plan. Expectations regarding the validity of strategic planning have already seen their heyday (see e.g. Mintzberg 2000) but the notion of planning illustrates the idealized version of a world in which one organization can plan its world while impeding planning in the world of its competitors. Because of open-endedness, this assumption is better replaced by one in which surprise is a natural consequence of agency and

4 Serendipity in Management and Organization Studies

55

emergence. Surprise happens, in addition to complexity and emergence, because our capacity to understand processes in scale and time is limited. Humans have difficulties with grasping even the most critical phenomena because of scale, such as time scale (Bansal et al. 2018). Given previous experience and expectations, we tend to have well defined solutions to well-known problems. The relation between problems and solutions is loose, at best, and organizations are often more interested in finding problems for their solutions rather than in finding solutions to their problems, as the garbage can model has explained in its metaphorical proposal—problems and solutions are dumped into the same space and associate in sometimes disordered ways (Cohen et al. 1972). For these reasons, organizations are often more interested in exploiting what they already know than in remaining alert to deviations or anomalies that might be indicative of cracks in organizing or to opportunities revealed in the absence of deliberate search. This tendency to normalize anomalies rather than to investigate them is problematic, as it has been associated with disasters and corruption (Ashforth and Anand 2003; Weick 2010). A number of motives have conspired to create less mindful organizations, i.e. organizations less likely to investigate anomalies. In some cases, in the extreme, organizations create coercive rules that are hard to counter even when they are dysfunctional. These forms of coercive structuring, including the well-known Kafkaesque bureaucracies (Clegg et al. 2016), create their own paradoxes (Berti and Simpson 2021) that lead to inaction and paralysis rather than to curiosity and discovery, disempowering their members and hindering rather than enabling work conditions, neutralizing the wisdom involved in serendipitous discovery (Copeland 2019). By making an effort to exercise control, organizations may exclude serendipity. Situations in which people pay attention to anomalies that might trigger serendipitous emergence may constitute an exception rather than a rule, when organizations promote obedience rather than curiosity. The decision to notice and to act in response to anomalies may be influenced by the organization’s power circuits (Clegg 1989). Maintaining an organization vigilant to unexpected opportunities, anomalies, deviations, requires a political infrastructure open to participation as well as a psychologically safe environment (Edmondson 2018). In case the organization is perceived as unsafe, serendipitous observations may be repressed rather than accepted. Given that deviations are not appreciated in mechanistic organization theory, correcting anomalies may be less risky than using them as opportunities for innovation. It is not surprising, therefore, that some organizations will fail to notice what better remains unseen. As de Rond and Lok (2016), observed, some things, once seen, cannot be un-seen; in contrast, other things may be kept unseen, namely serendipitous occurrences in machine-like organizations.

56

M. P. Cunha and M. Berti

Embracing Serendipity We now discuss approaches used by organizations to manage situations with a potential for serendipitous discovery. As noted above, we do not envisage serendipity as a pre-formed possibility nor the pure chance of encountering an opportunity waiting to be discovered. Rather, it is the result of sustained attention and openness to information, combined with perspicacity, a process in which people with some common interests potentiate their knowledge into unplanned areas (Lane et al. 2020). We discuss two domains that have been associated with serendipity: (1) Exposing the organization to serendipitous opportunities, and (2) Responding to serendipitous opportunities. These practices should not be viewed as representing forms of controlling or corralling serendipity in the directions desired by the management, but rather as possibilities for cultivating sensitivity to serendipity. Such practices, sustained over time, may increase ripeness for serendipity. Sensitivity may bear fruit or it may not. More predictable than any positive outcomes is the likelihood of expanding the organization’s attention beyond its habitual domains. It is important to highlight that serendipity—as a heuristic practice—is not casual. In their essay on serendipity Merton and Barber (2004) characterize the use of serendipity in science as the identification of data that is both unanticipated, anomalous, and strategic (i.e. allowing for new theorization). Such an achievement might be accidental but not truly unintentional: anomalies could be treated as observation mistakes, or ‘explained’ as non-anomalous recurring to ancillary theories in order to protect the paradigmatic core (Clegg et al. 2020; Lakatos 1978). Deciding to get ‘distracted’ by them, investing time and resources to further investigate these unanticipated occurrences and to leverage them to challenge existing theory, is an effortful accomplishment, predicated on the ability to maintain openness to unsettling inconsistencies. Thus, serendipity should be treated as a negative capability, a capacity to face mystery and doubt, and to embrace the inherently paradoxical nature of organizing (Smith and Lewis 2011). As a negative capability serendipity is accepted as a fact of life to be faced with stoic determination and frank curiosity, and as a means for the preservation of adaptability (Weick 2020). For this reason, instead of hiding from serendipity, organizations should expose themselves to it.

Exposing the Organization to Serendipitous Emergence Organizations can expose themselves to unsought discoveries by promoting “setting the stage” (Garud and Turunen 2021) or nurturing states of readiness (Dane 2020). Exposure will not offer any guarantees, as serendipitous discoveries imply effort and potentiality. In this section we discuss two possibilities previously envisioned in the literature: generative doubt and peripheral vision.

4 Serendipity in Management and Organization Studies

57

Generative doubt. One way to cultivate serendipity consists in living with generative doubt. Generative doubt refers to the cultivation of the experience of not knowing (Locke et al. 2008). Assuming an attitude of doubt is critical to be open for surprises, namely those of the serendipitous type. The problem with living in a condition of doubt is that organization consists in creating and perfecting routine and predictability, rather than in embracing ambiguity. For this reason, cultivating doubt can be difficult, but cases such as the Toyota Production system (Ohno 1988) seem to suggest that doubt and organization are not mutually exclusive. On the contrary, perfecting a system may benefit from an attitude of doubt. The relationship between generative doubt and the embrace of serendipity is still to be explored, but some possible antecedents may be considered. Leadership may play a critical role. New forms of leadership, such as servant/humble leadership, may be facilitative of a “learn-it-all”, growth type of mindset (Dweck 2008). Considering that in many organizations a “know-it-all” attitude prevails (Ibarra et al. 2018), cultivating generative doubt may provide the impetus for attention to serendipitous surprise. A “learn-it-all” type of attitude may promote serendipitous encounters because it encourages organizational members to create opportunities for planned happenstance. It can also be stimulated by bringing outsiders with a “doubt-based inquiry” approach to support people in questioning sense already made (Strike and Rerup 2016). Instead of assuming that deliberate search is the process that will help organizations to make valuable discoveries, doubt invites organizations to accept that valuable outcomes may reside in unexpected sites, namely the peripheries of organizational attention (Cunha and Chia 2007; Day and Schoemaker 2006). This mindset may stimulate unsought discoveries through the sharpening of peripheral vision. Peripheral vision. The emphasis on focus/clarity brings many advantages to management but it comes with a cost. MOS being a domain of focus, as reflected in the role of ideas such as segmentation/targeting/positioning or goals and goal setting. Focus is important as it directs organizational attention to critical ends with the corresponding gains of accountability and efficiency. Yet, these goals reduce attention to opportunities for learning. Focus may, for instance, inhibit serendipitous discoveries. Attention to existing issues, and a propensity to exploit existing knowledge “gets in the way of genuine learning” (Harrison et al. 2007, p. 338), creating an emotional attachment to the dominant organizational logic (Prahalad and Bettis 1986) and to subsequent inertia. Attachment to a dominant logic creates a form of “selective inattention” (Harrison et al. 2007, p. 339) that precludes potential discoveries, as anomalies are filtered as noise. This problem manifests also at individual level, as a trade-off between expertise in a domain and flexibility (capacity to apprehend and generate new ideas), which is however moderated by organizational factors, such as exposure to a dynamic environment and to other domains (Dane 2010). Thus, valuable information about developments that need to be considered may be obtained through the cultivation of peripheral vision, awareness of the spaces normally outside an organization’s attention. The notion of peripheral vision, introduced by Day and Schoemaker (2006), aims to educate managers and other organizational members to pay more attention to these

58

M. P. Cunha and M. Berti

areas outside the scope of organizational attention, where new developments incubate and new competitors grow. These areas (customers, users and uses, geographies, technologies) may present in different forms, such as new market spaces or different geographical locations. Erikson’s (2018) anthropological analysis of disease detection and containment during an Ebola epidemic in West Africa, for example, revealed unexpected utilization patterns of cellphones that differed markedly from those in other parts of the world. By engaging with users, one cannot predict that something valuable will be discovered, but the chance of unexpected discovery increases as curiosity and novel forms of thinking are sources of organizational re-creation. The periphery, in summary, is a valuable space to find what one is not expecting to find. This attitude of exploration with no intention other than exposing oneself to unplanned discoveries may thus constitute a wellspring of serendipitous emergence. Many organizations appear to have developed a cultivated distaste for the periphery because doing so is not efficient nor directed, but these are precisely the reasons why the periphery matters.

Responding to Serendipitous Opportunities Organizations may need to be able to respond to serendipity, as responding will be important to transform potentiality into discovery. Serendipity therefore requires not only perspicacity but also action and agency. To explore the process of responding, we consider improvisation, bricolage, error, and sites for innovation. Improvisation. As Heifetz and Linsky (2002, p. 66) have pointed out, “leadership is an improvisational art. You may be guided by an overarching vision, clear values, and a strategic plan, but what you actually do from moment to moment cannot be scripted.” To reap the fruits of serendipitous emergence, organizations need to be able to make planning and execution converge, i.e. they need to improvise in order to take advantage of fleeting serendipitous opportunities (Cunha et al. 1999) and to create forms of collective reflexivity that allow them to respond to identified opportunities (Abrantes et al. 2021). Improvisation, the substantive deliberate fusion of the design and execution of a novel production in an organisation, drawing on available resources (Cunha et al. 2017), is necessary, to move fast when a window of opportunity announces itself, or to respond to an unexpected threat such as a new coronavirus (Giustiniano et al. 2020). Dealing with improvisation implies a capacity to handle the unscripted dimensions of organizational functioning. The presence of improvisation in face of serendipitous emergence has been studied in cases such as Honda’s entry in the U.S. market (Cunha et al. 2015) or in IKEA’s embrace of the open warehouse (Jarrett and Huy 2018). The need to move fast because there is no time to devise new plans turns improvisation into a portal to serendipitous discovery. Bricolage. A reason why improvisation may lead to serendipity is because of its bricolage component. Not every instance of bricolage occurs in the context of improvisation but the presence of bricoleurs, people with intimate knowledge of

4 Serendipity in Management and Organization Studies

59

materials and willing to explore alternative uses, constitutes a stimulus for serendipitous discovery (Garud et al. 2018). The process was studied by authors such as Baker and Nelson (2005) and Busch and Barkema (2021), who explained how the creative engagement with materials allows entrepreneurs to make discoveries and even to gain organizational scale via the adoption of processes with serendipitous components. Bricolage implies a level of intimacy with materials that expands their affordances and enriches the organization’s knowledge of resources. Such intimacy with materials as well as the capacity to develop a tacit knowledge resulting from knowing how things work has been associated with the capacity to innovate. The ensuing innovations can be purposeful (Nonaka and Takeuchi 2007) but they may also originate occasions for serendipity. This may even follow what was originally perceived as an error. Errors. Errors are often viewed as expressions of incompetent actuation. In fact, honest mistakes can be instead understood as the outcome of explorations at the frontier of knowledge. When practices are pushed to the limits of the habitual and of established conventions, errors may happen but they may create opportunities for unplanned discoveries. Because mistakes are treated as learning imperatives, the system is able to transform mistakes into sources of learning from unexpected occurrences. Some of the most remarkable forms of serendipity have been those that explored the system to the extreme, testing its boundaries. This is sometimes literal, as in the case of the Columbus arrival in America, then unknown to the rest of the world, but also in cases in which materials are used creatively by bricoleurs to discover possibilities that may convert errors and experiments into sources of discovery (Baker and Nelson 2005). It is possible to hypothesize that errors are a fundamental ingredient of serendipitous emergence and discovery. Learning from errors is demonstrably important for organizational learning (Rashid et al. 2013), including of the serendipitous type. Instead of treating errors as faulty occurrences rendering the discovery of the culprit necessary, they can be windows into the world of serendipity. Sites of innovation. Organizations may also try to create opportunities for serendipitous emergence by designing serendipity-friendly spaces, by articulating different innovation sites and connecting communities of practice and interest. In this case, instead of limiting innovation to the space within the confines of a formal organizational boundary, the process is opened to outside participants, a form of collective brokering, as explained by Sauer and Bonelli (2020). This can be expressed in several ways. In some cases, organizations can expand their reach by opening innovation to outside agents by engaging the crowd (Oliveira and Cunha 2021) and adopting open forms of innovation (Chesbrough 2020). In this way organizations may obtain fast and unexpected inputs, namely from non-experts in critical moments such as an unfolding pandemic. Non-experts may bring innovative and unorthodox solutions because of, precisely, their lack of expertise. New sites for innovation, such as digital platforms, offer new forms of learning. Instead of planning, digital entrepreneurs expose themselves to the market in order to learn more about it on the basis of serendipitous discoveries. Given the nature of

60

M. P. Cunha and M. Berti

serendipity, even the design of spaces promotive of serendipitous encounters may not necessarily promote serendipity (Irving et al. 2020).

Future of Serendipity in Organization Studies Hafsi and Thomas (2005, p. 517) have argued that “the field of strategy has no future except to become close to reality”. We suggest that serendipity (and its unfortunate twin, zemblanity; see Giustiniano et al. 2016) may be one of the ways to get the discipline of MOS close to reality. The ideal of organizations as being in full control of their actions and of the evolution of their environments is an assumption globally dismissed by, for instance, a new coronavirus such as the one that caused the Covid-19 pandemic. The pandemic and consequent movement out of offices to work from home created obstacles to “important serendipitous face-to-face contact”, as observed by WL Gore’s strategic planner Annemarie Nicholson (in Hill 2020, p.15), but it might also create opportunities for other unwanted discoveries such as the possibility of new hybrid forms of work. It is better, in summary, to assume that reality cannot be controlled/predicted and that luck, happenstance and serendipity should be acknowledged by management and organization scholars. To speculate about the possible future of serendipity, we consider four possibilities that we call the four ‘P’s of serendipity. There is nothing particular about being four or Ps, but these themes allow us to compose possible entries for a future research agenda. These themes are: process, power, paradox and performativity. We do not claim that these themes exhaust the possibilities, only that they may expand our knowledge. But all these ideas are implied directly or indirectly in our definition of the concept. Process. We defined serendipity as an effortful process. This is an important element of the definition and one that deserves further scrutiny. If we take serendipity as an a-ha or a eureka type of phenomenon, something that happens explosively as a sudden event, we ignore its important processual nature. By taking serendipity as a process, one appreciates its unfolding over time, an in situ phenomenon that is affected by past choices and events. Words such as discovery, incubation, and effort, all incorporate a process dimension. They imply time and continuity. It is for this reason that we need to know more about the process, the sequence of actions in time and place that conduct serendipity. Cases such as Honda’s entry in the US market have a clear processual resonance (Mintzberg et al. 1996). Therefore, process studies may help to enrich the understanding of serendipity and serendipity may help illuminate process—especially if process is approached through the perspective of negative capability rather than linear teleology. Power. Serendipity, as the process that leads to an unsought and unplanned discovery, means that the agents involved will need to accept the unsought nature of the discovery. Honda assumed a strategy that was not the one planned beforehand. Yet the improvised strategy worked because people were determined and pragmatic, not because they were prescient. The courage to follow a serendipity-informed unplanned

4 Serendipity in Management and Organization Studies

61

strategy is, to some extent, a political act. In some contexts, it may be better to avoid the sort of experiments leading to serendipity rather than embrace them. Serendipity may be perceived by the formal structure as lack of control and, thus, as an unacceptable deviation from plans. It is thus possible to consider that to understand serendipity it will be necessary to explore the organizational power circuitry (Clegg 1989) where improvisation is nurtured or discouraged. More than a strictly technical endeavor, serendipity has important political dimensions (see also Soto, Chap. 11, Yadamseren Chap. 2, and Reviglio Chap. 8, in this volume). Understanding the politics of serendipity will be necessary to explain serendipity itself. Paradox. The notion of serendipity is paradoxical, as it involves a persistent tension between knowledge and ignorance. Paradox refers to the persistent opposition of mutually defining and mutually excluding contraries (Berti et al. 2021; Schad et al. 2016). This involves a tension between knowing and not knowing: discovery depends on previous knowledge but is activated to expand knowledge into unknown and unmapped territory. At the core of serendipity there is a paradoxical blend of exploitation and exploration (March 1991), using what one knows to discover what one does not. Future research can thus explore this interplay and how it unfolds over time: how does knowledge illuminate ignorance and how does ignorance create the impetus to engage in further knowledge creation? The lens of paradox may offer important insights on the nature of serendipity in organizations, namely in terms of persistence: how does the persisting tension between knowledge and ignorance contribute to the understanding of the attitude towards serendipity? (See Arfini, Chap. 7 of this volume, for a suggestion on how we can answer this question). Performativity. Finally, there is in serendipity an element of performativity that needs to be addressed. By assuming that serendipity is an acceptable approach, organizational agents may potentially start to look for opportunities to performatively enact serendipity. They may turn their theories on serendipity self-fulfilling (Marti and Gond 2018), by noticing anomalies and framing them as opportunities for serendipitous discovery. This suggests that it is not only the facts but also the language about the facts that needs to be considered in order to make sense of serendipity. Serendipity requires a specific mindset with the right words for it to be acceptable. ‘Luck’ and ‘serendipity’, for example are two very different ways of representing a given organizational phenomenon. We need to know more about the words used to communicate about events with serendipitous properties. An appreciation for the word ‘serendipity’ or a dislike of it may produce very different organizational consequences. Words do matter and, in a performative tradition of doing things with words (Austin 1975), the word ‘serendipity’ itself may be consequential. Which words in which contexts, render the context felicitous for serendipity to be taken seriously? These are only four possible themes to explore with regards to serendipity. The pathways they might offer will certainly contribute to understanding serendipity as something that people do in their circumstance rather than some mysterious event that happens without clear explanation. Theory being the explanation of why (Sutton and Staw 1995), we anticipate that novel theorizing can result from these four lines of inquiry, explaining why serendipity takes place and why organizational ripeness may inscribe serendipity in the strategy making process, as an ingredient of the

62

M. P. Cunha and M. Berti

strategy process (more on this in Busch and Grimes, Chap. 5 in this volume). At the same time, one needs to be aware of the political and social determinants of theorizing: theory is a collective construction and the way serendipity is treated in MOS involves considering the dynamics regulating ongoing research programs and their legitimation of serendipity or not (Clegg et al. 2020).

Conclusion There is an already rich literature on how to manage surprises, which often considers how to anticipate surprises in order not to be surprised by predictable surprises (Watkins and Bazerman, 2003). Some authors, however, have recommended the development of plans sensitive to surprises (Makridakis et al. 2010). We followed a different path: instead of inviting managers to plan to incorporate surprise in their planning, we offered an invitation to explore serendipity as negative capability. Serendipity may help managers and organizations to prepare for addressing challenges in an uncertain world. Instead of accepting that great organizing involves the exclusion of serendipitous emergence, we invite scholars to embrace serendipity as a window into new possibilities. Serendipity should thus not be excluded but embraced by management and organization studies. As a form of engagement with the surprising and the unexpected, it may be indispensable to operate in a surprising world. Acknowledgements We thank our editors and Christian Busch for their feedback on earlier drafts of this chapter. Miguel Cunha acknowledges support by Fundação para a Ciência e a Tecnologia (UID/ECO/00124/2019, UIDB/00124/2020 and Social Sciences DataLab, PINFRA/22209/2016), POR Lisboa and POR Norte (Social Sciences DataLab, PINFRA/22209/2016)

References Abrantes, A.C.M., A.M. Passos, M.P. Cunha, and C.M. Santos. 2021. Getting the knack for teamimprovised adaptation: The role of reflexivity and team mental model similarity. Journal of Applied Behavioral Science. Acemoglu, D., and J.A. Robinson. 2012. Why nations fail: The origins of power, prosperity, and poverty. New York: Currency. Anderson, P. 1999. Perspective: Complexity theory and organization science. Organization Science 10 (3): 216–232. Ansoff, H.I. 1975. Managing strategic surprise by response to weak signals. California Management Review 18 (2): 21–33. Ashforth, B.E., and V. Anand. 2003. The normalization of corruption in organizations. Research in Organizational Behavior 25: 1–52. Austin, J.L. 1975. How to do things with words. Oxford: Oxford University Press. Baker, T., and R.E. Nelson. 2005. Creating something from nothing: Resource construction through entrepreneurial bricolage. Administrative Science Quarterly 50 (3): 329–366.

4 Serendipity in Management and Organization Studies

63

Bansal, P., A. Kim, and M.O. Wood. 2018. Hidden in plain sight: The importance of scale in organizations’ attention to issues. Academy of Management Review 43 (2): 217–241. Berti, M., and A.V. Simpson. 2021. The dark side of organizational paradoxes: The dynamics of disempowerment. Academy of Management Review 46 (2): 252–274. Berti, M., and A.V. Simpson. 2020. On the practicality of resisting pragmatic paradoxes. Academy of Management Review, forthcoming. Berti, M., A.V. Simpson, S. Clegg, and M.P. Cunha. 2021. Elgar introduction to organizational paradox theory. Cheltenham: Edward Elgar. Birkinshaw, J., M. Gudka, and V. D’Amato. 2021. The blinkered boss: How has managerial behavior changed with the shift to virtual working? California Management Review 63 (4): 5–26. Bowles, M.L. 1990. Recognizing deep structures in organizations. Organization Studies 11 (3): 395–412. Brown, S.L., and K.M. Eisenhardt. 1998. Competing on the edge: Strategy as structured chaos. Boston, MA: Harvard Business Press. Brunsson, N., and B. Jacobsson. 2010. A world of standards. Oxford: Oxford University Press. Burns, T., and G.M. Stalker. 1961. The organization of innovation. Tavistock. Busch, C., and H. Barkema. 2020. Planned luck: How incubators can facilitate serendipity for nascent entrepreneurs through fostering network embeddedness. Entrepreneurship Theory and Practice, 1042258720915798. Busch, C., and H. Barkema. 2021. From necessity to opportunity: Scaling bricolage across resourceconstrained environments. Strategic Management Journal 42 (4): 741–773. Busch, C. & Grimes, M. (this volume). Serendipity in entrepreneurship strategy, and innovation: A review and conceptualization. Capelli, P. 2020. Stop overengineering people management. Harvard Business Review 98 (5): 56–63. Chesbrough, H. 2020. To recover faster from Covid-19, open up: Managerial implications from an open innovation perspective. Industrial Marketing Management 88: 410–413. Chia, R., and S. Morgan. 1996. Educating the philosopher-manager: De-signing the times. Management Learning 27 (1): 37–64. Clegg, S.R. 1989. Frameworks of power. London: Sage. Clegg, S., M.P. Cunha, I. Munro, A. Rego, and M.O. Sousa. 2016. Kafkaesque power and bureaucracy. Journal of Political Power 9 (2): 157–181. Clegg, S.R., M.P. Cunha, and M. Berti. 2020. Research movements and theorizing dynamics in management and organization studies. Academy of Management Review, 47 (3): 382–401. https:/ /doi.org/10.5465/amr.2018.0466. Cohen, M.D., J.G. March, and J.P. Olsen. 1972. A garbage can model of organizational choice. Administrative Science Quarterly 17 (1): 1–25. Copeland, S. 2019. On serendipity in science: Discovery at the intersection of chance and wisdom. Synthese 196 (6): 2385–2406. Coyne, W.E., and A.H. Van de Ven. 2021. Increasing the odds of maneuvering the innovation journey. Strategic Management Review. Cunha, M.P., and R. Chia. 2007. Using teams to avoid peripheral blindness. Long Range Planning 40 (6): 559–573. Cunha, M.P., S.R. Clegg, and S. Mendonça. 2010. On serendipity and organizing. European Management Journal 28 (5): 319–330. Cunha, M.P., S.R. Clegg, and A. Rego. 2012. Surprising organization. In Handbook of organizational and managerial innovation, ed. T. Pitsis, A. Simpson, and E. Dehlin, 295–316. Cheltenham: Edward Elgar. Cunha, M.P., A. Rego, S. Clegg, and G. Lindsay. 2015. The dialectics of serendipity. European Management Journal 33 (1): 9–18. Cunha, M.P., S. Clegg, A. Rego, and M. Berti. 2021. Paradoxes of power and leadership. London: Routledge. Cunha, M.P., J.V. Cunha, and K. Kamoche. 1999. Organizational improvisation: What, when, how and why. International Journal of Management Reviews 1 (3): 299–341.

64

M. P. Cunha and M. Berti

Cunha, M.P., A.S. Miner, and E. Antonacopolou. 2017. Improvisation processes in organizations. In The Sage handbook of process organization studies, ed. A. Langley and H. Tsoukas, 559–573. Los Angeles; CA: Sage. Dane, E. 2010. Reconsidering the trade-off between expertise and flexibility: A cognitive entrenchment perspective. Academy of Management Review 35 (4): 579–603. Dane, E. 2020. Suddenly everything became clear: How people make sense of epiphanies surrounding their work and careers. Academy of Management Discoveries 6 (1): 39–60. Day, G.S., and P.J. Schoemaker. 2006. Peripheral vision: Detecting the weak signals that will make or break your company. Boston, MA: Harvard Business Press. Day, G.S., and P.J. Schoemaker. 2008. Are you a vigilant leader’? MIT Sloan Management Review 49 (3): 43–51. De Rond, M., and J. Lok. 2016. Some things can never be unseen: The role of context in psychological injury at war. Academy of Management Journal 59 (6): 1965–1993. Dweck, C.S. 2008. Mindset: The new psychology of success. New York: Random House. Edmondson, A.C. 2018. The fearless organization: Creating psychological safety in the workplace for learning, innovation, and growth. New York: Wiley. Erikson, S.L. 2018. Cell phones /= self and other problems with big data detection and containment during epidemics. Medical Anthropology Quarterly 32 (3): 315–339. Feldman, M.S., and B.T. Pentland. 2003. Reconceptualizing organizational routines as a source of flexibility and change. Administrative Science Quarterly 48: 94–118. Fisher, C., O. Demir-Caliskan, M.Y. Hua, and M.A. Cronin. 2021. Trying not to try: The paradox of intentionality in jazz improvisation and its implications for organizational scholarship. In Interdisciplinary dialogues on organizational paradox. Research in the Sociology of Organizations, ed. R. Bednarek, M.P. Cunha, J. Schad, W.K. Smith (vol. 73B, 125–139). Emerald. Fleming, P. 2019. Robots and organization studies: Why robots might not want to steal your job. Organization Studies 40 (1): 23–38. Garud, R., J. Gehman, and A.P. Giuliani. 2018. Serendipity arrangements for exapting science-based innovations. Academy of Management Perspectives 32 (1): 125–140. Garud, R., and M. Turunen. 2021. Microfoundations of innovation as process: Usher’s cumulative synthesis model. In The Oxford handbook of organizational change and innovation, ed. by M.S. Poole, A.H. Van de Ven. Oxford University Press. Giles, D., and L. Walkowicz. 2019. Systematic serendipity: A test of unsupervised machine learning as a method for anomaly detection. Monthly Notices of the Royal Astronomical Society 484 (1): 834–849. Giustiniano, L., M.P. Cunha, A.V. Simpson, A. Rego, and S. Clegg. 2020. Resilient leadership as paradox work: Notes from COVID-19. Management and Organization Review 16 (5): 971–975. Giustiniano, L., M.P. Cunha, and S. Clegg. 2016. Organizational zemblanity. European Management Journal 34 (1): 7–21. Gratton, L. 2020. The pandemic’s lessons for managing uncertainty. MIT Sloan Management Review. https://sloanreview.mit.edu/article/the-pandemics-lessons-for-managing-uncertainty/. Hafsi, T., and H. Thomas. 2005. The field of strategy: In search of a walking stick. European Management Journal 23 (5): 507–519. Hahn, T., and E. Knight. 2021. The ontology of organizational paradox: A quantum approach. Academy of Management Review 46 (2): 362–384. Harrison, R.T., C.M. Leitch, and R. Chia. 2007. Developing paradigmatic awareness in university business schools: The challenge for executive education. Academy of Management Learning & Education 6 (3): 332–343. Heifetz, R.A., and M. Linsky. 2002. A survival guide for leaders. Harvard Business Review 80 (6): 65–74. Hill, A. 2020. Business in the time of Covid. Financial Times, 29 December, 15. Ibarra, H., A. Rattan, and A. Johnston. 2018. Satya Nadella at Microsoft: Instilling a growth mindset. London Business School case CS-18-008.

4 Serendipity in Management and Organization Studies

65

Irving, G.L., O.B. Ayoko, and N.M. Ashkanasy. 2020. Collaboration, physical proximity and serendipitous encounters: Avoiding collaboration in a collaborative building. Organization Studies 41 (8): 1123–1146. Jarrett, M., and Q.N. Huy. 2018. IKEA’s success can’t be attributed to one charismatic leader. Harvard Business Review, online 2 February Jones, B.F., and B.A. Olken. 2009. Hit or miss? The effect of assassinations on institutions and war. American Economic Journal: Macroeconomics 1 (2): 55–87. Keats, J. 1958. The letters of John Keats, 1814– 1821, ed. Hyder Edward Rollins. Cambridge, MA: Harvard University Press. Kendall, J.E., and K.E. Kendall. 1993. Metaphors and methodologies: Living beyond the systems machine. MIS Quarterly 17 (2): 149–171. Kim, B., S.H. Jung, S.H. Jang, B. Lee, E. Rhee, S.H. Cho, and S.M. Lee. 2014. Construction and initial validation of the Planned Happenstance Career Inventory. The Career Development Quarterly 62 (3): 239–253. Krumboltz, J.D. 1998. Serendipity is not serendipitous. Journal of Clinical Counseling 45 (4): 390–302. Lakatos, I. 1978. The methodology of scientific research programmes, J. Worrall, Trans. Cambridge: Cambridge University Press. Lane, J.N., I. Ganguli, P. Gaule, E. Guinan, and K.R. Lakhani. 2020. Engineering serendipity: When does knowledge sharing lead to knowledge production? Strategic Management Journal. https:/ /doi.org/10-1002/smj.3256. Leavitt, K., K. Schabram, P. Hariharan, and C.M. Barnes. 2020. Ghost in the machine: On organizational theory in the age of age machine learning. Academy of Management Review. Levien, R., and S. Tan. 1993. Double pendulum: An experiment in chaos. American Journal of Physics 61 (11): 1038–1044. Liu, C., and M. De Rond. 2016. Good night, and good luck: Perspectives on luck in management scholarship. Academy of Management Annals 10 (1): 409–451. Locke, K., K. Golden-Biddle, and M.S. Feldman. 2008. Perspective—Making doubt generative: Rethinking the role of doubt in the research process. Organization Science 19 (6): 907–918. Mackay, R.B., R. Chia, and A. Nair. 2020. Strategy-in-practices: A process philosophical approach to understanding strategy emergence and organizational outcomes. Human Relations, 74 (9), 1337–1369. Mainemelis, C. 2010. Stealing fire: Creative deviance in the evolution of new ideas. Academy of Management Review 35 (4): 558–578. Makridakis, S., R.M. Hogarth, and A. Gaba. 2010. Why forecasts fail. What to do instead. MIT Sloan Management Review 51 (2): 83. March, J.G. 1991. Exploration and exploitation in organizational learning. Organization Science 2 (1): 71–87. Marti, E., and J.P. Gond. 2018. When do theories become self-fulfilling? Exploring the boundary conditions of performativity. Academy of Management Review 43 (3): 487–508. Merton, R.K. 1936. The unanticipated consequences of purposive social action. American Sociological Review 1 (6): 894–904. Merton, R.K., and E.G. Barber. 2004. The travels and adventures of serendipity: A study in sociological semantics and the sociology of science. Oxford: Princeton University Press. Mintzberg, H. 2000. The rise and fall of strategic planning. New York: Pearson Education. Mintzberg, H., R.T. Pascale, M. Goold, and R.P. Rumelt. 1996. CMR forum: The “Honda effect” revisited. California Management Review 38 (4): 77–117. Mitchell, K.E., S. Al Levin, and J.D. Krumboltz. 1999. Planned happenstance: Constructing unexpected career opportunities. Journal of Counseling & Development 77 (2): 115–124. Miron-Spektor, E., A. Ingram, J. Keller, W.K. Smith, and M.W. Lewis. 2018. Microfoundations of organizational paradox: The problem is how we think about the problem. Academy of Management Journal 61 (1): 26–45. Morgan, G. 1986. Images of organization. Beverly Hills, CA: Sage.

66

M. P. Cunha and M. Berti

Nayak, A., R. Chia, and I. Canales. 2020. Noncognitive microfoundations: Understanding dynamic capabilities as idiosyncractically refined sensitivities and predispositions. Academy of Management Review 45 (2): 280–303. Nonaka, I., and H. Takeuchi. 2007. The knowledge-creating company. Harvard Business Review 85 (7/8): 162. Ohno, T. 1988. Toyota production system: Beyond large-scale production. Portland, OR: Productivity Press. Oliveira, P., and M.P. Cunha. 2021. Centralized decentralization or distributed leadership as paradox: The case of the Patient Innovation’s Covid-19 Portal. Journal of Change Management 21 (2): 203–221. Ou, L. 2011. Keats and negative capability. London: Bloomsbury. Parmigiani, A., and J. Howard-Grenville. 2011. Routines revisited: Exploring the capabilities and practice perspectives. Academy of Management Annals 5 (1): 413–453. Pentland, B.T., and M.S. Feldman. 2005. Organizational routines as a unit of analysis. Industrial and Corporate Change 14 (5): 793–815. Power, M. 1997. The audit society: Rituals of verification. Oxford: Oxford University Press. Prahalad, C.K., and R.A. Bettis. 1986. The dominant logic: A new linkage between diversity and performance. Strategic Management Journal 7 (6): 485–501. Rashid, F., A.C. Edmondson, and H.B. Leonard. 2013. Leadership lessons from the Chilean mine rescue. Harvard Business Review 91 (7–8): 113–119. Roberts, K.H. 1990. Managing high reliability organizations. California Management Review 32 (4): 101–113. Roberts, P.W., and K.M. Eisenhardt. 2003. Austrian insights on strategic organization: From market insights to implications for firms. Strategic Organization 1 (3): 345–352. Robbins, R., and B. Mueller. 2020. After admitting mistake, AstraZeneca faces difficult questions about its vaccine. The New York Times, 25 November, nyt.com. Ross, W., and F. Vallée-Tourangeau. 2020. Microserendipity in the creative process. The Journal of Creative Behavior 55 (3): 661–672. Sauer, S., and F. Bonelli. 2020. Collective improvisation as a means to responsibly govern serendipity in social innovation processes. Journal of Responsible Innovation, 7 (sup2): S44–S63. Schad, J., M.W. Lewis, S. Raisch, and W.K. Smith. 2016. Paradox research in management science: Looking back to move forward. Academy of Management Annals 10 (1): 5–64. Schoeneborn, D. 2011. Organization as communication: A Luhmannian perspective. Management Communication Quarterly 25 (4): 663–689. Shenhav, Y.A. 2002. Manufacturing rationality: The engineering foundations of the managerial revolution. Oxford: Oxford University Press. Smith, W.K., and M.W. Lewis. 2011. Toward a theory of paradox: A dynamic equilibrium model of organizing. Academy of Management Review 36 (2): 381–403. Strike, V.M., and C. Rerup. 2016. Mediated sensemaking. Academy of Management Journal 59 (3): 880–905. Sutton, R.I., and B.M. Staw. 1995. What theory is not. Administrative Science Quarterly 40 (3): 371–384. Taylor, F.W. 1911. Principles of scientific management. New York: W.W. Norton. Teece, D.J. 2007. Explicating dynamic capabilities: The nature and microfoundations of (sustainable) enterprise performance. Strategic Management Journal 28 (13): 1319–1350. Thompson, J.D. 1967. Organizations in action: Social science bases of administrative theory, 1967. New York: McGraw-Hill. Tsoukas, H. 2005. Complex knowledge: Studies in organizational epistemology. Oxford: Oxford University Press. Tsoukas, H. 2017. Don’t simplify, complexify: From disjunctive to conjunctive theorizing in organization and management studies. Journal of Management Studies 54 (2): 132–153. Tsoukas, H., and R. Chia. 2002. On organizational becoming: Rethinking organizational change. Organization Science 13 (5): 567–582.

4 Serendipity in Management and Organization Studies

67

Unger, R. 2004. False necessity: Anti-necessitarian social theory in the service of radical democracy. London: Verso. Volberda, H.W., S. Khanagha, C. Baden-Fuller, O.R. Mihalache, and J. Birkinshaw. 2021. Strategizing in a digital world: Overcoming cognitive barriers, reconfiguring routines and introducing new organizational forms. Long Range Planning, 54 (5): 102110. Von Krogh, G. 2020. Building capacity for empirical discovering in management and organization studies. Academy of Management Discoveries 6 (2): 159–164. Walker, M., P. Fleming, and M. Berti. 2021. ‘You can’t pick up a phone and talk to someone’: How algorithms function as biopower in the gig economy. Organization 28 (1): 26–43. Watkins, M.D., and M.H. Bazerman. 2003. Predictable surprises: The disasters you should have seen coming. Harvard Business Review 81 (3): 72–85. Weick, K.E. 2010. Reflections on enacted sensemaking in the Bhopal disaster. Journal of Management Studies 47 (3): 537–550. Weick, K.E. 2020. Sensemaking, organizing, and surpassing: A handoff. Journal of Management Studies, 57 (7): 1420–1431. https://doi.org/10.1111/joms.12617. Yaqub, O. 2018. Serendipity: Towards a taxonomy and a theory. Research Policy 47 (1): 169–179.

Miguel Pina e Cunha is the Fundação Amélia de Mello professor at Nova School of Business and Economics, Universidade Nova de Lisboa (Portugal). He studies organization as process and paradox and has recently co-edited Management, Organizations and Contemporary Social Theory (Routledge 2019) and, with Marco Berti among other colleagues, co-authored Paradoxes of Power and Leadership (Routledge, 2021) and Elgar’s Introduction to Organizational Paradox Theory (Edward Elgar, 2021). Marco Berti is Senior Lecturer in Management at UTS Business School, University of Technology Sydney. His research has been published in journals such as Academy of Management Review, Academy of Management Learning and Education, Management Learning, Organization, Journal of Management Inquiry, among others. He also authored Elgar’s Introduction to Organizational Discourse Analysis (Edward Elgar, 2018) and co-authored Paradoxes of Power and Leadership (Routledge, 2021) and Elgar’s Introduction to Organizational Paradox Theory (Edward Elgar, 2021).

Chapter 5

Serendipity in Entrepreneurship, Strategy, and Innovation—A Review and Conceptualisation Christian Busch and Matthew Grimes

Abstract Serendipity is at the core of many innovations, inventions, and entrepreneurial opportunities. However, despite its importance for organisations and individuals alike, research on the dimensions and antecedents of serendipity is surprisingly scarce. In this chapter, Christian Busch and Matthew Grimes review and synthesize research on serendipity in the entrepreneurship, strategy, and innovation context, and suggest a novel conceptualisation of the process of (cultivating) serendipity. They thereby provide the reader with a thorough and wide-ranging view of how serendipity has come into the fore in the field of organization and management, but also what possibilities it opens up for understanding and creating the conditions for entrepreneurial success. They advance a process-oriented model of serendipity that serves as a basis to elaborate factors that increase the chances for serendipitous encounters and how to capitalize on them. Amongst those, Busch and Grimes distinguish between individual (including reframing, extrovertedness and perseverance) and organizational factors (including systematic evaluations, iteration and team-based collaboration). Their paper, thereby, advances the conceptual understanding of serendipity as much as a theory of how to transfer this understanding successfully into the entrepreneurial context.

C. Busch (B) Marshall School of Business, University of Southern California, 3670 Trousdale Pkwy, Los Angeles, California 90089, USA e-mail: [email protected] M. Grimes Cambridge Judge Business School, University of Cambridge, Trumpington Street, Cambridge CB2 1AG, UK © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 S. Copeland et al. (eds.), Serendipity Science, https://doi.org/10.1007/978-3-031-33529-7_5

69

70

C. Busch and M. Grimes

Introduction “Chance favors the prepared mind only.” (Louis Pasteur) “The best education is one that prepares you for your own venture into the unknown.” (Lee C. Bollinger, President, Columbia University) Both entrepreneurs and organisational leaders tend to assume that market opportunities can be mapped out in advance, such that the process of strategy is frequently focused on developing targets and plans (Brown 2005). This focus is undergirded by the premises that individuals and organisations are able to anticipate possible outcomes a priori, and that activities and interactions can subsequently be coordinated around stable “strategic” objectives such as seeking particular resources (Engel et al. 2017; Hallen and Eisenhardt 2012). However, despite the wide acceptance of such premises, research into the practice of strategy and entrepreneurship offers evidence that in a fast-changing world it is often difficult to deliberately determine which resources, partners, or co-founders might be needed in the future (Busch 2021; Engel et al. 2017). As such, seminal studies have noted that firms’ and entrepreneurs’ strategies are often best understood as “emergent” (Mintzberg and Waters 1985; Mintzberg et al. 1996; Sarasvathy 2008), wherein intentionality is unclear and any corresponding formalised plans arise not as advanced directives but rather as tools for rationalising and justifying current action. In this way, positive outcomes such as opportunity discovery and (social) innovations and inventions (e.g., Viagra, microwaves, or postit-notes) are often a matter of serendipity rather than planning (Denrell et al. 2003, 2015; Grimes et al. 2019; Liu and de Rond 2016; Ramus et al. 2017). Building on a recent systematic review on serendipity in the management context (Busch 2022), we define serendipity as a surprising and valuable discovery originating from agentic responses to unplanned events. Thus, rather than being merely an event that happens to an individual or organisation, serendipity requires sagacity— i.e., it builds on the notion that positive discoveries are facilitated by “controllable” elements such as an open mind (Makri et al. 2014; Merton and Barber 2004; van Andel 1994). And yet, while serendipity as a concept has been occasionally referenced by strategy researchers (e.g., Graebner 2004; Kilduff and Tsai 2003), most prior research in entrepreneurship and management has interpreted serendipity as an exogenous structural, and thus uncontrollable, feature of spontaneous encounters (Casciaro et al. 2014; Feld 1981; Shipilov et al. 2014), and sometimes even as an “error” or type of “uncertainty” that needs to be avoided, rather than as something that can be beneficially managed (Brown 2005; Engel et al. 2017). Such depictions within the strategy and entrepreneurship literature of serendipity as uncontrollable may explain the clear divergence between the lack of academic exploration on the topic of serendipity and the frequency and consistency with which practitioners credit it for their success (Busch 2020a; Gyori et al. 2019). Given the seeming mismatch between the (theoretical and empirical) importance of the phenomenon and the lack of research on the topic, we embarked on

5 Serendipity in Entrepreneurship, Strategy, and Innovation—A Review …

71

an exploration of the role of serendipity in entrepreneurship, strategy, and innovation. Although our chapter is focused on serendipity as a general phenomenon, we also recognise prior distinctions from the literature that differentiate at least three types of serendipity based on the nature of the search process as well as the relation between the emergent solution and that search process (Busch 2020a; Yaqub 2018; also see Napier and Vuong 2013). They can broadly be clustered into three types: 1. Thunderbolt serendipity. No search for a solution to a specific problem is under way, but the actor unexpectedly (“thunderbolt”) comes across a new problemsolution dyad, often conceptualised as an ‘opportunity’. The problem and solution thus unexpectedly emerge at the same time, like in the example of the rolling suitcase: A traveling luggage company worker observed an employee in the airport, who rolled a heavy machine on a wheeled skid, while the traveler had to drag his heavy suitcases through the airport. When he realised that he could mount furniture casters on a travel suitcase, and put a strap on the front, the rolling suitcase was born (von Hippel and von Krogh 2016). 2. Archimedes serendipity. A search for a solution to a known problem is under way, but the solution comes from an entirely unexpected place. Example: in the apocryphal tale, Archimedes was trying to find out whether his king’s crown was made of pure gold, yet he unexpectedly found his answer when watching the water level rise as he lowered himself into a public bath, realising that submerging the crown in water could give him the solution to his problem (Busch 2020a). 3. Post-it note serendipity. A search for a solution to a known problem is underway, but in the process the social actor stumbles across a solution to a previously unrecognised or entirely different problem. Example: An inventor at 3M was initially looking for a stronger glue, but unexpectedly realised that a weaker glue, used in a different way, could result in an effective product.1 What unites each of these different types of serendipity is the presence of some unexpected event or trigger, a subsequent noticing and bracketing of weak cues, followed by the socio-cognitive and cultural effort involved in connecting that information to a potential problem or solution. And such serendipitous processes can be contrasted with more rational or non-serendipitous problem-solving processes, in which the actor has a clear initial problem, a search process that is directed toward proposing one or more solutions to that problem, while filtering out seemingly peripheral and/or unrelated information (Busch and Grimes 2023; Grimes and Vogus 2021; 1

Some researchers have differentiated between “real” and “fake”/“pseudo” serendipity (e.g., de Rond 2014), others (such as Dew 2009) focus on existing search. For example, a few researchers have contended that examples such as penicillin are “pseudo-serendipitous”. In this view, pseudoserendipity is about a situation in which you are looking for something already, and then come across something coincidentally that helps you reach the initial goal. In the case of Fleming’s penicillin, the team was somewhat prepared, as they were already interested in the antibiotic effects of substances. In this logic, “true” serendipity would require a change in objective (Roberts 1989). However, most researchers do not share this narrow notion, and rather look at serendipity in the broader sense—else, most of the documented serendipity stories would be “pseudo-serendipitous” (also see Busch 2022; Copeland 2018). In this paper, based on recent research, we cover the whole spectrum.

72

C. Busch and M. Grimes Serendipity Trigger

Bisociation*

Enactment Unexpected Positive Outcome

Systemic Enablers Expanded Search (von Hippel & von Krogh, 2016) Physical and Virtual Task Environment (McCay-Peet & Toms, 2010; Reinecke et al., 2010) Expanded Networks (Busch & Barkema, 2020)

Systemic enablers Collaboration (Cunha et al., 2010; Meyers, 2007) Theories of Value (Felin & Zenger, 2015; Zenger, 2013) Problem formulation (von Hippel & von Krogh, 2016)

Systemic Enablers Systematic Evaluations (Napier & Vuong, 2013) Corporate Culture (Cunha et al., 2010; de Rond, 2014)

Individual Enablers Extroversion (Wiseman, 2003) Curiosity and alertness (Diaz de Chumaceiro, 2004; Napier & Vuong, 2013) Positive Emotions (Baron, 2008; Helfat & Peteraf, 2015) Self-awareness (Daneels, 2011; Gyori et al., 2019) Humility (Gyori et al., 2019) Improvisation (Cunha et al., 2014)

Individual enablers Creativity (Christoff et al., 2009; Mason et al., 2007; Stock et al., 2017) Re-framing (Reinecke & Ansari, 2015; Wiseman, 2003) Analogous Thinking (Gentner & Markman, 1997; Gick & Holyoak, 1980)

Individual Enablers Perseverance (Austin 1978; Burgelman, 2003) Social skill (Busch, 2020)

Fig. 5.1 The process of (cultivation) serendipity

Yaqub 2018). They can also be contrasted against “garbage-can” models of decisionmaking or effectual models of innovation in which the actor starts with existing solutions, resources, or other means and subsequently searches for problems that might be addressed by way of those means (Cohen et al. 1972; Sarasvathy 2008). Throughout the article, we argue that for serendipity to be more systematically incorporated into entrepreneurship and business strategy, social actors must seek to increase (a) the likelihood of trigger events, (b) the likelihood of noticing and bracketing weak cues, and (c) cultural appreciation for and support structures which help to materialise unconventional solutions within and across organisations (Busch 2022). As per Fig. 5.1, in the following sections we consider how prior studies offer insight into the various factors that might constrain and enable such increases which may then result in serendipity. Our chapter thus attempts to move the conversation beyond a consideration of the related tensions such as those having to do with strategy versus luck, top-down causal planning versus bottom-up emergent or effectual action (e.g., Mintzberg et al. 1996; Sarasvathy 2008), and goal-directed activity versus deterministic structure (Engel et al. 2017; Porter and Woo 2015). In synthesising and building on this work, we illustrate how serendipity within the context of entrepreneurship and innovation can be best understood as a process (and related outcome) rather than an event. Our article contributes to the literature a review and conceptualisation of serendipity that questions the key assumptions of traditional “risk management” and “planning” approaches, showing how factors of unexpected innovation previously perceived as exogenous might instead be (partly) endogenous. This reframing allows us to then elaborate on the conditions required for the emergence of serendipity (Busch 2022; Gyori et al. 2019), thus opening up a number of fruitful avenues for further research.

5 Serendipity in Entrepreneurship, Strategy, and Innovation—A Review …

73

How Biases of Conventional Decision Making Constrain Serendipity Prior studies have shown how conventional decision-making approaches are often shaped by cognitive and behavioural biases, which can obscure the importance of weak cues and thus constrain serendipity (Denrell et al. 2003; Liu and de Rond 2016). Such biases include: underestimating the unexpected, self-censoring, illusion of control, and functional fixedness. First, social actors tend to have a particular—and potentially biased—view of the surrounding environment that shapes both expectations and attention. Events and information that are unexpected often go unnoticed or are discarded (Cunha et al. 2010). People also tend to overlook associated weak cues as they tend to focus on prominent features of the environment (Cunha et al. 2010, see also Ross, Chap. 9 in this volume). Second, as social actors engage in conventional thinking and decision-making, they are prone to self-censoring based upon normative pressure. The pressure to conform with such conventional wisdom is not only due to social desirability, but also with the desire to appear rational (Denrell et al. 2003). However, this pressure to conform with taken-for-granted insight or normative decisions can lead to the discarding or self-censoring of new ideas (Grimes and Vogus 2021, see also Arfini, Chap. 7 and Soto, Chap. 11 in this volume). This has shown to be particularly true in cases in which those ideas emerge unexpectedly, such as in the case of serendipity, due to those ideas’ lack of perceived legitimacy (Busch 2020a). Third, research shows that social actors tend to presume high degrees of control over the decision-making and innovation processes (e.g., Grimes et al. 2020; Sand and Jongsma 2020), such that when serendipity occurs, it is frequently airbrushed out of the ensuing narrative of change.2 Unfortunately, such perceived control is often an illusion. Research has shown that much of success is “unexplained variance”— it cannot be explained by traditional factors on which management tends to focus (Liu and de Rond 2016; McGahan and Porter 2002; also see Denrell et al. 2015; Henderson et al. 2012). For example, forecasts for fast-moving consumer goods (e.g., film box office revenues, company growth, toiletries) tend to have error rates of up to 70% (Coad 2009; Fildes et al. 2009; Geroski 2005). This is due to the fact that most situations and systems are too complex to be captured by models in every detail (Bansal et al. 2018). In addition, honest mistakes, unexpected events, and social dynamics tend to lead to outcomes that are different from those that were planned (Cohen et al. 1972; Hannan et al. 2003; Herndon et al. 2014). Given such illusions 2

Related research has shown that we tend to look for patterns where there are none: pareidolia (Sagan 1995; Voss et al. 2012). For example, in an experiment by behavioural psychologist Skinner (1948), a pigeon was placed inside a box, and food pellets were released at random intervals. While the pigeon had no way of predicting when pellets would drop (and indeed, no way of causing it), it began to behave as if it could. For example, if it received a pellet when walking in a circle, it started repeating that action, until the next one appeared. It began acting as if it could exercise control over it—even though it was an unpredictable event (also see Conrad 1958; Mishara 2010).

74

C. Busch and M. Grimes

of control, social actors are thus likely to not only minimize the role of serendipity in the past, despite evidence, but they are also likely to discount the possibility of similar serendipitous occurrences in the future. Finally, the process of professionalisation has led to increased degrees of specialised expertise but also functional fixedness—the quality of being mentally blocked from using tools and approaches in novel ways (Adamson and Taylor 1954; Duncker 1945). People who are deeply familiar with and skilled at particular methods tend to overlook opportunities for innovation (Allen and Marquis 1964; Arnon and Kreitler 1984). Conversely, by deviating from these “tried and true” methods and engaging in non-routine action, this encourages greater displays of creativity (Dane 2011; also see Arnon and Kreitler 1984). New experiences and unusual situations, in other words, tend to enhance cognitive flexibility, helping social actors to overcome functional fixedness. Intriguingly, this also introduces an argument for why serendipity might be less constrained in contexts where resources may be lacking, and where there are no particular methods, system, or tools to “unlearn” (Busch 2022; also see German and Barrett 2005; German and Defeyter 2000). For example, the MPesa money transfer system in Kenya emerged in a context lacking a reliable ATM network.

Step 1: Serendipity Triggers Although the biases that characterise conventional decision-making can often constrain serendipity by limiting attention to weak cues, other research suggests there are various individual- and organisational-level practices that have the capacity to both increase trigger events as well as overcome the aforementioned cognitive biases. In this chapter, we focused on those that have managerial relevance (for others, see for example, Wiseman 2003). Extroverted/introverted behaviours. Previous research has shown that displays associated with extroversion (the state of enjoying being with other people) can increase “fortunate” encounters by increasing the number and diversity of individual interactions, as well as by encouraging sustained engagement with those individuals (McCay-Peet et al. 2015; Wiseman 2003). Such extroverted displays often involve increases in culturally-inviting gestures, which have been shown to enhance the degree to which others feel more “attracted” to them (Wiseman 2003). Such attraction can thereby give rise to increased sharing of novel information, thus potentially surfacing unexpected solutions. However, given that the noticing and bracketing of peripheral information or weak cues may also require self-awareness, time, and inward-focus, serendipity may also arise from more introverted displays and practices, such as meditation or engagement with non-human sources of cultural engagement including the consumption of books, movies, or the internet (Beale 2007; Liang 2012). Curiosity and alertness. Being alert to a potentially meaningful trigger—and making sense out of it—is at the core of experiencing serendipity (Busch and Barkema

5 Serendipity in Entrepreneurship, Strategy, and Innovation—A Review …

75

2020; Cunha et al. 2010; Erdelez 1999; Kirzner 1979; Merton and Barber 2004). Research in psychology and management has shown that alertness and the desire to know or learn (“curiosity”) are paramount to noticing unexpected moments and events (Diaz de Chumaceiro 2004; Napier and Vuong 2013). Specifically, noticing and bracketing peripheral information without being cognitively constrained by goal-directed search processes helps social actors identify possibilities that might have previously been overlooked (Merton and Barber 2004; Cunha et al. 2010). Serendipity thus plays a major role in opportunity discovery (Corner and Ho 2010; Dew 2009), especially in the early stages of firm formation (Mirvahedi and Morrish 2017). Prior research has noted how such curiosity and alertness tends to vary based on domain experience and specialisation. For example, inexperienced founders tend to be more open to new information and demonstrate a high level of alertness, whereas more experienced ones tend to develop a high degree of focus which limits “distractions” (Busenitz 1996). Interestingly, this suggests possible unexpected benefits to inexperience in the early stages of a project, wherein the time-contingent importance of serendipity is presumed to be amplified (Cunha et al. 2010; also see Kornberger et al. 2005). Positive emotions. Positive emotions can be conducive to serendipity, as they increase alertness to outside stimuli, as well as responsiveness to external events, by broadening individuals’ action repertoire and scope of attention (Baron 2008; Cunha et al. 2010). This is particularly true of other-oriented, positive emotions such as compassion, wherein concern is expanded out from individual experience to account for others and their suffering. In turn, such positive emotions also increase a person’s capacity to make bisociations, because they can boost fluid and integrative thinking across topics (Isen et al. 1987). Conversely, negative emotions can diminish receptiveness to (potential) serendipity triggers, as they decrease receptivity to novel or unconventional information (Busch 2020a; also see Kahneman 2011). Self-awareness. Researchers have linked self-awareness to positive well-being and mental health (Fenigstein et al. 1975; Sutton 2016). The importance of selfawareness goes beyond psychological strength and affects performance, rumination, and interpersonal stress (Brinker et al. 2014; Feldman et al. 2014). A way to conceptualise self-awareness is through the practice of mindfulness (Brown and Ryan 2007). This practice is known to directly improve social interaction (Brown and Ryan 2003), which can contribute to our ability to notice serendipity triggers (Danneels 2011; Gyori et al. 2019). In addition, self-awareness plays an important role in the way social actors interact with their environment by reducing negative emotions and increasing the perception of one’s own potential (Kamenov 2013)— behavioural outcomes which may help increase receptivity to serendipitous trigger events. Humility. Psychologists have correlated humility with openness to alternative ideas and lack of dogmatism (Leary et al. 2017; McCray and Sutin 2009; Petrocelli et al. 2007; Seckler et al. 2021). This can be conducive to recognising serendipitous triggers because serendipity requires alertness to new connections (Krumrei Mancuso and Rouse 2017; McElroy et al. 2014). Humility is also associated with awareness

76

C. Busch and M. Grimes

of one’s intellectual “blind spots” and thus ensures increased openness to different persons, arguments, or ideas (Driver 1989; Spiegel 2012). In this way, social actors’ efforts to question their own assumptions can increase serendipity (Gyori et al. 2019; also see Cunha and Berti, Chap. 4 in this volume, for a great discussion of the role of “generative doubt”). Improvisation. Improvisation is about intentionally, quickly, and creatively reacting to a situation (Hmlieski and Corbett 2006; Magni et al. 2010; Weick 1998; Baker et al. 2003). The precursors that affect each individual’s potential to improvise are their skills, confidence, and self-efficacy, each contributing to the propensity of acknowledging serendipitous clues (Fisher and Amabile 2009; Fultz and Hmieleski 2021; Magni et al. 2010). (More on this can be found in Cunha and Berti, Chap. 4, this volume). In addition to these individual enablers, there are systemic enablers that support the emergence of serendipity triggers. Expanded search. Recent research in management similarly contended that narrowly defined problems can constrain serendipity triggers, as they limit the space for potential (unexpected) need-solution pairs to emerge (Stock-Homburg et al. 2021; Stock-Homburg et al. 2021; Von Hippel and von Krogh 2016). This research contends that adding more information to the respective problem allows for generating a broader range of solutions. For example, an appeal to “reduce costs” might result in people coming up with solutions such as buying less expensive raw materials or reducing headcount. If instead the problem was defined as “increase profit margins”, people might come up with additional suggestions such as raising the selling price, substituting the product with a more efficient option, among others (von Hippel and von Krogh 2016; also see Busch 2022). However, an organisation or individual is usually not able to provide all the potentially relevant information about the underlying need—and new information tends to emerge along the way as the problem-solving process unfolds (Tyre and von Hippel 1997; von Hippel and von Krogh 2016; von Hippel and Tyre 1996). Thus, “search strategies” that cast a wider net of possible problems and solutions potentially lead to a higher likelihood of serendipitous outcomes to occur (von Hippel and von Krogh 2016; McGahan et al. 2021). Expanded networks. Although social actors may vary in their openness to serendipity, much of the process of serendipity and even these aforementioned individual differences can be shaped by contextual factors, which enable or otherwise constrain serendipitous triggers. Social embeddedness, “the nature, depth and extent of an individual’s ties into an environment, community or society” (McKeever et al. 2014: 222; also see Portes and Sensenbrenner 1993), can facilitate or constrain action. It potentially gives access to resources (e.g., financial resources; Batjargal et al. 2013), status and legitimacy (Burt 1997), emotional support (Ozcan and Eisenhardt 2009; Schutjens and Stam 2003; Shane 2003), and learning benefits (Busch 2014)—all of which can be conducive to the surfacing of serendipity. Although social actors can also inform and shape their own networks (Busch 2014; Fligstein 2001) to coincide with their specific goals (e.g., Dhanaraj and Parkhe 2006; Hallen and Eisenhardt 2012; Provan and Kenis 2008), the uncertainty surrounding

5 Serendipity in Entrepreneurship, Strategy, and Innovation—A Review …

77

those goals is likely to constrain the actor’s capacity to properly evaluate existing and possible networks and related interactions (Alvarez and Barney 2007; Alvarez et al. 2013; Busch and Barkema 2020). Instead, the value of networks is often surfaced by way of serendipitous encounters and only recognised post-hoc. However, social actors may deliberately form innovation communities (Fleming and Waguespack 2007; Furnari 2014; Garud and Karnoe 2003), communities of practice (Wenger 1998), or social innovation communities (Toivonen 2016). While rituals and joint experiences may facilitate a feeling of belonging, which can foster serendipity (Merrigan 2019; also see Toivonen 2016), strong social networks and communities can also constrain individuals, as (over-) embeddedness can lead to the sedimentation of homogenous networks and lack of access to novel or diverse information (Di Falco and Bulte 2011; Khavul et al. 2009; Khayesi and George 2011; Kiggundu 2002; Maurer and Ebers 2006). As such, recent research (e.g., Busch 2022; Busch and Barkema 2020; Engel et al. 2017; Obstfeld et al. 2020) has highlighted the ways in which third-party organisations might act as boundary organisations, fostering serendipity by facilitating networks and resources for social actors such as entrepreneurs in contexts of high uncertainty. Specifically, given the importance of social networks for organisational survival and growth more generally, a number of organisations such as accelerators and incubators have been increasingly mobilised around the globe (Amezcua et al. 2013; Cohen et al. 2019). While some of these organisational sponsors provide highly structured support programs that specify which networks or resources are being offered (e.g., Rothaermel and Thursby 2005) and thus potentially “lock in” social actors (and their organisations), others have rejected this model in lieu of more open designs that actively encourage serendipity via mechanisms such as elevating commitment (e.g., appealing to an enlightened self-interest); agile platform design (e.g., supporting flexible space design); cultivating open-mindedness (e.g., fostering an openness to the unexpected); and highlighting emerging opportunities (e.g., developing adaptive support programs) (Busch and Barkema 2020; Giudici et al. 2018). Physical and virtual task environment. To increase the rate of serendipity triggers, prior research suggests that physical proximity matters. To the extent that entrepreneurs, innovators, and other stakeholders are co-located for an extended period of time, this will increase the likelihood of serendipitous trigger events and interactions. Supporting such assumptions, prior scholarship has highlighted how the physical task environment (as well as the type of work itself) has a major impact on the likelihood of serendipity occurring (McCay-Peet and Toms 2010; Reinecke and Ansari 2015). In companies, for example, it has been shown that small design changes such as placing couches next to doorways can increase the likelihood of serendipity, as they allow people to bump into each other (Lindsay 2013). Companies such as Pixar and Google have organised their headquarters to maximise “cross-pollinations” of data and people, across different areas. For example, the main buildings of Pixar— one of the highest grossing film studios of all time—were designed to maximise inadvertent encounters (Catmull 2008; Lehrer 2011). Instead of designing separate buildings for computer scientists, executives, and animators, the company developed

78

C. Busch and M. Grimes

a single big space with a big atrium as well as mail boxes, meetings rooms, and a coffeeshop at the center. This led to people “bumping into each other” in the atrium (Catmull 2008; Lehrer 2011). Research in the information sciences has shown that factors such as proximity play a major role for serendipity in virtual spaces, too. For example, it has been contended that smoother informal virtual communication between colleagues can increase serendipitous encounters (Guy et al. 2015; McKay-Peet and Toms 2018). Organisations have used approaches such as “randomised coffee trials”, in which people are randomly paired with strangers across the organisation to facilitate unplanned conversations (Busch 2020a; Soto, Chap. 11 in this volume). This is based on the idea that serendipity is governed by probability (Pirnot et al. 2013). The world’s biggest technologist gathering, Web Summit, provides a case study of how data scientists “engineer serendipity”, on-line and off-line (Cosgrave 2012). The conference hosts 50,000 participants, and uses complex systems and networks approaches such as eigenvector centrality (measuring the influence of a person in a particular network). For example, graph theory helps to “recommend” people on (potential) visitors’ Facebook feeds, and groups for pub-crawls are put together based on propensity to encounter commonalities (Cosgrave 2012).

Step 2: Bisociation Research has shown that bisociation—the connection of previously unrelated matrices of events, skills, or information (for example, linking a serendipity trigger to something relevant)—tends to be at the core of serendipity (Busch and Barkema 2020; von Hippel and von Krogh 2016). Often, these bisociations occur between problems and unexpected solutions to these problems, as in the Archimedes example discussed above. However, while problems may at times be formulated a priori, social actors might also “see” the problem and the solution at the same time (Busch 2020a; StockHomburg et al. 2021; von Hippel and von Krogh 2016), like in the rolling suitcase example mentioned previously. Here, the problem and solution “arrived” at the same time, via a sudden bisociation that lead to a serendipitous outcome. (Importantly, what is new to one observer might not be new to others; Felin and Zenger 2015). Consequently, innovation researchers von Hippel and von Krogh (2016) suggest to model problems/needs (e.g., a patient’s ailments and symptoms) on one landscape, and possible solutions to each problem/need on another (e.g., a doctor’s experiences, information, etc.). Problem-solving, then, is about linking a specific point on the problem landscape with a point on the solution landscape. This is where often creativity—the process of surfacing something new and valuable information—comes into play (see Ross, Chap. 9, this volume). Although creativity can be broadly useful to entrepreneurs and managers as they engage in goaldirected search for solutions to known problems, it is also essential to the process of serendipity wherein unexpected solutions must be creatively derived from bisociations – the perceived intersection of different and sometimes divergent perspectives,

5 Serendipity in Entrepreneurship, Strategy, and Innovation—A Review …

79

observations, and areas of application. Research in neuropsychology exposes the cognitive foundations of such creativity, illustrating how the “aha” effects tend to arise from the feeling of something (unexpectedly) making sense (Stock-Homburg et al. 2021; also see: Cosmelli and Preiss 2014; Schooler and Melcher 1995). These moments happen through a sudden gain in “processing fluency”—people fill in gaps in their own thinking that they did not even know existed (e.g., Cosmelli and Preiss 2014; Pelaprat and Cole 2011; Topolinski and Reber 2010). Specifically, the brain’s neural network tends to unconsciously integrate varieties of pieces of information over time (Ritter and Dijksterhuis 2015; Van Gaal et al. 2012)—and then, suddenly, a “eureka moment” occurs—a process which interestingly can be measured by analysing the brain’s electrical activity (Christoff et al. 2009; Mason et al. 2007; Stock-Homburg et al. 2021). Thus, what appears to be a spontaneous creative idea often is the result of previously forgotten insights and ideas that emerge to help social actors “connect the dots” in a particular moment. Research has shown that often the process of making this bisociation can take a long time, as one might not have initially bracketed an observation as important, or one might have missed a meaningful link. This lag between the triggering event and the bisociation is what some have referenced as the “incubation period,” requiring persistence and sagacity in order to eventually form the mental linkages (McCay-Peet and Toms 2010). Trivial activities such as browsing a book store can alert a person to something they might not have previously been aware of, and suddenly, a rapid, complete understanding of a solution—the eureka moment—emerges (Gilhooly and Murphy 2005). Incubation periods tend to take between five minutes and eight hours (Sio and Ormerod 2009), but can be much longer. Indeed, this delay between the triggering event and the related bisociation can make it difficult for an actor to recall and properly attribute the original source of the creative observation (Stock-Homburg et al. 2021). Previous research has discussed a number of approaches that help facilitate making these bisociations. Reframing situations. Research in psychology and management shows that how we perceive and categorise (“frame”) the world—and how we look at a particular situation from a different perspective (“reframe”)—plays an important role with regard to “seeing” opportunity in unexpected situations (Busch 2021; Busch and Barkema 2021; Reinecke and Ansari 2015). For example, “making the best out of what is at hand” (bricolage) can lead to creative solutions, as people look at a given object (or subject) afresh, recombine it with other ideas or objects, and identify opportunities that were not previously conceived as such (Busch and Barkema 2021; Baker and Nelson 2005). In a similar vein, research on frugal innovation highlights how, when social actors operate with limited resources yet reframe such situations as more abundant, this encourages those actors to identify unconventional solutions to perceived or unperceived problems (Prabhu 2017). Narrative theories of entrepreneurship also provide a basis from which to understand the importance of serendipity within the context of the entrepreneurial process. This builds on the notion that an entrepreneur’s role is to create new ideas and opportunities by way of frames or narratives that (re-) construct reality, reframing what

80

C. Busch and M. Grimes

was once deemed improbable as now possible (Grimes and Vogus 2021). In this way, new entrepreneurial narratives expose bisociations that were previously obscured. Analogous thinking. Approaches such as lateral thinking (focusing on nonobvious and unconventional cognitive links (de Bono 1985, 1992; also see Birdi 2005), disjunctive strategies (Gyori 2018), and analogous thinking (Gentner and Markman 1997; Gick and Holyoak 1980) can facilitate serendipity (Busch 2020a, b). The one most clearly linked to serendipity is analogous thinking, whereby information describing relationships from one domain of knowledge can be used to surface problem-solution dyads in another, perhaps unrelated domain (Cornelissen and Clarke 2010; Cornelissen et al. 2011; Gentner and Markman 1997; Gick and Holyoak 1980; Stock-Homburg et al. 2021). On the one hand, analogous thinking has been shown to require deep expertise (Ericsson and Staszewski 1989). This is most clearly evident when social actors attempt to draw temporal analogies, wherein the objective is to identify connections between current seemingly anomalous observations and future (or previous) experiences (Stock-Homburg et al. 2021). Yet conversely in the context of such deep expertise there is also the risk of “functional fixedness”, which can undermine much analogous thinking, which often requires general rather than specialised forms of expertise (Busch 2020a). This also raises the importance of intuition as a potential filter that helps form bisociations (Cunha et al. 2010). Intuition is a way of processing information that is fast, unconscious, and driven by our surroundings (Baldacchino et al. 2015). Besides simply being our “gut feeling” about a situation or person, it is the unintentional ability to create links between information (Cunha et al. 2010; Ezkinali and Giannopulu 2021). The ability to mindfully acknowledge and act on our intuition can thus be important for forming bisociations which can support more original and superior solutions to problems (Eubanks et al. 2010). However, while serendipity is often thought of as an individual-level phenomenon, it often emerges via collaboration, i.e., the interaction, resources, and skills of several people and teams (Meyers 2007; Cunha et al. 2010). For example, the team that discovered penicillin consisted not only of much-lauded “hero” Alexander Fleming. Ernst Chain, Howard Florey and others continued driving the train that Fleming set in motion, and received the Nobel Prize together (Copeland 2018; Cunha et al., 2010; Meyers 2007). Acknowledging that the understanding and bridging of observations tends to require the skills and resources of several people, the father of the scientific method, Francis Bacon, considered the ideal research organisation to include merchants of light (keeping up with the work of other organisations); pioneers (trying new experiments); inoculators (executing experiments with highest proficiency); interpreters (raising former discoveries into axioms); and mystery men (collecting earlier experiments into the state of the art) (Yaqub 2018). And in fact, research has shown that diversity often breeds serendipity, as the ability to form and then act upon bisociations depends on combining previously unrelated ideas or information (Hargadon and Bechky 2006; Napier and Vuong 2013). Often, the significance of events is only understood when people from other areas help explore the broader relevance of an unexpected moment. Then, “metaphorical leaps”—such as realising that the apple falling from the tree is not only about the apple falling down

5 Serendipity in Entrepreneurship, Strategy, and Innovation—A Review …

81

but that it might represent gravity’s pull on any object—become possible (Busch 2020a). Recent research has looked at companies and their practices to incentivise people to interact and create serendipity across the organisation, for example via means such as random coffee trials (e.g., NESTA) or learning lunches (e.g., HubSpot), which randomly pair people up to create “watercooler moments” (Busch 2020a). Other researchers have focused on the question of how to integrate people into teams from outside the organisation in an attempt to further broaden the potential opportunity space for need/solution pairs (von Hippel and von Krogh 2016). Such benefits of increased diversity and interactions, however, can only be realised when organisations ensure a strong collective culture (see below), which emphasises mutual interests, shared causes, experiences, or enemies. In this way, collective identities facilitate a general willingness to connect within teams despite strong differences in perspective (Foster and Ford 2003). Potential barriers to serendipity can thus be overcome by building diverse teams and including people early on in the process (Busch 2020a; Cunha et al. 2010). Importantly, while traditional innovation (and innovation research) for long has focused on intra-organisational processes, in a fast-changing world, varied customer demands increasingly require collaboration and co-creation across organisational boundaries. Thus, research has increasingly focused on the question of how effective networks of organisational actors—ecosystems—emerge, and how innovation is orchestrated within those ecosystems (Adner and Kapoor 2010; DeJordy et al. 2020; Kapoor and Agarwal 2017; Logue and Grimes 2019; Nambisan and Baron 2013; Thompson et al. 2018). Supportive organisational structures and ecosystems can also facilitate adaptation (i.e., change based on the initial function), as well as exaptation (i.e., use characteristics that evolved for other or no use, and co-opting them for their current role; Andriani et al. 2017; Gould and Vrba 1982). Designing organisational and ecosystem structures in ways that allow for exaptations to happen tends to drive serendipity (Andriani and Cattani 2016; Austin et al. 2012; Garud et al. 2018; Gould and Vrba 1982). For example, companies such as IDEO often accumulate interesting ideas without having a clear sense of how these ideas could be used later—the ideas are being loosely organised, and “search” can be activated whenever something that might be relevant comes along (Busch 2020a; Gould and Vrba 1982; also see Andriani and Cattani 2016; Austin et al. 2012). Problem-formulation-formulation. Especially in fast-changing environments such as startup companies, ill-structured problems tend to dominate, as situations can change rapidly, and there is often a lack of full information (Busch and Barkema 2020; Engel et al. 2017). Thus, researchers have explored alternative approaches by which entrepreneurs and innovators can facilitate the enactment of serendipity such as iterative problem formulation, whereby a problem is being approached repeatedly in a number of ways, in rapid succession, and quickly assessed for efficacy while lowering initial investment into any one specific solution. Companies such as the design group IDEO have developed related approaches such as rapid prototyping, where the problem-solver responds to initial challenges by immediately developing

82

C. Busch and M. Grimes

an easy-to-adapt working model. Users can then work with the prototype, and experiment and modify, before it goes back to the problem solver/designer, and a more refined prototype is being produced—then the cycle begins again (Thomke and Fujimoto 2000; von Hippel and von Krogh 2016; also see: Kurup et al. 2011; Nelson 2008; Shepherd, Seyb, and George 2021). This rapid prototyping approach tends to interpret each iteration of the prototype not as a “failure” but as a crucial and necessary step in the overall process (von Hippel and von Krogh 2016; also see Conboy 2009), allowing for serendipitous solutions to emerge. Theories of value. Companies develop firm-specific theories of value creation (bundles of market problems and architecture that guide the strategic direction of a company and help discover and filter opportunities; Felin and Zenger 2015; Zenger 2013) that do not limit but rather foster bisociation. Such theories of value creation can be used to formulate problems and select possible problem-solution pairs, making it not only unique to the respective company but also contrarian with respect to the broader field. Such unique and contrarian theories of value can potentially lead to new unexpected value creation possibilities that might be unforeseen by other companies (Felin et al. 2020; Grimes and Vogus 2021). Apple, for example, in contrast to companies such as Xerox realised the contrarian possibility of the graphical user interface, the “mouse”, and bit-mapping technology, as its theory allowed the company to recognise and pursue the potential value (Isaacson 2011). In a similar vein, Starbucks—as we know it today—emerged from Howard Schultz’s unexpected observation that replicating Italian coffee bar culture around the world could be a valuable business. His theory focused on the interplay between product sourcing, customer education, and store format, and this guiding frame propelled iterations and experimentation (Felin and Zenger 2015; Schultz 1998). However, as numerous other examples can attest, to the extent that such theories of value become narrow, over-specified, and conventional this can limit serendipity (Busch 2022). Research in neuroscience, psychology, and library science has shown that overly structured goals or narrowly defined problems can constrain serendipity, while less narrowly defined goals or aspirational objectives make serendipity more likely (Toms 2000; von Hippel and von Krogh 2016; also see McCay Peet and Toms 2010; Stock-Homburg et al. 2021; also see “expanded search” and “iteration”). In one experiment, participants were asked to interact with a reading device. Some participants were instructed to find some particular information, others were given no task at all. The first group often found the particular information they were instructed to seek out; the second group were more exploratory, and came away with interesting information that was not sought (Toms 2000; McCay Peet and Toms 2010). Related experiments have shown that individuals that face narrowly specified problems tend to be more closed to unexpected moments (and making related bisociations) than those that faced broader ones (Stock-Homburg et al. 2021; also see: Cosmelli and Preiss 2014; Schooler and Melcher 1995; Wiseman 2003).3 3

In one experiment (Wiseman 2003), for example, researchers gave participants a newspaper to read, and asked them how many photos were in it. Most of the participants needed around two minutes to flip through the newspaper, and some of the participants double checked—but given

5 Serendipity in Entrepreneurship, Strategy, and Innovation—A Review …

83

One response to this challenge is to ensure theories of value which are more highly abstract and connected to broader societal purposes, thus allowing for the possibility of positive forms of mission drift (Grimes et al. 2019). For instance, it has been shown that when firms maintain a broader “north star” (a broader purpose or ambition) while embracing emergent strategy, this can lead to an openness to the unexpected, allowing for serendipity to emerge (Gyori et al. 2019). The potential trade-offs between a prosocial purpose (e.g., eradicating malnutrition) and profitability have been used by some companies to generate a “creative tension” that can lead to (serendipitous) innovation (Busch 2020a, b). Related research has shown that often those at the frontlines (e.g., frontline workers) might generate new ideas through trial and error, heuristics, and informal contacts, while those at the center of an organisation often rely on deductive approaches, intelligence documents, and formal reports—which potentially discourages serendipity (Regner 2003). The integration of cross-organisational responsibilities is particularly important after an acquisition or a merger, given that over 50% of the value in acquisitions can emerge serendipitously—for example, the acquiring company might unexpectedly come across a beneficial technology that the acquired company used that it was previously unaware of (Graebner 2004). Research has shown that when people fill a role in both the subsidiary and the parent company, it helps them link different parts, and gives them the required standing in the parent organisation to “lobby” for ideas that emerge from the “periphery” (Graebner 2004). However, social actors may recognise new serendipitous opportunities for innovation, and yet still fail to enact that opportunity (Busch 2022; Ross 2022). Such failure can often be attributed to the fact that the process of innovation is fundamentally a social and organisational one, requiring “buy-in” and thus constrained by existing patterns of decision making and resource allocation.

Step 3: Enactment What are individual enablers that facilitate the enactment of serendipity? Social skill. Social actors tend to outweigh the costs of trying over the reward of potential positive outcomes and thus often focus on the potential risk of “unproductive accidents”, thereby preventing action even in cases of otherwise substantial benefit (Austin et al. 2012). Prior studies illustrate how innovation and change processes can be interrupted by fear of change, power dynamics, vested interests, or systemic biases within groups (Austin et al. 2012; Sting et al. 2019). This necessitates social skill—the ability of social actors to induce cooperation in others—which helps them frame and navigate their social context (Busch 2020a; also see Fligstein 2001). their focus on the photographs, none of them noticed the headline on page two that read “There are 42 pictures in this newspaper”, in large, bold letters. The participants also missed out on the chance to win £100—another large headline read “stop counting and tell the experimenter you see this and win £100”. When the researcher asked the participants to not focus on the photographs, they saw the messages immediately (Wiseman 2003).

84

C. Busch and M. Grimes

Perseverance. Furthermore, the emergence of unexpected solutions to perceived or unperceived problems often requires continuous experimentation and perseverance in the face of ‘negative’ outcomes (Austin 1978; Burgelman 2003). Perseverance and tenacity helps enable serendipity by allowing for increased incubation time of new ideas, which is often required for new and peripheral information to be properly noticed, bracketed, and implemented in such a way that it might be connected with perceived or unperceived problems (Busch 2020a; Napier and Vuong 2013). Factors that influence serendipity on the systemic level—and thus help materialize serendipity (Busch 2022)—are systematic evaluation and corporate culture. Corporate culture. Culture—the collective beliefs, principles, and values that guide our interactions—plays an important role in whether or not serendipity might be enacted (de Rond 2014; Cunha et al. 2010). In environments in which people feel “safe”, they tend to be less likely to self-censor ideas and are more vigilant to unexpected encounters and ideas (Cunha et al. 2010). Research has shown that serendipity increases in settings in which blame is being withheld, and where people are open to a diversity of ideas, as people feel “safe” to discuss unexpected findings or ideas that are not yet fully developed (Cunha et al. 2010; de Rond 2014; Napier and Vuong 2013). Studies on psychological safety have focused on how people can present themselves without fear of negative consequences related to career, status, and self-image (Edmondson 1999). Better-performing teams tend to talk more about emerging and failed ideas, while lower-performing teams tend to swipe them under the carpet, thus constraining knowledge sharing, learning, and trust (Edmondson 1999). Edmondson (1999) found that psychological safety can be increased by formulating shared meaning and expectations, giving people the feeling that their input is welcome, and expressing appreciation and sanctioning clear violations. High-creativity companies such as Pixar have used approaches such as opening meetings with sentences such as “Early on, all of our movies are bad!”, thus giving people the permission to ask critical questions in a “safe” environment (Catmull 2008). Systematic Evaluations. To the extent that organisations become more welcoming of peripheral and emergent insights, this can also increase the risk of potential information overload (McKay-Peet and Toms 2018). In such cases, the challenge for enacting serendipity becomes filtering in such a way that those firms can balance the need for clarity with the need for surfacing unexpected value (Busch 2020a). Several recent studies provide insight into how this balance within the filtering process might be struck. Napier and Vuong (2013) contrast flash evaluations of serendipity with systematic evaluations. Whereas a flash evaluation is a quick assessment that is based on a gut feeling about the new, unexpected information, a systematic evaluation is a more comprehensive analytical assessment that includes criteria such as risk tolerance, timing, and additional information that helps invalidate or substantiate the unexpected information (Napier and Vuong 2013). For example, companies such as white goods company Haier “place bets” and develop (decentralised) structures that allow for investment into unexpectedly emerging ideas (Gyori et al, 2019). Haier’s “micro-enterprise” model encourages employees to leverage company resources to spot and develop new ideas. Investment committees then bet on the best

5 Serendipity in Entrepreneurship, Strategy, and Innovation—A Review …

85

ideas. For example, employees within the organisation spotted that farmers unexpectedly used their washing machine to wash potatoes—which resulted in a potato washing machine. Other studies highlight the use of technology that screens for relevance (e.g., items that might be meaningfully related in unexpected ways) instead of similarity (Guy et al. 2015; McKay-Peet and Toms 2010, 2018). Some virtual platforms also allow users to defer serendipitous ideas and to bookmark items for later (McCay-Peet and Toms 2010).

Discussion Based on our review of the literature and our related work (e.g., Busch 2022; Busch 2020a; Busch and Grimes 2023), we developed a model of the process of cultivating serendipity that highlights the role of different individual and organisational practices in both enabling and constraining the various steps involved in that process. Our review thus established that serendipity is not a singular event, but a process (and related outcome) that requires sagacity. It can be influenced by noticing unexpected moments, and turning them into positive outcomes via proactive decisions (Busch 2022; Busch and Barkema 2020; Denrell et al. 2003). The process of serendipity includes a trigger (for example, a person making an unexpected observation), a bisociation (linking the trigger to something relevant), and the cultural and structural features that help to enact that bisociation into an unanticipated outcome (Busch 2020a, 2021; Copeland 2018; McCay-Peet and Toms 2018; Napier and Vuong 2013; also see Merton, 1948). While a specific random chance encounter is an event, serendipity is a process and related outcome (Busch 2022; de Rond 2014; Fine and Deegan 1996; McCay-Peet and Toms 2018; Merton and Barber 2004).4 The process—of trigger, bisociation, and enactment—unfolds at multiple levels of analysis (Busch 2021). Given that serendipitous bisociations often emerge from the interplay between agents and their environment, system-level conditions for serendipity are paramount. For example, these contextual factors can encourage people to question ideas and insights (Busch and Barkema 2020), foster people’s motivation to cooperate (Rauch and Ansari 2021), provide interactive physical and digital spaces that allow people to accidentally bump into each other (Amezcua, et al. 2013), legitimise serendipitous insights (Busch and Barkema 2020), and provide funding opportunities for new ideas with unknowable risks (Huang and Pearce 2015). For companies, we suggest that the ability to integrate, build and reconfigure internal and external competencies to facilitate serendipitous triggers, bisociations, and the enactment of serendipity can become a “dynamic capability” (Busch 2020a, b; de Rond et al. 2011). We suggest that it does so by enhancing the organisation’s 4

Trigger and bisociation may happen at the same time, and there can be feedback effects (Busch 2020a; also see Brown 2005; Busch 2022; Cunha et al. 2010; Merton and Barber 2004).

86

C. Busch and M. Grimes

“absorptive capacity”—its ability to encounter new information and to integrate it into existing structures and processes—which can amplify innovation and learning (Zahra and George 2002). In this way, companies can turn the acceleration of serendipity into a strategic advantage, for example by focusing employees’ attention on the important role of the unexpected. Limitations and future research. The purpose of this article was to give an overview of interesting serendipity-related research in the entrepreneurship, strategy, and innovation context. Our review is by no means exhaustive, and much works remains to be done in terms of conceptualising serendipity (see e.g., Busch 2022; Fultz and Hmieleski 2021). Furthermore, while we mapped serendipity as a linear process, it is clear that there are many opportunities for feedback loops within the process as well as the potential that steps within the process might happen simultaneously or, alternatively, draw out over years. Future research might thus explore some of the temporal dimensions of serendipity and the conditions that give rise to different temporal patterns. Our review of the literature opens up a number of other valuable areas for further scholarly inquiry. First, although we suggested that organisations’ efforts to cultivate serendipity might act as a type of dynamic capability (de Rond et al. 2011), how and under what conditions is this likely to hold? Similarly, while our study denotes a variety of individual and organisational practices that can foster serendipitous triggers, bisociations, and enactments, it is also likely that such practices may be more or less effective in different contexts and at different stages of organisational development (Busch 2022). What are those contingencies that explain the efficacy of the various practices? How can individuals and organisations cultivate “skilled luck” or “smart luck”? Furthermore, the emerging literature on entrepreneurial ecosystems and organisational sponsorship (c.f., Amezcua et al. 2013; Cohen et al. 2019; DeJordy et al. 2020; Hallen et al. 2020; Spigel 2017; Thompson et al. 2017) offers a setting within which to explore important tensions within the process of “engineered” serendipity. Much of the associated literature is focused on how systems of support can be structured in such a way so as to increase the likelihood of productive entrepreneurial and innovative outcomes. In essence, there is an implicit assumption that systems which foster serendipitous innovation can be designed, replicating for instance, the Silicon Valley or Silicon Fen phenomenon globally. And yet it is equally clear that some of the most prolific historic sites of innovation have been those in which the systems emerged with little top-down design over decades and even centuries. Future research might, therefore, explore the conditions under which systems of serendipity might be designed in top-down fashion, and the balance that is needed between structure and chaos or coordination and freedom. Also, what are the implications for success measures of organisational sponsors of entrepreneurship (e.g., celebrating “effective pivots” rather than the number of companies “graduating”)? Further research could also explore how local community leaders can be legitimised and enabled by policymakers to facilitate local serendipity-enhancing networks (as opposed to overly structured, centralised support programs—see also Soto, Chap. 11 in this volume).

5 Serendipity in Entrepreneurship, Strategy, and Innovation—A Review …

87

Furthermore, how can schools and universities integrate serendipity into their curricula? What is the role of approaches such as the Socratic method that focus on asking questions rather than solutions? How can scholarships be designed in more inclusive ways (e.g., not only monetary support but also including considerations around creating opportunity spaces for students)? Moreover, contexts of high uncertainty (e.g., emerging markets) could provide a fertile ground for further research. Although few studies of entrepreneurship make explicit reference to the concept of serendipity, much of the literature is oriented around understanding the related problem of uncertainty and its effects on entrepreneurial ideation and action. For instance, it has become a well-entrenched assumption within the entrepreneurship literature that the survival and growth of enterprises depends on their ability to deal with uncertainty (Alvarez and Barney 2007; McMullen and Shepherd 2006; Ramus et al. 2017). Because early-stage enterprises and entrepreneurs often face exceptionally high levels of uncertainty as to which partners, resources, or co-founders they might need in order to ensure success, they are often forced to frequently and radically change their assumptions about the problem that is worth solving and the solutions that might effectively address those problems (Grimes 2018). Amid such uncertainty, the process of discovering, constructing, and reconstructing the opportunity and its respective components is often a matter of serendipity (Busch and Barkema 2021). In this way the entrepreneur’s search to more clearly define a particular problem–solution dyad is subject to ongoing contingencies, which then lead to an emergent strategy (Harmeling and Sarasvathy 2013; Mintzberg and Waters 1985; Sarasvathy 2008). In larger companies, paying attention to weak signals allows managers to more quickly respond to emerging opportunities (Denrell et al. 2003; Liu and de Rond 2016; Teece et al. 1997; Winter 2003), which can play an important part, for example with regard to internationalisation (Kiss et al. 2020). Further research could explore these different contexts of uncertainty and how they might (or might not) provide a fertile ground for serendipity to emerge. This might be of particular relevance with regard to new technologies such as artificial intelligence (Busch and Grimes 2023). Additional avenues of research could explore how to operationalise and measure serendipity in ways that make it more accessible to larger-scale quantitative studies. Much of the extant research tends to be qualitative or experimental in nature. First attempts to measure serendipity (e.g., Busch 2020a; Busch 2022; Erdelez 1999; Fultz and Hmieleski 2021; McCay-Peet and Toms 2012; Makri and Blandford 2012) have focused on particular aspects of the process. Interesting insights could borrow but also distinguish from related constructs and concepts such as originality (e.g., Grant 2017), novelty (e.g., Toms 2000), interestingness (e.g., Andre et al. 2009), absorptive capacity (e.g., Zahra and George 2002), or unexpectedness (e.g., Adamopoulos and Tuzhilin 2015). Given that serendipity is a process, exploring counterfactuals might also be a worthwhile avenue for further research. Moreover, what is the link between serendipity and tackling global societal and environmental challenges? Given the complexity of societal and environmental issues (Busch and Barkema 2019), many of the solutions might be unknown a priori, and serendipitously emerge via experimentation (Busch and Hehenberger 2022). How

88

C. Busch and M. Grimes

can companies “prepare” for this? Related questions could focus on the link between serendipity and inequality. Blind luck, social connections, inherited wealth masking as skill (Piketty 2014), or unintended consequences often play a major role in success, and the possibility to encounter serendipity is not equally distributed, as financial and other pressures can sap attention (Mandi et al. 2013)—see also Soto, Chap. 11 of this volume. Given that base levels of potential serendipity are very different depending on the respective context, how could they be improved for those that did not win the birth lottery? Research could also look into the role of “negative serendipity” (“zemblanity”; Boyd 1998; Giustiniano et al. 2016), the faculty of making unlucky discoveries by design. This might be a particularly fruitful line of inquiry, as some individuals and organizations might have (subconsciously) cultivated an environment that fosters zemblanity, thus potentially setting them up for failure. Another fruitful area of exploration could be the role of culture in (cultivating) serendipity. How does the process of (facilitating) serendipity unfold differently across different cultural contexts? Given that local cultures and belief systems shape behaviours, attitudes, and values (Hofstede 1984; House et al. 2004), they presumably play a major role in the serendipity process. For example, in settings characterised by higher power distance (in which lower-ranking individuals tend to accept that power is distributed unequally), it might be more difficult to trigger serendipity, as hierarchical divisions might hinder the free flow of information and ideas. However, even in very hierarchical settings, innovative solutions can emerge (Nonaka 1991). These contextual nuances extend to whole industries—while in nuclear reactors failure tolerance is low, in more entrepreneurial settings it tends to be higher, and thus serendipity might be more favorable in the latter (Busch 2020a). Future research could explore related contextual questions. Last but not least, how could serendipity be integrated into policymaking? First experiments have shown that initiatives such as cross-council cultural collaborations, the development of communities of interest linking local areas, and communities such as “friends of park” and police-resident liaison groups can help increase diversity and connect groups that would usually not connect (Rowson et al. 2010; also see Chanan and Miller 2010). How can policymaking empower local communities to create their own “smart luck” by connecting with the right people at the right time? How can cities and regions be designed as “ecosystems” that help produce “unexpected productive collisions”?

Conclusion In this chapter, we aimed to revisit the planning vs emergence (and luck vs skill) debates by suggesting that not only is there room for synthesis in entrepreneurship, strategy, and innovation, but that it is critical to do so. The role of serendipity has often been discounted in organisational and management theory, even though it is a major driver for innovation and societal impact, and plays a crucial role in much of business and life. Thus, we recommend an integrated approach to education, training,

5 Serendipity in Entrepreneurship, Strategy, and Innovation—A Review …

89

and skills programs that bridges the demarcations of polarising predecessors. In a fast-changing world, nurturing serendipity is a dynamic capability necessary for companies and individuals alike to not only survive, but thrive. Acknowledgements We are grateful to Miguel Pina e Cunha, Martin Sand, and Samantha Copeland for their excellent feedback on earlier versions of this article.

References Adamopoulos, P., and A. Tuzhilin. (2015). On unexpectedness in recommender systems: or how to better expect the unexpected. ACM Transactions on Intelligent Systems and Technology, 5: 1– 32. Adamson, R.E., and D.W. Taylor. 1954. Functional fixedness as related to elapsed time and to set. Journal of Experimental Psychology 47 (2): 122–126. Adner, R., and R. Kapoor. 2010. Value creation in innovation ecosystems: How the structure of technological interdependence affects firm performance in new technology generations. Strategic Management Journal 31 (3): 306–333. Allen, T.J., and D.G. Marquis. 1964. Positive and negative biasing sets: The effects of prior experience on research performance. Administrative Science Quarterly 35 (4): 604–633. Alvarez, S.A., and J.B. Barney. 2007. Discovery and creation: Alternative theories of entrepreneurial action. Strategic Entrepreneurship Journal 1 (1–2): 11–26. https://doi.org/10.1002/sej.4. Alvarez, S.A., J.B. Barney, and P. Anderson. 2013. Forming and exploiting opportunities: The implications of discovery and creation processes for entrepreneurial and organisational research. Organisation Science 24 (1): 301–317. https://doi.org/10.1287/orsc.1110.0727. Amezcua, A.S., M.G. Grimes, S.W. Bradley, and J. Wiklund. 2013. Organisational sponsorship and founding environments: A contingency view on the survival of business-incubated firms, 1994–2007. Academy of Management Journal 56 (6): 1628–1654. Andre, P., A. Cazavan-Jeny, W. Dick, C. Richard, and P. Walton. 2009. Fair value accounting and the banking crisis in 2008: Shooting the messenger. Accounting in Europe 6 (1): 3–24. Andriani, P., A. Ali, and M. Mastrogiorgio. 2017. Measuring exaptation and its impact on innovation, Search and problem solving. Organisation Science 28: 320–338. Andriani, P., and G. Cattani. 2016. Exaptation as source of creativity, innovation and diversity: Introduction to the special section. Industrial and Corporate Change 25 (1): 115–131. Arnon, R., and S. Kreitler. 1984. Effects of meaning training on overcoming functional fixedness. Current Psychological Research and Reviews 3 (4): 11–24. Austin, J.H. 1978. Chase, Chance, and Creativity: The Lucky Art of Novelty. New York, NY: Columbia University Press. Austin, R.D., L. Devin, and E.E. Sullivan. 2012. Accidental innovation: Supporting valuable unpredictability in the creative process. Organisation Science 23 (5): 1505–1522. Baker, T., A.S. Miner, and D.T. Eesley. 2003. Improvising firms: Bricolage, account giving and improvisational competencies in the founding process. Research Policy 32 (2): 255–276. Baker, T., and R.E. Nelson. 2005. Creating something from nothing: Resource construction through entrepreneurial bricolage. Administrative Science Quarterly 50 (3): 329–366. Baldacchino, L., G. Ucbasaran, L. Cabantous, and A. Lockett (2015). Entrepreneurship research on intuition: A critical analysis and research agenda. International Journal of Management Reviews, 17 (2): 212–231. Bansal, P., A. Kim, and M.O. Wood. 2018. Hidden in plain sight: The importance of scale in organisations’ attention to issues. Academy of Management Review 43 (2): 217–241.

90

C. Busch and M. Grimes

Baron, R.A. 2008. The role of affect in the entrepreneurial process. Academy of Management Review 33 (2): 328–340. Batjargal, B., M.A. Hitt, A.S. Tsui, J.-L. Arregle, J.W. Webb, and T.L. Miller. 2013. Institutional polycentrism, entrepreneurs’ social networks, and new venture growth. Academy of Management Journal 56 (4): 1024–1049. Beale, R. 2007. Supporting serendipity: Using ambient intelligence to augment user exploration for data mining and web browsing. International Journal of Human-Computer Studies 65 (5): 421–433. Birdi, K.S. 2005. No idea? Evaluating the effectiveness of creativity training. Journal of European Industrial Training 29 (2): 102–111. Boyd, W. 1998. Armadillo: A Novel. New York: Vintage. Brinker, J.K., Z.H. Chin, and R. Wilkinson. 2014. Ruminative thinking style and the MMPI-2-RF. Personality and Individual Differences 66: 102–105. Brown, S. 2005. Science, serendipity and contemporary marketing condition. European Journal of Marketing 39: 1229–1234. Brown, K.W., and R.M. Ryan. 2003. The benefits of being present: Mindfulness and its role in psychological well-being. Journal of Personality and Social Psychology 84 (4): 822–848. Brown, K.W., R.M. Ryan, and J.D. Creswell. 2007. Mindfulness: Theoretical foundations and evidence for its salutary effects. Psychological Inquiry 18 (4): 211–237. Burgelman, R.A. 2003. Practice and you get luckier. European Business Forum 1 (16): 38–39. Burt, R.S. 1997. The contingent value of social capital. Administrative Science Quarterly 42 (2): 339–365. Busch, C. 2014. Substantiating social entrepreneurship research: Exploring the potential of integrating social capital and networks approaches. International Journal of Entrepreneurial Venturing 6 (1): 69–84. Busch, C. 2020a. The Serendipity Mindset: The Art and Science of Creating Good Luck. New York: Penguin Random House. LCCN 008801. Busch, C. 2020b. Toward an enlightened form of capitalism: The changing role of private organisations in the context of global Affairs. In The Future of Global Affairs, ed. C. Ankersen and W.P.S. Sidhu, 97–122. New York: Palgrave Macmillan. Busch, C. 2021. Connect the dots: The art and science of creating good luck. London: Penguin Life. ISBN: 9780241402122. Busch, C. 2022. Towards a theory of serendipity: A systematic review and conceptualization. Journal of Management Studies, forthcoming. https://doi.org/10.1111/joms.12890. Busch, C., and H.G. Barkema. 2022. Align or perish: Social enterprise orchestration in Sub-Saharan Africa. Journal of Business Venturing 37 (2): 106187. Busch, C., and H.G. Barkema. 2021. From necessity to opportunity: Scaling bricolage in resourceconstrained environments. Strategic Management Journal 42: 741–773. Busch, C., and H.G. Barkema. 2020. Planned luck: How incubators can facilitate serendipity for nascent entrepreneurs through fostering network embeddedness. Entrepreneurship Theory and Practice, in press. https://doi.org/10.1177/1042258720915798. Busch, C., and H.G. Barkema. 2019. Social entrepreneurs as network orchestrators. In Handbook of Inclusive Innovation, ed. G. George, P. Tracey, T. Baker, and J. Havovi, 464–486. London: Edward Elgar Publishing. Busch, C., and M.G. Grimes. 2023. Eureka!: Improving the effects of artificial intelligence on serendipitous innovation within organizations. Academy of Management Annual Meeting Proceedings, Boston. Busch, C., and L. Hehenberger. 2022. How to evaluate the impact of corporate purpose. MIT Sloan Management Review, Fall Issue. Busenitz, L.W. 1996. Research on entrepreneurial alertness. Journal of Small Business Management 34 (4): 35–44. Casciaro, T., F. Gino, and M. Kouchaki. 2014. The contaminating effects of building instrumental ties: How networking can make us feel dirty. Administrative Science Quarterly 59: 705–735.

5 Serendipity in Entrepreneurship, Strategy, and Innovation—A Review …

91

Catmull, E. 2008. How Pixar fosters collective creativity. Harvard Business Review, September. Chanan, G., and C. Miller. 2010. The big society: How it could work: A positive idea at risk from caricature. PACES; http://www.pacesempowerment.co.uk/pacesempowerment/Publicati ons.html Chen, L.Y. 2016. Tencent using internal competition in app push. https://www.bloomberg.com/ news/videos/2016-09-14/tencent-using-internal-competition-in-app-push Christoff, K., A.M. Gordon, J. Smallwood, R. Smith, and J.W. Schooler. 2009. Experience sampling during fMRI reveals default network and executive system contributions to mind wandering. Proceedings of the National Academy of Sciences 106 (21): 8719–8724. Clegg, S.R., J. Vieira da Cunha, and M.P. Cunha. 2002. Management paradoxes: A relational view. Human Relations 55 (5): 483–503. Coad, A. 2009. The Growth of Firms: A Survey of Theories and Empirical Evidence. Cheltenham: Edward Elgar. Cohen, M.D., J.G. March, and J.P. Olsen. 1972. A garbage can model of organisational choice. Administrative Science Quarterly 17 (1): 1–25. Cohen, S.L., C.B. Bingham, and B.L. Hallen. 2019. The role of accelerator designs in mitigating bounded rationality in new ventures. Administrative Science Quarterly 4 (64): 810–854. Conboy, K. 2009. Agility from first principles: Reconstructing the concept of agility in information systems development. Information Systems Research 20 (3): 329–354. Conrad, K. 1958. Die beginnende Schizophrenie. Versuch einer Gestaltanalyse des Wahns [The Onset of Schizophrenia: An Attempt to Form an Analysis of Delusion]. Stuttgart: Georg Thieme Verlag. Copeland, S. 2018. Fleming leapt on the unusual like a weasel on a vole: Challenging the paradigms of discovery in science. Perspectives on Science 26 (6): 694–721. Cornelissen, J.P., and J.S. Clarke. 2010. Imagining and rationalizing opportunities: Inductive reasoning and the creation and justification of new ventures. Academy of Management Review 35 (4): 539–557. Cornelissen, J.P., R. Holt, and M. Zundel. 2011. The role of analogy and metaphor in the framing and legitimization of strategic change. Organisation Studies 32 (12): 1701–1716. Corner, P.D., and M. Ho. 2010. How opportunities develop in social entrepreneurship. Entrepreneurship Theory and Practice 34 (4): 635–659. Cosgrave, P. 2012. Engineering Serendipity. https://blog.websummit.com/engineering-serendipitystory-web-summits-growth/. Embury-Denis, 2017. Cosmelli, D., and D.D. Preiss. 2014. On the temporality of creative insight: A psychological and phenomenological perspective. Frontiers in Psychology 5 (1184): 1–6. Cunha, M.P., S. Clegg, and S. Mendonça. 2010. On serendipity and organizing. European Management Journal 28 (5): 319–330. Cunha, M.P., S. Clegg, A. Rego, and P. Neves. 2014. Organisational improvisation: From the constraint of strict tempo to the power of the avant-garde. Creativity and Innovation Management 23 (4): 359–373. Dane, E. 2011. Paying attention to mindfulness and its effects on task performance in the workplace. Journal of Management 37 (4): 997–1018. Danneels, E. 2011. Trying to become a different type of company: Dynamic capability at Smith Corona. Strategic Management Journal 32 (1): 1–31. De Bono, E. 1985. Six Thinking Hats: An Essential Approach to Business Management. London: Little, Brown. De Bono, E. 1992. Serious Creativity: Using the Power of Lateral Thinking to Create New Ideas. New York: HarperBusiness. DeJordy, R., M. Scully, M.J. Ventresca, and W.D. Creed. 2020. Inhabited ecosystems: Propelling transformative social change between and through organisations. Administrative Science Quarterly 65 (4): 931–971. Denrell, J., C. Fang, and C. Liu. 2015. Perspective–chance explanations in the management sciences. Organisation Science 26 (3): 923–940.

92

C. Busch and M. Grimes

Denrell, J., C. Fang, and S.G. Winter. 2003. The economics of strategic opportunity. Strategic Management Journal 24 (10): 977–990. Denrell, J., C. Fang, and Z. Zhao. 2013. Inferring superior capabilities from sustained superior performance: A Bayesian analysis. Strategic Management Journal 34 (2): 182–196. De Rond, M. 2014. The structure of serendipity. Culture and Organisation 20 (5): 342–358. De Rond, M., Moorhouse, A., and M. Rogan. 2011. Make serendipity work for you. Harvard Business Review. https://hbr.org/2011/02/make-serendipity-work. Dew, N. 2009. Serendipity in entrepreneurship. Organization Studies 30 (07): 735–753. Dhanaraj, C., and A. Parkhe. 2006. Orchestrating innovation networks. Academy of Management Review 31 (3): 659–669. Diaz de Chumaceiro, C.L. 2004. Serendipity and pseudoserendipity in career paths of successful women: Orchestra conductors. Creativity Research Journal 16 (2–3): 345–356. di Falco, S., and E. Bulte. 2011. A dark side of social capital? Kinship, consumption, and savings. Journal of Development Studies 47 (8): 1128–1151. Driver, J. 1989. The virtues of ignorance. The Journal of Philosophy 86 (7): 373–384. Duncker, K. 1945. On problem solving. Psychological Monographs 58 (5): i–113. Edmondson, A. 1999. Psychological safety and learning behavior in work teams. Administrative Science Quarterly 44 (2): 350–383. Eisenhardt, K.M., M.E. Graebner, and S. Sonenshein. 2016. Grand challenges and inductive methods: Rigor without rigor mortis. Academy of Management Journal 59 (4): 1113–1123. Engel, Y., M. Kaandorp, and T. Elfring. 2017. Toward a dynamic process model of entrepreneurial networking under uncertainty. Journal of Business Venturing 32 (1): 35–51. Entman, R.M. 1993. Framing: Toward clarification of a fractured paradigm. Journal of Communication 43 (4): 51–58. Erdelez, S. 1999. Information encountering: It’s more than just bumping into information. American Society for Information Science 25: 25–29. Ericsson, K.A., and J.J. Staszewski. 1989. Skilled memory and expertise: Mechanisms of exceptional performance. In Complex Information Processing: The Impact of Herbert A. Simon, eds. Klahr, D. and K. Kotovsky. Hillsdale, NJ: Erlbaum. Ernst, and Young. 2016. The upside of disruption: Megatrends shaping 2016 and beyond. https://assets.ey.com/content/dam/ey-sites/ey-com/en_gl/topics/disruption/ey-megatr ends-final-onscreen.pdf. Eskinazi, M., and I. Giannopulu. 2021. Continuity in intuition and insight: From real to naturalistic virtual environment. Scientific Reports 11 (1): 1–12. Eubanks, D., S. Murphy, and M. Mumford. 2010. Intuition as an influence on creative problemsolving: The effects of intuition, positive affect, and training. Creativity Research Journal 22 (2): 170–184. Fan, J., J. Zhang, and K. Yu. 2012. Vast portfolio selection with gross-exposure constraints. Journal of the American Statistical Association 107: 592–606. Feld, S.L. 1981. The focused organisation of social ties. American Journal of Sociology 86 (5): 1015–1035. Feldman, G., E. Dunn, C. Stemke, K. Bell, and J. Greeson. 2014. Mindfulness and rumination as predictors of persistence with a distress tolerance task. Personality and Individual Differences 56: 154–158. Felin, T., Gambardella, A., and T.R. Zenger. 2020. Value lab: A tool for entrepreneurial strategy. SSRN Electronic Journal. Felin, F., and T.R. Zenger. 2015. Strategy, problems and a theory for the firm. Organisation Science 27 (1): 207–221. Fenigstein, A., M.F. Scheier, and A.H. Buss. 1975. Public and private self-consciousness: Assessment and theory. Journal of Counselling and Clinical Psychology 43 (4): 522–527. Fildes, R., P. Goodwin, M. Lawrence, and K. Nikolopoulos. 2009. Effective forecasting and judgmental adjustments: An empirical evaluation and strategies for improvement in supply-chain planning. International Journal of Forecasting 25 (1): 3–23.

5 Serendipity in Entrepreneurship, Strategy, and Innovation—A Review …

93

Fine, G.A., and J.G. Deegan. 1996. Three principles of serendip: Insight, chance and discovery in qualitative research. International Journal of Qualitative Studies in Education 9 (4): 434–447. Fisher, C.M., and T. Amabile. 2009. Creativity, Improvisation and Organisations. New York, NY: Routledge. Fleming, L., and D.M. Waguespack 2007. Brokerage, boundary spanning, and leadership in open innovation communities. Organization Science, 18 (2): 165–180. Fleming, L., and O. Sorenson. 2004. Science as a map in technological search. Strategic Management Journal 25 (8–9): 909–928. Flick, U. 2009. An Introduction to Qualitative Research (4th ed.). Sage. Fligstein, N. 2001. Social skill and the theory of fields. Sociological Theory 19 (2): 105–125. Foster, A., and N. Ford. 2003. Serendipity and information seeking: An empirical study. Journal of Documentation 59 (3): 321–340. Fultz, E.F., and K.M. Hmieleski. 2021. The art of discovering and exploiting unexpected opportunities: The roles of organisational improvisation and serendipity in new venture performance. Journal of Business Venturing. Furnari, S. 2014. Interstitial spaces: Microinteraction settings and the genesis of new practices between institutional fields. Academy of Management Review 39 (4): 4. Garriga, H., G. von Krogh, and S. Spaeth. 2013. How constraints and knowledge impact open innovation. Strategic Management Journal 34 (9): 1134–1144. Garud, R., and P. Karnøe. 2003. Bricolage versus breakthrough: Distributed and embedded agency in technology entrepreneurship. Research Policy 32 (2): 277–300. Garud, R., J. Gehman, and A. Giuliani. 2018. Serendipity arrangements for exapting science-based innovations. Academy of Management Perspectives 32 (1): 125–140. Gentner, D., and A.B. Markman. 1997. Structure mapping in analogy and similarity. American Psychologist 52 (1): 45–56. German, T.P., and H.C. Barrett. 2005. Functional fixedness in a technologically sparse culture. Psychological Science 16 (1): 1–5. German, T.P., and M.A. Defeyter. 2000. Immunity to functional fixedness in young children. Psychonomic Bulletin and Review 7 (4): 707–712. Geroski, P.A. 2005. Understanding the implications of empirical work on corporate growth rates. Managerial and Decision Economics 26 (2): 129–138. Gick, M.L., and K.J. Holyoak. 1980. Analogical problem solving. Cognitive Psychology 12 (3): 306–355. Gilhooly, K.J., and P. Murphy. 2005. Differentiating insight from non-insight problems. Thinking and Reasoning 11 (3): 279–302. Giudici, A., P. Reinmoeller, and D. Ravasi. 2018. Open-system orchestration as a relational source of sensing capabilities: Evidence from a venture association. Academy of Management Journal 61 (4): 1369–1402. Giustiniano, L., M.P. Cunha, and S. Clegg. 2016. Organisational zemblanity. European Management Journal 34: 7–21. Gould, S.J., and E.S. Vrba. 1982. Exaptation—A missing term in the science of form. Paleobiology 8 (1): 4–15. Graebner, M.E. 2004. Momentum and serendipity: How acquired leaders create value in the Integration of technology firms. Strategic Management Journal 25 (89): 751–777. Grant, A. 2017. Originals. New York: Penguin Random House. Grimes, M.G. (2018). The pivot: how founders respond to feedback through idea and identity work. Academy of Management Journal, 61 (5): 1692–1717. Grimes, M.G., and T.J. Vogus. 2021. Inconceivable! Possibilistic thinking and the sociocognitive underpinning of entrepreneurial responses to grand challenges. Organisation Theory 2 (2). Grimes, M.G., T.A. Williams, and E.Y. Zhao. 2020. Beyond hybridity: Accounting for the values complexity of all organisations in the dtudy of mission and mission Drift. Academy of Management Review 45 (1): 234–238.

94

C. Busch and M. Grimes

Grimes, M.G., T.A. Williams, and E.Y. Zhao. 2019. Anchors aweigh: The sources, variety, and challenges of mission drift. Academy of Management Review 44: 800–818. Guy, J.H., G.B. Deakin, A.M. Edwards, C.M. Miller, and D.B. Pyne. 2015. Adaptation to hot environmental conditions: An exploration of the performance basis, procedures and future directions to optimise opportunities for elite athletes. Sports Medicine 45 (3): 303–311. Gyori, B. (2018). Creating kismet: What artists can teach academics about serendipity. In Serendipity in Rhetoric, Writing and Literacy Research, Goggin, M.D., and P.N. Goggin, 247–256. Louisville, CO: Utah State University Press. Gyori, C., M. Brahman, C. Busch, L. Sharp, and T. Kazakova. 2019. Leaders on Purpose CEO Study. Washington, D.C.: Leaders on Purpose. Hallen, B.L., S.L. Cohen, and C.B. Bingham. 2020. Do accelerators work? If so, how? Organization Science 31 (2): 378–414. Hallen, B.L., and K.M. Eisenhardt. 2012. Catalyzing strategies and efficient tie formation: How entrepreneurial firms obtain investment ties. Academy of Management Journal 55 (1): 35–70. Hannan, M.T., L. Pólos, and G.R. Carroll. 2003. Cascading organisational change. Organization Science 14 (5): 463–482. Hargadon, A.B., and B.A. Bechky. 2006. When collective of creative becomes creative collectives: A field study of problem solving at work. Organization Science 17 (4): 484–500. Hargadon, A.B., and R.I. Sutton. 1997. Technology Bbrokering and innovation in a product development firm. Administrative Science Quarterly 42 (4): 716–749. Harmeling, S., and S. Sarasvathy. 2013. When contingency is a resource: Educating entrepreneurs in the balkans, the Bronx and beyond. Entrepreneurship Theory and Practice 37. Helfat, C.E., Finkelstein, S., Mitchell, W., Peteraf, M., Sing, H., and D. Teece. 2007. Dynamic Capabilities: Understanding Strategic Change in Organizations. Malden, MA: Wiley-Blackwell. Henderson, A.D., M.E. Raynor, and M. Ahmed. 2012. How long must a firm be great to rule out chance? Benchmarking sustained superior performance without being fooled by randomness. Strategic Management Journal 33 (4): 387–406. Herndon, T., M. Ash, and R. Pollin. 2014. Does high public debt consistently stifle economic growth? A critique of Reinhart and Rogoff. Cambridge Journal of Economics 38 (2): 257–279. Hmieleski, K.M., and A.C. Corbett. 2006. Proclivity for improvisation as a predictor of entrepreneurial intentions. Journal of Small Business Management 44: 45–63. Hofstede, G. 1984. Culture’s Consequences: International Differences in Work-Related Values. Beverly Hills: Sage. House, R.J., P.J. Hanges, M. Javidan, P.W. Dorfman, and V. Gupta. 2004. Culture, Leadership and Organisations: The Globe Study of 62 Societies. Thousand Oaks: Sage. Huang, L., and J.L. Pearce. 2015. Managing the unknowable: The effectiveness of early-stage investor gut feel in entrepreneurial investment decisions. Administrative Science Quarterly 60 (4): 634–670. Huldtgren, A., Mayer, C., Kierepka, O., and C. Geiger. 2014. Towards serendipitous urban encounters with sound track of your life. In Proceedings of the 11th Conference on Advances in Computer Entertainment Technology. New York: ACM Isaacson, W. 2011. Steve Jobs. New York: Simon and Schuster. Isen, A.M., K.A. Daubman, and G.P. Nowicki. 1987. Positive affect facilitates creative problem solving. Journal of Personality and Social Psychology 52 (6): 1122–1131. Jeppesen, L.B., and K.R. Lakhani. 2010. Marginality and problem-solving effectiveness in broadcast Search. Organizational Science 21 (5): 1016–1033. Kahneman, D. 2011. Thinking, Fast and Slow. London: Penguin. Kamenov, K. 2013. Self-awareness and behaviour within social structures. Economic Archive/ narodnostopanski Archiv 66 (4): 3–13. Kapoor, R., and S. Agarwal. 2017. Sustaining superior performance in business ecosystems: Evidence from application software developers in the IOS and Android smartphone ecosystems. Organization Science 28 (3): 531–551.

5 Serendipity in Entrepreneurship, Strategy, and Innovation—A Review …

95

Khavul, S., G.D. Bruton, and E. Wood. 2009. Informal family business in Africa. Entrepreneurship Theory and Practice 33 (6): 1219–1238. Khayesi, J.N.O., and G. George. 2011. When does the socio-cultural context matter? Communal orientation and entrepreneurs’ resource accumulation efforts in Africa. Journal of Occupational and Organizational Psychology 84 (3): 471–492. Kiggundu, M.N. 2002. Entrepreneurs and entrepreneurship in Africa: What is known and what needs to be done. Journal of Developmental Entrepreneurship 7: 239–258. Kilduff, M., and W. Tsai. 2003. Social Networks and Organizations. Sage. Kiss, A.N., W. Danis, S. Nair, and R. Suddaby. 2020. Accidental tourists? A cognitive exploration of serendipitous internationalization. International Small Business Journal 38 (2): 65–89. Kirzner, I. 1979. Perception, Opportunity and Profit. Chicago: University of Chicago Press. Kornberger, M., S.R. Clegg, and C. Rhodes. 2005. Learning/becoming/organizing. Organization 12 (2): 147–167. Krumrei-Mancuso, E.J., and S.V. Rouse. 2015. The development and validation of the Comprehensive Intellectual Humility Scale. Journal of Personality Assessment 98: 209–221. Kurup, U., P.G. Bignoli, J.R. Scally, and N.L. Cassimatis. 2011. An architectural framework for complex cognition. Cognitive Systems Research 12 (3–4): 281–292. Lane, J.N., I. Ganguli, P. Gaule, E. Guinan, and K.R. Lakhani. 2020. Engineering serendipity: When does knowledge sharing lead to knowledge production? Strategic Management Journal 42: 1215–1244. Leary, M.R., K.J. Diebels, E.K. Davisson, K.P. Jongman-Sereno, J.C. Isherwood, K.T. Raimi, S.A. Deffler, and R.H. Hoyle. 2017. Cognitive and interpersonal features of intellectual humility. Personality and Social Psychology Bulletin 43 (6): 793–813. Lehrer, J. 2011. Steve Jobs: Technology alone is not enough. New Yorker, 7 October 2011. Levinthal, D., and C. Rerup. 2006. Crossing an apparent chasm: Bridging mindful and less-mindful perspectives on organizational learning. Organization Science 17: 502–513. Lewis, W.S. 1965. Foreword. In Serendipity and the Three Princes, from the Peregrinaggio of 1557, ed. Remer T.G. Norman: University of Oklahoma Press. Laursen, E., and C. Salter. 2006. Open for innovation: The role of openness in explaining innovation performance among UK manufacturing firms. Strategic Management Journal 27 (2): 131–150. Liang, R.H. 2012. Designing for unexpected encounters with digital products: Case studies of serendipity as felt experience. International Journal of Design 6 (1): 41–58. Lindsay, G. 2013. Engineering serendipity. New York Times. https://www.nytimes.com/2013/04/07/ opinion/sunday/engineering-serendipity.html. Liu, C., and M. de Rond. 2016. Good night and good luck: Perspectives on luck in management scholarship. Academy of Management Annals 10 (1): 1–56. Logue, D., and M. Grimes. 2019. Platforms for the people: Enabling civic crowdfunding through the cultivation of institutional infrastructure. Strategic Management Journal. Magni, M., B. Provera, and L. Proserpio. 2010. Individual attitude toward improvisation in information systems development. Behavioral Information Technology 29: 245–255. Makri, S., A. Blandford, M. Woods, S. Sharples, and D. Maxwell. 2014. “Making my own luck”: Serendipity strategies and how to support them in digital information environments. Journal of the Association for Information Science and Technology 65 (11): 2179–2194. Makri, S., and A. Blandford. 2012. Coming across information serendipitously—Part 1. Journal of Documentation 68 (5): 684–705. Mandi, A., S. Mullainathan, E. Shafir, and Z. Zhao. 2013. Poverty impedes cognitive function. Science 341 (6149): 976–980. Mason, M.F., et al. 2007. Wandering minds: The default network and stimulus-independent thought. Science 315 (5810): 393–395. Maurer, I., and M. Ebers. 2006. Dynamics of social capital and their performance implications: Lessons from biotechnology Start-ups. Administrative Science Quarterly 51 (2): 262–292. McCay-Peet, L., and E.G. Toms. 2010. The process of serendipity in knowledge work. Association for Computing Machinery, Information Interaction in Context Symposium.

96

C. Busch and M. Grimes

McCay-Peet, L., and E.G. Toms. 2012. The serendipity quotient. In Proceedings of the American Society for Information Science and Technology. McCay-Peet, L., and E.G. Toms. 2018. Researching serendipity in digital information environments. Synthesis Lectures on Information Concepts Retrieval and Services 9 (6): 1–91. McCay-Peet, L., E.G. Toms, and E.K. Kelloway. 2015. Examination of relationships among serendipity, the environment and individual differences. Information Processing and Management 51: 391–412. McCrae, R.R., and A.R. Sutin. 2009. Openness to experience. In Handbook of Individual Differences in Social Behavior, ed. M.R. Leary and R.H. Hoyle, 257–273. New York, NY: Guilford Press. McElroy, S.E., K.G. Rice, D.E. Davis, J.N. Hook, P.C. Hill, E.L. Worthington, and D.R. Van Tongeren. 2014. Intellectual humility: Scale development and theoretical elaborations in the context of religious leadership. Journal of Psychology and Theology 42: 19–30. McGahan, A.M., and M.E. Porter. 2002. What do we know about variance in accounting profitability? Management Science 48 (7): 834–851. McGahan, A., M.L.A.M. Bogers, H. Chesbrough, and M. Holgersson. 2021. Tackling societal challenges with open innovation. California Management Review, 63(2): 49–61. McKeever, E., A. Anderson, and S. Jack. 2014. Social embeddedness in entrepreneurship research: The importance of context and community. In Handbook of Research on Small Business and Entrepreneurship, ed. E. Chell and M. Karatas-Ozkan, 222–235. Edward Elgar. McKeever, E., S. Jack, and A. Anderson. 2015. Embedded entrepreneurship in the creative reconstruction of place. Journal of Business Venturing 30 (1): 50–65. McMullen, J.S., and D.A. Shepherd. 2006. Entrepreneurial action and the role of uncertainty in the theory of the entrepreneur. Academy of Management Review 31 (1): 132–152. McNally, A., C.K. Prier, and D.W.C. Macmillan. 2011. Discovery of an α-amino C-H arylation reaction using the strategy of accelerated serendipity. Science 334: 1114–1117. Merrigan, T.W. 2019. How a Global Social Network Seeks Connection in Decentralization. Stanford E-corner, Stanford University. https://ecorner.stanford.edu/articles/how-a-global-social-net work-seeks-connection-in-decentralization/. Merton, R. 1948. The bearing of empirical research upon the development of social theory. American Sociological Review 13 (5): 505–515. Merton, R.K., and E. Barber. 2004. The Travels and Adventures of Serendipity: A Study in Sociological Semantics and the Sociology of Science. Princeton, NJ: Princeton University Press. Meyers, M. 2007. Happy accidents. New York: Arcade. Mintzberg, H., and J.A. Waters. 1985. Of strategies, deliberate and emergent. Strategic Management Journal 6 (3): 257–272. Mintzberg, H., R.T. Pascale, M. Goold, and R.P. Rumelt. 1996. The “honda effect” revisited. California Management Review 38: 78–117. Mirvahedi, S., and S. Morrish. 2017. The role of serendipity in opportunity exploration. Journal of Research in Marketing and Entrepreneurship 19 (2): 182–200. Mirvis, P.H. 1998. Variations on a theme: Practice improvisation. Organization Science 9 (5): 586–592. Mishara, A. 2010. Klaus Conrad, 1905–1961: Delusional mood, psychosis and beginning schizophrenia. Schizophrenia Bulletin 36 (1): 9–13. Nambisan, S., and R.A. Baron. 2013. Entrepreneurship in innovation ecosystems: Entrepreneurs’ self-regulatory processes and their implications for new venture success. Entrepreneurship Theory and Practice 37 (5): 1071–1097. Nonaka, I. 1991. The Knowledge-Creating Company. Harvard Business Review, November– December. Napier, N.K., and Q.-H. Vuong. 2013. Serendipity as a strategic advantage? In Strategic Management in the 21st Century, 1: Operational Environment, ed. T.J. Wilkinson and V.R. Kannan, 175–199. Oxford: Blackwell. Nelson, C.L. 2008. Intelligibility since, 1969. New York: John Wiley and Sons.

5 Serendipity in Entrepreneurship, Strategy, and Innovation—A Review …

97

Obstfeld, D., M.J. Ventresca, and G. Fisher. 2020. An assembly perspective of entrepreneurial projects: Social networks in action. Strategic Entrepreneurship Journal 14 (2): 149–177. Ozcan, P., and K.M. Eisenhardt. 2009. Origin of alliance portfolios: Entrepreneurs, network strategies, and firm performance. Academy of Management Journal 52 (2): 246–279. Pariser, E. 2012. The Filter Bubble: What the Internet is Hiding From You. New York: Penguin. Pelaprat, E., and M. Cole. 2011. ‘Minding the gap’: Imagination, creativity and human cognition. Integrative Psychological and Behavioral Science 45 (4): 397–418. Petrocelli, J.V., Z.L. Tormala, and D.D. Rucker. 2007. Unpacking attitude certainty: Attitude clarity and attitude correctness. Journal of Personality and Social Psychology 92: 30–41. Porter, C.M., and S.E. Woo. 2015. Untangling the networking phenomenon: A dynamic psychological perspective on how and why people network. Journal of Management 41: 1477–1500. Piketty, T. 2014. Capital in the Twenty-First Century. Cambridge, MA: Harvard University Press. Pirnot, M.T., D.A. Rankic, D.B.C. Martin, and D.W.C. Macmillan. 2013. Photoredox activation for the direct β-arylation of ketones and aldehydes. Science 339: 1593–1596. Pitsis, T.S., et al. 2003. Constructing the olympic dream: A future perfect strategy of project management. Organization Science 14 (5): 574–590. Portes, A., and J. Sensenbrenner. 1993. Embeddedness and immigration: Notes on the social determinants of economic action. American Journal of Sociology 98 (6): 1320–1350. Prabhu, J. 2017. Frugal innovation: Doing more with less for more. Philosophical Transactions of The Royal Society A Mathematical Physical and Engineering Sciences, 375 (2095): 20160372. Provan, K.G., and P. Kenis. 2008. Modes of network governance: Structure, management, and effectiveness. Journal of Public Administration Research and Theory 18 (2): 229–252. Ramus, T., A. Vaccaro, and S. Brusoni. 2017. Institutional complexity in turbulent times: Formalization, collaboration, and the emergence of blended logics. Academy of Management Journal 60 (4): 1253–1284. Rauch, M., and S. Ansari. 2021. From “publish or perish” to societal impact: Framing and serendipity in the emergence of a platform for responsible innovation. Journal of Management Studies, 59 (1): 1–91. Regner, P. 2003. Strategy creation in the periphery: Inductive versus deductive strategy making. Journal of Management 40 (1): 57–82. Reinecke, J., and S. Ansari. 2015. When times collide: Temporal brokerage at the intersection of markets and developments. Academy of Management Journal 58 (2): 618–648. Ritter, S.M., and A. Dijksterhuis. 2015. Creativity—The unconscious foundations of the incubation period. Frontiers in Human Neuroscience 8 (215): 1–10. Roberts, R. (1989). Serendipity. New York: John Wiley & Sons. Ross,W. 2022. Heteroscalar serendipity and the importance of accidents. In The Art of Serendipity, ed. W. Ross and S. Copeland. Palgrave MacMillan. Root-Bernstein, R.S. 1988. Setting the stage for discovery. The Sciences 28 (3): 26–34. Ross, W., and F. Vallée-Tourangeau. 2021. Microserendipity in the creative process. Journal of Creative Behavior 55 (3): 661–672. Rothaermel, F.T., and M. Thursby. 2005. Incubator firm failure or graduation? The role of university linkages. Research Policy 34: 1076–1090. Rowson, J., Broome, S., and A. Jones. 2010. Connected Communities: How Social Networks Power and Sustain the Big Society. RSA; www.thersa.org/discover/publications-and-articles/reports/ connected-communities-how-social-networks-power-and-sustain-the-big-society. Sagan, C. 1995. The Demon-Haunted World: Science as a Candle in the Dark. New York: Random House. Sand, M., and K. Jongsma. 2020. Scientists’ views on (moral) luck. Journal of Responsible Innovation 7: 64–85. Sarasvathy, S.D. 2008. Effectuation: Elements of Entrepreneurial Expertise. Cheltenham: Edward Elgar. Schooler, J.W., and J. Melcher. 1995. The ineffability of insight. In The Creative Cognition Approach, ed. S.M. Smith, T.B. Ward, and R.A. Finke, 97–133. Cambridge, MA: MIT Press.

98

C. Busch and M. Grimes

Schultz, H. 1998. Pour Your Heart into it: How Starbucks Built a Company One Cup at a Time. New York: Hyperion. Schutjens, V., and E. Stam. 2003. The evolution and nature of young firm networks. Small Business Economics 21 (2): 115–134. Schwenk, C., and H. Thomas. 1983. Effects of conflicting analyses on managerial decision making: A laboratory experiment. Decision Sciences 14 (4): 447–612. Seckler, C. 2021. Who adopts an error management orientation? Discovering the role of humility. Academy of Management Discoveries. Shane, S.A. 2003. A General Theory of Entrepreneurship: The Individual Opportunity Nexus. Edward Elgar Publishing. Shepherd, D.A., Seyb, S., and G. George. 2021. Grounding business models: Cognition, boundary objects, and business-model change. Academy of Management Review. Shipilov, A., R. Gulati, M. Kilduff, S. Li, and W. Tsai. 2014. Relational Pluralism within and between organizations. Academy of Management Journal 57 (2): 449–459. Skinner, B. F. (1948). Superstition in the pigeon. Journal of Experimental Psychology, 38 (2): 168–172. Silverman, D. 2013. Doing Quantitative Research. London: Sage. Simon, H.A. 1977. The New Science of Management Decision. Englewood Cliffs, NJ: Prentice-Hall. Sio, U.N., and T.C. Ormerod. 2009. Does incubation enhance problem solving? A meta-analytic review. Psychological Bulletin 135 (1): 94–120. Smith, R.P., and S.D. Eppinger. 1997. Identifying controlling features of engineering design iteration. Management Science 43 (3): 276–293. Spiegel, J.S. 2012. Open-mindedness and intellectual humility. Theory and Research in Education 10: 27–38. Stanek, W., A. Valero, G. Calvo, and L. Czarnowska. 2017. Thermodynamics for Sustainable Management of Natural Resources. New York: Springer. Sting, F., Fuchs, C., Schlickel, M., and O. Alexy. 2019. How to overcome the bias we have towards our own ideas. Harvard Business Review. Stock-Homburg, R.M., Heald, S.L.M., Holthaus, C., Gillert, N.L., and E. von Hippel. 2021. Needsolution pair recognition by household sector individuals: Evidence, and a cognitive mechanism explanation. Research Policy, 50: 104068. Sutton, A. 2016. Measuring the effects of self-awareness: Construction of the self-awareness outcomes questionnaire. Europe’s Journal of Psychology 12 (4): 645–658. Spigel, B. 2017. The relational organization of entrepreneurial ecosystems. Entrepreneurship Theory and Practice 41 (1): 49–72. Teece, D.J., G. Pisano, and A. Shuen. 1997. Dynamic capabilities and strategic management. Strategic Management Journal 18 (7): 509–533. Thomke, S.H., and T. Fujimoto. 2000. The effect of ‘front-loading’ problem-solving on product development performance. Journal of Product Innovation Management 17 (2): 128–142. Thompson, L. 2017. Research for better brainstorming. Harvard Business Review. https://hbr.org/ 2017/10/research-for-better-brainstorming-tell-an-embarrassing-story. Thompson, T.A., J.M. Purdy, and M.J. Ventresca. 2018. How entrepreneurial ecosystems take form: Evidence from social impact initiatives in Seattle. Strategic Entrepreneurship Journal 12 (1): 96–116. Toivonen, T. 2016. What is the social innovation community? Conceptualizing an emergent collaborative organisation. Journal of Social Entrepreneurship 7 (1): 49–73. Toms, E.G. 2000. Serendipitous information retrieval. In Proceedings of the 1st DELOS Network of Excellence Workshop on Information Seeking, Searching and Querying in Digital Libraries, 11– 12. Sophia Antipolis, France: European Research Consortium for Informatics and Mathematics. Topolinski, S., and R. Reber. 2010. Gaining insight into the ‘aha’ experience. Current Directions in Psychological Science 19 (6): 402–405. Tyre, M., and E. von Hippel. 1997. The situated nature of adaptive learning in organizations. Organization Science 8 (1): 1–107.

5 Serendipity in Entrepreneurship, Strategy, and Innovation—A Review …

99

Van Andel, P. 1994. Anatomy of the unsought finding. Serendipity: Origin, history, domains, traditions, appearances, patterns and programmability. The British Journal for the Philosophy of Science 45 (2), 631–648. Van Gaal, S., F.P. de Lange, and M.X. Cohen. 2012. The role of consciousness in cognitive control and decision making. Frontiers in Human Neuroscience 6 (121): 1–15. Vera, D., and M.M. Crossan. 2005. Improvisation and innovative performance in teams. Organization Science 16: 203–224. Volkema, R.J. 1983. Problem formulation in planning and design. Management Science 29 (6): 639–652. Von Hippel, E., and M.J. Tyre. 1996. How learning by doing is done: Problem identification in novel process equipment. Research Policy 24 (1): 1–12. Von Hippel, E., and G. von Krogh. 2016. Identifying viable ‘need–solution pairs’: Problem solving without problem formulation. Organizational Science 27 (1): 207–221. Voss, J.L., K.D. Federmeier, and K.A. Paller. 2012. The potato chip really does look like elvis! Neural hallmarks of conceptual processing associated with finding novel shapes subjectively meaningful. Cerebral Cortex 22 (10): 2354–2364. Weick, K. 1998. Improvisation as a mindset for organizational analysis. Organization Science 9: 543–555. Wenger, E. 1998. Communities of Practice. Cambridge: Cambridge University Press. Winter, S.G. 2003. Understanding dynamic capabilities. Strategic Management Journal 24 (10): 991–995. Wiseman, R. 2003. The Luck Factor. London: Random House. Yaqub, O. 2018. Serendipity: Towards a taxonomy and a theory. Research Policy 47 (1): 169–179. Zahra, S.A., and G. George. 2002. Absorptive capacity: A review, reconceptualization and extension. Academy of Management Review 27 (2): 185–203. Zenger, T.R. 2013. What is the theory of your firm? Harvard Business Review.

Chapter 6

Serendipity and the History of the Philosophy of Science Samantha Copeland

Abstract Samantha Copeland takes this chapter to delve into the history of philosophy of science, paying particular attention to the discussions around scientific discovery and the assumptions made by philosophers along the way about what parts of the discovery process can and cannot be studied. Copeland suggests that serendipity research might shed light on what has been left outside of philosophical investigation. She focusses in particular on the seeming ‘leap’ that scientists must take when discoveries happen, between a state of not-knowing to a state of recognizing the scientific value of an observation or event. Most philosophical accounts tend towards internalism (that is, assuming the important steps in discovery occur only in the mind), or the focus remains on what happens after an accident or chance encounter rather than on the encounter itself. Copeland offers an alternative interpretation from the perspective of her serendipity research, on what the interaction between chance and reason can tell us about scientific discovery more generally. That is, she argues, the intersection of chance and wisdom provides philosophy with the opportunity to better understand how our minds interact with the world to produce knowledge.

S. Copeland (B) Ethics and Philosophy of Technology Section, Department of Values, Technology and Innovation, Faculty of Technology, Policy and Management, Building 31, Office B4.140, Jaffalaan 5, 2628 BX Delft, The Netherlands e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 S. Copeland et al. (eds.), Serendipity Science, https://doi.org/10.1007/978-3-031-33529-7_6

101

102

S. Copeland

There is little work done in the philosophy of science1 that focusses explicitly on the nature of serendipity, that category of discoveries which occur at the intersection of chance and wisdom (Copeland 2019). Literature on serendipity and accidental discovery can be found more often in the social sciences. There, the contingencies of personal traits, context and culture are studied as driving forces behind scientific knowledge production, looking closely to the history and practice of science (instead of philosophy’s more narrow emphasis on the reasoning that scientists employ). Merton and Barber (2004) note in their seminal work on the word ‘serendipity’ that appreciation for the role of chance in the practice and progress of science has risen and fallen over time, varies from field to field, and that the opinions of scientists and the public may differ on whether it is a good or a bad thing that discoveries often occur accidentally. Some have argued that science is even mostly due to chance (e.g., see Chap. 5 in Trout 2016). Whereas others have judged a reliance on happy accidents to be a sign of laziness, arguing that a belief in progress through serendipity is for those who want science to be easy enough for anyone to do it, or to discredit those who consider it an elite, skilled enterprise (e.g., Merton and Barber 2004, Chap. 9). Indeed, because serendipity occurs in all areas of life and study, there seems little reason to think it marks something special about science itself, or scientists, and so it might well fall outside of the scope of interest for many, belonging to the mundane. For others, however, serendipity happens only to the ‘prepared mind’ and as such can indeed be a sign of deep skill and scientific ability (Cannon 1940; Copeland 2018; McAllister 2016). Western philosophers of science,2 however, most generally focus on teasing out the nature of what we call scientific reasoning, and explaining how we obtain knowledge through practice (see Strevens 2020 for a critique of this assumption). I argue that in the philosophical literature we see little concern for a deeper understanding of chance, and the accidental tends to be left out of accounts of how science happens—externalised,3 mentioned but unexplained, tangential or missing altogether from what counts for understanding discovery—with few exceptions.

1

There are other areas of philosophical study where luck and chance have played a more prominent role, namely in epistemology, where examples of ‘epistemic luck’ have been famously used to critique universal theories of knowledge acquisition (e.g., the Gettier examples, wherein people have knowledge but have gained it or are able to confirm it because of accidental features of the process, rather than rational, intentional processes in themselves). There is ongoing debate on epistemic luck (see particularly the work of and in response to Duncan Pritchard, on veritic and reflective luck; see also (Rescher 1995) for some similarities between serendipity in science and epistemic luck). 2 My focus here and elsewhere in my work so far remains unfortunately constrained by a Western approach; there is much to be done, and work I hope yet to do, toward understanding non-Western accounts of science in relation to chance. 3 For instance, accounts of science often consider the social context in which scientific practice and theory formation take place to have influence on its development, but I contend that describing the accidental as a condition for science to occur, is not the same as including it within our account of science itself. The accident is taken for granted as a precursor to the subject of interest, and thereby stipulated in such accounts rather than studied.

6 Serendipity and the History of the Philosophy of Science

103

Nonetheless, it is the case that in the study of science and discovery, experts have been unable to avoid at least mentioning the frequency with which accidents, error and chance do come into play. Discovery stories often mention the interference of a dream, forgetfulness or other disruption in the otherwise rational research process. Famous examples such as Fleming’s chance noticing of the nearly-discarded petri dish to observe penicillin in action, or Kekulé’s dream of the encircled snake to arrive at an explanation for benzene’s chemical shape, capture the imagination in popular collections of tales of discovery (e.g., Meyers 2007; Roberts 1989; van Andel 1994). These elements in such tales create a sense of magic around discovery that lends itself to storytelling. Merton may have been the first to really establish a theoretical use for Horace Walpole’s invented term, serendipity—coined to capture ‘discoveries made by accident and sagacity’ (see page xi this volume for the relevant passage). Merton’s ‘serendipity pattern’ (Merton 1948) describes what he considers a fairly common occasion in scientific practice, one of several ways in which events can affect theory fairly directly, when an ‘unexpected datum imposes itself’ upon an investigating scientist. Since then, various approaches to studying discovery in science have led theorists to conclude that the unexpected plays a key role, and that the strategies employed by scientists in response to the unexpected tell us something about scientific discovery in general (e.g., Dunbar and Fugelsang 2005, p.61). Again, most of these examples and theoretical approaches come from outside of philosophy. Part of the explanation for the relative lack of investigation into discovery (compared to the attention paid to explanation and justification) from within philosophy of science comes down to history: decades ago, a distinction was made between the context of discovery and the context of justification, with the latter thought of as being the proper subject of philosophical investigation, relegating the former to psychology, history and sociology. As a consequence, discovery-talk maintains a general vagueness about it, using concepts such as genius, chance, and serendipity, which seem to hold place for something ‘other than reason’ rather than offering meaningful alternative categories. That is, the moment of transition into discovery remains highly contingent and particular: mysterious and potentially ungraspable except in the very context in which it occurs (e.g., Okasha 2016, p.73–4). At the very least it is emergent and complex, and thus difficult if not impossible to make explicit, and easier to describe than to theorize. Yet we are drawn towards the ‘leap’ from ignorance to knowledge that discovery represents (see also Arfini, Chap. 7, this volume). Indeed, the distinction between discovery and justification has been increasingly maligned of late: we can hardly understand how theory comes about or how we might assess one theory over its rivals without a comprehensive picture of the process of science that includes discovery. In this chapter, I explore how this gap in our ability to explain what happens when discovery happens is part of why serendipity and accident have been so frequently left out of philosophical accounts of scientific reasoning, theory and practice. I argue that, in contrast, accident and chance, and especially serendipity are an important topic for philosophical study.

104

S. Copeland

Serendipity is especially relevant because it is more than mere luck—sagacity, a kind of wisdom accredited to the serendipitous (but not the merely lucky), separates it out as a different kind of thing, one that includes the contingencies of fortune but also the wise perception often attributed to a genius and marking skilled observers. Indeed, I hope to show that the inclusion of sagacity as one of its primary criteria makes serendipity a very useful, bridging category, suitable for helping us understand better how we make discoveries, and when and how we can be prompted to make the move from confirming what we already know into the realm of creation and innovation. Rather than a mystery, recent research points to sagacity and chance being interacting phenomena, an interaction we can not only come to understand but that we could purposefully develop and prepare for, and this in turn changes how we might think about the role that chance plays in scientific reasoning. Thus, in this chapter, I offer a review of key approaches to serendipity in science, from diverse disciplines and fields within them, with a focus on how the intersection of accidents or chance with sagacity or wisdom can provide philosophers with insight into the practice and progress of science and indeed provides a more comprehensive and correct picture of scientific progress than accounts which elide them could do. I propose, in turn, that integrating accident into our understanding of scientific practice is the right approach for philosophers of science—reason does not merely compensate for chance interruptions but rather takes them up into rational processes; if we take serendipity seriously, then we can see it as presenting a case study for the direct interaction of reason with chance.

The Context of Discovery Within philosophy of science, the decades of avoidance of the nature of discovery as a topic for study is often suggested to have been the result of the division between the contexts of discovery and of justification, suggested by Reichenbach but taken to heart by highly influential logical empiricists. Justification, according to Popper, was the proper subject of philosophy, the study of reasons and natural laws. “Every discovery”, on the other hand, “...contains ‘an irrational element’, or ‘a creative intuition’” (Popper 2005). Paraphrasing Einstein, Popper suggests thus that the path to discovering new laws through experience is not something we can rationally reconstruct, nor would we want to.4 And moreover: “If it is the processes involved in the stimulation and release of an inspiration which are to be reconstructed, then I should refuse to take it as the task of the logic of knowledge. Such processes are the concern of empirical psychology, but hardly of logic” (Popper 2005, p.8). In similar fashion should the investigation of serendipity, which according to this distinction 4

Even before Reichenbach, and as I discuss further on in the chapter, Whewell and others had suggested that the role of chance and genius in discovery meant that there could be no way to explicate the art of discovery (Schickore 2006; Silver 2015, p.249).

6 Serendipity and the History of the Philosophy of Science

105

may be seen as one of the least rational categories of discovery, be cast to the social sciences and out of philosophy. Brannigan (1981), in his review of the history following Popper’s distinction, points out that addressing the role of chance was a continuing problem for those seeking to pull discovery back under the cover of philosophy-worthy topics. For instance, an early critique by Hanson (1967, see p.322) of this contextual distinction and dismissal of discovery, notes the relationship between the non-rational nature of inspirational moments and concerns about the role of chance. Because chance frequently inspires discovery, that is, poor Popper assumed mistakenly that there can be no logic of discovery, in turn. For Hanson, however, the solution is fairly simple—reconsider the nature of discovery to separate chance out as an unimportant step. That is, how new ideas are generated is a reasonable thing to investigate, both for its importance to science but also because that reasonable part of the process can be differentiated from the ‘trigger’ or inspiration—not all of discovery is due to chance, and so the presence of chance ought not prevent us from investigating discovery at all. This represents one tactic for engaging discovery with philosophy and yet resisting the incorporation of chance and accident into accounts of scientific reasoning. Blackwell, who followed, gave a bit more room for chance within his analysis of scientific discovery: accidental discovery is one category of discovery, falling under what Blackwell labelled as ‘discovering that’, for those discoveries of things that exist, and how and what they do. In such cases, “the conceptual readiness for a puzzle solution is combined with the fortuitous circumstances of its realization” (as Brannigan paraphrases 1981, p.17). However, Blackwell still appeals to mystery in the guise of ‘genius’ in order to justify the seeming role of luck in enabling the right scientists to witness the right puzzles at the right time: “Certainly, it cannot be denied that Lady Luck has occasionally smiled upon the men of science by providing situations that may otherwise have been overlooked or never deliberately contrived” (quoted in Brannigan 1981, p.19/20). Thus, the separation of the accident itself from the discovery occurs here by moving the act of noticing into the realm of context— offering the conditions for noticing, rather than learned skills and practiced reason— and by keeping the mystery intact with vague concepts of the personal like ‘genius’, which serve somehow to translate those conditions into discovery. Kuhn’s concept of an anomaly, those interruptions and eventual disruptions of the problem-solving activities that mark periods of ‘normal science’, is perhaps the most significant move toward incorporation of chance into our understanding of discovery and the progress of science (Kuhn 1996). For Kuhn, an historian by training, accumulations of anomalies eventually lead to the introduction of a radically new framing for how to understand and explain the world—and, more specifically, the outcomes of our experiments and scientific work—that can incorporate them in a way that explains them (making them, that is, no longer anomalies). But, as Brannigan 1981, p.22) points out, Kuhn does not tell us what makes something become that kind of disruptive anomaly—how it passes from the threshold of dismissible mistake or

106

S. Copeland

outlier into the status of paradigm-shifting, undeterred fact.5 Rather, “Kuhn glosses this transition with an oblique reference to the ‘individual skill, wit or genius’ of the researcher” (Brannigan 1981, p.22, see also Celluci 2017, §11.3). The only difference between a noticed and an unnoticed anomaly, that is, is the noticing itself. When the scientific community is ready to witness and reason differently in response to an anomaly, then they will be able to do so: “as anomalies accumulate, a burgeoning sense of crisis envelops the scientific community” (Okasha 2016, p.76). Again, in such accounts a gap is allowed to remain, in the form of a threshold reached through this temporal process of accumulation that is pointed to but not itself described in detail. We are left without a clear account of what tips the balance at the level of the scientific community, nor of what leads individual scientists to reconsider errors as anomalies. The pattern I wish to attend to here is the tendency, even within proponents of logics of discovery, to presume that neither chance itself nor the initial response to it are possible topics for philosophical investigation in relation to scientific discovery. That is, attempts to rescue discovery as a topic for philosophical study have generally and nonetheless left the constituents of serendipity somewhat to the side, even while acknowledging the importance of chance and the unexpected in the progress of science. Chance, accident, genius, noticing and surprise, are concepts that, as used, leave a gap and an air of mystery in such accounts, ultimately failing to resolve the dichotomy of contexts they critique. To see what closer examination of serendipity can add, consider further this idea of an anomaly, a status given only when enough encounters with similar results creates a pattern that can be recognized as significant. That is, a paradigm shift does not result from a single individual observation or instance: Anomalies that shift paradigms in Kuhnian scientific revolutions do not act alone. Like with the statistical significance of adverse side effects presenting when a drug is under clinical trial, there is a threshold of importance: at a certain point, once so many such reactions have been observed, action in response to the unexpected result is demanded from the experimenters, the trial stopped or the drug recalled. And if we turn to empirical work done with practicing scientists, to observe their reasoning and responses to error in real time, we can see that iterative exchanges between colleagues generate a threshold of interest and the need for further investigation. Possible explanations for the anomaly as error or outlier are themselves interrogated and dismissed through deliberation, debate and experimentation (Dunbar and Fugelsang 2005). This aspect of discovery, that it can operate via threshold, is closely tied to accounts of serendipity 5

Here I attend to critiques of Kuhn in respect to the implications of the accidental in how we reason about anomalies toward discovery rather specifically. The reader can note that what Kuhn gets right, particularly in respect to my other work that focusses on the role of network interactions in scientific discovery and serendipity, is his emphasis on the nature of discovery as a process— thus, there is no definable ‘moment’ of transition in his theory to explain (see especially Kuhn 1996, p.55–57). As I point out here, however, this leaves the transition itself an empty notion, a retrospective explanation of something already observed rather than an observation of the thing itself. In the following paragraphs I turn to how the process dimension of discovery intersects with what we know of serendipity in science.

6 Serendipity and the History of the Philosophy of Science

107

as well, which highlight that while an individual experiences the surprise of the unexpected, potential value, an accident is not serendipitous science until it is accepted as a discovery (Copeland 2015; Kantorovich 1993). This multi-scalarity, acting at both the individual and community levels, in the moment and over time, is a feature of discovery that investigations into serendipity can elucidate. In sum, efforts to defer rather than resolve the issue merely push the borderline further to one side rather than showing us how to bridge the gap that the distinction between the contexts of discovery and justification represents. Seen through the lens of how chance is dealt with, the split between discovery and justification has been a matter of debate, but it represents something different than simply our ability to explicate rational processes (with this capacity marking the threshold between the two and the border of the realm of logic), but rather where we feel contingency or chance rather than reason and agency dictate the turn of events. I posit that philosophers have tended to shy away from analysing discovery before theory choice because chance and contingency are seen as things human reason must deal with or work around— as constraints or confounding variables, you could say—and philosophy concerns human reason itself. However, research on serendipity has shown that rational agency and chance are intertwined in a way that reveals how arbitrary or forced the division must be when we leave these aspects of discovery out of our accounts. The accounts of philosophers of science who call into question this division would benefit from even more radical moves than the shifts and shuffles of the boundary described above. Rather, I suggest that a suitable type of rational agency to elucidate what happens in scientific discovery would consider chance from within the account, an account that in turn must be much broader than those entertained so far in philosophy of science. I provide in what follows, therefore, a complementary reading of the role of accident and chance and, in turn, of the philosophy of discovery, through the lens of serendipity as a bridging concept. Entailing the integration of accident and sagacity, the use of this concept to describe discoveries marks our acknowledgement that some kind of skill or intelligence is related to the recognition of value in a chance or unexpected encounter. Chance does not have to be left out of our understanding of discovery, as a trigger or precursor, nor does our engagement with it need to remain at the descriptive level, and it is neither mysterious nor epiphemonenal in respect to the aspects of discovery thereby more worthy of study. To show this to be true, I will first put this ‘gap’ or dichotomy further into context, and then we will see what the ‘friends of discovery’ can offer to our understanding of serendipity in science. In the final section, I propose an epistemology of serendipity as a way to fill this gap.

Serendipity as a Bridge Nickles (2009, p.174) points out that leaving aside aspects of discovery—namely the movement from observation or encounter, to reason and knowledge—means assuming that when we hypothesise, we take (blind) risks in testing those hypotheses rather than having reasons for selecting them. Those who leave discovery out of

108

S. Copeland

the scope of science, that is, must concede that it does not matter where our ideas and solutions originate—and philosophy’s task is reduced to understanding how ideas are justified and tested, once they are thought. While more recent approaches to how scientific ideas are generated may push the boundary back to include the making of a connection between previously thought ideas to generate a novel idea (e.g., combinatorial approaches, like those of Thagard and Simonton), they leave the gap I discussed in the previous section virtually unaddressed. Silver (2015, p.236) argues that this gap, problematic ‘conceptual leap’, or what Duffin (2000) calls “the heuristic quantum leap of a discovery” (p.159), has a long history of complicating our conception of science. Indeed, this was what the turn to method was supposed to correct: so long as the key to science is not the moment of discovery but what we do with it to generate knowledge, then the right topic for study is method, no mystery. In his review of the history of philosophical approaches to discovery, Nickles suggests that, “For both Bacon and Descartes, method was the very antithesis of chance” (Nickles 2009, p.168). Method was meant to bring the practice of science into maturity, out of the state of lurching in the dark and hoping for discoveries to happen across, and into an enlightened application of clear reasoning for sorting out the world into “causes and deterministic laws” that could then be applied with success. You could have either chance or method, with no overlap, and with the latter being the ultimate aim.6 While it stems from an attempt to understand how to generate scientific progress through discovery, this dichotomy ultimately underlies the later distinction between discovery and justification discussed above. But the advent of method did not mean the end of chance, nor prevent serendipity from playing a role in discoveries nonetheless widely accepted by the scientific community; the focus on method did not prevent even philosophers from seeing science as a practice imbued with accident and chance (Trout 2016, p.391). As Silver (2015) points out, Bacon’s imposition of method as the way to do what we now consider science did not eliminate the gap between searching and finding that accidental discovery highlights. Indeed, Silver traces Walpole’s invention of the term back through Bacon’s use of the hunt as a metaphor, particularly in his telling of the Fable of Pan. Pan offers us a case of serendipity under this interpretation: Having refused to participate in the search for the goddess Ceres, Pan nonetheless happens upon her while pointedly on the hunt for deer instead. Echoed later in Walpole’s words for serendipity as finding something valuable while à la chasse of something quite different (see the passage from Walpole’s letter, page xi this volume), the hunt plays an important role in Bacon’s explanation of how science progresses. However, with Bacon as with those who came later, the accidental discoveries that mark early advances in the arts should be left there, and reasonable search be the province of science thenceforth: “‘The mechanical arts draw little light from philosophy,’ Bacon 6

As Nickles (2009) puts it: “Chance and luck are the very things that method traditionally is supposed to exclude…If luck is unavoidable in inquiry, if inquiry presupposes luck, then the classical discovery program is doomed from the start” (p.178). Or from Trout (2016): “Under an ideology that equates science with prediction and control, the role of luck or fortune seems incompatible with great discovery. We desire to find something out and, having gathered and evaluated the evidence, we have the feeling that the evidence is now under our cognitive control-that we understand” (p. 394).

6 Serendipity and the History of the Philosophy of Science

109

lamented, precisely because they take their lesson from accidents” (Silver 2015, p.247). The discoveries by accident that lead to the development of arts and technology differ from science, according to Bacon, because they produce ways of engaging with the world instead of knowledge about the world. We are unable, that is, to move from the particulars of an accident—a situation of contingency—to generalizable truths, insofar as accidents produce the wrong kinds of methods. The methods of art deal with specific contexts and interactions, not with logic or laws, according to this distinction…in much the same way that the distinction between discovery and justification separated out the contingencies of noticing from the knowledge gained through testing and validation. Silver hones in on this underlying problem of how particulars can lead to generalities as a fundamental problem in the philosophy of science, unresolved by Bacon. I suggest here that siphoning off the contingent and accidental aspects of discovery from the advancement in knowledge that discovery represents provokes a similar dilemma. What we learn through serendipity are not only new facts about the world, but about new ways to do science (or, new ways to engage with our world, through accidental success). Thus, serendipity offers a broader understanding of scientific discovery that can take in processes such as paradigm shifts without relying on straightforward induction. Understanding serendipity can help us to understand how we make the leap through discovery into the realm of method. Conceptualising general truths when one encounters a particular case, for instance, is an aspect of the sagacity involved in serendipity. As Silver notes, serendipity “names the way concepts emerge from the unexpected bumps and nudges of the material world, and it therefore isolates a critical tension in the method of the sciences” (2015, p.236). Merton’s description of the serendipity pattern as a common method for developing theory from empirical data, includes the ‘strategic’ application to theory of the ‘unexpected datum’ encountered (1948, p.507). As I have noted, however, the focus from a philosophical perspective has been on what happens after the ‘bump’, ‘nudge’, or datum ‘imposes itself’ upon the observer. That is, the focus has been mainly on what happens in the head, during or to signal a discovery, carrying over the division between contexts that Popper embraced—the psychological trigger of surprise to a novel combination, and then, after that, the discovery described as rational processes of search, abduction and justification. In the next section, I pull together philosophy of scientific discovery and serendipity research to show how we can understand sagacity in cases of serendipity as the kind of reasoning and perception we also find in the early stage instances of noticing, or intuition, for example.

From the Hunt to Heuristics Philosopher of science Whewell thought that there could be no explication of the art of discovery, because it began with a ‘happy thought’. Such thoughts were happy because of their value, and as prompts to discovery they were unplanned emergences

110

S. Copeland

from scientific activity (see Schickore 2006, p.61). They were also happy and fortunate because they occurred in the mind of someone ready to have them, and able to see their value; they arise as ‘colligations’ from the knowledge already present in the mind of a genius with the talent to perceive the value of a new connection. As discoveries, then, they are not really accidental and certainty not lucky. Thus does the role of sagacity mark serendipity in science different from mere luck. The requirement of what many call ‘the prepared mind’ after Pasteur’s famous comment highlights the fact that some kind of perception, wisdom, or expertise is involved, as necessary as chance, luck or accident to serendipity.7 But, as Dunbar and Fugelsang have noted: “It is interesting that, other than quoting Pasteur by stating that ‘Chance favours the prepared mind,’ scientists have not given many insights into what the prepared mind is; neither do they provide a clear account of what strategies they use when they encounter results that were not expected” (2005, p.61–2). The difficulty in understanding the prepared mind in relation to accidental discoveries or chance observations lies in the fact this can only really be done retrospectively. Taleb (2010), for example, describes unexpected events with negative impact as Black Swan events, which after the fact may be fully explainable in terms of cause and effect and thereby seem inevitable, although before the event no one could have predicted it would happen. Thus, the circumstances for which we must be prepared remain necessarily unknown; every instance of serendipity will be different. Given the particularistic nature of what counts as preparation, then, it is not surprising that the concept of the prepared mind itself remains quite vague.8 More recent work has recognised the role of chance in relation to the prepared mind in cognitive, representational terms. These accounts tend to focus on how accidents can trigger discoveries, separating out the discoveries themselves to be described in cognitive, representational terms, as novel combinations thereof. Harnad (2007, p.174), e.g., argues that while discoveries can be the result of ‘fortuitous’ combinations made by trial and error, these must also be recognized as valuable by an experienced mind. He calls his own account ‘cerebral’, focussing on combinations of mental representations, which constitute the creative acts of discovery he describes (p. 171–2). Thagard similarly develops a theory of discovery as the generation of novel combinations of cognitive, representational content (Thagard 2012). In his approach (e.g., Thagard 1998, 2002), serendipity is a kind of discovery, one that is triggered by surprise—but otherwise, not really different than those arrived at by search. Brannigan (1981) comments (p.5) that the positivist ideal, that scientists ought to apply logic and theory in their practice, resulted in a separation of the behavioural approaches to scientific practice from the philosophical approaches to scientific theory. He proposes instead to flip the understanding of discovery over to a purely 7

One could say that this adds only further layers of chance. That is, the idea not only has to occur unplanned, but in the right person and at the right time in their development of expertise and awareness. I simply sidestep this regress here; it does not lead anywhere, so I take the more fruitful route. 8 See Glaveanu (2022) for an alternative and contemporary example of improving focus on this from a creativity study perspective.

6 Serendipity and the History of the Philosophy of Science

111

social category (rather than being about individual experience), with his ‘attributive’ model—but these types of approach also fall short as explanations of what kind of discovery serendipity is (Simonton 2022). That is, purely social accounts fail to take up the multiscalar aspects of serendipity, which bridges the individual and the social. A similar issue is raised by the account offered by philosopher of science Kantorovich (1993). Serendipity earns pride of place in his evolutionary account of scientific progress, playing the role of random mutation—the driving force behind novelty and change in science. However, this can also be seen as another way of externalizing rather than intertwining chance with reason. That is, serendipity includes sagacity in Kantorovich’s account, but not intentionality or agency; chance opportunities become scientific discoveries by way of a naturalistic rather than a reasoned process—it can be described but not applied. The scientific community (as a whole) acts like a prepared mind, but only insofar as it selects the opportunities that arise by accident and chance for uptake into shared scientific knowledge, just as in classic evolutionary theory natural environments ‘select’ animals with mutations that better enable them to survive and reproduce. Thus, the interaction between individual reasons for following up on unexpected observations and the community’s validation of that effort by accepting it as a serendipitous discovery is left outside the realm of reason, a gap that must be crossed with a leap. However, serendipity seems to rather entail overlap between the phenomenological and theoretical aspects of discovery—cases of serendipity, as they are told, generally include a response of surprise, an ‘aha’, or even a ‘eureka’ in the moment. There is an emotional or affective valence to the shifting of expectations that occurs when serendipity is recognized and a new direction opens up by chance, which denotes a distinctly personal aspect to serendipitous events (Sauer and Copeland 2021).9 Further, consider that, despite an otherwise combinatorial account of creativity as ‘bisociation’—“the synthesis of a single idea with two apparently inconsistent contexts”—Koestler’s explanation of discovery also says that it is “characterized by a mix of elation and catharsis” (in Brannigan 1981, p.27–8). Bisociation is one of the more common terms used to describe how serendipity happens. For de Rond, for instance, serendipity is not an event so much as a capability for bisociation, for identifying useful ‘matching pairs’ of events (de Rond 2014). I suggest that serendipity offers either a dilemma or a bridge. It straddles the borderline between phenomenal experience and the empirical, perceptual encounter with the world, and the reasoned processing and evaluation of one’s observations, thoughts and encounters as potentially valuable and pursuitworthy. If regarded as a dilemma, the integration of chance means that we lose the ability to examine serendipitous discoveries holistically, and have to split it in two, as ‘accidents’ plus 9

Although I don’t have room to explore further here, it is notable that such ‘aha’ responses are also generated by stories of serendipity—listeners and readers experience joy at making the same realization of how fortunate an accident turned out to be, perhaps as much as those who perceived it first may have felt, which goes far in explaining why serendipity stories are such popular literature. The link to be followed up on this is with Erdelez’s comment that super-encounterers, those who experience intentional and surplus serendipity in their seeking activities, often also create serendipitous opportunities for others while they are at it (Erdelez 1999, p.26, and Chap. 12 in this volume).

112

S. Copeland

‘sagacity’. The prepared mind sets the stage for the rational uptake of the chance moment, but the accident in this kind of account amounts to an interference in a rational process rather than part of that process. This accords with Walpole’s and later proponents’ conception of serendipity as occurring only while we are engaged already in search or other such activity when the accident occurs (Chumaceiro 1999). But Walpole also calls it ‘accidental sagacity’, suggesting that the two elements are conjoined under one conception, rather than separable and distinctly conceived.10 In Merton and Barber (2004), we find the idea of sagacity traced back through Samuel Johnson to John Locke, who suggests that the sagacious are wise in the sense of being able to perceive the links between ideas; leaning on perception in this way, according to Merton and Barber, lends the use of sagacity the sense of being “a psychosocial cognitive mechanism” (Merton and Barber 2004, p.261 Footnote 38). Thus, serendipity indeed is a kind of trigger, but one that simultaneously reveals a link, perceived because and only once chance draws one’s attention toward it. As an act of perception, accidental sagacity does not seem separable from the recognition of potential value, and so chance, attention and recognition (or ‘noticing’) work together rather than one following the other. According to Bacon, the wisdom of Pan on the hunt is a combination of “sagacious experience and general knowledge of nature” (quoted in Silver 2015, p.243).11 Sagacious experience, if sagacity is a matter of noticing links that others (would) have missed, in this case might entail a perception of the connections within nature. A good hunter sees tracks, signs of movement, and comes to understand the environment she moves within as a set of patterns (waterholes for drinking from, trails to and from burrow to food…) that can lead her reliably to her prey. Making cognitive connections between the observations one makes is insufficient for discovery however; not just any connection will do, and mentalistic accounts will have a difficult time distinguishing between the processes that generate discovery and those that do not (hence the tendency to rely on additional factors, such as ‘genius’ to fill the explanatory gap). The metaphor of the hunt draws our attention to the fact that when these perceptions are insights into the workings of the world, then our engagement with that world is what tells us when our connections are well made; if the pattern I perceive isn’t really there, I will find neither deer nor goddess on my hunt. In addition, arguments from creativity theory as well as within serendipity research have connected the emergence of serendipities with what is often called an ‘incubation’ period (e.g., process models of serendipity by Makri and Blandford 2012; McCay-Peet and Toms 2017; McCay-Peet and Wells 2016, and specifically on this topic, Gilhooly 2022).12 As Scheffer et al. (2015) carefully note, an important part 10

This is noted as well by authors March and Vallee-Tourangeau in their contribution to the Art of Serendipity (eds. Ross and Copeland, see references for Simonton, Gilhooly and Glaveneau 2022, at the end of this chapter), but besides this noted point about inseparability the approach I take here in respect to ‘accidental sagacity’ differs from theirs. 11 Silver further offers evidence that this idea of sagacity would indeed be the one picked up by Walpole later, when he wrote the famous letter to Mann (Silver 2015, p.244). 12 Note that I cite here serendipity scholars, mostly, and that this represents only an echo and not a reflection of the work done in psychology, for instance, on the nature of incubation.

6 Serendipity and the History of the Philosophy of Science

113

of science includes ‘priming’ for discoveries to be made in this way: allowing time for associations between ideas and memories to occur (or, in mind wandering, to collide with seemingly random thoughts). But, as with other cognitive models, this approach keeps our opportunities for discovery very much within the confines of the prepared mind. To combat this, Scheffer et al. suggest diversifying the information we encounter, including drawing from the arts for “finding interesting input for our hypothesis testing machinery” (Scheffer et al. 2015, p.3). Thus, despite the integration of intuitive and creative thinking into accounts of scientific reasoning, there remains a tendency to still relegate chance and accident to external triggers of the reasoning processes involved in ‘actual’ discovery. Nickles, however, offers a promising account of heuristic appraisal or ‘HA’. This is the form of (often unconscious) reasoning which “evaluates the promise or potential fertility and feasibility” of something (Nickles 2006, p.159). In relation to cases of serendipity, I have previously called this the perception of “potential value” (Copeland 2019)—this is value that is not yet realized, and so it therefore really represents an assessment of the fecundity of the chance observation at hand. As a method, “HA has one foot in context of discovery and one foot in context of justification yet belong to neither as defined by the standard [positivist] distinction” (Nickles 2006, p.160). Indeed, Nickles takes up heuristic appraisal as the key form of reasoning in science partly because, “In [his] view, all innovation involves at least small elements of luck or serendipity, a view that makes HA even more obviously indispensable—and more difficult” (2006, p.178 Footnote 9). The very practical and experiential nature of HA draws out the role of our tacit relationships with our environment and each other in such reasoning processes: for example, we incorporate considerations of how to most economically use our time and resources, implicitly, when we decide whether to follow up on something that catches our eye or to keep heading in the direction we are already going (e.g., Barber and Fox 1958). This overlap of appraisal with action, I suggest, captures more closely the nature of sagacity in cases of serendipitous discovery, setting us in the right direction.

An Epistemology of Serendipitous Discovery What is particularly attractive in an evolutionary account such as the one proposed by Kantorovich above, or Campbell whom he follows, is that it seems to incorporate chance into the trajectory of scientific process (as random mutation, and thus still independently and externally from the context that ultimately gives it value) without giving up reasoning processes that prevent an equivalence between error and discovery (e.g., the ‘selection’ of good ideas out from the bad). That is, we can generalize our explanation of chance (Nickles 2009, p.182)—but, as I argued above, it offers little in the way of understanding sagacity better. Serendipitous discoveries can be thought of as those moments in science when we make theory fairly directly out of the clay given to us by the world (as Merton suggests), when we move from the particular and contingent to the general and true (as Silver puts it). Discovery

114

S. Copeland

by chance may be contingent and random from one perspective, but causal explanations are available from another (as with Taleb’s Black Swan). Indeed, as Nickles notes, in a biological system it is precisely the noise in the system that provides the means for progress and change—innovation—and discoveries require what Max Dulbruck called the ‘principle of measured sloppiness’ (Nickles 2009, p.185). This is in keeping with other accounts of serendipitous discoverers: Fleming, for instance is often commended for leaving his petri dishes unwashed, with the conviction that if he had kept a tidier bench, penicillin would have remained undiscovered (Copeland 2018). Thus, personality attributes that lead to an increase in overall noise and/or perceptive abilities tend to be used to explain ‘sagacity’, focussing again on chance (being the ‘right kind of person’) rather than providing an explanation of the reasoning processes that may lead up to and include a recognition of potential value. Recent proposals for how to increase serendipity in a variety of contexts illustrate this belief that increasing the number of chance connections made is the key, so that they may more likely befall the ‘right person’ to recognize their value. For instance, building designs that promote random interactions between people within them are exemplified by Google’s New York campus, a “workplace designed to encourage ‘collisions’”,13 or the Pixar building designed similarly by Steve Jobs, where “you can’t avoid running into people”.14 Increasing ‘collisions’ between people at the water cooler or elsewhere, that is, is commonly conceived as sufficient for increasing serendipity in an institution (see Busch & Grimes, Chap. 5, this volume, for more examples). There are two things I would like to note about this approach: First, that it illustrates the ubiquity of the role of chance in discovery and progress. Innovation is known to arise by chance and accident on a regular basis, and the growing desire to somehow ‘engineer serendipity’ highlights the wide recognition of this fact. Second, more than increasing the number of random encounters that happen is needed for serendipity to follow from such measures. The second point calls for further elaboration. Sagacity is required in addition to accidents for serendipity. But, as I have pointed to above, the strategy is frequently to internalize sagacity and externalise chance in order to understand the former as a rational process, which leads to an individualised model of the discoverer as having innate characteristics, such as genius or having the right personality. Other traits assigned to the sagacious in serendipity have included being a maverick or being a novice, and hence susceptible to new ideas in a way that trained experts full of expectations are not (e.g., Meyers 2007). In others, persistence in the face of opposition and uncertainty denote their special talent for taking an unexpected observation to the level of scientific discovery, as with Barry Marshall, commended (in hindsight, 13

From an online article accessed in May 2022: https://www.anitainsights.com/blog/building-forserendipity-at-google-nyc/, published for a blog on Workspace Innovation, November 7, 2016 by Time.com, “Building for Serendipity at Google NYC”. 14 From a Forbes online blog by Robert C. Wolcott, published January 22, 2021 and accessed May 2022: https://www.forbes.com/sites/robertwolcott/2021/01/22/vr-storytelling--ser endipity-pixar-co-founder-ed-catmull-and-composer-harold-oneal/.

6 Serendipity and the History of the Philosophy of Science

115

of course) for his insistence to the point of self-experimentation15 against prevailing wisdom that bacteria could in fact live in the acidic environment of our stomach, as he had observed (e.g., Dreger 1999). These talents, however, seem quite separable as such from the scientific skills of the main characters in the story; like the accident or surprising observation that started it all, the success of the discovery in these accounts relies on the ‘right person’ having made the observation, with the success criteria for this being personality and psychological traits, or the generic and vague ‘prepared mind’, rather than developed and refined skills of scientific practice.16 Again, we see serendipity ‘outsided’ from scientific activities, influencing by interfering with, rather than seen as an integral part of discovery processes. An epistemology of serendipitous discovery, however, would address the interaction between wisdom and chance directly instead of side-stepping it. One way to do this is to assume that much or at least a significant part of scientific progress is simply determined by contingency and chance (Trout 2016). Evolutionary approaches adopt this assumption, but also social scientists and historians of science have addressed the role of contingency by looking to examples to say more about how discoveries do happen, than how they ought to happen. For instance, Holton et al. (1996) write about how a discovery ‘is made’ through the case study of high-temperature superconductivity: “Our findings emphasize the great importance for scientific research of unintended interactions or applications” (p.373). That is, when an observation is made or results obtained, their eventual use is not yet obvious—often, the value of scientific knowledge emerges from interactions and ‘borrowing’, allowing it to be developed in a context and for a purpose quite different than originally intended (Holton et al. 1996). For Kantorovich (1993), this represents the process of selection by the scientific community; as knowledge is exchanged, it is selected for or against by other scientists and either dropped, or shaped into a discovery and accepted as new knowledge through such exchanges. But this diffuses sagacity across the collective ‘scientific mind’, thereby eliding the role that individual scientists do play when they agree or disagree about the (potential) value of a (potential) discovery.17 And while the original intentions in producing results that were later found to be relevant to super-conductivity did not give hints as to that ultimate value, the nature of the ‘borrowing’ that Holton et al. (1996) describe, despite resulting often from chance encounters with other scientists and their results, suggests that there was some intentionality involved—this process was not entirely due to chance without thought, nor did it lay outside of the range of scientific skills that the borrowers had to hand. 15

And later, critiqued for this particular part of the story. Notably, Thagard (1998) offers in two companion articles a far more thorough history for the perception of Marshall as a maverick, illustrating the professional relationships and use of scientific skill that went into making his observation a renowned discovery. As I note elsewhere in this chapter, the observation itself remains a trigger in Thagard’s account, but the role of contingency and complexity in context is captured with care. 17 In (Copeland 2018) I mention that this also diverts attention to how responsibility as well as credit ought to be assigned to those who influence that scientific mind, which is problematic. 16

116

S. Copeland

Indeed, borrowing the knowledge of others to put it to new use is, one might argue, a fundamental practice for all sciences and a basic scientific skill.18 In accounts such as Kantorovich’s, where chance and accident is emphasised as the necessary trigger to initiating the processes of selection and, thus, the root of scientific progress, the ideas expressed about science and sagacity raise deep questions about the merit-based system we employ to ensure science continues to progress. That is, if success is due mainly to chance, then discoveries are made by luck—in turn, systems that reward scientists with funding and accolades ought to take this role of luck more seriously (see e.g., Sand and Chiapperino, Chap. 10, this volume). Others, in reflection on the prominence of chance, have gone so far as to suggest that we ought to abandon our claims to reward only or primarily merit altogether, and acknowledge the influence of luck by adopting lottery systems for scientific funding, for instance (Gillies 2015; Sand and Copeland 2020). From Holton et al. (1996) we get the suggestion that, “the chance of serendipitous encounters with key ideas is increased by permitting research to proceed at an unforced pace” (p.375). They found it striking that many of the interactions that led to the discoveries they examined occurred because scientists were working in an environment that allowed for ‘projects with long gestation periods’—demonstrating a “willingness to give good people the freedom” to follow new directions that may not lead to immediate rewards. Ironically, perhaps, this kind of freedom is often rewarded to those who already have a long track record of success, possibly even reinforcing the arguments of proponents for funding lotteries who claim that it is privilege and not merit being most often rewarded.19 However, I would argue rather that the fecundity of research without pressure to produce value quickly shines light upon an underlying reason for why ‘freedom’ is so valued by proponents of serendipity in science; it takes time for chance and accidental events to develop into a discovery (Solomon 2016). The role of incubation was mentioned already, but the importance of interactions—social, cognitive and physical, as well as through the tacit or explicit communication of ideas—to serendipity offers an additional way to understand how long it might take for contingent circumstances to arise and evolve, through such interactions, to allow serendipity to emerge. Emergence is a key concept for understanding serendipity at all levels (Copeland 2019). Key to understanding emergence is the idea of unpredictability: An emergent phenomenon is (more or less) one that could not have been predicted, given what we do or could know about the conditions that ultimately caused it. Pickering, for instance, gives an account of science in terms of the ‘mangle’ of complexity we engage through practice, and he describes the experience of emergence as our experience of ‘brute chance’ while we are so engaged: 18

In fact, in their final point Holton et al. (1996) say that any given discovery ultimately requires a breadth of disciplines involved in the exchange of knowledge, and that we cannot neatly separate between pure and applied science, or draw a sharp distinction between curiosity-and mission-driven research (we should rather support a ‘seamless web of research’ for progress—see esp. p.375). 19 Merton noted something similar with his concept of the Matthew Effect—notoriety can also lead to eminent scientists getting more credit for their discoveries than novices do for theirs.

6 Serendipity and the History of the Philosophy of Science

117

I have emphasized that brute contingency, sheer chance emerging in time, is integral to practice—in the tentative fixing of goals, in the emergence of specific resistances, in the substance of particular accommodations and their success or failures. (Pickering 1995, p.209)

For Pickering, the production of scientific knowledge is best described as an encounter; like in Merton’s serendipity pattern, the world imposes itself upon the scientist, for whom the capacities and nature of that world emerge (p.18).20 From this perspective, too much planning can interfere with our ability to perceive such emergences (e.g., see Bush 1945). To induce, or rather, to rush science is to preclude such chance discovery (see also the discussion around superconductivity following Holdon et al., above).21 Picking up on Pickering’s emphasis on the encounter, one way to address the dichotomy of the accidents—and sagacity—account of serendipity that we are led into when we offer an overly cognitive account of sagacity, is to resist it. As others have said before, it is paradoxical to bring an element of control into serendipity, contradicting its definitional dependence on chance (e.g., Nickles 2009, p.179). But we can work from the other direction, bringing our concept of sagacity in line with other rationality-based approaches to how we actively engage chance, in science and in our everyday lives. That is, I think we can in fact take up the assumption that a significant portion of scientific progress is due to accidental encounters without letting go of the individual sagacity or skills that must play a role in discovery and progress. For this, I propose we widen the understanding of sagacity beyond the cognitive and individual, and yet resist dispersing it across the whole of science or the contingent world—that is reducing it to neither logic nor luck, but rather trying to understand it as a kind of reasoning of its own, suitable for situations where the two meet. The pragmatist Peirce most famously conceptualised this kind of reasoning as abduction, an approach taken up in philosophy of science and serendipity research alike to explain the rationality of discovery. For Peirce, abduction is closely tied to surprise. He offers the following formula: “The surprising fact, C, is observed; But if A were true, C would be a matter of course; Hence, there is reason to suspect that A is true.” (Peirce 1903). Again, we see the surprising fact playing the role of trigger to thought, or to inference about the best or at least a likely explanation for that fact. We see the dichotomy arise in many uses of abduction in relation to serendipity—for instance, van Andel writes, “serendipity is an apt description for the observation of 20

Note that Pickering does differ from Merton in a fundamental way; while Merton’s narrative implies a realism about the world, which will thereby reveal its nature to us, Pickering adopts a more contemporary approach to the interaction between constructivism and realism in scientific practice. There isn’t room to go into this here, but suffice to say that the implications for realism in science of different approaches to serendipity is another topic that could do with a closer look by philosophers of science. 21 Also along these lines, Nickles (2009) suggests that, next to trial and error, serendipity may be the most economical of scientific methods: “…check instead for any coherent or sufficiently interesting manuscript, whether previously anticipated or not. Prespecifying a goal and rigidly sticking to that research plan not only reduces the probability of hitting something interesting but also limits the innovation to what we currently think we know or can plausibly imagine” (p.191).

118

S. Copeland

a surprising fact followed by a correct ‘abduction’” (1993, p.692). Rather than both describing the same phenomenon, surprise and sagacity in these accounts arrive one after the other. But contemporary scholars have thoroughly considered what abduction is meant to capture, and offer additional considerations.22 For instance, consider computational philosophers such as Magnani (2006), who uses ‘chance discovery’ as a frame to investigate the role of what he calls manipulative abduction in scientific discovery. More than an inference or creative invention of possible hypotheses (what he calls ‘theoretical abduction’, which lies entirely within the cognitive realm), there are “many cases of explanations occurring in science when the exploitation of environment is crucial” (p.1747; see also Magnani et al. 2016). In cases of manipulative abduction, “action can provide otherwise unavailable information that enables the agent to solve problems by starting and by performing a suitable abductive process of generation or selection of hypotheses” (p.1748). Importantly, in this account, abduction is part of the observation process, insofar as the action is a performance of both encountering the (surprising) fact and generating a hypothesis about it; by both ‘starting’ and ‘performing’ the abductive process itself. In other words, Magnani describes manipulative abduction as ‘thinking through doing’ (2006, p.1751). Similarly, where chance and wisdom, or accident and sagacity come together, there is also an intersection of cognition and environment (Arfini et al. 2020). Grinnell, philosopher and scientist, describes Peircean abduction as he sees it happening in scientific practice as “a surprising observation that becomes reconfigured as an unintended experiment about an entirely new research problem” (2019, p.225; see also Grinnell 2009). The phenomenology of this experience is like having one’s attention ‘abducted’ from the original problem and set upon a new one in a response to surprise (Grinnell 2019, p.224–5). In such cases, Grinnell suggests that the surprise effects a ‘gestalt switch’, where the scientist sees the situation at hand in two different ways: in respect to the original problem, as a problem or anomaly; in respect to the abduction or novel explanation for the surprising observation, as a completely new problem worthy of investigation (Grinnell 2019, p.221). In this way, one sees the unexpected results from the perspective of original intentions instead as an expected result, given the new explanation for what actually resulted from one’s experimental set-up. Rather than following the surprise, as van Andel suggests in his explanation of serendipity, the abduction in this case constitutes the surprise in the form of a gestalt switch in perspectives on what has occurred and been observed. In an analogous fashion, some theorists have used affordances to explain how serendipity and discovery occurs at the site of interaction between reason and experience. Björneborn describes serendipity as an affordance, “i.e., as a three-way relationship between an environment, a human being, and a potential activity” (Bogers and Björneborn 2013, p.207). Following Gibson,23 affordances are like enacted clues. 22

As Grinnelll (2019) notes, this discussion has much more nuance than the usual use of abduction in common parlance, which tends to mean simply ‘inference to the best explanation’. 23 To quote Gibson (1979, p.127): “The affordances of the environment are what it affords the animal, what it provides or furnishes, either for good or ill. The verb to afford is found in the

6 Serendipity and the History of the Philosophy of Science

119

The way in which an object or environment is perceived, that is, conjoins with the potential actions or behaviours that it allows. A chair, for instance, is perceived as ‘something to sit upon’ by someone who has an interest in sitting (as well as the right kind of body to accomplish ‘sitting upon a chair’). The interest in sitting and the perception of the chair are inseparable in this account; the chair affords sitting, and that is how it is perceived by me. A cat, for instance and in contrast, may perceive a landing pad that affords jumping up onto the counter. Insomuch as it is an affordance, serendipity affords the perception of an accident, for instance, as potentially valuable, or as an opportunity. Rietveld and colleagues (e.g., Bruineberg et al. 2021; Kiverstein and Rietveld 2018; Yakhlef and Rietveld 2019) have recently furthered the approach via affordances in relation to innovation by developing a theory of “innovative action as skilled affordance-responsiveness” that also looks at how individuals respond via both learned and novel behaviours to unexpected and ‘unconventional’ affordances given to them by their environment.24 They describe this in terms of ‘skilful coping’, wherein the experiential background and skills possessed by individuals couple with the environment through affordance perception. When the environment offers unexpected or novel affordances, the specific skills and thus expertise of acting individuals may enable their perception of those affordances. Given the novelty of the affordance situation, such individuals may then respond to those perceptions by acting in innovative ways within their environment; where atypical affordances are perceived, then typical behavioural responses do not automatically follow. This approach thereby avoids the temporal and physical dichotomies, such as observation as followed by evaluation, as discovery followed by justification, or surprise as a trigger followed by cognition, and offers instead a way to examine philosophically the intersection of perception and reasoning (like surprise and abduction) through examining the nature of rational, human responses to the unexpected.25 These avenues for investigation all suggest that there is room in philosophy of science for looking closely at serendipity, but also and importantly for seeing it as a bridging concept or a concept that captures the intersection of what are normally considered separately, and that will give us real insight into the nature of discovery. In my current work, I have been examining the potential of a type of reasoning from the categories of reasoning of ancient Greece, metis, as a frame for understanding dictionary, but the noun affordance is not. I have made it up. I mean by it something that refers to both the environment and animal in a way that no existing term does. It implies complementarity of the animal and the environment”. 24 In the 2019 article cited, the authors’ focus is a bit more on the other direction: how to cultivate novel affordances in otherwise behaviourally deterministic environments. 25 Also notable is the correlation between the way Yakhlef and Rietveld (2019) describe the rise of affordances as ‘soliciting and prompting action from the body’ (p.5) and Merton’s description of the serendipity pattern (1948), wherein the surprising datum imposes itself upon the investigator, whose perception of it is strategic, that is intimately tied to its potential value in relation to theory. The dual nature of ‘strategic’ in Merton’s account, that is, offers a similar bridging of reason and perception as these theories, along with the sense that the world directly affects the capacity of an individual to perceive and reason about her environment (e.g., the skills related to her affordance-responsiveness in that particular situation).

120

S. Copeland

how embodied and situational engagement with the world can be explained in philosophical terms as a form of rationality. Metis is particularly interesting because of the way it has been left out of Western philosophical discussions of rationality that follow the Greeks but focus only on episteme, techne and phronesis; metis is left to the side of these discussions much as serendipity is elided or avoided in philosophical approaches to science. But metis offers a way to capture the reasonable response to a changing landscape of affordances in philosophical terms: ‘metis’ is the cunning wisdom of agonistic encounters as well as sophistry, and democracy as well as the defensive strategies we employ in a changing environment (Copeland 2022). Sagacity, when part of serendipity, highlights the relationship between innovation and discovery not only in these dynamic contexts but also within scientific practice, itself a dynamic way of engaging with the world through action.

Concluding Comments The epistemology of serendipity I have sketched above is more a set of new directions to take philosophy of science than it is a detailed examination of how serendipity produces knowledge. In looking closely at the nature of serendipity as well as its rhetorical significance in the stories we tell about science, as a term that points to the role of wisdom in even accidental discovery, I have found that rather than a mysterious gap between chance observations and the production of knowledge, serendipity offers philosophy of science a bridging concept. Philosophy of science, rather than avoiding or deferring investigation of what happens when that gap is crossed—in the mind of the discoverer as well as in the world so discovered— has missed opportunities to understand how discoveries are made, more generally speaking. Contemporary philosophers, social scientists and science and technology studies scholars have taken up the cause, portraying scientific progress as partially or mostly the result of contingencies and chance (see Strevens 2020, for but one example) but they are only beginning to explore how rationality and chance truly intertwine in both ‘aha’ moments and in lengthy processes of scientific discovery. For this reason, serendipity science, I suggest, still has much to offer philosophy of science toward the clarification of what sagacity consists in, and how it enables us to recognise potential value in the unexpected.

References Arfini, S., T. Bertolotti, and L. Magnani. 2020. The antinomies of serendipity how to cognitively frame serendipity for scientific discoveries. Topoi 39 (4): 939–948. https://doi.org/10.1007/s11 245-018-9571-3. Barber, B., and R. Fox. 1958. The case of the floppy-eared rabbits: An instance of serendipity gained and serendipity lost. American Journal of Sociology 64 (2): 128–136.

6 Serendipity and the History of the Philosophy of Science

121

Bogers, T., and L. Björneborn. 2013. Micro-serendipity: Meaningful coincidences in everyday life shared on twitter. Iconference 13: 196–208. https://doi.org/10.9776/13175. Brannigan, A. 1981. The Social Basis of Scientific Discoveries. Cambridge University Press. Bruineberg, J., L. Seifert, E. Rietveld, and J. Kiverstein. 2021. Metastable attunement and real-life skilled behavior. Synthese 199 (5–6): 12819–12842. https://doi.org/10.1007/s11229-021-033 55-6. Bush, V. 1945. Science: The endless frontier. Transactions of the Kansas Academy of Science 48 (3): 231–264. Cannon, W.B. 1940. The Role of Chance in Discovery. The Scientific Monthly 50 (3): 204–209. Celluci, C. 2017. Rethinking knowledge: The heuristic view. In European Studies in Philosophy of Science (Vol. 4). Springer International. https://doi.org/10.3917/rac.035.0102. Chumaceiro, C.L.D. 1999. Sortes Walpolianae: Discoveries “almost” like serendipity. Creativity Research Journal 12 (4): 339–341. https://doi.org/10.1207/s15326934crj1204. Copeland, S. 2019. On serendipity in science: discovery at the intersection of chance and wisdom. Synthese 196 (6). https://doi.org/10.1007/s11229-017-1544-3 Copeland, S.M. 2018. “Fleming leapt on the unusual like a weasel on a vole”: Challenging the paradigms of discovery in science. Perspectives on Science 26(6). https://doi.org/10.1162/posc_ a_00294. Copeland, Samantha. 2015. The Case of the Triggered Memory: Serendipitous Discovery and the Ethics of Clinical Research [Dalhousie University]. http://hdl.handle.net/10222/63140. Copeland, Samantha. 2022. Metis and the Art of Serendipity. In The Art of Serendipity: Springer International Publishing. https://doi.org/10.1007/978-3-030-84478-3. de Rond, M. 2014. The structure of serendipity. Culture and Organization 20 (5): 342–358. https:/ /doi.org/10.1080/14759551.2014.967451. Dreger, A. 1999. A call for passion in the realm of discovery. New York Times. https://www.nyt imes.com/1999/01/12/science/essay-a-call-for-passion-in-the-realm-of-discovery.html. Duffin, J. 2000. Poisoning the spindle: Serendipity and discovery of the anti-tumor properties of the Vinca alkaloids. Canadian Bulletin of Medical History = Bulletin Canadien d’histoire de La Médecine 17 (1–2), 155–192. https://doi.org/10.3138/cbmh.17.1.155 Dunbar, K.N., and J.A. Fugelsang. 2005. Causal thinking in science: How scientists and students interpret the unexpected. In Scientific and Technological Thinking, ed. M.E. Gorman, R.D. Tweney, D.C. Gooding, and A.P. Kincannon, 57–80. Lawrence Erlbaum Associates. Erdelez, S. 1999. Information Encountering: It’s More Than Just Bumping into Information, 25–29. February/M: Bulleting of the American Society for Information Science. Gibson, J.J. 1979. The Ecological Approach to Visual Perception. Houghton. Gilhooly, K. 2022. Problem solving, incubation and serendipity. In The Art of Serendipity, ed. W. Ross and S. Copeland, 267–292. Palgrave Macmillan. Gillies, D. 2015. Serendipity and chance in scientific discovery: Policy implications for global society. In The Handbook of Global Science, Technology, and Innovation, ed. D. Archibugi and A. Filipetti, 525–539. WILEY Blackwell. Glaveanu, V.P. 2022. What’s “inside” the prepared mind? Not things, but relations. In The Art of Serendipity, ed. W. Ross and S. Copeland, 23–40. Palgrave Macmillan. Grinnell, F. 2009. Everyday Practice of Science: Where Intuition and Passion Meet Objectivity and Logic. Oxford University Press. Grinnell, F. 2019. Abduction in the everyday practice of science: The logic of unintended experiments. Transactions of the Charles S Peirce Society 55 (3): 215–227. https://doi.org/10.2979/ trancharpeirsoc.55.3.01. Hanson, N.R. 1967. An Anatomy of Discovery. Journal of Philosophy 64 (11): 321–352. Harnad, S. 2007. Creativity. Method or magic? Consciousness and Cognition 20: 127–137. https:/ /doi.org/10.1016/B978-012373734-2/50010-8. Holton, G., H. Chang, and E. Jurkowitz. 1996. How a scientific discovery is made: A case history. American Scientist 84 (4): 364–375. Kantorovich, A. 1993. Scientific Discovery: Logic and Tinkering. SUNY Press.

122

S. Copeland

Kiverstein, J.D., and E. Rietveld. 2018. Reconceiving representation-hungry cognition: An ecological-enactive proposal. Adaptive Behavior 26 (4): 147–163. https://doi.org/10.1177/105 9712318772778. Kuhn, T. 1996. The Structure of Scientific Revolutions (4th ed.). The University of Chicago Press. https://doi.org/10.5840/philstudies196413082. Magnani, L. 2006. Prefiguring ethical chances: The role of moral mediators. In Chance Discoveries in Real World Decision Making, ed. Y. Ohsawa and S. Tsumoto, 205–229. Studies in Computational Intelligence: Springer. Magnani, L., S. Arfini, and T. Bertolotti. 2016. An argument for ignorance-based chance discovery. International Journal of Advanced Intelligence Paradigms 8 (3): 327–342. https://doi.org/10. 1504/IJAIP.2016.077500. Magnani, L., Nersessian, N. J., and P. Thagard. (eds.). 1999. Model-Based Reasoning in Scientific Discovery. Springer Science. Makri, S., and A. Blandford. 2012. Coming across information serendipitously—Part 1: A process model. Journal of Documentation 68 (5): 684–705. https://doi.org/10.1108/002204112112 56030. McAllister, J.W. 2016. Rhetoric of effortlessness in science. Perspectives on Science 24 (2): 396– 415. https://doi.org/10.1162/POSC_a_00213. McCay-Peet, L., and E.G. Toms. 2017. Researching Serendipity in Digital Information Environments. Morgan & Claypool. McCay-Peet, L., and P.G. Wells. 2016. Serendipity in the sciences: Exploring the boundaries. Proceedings of the Nova Scotian Institute of Science 49 (1): 97–116. Merton, R.K. 1948. The bearing of empirical research upon the development of social theory. American Sociological Review 13 (5): 505–515. Merton, R.K. and E. Barber. 2004. The Travels and Adventures of Serendipity: A study in sociological semantics and the sociology of science. Princeton University Publishing. Meyers, M.A. 2007. Happy Accidents: Serendipity in Modern Medical breakthroughs. Arcade Publishing. Nickles, T. 2006. Heuristic appraisal: Context of discovery or justification? In Revisiting Discovery and JUSTIFICATION, ed. J. Schickore and F. Steinle, 159–182. Springer. Nickles, T. 2009. The strange story of scientific method. In Models of Discovery and Creativity, ed. J. Meheus and T. Nickles, 167–208. Springer Science. Okasha, S. 2016. Philosophy of Science: A Very Short Introduction. Oxford University Press. Peirce, C.S. 1903. Harvard Lectures on Pragmatism, Collected Papers, vol. 5. Harvard University Press. Pickering, A. 1995. The Mangle of Practice: Time, Agency and Science. University of Chicago Press. Popper, K. 2005. The Logic of Scientific Discovery (1935/1959). Taylor & Francis. Rescher, N. 1995. Luck: The Brilliant Randomness of Everyday Life. Univeristy of PIttsburgh Press. Roberts, R. 1989. Serendipity—Accidental Discoveries in Science. John Wiley & Sons, Ltd. Sand, M., and S. Copeland. 2020. Luck as a challenge for the responsible governance of science and technology. Journal of Responsible Innovation 7 (Special Issue), S1–S11. https://doi.org/ 10.1080/23299460.2020.1848848. Sauer, S., and S. Copeland. 2021. Wa/ondering with data—or, Responsibly measuring sociotechnical serendipity in the urban environment. In Proceedings of the IEEE International Smart Cities Conference (ISC2). Scheffer, M., J. Bascompte, T.K. Bjordam, S.R. Carpenter, L.B. Clarke, C. Folke, P. Marquet, N. Mazzeo, M. Meerhoff, O. Sala, and F.R. Westley. 2015. Dual thinking for scientists. Ecology and Society 20 (2): 1–4. https://doi.org/10.5751/ES-07434-200203. Schickore, J. 2006. A forerunner?—Perhaps, but not to the context distinction. William Whewell’s Germano-Cantabrigian history of the fundamental ideas. In Revisiting Discovery and Justification, eds. Schickore, J., and F. Steinle, 57–78. Springer.

6 Serendipity and the History of the Philosophy of Science

123

Silver, S. 2015. The prehistory of serendipity, from Bacon to Walpole. Isis 106 (2): 235–256. https:/ /doi.org/10.1086/681977. Simonton, D.K. 2022. Serendipity and creativity in the arts and sciences: A combinatorial analysis. In The Art of Serendipity, ed. W. Ross and S. Copeland, 293–320. Palgrave Macmillan. Solomon, Y. 2016. Temporal aspects of info-serendipity. Temporalités: Revue de Sciences Sociales et Humaines 24, 1–17. http://temporalites.revues.org/3523. Strevens, M. 2020. The Knowledge Machine: How Irrationality Created Modern Science. Liverlight Publishing. Taleb, N.N. 2010. The Black Swan: The Impact of the Highly Improbable. Random House Trade Paperbacks. Thagard, P. 1998. Ulcers and Bacteria I: Discovery and acceptance. Elsevier Studies in History and Philosophy of Science Part C: Studies in History and Philosophy of Biological and Biomedical Sciences 29 (1): 107–136. Thagard, P. 2002. Curing cancer? Patrick Lee’s path to the reovirus treatment. International Studies in Philosophy of Science 8595 (16): 179–193. https://doi.org/10.1080/02698590120118846. Thagard, P. 2012. Creative combination of representations: Scientific discovery and technological invention. In Psychology of Science: Implicit and Explicit Processes, eds. Proctor, R.W., and E.J. Capaldi. Oxford Scholarship Online. https://doi.org/10.1093/acprof:oso/9780199753628. 003.0016. Trout, J.D. 2016. Wondrous Truths: The Improbable Triumph of Modern Science. Oxford University Press. van Andel, P. 1993. Serendipity: Its origin, history, domains, traditions, patterns, appearances and programmability. In Precision Process Technology, ed. M.P.C. Weijnen and A.A.H. Drinkenburg, 689–704. Kluwer Academic. van Andel, P. 1994. Anatomy of the unsought finding. Serendipity: Orgin, history, domains, traditions, appearances, patterns and programmability. British Journal for the Philosophy of Science 45 (2), 631–648. https://doi.org/10.1093/bjps/45.2.631. Yakhlef, A., and E. Rietveld. 2019. Innovative action as skilled affordance-responsiveness: An embodied-mind approach. Creativity and Innovation Management, 1–13. https://doi.org/10. 1111/caim.12345.

Chapter 7

Serendipity and Ignorance Studies Selene Arfini

Abstract Selene Arfini seeks to resolve a long-standing paradox and the seemingly exclusive dichotomy between knowledge and ignorance through the concept of serendipity. How can we find new knowledge when we do not know what are we looking for? This question is a brief version of Meno’s or the Learner’s Paradox, which still manages to be upsetting in contemporary philosophy, despite having been discussed since Plato’s times. Arfini believes that the paradox still upsets because we strongly connect the explicit act of searching to the event of finding, in the same way as we describe the ideas of knowledge and ignorance as opposite and unrelated. Arfini argues that these assumptions live on, thanks to a cluster of misconceptions that still envelop the ideas of discovery, which could be overcome by utilizing insights from recent studies in serendipity, and by reframing ignorance from a cognitive perspective. MENO: And how will you search for something, Socrates, when you don’t know what it is at all? I mean, which of the things you don’t know will you take in advance and search for, when you don’t know what it is? Or even if you come right up against it, how will you know that it’s the unknown thing you’re looking for? (Plato 2005, p. 113).

Meno’s Paradox, in its simplicity, still troubles and upsets Plato’s readers when they stumble upon it. How can we find new knowledge when we do not know what are we looking for? This question is upsetting, not only to philosophers (Pirocacos 2020) but also to cognitive scientists (Paavola and Hakkarainen 2005) because our epistemological intuitions still strongly connect the explicit act of searching to the event of finding, just as they see the ideas of knowledge and ignorance as opposites and unrelated. Moreover, these two assumptions depend on each other: if knowledge and ignorance are antithetically opposed and mutually inaccessible, we could not jump from the latter to the former—at least not without knowing all in advance, as Plato’s solution suggests. In the same way, even if we accept the concept of serendipity— which can be briefly described as an unsought but beneficial discovery (Merton and Barber 2004)—it would only provide a “workaround” (Ingraham 2019) to the first S. Arfini (B) Department of Humanities—Philosophy Section, University of Pavia, Pavia, Italy e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 S. Copeland et al. (eds.), Serendipity Science, https://doi.org/10.1007/978-3-031-33529-7_7

125

126

S. Arfini

part of Meno’s paradox: the problem of how we could recognize new knowledge when we were previously ignorant about it would still remain. In this chapter I argue that these assumptions live on, thanks to a cluster of misconceptions that still envelop the idea of discovery, and which could be devaluated and overcome thanks to the recently advanced studies on the concept of serendipity and a reframed analysis of ignorance from a cognitive perspective. The positive endorsement of both these research directions comes from the adoption of an eco-cognitive approach to epistemological topics, which aims at configuring epistemological analysis by taking into consideration the cognitive agents’ possibilities and limitations in their ever-changing relationship with their body and environment. In this perspective, treating the concept of “ignorance” or “discovery” as theoretical abstracts that have little to do with these agent-environment interactions would reduce their meaning to clear but unrealistic ideas; they could well depict the prospects of an ideal agent but that would not encompass the nuanced experiences of an actual human agent. Thus, this chapter will aim at providing some reasons to acknowledge the importance of researching both serendipity and ignorance to advance our understanding of the concept of discovery in epistemology and cognitive science—without, of course, the ambition of closing this investigation. To open the discussion around the common misconceptions regarding the idea of discovery, in the first section I will address the hypothetical question “what if there was a way to generate serendipitous events without effort?” The answer will be an imaginary case called the Serendipity Machine.1 By discussing the issues that scenario brings out, I will point out that we need to consider both serendipity and ignorance to discuss the core premises of our capacity to discover something. In the second section, I will argue that we need to consider both serendipity and ignorance as processes rather than events or conditions that are action-driven, and that help us make sense of how we encounter and become familiar with new information. In this section I will also give details about which kind of ignorance we should take into consideration when we discuss serendipity and why hypothetical reasoning can shed some light on how the agent can go from ignorance to potential knowledge. In the third section, I will defend the thesis that, for investigating how discoveries happen, looking at the individual agent as a brain without a body and an embedding context is not enough. Adopting an eco-cognitive perspective and referring to recent literature on serendipity and ignorance studies (such as some of my previous work i.e. Arfini 2019; Arfini et al. 2020), I will point out the importance of acknowledging the role of embodied and distributed cognitive features in the emergence of chancebased discoveries and of related recognition of ignorance. In the conclusion, I will recap the discussion, acknowledging the misconceptions that still burden our idea of discovery and pointing out how the recent advancement of serendipity and ignorance studies could help us overcome them once and for all, 1

I resolved calling this thought experiment “the Serendipity Machine” before discovering that there was already a homonymous app that randomly connects people together (Serendipity Machine, n.d.) and a controversial book of the same name but that does not actually discuss serendipity (Olma, 2012). Of course, this argument has nothing to do with either of them.

7 Serendipity and Ignorance Studies

127

even diminishing our temptation to fall for easy ways out from the discussion, such as the one offered by the Serendipity Machine.

The Serendipity Machine and Why It Could not Work As already stated, the question of how we can find new knowledge without knowingly looking for it may be partially overcome by the reference to serendipity. The tricky part of involving the concept of serendipity in this discussion is that we should acknowledge its features, complexities, and subprocesses rather than addressing it as a cognitive miracle, that happens out of the blue without apparent reason. This part is tricky because, for some time, the concept of serendipity was a label used to designate, without explaining, some phenomena that simply involved lucky events— for example, the Oxford Learner’s Dictionary (2020) still refers to serendipity as “the occurrence and development of events by chance in a happy or beneficial way”. In recent times the definitions of serendipity provided by scholars of different fields have become more complex, featuring ambiguities (Copeland 2019; McCay-Peet and Toms 2015) and paradoxical connotations (Arfini et al. 2020; McBirnie 2008). These more nuanced definitions have marked an emerging need to comprehend not only the impact of the lucky event in the serendipitous occurrence, but also the role of the qualities of the agents who exploit it, such as sagacity and wisdom (Copeland 2018, 2019; Rivoal and Salazar 2013), and of the context in which they are embedded (Grundy and Hosking 1998; McCay-Peet et al. 2015). Still, the idea that epistemological, cognitive, and psychological research have extrapolated complex nuances from the concept of serendipity does not utterly defeat the assumption that a single-bite definition could suffice to encompass the phenomenon. While it is a dubious argument, thinking that serendipity could be simple and simply attained is pragmatically motivating. If we see serendipity as a positive event that merely “happens”, bringing game-changing outcomes, and we compare it with the slow and small results of heavy and hard-working research, then it is truly tempting to reflect about ways to optimize our chances to make serendipity happen, maybe automatically—or even effortlessly! I argue that the suppositions that back up this idea, and even how appealing it appears, encapsulate certain misconceptions and simplifications around not only the concept of serendipity but also about the very idea of discovery, to which we hold on despite the most recent research on cognition and reasoning. To be fair, I maintain that they are also encouraged by how we refer to our ignorance, both academically and commonsensically, as, simply, a lack of information. Indeed, the two most common academic definitions of ignorance are: lack of true beliefs (Peels 2012) or lack of knowledge as true and verified beliefs (Le Morvan 2013). So, if ignorance is just composed by a lack of true beliefs and serendipity is the event of finding believable and unexpected data, then serendipitous encounters could happen whenever we form true beliefs about something that we were previously ignorant about. Serendipity would become what happens when we read the newspaper, when we scroll through the newsfeed of our favorite social

128

S. Arfini

media, or when we see a video of instructions.2 If this implication were correct, we could even think, with a push from our imagination, that we could design a way to make serendipity easily happen with a sort of Serendipity Machine. To discuss the potential implications of this idea, let us imagine that a brilliant inventor designs that incredible machine, with just one small display on it upon which, once every minute, a sentence pops out, useful to us and about which we were previously ignorant. Since serendipity involves an “happy” event, it would report to us only truthful statements,3 which we could believe because we know that the machine could not lie. Would we say that this machine could make serendipity happen for us more frequently than it does naturally? Now, of course, there are many problems that pertain to this scenario. Some of them are relatively simple if we remember that this is a thought experiment and not an actual proposal for machine construction: How does the machine know what are we ignorant about? How does the machine know what is true and what is false? Would the machine be able to report to us only truthful statements that are already known to other people or also truthful statements that are unknown to the rest of humankind? Is the machine subject to classic logic limitations? All these questions and many more concentrate on pragmatic issues that relate to if and how we can construct such a machine. Even if nowadays some theories are actually emerging about the automatization of processes that could favor serendipity, I don’t think this machine could in any case represent a good candidate for that job. Before discussing the theoretical problems of this scenario, we should acknowledge why it is a good candidate to present the misconceptions that enclose our idea of discovery. Indeed, on the one hand, the Serendipity Machine exploits some features that actually define chance-based discovery—in the words of Anciaux (1993) “the gift, the discovery, the accident, the things, the non-search”—making it seem a close representation of what happens during a serendipitous encounter. On the other hand, there is also something wrong in this depiction: The same features that make the Serendipity Machine work do not seem enough to capture all of what happens when

2

I am not arguing that people cannot experience serendipitous occurrences in these contexts, but that it is not enough for them to be in these situations, in which they may find out new and unexpected information by chance, to experience serendipity. In the rest of the paper, I will argue that what we need to encompass in our definition of serendipity does not refer only to the epistemic status of the agents, but also to their embodied cognitive processes and their interactions with the environment. This point has been put forward also to understand cases of what Ross and Vallée-Tourangeau (2020, p. 3) call “microserendipity” [also investigated by Beghetto (2013) and Bogers and Björneborn (2013)]. In their study, they indeed employ a “dynamic, action-based account of creativity [that] invites tracking the smaller, micro moments that arise in the sometimes mundane practice of artistic creation.”. 3 I’m here considering truthfulness as a necessary but not sufficient condition to produce “happy” serendipitous events. Of course, this reduces the meaning of “happy” to a very specific quality, but this choice of words relies on the idea, commonly accepted in the philosophical literature, that it is always preferable to discover and come to believe in something true instead of something false. The apparent triviality of this assumption is nonetheless challenged by the many uses of the word “fiction,” which always manages to spur irritating reactions in serious epistemological circles.

7 Serendipity and Ignorance Studies

129

someone finds something unexpectedly. In few words, the Serendipity Machine is both an attractive and dubious representation of serendipity. To make more explicit why it has these two characteristics, we could refer to a renowned example of serendipitous discovery and test our intuitions about this thought experiment. Fleming famously discovered the antibacterial properties of penicillium mold after observing a petri dish in which staphylococcal bacteria were failing to reproduce in proximity of mold spores that, by accident, happened to have grown there first. So, thanks to this observation, he found out something true that he did not previously know. What if, instead of the petri dish, Fleming had just observed his Serendipity Machine and read the following sentence: “Mold of the penicillium genus has antibacterial properties”? In that case, would he have considered the discovery important and exploited its possibilities? It would be tempting to say yes, but, in this chapter, I will show you there are good reasons to resist this impulse. I argue that even if we could find a way to create a Serendipity Machine, it would still not make serendipity happen effortlessly for anyone. So, to make the following arguments clear, I need to say that, by discussing the Serendipity Machine I don’t aim at finding a solution to Meno’s paradox, but to argue that we should not bother to find a solution to what is really a non-problem. What the idea of the Serendipity Machine can do is convince us that Meno’s paradox is less troublesome than it appears to be. Less a paradox, even, and more a conceptual trap, which should help us notice that strict and internalist ideas about what discovery and ignorance create contradictory implications. To this aim, I use the Serendipity Machine as a thinking tool, to reveal several misconceptions we still hold on to around the ideas of discovery, serendipity, and ignorance.

Not Just a Happy Event First, categorizing statements as “truthful” and “of which we are personally ignorant about” is not enough to make them relevant and serendipity-worthy for us. Focusing on this part, we could derive two sets of misconceptions regarding the idea of discovery that we should address and break down. The first regards the assumption that in a process of discovery we go from a state of ignorance to a state of knowledge, as if ignorance were just an epistemic condition of human agents that passively affects their comprehension of a situation, as a missing piece of a knowledge-based jigsaw puzzle. The second misconception is that a chance-related discovery is primarily an event that happens to a naively and blissfully unaware agent.

130

S. Arfini

Rational, Willful, and Aching Ignorance So, first of all, let us talk about ignorance by admitting something uncomfortable: We do not always aim at reducing our ignorance and that is because it is not something that is inherently bad for us. On the contrary, we are used to being in a condition of ignorance and, since we are finite beings, we should not always aspire to escape it. Some conditions of ignorance are even presented as rational in contemporary theories of behavioral economics, in particular in rational choice theory (Somin 2015). For example, a condition of ignorance usually presented as rational is when the costs of acquiring particular knowledge outweigh the benefits of possessing it (Williams 2021). To make a simple example for it, if I am not, nor I am studying to be, a medical professional, it is both easier and more rational to rely on the advice of my family doctor if I need to decide what medicine I should take for an ear infection rather than acquiring the appropriate knowledge on my own. Since we need to be selective in choosing what deserves our attention, study, and epistemic efforts, it is not a given that what we personally do not know would be automatically interesting and useful for us. Sometimes ignorance is even the result of considered choices that are situationally or relationally motivated. Copeland (2022), for instance, discusses how sometimes our epistemic relations imply we preserve willful and beneficial ignorance because other people know things for us. There are times and situations in which we consciously preserve parts of our ignorance, avoiding situations that would give us clues and evidence that would provide us information and knowledge—even if this can lead us to partial or downright states of self-deception (Grossman and Van Der Weele 2017). So, implicitly, a serendipitous discovery is not just about something the agent was not prepared to find, but also something that would not fit into her/his rational ignorance (so that it would be no longer rational to remain ignorant about). Moreover, common sense and empirical studies (Gigerenzer and Garcia-Retamero 2017) confirm that willful—and not simply rational—ignorance exists, and it can be a way to avoid dangerous or painful unexpected news, or to preserve surprise, suspense and positive emotions (Williams 2021). In few words, not only is ignorance not inherently bad for us, but there are also many cases in which acknowledging and managing our ignorance is far better than reducing it. Without necessarily discussing controversial cases, for example situations in which people could choose to not know if they have a high risk of suffering from a genetic disease (Wahlin 2007), there are ordinary types of ignorance that we preserve, for example to maintain a fair judgment. In academia we use the double blind peer review to judge the products of our research for this reason, but we also commonly avoid spoiling anticipated surprise or marvel—information about a plot given too early is called “spoiler”, after all—or to avoid regret. This type of willful ignorance should also be taken into consideration to provide conceptual specifications to our idea of discovery: We would not always be happy to find out something that we would be epistemically, but also mentally and emotionally, unprepared to uncover.

7 Serendipity and Ignorance Studies

131

In that sense, it is now transparent that “something of which we were previously ignorant about”, or personally relevant ignorance, is not the same as valuable ignorance. Moreover, at this point it would be necessary to acknowledge that ignorance cannot simply describe a condition that we passively bear. The way we, in part, preserve, organize, and consciously deal with our ignorant states—through actions, but also epistemic attitudes such as belief, doubt, confidence, etc.—are relevant to the idea of ignorance as much as the simple condition of not knowing something. Indeed, we are used to referring to ignorance as Meno’s paradox presents it, as something that we cannot see, comprehend, nor reach. However, we should acknowledge the fact that not only can we preserve healthy and rational states of ignorance, but we can also act on the grounds of what we are unsure of or doubtful about, in order to look for information and to improve our assumptions and insights. In that sense, serendipity events exploit not a general type of ignorance, but one that we ache to overcome, that we had not the right tools to uncover before. For this reason, when we think about famous examples of serendipitous encounter it’s hard to separate the lucky event from the subsequent acts and efforts that agents adopt to make sure they got it right. In this regard, we should specify that, to have a clear picture of what discovery entails, in the same way in which we should begin addressing ignorance as more than a sadly and passively borne epistemic condition, we should also address serendipity as more than a simple event (or accident).

Before and After the Accident It’s not hard to explain why a potentially complex epistemological notion such as serendipity has obtained popularity in lay culture: Indeed, the term was voted the UK’s favorite word in 2000 (Rubin et al. 2011), and it has been described both as a “vogue” and a “vague” word in common understanding (Merton and Barber 2004). Indeed, as the Wizard of Oz of research, or the wonderful and hidden trickstergenius of all science, it’s almost too easy to think about colorful anecdotes of genius intuitions and marvelous discoveries when referring to examples of serendipity, since it has been for so long referred to as a “happy accident”. As the literature review of Foster and Ellis (2014) presents, the importance of serendipity emphatically points out the role of coincidence and luck in the intricate processes of scientific research. Of course, recent serendipity studies have provided ways to lift the curtains that hide the process and take a hard look at the real face of the Wizard. The research in this direction has not diminished the role of chance in the process of serendipity, but has rather given meaning to what happens also before and after the accidental event occurs.

132

S. Arfini

Before: Some Ignorance-Based Considerations on the Prepared Mind There is an intense debate on which conditions are a good prelude to a serendipity event. Theories regarding the so-called “prepared mind” (Gl¯aveanu 2022) demand that we acknowledge the contribution of background knowledge, interests, and individual characteristics in the epistemic process of recognizing something as a valuable accident. In brief, even if the agent does not expect something to happen, if they react appropriately by recognizing a good opportunity, they must have been epistemically and psychologically prepared for that eventuality. A reflection that we could advance now regards the role that “aching ignorance” may have in the prepared mindset or, better, relational context—as Ross (Chap. 9, this volume) eloquently suggests. Briefly, we could say that the same mindset and the same aching ignorance that brings us to subscribe to a newsletter could also prepare us for a serendipitous event. To advance a more composed explanation for this consideration, we should briefly return to the feasibility of the Serendipity Machine. Indeed, while considering it, a similar and more familiar construct may come to mind: The newsletter format, or the periodically received news from an organization to which we subscribed. Would these types of recurring news be realistic versions of the Serendipity Machine? Would the fact that sometimes they work as Serendipity Machines change our mind regarding that thought experiment? I would argue that newsletters are perfect examples of systems that favor serendipity without falling prey to the theoretical problems of the Serendipity Machine. Indeed, they work and they are different from a Serendipity Machine because they are people-based: The news we receive is created, managed, and updated by other human agents—so not the supernatural I-know-what-you-are-ignorant-of agents from our imaginary example. They are people-based also in the sense that we choose to subscribe to certain newsletters and not others. So, newsletters are not rustic Serendipity Machines: they may favor serendipity, but they are far from perfect tools for it. Indeed, we often waste a lot of time trying to unsubscribe from some of them because we stopped finding them useful. In that sense, we do exercise a form of control over the serendipitous occurrences that may emerge from our newsletter subscriptions. In particular, when we subscribe to a newsletter we anticipate, with good reason, serendipitous events. This is, of course, not controversial if we take into consideration how scientific research is conducted: As colleagues and myself have already argued (in Arfini et al. 2020), researchers always have a broad range of anticipated events that they take into consideration when research planning. These anticipated occurrences can be either explicitly or implicitly acknowledged and they may have to do with the possible success or failure of rounds of experiments or about the way an investigation could proceed forward after them. This range potentially distinguishes reasonably unexpected circumstances from wildly unexpected ones. Defining some situations as “wildly unexpected” depends heavily, of course, on the context. Even the hypothesis of “little green men who tamper with the data” can be considered, reasonably, as part of the equation in the right circumstances. For example, astrophysicist Jocelyn

7 Serendipity and Ignorance Studies

133

Bell Burnell reported that she considered this possibility before serendipitously discovering the existence of pulsar stars: In the fall of 1967, I was conducting a routine mapping project studying the radio scintillation of quasars for my doctoral thesis at Cambridge University, under the direction of my adviser, Antony Hewish. Investigation of a puzzling weak signal showed it to be a string of pulses, 1.33 seconds apart. We spent a month trying to find out what was wrong, so unexpected was the signal; and we nicknamed it “Little Green Men” (LGM). At the end of that month, I found a second pulsar, killing the LGM hypothesis and indicating a new kind of astronomical source (Burnell 2004, p. 489).

The LGM hypothesis was a possible solution, even if highly unexpected, because the scientists at the time, as Burnell (2004, p. 489) continues, thought they had enough “time to understand the instrument, recognize real and spurious signals, and investigate the anomalous or unexpected”. In that sense, they thought they would have been able to recognize even alien signals if the occasion would have arisen and so they would have transformed an unfamiliar and unexpected event into a familiar, even if strange, one. Following the same assumption, we can reflect on how we feed our aching ignorance when we consider what is to be expected in certain situations: Hypotheses regarding how an experiment could fail or which kinds of tools could help us collect certain data are necessary parts of scientific planning but they do not automatically exclude surprises and unexpected occurrences. I need to point out that the idea of “aching ignorance” is profoundly connected to what Ross (Chap. 9, this volume) calls the “frustrated mind,” which she defines as a state in which “the agent is aware of her shortcomings and is unable to resolve that epistemic abyss through continued cogitation.” It is interesting to note how both these terms describe what prima facie should suggest a negative state involving a lack of explanations, knowledge, or planned strategy. At the same time, they also invoke the idea that the agent is directed or oriented toward something. “Aching,” as longing or wishing, is specific: it is not a general negative feeling, but the need for a specific remedy to an unwanted situation. The term “frustrated” refers to a similar situation in which the negative sensation is caused by a specific issue. As Ross also argues, each of these features (the direction of the feeling and its negativity) contribute to make them the perfect drives of a serendipitous occurrence. Another example from recent philosophy of science that can easily explain the kind of epistemic flexibility that brings forward serendipitous occurrences is given by epistemological reflections on interdisciplinarity. Indeed, it is beyond doubt that collaborative interdisciplinarity is linked to the occurrence of serendipitous events (Holbrook 2017): More than that, different scholars have argued that serendipity can boost interdisciplinary and transdisciplinary collaborations (Darbellay et al. 2014; Rocca et al. 2019) and, vice versa, interdisciplinarity attracts serendipitous occurrences (Busch and Barkema 2020; Lyall 2020). Without much speculation, this relation can be explained by discussing the role of ignorance, especially “aching” ignorance, in the establishment of good interdisciplinary collaborations. Indeed, to establish an interdisciplinary project, the people involved need to acknowledge the disciplinary boundaries that limit their research: They need to know that they do not have the right conceptual or empirical tools, or both, to reach certain

134

S. Arfini

goals (Buller 2009). So, they need to make assumptions about what they do not know and how someone from another specialization could help. When the project begins, various “exhausting” and “messy” (Donaldson et al. 2010) interactions between disciplinary experts may help change, manage, and integrate those assumptions, increasing the range of anticipated expected and unexpected outcomes from the research (Townsend and Mikkonen 2019). So, the perfect conditions for discovery are prepared, in the shared context of interdisciplinary “aching” ignorance. Thus, how a discovery is anticipated, especially if serendipitously done, matters as much as how the discovery events occur. More than that, the aftermath of a discovery is even more relevant, especially if we discuss the reasoning that accompanies the agents’ steps through the recognition of a discovery as such.

After: Abduction and Post-Hoc Realizations It is now a common assumption that in the context of discovery and even serendipity, luck describes just the trigger of discovery, not the discovery itself. Indeed, the tight relationship between a propitious occurrence and the actual exploitation of it heavily relies on the researchers’ work after a particular opportunity has been uncovered. To comment on what happens, we should refer again to recent studies on serendipity, as well as to Fleming’s discovery and why a written input from the Serendipity Machine could not have equally likely been the trigger of a natural serendipitous occurrence for him. In the first section I argued that the Serendipity Machine could have stated “mold of the penicillium genus has antibacterial properties” to equally match Fleming’s serendipitous discovery. To be fair, though, Fleming could not have immediately thought this sentence when he found out the compromised petri dish: Indeed, if it were so, others before him could have made the same discovery. As Copeland (2019) has commented, Fleming was aware that this kind of accident could not have been an isolated occurrence and that “inhibition of one microbe by another was commonplace”, as he stated in his Nobel prize speech. Others, including his contemporary André Gratia, could have worked out the groundbreaking importance of the antibacterial properties of P. Notatum mold, if the sole observation of the phenomenon had been enough to make the connection. The point is: The “lucky” occurrence of a petri dish with inhibited bacteria as well as its observation are not enough to make the sentence “mold of the penicillium genus has antibacterial properties” pop out as significant in the mind of that period’s scientists. In other words, there is correlation, not causation at the heart of the machine’s serendipitous process, between the occurrence of a chance-based event and the recognition of its importance. What Fleming did, which uniquely brought out the actual discovery of penicillin properties, was to be surprised by the occurrence he observed and to perform abductive reasoning to explain it. Abduction is an epistemological concept that, in brief, refers to fallible (in the classical logic sense) but useful kind of reasoning that connects a surprising event to a general rule (Magnani 2009). It is useful, especially in creative and discovery processes (Magnani 2017), because it exploits the state of surprise to

7 Serendipity and Ignorance Studies

135

shape an explanatory process: It basically follows the “why is that so?” exclamation and it pushes the agent to propose and test hypotheses and conduct further empirical investigations. It is also the reason why the act of reading the sentence “mold of the penicillium genus has antibacterial properties” would not have triggered a serendipitous process for Fleming. That sentence is rather the conclusion of an explanatory reasoning that occurred because a strange and unexplored fact was bothering Fleming with surprise and “aching” ignorance. It became significant for him because the first observation begged for an explanation and led him to conduct a series of experiments and to formulate hypotheses to make sense of that original, neglected petri dish. Without the epistemological work performed after the lucky event, the simple sentence of explanation would not have been useful or referred to as significant: In other words, it would have given an answer without specifying to which question it responds. Moreover, the emotional words I used in this section, such as surprise, bothering, aching, are not there to provide metaphorical embellishments to the context of discovery. Studies on both serendipity (Ross and Vallée-Tourangeau 2020) and ignorance (Arfini 2021) are now invoking the need to consider more than just the conceptual process of data elaboration to address and understand cognitive phenomena. These additional considerations provide other reasons to abandon common misconceptions regarding how discoveries happen and why the Serendipity Machine is a flawed representation of a more complex but still explorable phenomenon.

The Embodied and Distributed Aspects of Discovery The process of discovering something, as already argued, cannot be correctly described as just the activity of finding out some pieces of information the agent did not have before. Even acknowledging the importance of the pre- and post-hoc processes that prepare and follow the event, the actual discovery of new information does not happen as a simple combination of new information in the mind of the agent, like the positioning of missing pieces of a jigsaw puzzle.4 Moreover, this metaphorical representation is not insignificant: if our mind worked like a jigsaw puzzle and so we would accumulate knowledge by filling up blank spaces—which would represent our ignorance—then Meno’s paradox should rightfully upset us. How could we know which piece could fit the blank? And if we knew it in advance, why should we bother to fill it at all—since we would already know? If that were the case, then the Serendipity Machine would also properly work, offering ready-made pieces we didn’t know were missing. But, if it is not a reliable representation—and we are arguing for exactly that—how should we address the act of discovery? 4

Here I’m not arguing that a sort of internal “conceptual combination,”–as Thagard presents, “the process in which new theoretical concepts arise by putting together old ones” (2012, p. 109)–does not occur, but I’m defending the idea that this is not the only nor the most important part of the cognitive process that brings a researcher to perform a serendipitous discovery.

136

S. Arfini

Thinking about discovery as merely finding out new data not previously elaborated would mean to address the discovering agents as little more than brains-in-a-vat5 : Without recognizing their epistemological attitudes as incorporating action-based reasons and discussing at least their embodied and distributed cognitive performances, we would not simply have an “abstract” view of what happens when someone discovers something, but a conceptually compromised construct. Both in serendipity and ignorance studies, the idea that we should approach these themes from an externalist perspective have started to emerge and provide new ways to approach the idea of discovery (Lavazza and Manzotti 2013). In this section, I will provide comments on the embodied and distributed features of discovery and I will argue that, by acknowledging the importance of discussing these traits, we allow ourselves to abandon certain misconceptions that still permit us to become bothered with Meno’s paradox and that push us to look for and believe in simplified versions of serendipity.

The Body Matters: Epistemic Feelings and Discovery Acknowledging the interrelation between cognitive processes and bodily feelings is not a new idea, even if it’s still considered a radical one in certain academic circles (Silva and Ferreira 2019). The critics of this view mainly question the assumption that feelings and bodily cues are integral parts of cognitive actions and processes (so, not just either their result or their trigger), which contribute to and compromise the propositional contents of reasoning and thought (Chemero 2011). Recent perspectives on discovery allow us to recognize the roles that surprise, irritation, and relief play in the epistemological process (Arfini 2019; Cleverly et al. 2012). A way to approach these topics is to analyze and discuss the specific features of what the literature now calls “epistemic feelings” (Arango-Muñoz 2014). Epistemic feelings are phenomenological experiences regarding the agent’s own cognitive abilities, conditions or processes. To give a fairly common example, we feel we recognize someone in a crowd because we have seen him/her somewhere. The feeling arises in correlation to a cognitive condition—a belief, an assumption, a doubt, etc.—that enables us to say “I’ve seen her/him somewhere”. So, a bodily reaction accompanies these experiences—for example, we can remember the feeling 5

The brain-in-a-vat hypothesis has been advanced by Harman (1973), to theoretically defend a radical form of skepticism. The argument says that if we were just disembodied brains that are kept alive by a mad scientist who feeds us electrochemical impulses that deceive us in thinking that we have a body, a job, and a reality around us, we would not be able to notice it. In this case I used the expression brain-in-a-vat, as others do in philosophical discussions (most famously, Hilary Putnam), to refer to an idea of cognition and mind which is, in theory, disembodied, as though all those processes that shape our cognitive lives were “all in our head,” as Chalmers and Clark (1998) famously stated. The 4E cognitive theories, which I personally endorse, do not find the brain-in-a-vat thought experiment particularly convincing because they defend a conception of cognition which is profoundly connected to embodied features of the agents and to processes distributed in the social and material environment.

7 Serendipity and Ignorance Studies

137

of having something ‘on the tip of our tongue’ (Dellantonio and Pastore 2019), which relates to epistemic conditions. So, even if it is a given that any epistemic experience comes with not only cognitive and metacognitive processes and states but also with non-cognitive reactions, it’s useful to keep in mind that the range of feelings that interact with our epistemic processes is fairly articulated. Furthermore, epistemic and emotional feelings can create loops between each other without the conscious approval of the cognitive subjects involved. The epistemic processes that lead to discovery, of course, are not immune to those loops. We already discussed the profound effect that surprise has on the cognitive reaction of the agent. It triggers that “aching” ignorance that makes us want to know something that is temporarily out of our cognitive grasp. But if we take serendipity into account as an interesting event related to the act of discovery, another specific feeling that marks the start of the discovery journey immediately comes to mind: the aha-moment (Kounios and Beeman 2009). In this sense, if surprise contributes to the epistemological engagement with the problem, the aha-moment could have at least two particular effects. One has been described by American pragmatist Charles Sanders Peirce (1878), as the fixation of belief: It basically describes the state of being content with a piece of information, which is treated as solid and truthful. The agents “believe” that information enough to start depending on it to act and react to their environment “as if it were true”. This confidence permits the agents to act upon what they believe is true even when it is not, sometimes creating epistemic illusions that may affect their subsequent judgments, e.g., epistemic bubbles (Woods 2005). Of course, the aha-feeling may also allow the agent to want to prove the hypothesis that made them say “aha!”, through hard work in the form of making observations, performing series of experiments, and then by constructing around that aha-feeling a coherent theory supported by further hypotheses and theses. This is part of the post-hoc scenarios that surround discoveries in science and that would probably not happen if the Serendipity Machine would simply present a concise description of a state of affairs about which we did not know anything. In brief, if anticipatory processes prepare the discovery, and abductive explanatory reasoning follows it, the embodied features that we channel mediate our experience, making the potential discovery more or less productive. For this reason, recent studies on serendipity and ignorance are now focusing on the extended, embodied, and distributed framework in which the agents dwell: Without this enclosing narrative, discovery would be miraculous, and the history of humankind would seem just a series of terribly fortunate events (see also Copeland, Chap. 6 of this volume).

138

S. Arfini

Distributed Serendipity: Interdisciplinarity and Disciplinary Jars To strengthen the argument that we should consider the concept of serendipity and ignorance from an externalist perspective, we can reflect once more on the productive relationship between serendipity and interdisciplinarity and what kind of “aching” ignorance could favor them both. Of course, in collaborative settings ignorance is more subtle and difficult to catch than in individual cases (indeed, especially in analytic philosophy, ignorance is analyzed as belonging to singular agents, not to interacting groups), but it is implied, nonetheless. To get an idea of the kind of ignorance that is specifically involved in interdisciplinary collaborations, it could be useful to direct our attention to the idea of “disciplinary boundaries”. Relevantly, Darbellay et al. (2014, p. 1) start their article with the translation of this interesting quotation: Each discipline can be compared to a jar. Once we are in the jar, it takes genius to get ourselves out of it or to innovate: we feel that the jar’s boundaries are natural (Veyne 1983, p. 127).

When reading a sentence like that our first instinct could lead us to embrace the metaphor as representing abstract objects and features. In this case, it is easy to think about the jar of anyone’s discipline as representing conceptual boundaries, the theoretical framework in which people who engage one discipline find themselves in. Although that seems not wrong, discussing interdisciplinarity by just thinking about theoretical boundaries would not be enough. Indeed, when engaging with a discipline, we get used not only to adopting certain technical concepts and figures, but also material tools; we interact with a particular community of more or less likeminded people, and we get to know some environments in which, through education and efforts, after a few years we begin feeling at home. Thus, the jar cannot be just a metaphor for conceptual boundaries, but also for the mix of feelings of familiarity, as well as the common environment, community, and tools that come with the disciplinary perspective. Following further the metaphor, as already said, we assume that it is not enough to know our own jar: we often need someone else’s expertise and experience to reach a particular scientific goal. However, it is difficult for us to actually understand how much we don’t know about others’ jars and about the efforts they are putting into reaching out to understand ours. Indeed, evidence from countless case studies on interdisciplinary collaborations that did not click, at least for a while, report that people often have problems with other teams’ timelines (since they expect that some parts of the project taken by other disciplinary teams would take more or less time than it actually ends up taking), misuse resources and funds, and mostly blame the team members from other disciplinary backgrounds (MacLeod and Nersessian 2016). The researchers usually attribute these problems to various causes, and lack of communication is always on top of their list. Nevertheless, it is not a wild guess that these problems ultimately depend on unacknowledged ignorance, in other words, the difficulty of understanding others’ priorities, tools, needs and so on – also called “domain specificity” by MacLeod

7 Serendipity and Ignorance Studies

139

(2018). Now, for interdisciplinarity it is easy to see where this is going. This kind of ignorance is the effect of the boundaries of disciplinary jars. We believe we can form realistic expectations about others’ work and methodology, even when we know that we are working together exactly because our methods, tools, and distributed expertise (distributed through the methods, material tools, artifacts, and environments we expertly use) are different. Thus, exactly for this reason, discussing the kind of ignorance that is involved in discovery, and serendipitous discoveries in particular, can be useful. As already argued, when we enact a serendipitous discovery, we realize that we could not prepare for it, but we anticipate it as a possible but unlikely event. Expectations regarding serendipity should be taken into consideration here exactly because we know ignorance is a key part of the serendipity game, but we can be vigilant regarding it. The same thing goes for interdisciplinary collaborations: Once the researchers adopt the perspective that their jar is limited and they can only prepare up to a point for what is outside of it, they also become ready to negotiate more, which in turn makes them also more prepared for chance-related discoveries. Recognizing the role that distributed and relational cognition plays in the development of our cognitive processes, especially in the context of discovery, should open our eyes even more to the limitations of the kind of perspective we would embrace if we let Meno’s paradox upset us. Indeed, if we resist acknowledging the complexities of serendipity, the role of ignorance and of embodied and distributed cognition in the performance of a discovery, we should be upset that our scientific endeavors have made any progress at all.

Concluding Thoughts So, how can we find new knowledge when we do not know what are we looking for? This chapter addressed how and why this question misrepresents how discovery works. Without pretending to close the topic to further inquiries, my goal here was to propose recent studies in ignorance and serendipity as guides to properly discussing interesting issues pertaining to how we find out about new knowledge. To direct my arguments, I used a thought experiment, the Serendipity Machine, which concentrated, in a brief scenario, some misconceptions that we still hold on to when referring to the concept of discovery. The misconceptions I referred to regard the idea that discovery is simply the assertion of new, as in previously unknown, information in the mind of the researcher. I also argued that these misconceptions, in turn, depended on: • The idea that serendipity is a simple event that involves a lucky occurrence that comes out of the blue without requiring epistemological efforts from the agents involved;

140

S. Arfini

• The description of ignorance as the opposite of the state of knowledge or true belief, as an unwanted state that can be simply described as a lack of true beliefs or true and verified beliefs. I argued instead that serendipity is more than a happy event that only modifies the propositional beliefs of the agent, and ignorance cannot just be a state passively borne by the agents, also affirming the necessarily interconnectedness between the embodied and distributed context of the agent and the action-based processes that compose the concept of discovery. Indeed, in a serendipitous process we interact with our environment, following our epistemic feelings and managing our distributed cognition to perform certain processes and adapt our cognitive tools to explain those surprising facts that we could not previously approach until they finally make sense for us. As Copeland (2018) already suggested, epistemological and cognitive explanations regarding how serendipity is enacted may challenge the long-lasting idea that discovery is a “mysterious” process, and that epistemologists should instead focus on how proper justifications are provided in “ordinary” methodology (see also Copeland, Chap. 6 of this volume). The point that these cognitive-based reflections should advance is that there is nothing “out-of-the-ordinary” about discovery, except for its positive impact on research efforts: thus, it should follow the same rules that we apply when we focus on how regular cognitive activities lead us to explanatory or analytic reasoning. Indeed, discussing how serendipity emerges within the ordinary process of discovery is now the subject of ongoing new research directions regarding the specific role of ignorance and of 4E cognition.6 At this point, the one thing that can be assured it is that no further answers could be found by pondering about something like the Serendipity Machine. Acknowledgements I need to express my deepest thanks to Samantha Copeland and Wendy Ross for too many thought-provoking conversations on serendipity, chance-based reasoning, and ignorance, and their invaluable comments on an earlier draft. Research for this article was supported by the PRIN 2017 Research 2017 3YP4N3—MIUR, Ministry of University and Research, Rome, Italy. Author affiliation.

References Anciaux, A. 1993. The serendipitous effects of disaster management by nonprofit organizations: Studie cases in Jamaica, Costa Rica and Dominican Republic. Society for Applied Anthropology Annual Meeting, 1–12.

6

For example, one good question to ask is how to differentiate serendipity from deep understanding of new information, which also requires work, preparation, and ‘aching’ ignorance. While I’m grateful to Wendy Ross for pointing this out in her spot-on review, I have to confess that I do not have an easy answer for this question yet. If nothing else, trying to find a well-thought answer to this question is however a good reason to continue the research on the impact of 4E cognition and the role of ignorance in serendipitous occurrences.

7 Serendipity and Ignorance Studies

141

Arango-Muñoz, S. 2014. The nature of epistemic feelings. Philosophical Psychology 27 (2): 193– 211. https://doi.org/10.1080/09515089.2012.732002. Arfini, S. 2019. Ignorant Cognition. A Philosophical Investigation of the Cognitive Features of Not-Knowing: Springer. Arfini, S. 2021. Situated ignorance: The distribution and extension of ignorance in cognitive niches. Synthese 198: 4079–4095. Arfini, S., T. Bertolotti, and L. Magnani. 2020. The antinomies of serendipity how to cognitively frame serendipity for scientific discoveries. Topoi 39: 939–948. Beghetto, R.A. 2013. Nurturing creativity in the micro-moments of the classroom. In Creatively Gifted Students are Not Like Other Gifted Students, 3–15. Brill Sense. Bogers, T., Björneborn, L. 2013. Micro-Serendipity: Meaningful Coincidences in Everyday Life Shared on Twitter. Buller, H. 2009. The lively process of interdisciplinarity. Area 41 (4): 395–403. Burnell, J.B. 2004. So few pulsars, so few females. Science 304 (5670): 489–490. Busch, C., and H. Barkema. 2020. Planned luck: How incubators can facilitate serendipity for nascent entrepreneurs through fostering network embeddedness. Entrepreneurship Theory and Practice. Chance, Z., and M.I. Norton. 2015. The what and why of self-deception. Current Opinion in Psychology 6: 104–107. Chemero, A. 2011. Radical Embodied Cognitive Science. MIT press. Clark, A., and D.J. Chalmers. 1998. The extended mind. Analysis 58 (1): 7–19. Cleverly, J., Heath, C., vom Lehn, D., and P. Luff. 2012. Revealing Surprise: The Local Ecology and the Transposition of Action. Copeland, S. 2018. “Fleming leapt on the unusual like a weasel on a vole”: Challenging the paradigms of discovery in science. Perspectives on Science 26 (6): 694–721. Copeland, S. 2019. On serendipity in science: Discovery at the intersection of chance and wisdom. Synthese 196 (6): 2385–2406. Copeland, S. 2022. Relational Ignorance. In Embodied, Extended, Ignorant Minds. New Studies on the Nature of Not-Knowing, eds. by Arfini, S., and L. Magnani. Springer. Darbellay, F., Z. Moody, A. Sedooka, and G. Steffen. 2014. Interdisciplinary research boosted by serendipity. Creativity Research Journal 26 (1): 1–10. Dellantonio, S., and L. Pastore. 2019. How can you be sure? Epistemic feelings as a monitoring system for cognitive contents. In Model-Based Reasoning in Science and Technology, eds. by Nepomuceno-Fernández, Á., Magnani, L., Salguero-Lamillar, F.J., Barés-Gómez, C., and M. Fontaine, 407–426. Springer International Publishing. https://doi.org/10.1007/978-3-03032722-4_23 Donaldson, A., N. Ward, and S. Bradley. 2010. Mess among disciplines: Interdisciplinarity in environmental research. Environment and Planning A: Economy and Space 42 (7): 1521–1536. Fingarette, H. 2000. Self-deception: With a New Chapter. University of California Press. Foster, A.E., and D. Ellis. 2014. Serendipity and its study. Journal of Documentation 70 (6): 1015– 1038. Gigerenzer, G., and R. Garcia-Retamero. 2017. Cassandra’s regret: The psychology of not wanting to know. Psychological Review 124 (2): 179. Gl¯aveanu, V.P. 2022. What’s ‘inside’ the prepared mind? Not things, but relations. In The Art of Serendipity: Insight, Innovation and Inspiration, eds. Ross, W., and S. Copeland. Palgrave Macmillan. Grossman, Z., and J.J. van der Weele. 2017. Self-image and willful Ignorance in social decisions. Journal of the European Economic Association 15 (1): 173–217. https://doi.org/10.1093/jeea/ jvw001. Grundy, J.C., and J.G. Hosking. 1998. Serendipity: Integrated environment support for process modelling, enactment and work coordination. Automated Software Engineering 5: 27–60. Harman, G.H. 1973. Thought. Princeton University Press.

142

S. Arfini

Holbrook, J.B. 2017. Peer review, interdisciplinarity, and serendipity. In The Oxford handbook of interdisciplinarity, ed. by Frodeman, R., 485–498. Ingraham, C. 2019. Serendipity as cultural technique. Culture, Theory and Critique 60 (2): 107–122. Kounios, J., and M. Beeman. 2009. The Aha! moment: The cognitive neuroscience of insight. Current Directions in Psychological Science 18 (4): 210–216. Lavazza, A., and R. Manzotti. 2013. An externalist approach to creativity: Discovery versus recombination. Mind & Society 12 (1): 61–72. Le Morvan, P. 2013. Why the standard view of ignorance prevails. Philosophia 41 (1): 239–256. Lyall, C. 2020. Facilitating serendipity for interdisciplinary research. Integration and Implementation Insights. https://i2insights.org/2020/01/21/serendipity-for-interdisciplinarity/. MacLeod, M. 2018. What makes interdisciplinarity difficult? Some consequences of domain specificity in interdisciplinary practice. Synthese 195 (2): 697–720. MacLeod, M., and N.J. Nersessian. 2016. Interdisciplinary problem-solving: Emerging modes in integrative systems biology. European Journal for Philosophy of Science 6 (3): 401–418. Magnani, L. 2009. Abductive cognition. The epistemological and eco-cognitive dimensions of hypothetical reasoning: Springer. Magnani, L. 2017. Abductive Structure of Scientific Creativity. Springer. McBirnie, A. 2008. Seeking serendipity: The paradox of control. Aslib Proceedings: New Information Perspectives 60 (6): 600–618. McCay-Peet, L., and E.G. Toms. 2015. Investigating serendipity: How it unfolds and what may influence it. Journal of the Association for Information Science and Technology 66 (7): 1463– 1476. McCay-Peet, L., E.G. Toms, and E.K. Kelloway. 2015. Examination of relationships among serendipity, the environment, and individual differences. Information Processing and Management 51 (4): 391–412. Merton, R.K., and E. Barber. 2004. The Travels and Adventures of Serendipity: A Study in Sociological Semantics and the Sociology of Science. Princeton University Press. Olma, S. 2012. The serendipity machine. A Disruptive Business Model for Society 3. Paavola, S., and K. Hakkarainen. 2005. Three abductive solutions to the Meno paradox–with instinct, inference, and distributed cognition. Studies in Philosophy and Education 24 (3): 235–253. Peels, R. 2012. The new view on ignorance undefeated. Philosophia 40 (4): 741–750. Peirce, C.S. 1878. How to Make Our Ideas Clear 2000: 77–87. Pirocacos, E. 2020. False Belief and the Meno Paradox. Routledge. Plato. 2005. Meno and Other Dialogues Charmides, Laches, Lysis, Meno (R. Waterfield, Trans.). Oxford University Press. Rivoal, I., and N.B. Salazar. 2013. Contemporary ethnographic practice and the value of serendipity. Social Anthropology 21 (2): 178–185. Rocca, E., S. Copeland, and I. Ralph Edwards. 2019. Pharmacovigilance as scientific discovery: An argument for trans-disciplinarity. Drug Safety 42 (10): 1115–1124. https://doi.org/10.1007/ s40264-019-00826-1. Ross, W., and F. Vallée-Tourangeau. 2020. Microserendipity in the creative process.The Journal of Creative Behavior, 1–12. https://doi.org/10.1002/jocb.478 Rubin, V.L., Burkell, J., and A. Quan-Haase. 2011. Facets of serendipity in everyday chance encounters: A grounded theory approach to blog analysis. Information Research 16 (3), 488. Serendipity. n.d.. In Oxford Learner’s Dictionary. Oxford University Press. Retrieved 21 December 2020, from https://www.oxfordlearnersdictionaries.com/definition/english/serendipity. Serendipity Machine. n.d. in Serendipity Machine. Retrieved 1 February 2021, from https://serend ipitymachine.com/. Silva, M., and F. Ferreira. 2019. Special issue on radical views on cognition: Introduction. Synthese, 1–4. Somin, I. 2015. Rational ignorance. Routledge International Handbook of Ignorance Studies, 274– 281.

7 Serendipity and Ignorance Studies

143

Thagard, P. 2012. The Cognitive Science of Science: Explanation, Discovery, and Conceptual Change. MIT Press. Townsend, R., and J. Mikkonen. 2019. Serendipity as a Catalyst. Knowledge Generation in Interdisciplinary Research. The Design Journal 22 (sup1), 1853–1869. https://doi.org/10.1080/146 06925.2019.1595038 Veyne, P. 1983. Les Grecs croyaient-ils à leurs mythes. Editions du Seuil. Wahlin, T.-B.R. 2007. To know or not to know: A review of behaviour and suicidal ideation in preclinical Huntington’s disease. Patient Education and Counseling 65 (3): 279–287. Williams, D. 2021. Motivated ignorance, rationality, and democratic politics. Synthese 198: 7807– 7827. https://doi.org/10.1007/s11229-020-02549-8. Woods, J. 2005. Epistemic bubbles. In We Will Show Them! Essays in Honour of Dov Gabbay, Volume Two, eds. by Artemov, S., Barringer, H., d’Avila Garcez, A., Lamb, L.C., and J. Woods, 731–774). College Publications.

Chapter 8

Serendipity as a Design Principle of Personalization Systems—Theoretical Distinctions Urbano Reviglio

Abstract Urbano Reviglio elaborates in this chapter on a framework for the design of serendipity in human–computer interactions and digital environments, especially for personalised online media and (news) recommender systems. His contribution is a theoretical and interdisciplinary investigation of digital serendipity in the context of personalization systems. Serendipity is deeply involved in various sub-domains of information science, such as information retrieval, information literacy, knowledge management and knowledge acquisition. It is considered a design principle able to reduce the risks of redundant and overly consonant information consumption (also referred to as filter bubbles and echo chambers). Yet, Reviglio shows that the theoretical and ethical analysis of the concept is still limited. This article attempts to fill the gap by providing a taxonomy of serendipity as it might be used in the design, or found through the use of personalization systems. Based on an interdisciplinary literature review, distinctions between pseudo-personalised and hyperpersonalised serendipity, between individual and political serendipity, and between fake and manipulative serendipity are described. Finally, Reviglio discusses how this taxonomy can encourage the design of more ethically sound digital systems that incorporate serendipity.

Every single day billions of people search, browse and scroll information online for a significant amount of time. Such information is mostly recommended algorithmically. Through data collection and analysis, digital environments, in fact, allow the development of increasingly personalised services; for example, when Google suggests results, Netflix suggests movies, Youtube videos, Facebook posts, Spotify music and Amazon suggests related products. Our time and attention are scarce

U. Reviglio (B) University of Milan, Milan, Italy e-mail: [email protected]

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 S. Copeland et al. (eds.), Serendipity Science, https://doi.org/10.1007/978-3-031-33529-7_8

145

146

U. Reviglio

resources, whereas the quantity of potentially relevant information is immense.1 Our attention is thus mostly outsourced to these powerful algorithms that perform a fundamental role in knowledge management; these are both socio-technical systems that mediate and influence interactions with the social and economic environment, and technologies of construction of the self that can significantly affect who we are and who we think we are. This naturally raises questions of how to develop ethical personalization systems, what are (and ought to be) the values inscribed into them, and eventually how to operationalise them. In the last decade, there have been concerns that personalization systems can reduce media diversity, creating filter bubbles (Pariser 2011) and echo chambers (Sunstein 2017a, b),2 but also stimulate compulsive engagement and, eventually, politically polarize and even radicalize. While research is still unconvincing and inconclusive (Bruns 2019), there is evidence that these algorithmic systems can increase political polarization, disinformation and influence public opinion formation in ways that are not yet well understood. One of the solutions to mitigate these concerns is to design these systems to account more and more explicitly for serendipity, generally intended as unexpected and meaningful recommendations that intersect users’ presumed interests. The main goals are to reduce the risks that a redundancy of an overly personalised environment could cause and, conversely, to expand users’ opinions, interests and knowledge. At first sight, this idea might sound either obvious or counterintuitive. Serendipity is indeed the epitome of the Internet. Media theorist Steven Johnsons described the Internet as “the greatest serendipity engine in the history of culture” (2006). Everyday Internet users already discover information that changes their opinions and ideas, or even their beliefs and worldviews. Social media, search engines, news aggregators, social bookmarking and, more generally, browsing and hyperlinks are powerful sources through which to experience serendipity. Serendipity, however, is not only a widespread by-product of digital environments; it can be intentionally inscribed into personalization algorithms and designed through affordances (e.g., trending topics and hashtags). The design of serendipity in digital environments is indeed receiving increasing attention in respect to its role, its potential cultivation, the threat of its loss, and even its manipulation.

1

Take two paradigmatic examples of personalization online: Facebook and Youtube. On Facebook, the posts encountered by the average user everyday are approximately 350 on at least 1.500 potential posts. Thus, roughly 25% are algorithmically filtered and prioritised and 75% are ultimately hidden. While Youtube’s algorithm already drives more than 70% of the time spent in the platform. Indeed, 90% of the ‘related content’ in the right bar is personalised (see Reviglio 2019). 2 These are two intertwined phenomena; at the individual level filter bubbles and at the group and social level echo chambers. While the first refers to the fact that a user continues to receive and consume content that reinforces their opinions and interests, the latter refers to a group situation where information is uncritically spread and amplified, whereas dissonant information is mostly ignored. These concepts are, however, poorly defined and, in fact, are used more as simplifying metaphors (Bruns 2019).

8 Serendipity as a Design Principle of Personalization …

147

Serendipity has thus become a desirable design principle for information societies, especially in personalization systems and, more generally, in digital environments (Reviglio 2019). Despite multiple dimensions of serendipity, the concept is often used as an umbrella term to argue against determinism in personalization algorithms practices. Theoretical distinctions are thus needed. This article attempts to fill the gaps to enrich the conceptualization of “digital serendipity” and, eventually, contribute to improve its cultivation as well as the development of more responsible personalization systems. The chapter is organised as follows: in the first part, I introduce serendipity and its development in digital environments. In the second part, I develop a conceptual taxonomy of how serendipity has been mainly conceived, introducing meaningful distinctions emerging from the literature review. Subsequently, I discuss how this taxonomy provides a point of departure for a more granular and informed design of serendipity in digital environments.

An Introduction to Digital Serendipity Let’s start by clarifying the phenomenon of interest: A general definition of serendipity by Merriam-Webster is “the faculty or phenomenon of finding valuable or agreeable things not sought for”. Serendipity can be a route to new knowledge, problem-solving, belief change and, more generally, creativity and innovation. In this process, subjects involved commonly experience unexpectedness, surprise and meaningfulness. These characteristics can be cultivated in physical as well as digital environments, for what is called “artificial serendipity”, that is, the ability to create the necessary conditions for serendipity to occur (de Melo 2018). Instead, “natural serendipity”—meaning the serendipity that occurs naturally in the world—is much more unpredictable, as the number of factors and variables that create it may be difficult if not impossible to calculate. Various terms have been used to describe the phenomenon of serendipity in digital information environments such as incidental information exposure (Yadamsuren and Erdelez 2016), accidental information discovery (Race and Makri 2016), automated serendipity (Fletcher and Nielsen 2018), micro-serendipity (Bogers and Björneborn 2013) or, more broadly, information encountering (Erdelez and Makri 2020). Different definitions approach the phenomenon from different perspectives and assign different weights to personal and environmental factors, between individual responsibility and lucky chance (McCay-Peet and Toms 2017).3 Digital serendipity has been seen in particular from three main perspectives: (1) the inevitable serendipity that large amounts of personalised information provide (e.g., automated serendipity), 3

For example, information behavior research approaches serendipity as a quality of someone. Research relating to recommender systems (RSs) and search engines, on the other hand, approach serendipity primarily as a quality of an event or something. And, lastly, information science and human–computer interaction describe it as an experience or a process.

148

U. Reviglio

(2) the accidentality of encountering information online (e.g., Incidental Exposure to Online News) and, relatedly, the ability of users to experience serendipity and (3) the opportunity to design for serendipity (e.g., how serendipitous is a recommended item to a user). While being approached as either an individual proclivity and experience, or as a phenomenon or event, the most holistic approach to can be found in information science and Human–Computer Interaction research, where serendipity is more often approached as a process. This perspective is the one mainly adopted in this article. Skepticism about designing and especially programming (“pure”) serendipity has long been widespread (van Andel 1994; Krotoski 2011; Carr 2016). As the argument goes, once you create an engine or a system to produce serendipity, if ever, you destroy its very essence. If serendipity could be controlled, then an event would no longer be serendipitous, but predictable or reproducible, and therefore manipulative, “fake serendipity” (Erdelez et al. 2019, see also Erdelez, Chap. 12, this volume). As I argue here, despite theoretical speculations and empirical challenges, it is actually possible to cultivate serendipity (Race and Makri 2016). Moreover, the phenomenon of serendipity in digital contexts has been reframed as commonplace, rather than rare or random. That is, despite classic narratives that highlight how wondrous and rare serendipity is, in today’s information societies it is actually very common, even if we do not always recognise it when it happens. Serendipity is indeed an integral part of everyday information behaviour, and there are varying degrees of serendipitous encounters. They may range from scientific discoveries (e.g., paradigm shifts) to mundane discoveries (e.g., memes). Ideally, these serendipitous encounters ought to be insightful and arise from accurate information because otherwise they could result in outcomes that only seem beneficial to the individual (e.g., infotainment4 ), and may be even harmful to society at large (e.g., misinformation).

A Brief History of Digital Serendipity Traditionally, most of the studies on serendipity have focused on scientific discovery and innovation studies (Merton and Barber 2006; Foster and Ellis 2014). According to Merton and Barber (2006), serendipity is one of the main forces that has steered the progress of science. There are numerous renowned examples of serendipitous scientific discoveries, including penicillin, DNA, string theory and mirror neurons. The role of serendipity in science is undisputed (see also Copeland, Chap. 6 and Sands and Chiapperino, Chap. 10, this volume).

4

Infotainment, a portmanteau of information and entertainment, refers to ‘soft news’ that aims to attract viewers with short attention spans. More often than not, such information is centered on celebrity culture and laden with infotainment content that tends to simplify and sensationalise information.

8 Serendipity as a Design Principle of Personalization …

149

Over time, the interest for serendipity grew also in the realm of library and information science, initiated first by Bernier (1960), as well as in the realms of art and computing with the famous 1968 Cybernetic Serendipity exhibition at the Institute of Contemporary Arts in London. Early practical work on serendipity was done in relation to browsing Online Public Access Catalogs (O’Connor 1988). Functionality for supporting serendipitous discovery included browsable search indexes and similar article citation retrieval (Rice 1988). Eventually, serendipity became a subject in information studies, design theory, digital libraries and learning theories (Race and Makri 2016; McCay-Peet and Toms 2017, Burrows and Verhoeven, Chap. 3, this volume). With the advent of the Internet, it was acknowledged that computer search tended to facilitate purpose-driven information seeking instead of opportunistic and accidental finding—with the latter more typical when browsing printed information.5 Before the emergence of search engines, the journey on the Web was initially intended as discovering what was out there—accidentally—not on finding specific content.6 The advent of systematizing algorithms led journalist Ted Gup to argue that we would have experienced “[t]he end of serendipity” (1997). Social media, however, are much more serendipitous than search engines: you don’t know what you will encounter and the quantity of content you are exposed to will likely make you stumble upon serendipitous content. Consider, in fact, that the main reason why Facebook users consume news on the platform is actually incidental: 32% see news while they are there for other reasons (Andi 2021). The problem is that editorial choices in legacy media recognised a tension between a generalised relevance—what people are expected to want—and serendipity—what people may like— whereas algorithmic personalization can shift this balance from a more generalised relevance to a more individually personalised one. A trade-off between accuracy and serendipity can also emerge in recommender systems (RSs) (Kotkov et al. 2020). Furthermore, with the development of the Web 2.0, a number of social media platforms monopolised the market and heavily influenced worldwide information consumption, notably by favoring a more individualistic and attention-capturing personalization. As a reaction, some scholars metaphorically advocated to ‘save serendipity’ as a fundamental experience for individual progress (Meckel 2011) and to ‘cultivate serendipity’ as a necessary experience for democracies (Sunstein 2017a, b). Ethan Zuckerman (2013) also stressed how the development of a cosmopolitan culture through serendipitous encounters could be limited by georeferenced algorithms, which personalise content based on one’s location. Morozov (2012) also feared the “death of the cyberflaneur”—the user who aimlessly surfs the web – and

5

In a sense, this observation is intimately connected with the provocation of the great artist Pablo Picasso who famously said already in 1968: “Computers are useless. They can only give you answers”. 6 As a matter of fact, the metaphor ‘surfing the Internet’ was chosen to refer to a fun feeling and “something that would evoke a sense of randomness, chaos, and even danger” (Polly 1992).

150

U. Reviglio

that "playfulness, intrigue, and serendipity" would vanish because of mainstream social media. Over time there have been an increasing interest and recognition of the value of serendipity. Attempts to pursue serendipity as a design goal include, amongst others, via information retrieval (e.g. Campos and Figuereido 2002), social media (e.g. Burkell et al. 2012), music discovery (e.g. Zhang et al. 2012), business and ICT development (e.g. Saini and Khurana 2013), information behavior strategies to enhance online discovery, in everyday searches (e.g. Buchem 2011) or even in academic scholarly research (e.g. Maloney and Conrad 2016, Burrows and Verhoeven, Chap. 3, this volume). Also, research on how to design RSs ‘beyond-accuracy’ metrics received increasing interest (Kotkov et al. 2020). The Internet more generally, and personalization systems more specifically, could be both quantitatively and qualitatively much more serendipitous than they currently are. Under the current business model of mainstream personalised services, personalised relevance and quantity (i.e. engagement optimization) outstrip serendipity and quality, providing more an ‘illusion of serendipity’ (Erdelez and Jahnke 2018). This also stems from the limited opportunities for users to control design choices and expressed preferences, and to proactively navigate information.

Digital Serendipity and Its Nuances Given the interdisciplinary, multifaceted and elusive nature of serendipity, different disciplines and scholars have interpreted and employed the concept of serendipity in digital environments in different ways. In serendipity studies, diverse and at times unclear conceptions have been developed with overlapping meanings. For this reason, I will first propose a taxonomy of data-driven serendipity based on an interdisciplinary literature review. I will begin by discussing this subject from three intertwined but different perspectives: theoretical (serendipity from an epistemological perspective), technical (serendipity as a design principle) and educational (serendipity as a capability).

Serendipity from a Theoretical Perspective Serendipity is a fascinating and powerful idea related to several renowned philosophical concepts such as Aristotle’s poiesis, Jung’s synchronicity, Pierce’s abduction or Nietzsche’s amor fati. At the same time, it can also be considered as a philosophical attitude that values curiosity, creativity, diversity, lateral thinking, accidentality, luck, randomness, mistakes, glitches, the unexpected and the challenging and, more generally, the ability to observe and the pleasure derived from experiencing all these. It is indeed relatable to virtue ethics and epistemology as well as to the idea of “upgrading”, that is, “turning information into knowledge” (Floridi 2011).

8 Serendipity as a Design Principle of Personalization …

151

Serendipity sheds light on the desirable opportunity to preserve and cultivate a degree of chance opportunities in life, more generally, and in personalization practices, in particular: for something that is not too personalised and thus truly unexpected. Though pleasurable and useful, the main risk of excessively accurate personalization systems is indeed a redundancy of information, which can ultimately limit the breadth and depth of the information we are exposed to. It becomes therefore relevant to cultivate information experiences that go beyond deterministic goal-driven ones. McKay et al. (2020) call this endeavor ‘bounded chaos’. Computer scientists and philosophers have also already argued that it will be necessary to leave some things open to chance in future algorithms (Domingos 2015; Stiegler 2017). After all, because of evolution and the ability to adapt to environmental changes, human beings can be considered as “antifragile” (Taleb 2012): They benefit and get better from shocks, randomness and uncertainty. Personalization algorithms might indeed overly reduce pain and uncertainty (Gal 2017), make people more conservative and, eventually, create only an illusion of knowledge, mainly sustained by selective exposure and confirmation bias. Valuing serendipity helps to internalise the significance of the ‘unknown unknown’: what we don’t know that we don’t know. This approach values intellectual humility, tolerance towards uncertainty, but also audacity (Reviglio 2019). Serendipity-aware digital environments can also stimulate exploratory information behavior. Serendipity thrives on curiosity, on a will to navigate information proactively but also with a prepared mind for the unexpected, even if potentially challenging and undesirable (see also Arfini, Chap. 7, and Ross, Chap. 9, this volume). This is true in general, but it applies also to the context of political news consumption and democratic deliberation (see §5.2 Political Serendipity). Ultimately, serendipity has a semantic value. It is a welcome addition to our current conceptual toolbox, which may be insufficient to address unprecedented digital challenges. As Luciano Floridi argues (2015), this is a risk because “the lack of a clear conceptual grasp of our present time may easily lead to negative projections about the future: we fear and reject what we fail to semanticise” (p.3). In this sense, serendipity can represent a common good as well as a narrative for information societies in which most information is automated and tailored to previously assumed individuals’ preferences (Reviglio 2019).

Serendipity as a Design Principle The theoretical interpretations of serendipity are relevant, yet serendipity in digital environments is mainly conceived of as a technical challenge. Design or programming for serendipity indeed sounds like an oxymoron and, to some extent, it is. Serendipity is, in fact, a subjective experience that is only discernible in hindsight and for which the unexpected plays a fundamental role. As said, while natural, “pure serendipity” is not amenable to generation by a computer (van Andel 1994), artificial serendipity can actually be cultivated by creating opportunities for it through

152

U. Reviglio

the design of physical, digital, and learning environments (Race and Makri 2016; de Melo 2018). Fundamentally, the design for serendipity in digital environments occurs on two levels; one that is invisible—algorithmic—and one that is apparent—interface design. On the one hand, the complex ‘recipe’ for automating serendipity algorithmically remains open-ended. In the RSs community, there is no consensus on what serendipity means, probably due to partly-overlapping concepts making definition difficult. For example, some consider serendipity to be relevant and unexpected, for others it is novel and unexpected. Kotkov et al. (2020) propose serendipitous items to be relevant, novel and unexpected at the same time. In addition, they could also be more diverse.7 Serendipitous items in RSs have also narrowly focused on increasing user satisfaction, broadening user preferences and overcoming the over-specialization problem. Yet, these could also expand a user’s social network and knowledge, promote belief change and, more generally, educate users. Broader conceptions of serendipity could make serendipity in RSs more aligned to the above theoretical interpretations. On the other hand, serendipity is considered an affordance, an opportunity that the environment offers to the user who is able to grasp it, as an intuitive invitation by design (Gibson 2014). At the same time, there are affordances for serendipity itself (Björneborn 2017). The main challenge is how to operationalise these ideal conditions. Even though there is no commonly shared set of requirements for a system to facilitate serendipity in digital environments—and probably there will never be— most scholars agree that the main functional requirements are to enable the “anomaly” or the chance encounter to trigger an event, and to support the user in “connecting the dots” so as to reach a significant surprise outcome (McCay-Peet and Toms 2017). This can be done in several ways. To provide some general approaches: increasing the diversity, unexpectedness, relevance and novelty of information; cultivating weak ties—to connect with people that connect us to other, more diverse social networks; improve information explorability and information findability; exposing users to (and equally, affording them the ability to expose themselves to) (pseudo)random information; introducing noise and uncertainty and also valuing what in statistical jargon are called outliers—deviant cases that do not follow the statistical model—in order to take into account the “long tail content” of the system.8

7

In RSs discussions these metrics are generally intended as following: • Novelty is when an item is unknown to the user. • Relevance is when an item is useful or meaningful to the user. • Unexpectedness is when an item is not intentionally looked for by the user or it is different from the user’s expectations. • Diversity is when there is difference between the current recommendation and the user’s profile or the system’s prior recommendations. 8 In e-commerce, the long-tail refers to the idea that there are a lot of unpopular products that can be offered to the interested ones. Likewise, the long tail content refers to the "lesser known" content in the system, that is less popular and therefore seen by less users. It is a form of hyper-personalization of unpopular items that could be serendipitous to a user.

8 Serendipity as a Design Principle of Personalization …

153

Furthermore, the value of designing for serendipity has to be considered in the light of future cultural and technological developments. It is likely that most people will spend more and more time exposed to more and more personalised information. Advances in artificial intelligence have the potential to enable ever more sophisticated recommendations, whereas virtual reality, augmented reality and the Internet of Things will continue to blur the distinction between online and offline, giving rise to the vision of ‘ambient intelligence’,9 or even leading to a ‘social physics’10 (Pentland 2015). In sum, serendipity as a design principle is a technical challenge that particularly concerns designers, data and computer scientists whose intentions, however, can be various and might not always match the ideal of serendipity as a common good (Smets 2022).

Serendipity as a Capability Serendipity is also intended as a capability, namely “the ability to recombine any number of observations and deduce ‘matching pairs’, or sets of events, that appear to be meaningfully related” (de Rond 2014, p.344). It is arguably a desirable educational goal: a serendipity mindset can indeed stimulate proactive information seeking as well as sustain creativity and innovation. Sagacity and curiosity are indeed often acknowledged as fundamental virtues for experiencing serendipity. The ability to see things from different perspectives but also to ‘see ourselves see differently’ are also fundamental (Lotto 2017). There is, moreover, a strong correlation between a serendipity mindset and the psychological trait of openness to experience (Heinström 2010). More generally, the experience of serendipity is strictly related to a series of intertwined attitudes and skills that can be eventually cultivated in education, such as skepticism, intellectual humility, belief revision ability and tolerance towards uncertainty. Importantly, research also investigates serendipity as a capability for selflearning—also referred to as “accidental learning” (Kop 2012). In this context, the role of serendipitous news consumption online (e.g., Incidental Exposure to Online News) is investigated. This work explores how and why individuals obtain information opportunistically, rather than purposively, as well as what are the effects of such behavior for individuals and public opinion (Yadamsuren and Erdelez 2016). As said above, a serendipity mindset can help individuals to navigate information overload more resiliently. In the informal learning that derives from human– computer interactions, it also represents a powerful soft-skill and, more generally, 9

Ambient intelligence refers to the eventual future vision in which automatic smart online and offline environments interact with each other and take an unprecedented number of decisions for us and about us in order to cater to our inferred preferences. It may actually represent a new paradigm in the construction of knowledge (Hildebrandt and Koops 2010). 10 Pentland (2015) anticipates the emergence of a Comtian “social physics”, the possibility to create a data-driven mathematical model of social behaviours from big data. The broader challenge is indeed to design serendipity at the intersection of physical and digital environments.

154

U. Reviglio

a new fundamental idea that pushes educators to rethink the current approaches to education. For example, emerging theories of learning such as ‘connectivism’ acknowledge that learning is becoming a continual process where control is shifting from the tutor to an increasingly more autonomous learner that is not even entirely in control (Siemens 2014). In this context, a serendipity mindset is welcome, whereas the design for serendipity can also help to refine techniques (such as gamification) to even improve learning itself. More generally, scholars have already remarked that the researcher of the future—and analogously the citizen of information societies—has to be able to work in several fields over time and to find similarities between apparently heterogeneous concepts. Serendipity indeed more often occurs among interdisciplinary scholars (Foster and Ellis 2014) and, similarly, boosts interdisciplinary research (Darbellay et al. 2014). Serendipity as a capability ultimately represents a desirable and innovative educational endeavor to cope with our complex, fast-paced and information-rich digital societies.

Two Routes to Design for Digital Serendipity Not all serendipitous experiences are triggered the same way, of course. Today’s Internet mainstream services certainly provide serendipitous recommendations. Yet, they tend to employ convergent systems—the capacity to discover the right thing at the right time, to cater to the user’s perceived intentions, interests, tastes—rather than divergent systems—increasing the diversity of information in order to expand user’s horizons and to help them make surprising discoveries (de Melo 2018). These are related to two ideal conceptions to design for serendipity, two opposing ideal routes to cultivate serendipitous encounters, what I define here as “hyper-personalised serendipity” and “pseudo-personalised serendipity”. These are two extremes on a continuum based on the degree of profiling reliance and, therefore, individuals’ preferences predictability. In other words, these concepts refer to the extent to which personalised content is accurate and based on individual’s supposed preferences. This shall be further elaborated in the following section.

Hyper-Personalised Serendipity Hyper-personalization is generally intended as the use of data to provide more personalised and targeted products, services and content. It is the paradigm of convergent systems. To give an example: one watches a movie and while scrolling a social media feed is recommended a review of that particular movie. That would not be accidental but predictable. Hyper-personalised recommendations could be timely, useful, meaningful and surprising. Hyper-personalised serendipity represents what has been termed as “serendipity on a plate” (Makri et al. 2014), namely highly personalised information mainly based on implicit personalization—behavioral data

8 Serendipity as a Design Principle of Personalization …

155

inferred from previous activity. The same idea has been expressed by former Google CEO Eric Schmidt, when he envisioned a future where people and technology come together to create a serendipity engine, “where you don’t even have to type” (2006). This conception of serendipity is more akin to RSs with commercial objectives, which have been the main focus of research in this area these last 20 years (Milano et al. 2020). It is primarily driven by principles of delegation and persuasion on the one hand, and efficiency and expertism on the other. Indeed, it can also help people to specialise in, and deepen their knowledge of certain, perhaps rather specific, subjects. Eventually, it presents a trade-off with privacy. As will be discussed later on, excessively relying on this approach could also reduce individual media diversity (e.g. creating filter bubbles and echo chambers), offer a limited experience of serendipity, and negatively influence the formation of preferences and opinions.

Pseudo-Personalised Serendipity Pseudo-personalised recommendations are those recommendations that rely less on preferences implicitly and individually inferred (e.g., collaborative filtering, which predicts the interests of a user by collecting preferences information from many users). Therefore, while they might be less accurate in matching users’ presumed preferences, they might lead to more unexpected serendipitous encounters. To some extent, serendipity is thereby already an implicit by-product of current personalization practices, for they are probabilistic and prone to misjudgment. Also, this limitation of RSs has long been acknowledged, and it originates from the analogous “over-specialization problem”, which occurs when systems only recommend items akin to those already favored by the user. Nonetheless, pseudo-personalization ideally represents a more hazardous attempt to trigger serendipitous encounters: On the one hand, from an economic perspective, it could decrease user’s engagement and therefore go against companies’ pursuit of profit maximization. On the other hand, from an individual perspective, it could be distracting and frustrating and, ultimately, undermine the original attempt to reduce information overload. Nevertheless, considering how personalization is usually designed to filter vast amounts of information, applying more ‘pseudo-personalised’ recommendations to a number of items can be beneficial. One of the main goals of pseudo-personalised recommendations is to overcome the potential determinism implicit in profiling—which can result in redundancy and limited content discovery.

156

U. Reviglio

Individual and Political Serendipity: Two Sides of the Same Coin In the previous literature review on the apparent threat to or loss of serendipity, I have identified two main conceptions of serendipity: individual and political serendipity, which respectively refer to the exposure to new ideas and potentially new interests from outside one’s filter bubble, and facts, views and arguments from outside one’s echo chambers.

Individual Serendipity Individual serendipity can be intended as the exposure to information that intersects one’s digital profile, so that one can discover new topics, ideas and, eventually, interests with greater likelihood. It is generally related to the filter bubble problem. This notion also overlaps with the concepts of ‘content diversity’, ‘source diversity’ and ‘cultural diversity’, thus generally referring to the media policy goal to consume a diverse information diet.11 In the context of RSs, it widely overlaps with serendipity as a metric to assess the quality of recommended items (Kotkov et al. 2020). Individual serendipity may not be particularly problematic as long as the system adapts to the individual’s serendipitous disposition. Individuals, however, could be misinterpreted and, eventually, discriminated against. To begin with, there might be a mere information asymmetry. A user could think they are being exposed to decent amounts of serendipity while still being exposed to a limited portion of the serendipity his/her social networks can actually offer. Relatedly, there is a growing epistemic and social connectivity inequality: A privileged group of users has higher (digital) literacy and knowledge, as well as more diverse social networks, is effectively able to strike a balance between relevance and serendipity, and the RSs would indeed adapt to their serendipitous proclivity. In contrast, a larger group of users risk exposure to only a minimum, qualitatively inferior pool of information, without even noticing. Similarly, there might be another asymmetry between ‘first’ and ‘second order preferences.’12 Individual serendipity can help to tackle the mismatch between the necessity for the “aspirational self” to uncover and the intrinsic identity reductionism of profiling. Designing for individual serendipity affords users the ability to both manage their information consumption for aligning it to their preferences and yet ultimately sustain the critical capacity to imagine themself as a different person. 11

There is a significant debate on the conceptualization and operationalization of media diversity and its various sub-dimensions in various disciplines (see Helberger 2011; Hoffman et al. 2015; Loecherbach et al. 2020; Hendrickx et al. 2020). 12 Individuals have different “orders” of preferences; “first-order preference” is expressed in how we behave in the moment that a stimulus or temptation affects our consciousness. In contrast, “second-order preference” is the choice we make for ourselves upon further reflection, generally separated from the immediate temptation and it has to do with an “aspirational self”.

8 Serendipity as a Design Principle of Personalization …

157

Affordances to support individual serendipity are, arguably, limited only by imagination, yet some may be more serendipitous than others, and indeed there are attempts to systematise effective affordance categories (Björneborn 2017).

Political Serendipity Political serendipity can be seen as the attempt to cultivate politically constructive online debates by nudging users to experience information that is ethically challenging and politically transformative. As such, it mainly concerns news and platform RSs, and the risks deriving from so-called echo-chambers. Generally speaking, its design can be intended in two main ways: (1) preventive, for example balancing a user’s political information consumption (i.e. increasing diversity exposure) mainly to prevent political polarization, and (2) proactive, for example supporting conflict transformation in specific contexts in order to reduce political polarization (i.e. depolarization). Similarly, political serendipity can be intended as the right to be confronted with challenging viewpoints. The design for political serendipity aims to cultivate a set of democratic principles. Indeed, political serendipity can be implicitly valued by all democratic theories, even though for different reasons (Bozdag and Van den Hoven 2015; Helberger 2019). For example, in the liberal model of democracy, its value is to be found particularly in the access each individual should have to the ‘marketplace of ideas’, while in the deliberative model it is valued especially for the promotion of a common ground for discussion (Helberger 2019). In current media policy discussions, political serendipity is to be found particularly in the concept of ‘viewpoint diversity’ (Helberger 2011; Hoffmann et al. 2015).13 In this context, it is fruitful to apply the introduced distinction between hyperpersonalised and pseudo-personalised serendipity. Political serendipity can indeed be achieved in a hyper-personalised manner in order, for example, to facilitate belief change (e.g. political microtargeting) or in a pseudo-personalised manner so as to better preserve individual autonomy. On the one hand, hyper-personalised political serendipity is concerning for a number of reasons. First, it can become intolerably manipulative or paternalistic. The (presumed) most persuadable individuals can indeed be targeted in alarming ways (e.g., the Cambridge Analytica scandal). Second, it can be ineffective or even worsen polarization. In fact, under certain conditions politically polarised individuals could radicalize even further (e.g., backfire effect). Yet, a reasonable degree of hyperpersonalised political serendipity could empower users and even help to prevent them from radicalising in the first place. For example, one can expose users to scientific

13

‘Diversity’ in media discussions has several sub-dimensions, among them, media ownership diversity, source diversity, content diversity, cultural diversity, viewpoint diversity, exposure diversity and diversity experience (what are you exposed to and what you actually consume).

158

U. Reviglio

facts that debunk a specific false conspiracy theory, or if a previously consumed news has been found inaccurate or misleading the system can expose users to corrections. On the other hand, pseudo-personalised political serendipity can be reached through increasing exposure diversity, but also by removing polarizing content or amplifying civil arguments and demoting uncivil ones (Helberger 2011; Hoffmann et al. 2015; Stray 2021). It can also better preserve autonomy by educating users to retrieve information from more and diverse sources and to engage in more constructive online debates by providing them tools, filters and affordances by design. There are plenty of practical or experimental examples. For instance, Cass Sunstein (2017b) proposed that social media could afford a “serendipity button” for news and opinions, allowing people to opt in, especially during electoral times. Related stories at the bottom of a post can indeed help in counteracting misinformation or simply enriching a user perspective in a serendipitous way (Bode and Vraga 2015). Similarly, plug-ins such as Flipfeed afford the possibility to scroll the feed of a random individual from another ideological background, whereas social media news aggregators such as Gobo offer sliders which can be controlled by users to filter information (e.g. political perspectives from left to right). These are all examples of ‘perspective widening tools’ (Delacroix and Veale 2019) that, eventually, would support pseudo-personalised political serendipity.

Discussion Personalization algorithms employed in social media are revolutionizing the processes of formation of individual’s preferences and identities, of public opinon and, more broadly, of culture. Yet they can embed developers’ biases, reproduce power structures and be exploited to manipulate, alienate and deprive people of their dignity. Eventually, they are able to suppress the aspirational and relational self, thus individual progress and societal cohesion. It is indeed necessary to better understand the politics and epistemology behind such technologies in order to avoid their use as weapons of mass distraction (Chomsky 2002), or to amuse ourselves into a “culturedeath” (Postman 1986). Personalization systems—especially these of mainstream social media—can indeed be used as tools of control, oppression, isolation, and entertainment as much as tools of emancipation, progress and self-actualization. Hyperpersonalized serendipity, for example, contributes to make the Internet an addictive, manipulative and, ultimately, relatively unquestioned machine (see also Erdelez, Chap. 12, this volume), whereas the design for pseudo-serendipity can help innovate our perceptions. Neuroscience is indeed demonstrating how adaptive change, either individual or social, is possible by adding ‘noise’ (e.g., random elements) to a system, and then by celebrating uncertainty and openness to possibilities, adopting a playful attitude (Lotto 2017). Similarly, a thoughtful design of serendipity can also contribute to the development of a ‘hyper-culture’, a global culture in which “cultural spaces overlap and penetrate each other”, without homogenizing themselves (Han 2022).

8 Serendipity as a Design Principle of Personalization …

159

The presented taxonomy, and more generally the design for serendipity, offer fertile perspectives to understand current challenges and opportunities in the development of personalization systems in digital environments. The need for serendipity, above all, highlights the fact that personalization algorithms are unable to perfectly predict individuals and their preferences. Personalization algorithms indeed usually rely on the economic paradigm of “revealed preferences”: they assume people know what they want, which is often untrue (Thornburn et al. 2022). They are also essentially behaviorist, in the sense that they only analyze a limited set of quantifiable users’ behaviors to influence their future behaviors (Greene and Shmueli 2019). These theoretical premises are reductionist. They stem from a commercial view that pretends that everything is calculable and thus uncritically relies on data and algorithms as a source of valuable insights. Yet, there will never be perfect personalization systems, as human beings are ultimately undetermined and even strive to be so. The design for serendipity could thus help to further develop responsible personalization systems in which identity is considered in its complexity, dynamicity and multiplicity—not something reduceable to a single profiling driven by psychometric, real-time, and behavioral data (Hildebrandt 2017). Personalization systems, in fact, nowadays occur mainly beyond the control of users as they are mostly based on implicit inferences rather than on deliberate and expressed preferences. It is, therefore, of paramount importance to balance these two routes (implicit and explicit personalization) to model the preferences of users and nudge them to consciously align these to eventually support self-actualization (Sullivan et al. 2019). Hyper-personalization is a natural feature of RSs and other personalization systems, and can certainly trigger serendipitous encounters. It is essential for social media platforms, and as a tool for news organizations and companies to (re)connect with audiences. However, encountering mostly hyper-personalised information may only create an illusion of serendipity (Erdelez and Jahnke 2018). Hyperpersonalization follows the principle of kairos14 —one of the principles of captology (the study of computers as persuasive technologies)—that refers to the ability to provide the right content to the right person in the right moment (Fogg et al. 2002). Of course, if such kairos is benevolent, it could be immensely useful and serendipitous. Otherwise it could be exploited in a manipulative manner, for example, by creating compulsion loops (Deibert 2019). These could work via a variable or random ratio reinforcement, in which rewards are delivered in an unpredictable or random fashion. The random character of a reward can indeed act as a positive reinforcement, as behaviorists demonstrated long ago. The order of the content can indeed affect usage times and serendipity experiences (Gundlach et al. 2022). Thus, hyper-personalised serendipity can be purposefully offered among predictably uninteresting content in a way that increases engagement. In this case, it would be possible to talk of ‘manipulative serendipity’. This would hardly lead to reliable and meaningful serendipitous experiences. Instead, it would more often result in the consumption of ephemeral, sensational, misleading, and false content. To date, this is more of a speculation 14

Kairos is an ancient Greek word that traditionally is associated with the homonymous Greek divinity, personification of the “opportune moment”.

160

U. Reviglio

which highlights the need for users to critically evaluate the quality of personalized information to avoid manipulation. Erdelez et al. (2019), for example, have developed a test to differentiate between genuine and constructed serendipity, namely if it is a result of ‘chance’ (i.e. pseudo-personalised serendipity) or a result of algorithmic interference (i.e. hyper-personalised serendipity) (see also Erdelez, Chap. 12, this volume). This can represent a fruitful step for the cultivation of serendipity as a capability. Conversely, pseudo-personalised serendipity can trigger more serendipitous encounters (greater in quality, rather than number). At a theoretical level, its pursuit underpins several intertwined assumptions (Reviglio 2019). It acknowledges that wisdom (e.g. for selecting the most serendipitous information) is something that should not be fully and uncritically delegated to machines. One of the main consequences of these assumptions is that in order to properly sustain pseudo-personalised serendipity, users must be always capable of acting upon profiling construction, personalization choices and eventually opt-out from any algorithmic curation. Design and algorithms always embed certain values—they are never neutral. In order to cultivate serendipity responsibly, designers should, above all, (1) balance pseudo-personalization and hyper-personalization—both algorithmically and by interface design—(2) take into account both individual and, if needed, political serendipity, and (3) prevent manipulative serendipity. Without such distinctions, I argue, the design for serendipity may become unethical. Otherwise, with its superficially positive connotation, serendipity might be easily commercialized (Smets, 2022), or even become a design principle for ethics blue washing,15 for example by claiming to design serendipity to combat filter bubbles and echo chambers while actually designing hyper-personalised and manipulative serendipity. Admittedly, however, it is the business model of current mainstream social media based on “engagement optimization” that ultimately starves the blossom of serendipity in many ways. For example, the goal to keep users inside their “walled-gardens” inevitably decreases the chances users have to serendipitously surf the web and eventually discover new sources of information. The presented taxonomy supports the development of pragmatic methodologies that make serendipity an integral part of research and development in digital environments, especially personalization systems. The outlined concepts can act as normative constraints, supporting the move from the ‘what’ (ethical principles) to the ‘how’ (technical requirements), offering a more granular approach to cultivating serendipity. To formalize such a normative ideal, conceptual clarity is fundamental at all levels. And to achieve such clarity, interdisciplinary communication and collaboration are essential. Serendipity has to be further promoted to influence practitioners’ understanding of the phenomenon, from designers to engineers. Importantly, in the RSs community a new and broader conception of serendipity could be developed 15

“Ethics blue washing” is defined as “the malpractice of making unsubstantiated or misleading claims about, or implementing superficial measures in favour of, the ethical values and benefits of digital processes, products, services, or other solutions in order to appear more digitally ethical than one is (Floridi 2019, p.3)”.

8 Serendipity as a Design Principle of Personalization …

161

and, eventually, even become an additional metric to assess the ethics of personalised media systems. Ultimately, the design for serendipity can also sustain the fundamental right to (receive) information.16 This theoretical framework resonates with the core function of traditional media, that is to provide ‘reliable surprises’ in order to balance familiarity and chaos (Schönbach 2007), and the function of public service media to provide universalism, quality and serendipity17 . Likewise, it resonates with the dynamic relation between randomization, personalization and generalization suggested by Carr to cultivate serendipity in libraries (2015), as well as the balance between convergent and divergent systems in digital environments (de Melo 2018). Moreover, it can also help to reconceptualize the public service media principle of “universalism” which aims to provide a wide range of content that is relevant, diverse, and inclusive (see Savage et al. 2020)18 . Ultimately, an ‘institutionalised serendipity’ (Merton and Barber 2006) could enable the construction of agonistic personalization systems that account for the multiplicity of our identities (Hildebrandt 2017) and that eventually offer ‘perspective widening tools’ (Delacroix and Veale 2019). To further develop this theoretical framework and taxonomy, future studies could, among many research directions: corroborate the concepts identified in this taxonomy, including the differences between hyper- and pseudo-personalization as well as the existence of manipulative serendipity; assess the capacity of serendipity to help to burst filter bubbles and weaken echo chambers; find a consensus of what serendipity really means in (news) recommender systems studies; better understand how design affordances support serendipity; improve measurement-techniques; explore the relationship between serendipitous personalization and engagement; develop auditing techniques that provide more insights into black-boxed algorithms. 16

Article 10 of the European Convention on Human Rights can prove an important point of departure to realise democratic values in the personalised media landscape (Eskens et al. 2017). This also declares that: Everyone has the right to freedom of expression. This right shall include freedom to hold opinions and to receive and impart information and ideas without interference by public authority and regardless of frontiers. This article shall not prevent States from requiring the licensing of broadcasting, television or cinema enterprises.

17

The first director-general of the BBC John Reith argued that public service media should be targeted at everybody, irrespective of their status or location, with a mixed programming created to the highest possible standard, a miscellany of genres in which every listener could find his or her particular interests. Ultimately, he believed that “few know what they want, and very few know what they need. […] In any case it is better to over-estimate the mentality of the public than to under-estimate it” (Reith 1924, p.34 in Savage et al. 2020). 18 Universalism “indicates an assumption and claim that all the diversity of reality as a whole can be traced to a single principle or law of order. From this, it follows, that ideas, ideals, rights, and obligations apply in principle to all human beings. Universalism is therefore a perspective that prioritises the whole of an entity above singularity, and generality above specificity. The concept contends that some principles and norms are valid for all human beings.” (Thomas 2020, p.26 in Savage et al. 2020).

162

U. Reviglio

Furthermore, the opportunity to educate for and develop a serendipitous mindset is significant. Cognitive factors involved in the serendipitous process are another promising research area in this context (eg. Arfini, Chap. 7, and Ross, Chap. 9, this volume). The presented taxonomy might stimulate these lines of research which can advance research on it in return.

Conclusions In this chapter I have investigated the phenomenon of serendipity in digital environments to map the conceptual space of ‘digital serendipity’. In this pursuit, I carried out an interdisciplinary literature review and discussed the most significant conceptions of serendipity for digital environments. Digital serendipity is indeed a powerful philosophical concept, a technically volatile and conceptually nuanced design principle, and a fundamental capability for citizens of information societies. Essentially, the design for serendipity – especially in news and platforms personalization systems – helps to navigate more effectively the breadth and depth of all the available information and, eventually, widen our political perspectives, increase our individual discoveries and even improve our learning experiences. As serendipity becomes widely acknowledged as a desirable design principle, it is essential to keep asking what we mean by serendipity in different contexts, how to design for it and how much of it is desirable. In order to increase serendipity in digital environments, designers should recognise a number of conceptual nuances and dynamics: not only the potential trade-off between serendipity and accuracy but, eventually, the tensions between implicit and explicit personalization, and between hyper-personalised and pseudo-personalised recommendations. Moreover, it has to be acknowledged how serendipity can be faked. As many research gaps still remain in the study of digital serendipity, the taxonomy developed in this article provides only a point of departure for one that can cultivate a more ethical design for serendipity in digital environments. To achieve this, it is essential to foster inter-disciplinary communication between political philosophy and design, media, algorithmic, information and serendipity studies. To conclude, personalization algorithms and the serendipity they afford are fundamental for epistemic well-being and epistemic justice. They can indeed be a social force for individual and social change. And yet, these systems can be used both as private tools of oppression and as democratic tools of emancipation. The plea of this preliminary theoretical framework is indeed to stimulate a more critical discussion on how to design personalization systems that recognise and ultimately sustain serendipity in all its beneficial nuances. Acknowledgements Research for this article has been partially funded by the project "“Public Perception of Algorithms in Society: Accounting for the Algorithmic Public Opinion—ALGOCOUNT” (2020–2022), funded by Fondazione Cariplo. More information is available at: www.alg ocount.org.

8 Serendipity as a Design Principle of Personalization …

163

References Abbott, A. 2008. The traditional future: A computational theory of library research. College and Research Libraries 69 (6), 524–545. Andı, S. 2021. How and why do consumers access news on social media. Reuters Institute Digital News Report, 2020–06 Björneborn, L. 2017. Three key affordances for serendipity: Toward a framework connecting environmental and personal factors in serendipitous encounters. Journal of Documentation 73 (5): 1053–1081. Bogers, T., and L. Björneborn. 2013. Micro-serendipity: Meaningful coincidences in everyday life shared on Twitter. iConference, 196–208 Bode, L., and E.K. Vraga. 2015. In related news, that was wrong: The correction of misinformation through related stories functionality in social media. Journal of Communication 65 (4): 619–638. Bozdag, E., and J. van den Hoven. 2015. Breaking the filter bubble: Democracy and design. Ethics and Information Technology 17 (4): 249–265. Bruns, A. 2019. Filter bubble. Internet Policy Review 8 (4). Buchem, I. 2011. Serendipitous learning: Recognizing and fostering the potential of microblogging. Form@ re-Open Journal per la formazione in rete 11 (74), 7–16 Campos, J., and A.D. Figueiredo. 2002. Programming for serendipity. In Proceedings of the AAAI Fall Symposium on Chance Discovery—The Discovery and Management of Chance Events Carr, P.L. 2015. Serendipity in the stacks: Libraries, information architecture, and the problems of accidental discovery. College and Research Libraries 76: 831–842. Chomsky, N. 2002. Media Control: The Spectacular Achievements of Propaganda. Seven Stories Press. Darbellay, F., Z. Moody, A. Sedooka, and G. Steffen. 2014. Interdisciplinary research boosted by serendipity. Creativity Research Journal 26 (1): 1–10. Delacroix, S., and M. Veale. 2019. Smart technologies and our sense of self: Going beyond epistemic counter-profiling. Law and Life in the Era of Data-Driven Agency, O’Hara & Hildebrandt (eds.) de Melo, R.M.C. 2018. On serendipity in the digital medium: Towards a framework for valuable unpredictability in interaction Design. Doctoral dissertation. Portugal: Universidade do Porto. De Rond, M. 2014. The structure of serendipity. Culture and Organization 20 (5): 342–358. Domingos, P. 2015. The Master Algorithm: How the Quest for the Ultimate Learning Machine will Remake Our World. New York: Basic Books. Edward Foster, A., and D. Ellis. 2014. Serendipity and its study. Journal of Documentation 70 (6): 1015–1038. Erdelez, S., and I. Jahnke. 2018. Personalized systems and illusion of serendipity: A sociotechnical lens. In Workshop of WEPIR 2018. Erdelez, S., Agarwal, N.K., and I. Jahnke. 2019. Serendipity and critical thinking: Fighting disinformation in a socio-technical society. In SIG Social Informatics Research Symposium, 82nd ASIS&T Annual Meeting. Fletcher, R., and R.K. Nielsen. 2018. Automated serendipity: The effect of using search engines on news repertoire balance and diversity. Digital Journalism 6 (8): 976–989. Floridi, L. 2015. The On Life Manifesto. Cham: Springer. Floridi, L. 2016. Tolerant paternalism: Pro-ethical design as a resolution of the dilemma of toleration. Science and Engineering Ethics 22 (6): 1669–1688. Floridi, L. 2019. Translating principles into practices of digital ethics: Five risks of being unethical. Philosophy and Technology 32 (2): 185–193. Fogg, B.J., E. Lee, and J. Marshall. 2002. Interactive Technology and Persuasion. The Handbook of Persuasion: Theory and Practice. Thousand Oaks: Sage. Gal, M.S. 2017. Algorithmic challenges to autonomous choice. Michigan Telecommunications and Technology Law Review. Gandini, A., Reviglio, U., and S. Keeling. (forthcoming). Conceptualizing the Algorithmic Public Opinion.

164

U. Reviglio

Greene, T., and G. Shmueli. 2019. How Personal is Machine Learning Personalization? arXiv preprint arXiv:1912.07938. Gibson, J.J. 2014. The Ecological Approach to Visual Perception. Classic. Hove: Psychology Press. Gundlach, J., Baumann, A., and H. Krasnova. 2022. Coming Back for More: The Effect of News Feed Serendipity on Social Networking Site Usage. In PACIS 2022 Proceedings, 271. Gup, T. (1997). Technology and the end of serendipity. The Chronicle of Higher Education, 44(21) Heinström, J. 2010. From Fear to Flow: Personality and Information Interaction. Elsevier. Han, B.C. 2022. Hyperculture: Culture and Globalisation. John Wiley & Sons. Harambam, J., N. Helberger, and J. van Hoboken. 2018. Democratizing algorithmic news recommenders: How to materialize voice in a technologically saturated media ecosystem. Philosophical Transactions A 376 (2133): 20180088. Helberger, N. 2011. Diversity by design. Journal of Information Policy 1: 441–469. Helberger, N. 2019. On the democratic role of news recommenders. Digital Journalism, 1–20 Hendrickx, J., Ballon, P., and H. Ranaivoson. 2020. Dissecting news diversity: An integrated conceptual framework. Journalism. Hildebrandt, M. 2017. Privacy as protection of the incomputable self: Agonistic machine learning. Available at SSRN 3081776. Hildebrandt, M., and B.J. Koops. 2010. The challenges of ambient law and legal protection in the profiling era. The Modern Law Review 73 (3): 428–460. Hoffmann, C.P., C. Lutz, M. Meckel, and G. Ranzini. 2015. Diversity by choice: Applying a social cognitive perspective to the role of public service media in the digital age. International Journal of Communication 9 (1): 1360–1381. Johnson S. 2006. Can We Please Kill This Meme Now. stevenberlinjhonson. https://stevenberlin johnson.com/can-we-please-kill-this-meme-now-117067f1f999?gi=8b3206c1d5e1. Kop, R. 2012. The unexpected connection: Serendipity and human mediation in networked learning. Journal of Educational Technology & Society 15 (2): 2–11. Kotkov, D., J. Veijalainen, and S. Wang. 2020. How does serendipity affect diversity in recommender systems? A serendipity-oriented greedy algorithm. Computing 102 (2): 393–411. Krotoski, A. 2011. Digital serendipity: Be careful what you don’t wish for. The Guardian International Edition. https://www.theguardian.com/technology/2011/aug/21/google-serendipity-pro filing-aleks-krotoski. Lotto, B. 2017. Deviate: The Science of Seeing Differently. Hachette UK. Merton, R.K., and E. Barber. 2006. The Travels and Adventures of Serendipity: A Study in Sociological Semantics and the Sociology of Science. Princeton: Princeton University Press. Makri, S., A. Blandford, M. Woods, S. Sharples, and D. Maxwell. 2014. “Making my own luck”: Serendipity strategies and how to support them in digital information environments. Journal of the American Society for Information Science and Technology 65 (11): 2179–2194. McKay, D., Makri, S., Chang, S., and G. Buchanan. 2020. On birthing dancing stars: The need for bounded chaos in information interaction. In Proceedings of the 2020 Conference on Human Information Interaction and Retrieval, 292–302 Maloney, A., and L.Y. Conrad. 2016. Expecting the Unexpected: Serendipity, Discovery, and the Scholarly Research Process (white paper), Thousand Oaks: SAGE. Meckel, M. 2011. “SOS—Save our serendipity”. https://www.miriammeckel.de/2011/10/11/sossave-our-serendipity. McCay-Peet, L., and E.G. Toms. 2017. Researching serendipity in digital information environments. Synthesis Lectures on Information Concepts, Retrieval, and Services 9 (6): i-i91. Milano, S., M. Taddeo, and L. Floridi. 2020. Recommender systems and their ethical challenges. AI and Society 35 (4): 957–967. Morozov, E. 2012. The Death of the Cyberflaneur. New York Times. https://www.nytimes.com/2012/ 02/05/opinion/sunday/the-death-of-the-cyberflaneur.html. O’Connor, B. 1988. Fostering creativity: Enhancing the browsing environment. International Journal of Information Management 8 (3): 203–210.

8 Serendipity as a Design Principle of Personalization …

165

Pariser, E. 2011. The Filter Bubble: How the New Personalized Web is Changing What We Read and How We Think. New York: Penguin. Pentland, A. 2015. Social Physics: How Social Networks can Make Us Smarter. New York: Penguin. Postman, N. (1986). Amusing Overselves to Death: Public Discourse in the Age of Show Business. ThriftBooks Race, T., and S. Makri. 2016. Accidental Information Discovery: Cultivating Serendipity in the Digital Age. Cambridge: Chandos Publishing. Reviglio, U. 2019. Serendipity as an emerging design principle of the infosphere: Challenges and opportunities. Ethics and Information Technology 21 (2): 151–166. Rice, J. 1988. Serendipity and holism: The beauty of OPACs. Library Journal 113 (3): 138–141. Saini, A.K., and V.K. Khurana. 2013. Using serendipity for ICT development. Global Journal of Enterprise Information System 5 (2). Savage, P., Medina, M., and G.F. Lowe. 2020. Universalism in Public Service Media: RIPE@ 2019. Schmidt, E. 2006. Google Press Day 2006. Google Blogoscoped. http://blogoscoped.com/archive/ 2006-05-10-n76.html. Schönbach, K. 2007. ‘The own in the foreign’: Reliable surprisean important function of the media? Media, Culture and Society 29 (2): 344–353. Seaver, N. 2019. Captivating algorithms: Recommender systems as traps. Journal of Material Culture 24 (4): 421–436. Smets, A. 2022. Designing for serendipity: a means or an end?. Journal of Documentation. Siemens, G. 2014. Connectivism: A learning theory for the digital age. http://www.itdl.org/Journal/ Jan_05/article01.htm. Sullivan, E., Bountouridis, D., Harambam, J., Najafian, S., Loecherbach, F., Makhortykh, M., and N. Tintarev. 2019. Reading news with a purpose: Explaining user profiles for self-actualization. In Adjunct Publication of the 27th Conference on User Modeling, Adaptation and Personalization, 241–245. Stiegler, B. 2017. The new conflict of the faculties and functions: Quasi-causality and serendipity in the anthropocene. Qui Parle 26 (1): 79–99. Stray, J. 2021. Designing Recommender Systems to Depolarize. arXiv preprint arXiv:2107.04953. Sunstein, C.R. 2017a. #Republic: Divided Democracy in the Age of Social Media. Princeton: Princeton University Press. Sunstein, C.R. 2017b. In praise of serendipity. The Economist. https://www.economist.com/booksand-arts/2017b/03/09/in-praise-of-serendipity. Thornburn L., Stray J., and P. Bengani. 2022. What does it mean to give someone what they want? The nature of preferences in recommender systems. Medium. https://medium.com/understan ding-recommenders/what-does-it-mean-to-give-someone-what-they-want-the-nature-of-prefer ences-in-recommender-systems-82b5a1559157. Taleb, N.N. 2012. Antifragile: Things that Gain from Disorder, vol. 3. New York: Random House. van Andel, P. 1994. Anatomy of the unsought finding. serendipity: Origin, history, domains, traditions, appearances, patterns and programmability. The British Journal for the Philosophy of Science 45 (2), 631–648. Yadamsuren, B., and S. Erdelez. 2016. Incidental exposure to online news. Synthesis Lectures on Information Concepts, Retrieval, and Services 8 (5): i-i73. Zuckerman, E. 2013. Rewire: Digital Cosmopolitans in the Age of Connection. New York: W. W. Norton & Company.

Chapter 9

Serendipitous Cognition—The Systematic Consideration of the Accidental Genesis of New Ideas Wendy Ross

Abstract Wendy Ross proposes a method to examine serendipity systematically and disentangle the occurrence of serendipity from the sense of serendipity, which relies on subjective experience. She does so while acknowledging that this perspective on serendipity from the outside can only complement rather than replace existing ideas of serendipity as networked and extending across time. The research she presents is firmly rooted in the tradition of extended and distributed cognition. The proposal builds on the work of Victoria Rubin and colleagues and examines how each part of their three-part model may be investigated systematically. She goes on to describe the work that she is conducting as part of a long term research programme into serendipitous cognition, which relies on first order cognitive tasks and finely grained video analysis to identify moments of microserendipity, allowing for a focus on the precipitating conditions for the noticing of contingent material agency.

Much of the empirical research in serendipity to date has taken place in the realm of information encountering (e.g. Erdelez et al. 2016; Foster and Ford 2003). In these classic descriptions, serendipity only happens when it is recognised as such by the people making the discovery. It is typically identified in hindsight, as the final part of a process which relies on the person’s judgement at the point at which they are asked to consider the event (what I would call the “sense of serendipity”). For example, Makri and Blandford’s (2012) model requires that the person experiencing the moment considers it to be serendipity. In the words of one of their interviewees: “Serendipity doesn’t exist until you have hindsight, until you’ve gone through it and thought about it” (p. 696). McCay-Peet and Toms’s (2015) model picks up on Makri and Blandford’s and combines it with Rubin et al.’s (2011) model of noticing to highlight the importance of the perception of the event as one which is serendipitous, although they note that it is not always essential that this judgement happens immediately. W. Ross (B) Psychology Department, London Metropolitan University, 166-220 Holloway Road, London N7 8DB, England e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 S. Copeland et al. (eds.), Serendipity Science, https://doi.org/10.1007/978-3-031-33529-7_9

167

168

W. Ross

While the empirical study of the importance of serendipity in this personal and retrospective manner is gaining momentum as a research field, this understanding of serendipity can hardly be examined in a systematic way in the cognitive psychologist’s lab. There is blurring of the line between serendipity and the sense of serendipity which renders the phenomenon subjectively dependent on individual interpretation. And so, despite its reported importance, empirical research into serendipity has been inconclusive with regards to the underlying cognitive mechanisms and necessary features; additionally, serendipity’s inherently contingent and subjective nature defies easy experimental manipulation and identification of its causes. This means that its impact cannot be reliably explored or enhanced from a systematic perspective. Additionally, through reliance on narrative reconstruction, we fail to consider the nature of the mediated environmental agency necessary to fully understand serendipitous moments. My research aims to bridge this gap by borrowing methods from the study of cognitive interactivity so that the subjective sense of serendipity (whether that sense is located in the person or the wider socio-cultural domain) can be disentangled from its actual occurrence. This is especially important as that perception appears to be idiosyncratic and not necessarily related to the precipitating conditions (McCayPeet and Toms 2015). It is important to note that the work I describe in this chapter deals with serendipity on a microscale, what I have called microserendipity (Ross and Vallée-Tourangeau 2021a, 2021c; Ross, 2022). Serendipity itself arises from networks of people and things. It is necessarily distributed across networks and time. The aim of the work described here is to provide depth to that broader understanding but not to replace it. Serendipity is multiple and, so, defies a stable definition; it is an experience, triggered by an event which serves a rhetorical function. Such complexity requires a plurality of analysis. Thus, a cognitive approach to serendipitous thinking is an appeal to methodological pluralism and has as an aim to establish a complementary research programme to scaffold understanding alongside the existing empirical approaches. This pluralistic perspective suggests that both narrow and wide views of human behaviour can yield results that increase our understanding of a particular psychological phenomenon. This multiplicity of perspectives is to be welcomed because it seems unlikely that any one viewpoint will be able to explain the complexity of human behaviour. Additionally, it is likely that the appropriate level of analysis will change according to the purpose of the enquiry. A pragmatic approach to the question of the level of analysis is illustrated by Giere (2006)in relation to the composition of fluids. For the molecular chemist the appropriate level of analysis is different than for the engineer who needs to model the flow of water through a system of pipes. A plurality of approaches is not in itself antithetical to scientific progress, but each level must be meaningful. I argue that as it is the accident which is the necessary event in serendipitous thinking (see also Ross 2022), a research programme which focuses on the cognitive effect of this accident is likely to yield beneficial observations, but I do not dismiss the usefulness of other approaches. The research outlined in this chapter also requires a shift in the understanding of cognition. In my work, cognition is not simply an internal process, rather I place my

9 Serendipitous Cognition—The Systematic Consideration …

169

perspective firmly in externalist approaches and indeed, with its focus on distributed agency, I align myself with some of the more radical approaches.1 Taking serendipitous thinking seriously forces a shift from a computational, sequestered and internalist approach into understanding cognition as an ecosystem arising from an active engagement between people and things (Malafouris 2013). The main argument of my work to date is that taking this seriously requires us to examine the moments when human, material and environmental agency2 combine to create an emergent co-agency. Research which takes environmental engagement seriously in understanding cognition needs to accommodate evidence which shows that engagement is non-linear and idiosyncratic. This is not always easy and some doubt has been cast on whether it is even possible. Thus, I foresee benefits across both research fields as the focus of cognitive agency shifts. By focusing in on the trigger event, the “accident”, I hope to be able to identify microscale moments of serendipity as they happen. So called “real world”, serendipity is marked by survivorship bias. That is, we study products which are already validated as valuable by the field and work backwards from this valuable outcome. Alongside the use of both serendipity as narrative and rhetorical devices, this makes it hard to distentangle what might be causal variables because we cannot filter which of the antecedants were in fact necessary but insufficient and so failed to generate serendipitous outcomes. If we study situated and serendipitous accidents outside of criteria of success, we can better focus on the reactions to that accident both individual and systemic before assessing long term impact, which may allow us to draw broader conclusions free of the concerning circularity. We can also move forward from the pivot event rather than backwards, so the work here shifts the traditional direction of research both in serendipity and cognition; rather than moving from outcome to (presumed) origin, we will move from hypothesised cause to outcome. This will allow us to track uneven trajectories and fully assess the importance of that cause. Serendipitous Cognition Below I shall outline a proposal for the systematic study of serendipitous cognition. There are many models of the serendipitous process (e.g. Lawley and Tompkins 2008; Makri and Blandford 2012; McCay-Peet and Toms 2010), but the conceptualisation here is designed with experimental research in mind. It is therefore not a 1

Most approaches to 4E cognition center the agent as an active recruiter of the external world and maintain this biological agent at the centre of the models. More radical approaches such as mine resist this centering. 2 This position states that human and material agency are entangled such that intentionality and agency are neither properties of human nor material but are rather only the property of the humanmaterial system which arises during material engagement. Agency and intentionality on this view are cast as “open” concepts which arise in action rather than a universal and identifiable essence. “Agency and intentionality may not be innate properties of things, but they are not innate properties of humans either; they are emergent properties of material engagement” (Malafouris 2013, p. 149). I am aware of the controversy that can lie in endowing inanimate objects with agency but choose here to use this term to indicate their active involvement in human cognitive processes.

170

W. Ross

descriptive review—one which reports research which has already been conducted— but a prospective one, one which suggests possible avenues of further investigation and potential underlying precipitating cognitive mechanisms. Some of these I have already started exploring, others remain as yet unconsidered. The Importance of Accident Serendipity is naturally extended across people and things situated in an extended timeline (Copeland 2019). Serendipitous cognition, however, is more likely to depend on microserendipity, that is the identifiable moments when cognitive agency and epistemic credit are dissipated across person and environment. An externalist view on cognition requires us to take the material objects seriously and with those their agency in the form of affordances and accidents (Malafouris 2023). Elsewhere, I have defended three characteristics of the serendipitious accident: It arises from unintentional action, it results in the act of noticing, and it leads to change in behaviour (Ross 2022). A serendipitious accident is necessarily contextualised, which should reassure those who would consider such an approach too severely reductionist. A serendipitious accident is also situated; it cannot be understood without a reference to the state of the broader surrounding system before it occurred, and it is only serendipitious if it is later enacted. This means we keep the sequential nature of human action while still maintaining that some aspects are unexpected and unpredictable. Accidents thus add texture to the topography of human cognitive trajectories. A focus on the accident allows us to shift research from cataloguing the person or the system who exploits the accident and also opens up the space to systematically manipulate the type of accident that occurs. While accidents may be necessarily contingent when they occur in the normal narrative arc of a serendipitious story, the only possible way to develop a systematic approach to serendipity is to isolate this moment. As Merton and Barber (2004, p. 261) note, the precipitating cognitive states and subsequent network sedimentation of serendipity are likely to be similar for all discovery: “[I]t is the other component, the ‘accident’, the unanticipated circumstance that provides the opportunity for putting that sagacity to work, that gives ‘serendipity’ its altogether distinctive meaning”. Therefore, serendipitous cognition is centred around the accident as a trigger event. In the work that I do and which is outlined here, the trigger event is a material one, but there is no reason why the work here could not be extended also to social interactions that extend the epistemic landscape in unexpected ways. The accident is serendipitous because it introduces novelty to the system and forces a change in the epistemic state of that system. The key point is that the knowledge was unknown to the system before the accident; whether this knowledge is introduced by material agency or interpersonal communication, the ideas outlined here are still valid. However, given that the broader focus of my work is the study of thinking as distributed across a single agent and her interactions with cognitive objects, the work I describe here is necessarily object bound. In serendipitous cognition, the accident is reframed as a disruption in the epistemic state of the system that forces a noticeable change. This moment of change is the focus to investigate how that change occurs and what the impact of that change is. This

9 Serendipitous Cognition—The Systematic Consideration …

171

Fig. 9.1 Facets of serendipity in everyday chance encounters (Adapted from Rubin et al. 2011)

concept of serendipitous thinking draws heavily on the four aspects of serendipity outlined by Rubin et al. (2011; Fig. 9.1): a prepared mind, the act of noticing, chance and a fortuitous outcome. These facets feed into the notion of serendipitous thinking and I shall use them to structure the discussion because they echo many existing categories which are used to discuss serendipity. Thus, while I borrow the categories, I expand the original analysis somewhat. The combination of facets does not yet appear to be sufficient for serendipity, but it seems likely that each part is to some extent necessary. The systematic approach laid out here aims to test that and quantify how necessary each part might be. Facet A: The Prepared Mind3 Much has been made of the prepared mind and how it interacts with serendipity. It is somewhat ironic that it comes from a simple aside made by Louis Pasteur4 and that it sometimes seems forgotten it is only a theoretical concept rather than a measurable, physical thing. Instead, it has been granted a status which has been solidified by subsequent theorising. In the research field of serendipity, it is the prepared mind which turns serendipity from a story of happy accidents into one of accidents and sagacity. Indeed, there are some such as de Rond who see serendipity as a capability 3

The prepared mind in this description is mainly understood in terms of individual traits. This reflects the micro level of analysis which is the focus of this chapter and the research programme laid out herein. However, serendipity arguably requires an understanding of the prepared mind as extended, that is, that the securing of a serendipitous moment of discovery requires sagacious networks of human and non-human actants. Again, a commitment to a plurality of analysis allows both levels to co-exist. 4 As Merton and Barber (p. 163) write this happened “almost incidentally”.

172

W. Ross

rather than an event, the happy faculty of identifying matching pairs (de Rond 2014). It is certainly clear that it is this interaction with the prepared mind which lifts serendipity above chance fluctuations and makes it a subject of study. However, the prepared mind is not constantly prepared like an ever-vigilant boy scout (Glˇaveanu 2022). Rather, it is prepared for the event. It is an inherently relational state. Part of the magic of serendipity is this overlay of person and event: Serendipity requires the right person, at the right time and in the right place. I propose, therefore, that we start our understanding of the prepared mind as a relational phenomenon. Rather than a static attribute of a great individual, let us centre the accidental event in the picture and place the prepared mind in relation to that accident. Understanding this may help us unwind some of the contingencies of serendipity. Take, for example, the following description of Vera Rubin as she makes the discovery that stars belonging to the same galaxy do not rotate in the same way. Csikszentmihalyi (1996, p. 2) describes it thus: As is the case with many discoveries, this one was not planned. It was the result of an accidental observation of two pictures of the spectral analysis of the same galaxy obtained a year apart. By comparing the faint spectral lines indicating the positions of stars in the two pictures, Rubin noticed that some had moved in one direction along the interval of time, and others had moved in the opposite direction. Rubin was lucky to be among the first cohort of astronomers to have access to such clear spectral analyses of nearby galaxies—a few years earlier such details would not have been visible. But she could use this luck only because she has been, for years, deeply involved in the micromovements of stars.

Here, it is clear that the discovery resulted from the accidental moment, the wider socio-historic situation and also a personal sagacity built on years of expertise (see the heteroscalar model of serendipity I discuss in Ross (2022), which approaches each level in more detail). The sagacity consisted in having the skills and understanding to recognise the implications of the rotation. It is unlikely that a lay person could have understood the significance of the traces. This full description of the person in terms of expertise and personality traits may help us to understand the overall ability to notice, enact and sediment the accidental overlay. However, these skills do not fully explain moments of microserendipity, which seem contingent not on underlying cognitive traits but also temporal coincidences, in other words shifting cognitive states. Take, for example, the problem solver reported in a case study in Steffensen et al. (2016). She solves the problem after noticing an accidental overlap in the experimental materials. The overlap, however, occurs prior to the moment of noticing, is unmade and is then made again so the accident (the overlap) is first non-serendipitous and then serendipitous over the course of the problem solving trajectory. A research programme which deals with a series of static personality traits will fail to capture the shifting epistemic state of the agent in the micro-moment, which may be an important way to increase our understanding of the prepared mind. If we recast the prepared mind as more of a state than a trait, we can explore three broad states in relation to the accident-generating matter. These are not meant as an exhaustive list and it is likely that movement will happen across them; I hope rather to generate a discussion on the epistemic state most likely to precipitate serendipity.

9 Serendipitous Cognition—The Systematic Consideration …

173

In summary, a mind in interaction with material can either be in synchronous flow, frustrated by an elusive capture of material agency and conscious of this frustration, or unconnected and neutral. This latter state is perhaps the closest to “pure” serendipity but as we shall see it is the least tangible. Each of these states is necessarily disrupted by the accident. The Neutral Mind In a traditional understanding of serendipity, the idea of the neutral mind would constitute what van Andel (1994, p. 646) calls “pure serendipity”. In the original framing of the word serendipity, Walpole writes that the princes found by accident and sagacity something that “they were not in quest of” (p xi, this volume). A neutral mind is one that is surprised by an accident, an observation which it was not in search of and from this surprise generates a novel theory. This roughly approximates to what Yaqub (2018) terms either Walpolian or Bushian serendipity. In Walpolian serendipity, investigators are searching in one problem space when the answer arises in another, whereas in Bushian serendipity there is no search underway at all when the unanticipated discovery occurs. Of course, such an unexpected discovery does not require a naïve mind. The unanticipated datum must still be noticed and enacted and this requires some level of sagacity. What is important is that the ongoing activity leading up to the accident is not directed towards solution of the problem it eventually solves. The Flow State The flow states describe a full and uninterrupted connection of the human agent with the material. It remains to be seen if such a state can generate serendipity, and there is some evidence that it may not be possible. March and Vallée-Tourangeau (2022) argue that an accident (which, as we have seen above, is necessary for serendipity) is impossible in a system where the human and the non-human work together in knowledge co-creation. Lock and Sikk (2022) argue that the marker of skilled improvisation is when accidents are erased from the performance because the boundaries between correct and incorrect become irrelevant. Considering why such a state is unconducive to serendipity sharpens our sense both of what serendipity is and what it requires. Serendipity requires disruption of the original state by something outside of the cognitive ecosystem and an introduction of multiple perspectives. A flow state implies a system with flexible boundaries, so intrusions are unlikely to trigger this act of noticing because they are not experienced as intrusions. For an accident to arise in a flow state, the flow state must be disrupted, and the human agent become aware of a disconnect between herself and the material world. This emphasis on disruption can be seen in qualitative reports from artists such as those reported in Gl˘aveanu et al. (2013) who speak of accidents as forcing “moments of distance and reflection” (p. 7). Accidents when they occur in a flow state need to be extreme enough that they disrupt and change that state (otherwise they would not be noticed, see below). An accident in this instance would reinstate the binary division between human and material that flow collapses. Therefore, it is likely that

174

W. Ross

the response to the accident would be to discard it as an impediment to the smooth functioning of the system and, I hypothesise, in this instance less likely to be enacted. The Frustrated Mind It is my supposition that the frustrated mind is the one which is most open to serendipitous occurrences. The state that I suggest we explore here is described much more eloquently by Arfini (Chap. 7 in this volume) as “aching” ignorance. In this state, the agent is aware of her shortcomings and is unable to resolve that epistemic abyss through continued cogitation. She is at what researchers in creative cognition would term an impasse, that is, she has exhausted the search space for solutions. It is related to what Yaqub (2018) calls Mertonian serendipity in a nod to the great sociologist who suggested that Walpole’s definition was too narrow for how discovery actually unfolds in scientific research. This state of mind is firmly related to the state described in Seifert et al.’s (1994) Opportunistic Assimilation Hypothesis. This hypothesis suggests that encountering an impasse sets the stage for any external information triggers because “the tentative partial representation of a problem will have an appropriate stable form, like a nearly completed jigsaw puzzle, ready to receive other crucial missing pieces” (p. 87). This suggests that previous experience works as a series of red flags which are triggered by important pieces of information. This also ties in with research on incubation which suggests that a period of time away from problem solving increases solution likelihood (see Sio and Ormerod 2009 for a review). Important for the methodological argument laid out here is that there is preliminary evidence that this incubation effect is more effective in a first order problem solving environment (Henok et al. 2020). A first order problem-solving environment is one in which the problem-solver is able to interact with and engage with movable problem representations (Vallée-Tourangeau and March 2019). It remains to be seen to what extent the benefits of this materially rich environment are on incubation, although the research conducted used an insight problem (the cheap necklace problem) that other research has suggested is solved around a third of the time through accident in a first order environment (Fioratou and Cowley 2009). As the research programme into serendipitous cognition matures, both the flow and neutral states will remain those from which it is hardest to generate clear recognisable moments of serendipity in controlled conditions. Both require large scale disruptions which are rare; for the flow state, the changes in epistemic state go unnoticed by the system and so do not attain the force of an accident, for the neutral state the accident must generate not only noticing but also a receptive cognitive state. The view of serendipity presented here hinges on the idea of an unanticipated datum and it is most likely to occur when there is a receptive search. Serendipitous cognition as it unfolds in studies of scientific discovery is less magical than the ideal of unanticipated inspiration. Rather, it results from continual work motivated by a sense of frustration and uncertainty and a desire to resolve those uncomfortable states (Ormerod 2023). When an internal cognitive space is exhausted, the work spills into the external cognitive space through speculative experimentation (a sort of Kettering principle, cf. Austin 2003) and the accidents triggered here are most

9 Serendipitous Cognition—The Systematic Consideration …

175

likely to fall on receptive mental states. This means that they are more likely to be noticed, which as we shall see is key to the moment of serendipity. Facet B: The Act of Noticing The act of noticing is central to process models of serendipity. Erdelez (1998) suggests that ‘noticing’ is the first functional aspect of information encountering (a concept closely related to serendipity) while Lawley and Tompkins’ (2008) model requires the recognition of potential (a form of noticing) as the first post-event stage in serendipity. The idea of an ‘act of noticing’ was explicitly extracted by Rubin et al.’s (2011) model and features as the starting point in Sun et al.’s (2011) model. However, not all models centre the act of noticing: Makri and Blandford’s (2012) model begins the serendipitous process with the idea of making a connection which is close to the act of noticing but places more agency in the noticer to use the information (still an important part of the model proposed here, see below). McCay-Peet and Toms (2015) change the noticing aspect to a “trigger” which conversely places the agency in the trigger to generate activity. This points to the relational aspect of the reaction to the accident – the behavioural response is both directed towards and generated by the accident. There are two ways in which the act of noticing is important: First, there is preliminary evidence to suggest that helpful opportunities for action are regularly ignored or missed (Ross and Vallée-Tourangeau 2020a, 2021a). In a series of experiments, I presented participants with letter tiles either to solve anagrams or to generate as many words as they could in a task akin to scrabble. Participants were invited to call out when they had found a solution or a word. The letter tiles allowed the mapping of moments when an answer word or a word close to the answer was formed. This happened surprisingly frequently without being noticed by the participant. Without a conscious noticing, no matter how rich the environmental possibilities, the serendipitous moment is lost. The act of noticing is, therefore, a non-trivial and necessary element in models of the serendipity process. In short, without noticing, environmental opportunities are meaningless no matter what the characteristics of the material and human agent. The act of noticing becomes then the contingent link between the qualities of the environment and the characteristics of the human. The act of noticing is likely to be related to the states outlined in the previous section but may also be informed by other underlying traits such as attentional field (Agnoli et al. 2015). Second, an act of noticing implies a disruption and a surprise. Indeed, I suggest here and elsewhere (Ross 2022) that surprise is a key phenomenological marker of a serendipitous accident. Evidence from qualitative work suggests surprise as an important marker of serendipity. It is noteworthy that surprise is also a key factor in a cognitive event related to serendipity, creative insight (Makri and Blandford 2012; Ross and Arfini forthcoming). A recognition of this can allow us to draw on techniques from the study of creative cognition in our examination of serendipitous discovery. However, it is also worth noticing that a key difference between insightful surprise and serendipitous suprise is that insight is theorised to be an end state, whereas serendipitous surprise is rather the start of knowledge exploration and

176

W. Ross

ignorance reduction. Where insight grants epistemic closure, serendipity represents an epistemic opening. Therefore, it is likely that some of the underlying cognitive mechanisms will be similar while others are less so. Integral to the act of noticing is that it shifts not only the landscape of the agent at that point in time but also her understanding of the action and epistemic state leading up to the moment of epistemic disruption. Existing ignorance is thrown into sharp relief and can no longer be understood without reference to the moment of change. It becomes clear not only that she is ignorant but of what she is ignorant. This then returns us to the question of whether a flow of unceasing and gradual epistemic change can ever generate enough disruption to be called serendipity. It is this requirement of noticing which informs my hypothesis that the frustrated state which is already aware of its own ignorance would be most congenial to serendipity. This also has methodological implications. Given that the accident necessarily changes the phenomenology of the cognitive trajectory, the descriptive direction should be forward from the accident rather than in reverse from the outcome. This allows us to isolate potential serendipity from a sense of serendipity. Enacting Chance For serendipitous cognition, the act of noticing has two aspects: the noticing as described above and the act. Noticing is not enough; it also has to result in action to shore up the changes. Noticing also has to be a conscious act, not only to generate intentional action which moves to shore up the potentials generated by the accident but also to be enfolded in the model of serendipity. All discovery is a messy interplay of chance and unintended action. This is often smoothed out to uphold the illusion of linearity to adhere to traditional models of discovery. What marks narratives of serendipitous discovery is that the accident is important enough to be noticed and becomes game changing (Arfini et al. 2018). We see that, while all noticed accidents result in an updated epistemic state, serendipity arises when that disruption consciously leads to novelty or discovery through action. That action can be a simple reaction and detour but is far more likely to be a selection for action. This is the game changing element of serendipity which is highlighted by Arfini et al. (2018). It is also encompasses the aspect of information encountering called “capturing” (Erdelez 2005) or “follow up” (McCayPeet and Toms 2015). Race and Makri (2016) too, describe innovation generated by serendipity as a change in action. In other words, the act of noticing extends beyond the simple fact of passive epistemic change (mere reception) to using this change to direct future action. In one of the few close examinations of serendipity, Barber and Fox (1958) note that two different scientists made the accidental observation that an injection of papain unexpectedly made rabbit ears go floppy. One had the time and resources to follow this up and the other did not. One experienced serendipity, the other did not. Serendipitous cognition is thus deliberately posited as open-ended.

9 Serendipitous Cognition—The Systematic Consideration …

177

The agent, after noticing and assessing the value of the unanticipated datum, can take three courses of action: She can disregard it for the moment,5 respond in a passive manner (something like the confirmation in Makri and Race’s (2016) model) or enact it, that is, change their course of action to sediment and fully exploit the opportunity. The methodological focus on the accident as a pivot moment allows us to track forwards to see the effects of each action choice. Different actions selected by different people also allow the systematic and controlled assessment of the impact of action. Taking into account the relational and entangled nature of the phenomenon I have described, it is likely the state of mind prior to the accident will also influence the actions taken. However, underlying all of this is a keen awareness of the contingency of the moment of enaction, which needs much more thorough investigation in the future. Facet C: Chance The prepared mind is a necessarily situated mind. Indeed, the often-cited presence of serendipity in stories of creativity and discovery is a convincing argument for including the external world in cognitive models of the genesis of a novel idea. Traditionally, research into the internal mental processes leading to inventions and creativity rests on the notion that there is a creative spark, a novel idea that can be understood in isolation from surrounding material and social environments. Serendipity undermines this notion by suggesting that the skills that lead to discovery cannot be understood when lifted out of the surrounding flow of chance encounters. Skills as skills remain unnoticed without the additional chance encounter (Sand 2020). It posits a view of cognition directed externally, a co-ordination with the environment in which cognition is not tasked with resolving internal states of indeterminacy but rather navigating an uncertain world. This resolution naturally changes the nature of that world and so the two, cognition and world, are co-determined (van Dijk 2021). This view of cognition and human activity is the only way to make sense of the role of serendipity. As argued in other papers (Ross 2020, 2022; Ross and Arfini forthcoming; Ross and Vallée-Tourangeau 2020b, 2021a, 2021c), engaging with the material world requires losing an agent-focused view of external cognition and engaging with distributed cognitive agency. I follow Malafouris (2013) and Pickering (1995) in decoupling the sense of agency from agency itself. This decoupling allows a space for material agency without intentionality. In the research which underlies my overall programme this material agency is manifested through accidents. The accident is a necessary aspect of engaging with a material world in motion and out of the conscious control of the human agent. 5

Yaqub (2018) has a fourth category of serendipity which was not mentioned above, that of Stephanian serendipity in which the unanticipated information solves a problem yet to be posed. It is clear that serendipity encounters can be incubated for a length of time (see McCay-Peet and Toms 2015). This is why in this model the information can be discarded temporarily until a future event causes a disruption and triggers its revival, which provides an interesting avenue of exploration.

178

W. Ross

The accident is not necessarily negative (at its root an accident is not necessarily a negative event) but it does represent an unanticipated disruption to a planned cognitive trajectory even if it is positive. In the of serendipitous cognition outlined here, the external world is not recruited in the service of an omniscient human agent who selects to scaffold internal processes, but, rather, intrudes on those internal processes and derails them. This interruption forces a noticing accompanied by surprise and so is already relational. There is a tricky circularity here; the accident can be described as the triggering cause (Dretske, 2010) which may serve to set in motion the act of noticing, yet it is the act of noticing which grants the intrusion its accidental status. There has yet to be a systematic investigation of the nature of the accident itself but there is an increasing interest in the nature of the environments which might increase the changes of serendipity occurring. These are always relational phenomena. As Björneborn (2017, p.18) suggests, serendipity can be seen as akin to an affordance—that is something which “does not reside inside the environment alone nor inside people alone but reside in the relation—the correspondence—the encounter—between people and a given environment”. However, I wish to move beyond focusing on environments to identifying the accidental event itself. It may be, indeed, it is rather likely, that accidents are more likely to occur in particular environment-person couplings, but serendipitous cognition aims to look not at which environments generate accidents, rather to track outwards from the ripple of that accident. Paradoxically, a focus on the accident as an event may help underline the importance of the surrounding system to more macro levels of serendipity. We know that neither the accident nor the sagacity are sufficient for serendipitous discovery. After all, many wise people have accidents on a regular basis. If it is an accident which distinguishes serendipitous discovery from other forms of discovery, assessing whether post accident activity leads to a positive outcome will allow us to assess the relative important of the enactment of the accident. This would be possible both in the moment and extending beyond in generating the required fortuitous outcome. As suggested before, a focus on accident will reveal that it is not the accident itself but the distributed work which it stipulates that is most important in understanding the creation of novel knowledge. Facet D: A Fortuitous Outcome Serendipity requires Facet D—a fortuitous outcome and the recognition of the event as serendipitous by the agent. It is not only a function of the methods used that serendipity is an experience which is retold but an integral part of the definition of serendipity. This also makes it unstable in the moment because it is simply impossible to state whether a moment is serendipitous or not, and further frustrates experimental or observational research. Take, for example, penicillin: The moment at which Fleming noticed that the mould inhibited bacterial growth was not the moment that penicillin was discovered. Indeed, penicillin was not discovered as we would commonly understand discovery; it was brought into being by networks of people and things which reified this accidental observation (Copeland 2019). Serendipity, thus,

9 Serendipitous Cognition—The Systematic Consideration …

179

can only be properly understood if we accept its essentially distributed nature—across networks of people and actants and further, across time. Serendipity cannot be understood outside of the person and the wider society which is experiencing it. The methods from cognitive psychology are necessarily reductive because they isolate the cognitive unit (person or group or person/object system) from its historical timeline and wide society and require interchangeable units which generate generalisable nomothetic laws. At first glance, the distributed, subjective and networked nature of serendipity make it hard to see how it can be understood using the tools of cognitive science. And indeed, the suggestions and recommendations outlined here do not intend to reduce serendipity to a moment (see also Ross, 2022) but rather to zoom in on one aspect of the experience. To fully understand serendipity, I propose we shift the focus from the post event realisation and reification of serendipity and examine precipitating conditions for the noticing of contingent material agency; this relies on a recognition of moments of microserendipity rather than aiming to explain the whole unstable concept. However, there is a place for post event understanding. Importantly, this understanding also changes the nature of the event which, as I suggest above, will frustrate prospective research. In order to avoid this inversion, that is, a change in the understanding of the process because of the outcome (Latour 1999; Vallée-Tourangeau forthcoming), we should move forward from the accident to potential discovery. Methodological Recommendations The concept of serendipity outlined above was developed with a long-term research programme in mind. This work is at a youthful stage and the various lines of enquiry are open and often lead to more questions than they may initially solve. Such a situation is to be expected with a theoretical and methodological shift of the kind I am calling for. The discomfort of uncertainty may be necessary and discovery will perhaps only arise through doing and action. Only a poor serendipitologist expects to have the answer before starting on the search! Serendipity necessarily distributes cognitive agency and to map this accidental thinking requires rethinking some of the methods of cognitive science. First and Second Order Cognition Despite paying lip service to embodied cognition, mainstream cognitive science tends to search for the locus of discovery in the mind of the biological agent or in her unique abilities. This approach is also supported by the existing methodological decisions. Cognitive tasks tend to be mental tasks—after all, the target phenomenon is a mental one—requiring mental manipulation of often verbal problems. Vallée-Tourangeau and March (2019) suggest that we call these types of problem “second order”-problems because they rely on a mentally represented world. These problems are inescapably mental and are often presented in materially impoverished, sequestered environments (Vallée-Tourangeau and Vallée-Tourangeau 2014). There is little room for the situated cognition—reliant on material or social interactions—that is necessary for serendipity.

180

W. Ross

Fig. 9.2 Five volumes of books each consisting of 150 pages. If a bookworm eats its way from the first page of the first volume to the last page of the second volume, how many pages does it eat through?

Vallée-Tourangeau and March contrast these problems with “first order”-problems for which the participant is invited to move and interact with objects. For example, take the riddle: “Ryan has a box set of books, arranged in order from Vol. 1 on the left to Vol. 5 on the right. Each book is 150 pages long. A bookworm had eaten its way in a straight line from the first page of Vol. 1 to the last page of Vol. 2. How many pages did it damage?”. Even with a picture of the books (such as in Fig. 9.2), it is hard to suppress the prepotent feeling that when moving from left to right the first page of volume 1 is to the left of the row of books and so the worm munched his way through the 150 pages of Volume 1 and the 150 pages of Volume 2, 300 pages in total. However, when books are ordered in this way the first page is always to the right of the volume and the last page to the left, thus the answer become immediately clear: No pages are eaten. Not only external representations but actions over objects reveal the answer in a form of “outsight”, where the answer unfolds in front of the problem solver rather than as a form of internal processing; objects and understanding combine through movement. Such a perspective with its emphasis on engagement with the environment is a far more promising approach to understanding serendipity as it unfolds in interaction with the environment. The initial work on first order problem solving, or interactivity, sought to establish interactivity’s scaffolding effect on problem solving and was based on an agent centric model where objects were recruited but lacked agency—they were viewed as cognitive offloads, necessarily augmentative expansions of the cognitive workplace. This led to experimental designs which contrasted conditions with and without objects to seek to illustrate the augmentative aspect of embedding human problem solvers in the world. My own work aims to shift this view and redistribute the agency across people and things endowing the objects with cognitive agency through their material characteristics and through the role of accidents. The nature of the shift from a predictive, experimental paradigm to a descriptive one also turns the objects from simple cognitive offloads to markers and traces of cognition. A theoretical position which suggests thoughts and objects are co-created also supports the tracing of thoughts through the tracing of movements, as the difference between the two collapses. Such a microgenetic analysis allows the observation of moments of accident and therefore, moments of serendipity.

9 Serendipitous Cognition—The Systematic Consideration …

181

Methodological Details The work described here draws heavily on a qualitative, microgenetic method which is explored more fully in other published work (e.g. Ross and Vallée-Tourangeau 2020a, 2020b, 2021a; Vallée-Tourangeau et al. 2020) but shall be briefly outlined here.6 In short, this approach does not only assume that objects are cognitive and form vital aspects of cognitive ecosystems but that they can also function as cognitive traces which can be mapped and catalogued. In such a way, cognition is located not in the mind but in people’s behaviour. Once the distinction between thought and behaviour is revoked, behaviour can be conceived not as reflection of internal thought processes but as thought itself. The method that has been developed across the course of our research in this methodological tradition to date has two key features: An experimental set-up and a mixed methods approach. The first allows strict boundaries and definitions of success, stabilising somewhat the contingency of the outcome and allowing manipulations of chance and accident; the methodological pluralism combines deductive quantitative analysis with inductive qualitative analysis, and systematicity with idiosyncratic description. Thus, it lies at an uncomfortable intersection of research traditions. However, as I have indicated throughout, a pluralist approach is important. This novel method relies heavily on Cognitive Event Analysis (Steffensen et al., 2016) and cognitive ethnography. This analysis allows us to track the moments in which serendipity arises by following object-thought couplings (or mutualities; Vallée-Tourangeau forthcoming) as they arise in action and through action. It is also rooted in a pragmatic approach to cognition which suggests that intentionality can be witnessed in the actions of people. Using this method, I invite participants to solve problems using movable artefacts and at times prompt them to shuffle and rearrange the artefacts at random. For example, I asked participants to generate as many words as they could from a set of seven letters. The fortuitous outcome was clear—generating a word was a good thing—and unambiguous. However, although initial strands of our research focused on describing the events which led up to a successful solution—a serendipitous outcome—we soon learnt that a focus on the process made this reverse engineering of a solution difficult to sustain because similar processes lead to different outcomes and similar outcomes were sparked by different processes (Ross and Vallée-Tourangeau 2021a; Vallée-Tourangeau forthcoming). A correct solution is only generated at the moment of solution, so to attribute causal agency to the process before then is dangerous. Rather than working backwards from outcome, the work we follow here moves forward, in the case of the work on serendipity, from an accident to the outcome. In other words, we do not look for what caused a positive outcome but rather what is caused by a move in the world. Accidents are generated both naturally (through the flow of lettered tiles as they are moved around) and also artificially by asking participants to shake and randomly 6

The methodological musings are presented here as a fully sedimented programme but that is a necessary smoothing of the process of scientific discovery. Rather, the method here is open and it is likely that it will be continually refined.

182

W. Ross

rearrange the tiles. Thus, a level of control is applied to the situation which violates the traditional disciplinary parameters of ethnography, but which allows a micro analysis of the interaction of cognitive movements across large data sets. The instrumentalization of the procedure is important, numbered or lettered moveable pieces enable the easy tracking of object-thoughts and action-thinking. When less attention is paid to the instrumentalization of the problem space then the data generated are no less interesting but hardly lend themselves to process tracing. For example, we conducted research using interchangeable socks as objects, rather than abstracted moveable representations, which yielded interesting data but told us little about the movement of objects in the world (Ross and Vallée-Tourangeau 2021b). Similarly, in the search for accidental cognition, the experimental set up has to allow for the widest range of unplanned actions and also elicit a range of non-intentional moves such as fiddling. Certain set ups such as touch screens will necessarily limit the accidents available even though they bestow a level of experimental control (March and Vallée-Tourangeau 2022). Three levels of analysis are conducted on the data. First, group level aggregate scores look overall at the most beneficial strategy to reach problem solving success. Second, process-based analysis, which uses interaction analysis (Bakeman and Gottman 2009; Bakeman and Quera 2011), to assess cognitive actions across the whole data set. Finally, a selected subset is used to produce detailed case studies. This final level of analysis draws on the methods of cognitive ethnography (e.g. Hutchins 2010) and is idiographic and qualitative, following the principle of cognitive probatonics, a focus on the individual cognitive trajectory (Steffensen 2016). This work allows us to identify and manipulate the moment of serendipity through describing moments of accident and also through generating accidents in a controlled manner. This also offers the opportunity to track forward from the change rather than track back from the outcome. In other words, rather than starting the mapping of the cognitive trajectory from the outcome which would be a traditional method of approaching cognitive research, the method outlined here tracks forward from a controlled pivot moment. This will allow us to be clearer about the causal efficacy of the moments we identify. For example, our research has already shown that many accidents are not noticed by the problem solvers. Such moments of “missed serendipity” could not be identified in any other way. Discussion It remains to be seen to what extent this semi-reductionist approach can yield useful information about an aggregate and emergent phenomenon. At times, the research described here is full of hope, at other times it struggles a little under some fundamental inconsistencies. My suggestion is that those inconsistencies may resolve themselves through the action of doing and exploring through empirical research. I also suggest that many, more mature, research fields progress somewhat burdened by inconsistencies. For example, psychological research into creativity relies on a bipartite standard definition of creativity which requires something both novel and valuable. Each of those terms is relational and the creative status of an object, act or person can change over time. Creative cognition, therefore, is in a curious position

9 Serendipitous Cognition—The Systematic Consideration …

183

in cognitive science perhaps analogous to my work on serendipitous cognition. In selecting creativity, it is tasked with understanding not only a latent phenomenon but one which is conceptually contingent and ontologically unstable. However, research in this area has carved out a niche which yields some useful empirical data. I do not suggest that a focus on microserendipity and serendipitous cognition can replace the broader focus on macro levels of serendipity nor that unpicking the black box will necessarily lead us any closer to an understanding or even that such an understanding is desirable or possible.7 However, such a complex and unstable phenomenon requires a plurality of analyses and a plurality of approaches. I believe that only with a multiplicity of perspectives can a comprehensive understanding of the phenomenon be developed. I have also provided a description of a phenomenon with sharp edges. I have presented three clear precipitating mental states, three possible actions, a neatly delineated accident. Of course, this description is an artificial rendering of a fluid and complex process. Boundaries and binaries collapse whenever they are examined at their liminal edges, and it is an unrealistic research programme which would suggest that such categories are necessarily neatly compartmentalised. However, there is a very real danger of stagnation if we focus only on testing the places where these things break down. These places may be easy to find but I argue that we will progress further in our understanding even of the limitations of the models and boundaries if we sit with the discomfort of potentially collapsing boundaries. I have made a similar argument elsewhere (Ross 2022) about the necessity of resisting examining thresholds in order to make practical progress. Threshold definitions are rarely sustainable when closely examined but collapsing them can lead to a dangerous dissolution and conceptual spread which frustrates empirical research. By focusing on the accident as it happens and mapping the actions and consequences that flow from that event, we can perhaps begin a science of prospective rather than retrospective description. Such a detailed and granular method will likely add to our understanding of this complex and multifaceted phenomenon. Acknowledgements I thank Frédéric Vallée-Tourangeau, Samantha Copeland and Selene Arfini for many inspiring conversations which have informed the ideas laid out here and Martin Sand for close and useful feedback on a previous version of this manuscript. Thomas Ormerod has been invaluable in supporting me to thoroughly test the empirical limits of these ideas.

7

I am aware that we run the very real risk of disassembling the black box of serendipity and find ourselves none the wiser. Serendipity happens in process and it is not clear that reducing its complexities to a series of blocks will bring us closer to understanding anything about this contingent and emergent process. However, discounting this as a possibility would at least advance the field of research.

184

W. Ross

References Agnoli, S., L. Franchin, E. Rubaltelli, and G.E. Corazza. 2015. An eye-tracking analysis of irrelevance processing as moderator of openness and creative performance. Creativity Research Journal 27 (2): 125–132. https://doi.org/10.1080/10400419.2015.1030304. Arfini, S., T. Bertolotti, and L. Magnani. 2018. The antinomies of serendipity: How to cognitively frame serendipity for scientific discoveries. Topoi. https://doi.org/10.1007/s11245-018-9571-3. Austin, J.H. 2003. Chase, Chance, and Creativity: The Lucky Art of Novelty. MIT Press. Bakeman, R., and J. Gottman. 2009. Observing Interaction: An Introduction to Sequential Analysis. Cambridge University Press. Bakeman, R., and V. Quera. 2011. Sequential Analysis and Observational Methods for the Behavioral Sciences. Cambridge University Press. Barber, B., and R.C. Fox. 1958. The case of the floppy-eared rabbits: An instance of serendipity gained and serendipity lost. American Journal of Sociology 64 (2): 128–136. https://doi.org/10. 1086/222420. Björneborn, L. 2017. Three key affordances for serendipity: Toward a framework connecting environmental and personal factors in serendipitous encounters. Journal of Documentation 73: 1053–1081. https://doi.org/10.1108/JD-07-2016-0097. Copeland, S.M. 2019. On serendipity in science: Discovery at the intersection of chance and wisdom. Synthese 196: 2385–2406. https://doi.org/10.1007/s11229-017-1544-3. Csikszentmihalyi, M. 1996. Creativity: The Psychology of Disovery and Invention. Harper Collins. de Rond, M. 2014. The structure of serendipity. Culture and Organization 20 (5): 342–358. https:/ /doi.org/10.1080/14759551.2014.967451. Dretske, F. 2010. Triggering and structuring causes. In A Companion to the Philosophy of Action, 139–144, ed. T. O’Connor and C. Sandis. Wiley-Blackwell. http://doi.wiley.com/10.1002/978 1444323528.ch18. Erdelez, S. 1998. Towards understanding information encountering on the web. In Proceeding of the 63rd Annual Meeting of the American Society for Information Science, 363–371. Erdelez, S. 2005. Information encountering. In Theories of Information Behaviour, ed. K.E. Fisher, S. Erdelez, and L. McKechnie, 179–185. Information Today. Erdelez, S., J. Heinström, S. Makri, L. Björneborn, J. Beheshti, E. Toms, and N.K. Agarwal. 2016. Research perspectives on serendipity and information encountering. Proceedings of the Association for Information Science and Technology 53: 1–5. https://doi.org/10.1002/pra2.2016.145 05301011. Fioratou, E., and S.J. Cowley. 2009. Insightful thinking: Cognitive dynamics and material artifacts. Pragmatics & Cognition 17 (3): 549–572. https://doi.org/10.1075/pc.17.3.04fio. Foster, A.E., and N. Ford. 2003. Serendipity and information seeking: An empirical study. Journal of Documentation 59: 321–340. https://doi.org/10.1108/00220410310472518. Giere, R. 2006. Perspectival pluralism. In Scientific Pluralism, ed. S.H. Kellert, H.E. Longino, and C.K. Waters, 26–42. University of Minnesota Press. Gl˘aveanu, V.P., T. Lubart, N. Bonnardel, M. Botella, P.-M. de Biaisi, M. Desainte-Catherine, A. Georgsdottir, K. Guillou, G. Kurtag, C. Mouchiroud, M. Storme, A. Wojtczuk, and F. Zenasni. 2013. Creativity as action: Findings from five creative domains. Frontiers in Psychology 4: 176. https://doi.org/10.3389/fpsyg.2013.00176. Glˇaveanu, V.P. 2022. What’s inside the prepared mind. In The Art of Serendipity, ed. W. Ross and S. Copeland. Palgrave. Henok, N., F. Vallée-Tourangeau, and G. Vallée-Tourangeau. 2020. Incubation and interactivity in insight problem solving. Psychological Research Psychologische Forschung 84 (1): 128–139. https://doi.org/10.1007/s00426-018-0992-9. Hutchins, E. 2010. Enaction, imagination and insight. In Enaction: Towards a New Paradigm for Cognitive Science, ed. J. Stewart, O. Gapenne, and E. Di Paolo, 424–450. MIT Press. Latour, B. 1999. Pandora’s Hope: Essays on the Reality of Science Studies. Harvard University Press.

9 Serendipitous Cognition—The Systematic Consideration …

185

Lawley, J., and P. Tompkins. 2008. Maximising serendipity: The art of recognising and fostering unexpected potential. First Presented at the Developing Group. Lock, G., and J. Sikk. 2022. Accident and serendipity in music composition, improvisation and performance art. In The Art of Serendipity, ed. W. Ross and S. Copeland. Palgrave MacMillan. Makri, S., and A. Blandford. 2012. Coming across information serendipitously—Part 1: A process model. Journal of Documentation 68: 684–705. https://doi.org/10.1108/00220411211256030. Malafouris, L. 2013. How Things Shape the Mind: A Theory of Material Engagement. MIT Press. Malafouris, L. 2023. Enactychism: Enacting chance in creative material engagement. Possibility Studies and Society 1 (2). March, P.L., and F. Vallée-Tourangeau. 2022. Briefing for a systemic dissolution of serendipity. In The Art of Serendipity, ed. W. Ross and S. Copeland. Palgrave MacMillan. McCay-Peet, L., and E.G. Toms. 2010. The process of serendipity in knowledge work. In Proceeding of the Third Symposium on Information Interaction in Context—IIiX’10, 377–382. https://doi. org/10.1145/1840784.1840842. McCay-Peet, L., and E.G. Toms. 2015. Investigating serendipity: How it unfolds and what may influence it. Journal of the Association for Information Science and Technology 66: 1463–1476. https://doi.org/10.1002/asi.23273. Merton, R., and E. Barber. 2004. The Travels and Adventures of Serendipity: A Study in Sociological Semantics and the Sociology of Science. Princeton University Press. Ormerod, T.C. 2023. Possible, yes, but useful? Why the search for possibilities is limited but can be enhanced by expertise. Possibility Studies and Society 1 (1). Pickering, A. 1995. The Mangle of Practice: Time, Agency, and Science. University of Chicago Press. Race, T.M., and S. Makri. 2016. Introducing serendipity. In Accidental Information Discovery: Cultivating Serendipity in the Digital Age, ed. T.M. Race and S. Makri, 1–14. Elsevier: Chandos Publishing. Ross, W. 2020. Serendipity. In The Palgrave Encyclopedia of the Possible, ed. V.P. Glˇaveanu. Palgrave MacMillan. Ross, W., and S. Arfini. 2023. Serendipity and creative cognition. In Routledge Handbook of Creative Cognition, ed. L.J. Ball and F. Vallée-Tourangeau. Ross, W., and F. Vallée-Tourangeau. 2020a. Catch that word: Interactivity, serendipity and verbal fluency in a word production task. Psychological Research Psychologische Forschung. https:// doi.org/10.1007/s00426-019-01279-y. Ross, W., and F. Vallée-Tourangeau. 2020b. Microserendipity in the creative process. Journal of Creative Behavior 55 (3): 661–672. Ross, W., and F. Vallée-Tourangeau. 2021a. Accident and agency: A mixed methods study contrasting luck and interactivity. Thinking & Reasoning 28 (4): 487–528. Ross, W., and F. Vallée-Tourangeau. 2021b. Rewilding cognition: Complex dynamics in open experimental systems. Journal of Trial and Error 2 (1): 30–39. Ross, W., and F. Vallée-Tourangeau. 2021c. Catch that word: Interactivity, serendipity and verbal fluency in a word production task. Psychological Research Psychologische Forschung 85 (2): 842–856. https://doi.org/10.1007/s00426-019-01279-y. Ross, W. 2022. Heteroscalar serendipity and the importance of accidents. In The Art of Serendipity, ed. W. Ross and S. Copeland. Palgrave MacMillan. Rubin, V.L., J. Burkell, and A. Quan-Haase. 2011. Facets of serendipity in everyday chance encounters: A grounded theory approach to blog analysis. Information Research 16(3): 488. Retrieved from https://informationr.net/ir/16-3/paper488.html. Sand, M. 2020. Did Alexander Fleming deserve the Nobel prize? Science and Engineering Ethics 26 (2): 899–919. https://doi.org/10.1007/s11948-019-00149-5. Seifert, C.M., D.E. Meyer, N. Davidson, A.L. Patalano, and I. Yaniv. 1994. Demystification of cognitive insight: Opportunistic assimilation and the prepared-mind hypothesis. In The Nature of Insight, ed. R.J. Sternberg and J. Davidson, 65–124. MIT Press.

186

W. Ross

Sio, U.N., and T.C. Ormerod. 2009. Does incubation enhance problem solving? A meta-analytic review. Psychological Bulletin 135 (1): 94–120. https://doi.org/10.1037/a0014212. Steffensen, S.V. 2016. Cognitive probatonics: Towards an ecological psychology of cognitive particulars. New Ideas in Psychology 42: 29–38. https://doi.org/10.1016/j.newideapsych.2015. 07.003. Steffensen, S.V., F. Vallée-Tourangeau, and G. Vallée-Tourangeau. 2016. Cognitive events in a problem-solving task: A qualitative method for investigating interactivity in the 17 Animals problem. Journal of Cognitive Psychology 28 (1): 79–105. https://doi.org/10.1080/20445911. 2015.1095193. Sun, S., S. Sharples, and S. Makri. 2011. A user-centred mobile diary study approach to understanding serendipity in information research. Information Research 16 (3): 16–23. Vallée-Tourangeau, F. 2023. Insight in the kinenoetic field. In Routledge Handbook of Creative Cognition, ed. L.J. Ball and F. Vallée-Tourangeau. Routledge. Vallée-Tourangeau, F., and G. Vallée-Tourangeau. 2014. Diagrams, jars, and matchsticks: A systemicist’s toolkit. Pragmatics & Cognition 22 (2): 187–205. https://doi.org/10.1075/pc.22. 2.02val. Vallée-Tourangeau, F., and P.L. March. 2019. Insight out: Making creativity visible. The Journal of Creative Behavior. https://doi.org/10.1002/jocb.409. Vallée-Tourangeau, F., W. Ross, R.R. Rech, and G. Vallée-Tourangeau. 2020. Insight as discovery. Journal of Cognitive Psychology 33 (7): 1–20. https://doi.org/10.1080/20445911.2020.1822367. Van Andel, P. 1994. Anatomy of the unsought finding. The British Journal for the Philosophy of Science 45: 631–648. van Dijk, L. 2021. Psychology in an indeterminate world. Perspectives on Psychological Science 16 (3): 577–589. https://doi.org/10.1177/1745691620958005. Yaqub, O. 2018. Serendipity: Towards a taxonomy and a theory. Research Policy 47: 169–179. https://doi.org/10.1016/j.respol.2017.10.007.

Wendy Ross studies the role of material serendipity in higher cognitive processes such as insight problem solving and creativity. She draws on a range of methods from eye-tracking and experimental psychology to focused cognitive ethnography. She has recently co-edited the collection on serendipity: The Art of Serendipity (Palgrave). She is Co-Chair of the Serendipity Society and Vice President of the Possibility Studies Network as well as an elected member of the BPS Cognitive Section Committee. In 2021 she was awarded the Frank X Barron prize by Division 10 of the APA.

Chapter 10

Serendipity, Luck and Collective Responsibility in Medical Innovation—The History of Vaccination Martin Sand

and Luca Chiapperino

Abstract Martin Sand and Luca Chiapperino find in the concept of serendipity a versatile umbrella term to reassess their previous work on moral luck and collective responsibility. Moral luck supposedly occurs when someone receives praise or blame for things beyond control. Given the ubiquity of luck, this seems to be a seriously disquieting aspect of ordinary morality. The rewards and recognition for serendipitous discoveries fall into exactly this category. That is: more than the intentions, actions, and characters of scientists matters for discoveries to obtain, just as in cases of moral luck something beyond morality affects our moral judgments. Even if a theoretical way of resolving the conceptual ambiguities that underlie this debate were found, there remain practical questions of how to perform stratification in science and innovation in ways that both hinge on, and yet refrain from, considerations of desert and achievement. With the example of Edward Jenner’s luck- and serendipity-infused discovery of vaccination, the authors attempt to better understand the intricate value trade-offs that underlie stratification policies in science, which have to be constantly re-negotiated to maintain their legitimacy. Thereby, Sand and Ciapperino aim to take a bold step towards understanding the ethics of serendipity.

In this chapter, we consider the history of vaccination and discuss its relation to the notions of serendipity and luck. Edward Jenner is usually credited with the discovery of the basic mechanism of vaccination that led to the eradication of smallpox: However, when dissecting this case in more detail, we will find that luck was involved in various shapes and forms throughout this process. Given the conceptual proximity between luck and serendipity, the discovery of vaccination has also been considered serendipitous (Roberts 1989). It has been suggested, that is, that Jenner was made M. Sand (B) Department of Values, Technology and Innovation Faculty of Technology, Policy and Management, Jaffalaan 5, Delft 2628 BX, The Netherlands e-mail: [email protected] L. Chiapperino Faculty of Social and Political Sciences, Institute of Social Sciences (ISS), UNIL-Mouline, Bâtiment Géopolis, Lausanne, Suiss CH-1015, Quartier, Switzerland © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 S. Copeland et al. (eds.), Serendipity Science, https://doi.org/10.1007/978-3-031-33529-7_10

187

188

M. Sand and L. Chiapperino

aware of the immunizing effect of cowpox by a milkmaid during a rather accidental encounter. This account, which identifies luck in the antecedent conditions of the finding— conditions which he did not control—has been challenged by historians. Serendipity is often understood as “finding the unsought” (van Andel 1994) or “looking for something and finding another” (Gillies 2015). In light of these notions, Jenner’s exposure to rumors in the dairies makes vaccination indeed seem like an unsought finding, but much of this assessment hinges on information that historians are lacking, as we will show. In contrast to this traditional account, we will suggest it is better not to identify serendipity as appertaining to Edward Jenner’s realizing that vaccination might be an effective medical procedure (after receiving the hint), but rather to appeal instead to a notion of serendipity as an emergent property of the collective effort that ultimately made vaccination a global success (Copeland 2018): That is, vaccination could become a global standard for disease prevention only through advancements in microbiology and sterilization techniques (to which Ignaz Semmelweis (Gillies 2005) and later Joseph Lister made major contributions in the second half of the nineteenth century (Lister 1867), and that formed the basis of Pasteur’s work that followed suit), as well as the global effort of the World Health Organization (WHO). Jenner’s early experiment is only one building block in the monument that is today’s common practice of vaccination. Serendipity is, thus, present as an emergent property of this process rather than being confined to the “moment of discovery”. Describing these events with a slightly different terminology, one might suggest that luck has influenced the development of vaccination—luck provided the circumstances and contributed to the results that made vaccination a global success story.1 Surely, luck is not always related to discovery: Luck in gambling is not serendipitous. But all serendipitous discoveries contain luck: If something unsought is found, it has not been pursued and something that has not been pursued was not (intentionally) under control. Luck is oftentimes understood to be a significant event beyond someone’s control.2 There is clearly an affinity between serendipity and luck (Sand and Copeland 2020). Consider that at the moment of his first vaccination experiment—the first documented vaccination experiment with systematic intent3 —Edward Jenner could by no 1

Paul Thagard seems to use both terms interchangeably. He puts “questioning” in the centre of his psychological model of discovery and suggests that serendipity can lead to surprise and thus further questioning. Thus, serendipity is not a combination of wisdom and chance, but comes entirely from outside the agent (Thagard, 2011). In this way, it could be replaced with the concept of luck. 2 Luck is conceptually at least as fuzzy as serendipity (Hales 2020). Clearly, many things that are significant and beyond one’s control are not things we consider to be luck: The fact that the sun rises every day ensures survival and is beyond our control, but not really lucky. Hence, philosophers wrestle with adding qualifying conditions such as likelihood to exclude such cases from the extension of the concept. However, even this does not seem to make the concept of luck any clearer: The chances for surviving a round of Russian roulette are pretty high, yet, most of us would consider themselves lucky when surviving. 3 It is important to underscore that previous vaccination experiments have been undertaken without medical or systematic intent by Benjamin Jesty (Pead 2003, 2006). Unlike Jenner, Jesty indeed undertook these “experiments” on his family members.

10 Serendipity, Luck and Collective Responsibility in Medical …

189

means be certain that the vaccine had induced immunity in the eight-year-old James Phipps when he infected him with smallpox. In other words, he put the boy at risk of dying. Had that happened, current judgments about his service to medicine would be very different from what they actually are (Sand et al. 2019). Jenner would have probably vanished from the annals of medical history. In his own reflections on the experiment, he expresses “anxiety” and uncertainty: He was not entirely confident in the success of the procedure and was aware of the accompanied risks (Jenner 1801). Luckily, the procedure worked and James Phipps survived.4 But, vaccination as a medical practice continued to face resistance. The procedure had several shortcomings based on misunderstandings that Jenner himself induced, and it took a long time to convince the general public of its merits (we will return to this later on). Only through fervent advocacy by influential aristocrats, the invention of modern sterilization methods and the global effort of the WHO, could vaccination lead to the global eradication of smallpox. These too are instances of luck—matters beyond his control—that nonetheless have had some striking effects on our (moral) assessment of Jenner and his perceived role in the vaccination revolution: He is often mentioned as the most important agent towards this unique medical innovation. In this manner, we will read the case as one of moral luck and take this as a backdrop to problematize the usual means of stratification in science and policies of publicly attributing merit and commemorating individual achievements. That is, we argue that Jenner is elevated to the Pantheon of medical heroes for vaccination’s global success in spite of the contributions of numerous other actors. In summary, we suggest that Jenner can count himself lucky in two distinct ways: First, he undertook an experiment that could have failed dramatically, but succeeded instead. Second, without a number of contingent factors (including additional medical progress and efforts by other actors) melding in the right way, his finding could not have become the global success for which he is hailed. Both of these were beyond his control, which is in our view a constitutive condition for luck (see footnote 2).5 We will start by sketching the forebear of vaccination and the historical context of Jenner’s first vaccination experiment. Vaccination, Serendipity and Moral Luck Roughly speaking, vaccination is a medical practice, whereby one receives an injection of a mild form of a certain disease in order to activate the body’s immune-system to produce antibodies to later resist stronger versions of the same disease or diseases of the same family. In the eighteenth century, the working mechanism behind these 4

Jenner might not have been aware, despite acknowledging the risks involved, that this was in fact a matter of luck. To him, the procedure—and its eventual success—might have been the logical and necessary consequence of the conviction that his hypothesis was right. Why the hypothesis was right, though, he could not have fully understood at the time. Clearly, how strongly one is convinced of a hypothesis and how far one decides to go to prove it, is not only a matter of epistemology, but also of research integrity. We thank Samantha Copeland for bringing this point to our attention. 5 We believe that there has been agential contribution to this process, which might have been necessary, but was insufficient for the success of vaccination. In a deterministic universe, in contrast to this view, all causality is necessary and sufficient.

190

M. Sand and L. Chiapperino

facts—the role of the immune systems and antibodies—were still unknown (Helmstädter 2008). What existed, then, was a pre-form of vaccination called “variolation”, which was already practiced in ancient China and the Middle East, according to several historical accounts, and reached England in the early eighteenth century (Boylston 2012, 2013; Williams 2010): Lady Montague brought the practice from Turkey to London and applied it to her kids. Robert Sutton from Suffolk in England was a doctor who practiced variolation systematically and also applied it to his kids (Gross and Sepkowitz 1998). The Suttonian method of variolation was widely known and a source of great fortune for the Suttons (Bazin 2000). Naturally, when Jenner’s method got traction, the Suttons had the strongest reservation against vaccination. When he was just a child, Jenner himself underwent the Suttonian treatment. Variolation works by injecting a mild form of smallpox into the upper layer of the skin. The skin cells will send all necessary information about the bacteria to the immune system, which immediately starts developing anti-bodies. By the time the smallpox reaches the bloodstream, the body has already developed all necessary defences and can easily fight off the main attack. The problem with variolation is that if it is practiced wrongly and smallpox is infused too deep into the skin, there is the risk that the patient suffers a full-blown smallpox infection, which at the time was highly lethal and left many people disfigured (Baxby 1999; Williams 2010). Furthermore, by inoculating people with the smallpox, you immunize them against the disease, but while they are infected, they remain infectious to others and could become a source of bigger outbreaks.6 Therefore, while variolation helped to immunize some people it also meant procreating the disease. Something better was required. Against this historical backdrop, Edward Jenner was the first to systematically investigate and describe a different method, which he called “variolæ vaccinæ”. The treatise An Inquiry into the Causes and Effects of the Variolæ Vaccinæ, in which he suggests that cowpox protects from contracting smallpox was first privately published in 1798 and in a revised version by Ashley & Brewer in 1802 (Jenner 1802a). In Jenner’s time, the distinction between medical science and medical practice was much less clear (Jewson 1976). In the practice of what Jewson calls—“bedside medicine”—statistical methods and systematic experimentation were not yet fully developed. Medical knowledge was primarily gained on the basis of patients’ accounts of a disease experience (Jewson 1976). Despite this, the original Inquiry, which Jenner recommended to the Royal Society for publication in 1798, was even in his time considered to be based on too much circumstantial knowledge: Sir Joseph Banks, president of the Royal Society 6

In particular, the latter of these two points has been disputed by Peter Razzell (1965). Razzell suggests that the contagiousness of the person diminishes with the severity of infection. Since inoculated patients sustained only mild forms of disease, they remained allegedly non-infectious. Furthermore, he claims that initial vaccination experiments were misinterpreted: Jenner mistakenly thought that he had conferred cowpox on his patients, while it was in reality a variation of smallpox. Rezzal’s paper has been controversially discussed and published with an editorial note and comment. Our account is largely based on Gareth William’s research (Williams 2010). It is noteworthy that before proper methods of sterilisation the method of inoculation suffered from the same issues of transferring other forms of bacteria from human to human.

10 Serendipity, Luck and Collective Responsibility in Medical …

191

in London, rejected the original treatise because Jenner’s account relied on too few observed cases (Baron 2014; Bazin 2000, p. 39). While the Inquiry contains all essential ingredients to kick off a medical revolution, it is often considered a somewhat careless and prematurely published study (Baxby 1999; Beale and Beale 2005). It contained a number of flaws that haunt the vaccination debate until today: First, in the original version, Jenner acknowledged only in passing that there are different types of ulcers to be found on the utters of cows, some which he considered “spurious cowpox” and some “true cowpox”. The “spurious cowpox” had a different virological nature and could be harmful to humans. It did not produce the sought immunity against smallpox. Hence, distinguishing the two was of utmost importance, which Jenner failed to do with the required precision. Second, Jenner assumed that vaccination would protect people for the rest of their lives from contracting smallpox, which is not true. The effect wears off after some time and has to be refreshed. He drew a premature conclusion from his early studies (Baxby 1999, p. 305). Third, he asserted that one could perfectly well inoculate patients with bacteria from other patients, who had been infected with cowpox, instead of using the pox material directly from cows. Given the lack of proper aseptic techniques at the time, this meant that the procedure likely transmitted other infectious diseases like syphilis from human to human (Baxby 1999), a flaw that appertained equally to the practice of variolation. A serious side-effect of the procedure until at least the beginning of the twentieth century was, hence, that a significant number of people developed bacterial infections through non-sterile vaccination equipment. This—although not directly caused by vaccination—fuelled vaccination opposition and gave rise to prejudices that did not easily vanish even after proper hygiene techniques were developed. Fourth, Jenner’s idea of inoculating humans with matter coming from the bodies of humble dairymaids (like Sarah Nelmes, whose case was reported in his Inquiry) or even the udder of cows was in many ways subversive. On the one hand, it challenged established social orders by rendering the bodies of the bucolic and ignorant members of eighteenth century British society essential to the well-being of their social superiors (Fulford and Lee 2000). The owners of the lands and cattle now depended on the bodies of the women and men whose labour they condescended to. On the other hand, Jenner’s technique affirmed also a fundamental unity of the living, which was problematic for the medical establishment of that time. Cattle and humans did not only display similar diseases, in Jenner’s view, but these could actually travel from one to the other. This is the fundamental reason why Jenner’s ideas were received with mockery and fantasies of transmogrification of humans into sick beasts on the side of the general public (Stern and Markel 2005), and as “Jacobin innovation” on the side of the Royal Society (Fulford and Lee 2000, p. 142). In this historical context, and given the outspoken opposition of the Suttons and other sceptics of Jenner’s work who made a living from variolating, and given the fragility of the case for vaccination (Baxby notes that “Jenner’s actual experimental evidence as presented in the Inquiry was not great and was not to be significantly increased in his later monographs” [Baxby 1999, p. 306]), euphoria for the new preventative treatment could have easily ebbed away. Luckily, Jenner had a number

192

M. Sand and L. Chiapperino

of influential, aristocratic proponents, who carried the fire of vaccination on (Baron 2014), and future medical progress provided a remedy for some of the early flaws of the method. This shall be discussed in more detail now from the perspective of serendipity. Vaccination and Serendipity as an Emergent Property Let’s turn to the question of serendipity. The term “serendipity” was coined by Horace Walpole, who wrote to Horace Mann: “I once read a silly fairy tale, called The three Princes of Serendip: as their Highnesses travelled, they were always making discoveries, by accidents and sagacity, of things which they were not in quest of”. (Merton 2004, p. 2; own emphasis) The statement suggests two necessary conditions for a discovery to be considered serendipitous. The first is the presence of wisdom or sagacity along with the accident: A person or collective has to have the right mindset or the “keen eye” to recognize the value of what they stumbled over (Macfarlane 1984). The words attributed to Louis Pasteur, who claimed that “chance only favors the prepared mind”, resonate with this (Bennett and Chung 2001). The second condition is the role of accident: That this “stumbling over something” is indeed accidental—whatever is found has not been sought and comes as a surprise (van Andel 1994). The traditional serendipity perspective applied to the case is that Jenner was made aware of the potential of vaccination by a milkmaid, who contracted cowpox and allegedly never contracted smallpox despite nursing smallpox patients (Roberts 1989). In an article by Edward Compere from 1957 in the Journal of the American Medical Association, we find this traditional account of the so-called “milkmaid myth”, which is adopted by Royston Roberts and many others (Barquet and Domingo 1997; Compere 1957b). Roberts relies in his account entirely on Compere’s, as he cites the following passage (p. 20): Jenner did not discover his vaccine as a result of long and arduous work in a laboratory. At the age of 19 years he was told by a former milkmaid that she could never have smallpox because she had had cowpox. Jenner recalled this statement when later, as a physician, he realized the futility of trying to treat the disease. […] This was true serendipity. The fact that cowpox gave immunity to smallpox came to him without effort on his part. He had the good judgement to recognize its value and to make use of it. (Compere 1957a)

In his analysis, Compere underscores that both conditions are met and stresses particularly the first: He emphasizes that Jenner had made no effort to find such insight. He bumped by accident into the milkmaid, who gave him the clue. This account of the events has been hawked by Jenner’s biographer John Baron (Baron 2014). Current historians of medicine, most outspokenly Arthur Boylston, contest it (Boylston 2018). It is more likely that Jenner learned of the possible immunizing effects of cowpox from Andrew Fewster during his years as a student physician in 1768. Jenner himself suggested in 1801 that his work on cowpox “commenced upwards of twenty-five years ago”. (Jenner 1801) He himself confirmed in these publicized reflections that the milkmaids’ immunity was “known among the dairies” at the time: “On inquiry, it appeared that it had been known among the dairies time immemorial,

10 Serendipity, Luck and Collective Responsibility in Medical …

193

and that a vague opinion prevailed that it was a preventive of the Small Pox”. (Boylston 2018; Jenner 1801, p. 505 f.). In the original Inquiry, Jenner vaguely refers to “prevailing notions” of such causal relation (Jenner 1801, p. 506, 1802a). But, what does that tell us about the accidentalness of the finding? On the one hand, we might say that it can still be regarded as an accident that it was Jenner, who was around the dairies at the time. He has not chosen to be born and practice medicine in Gloucestershire in the second half of the eighteenth century in order to find a smallpox vaccine. In this manner, one might consider his luck circumstantial—he was at the right place, at the right time (Sand 2020; Williams and Nagel 1976). This relates to another sub-question about the intent (or lack thereof) that might have motivated his behavior at the time: Had he been consciously receptive of those notions? Keeping an open eye might as well count as an effort on his part. Walpole suggests with his illustrative example, taken from a fairy tale that was popular at the time about the Princes of Serendipity; the Princes were “not in quest of” the things they found on their journeys, and Compere assumes that since Jenner has made no effort to meet the milkmaid, he hadn’t been on the lookout for vaccination either. But this is a psychological hypothesis. Even if the “milkmaid myth” hawked by Baron and Compere, were true—which it probably isn’t—we would have to determine, whether Jenner has been “in quest of a preventative treatment for smallpox” or in quest of anything at all, to definitely determine, whether this is a case of serendipity in the Walpolean sense. Answering this psychological question from our historical vantage point seems impossible. It is not unlikely, though, that he—like many other practicing physicians at the time—was very aware of the shortcomings of the established method of variolation and always more or less dedicatedly on the lookout for a better procedure—even if not necessarily in the place where he eventually found it. Thus, one might claim that he indeed “sought of” a way of improving the method of variolation, when (by accident) crossing the path with Andrew Fewster and the rumours that circulated at the time (though he might not have sought exactly the type of method that he found). Donald Gillies presents a similar interpretation of Fleming’s discovery of Penicillin: Fleming was on the lookout for antiseptic agents, though he didn’t expect to find them in a mould that accidently showed up in his laboratory (Gillies 2015). That Jenner’s sagacity played a role in the discovery seems less controversial: Jenner recognized the potential of the insight and dedicated to it his systematic attention, to become the most important and outspoken proponent of the procedure at the time (Baxby 1999). There is another understanding of serendipity that equally—if not better—lends itself to this case: Serendipity as an emergent property of collective action (Copeland 2017). As mentioned above—in one way or another—Jenner was on the receiving end and entered in at a later stage of a discovery process that had originated in folk knowledge. Whether by repeated witnessing of farmers bragging about never having “an ugly pockmarked face”, by being confronted with rumours in the dairies or contact with Fewster, Jenner was led by others on the trail of cowpox inoculation (Barquet and Domingo 1997, p. 639). Furthermore, aside from the previously

194

M. Sand and L. Chiapperino

mentioned patronage that helped to secure continuous support of Jenner’s ideas (they were fragile in their beginnings to say the least) (Baron 2014; Bazin 2000; Burrell and Kelly 2014, p. 876), this points to a collective genealogy of the ‘eureka’ (serendipitous) moment in this discovery. Further, making vaccination a commonplace in modern times included much more than detecting its basic principle: It relied on the experiences, lore, support and patronage of other unacknowledged people. That is, the global success of vaccination that ultimately led to the first and only eradication of a disease required more than just the discovery of the basic technique of vaccination or even a number of prominent proponents: As Andrea Rusnock (2009) reminds us, it was no small feat to actually gather and deliver suitable cowpox material to provide vaccination in the early days, another reason to adhere to a human-to-human transmission of pox material (see also Bazin 2000, p. 185): The concerted effort to spread cowpox as a beneficial prophylaxis thus provides a very different perspective on the globalization of disease. Exploring how Europeans and their colonial allies transported and maintained cowpox in new environments is a social and technological story involving a broad range of individuals from physicians and surgeons to philanthropists, clergy, and colonial administrators. (Rusnock 2009, p. 18)

To cope with the initial flaws of the procedure—its susceptibility to transmit other viruses and bacteria—improvements in hygiene and new sterilization techniques were required. Those were discovered around the middle of the nineteenth century by Ignaz Semmelweis and Joseph Lister (Lister 1867). It nevertheless took until the early twentieth century before those flaws were entirely eliminated from the procedure. Most importantly, the concerted and global effort beginning in the 1960s of the WHO included training laypeople how to vaccinate and policies that managed to reach the most rural areas of the world (Williams 2010). This was a “unique” and exceptional global effort that could have failed at more than one point and, in fact, almost did fail (Barrett 2007). Only through these collective and incremental efforts (in a diachronic and synchronic sense) could vaccination against smallpox become a pillar of global health. In this manner, as Copeland writes, serendipity becomes an emergent property of a complex process that involves various actors and their diverse relationships (Copeland 2017, 2018). Seen from this intriguing perspective, we recognize that the medical procedure of vaccination itself is not a static, separate artefact that maintained a concise form or shape over the centuries. Through its embedding in society, its various regulations and the impact of related medical practices and knowledge, its guise and impact on society changed over time. Vaccination is in a sense a dynamic artefact and a constant production of many hands (van de Poel 2015). In many instances, in which individuals are singled out to be attributed the status of the sole discoverer, this represents the forceful reduction of the historical complexities of the process of discovery (Schaffer 1986). What is considered a discovery, a finding that is in some way significant, can be acknowledged as being significant only after it is contextualised by others, calibrated by others in relation to existing knowledge and artefacts, being credited by others as a novelty and diffused by others to receive

10 Serendipity, Luck and Collective Responsibility in Medical …

195

wider social recognition. A discovery is in this manner always distributed amongst the behaviour of several individuals, various locations and moments in time. In any of the aforementioned phases, luck can—and most likely will—play a role. Consequently, we can assume that serendipitous discovery is rather the norm than the exception: The majority of known discoveries went through the aforementioned steps, thus containing facets that were uncontrolled and unforeseen (Trout 2016, 2019). Moral Luck and Individual Responsibility In the following, we will focus on the aspect of luck—not so much in the unexpected occurrence of an insight that Jenner managed more or less well to capitalize on—but also the plain luck that this was indeed not just a baseless hunch that one disease could protect from contracting another, but a real insight whose underlying mechanism was only decades later fully understood. Furthermore, we want to consider the luck that led vaccination to become a global success story despite the procedures’ initial flaws, some of them caused by Jenner himself. Each of these developments seems to have affected positively the way in which we evaluate Jenner’s decision today and elevated his recognition and status as a scientist. Had the experiment not succeeded, which— as he himself anxiously recognized—was a very real possibility and which was beyond certainty and control, we would assess him differently today. As mentioned before, it seems that this luck has a moral dimension as it affected our ultimate evaluation of Jenner’s scientific merits and desert, which are expressed in the public commemorations of his achievements. We shall begin with the first recorded vaccination experiment from 1796, which he describes in his Inquiry. The experiment is described in Jenner’s original publication from 1798 on page 20 to 22 as case 17. The case descriptions before this one document circumstantial observations: They present Jenner’s recollection of patients, who contracted the cowpox and who assured him that even during latent smallpox surges, they didn’t contract the disease. Based on this, he forms the intention to prove that there is indeed a connection between contracting cowpox and being immune to smallpox. We can read: The more accurately to observe the progress of the infection, I selected a healthy boy, about eight years old, for the purpose of inoculation for the Cowpox. The matter was […] inserted, on the 14th of May, 1796, into the arm of the boy by means of two superficial incisions […]. On the seventh day he complained of uneasiness in the axilla, and on the ninth he became a little chilly, lost his appetite, and had a light headache […]. In order to ascertain whether the boy, after feeling so light an affection of the system from the Cow Pox virus, was secure from the contagion of the Small Pox, he was inoculated on the 1st of July following with variolous matter. Several light punctures and incisions were made on both his arms, and the matter was carefully inserted, but no disease followed. (Jenner 1802b, pp. 20-22)

This experiment can rightly be considered dubious: On a thin evidential basis (Baxby calls it “scanty”; Baxby 1999, p. 302,), Jenner put the life of a boy at risk. Children are highly vulnerable research subjects, since they cannot properly assess the risks entailed in a study—hence, their capacity for informed consent is limited. Furthermore, children were much more likely to die from smallpox than adults (Baxby 1999), of which Jenner must have been aware. So, why choose a healthy child? In

196

M. Sand and L. Chiapperino

his own account presented in On the origin of the Vaccine Inoculation, he notes the anxiety and uncertainty that preceded the experiment: During the investigation of the Cow-pox, I was struck with the idea that it might be practicable to propagate the disease by inoculation, after the manner of the Small-pox, first from the Cow, and finally from one human being to another. I anxiously waited some time for an opportunity of putting this theory to the test. At length the period arrived. [After the experiment,] I could scarcely persuade myself the patient was secure from the Small-pox. (Jenner 1801, own emphasis, p. 507)

Bazin, a chronicler of smallpox, suggests that “what we know of [Jenner’s] work in no way indicates that he transgressed the morality of his period (and probably also of the present day), considering what was known at that time”. (Bazin 2000, p. 186) Yet, Bazin notes that “[t]here had been a little more risk for James [more than for Sarah Nelmes—the milkmaid, from whom the cowpox material was taken] but it was, nevertheless, fairly limited” (p. 38). In general agreement with this, Hugh Davies argues that in this case the benefits clearly outweighed the risks. He writes that Jenner’s experiment “addressed a major risk to the health of the community, and, given the devastating nature of smallpox and the significant risk of variolation, the only alternative preventative measure, Jenner’s study had purpose, justification and a base in the practice of the day”. (Davies 2007, p. 174) It should be clear that such utilitarian reasoning would by no means be acceptable nowadays to justify unconsented medical experiments with vulnerable populations. Has it been during his day and age? Furthermore, even if we accept the general view that some major potential benefits outweigh some minor risks, this only justifies the general motivation behind the experiment and not the manner in which it was conducted. There were alternatives that would have mitigated the risks to some extent—we will allude to those in a moment. Clearly, one might object that we are assessing him uncharitably according to rather recently established bioethical standards, which did not prevail during his time. Considering his work from the vantage point of today means doing him a great injustice. We argue, however, while biomedical regulations and the method of systematic animal trials did not exist at the time, a debate about good medical practice had begun to flourish: In the treatises of John Gregory, a Scottish contemporary of Jenner, we find admonitions to doctors to not capitalize on the vulnerability of their poorer patients (McCullough 1998; Strätling 1997; Truman 1995). Gregory also suggested that doctors should never recommend medical treatments that they wouldn’t use on themselves or their own children. Jenner abided by neither of those admonitions. He tested vaccination on Phipps two years before he tested it on his own children and Phipps was the son of his penniless gardener (Williams 2010). Whether Jenner offered a financial incentive to participate is unknown. It is unlikely though, given the dependence of medical practitioners on their oftentimes financially affluent aristocratic patrons and donors, that Jenner would have chosen a child from a more affluent family for the experiment (Jewson 1974). It is not unlikely that there was at least implicit pressure for Phipps’ parents to let their child participate, which capitalizes on their vulnerability and undermines their autonomy for informed consent.

10 Serendipity, Luck and Collective Responsibility in Medical …

197

One might compare Jenner’s approach to that of Lady Montague, who decided to prove her conviction in the efficacy of variolation by first variolating her own child.7 This left a decent impression on the British aristocracy at the time. However, the royal family of George III. still did not entirely trust the procedure. The devastating outlook of contracting smallpox did not convince everyone to try whatever new procedure was in fashion.8 They decided to test the efficacy of variolation on several adult prisoners first (Bazin 2000, p. 13). What does that show? On the one hand, that adults, if they were not inoculated before, were to some extent considered as suitable research subjects, too. It shows, on the other hand, that the danger of these sorts of medical “experiments” were known and people consciously considered how to mitigate the risks for those societal members who allegedly mattered most—in this case the royal children. Exposing people to such risks was reserved for those who mattered less in the public eye: Prisoners awaiting capital punishment belonged to this group.9 Surely, that is not an ethical choice either: Here too, pressure has likely undermined the autonomy of the research subjects for legitimate consent. But the case serves to show that the awareness of the risks for research participants and mitigating strategies existed at the time. In 1796, Jenner had been waiting for some time for the cowpox to return to Gloucestershire. To not miss the opportunity, he probably took the next best subject available to him—someone who had not yet been variolated by the Suttons: It was a choice of impatience and maybe convenience. Conclusively, it is fair to say that Jenner imposed a considerable risk on the research subject. Alternative designs for the experiment, which might have mitigated the risks involved in the procedure also from his viewpoint, were comfortably available to him; a more mindful approach to the testing of inoculation was thus a concrete option he discounted for the sake of pursuing success (Sand et al. 2019).10 As the experiment succeeded and Phipps survived and became immune to smallpox, the public recollection tends to focus primarily on celebrating the massive, positive impact Jenner’s innovation had on public health. The dubiousness of this practice has vanished from public recollection. Jenner received all the big honours suitable for someone who is considered a medical revolutionary. A society and an institute were dedicated in his honour, towns and streets were named after him, several statues were erected (most notably the one in Kensington Garden, which stood originally at Trafalgar Square) and several biographies have been written. Current 7

“On returning to England, Lady Montagu had her daughter inoculated with smallpox in public, by her surgeon, “Doctor” Maitland, in order to show, by example, the advantage of the method. […] Lady Mary was a capable woman- and brave, for the risks were very real!” (Bazin 2000, p. 12 f.). 8 In times of enormous qualitative diversity in medical practice, awareness of the widespread charlatanry was important (Jewson 1974). 9 As the royal family still didn’t feel 100% sure about the procedure, they accepted orphans as research participants in the following step. 10 Note, this does not imply making Jenner a moral and scientific outcast and erasing him from scientific history. Our plea for a balanced assessment is most difficult to convey as it draws a picture beyond a binary scheme of good and bad: The dubiousness that we underscore must be as much part of the evaluation as his awareness and recognition of the potential of the method at hand.

198

M. Sand and L. Chiapperino

commentators—even those who highlight the role of the other contributors to the success of vaccination—predominantly praise Jenner and argue that he deserves to be called a medical “hero” amongst others (e.g. Andrew Fewster) (Gross and Sepkowitz 1998), and his “life story remains an inspiration to physicians”. (Willis 1997). In an overall extremely balanced account of the history of smallpox Angel of Death, we cannot find any hint that Jenner’s experiment was morally dubious (Williams 2010). It is only Royston Roberts who wonders “how he [Jenner] persuaded the boy and his parents to take such a chance”. (Roberts 1989, p. 21). Conclusively, to put it in terms of a famous notion of Bernard Williams; success gave him right (Williams and Nagel 1976). James Phipps was not harmed—though the possibility of him passing away due to an unknown health issue is easily conceivable. Because he was not harmed, Jenner using him as a guinea-pig seems less relevant or even irrelevant to an evaluation of his doing so. The fact that Jenner succeeded in an experiment of which he couldn’t have known with certainty that it would succeed, makes him morally lucky. The “success though lucky” state overshadows his thoughtless experimental design and the prematurity of his 1798 publication. Are these such evaluations in hindsight justifiable, given the lack of control that he had over the subsequent events including the necessary contributions by others to the success of vaccination? What weight should these things beyond his control have on the way we commemorate him? It is clear that in a counterfactual world, in which vaccination would have turned out to be a fluke and Phipps had died, or vaccinations’ advocacy had ebbed away due to the devastating side-effects of the original procedure, his reputation would be a different one: No Jenner institute, no fame, no statues. Can such a paradox of moral luck be avoided? Moral Luck and Value Trade-offs in Science Stratification Philosophers have wrestled with the problem of moral luck over millennia: People ought not to be assessed for things beyond their control, but they apparently are assessed for exactly such things (Williams and Nagel 1976). Jenner is commemorated for the success of vaccination, although he could not have been certain, nor has he alone brought it about. Luck is in this manner often seen as an adversary to moral responsibility: If accidents occur, no one should be held responsible. Thus, if serendipitous discoveries are in fact substantively accidents, we must refrain from praising people for “making” such discoveries. And why should we refrain from assessing someone for things beyond her control—for being lucky? It might be suggested that this would be unfair: It is unfair that some people fail in their endeavors and others succeed owing to luck and we take this into consideration in our allocation of praise and blame—some are not in quest of something, but find it, others are also not in quest of something and it is only lack of luck that stops them stumbling over it. In this particular case, it is difficult to imagine who the complainant would be: If fairness is a problem, it seems less so in cases of praise and commemoration. There is an asymmetry between praise and blame in this regard: Prima facie, praise does not really harm anyone and neither does the omission of praising (King 2014). Hence, unlike someone who is punished for things beyond control, Jenner has no reason for complaint: He was not assailed by anything adverse to his interests for things beyond

10 Serendipity, Luck and Collective Responsibility in Medical …

199

his control. Neither did anyone else have to incur harm as a result of the praise he received. Or so it seems: But, what about other people, whose relative standing might be denigrated in comparison? Clearly, there is no counterfactual Jenner, who is identical in all regards to the actual Jenner, and whose experiment failed, and who could now complain that he was just unlucky and must be treated equally for fairness reasons. This could mean that neither the actual nor the counterfactual Jenner receives praise and rewards, or that both do. Surely, in real life, such comparisons between real and fictitious people are highly speculative: Was there another keen-eyed, sagacious physician in Europe at the time, who would have rightly assessed the gravity of this notion, had she been in contact with the rumors in the dairies? Someone who thus deserves to be praised too, given that this was a matter of circumstantial luck—beyond both their control. Whether such a person existed, we will never know. We write the history of medical innovation after the fact, once the success has occurred. A successful innovation, even if partially lucky, provides evidence that indeed someone has had the laudable skills and virtues that were required to recognize its value. Luck, thus, provides evidence for the existence of admirable traits—such as Jenner’s sagacity. Without such evidence, without having a recognizable social impact of an innovation, we can only speculate about an agent’s laudable traits (Rescher 1995). But, it doesn’t seem that Nobel Prizes are granted only for such epistemic reasons to those who have “conferred the greatest to mankind”. (Nobel 1895) The rewards fulfill important societal functions and here the visible impact on society might be crucial for their efficacy. Prizes are supposed to encourage and incentivize the pursuit of desirable research and the best way to achieve this might be to link them directly to such type of research, lucky or not (Sand 2020). This is not limited to Nobel Prizes—it might also appertain to the aforementioned ways in which Jenner was (and still is) publicly commemorated with statues and streets named after him. Harriet Zuckerman has suggested that “[rewards in science] serve much the same purpose as they do in other institutional spheres. They validate past performance and provide a degree of motivation for the future [own emphasis]. They bring attention to performance judged to be of high quality, thereby reinforcing the standards by which performance is to be assessed”. (Zuckerman 1970, p. 252) Equally, in Responsible Research and Innovation (RRI), which has been a leading science and innovation governance programme in the European Union, the explicit goal is to “invite, accommodate, stimulate, enhance, foster and incentivize responsible action”, following the Directorate General for Research and Innovation (van de Poel and Sand 2021). It might seem, thus, that there is a justification for focussing on the outcomes of Jenner’s work despite the involvement of luck. Not only is it difficult if not impossible to identify those sagacious researchers and innovators, whose work remained incomplete and unsuccessful for absence of evidence that luck could have made a difference. Serendipitous discoveries can also be used to showcase the type of endeavours and outputs that society values and aims to foster. There remains a caveat though: Zuckerman also suggests that stratification validates past performance. But, Jenner’s past performance, as we have seen, boils

200

M. Sand and L. Chiapperino

down to the sagacity of acknowledging the potential of vaccination and prematurely publishing results of his experiments, which were impatiently and thoughtlessly cobbled together. This sounds less exuberant and much more sober than viewing him hyperbolically as a “medical hero”. Reward policies such as memorials, prizes and statutes aim at incentivizing and validating individuals’ responsibility. These tools are too crude to convey a balanced assessment of their praiseworthiness (see also footnote 10). It seems that such instruments fulfil their societal function inevitably by distorting or at least simplifying the complex truth regarding people’s desert. A notable constructive suggestion to remedy some of those downsides is this: Validations of past performances could at least acknowledge that luck and serendipity in terms of the influence of other individuals, the accumulated knowledge available, the circumstances and results that fall beyond scientists’ control, all contribute to successful “discoveries”. This might still not convey a balanced assessment of Jenner’s individual responsibility: A thick evaluation of his moral reasoning may even be beyond the grasp of present insight, as we suggested above. Yet, highlighting the role of luck and serendipity in this discovery could at least do justice to the complex historical, social and ethical genealogy of vaccination (Crawford 1998). We could, for instance, put Andrew Fewster, from whom Jenner learned about the possible relationship between cowpox and smallpox, up in the list of people who deserve more credit for their role in the history of vaccination. Or we might add early patrons of Jenner, who promoted the procedure and increased its public acceptance, to the list. While this seems a creditable move towards a fairer distribution of responsibility, we must beware that it won’t eradicate luck from the equation. Luck equally provides the preconditions required to bring an innovation collectively to fruition (Sand 2020). Scott Barrett points out that “[e]radication is a high stakes game, and could fail for any number of reasons. The ongoing effort to eradicate the regional disease, Guinea worm, for example, has been stopped in its tracks by the decades-long civil war in Sudan […]. The smallpox campaign was luckier. A ten-year cease-fire just happened to be in effect in Sudan during a key stage of the eradication effort: otherwise, eradication may not have been possible”. (Barrett 2007, p. 180) Thus, there have been innumerable types of circumstantial and resultant luck involved in the collective endeavor of advocating and improving vaccination and fighting smallpox. In this manner, serendipity as a collective endeavor is no less accidental than the paradigmatic individual strike of serendipity (Chiapperino 2020). Here too, it will remain a constant challenge for the future of science stratification with regard to collective innovations to balance those different aims: being proportionate to the desert of individuals in those endeavors and showcasing what is valuable to effectively incentivize future innovation. It seems that this will always require drawing more or less arbitrary boundaries between an agent’s contributions to an event that is societally relevant and desirable to be reproduced in kind, and the causal factors (including the actions of others) that were beyond this agent’s control yet without which the event would not have occurred.

10 Serendipity, Luck and Collective Responsibility in Medical …

201

Conclusions Serendipity and luck give rise to normative questions, of how we look at the moral standing of serendipitous individuals, how we separate their sagacity from the luck that led them to make their discoveries and the contributions of others that made their projects flourish. Incorporating this insight into the governance of science and science stratification is a challenge. Although it is suggested that focusing on luck can help us to make policies better to accommodate or eradicate it, it is by no means clear how to do this (Sauder 2020). There is a natural request that policies ought to be effective on the one hand and fair on the other: If they ought to be fair, they seem to require the eradication of luck from the picture. But, this seems impossible given luck’s ubiquity that also appertains to collective efforts. To sum: Fairness would require rewards based on merit, but merit is not sufficient for reward (sometimes not even necessary); therefore, rewards are not fair, contrary to their meaningfulness. An ethics of serendipity will have to illuminate these issues on both a descriptive and normative level. On the descriptive level, we need a better understanding of the variety of contexts in which serendipity occurs and the manifold ways in which luck affects processes of discovery. Further, we need a more concise idea of how to identify the agential contributions to discovery and how to expand their scope of control. On the normative level, we need to consider the values of fairness and effectiveness and develop a theory that balances these and other values. It is unlikely that fairness is always overriding (Sand and Klenk 2021). Hence, a theory of just reward that takes serendipity and luck into account might at least provide a heuristic that indicates when fairness can reasonably be overridden and when not. These are the key challenges for any future ethics of serendipity.

References Baron, J. 2014. The Life of Edward Jenner M.D.: With Illustrations of his Doctrines, and Selections from his Correspondence, vol. 1. Cambridge: Cambridge University Press. Barquet, N., and P. Domingo. 1997. Smallpox: The triumph over the most terrible of the ministers of death. Annals of Internal Medicine 127 (8 Pt 1): 635–642. https://doi.org/10.7326/0003-4819127-8_part_1-199710150-00010. Barrett, S. 2007. The smallpox eradication game. Public Choice 130 (1/2): 179–207. Baxby, D. 1999. Edward Jenner’s Inquiry; a bicentenary analysis. Vaccine 17 (4): 301–307. https:/ /doi.org/10.1016/S0264-410X(98)00207-2. Bazin, H. 2000. The Eradication of Smallpox: Edward Jenner and the First and Only Eradication of a Human Infectious Disease. San Diego, California: Academic Press. Beale, N., and E. Beale. 2005. Evidence-based medicine in the eighteenth century: The Ingen HouszJenner correspondence revisited. Medical History 49 (1): 79–98. https://doi.org/10.1017/s00257 27300008292. Bennett, J.W., and K.-T. Chung. 2001. Alexander Fleming and the discovery of penicillin. In Advances in Applied Microbiology, vol. 49, 163–184. Academic Press. Boylston, A. 2012. The origins of inoculation. Journal of the Royal Society of Medicine 105 (7): 309–313. https://doi.org/10.1258/jrsm.2012.12k044.

202

M. Sand and L. Chiapperino

Boylston, A. 2013. The origins of vaccination: Myths and reality. Journal of the Royal Society of Medicine 106 (9): 351–354. https://doi.org/10.1177/0141076813499292. Boylston, A.W. 2018. The myth of the milkmaid. New England Journal of Medicine 378 (5): 414–415. https://doi.org/10.1056/NEJMp1715349. Burrell, R., and C. Kelly. 2014. Public rewards and innovation policy: Lessons from the eighteenth and early nineteenth centuries. The Modern Law Review 77 (6): 858–887. Chiapperino, L. 2020. Luck and the responsibilities to protect one’s epigenome. Journal of Responsible Innovation 7 (sup2): S86–S106. https://doi.org/10.1080/23299460.2020.1842658. Compere, E.L. 1957a. Research, serendipity, and orthopedic surgery. Journal of the American Medical Association 165 (16): 2070–2073. https://doi.org/10.1001/jama.1957.029803400 36010. Compere, E.L. 1957b. Research, serendipity, and orthopedic surgery. JAMA 165 (16): 2070–2073. https://doi.org/10.1001/jama.1957.02980340036010. Copeland, S. 2017. On serendipity in science: Discovery at the intersection of chance and wisdom. Synthese. https://doi.org/10.1007/s11229-017-1544-3. Copeland, S. 2018. “Fleming leapt on the unusual like a weasel on a vole”: Challenging the paradigm of serendipity in science. Perspectives on Science 26 (6): 694–721. Crawford, E. 1998. Nobel: always the winners, never the losers. Science 282 (5392): 1256–1257. https://doi.org/10.1126/science.282.5392.1256. Davies, H. 2007. Ethical reflections on Edward Jenner’s experimental treatment. Journal of Medical Ethics 33 (3): 174–176. https://doi.org/10.1136/jme.2005.015339. Fulford, T., and D. Lee. 2000. The jenneration of disease: Vaccination, romanticism, and revolution. Studies in Romanticism 39 (1): 139–163. https://doi.org/10.2307/25601434. Gillies, D. 2005. Hempelian and Kuhnian approaches in the philosophy of medicine: The semmelweis case. Studies in History and Philosophy of Biological and Biomedical Sciences 36 (1): 159–181. https://doi.org/10.1016/j.shpsc.2004.12.003. Gillies, D. 2015. Serendipity and chance in scientific discovery: Policy implications for global society. In The Handbook of Global Science, Technology, and Innovation, ed. D. Archibugi and A. Filippetti, 525–539. Wiley-Blackwell. Gross, C.P., and K.A. Sepkowitz. 1998. The myth of the medical breakthrough: Smallpox, vaccination, and Jenner reconsidered. International Journal of Infectious Diseases 3 (1): 54–60. Hales, S.D. 2020. The Myth of Luck—Philosophy, Fate and Fortune. London: Bloomsbury Academic. Helmstädter, A. 2008. Zur Geschichte der aktiven Immunisierung. Vorbeugen ist besser als Heilen. Pharmazie in Unserer Zeit 37 (1): 12–18. https://doi.org/10.1002/pauz.200700247. Jenner, E. 1801. On the origin of the vaccine inoculation. The Medical and Physical Journal 5 (28): 505–508. Jenner, E. 1802a. Inquiry into the causes and effects of the variolae vaccinae. In ed. D.S. Cooley. Springfield: Ashley & Brewer. Jenner, E. 1802b. Inquiry into the causes and effects of the variolae vaccinae. A disease discovered in some of the western counties of England, particularly Gloucestershire, and known by the name of the cow pox. In ed. D.S. Cooley. Springfield: Ashley & Brewer. Jewson, N.D. 1974. Medical knowledge and the patronage system in 18th century England. Sociology 8 (3): 369–385. https://doi.org/10.1177/003803857400800302. Jewson, N.D. 1976. The disappearance of the sick-man from medical cosmology, 1770–1870. Sociology 10 (2): 225–244. https://doi.org/10.1177/003803857601000202. King, M. 2014. Two faces of desert. Philosophical Studies 169 (3): 401–424. https://doi.org/10. 1007/s11098-013-0188-5. Lister, J. 1867. On the antiseptic principle in the practice of surgery. British Medical Journal 2 (351): 246–248. https://doi.org/10.1136/bmj.2.351.246. Macfarlane, G. 1984. Alexander Fleming: The Man and the Myth. London: Chatto & Windus, The Hogarth Press.

10 Serendipity, Luck and Collective Responsibility in Medical …

203

McCullough, L.B. 1998. John Gregory and the Invention of Professional Medical Ethics and the Profession of Medicine, vol. 56. Dordrecht: Kluwer Academic Publishers. Merton, R.K. 2004. The Travels and Adventures of Serendipity: A Study in Sociological Semantics and the Sociology of Science. New Jersey: Princeton University Press. Nobel, A. 1895. Will. Retrieved from https://www.nobelprize.org/alfred-nobel/full-text-of-alfrednobels-will/. Accessed on 20 Aug 2018. Pead, P.J. 2003. Benjamin Jesty: New light in the dawn of vaccination. The Lancet 362 (9401): 2104–2109. https://doi.org/10.1016/s0140-6736(03)15111-2. Pead, P.J. 2006. Benjamin Jesty: The first vaccinator revealed. The Lancet 368 (9554): 2202. https:/ /doi.org/10.1016/S0140-6736(06)69878-4. Razzell, P.E. 1965. Edward Jenner: The history of a medical myth. Medical History 9 (3): 216–229. https://doi.org/10.1017/S0025727300030714. Rescher, N. 1995. Luck: The Brilliant Randomness of Everyday Life, 1st ed. New York: Farrar Straus & Giroux. Roberts, R.M. 1989. Serendipity: Accidental Discoveries in Science. New York: Wiley. Rusnock, A. 2009. Catching cowpox: The early spread of smallpox vaccination, 1798–1810. Bulletin of the History of Medicine 83 (1): 17–36. Sand, M., A.L. Bredenoord, and K.R. Jongsma. 2019. After the fact—the case of CRISPR babies. European Journal of Human Genetics 27 (11): 1621–1624. https://doi.org/10.1038/s41431-0190459-5. Sand, M. 2020. Did Alexander Fleming deserve the Nobel Prize? Science and Engineering Ethics 26: 899–919. https://doi.org/10.1007/s11948-019-00149-5. Sand, M., and S. Copeland. 2020. Luck as a challenge for the responsible governance of science and technology. Journal of Responsible Innovation. https://doi.org/10.1080/23299460.2020.184 8848. Sand, M., and M. Klenk. 2021. Moral luck and unfair blame. The Journal of Value Inquiry. https:// doi.org/10.1007/s10790-021-09856-4. Sauder, M. 2020. A sociology of luck. Sociological Theory 38 (3): 193–216. https://doi.org/10. 1177/0735275120941178. Schaffer, S. 1986. Scientific discoveries and the end of natural philosophy. Social Studies of Science 16 (3): 387–420. Stern, A.M., and H. Markel. 2005. The history of vaccines and immunization: Familiar patterns, new challenges. Health Affairs 24 (3): 611–621. Strätling, M.W.M. 1997. John Gregory (1724–1773) and his lectures on the duties and qualifications of a physician establishing modern medical ethics on the base of the moral philosophy and the theory of science of the empiric British enlightenment. Medicina Nei Secoli-Arte e Scienza 9 (3): 455–475. Thagard, P. 2011. Patterns of medical discovery. In Philosophy of Medicine, ed. F. Gifford, 187–202. Elsevier. Trout, J.D. 2016. Wondrous Truths: The Improbable Rise of Modern Science. New York: Oxford University Press. Trout, J.D. 2019. Luck in science. In The Routledge Handbook of the Philosophy and Psychology of Luck, ed. I.M. Church and R.J. Hartman, 391–400. New York: Routledge. Truman, J.T. 1995. The compleat physician: John Gregory MD (1724–1773). Journal of Medical Biography 3 (2): 63–70. https://doi.org/10.1177/096777209500300201. van Andel, P. 1994. Anatomy of the unsought finding. Serendipity: Origin, history, domains, traditions, appearances, patterns and programmability. British Journal for the Philosophy of Science 45(2): 631–648. van de Poel, I. 2015. The problem of the many hands. In Moral Responsibility and the Problem of Many Hands, ed. I. van de Poel, L. Royakkers, and S.D. Zwart, 50–92. New York: Taylor & Francis. van de Poel, I., and M. Sand. 2021. Varieties of responsibility—two problems of responsible innovation. Synthese 198 (19): 4769–4787. https://doi.org/10.1007/s11229-018-01951-7.

204

M. Sand and L. Chiapperino

Williams, B., and T. Nagel. 1976. Moral luck. Proceedings of the Aristotelian Society, Supplementary Volumes 50: 115–151. Williams, G. 2010. Angel of Death: The Story of Smallpox. Basingstroke: Palgrave Macmillan. Willis, N.J. 1997. Edward Jenner and the eradication of smallpox. Scottish Medical Journal 42 (4): 118–121. https://doi.org/10.1177/003693309704200407. Zuckerman, H. 1970. Stratification in American science. Sociological Inquiry 40 (2): 235–257. https://doi.org/10.1111/j.1475-682X.1970.tb01010.x.

Martin Sand is an Assistant Professor of Ethics and Philosophy of Technology at TU Delft. In 2020, he was a member of a theme group on “Accountable and Explainable Medical AI” at the Netherlands Institute for Advanced Study (NIAS). Before, he undertook a two-year project on the topic “Moral Luck in Science and Innovation” as a Marie Skłodowska-Curie-Fellow. He is a member of the scientific advisory board of the Journal for Technology Assessment in Theory and Practice and an editorial board member of the journal Philosophy of Management. Luca Chiapperino graduated in moral philosophy at the University of Rome “La Sapienza”, before doing a PhD in Bioethics and STS at the Università degli Studi di Milano and the European School of Molecular Medicine (SEMM), on the relationship between the science of epigenetics and claims of responsibility, and empowerment in healthcare policy-making. Currently, he is appointed Lecturer at the Institute of Social Sciences, Faculty of Social and Political Sciences, of the University of Lausanne, Switzerland.

Chapter 11

Serendipity Across Contexts: From Offices to Post-conflict Settings Michael Soto

Abstract Michael Soto’s life and work spanning from Colombia to the United Kingdom has been full of serendipity and he aims to keep it this way. In this contribution, Soto takes the reader on a journey between these two places and outlines how he became involved in the practice of institutionalizing serendipity, developing an organizational process known as Randomised Coffee Trials that has been used across the globe by nonprofits, governments, and companies. He highlights the role of networks in serendipity, while emphasizing that it depends not just on the patterns of interactions as might be represented in a network map, but also on how individuals interact. The chapter juxtaposes two distinct contexts; offices and post conflict settings, which helps make salient the skill-, and resource-inequities between people in those settings and their ways of dealing with vulnerability and shortage. In conclusion, Soto argues that serendipity can be institutionalized and is more likely to happen when we understand the process behind it, and that positions in networks are intertwined with privilege and inequity, which affects the scale of impact of serendipity in those different contexts.

My walks with Pedro Medina provide a closeup view of serendipity in action. Over the course of several weeks in 2010, I would meet at his apartment in Bogotá, Colombia and we would walk together the ten or so blocks to his office. The entire way there, Pedro would speed up or slow down to start up conversations with strangers on the sidewalk. There was a similar script, nearly always starting with “Hello neighbor!” which he subsequently adapted to the person’s varying degrees of confusion. When he would walk into an elevator, he would turn to the others and ask, “Is this an elevator where it is permissible to speak?” People would laugh and the ice was broken. Pedro often talked about the power of serendipity and social capital, and it was through his mentoring that I too became interested in serendipity. Pedro encouraged me and others to discover the unexpected wonders that lay hidden at our fingertips. He would often say that if we would only start a conversation with the person next M. Soto (B) University of Minnesota Twin Cities, Minneapolis, MN, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 S. Copeland et al. (eds.), Serendipity Science, https://doi.org/10.1007/978-3-031-33529-7_11

205

206

M. Soto

to us on the bus, or in line at the store, we would be amazed by the resources and opportunities at our disposal. Pedro encouraged me, and others, to build relationships with those around us, and to approach those we knew with constant curiosity. Pedro has been instrumental in shaping how I think of serendipity. Over the past decade I have become a proponent of “institutionalising serendipity” via a process playfully called Randomised Coffee Trials (RCTs). Increasingly, I have emphasised the role of networks in fostering serendipity within organisations. Building on Pedro’s teachings, I have incorporated insights on social networks and collective memory that focuses on how the pattern of ties between people shape what they know and do not know. I have focused on how connecting disparate parts of social networks often leads to new insights. In 2013, I co-founded Spark Collaboration, a software platform to streamline a networking process I helped developed at Nesta, the UK innovation foundation. Since then, we have helped thousands of individuals make more unexpected discoveries via coffee, tea, or video conference chats with a coworker. In 2016, I began working towards a Ph.D. in Sociology focused on the social reintegration of ex-combatants and have conducted research in Colombia and Northern Ireland. Serendipity is not a central concept of my dissertation research but in many ways, it has been implicit in my approach to understand how relationships form between ex-combatants (FARC, IRA, UDF, UVF) and other civilians following peace agreements. I have generally avoided speaking of serendipity in relation to post-conflict studies because it seemed naive to link coffee between co-workers and civilian encounters with former combatants. But in this chapter, I will do just that. I first review some of the literature that shapes my understanding of serendipity and networks. I then describe how I became a proponent for institutionalising serendipity and then use this as a lens to re-examine my doctoral fieldwork research. The sharp turn in my personal experience from promoting collaboration in offices to examining ex-combatant reintegration provides a vantage point to juxtapose these two very different contexts, which in turn presents an opportunity for further understanding serendipity. Serendipity, Relationships, and Networks In the eighteenth century, Horace Walpole coined Serendipity in describing how the three princes of Serendip “… were always making discoveries, by accidents and sagacity, of things which they were not in quest of…” (Walpole, as quoted in Merton and Barber 2004, p. 2) Much of the focus since then has been on the unexpected and uncontrollable (luck) and the skill of the individual to make the most of the circumstances (sagacity). This representation focuses on the present moment and makes planning for serendipity appear paradoxical (Sand and Copeland 2020). A further complication is that serendipity, as with beauty, is in the eye of the beholder. The same action or event may be seen as serendipitous by one but not by another. Taking this even a step further, an individual may describe a course of events serendipitous at one moment but later decide it is not. Consequently, much of the focus has been on the individual experiencing serendipity. As Samantha Copeland indicates,

11 Serendipity Across Contexts: From Offices to Post-conflict Settings

207

Analyses of serendipity commonly focus on an insightful, cognitive connection made by a ‘serendipitous’ individual. However, serendipity as experienced by scientists and by others consists of multiple kinds of connections, including social relationships and timeliness. (Copeland 2019, p. 2386)

Building on Copeland (2019), I too wish to expand the time horizon and emphasise the role of social relationships. While the cognitive focus in the scholarship is in line with Walpole’s initial description of sagacity and the powers of observation exemplified by the three princes, it leaves unopened the black box of what leads to the cognitive experience. Rather than seeing it as a mysterious and magical moment, many writers emphasise how serendipity can be planned, in the sense that the likelihood of it happening can be systematically increased (Busch 2020; Copeland 2019; Kennedy 2016; Lindsay 2013, 2014; Olma 2012; Soto 2013, 2016, 2020). My take on serendipity places relationships, networks, and ways of interacting at the center. Serendipity most often occurs either when individuals from relatively disconnected parts of a network connect, or new meaning is given to a prior connection. As Ron Burt has put it, “The creative spark on which serendipity depends, in short, is to see bridges where others see holes” (Burt 2004, p. 351). The etymology of serendipity stems from a story of the three Princes of Serendip who described a lost donkey so well they were subsequently suspected of having stolen it. They claimed never to have seen it, yet described it as lame, and carrying butter on one side and honey on the other. When asked to explain how they knew they said that the animal left a trail of three sets of hooves and dragged the fourth. Additionally, ants clustered around one side of the path and flies around the other. For the three princes this seemed like an obvious conclusion based on observation and deduction, but the audience was in awe. What is left out of the story is, how they knew. How did they go from their observations to those conclusions? (see also Arfini, Chap. 7 in this volume, on this). Their “sagacity” rests on an embodied set of experiences and relationships. They would not have been as observant had they been fleeing for their lives or had they been without food and water for days. Nor would someone unfamiliar with the countryside have been able to draw those conclusions from the separate trails of ants and flies. One’s life trajectory is a composite of experiences and relationships, shaped by the social contexts one inhabits. Serendipity can occur across a range of contexts. It is these contexts that shape one’s ability to explore paths, especially those that appear unlikely to provide an immediate return, and when they are explored, the potential scale of impact of any serendipitous discoveries made along the way. Just as social inequities are reflected in spatial arrangements from neighborhoods, to schools, to workplaces, the corresponding social networks an individual is a part of, in turn, reflect those inequities. The opportunities for serendipity, while not determined, are heavily influenced by the social networks individuals are a part of and those within reach. The adjacent literature on idea generation provides a useful lens to examine serendipity. In Where Good Ideas Come From, Steven Johnson (2010) analyses the pseudo-eureka moment that is associated with inventors. Instead of conceiving this moment of epiphany as the realisation of one individual in a particular moment, he

208

M. Soto

shows how the epiphany process extends over a longer period of time and involves others. He talks about ideas in a social and interactive way. People carry around hunches, and Johnson suggests that “[h]unches that don’t connect are doomed to stay hunches” (Johnson 2010: 76). Contrary to the conception of the eureka moment Johnson shows by analysing Darwin’s diary that the roots of Darwin’s groundbreaking theory of evolution were present in entries long before he ties the different pieces together and sees it as a breakthrough. Similarly, Thomas Kuhn’s The Structure of Scientific Revolutions (1962) emphasises scientific discovery as a process rather than an eureka moment. While not all eureka moments are serendipity, and serendipity does not always happen in a moment, it is equally useful to examine serendipity as a process rather than a moment. Following Copeland (2019), I discuss serendipity as not just a snapshot in time between a set of actors, but as a longer process often involving multiple people and points in time. Information and ideas travel through relationships that serve as metaphoric bridges between individuals. The patterns of relationships between individuals are not evenly distributed, there are clusters among some that are closer and more interconnected, and there are gaps between those who have no immediate connection.1 The result is that some ideas can be commonplace within one cluster but novel in another. Consequently, Ron Burt has said, “creativity is an import–export game. It’s not a creation game” (Erard 2004). There is often a silo effect when interacting with people within one’s own cluster, where one is constantly exposed to the same set of ideas; unexpected discoveries are most likely to occur when interacting with others beyond one’s cluster. Ron Burt calls it a “structural hole” (2004) when there are no connections between clusters. It is often through an individual’s weak ties (Granovetter 1973; Maddox 2015) or dormant ties (Grant 2013, p. 50–52) that new information arises. The people an individual is closest with (strong ties) tend to also know each other (triadic closure) and are thus less likely to be sources for novel information. A useful illustration of social networks and the strength of weak ties can be seen by examining Stanley Milgram’s classic Small World phenomenon (1967). If I asked you to mail a letter to a specific person in Alaska, Paris, or São Paulo, your first reaction might be to doubt the purpose of the exercise. But if you were asked to send the envelope to someone closer, then you might remember that cousin, who likes to travel, or your friend learning French or Portuguese. In Milgram’s research study, he provided envelopes to individuals across the US with instructions to deliver it to an individual in Massachusetts, and if the person could not send it directly, they were asked to send it to someone who might be better positioned to getting it there. The catch was: participants could only send it to someone they knew personally. Through this study the “six degrees of separation” concept emerged; those envelopes that arrived tended to be within six intermediaries from the chosen destination.

1

In network terms, one can analyse the density of a subsection of a network by dividing the number of ties by the total possible number of ties. And the gaps between clusters are called “structural holes”.

11 Serendipity Across Contexts: From Offices to Post-conflict Settings

209

The social network that connects us to unknown individuals is not in the fore of most people’s minds, which is what generates a sense of surprise when people discover unexpected links. While Milgram does not use the word serendipity, the anecdotal story he begins his text with shares commonalities with serendipity stories. He describes two strangers in Tunis who discover they have a shared acquaintance in Detroit and then exclaim: “What a small world!”. Milgram states: “The importance of the [small world] problem does not lie in these entertaining aspects, but in the fact that it brings under discussion a certain mathematical structure in society, a structure that often plays a part, whether recognised or not” (Milgram 1967, p. 62) New opportunities always lurk beneath the surface, and a simple conversation helps them become more visible. For those interested in serendipity, this sheds light on how serendipity happens, and consequently how it can be made more likely to happen. We may think we know who in our network has information, resources, or contacts that may be useful for our objectives. However, there are often many opportunities hidden from view, due to assumptions developed through the interactions we have had with others. Mingling with strangers, and small talk with those we know can surface unexpected discoveries that would be missed were we exclusively goal-oriented in our interactions. Relationships with others drive performance and innovation. One of my favorite studies is by Brian Uzzi and Jarrett Spiro (2005), in which box office sales were taken as a success metric when comparing Broadway shows. Uzzi and Spiro discovered that the best performing shows were put together by a mix of actors—some who had previously worked together and some who had not. If a set of actors worked together for too long, they lost the spark which was necessary to make the show a hit, whereas actors were not able to click in the right way, had they never worked together. By having a mix of actors who knew each other well with new additions to the cast, the show had the energy and confidence to be a success. These results are too schematic to provide a recipe to cast the next Broadway hit. However, the study suggests a relation between a show’s success and cast diversity in terms of experience in collaborating; neither too much nor too little, but just enough is the sweet spot. Two of the key factors that shape whether people connect are their job roles (e.g. working in the same Broadway play) but also spatial layouts of firm buildings: People who are in close physical proximity are much more likely to speak to each other (see Busch and Grimes, Chap. 5 in this volume). In the 1970’s, Thomas Allen discovered the Allen Curve which demonstrates a negative correlation between physical distance and communication frequency in firms. Several authors have discussed how building architecture and interaction impacts organisational performance (Pentland 2012; 2014; Sailer 2015; Waber et al. 2014). In a New York Times article entitled “Engineering Serendipity”, Greg Lindsay (2013) recounts numerous instances of value that arose from accidental encounters, shaped by spatial configurations within organisations. Examples include the discovery of asbestos at a French pharmaceutical company, whose firm building forced the relocation of teams. Another example was an MIT building, which was constructed in ad-hoc phases, resulting in odd hallways and unlikely adjacencies

210

M. Soto

of university departments. In another piece for the Aspen Ideas Festival, Lindsay alluded to Steve Jobs’ proposed office design to limit bathrooms at Apple when he asked: “But how can we do a better job of bringing people together than installing bigger cafeteria tables, adding another coffee machine, or locking all the bathrooms but one?” (Lindsay 2014). Much of the aforementioned examples assume that the network clusters and gaps are just due to interaction and/or a lack of interaction. Formalist network scholars see networks as bridges and information as cars that go through them. But the field of social networks is just one way of examining the relationships between individuals. In social network analysis it has become common to reduce relationships into a set of lines between nodes. This is powerful, but taken to its extreme limits, this perspective suggests that the only thing that matters is the information that passes through nodes. As Alex Pentland puts it: “[S]ocial physics is a quantitative social science that describes reliable, mathematical connections between information and idea flow on the one hand and people’s behavior on the other” (Pentland 2014, p. 5). However, Pachuki and Breiger (2010) have highlighted this historical tension and potential synergy between social networks (that focus on patterns of interactions, e.g., network maps) and cultural approaches (that focus on the meanings ascribed to and that shape interactions). Emily Erickson calls these two camps formalists and relationists, describing them as follows: “for relationists, meaning is inseparable from the study of social networks because relationships are created out of meaning. For formalists, in contrast, meaning is something that flows through relationships and should be separated out from structural analysis whenever possible” (Erikson 2013, p. 227). This distinction matters for two reasons. First a pattern of interactions on its own does not convey how individuals think about each other, whether they like each other or not, or whether they share the same ideological convictions or not. The meanings ascribed to the relationship are context dependent, and thus the relevancy of the tie depends upon the matter at hand. Where a formalist approach would focus on network maps, the presence, absence, and density of ties, a relationist approach emphasises how that categorisation is inseparable from the specific meanings that individuals ascribe to those relationships. Human interaction is not so easily quantified and involves more than the connections between nodes: How people connect matters. Serendipity is not just about network maps but also how we behave and interact with each other. The way individuals interact shapes what they talk about and how they interpret each other. These in turn shape whether these individuals interact again and how. I thus draw from relational sociology (Emirbayer 1997; Erikson 2013; Mische 2011; Pachucki and Breiger 2010), which integrates cultural sociology approaches with the traditional views of social network analysis that prioritised structure. Relational Sociology understands networks as co-constituted by both agency and structure. Rather than seeing networks as a structure in which individuals are embedded, this approach emphasises how a network both shapes and is shaped by the meanings individuals ascribe to their relationships.

11 Serendipity Across Contexts: From Offices to Post-conflict Settings

211

An illustrative example can be found in the business world concept of “structured unstructured time” (Haas and Mortensen 2016; Larson 2020), which advises managers to allot time in meetings for small talk and other unstructured interaction. So much about work is focused on projects, it is goal oriented action. However, it is often the interruption, walking over to the watercooler or the coffee breaks that cause serendipitous breakthroughs. As Larson describes the power of ‘structuring unstructured time’ she writes: “I constantly hear… about things that came up in those sort of watercooler moments” (Larson 2020). Similarly, Kevin Kniffin found that eating together improved firefighters’ performance (Kniffin et al. 2015), likely since this is a different setting of interaction allowing individuals to connect in new ways. It is not just about connecting, the frequency or patterns of interactions, but also how we connect. This has become particularly significant during the recent pandemic, which forced many to work remotely. The shift to remote work has made organisations more aware of what they gain through live interaction (Baym et al. 2021). While traditionally work environments are project focused, being in person inevitably entails a range of activities and interactions unrelated to work. Taylorist management practices see these as inefficiencies to snub out, but a hyper goal-oriented approach reduces the likelihood of serendipitous discovery (see de Cunha and Berti, Chap. 4 in this volume). Christian Busch highlights a similar point when he indicates that defining a problem too clearly restricts serendipity (Busch 2020). Similarly, a basic innovation practice is to redefine the problem, all of which is less likely to occur if one is resolutely focused on solving the problem and taking its core parameters as assumed. Additionally, information is not transmitted between nodes without interpretation. Our perceptions are shaped by our understanding of the past and present. Scholarship on collective memory—a core field of my doctoral research – is a subfield of sociology which has also focused on social relationships. Beginning with Maurice Halbwachs’s work (1992 [1952]), the study of collective memory has focused on the role of interactions and groups in shaping how we remember the past, and more broadly on how we understand the social world. Collective memory is not just about “information”, but also about how we see the world and interpret it. Within a single shared setting, like an office or a laboratory, individuals have a multitude of shared, yet tacit assumptions. Serendipity is a buzz word in corporate America and the tech scene, but not often related to the context of post-conflict settings, where people have been killed and tortured for decades. What lessons could possibly be drawn for that context from experiences with a coffee initiative among coworkers? My confidence in employing the concept of serendipity in post-conflict settings grew substantially when I discovered that a leading thinker in peacebuilding studies, John Paul Lederach, had dedicated a chapter of his book (Lederach 2005) to serendipity. He writes: Serendipity, it seems, is the wisdom of recognizing and then moving with the energetic flow of the unexpected. It has a crablike quality, an ability to accumulate understanding and create progress by moving sideways rather than in a direct linear fashion. Serendipity requires

212

M. Soto

peripheral vision, not just forward-looking eyesight. It is the single greatest antidote to static politics and tunnel vision (Lederach 2005, p. 115).

Lederach’s work has been critical in my recognition of the possibility of introducing the concept of serendipity fruitfully into research on post conflict settings. In this spirit, I now turn to my own crablike trajectory, moving sideways rather than in direct linear fashion. I narrate my path to ‘institutionalising serendipity’ through Randomised Coffee Trials in organisations around the globe and then turn to my doctoral research on the reintegration of FARC ex-combatants in Colombia. In Search of Serendipity Pedro Medina encouraged me to look around and to start conversations with strangers. Just wait until you have a positive sensation in your gut and go for it, he would say. In a Harvard Crimson article he wrote, “When one sees in people experiences, opportunities, beauty, and life, one is open to meeting them, and often willing or eager to break the ice. When one sees threat in people, then all the circuits block” (Medina 2002). He derided norms that discouraged people from talking with strangers and encouraged rethinking why not. The first several times I tried it was painfully awkward, but it got better with time. To my amazement, it repeatedly resulted in unexpected discoveries, serendipity, as if by recipe. Even if it didn’t extend beyond a coincidental ride shared on a public transit bus, the moment was special, and the day transformed just for having experienced the transition from silence to a warm goodbye from a stranger. Pedro not only taught me about serendipity but also about the possibility of planning for the unexpected to occur. Implicitly a key change was taking place in my understanding: rather than seeking a desired outcome, this approach encouraged one to be open to any desired outcome, thereby increasing the possibility of success in finding at least one. This was a key building block toward my work on “institutionalising serendipity”. Over the course of 2010, Pedro continued to mentor me and countless others, seeming to always have an incredible introduction to offer regardless of the topic. As President of the Yo Creo En Colombia foundation, his mission was to connect, inspire, and educate. Countless social change agents crossed paths with Pedro and were aided by introductions that allowed them to quickly find new collaborators that they previously did not even know existed. Pedro was providing a public service by making the city magical, he constructed new social bridges that helped people discover connection and opportunities among strangers. About a year later, I was in London working at Nesta, the UK innovation foundation, where I applied this experience in a new context. I like to joke that Nesta confines its boundaries to just those things that are shiny and new—meaning that its boundaries were fluid and determined ad hoc. The challenge of this breadth was that we constantly had new projects starting, and rotating staff. I started talking about this with Jon Kingsbury, a director at Nesta at the time, and through side conversations up and down the stairs, along the walk between the office and to his underground station, we came up with the idea for Randomised Coffee Trials (Kingsbury 2013). Nesta regularly conducted randomised, controlled trials to assess the impact of various

11 Serendipity Across Contexts: From Offices to Post-conflict Settings

213

policy initiatives, so we thought it would be funny to remove the control and add some coffee for good measure. The idea was simple, that co-workers at Nesta be introduced at random with someone else in the office each week and invited to get coffee together. The idea was a hit from the get-go, and I did my best to send out introductions at the same time every week. But as this was a side project, there were occasionally work-related meetings that got in the way, much to the dismay of my co-workers, some of whom would be waiting every Friday afternoon to see who they were paired with for the coming week. Towards the end of my time at Nesta, Pedro prodded me to write up my experience, if nothing else to help preserve it for myself. I reluctantly followed his guidance, and it was ultimately published as a blog post on Nesta’s website as “Institutionalising Serendipity Via Productive Coffee Breaks” (Soto 2013). In my bubbling enthusiasm, I ended the post with an invitation for readers to contact me to share what they were doing or to learn more about how they could start their own Randomised Coffee Trial (RCT). I didn’t expect at the time that this would become the single most read piece of content on Nesta’s site in its history, nor that I would end up having over 70 video calls with people from across the globe within a couple weeks. I thought this was an amazing finish to my time at Nesta, without realising it was simultaneously very much the beginning of something new. During these calls I walked folks through the process I had done manually and shared the excel template I used. I attribute much of the subsequent spread of the idea to my copy-left approach. Rather than try to maintain ownership or hide the secret sauce, I did my best to fully answer questions and gave advice to people I had never spoken to, and, in many cases, have not spoken to since. But it is undeniable that the initial interest built upon Nesta’s reputation and networks, and similarly, the second order networks, those of the 70 people and organisations with whom I set up meetings. In the moment, one often isn’t aware of one’s context just as a fish may not know it swims in water. One of those initial responses was from an innovation manager at a global pharmaceutical company who was based in my hometown in Princeton, New Jersey. Upon seeing his email, I let him know that while I was setting up video conference calls with other folks, he was coincidentally in my hometown where I would be returning the following week, “How about if we meet in person the following week?” He happily agreed, and we began to meet on a regular basis as I began to manually run their RCT initiative. I did not charge them anything, because I was not sure what to charge and assumed that a successful pilot with a multinational pharmaceutical company would be good for subsequent business. To this day, I don’t believe he knows that I was working two part time jobs as a waiter and a clerical worker to stay afloat. At some point months later, the innovation manager said that he had never had a similar relationship with “a vendor”—in part he surmised because I never asked for money. To repay me for all I had done, he introduced me to someone, Sid Shah, who he thought I could work on this with further and in fact the two of us co-founded Spark Collaboration six months later. There have been countless serendipitous “aha moments” since publishing the blog post at Nesta. Each one brought immense joy and surprise to me, as things just

214

M. Soto

happened to be working out. Over the years, I have frequently heard of initiatives taking place across the globe, led by individuals that I had never met before. They had read the blog post or heard about it from someone who had participated in an initiative. At one point I was sent a photo from an elevator in the UK Department of Education promoting their Randomised Coffee Trials Initiative—I had never met anyone there nor knew they were running an RCT. UNDP Cyprus was inspired by Randomised Coffee Trials and developed a community development initiative with Cypriot handicraft women that helped build ties between Greek Cypriot and Turkish Cypriot Women. Spark Collaboration went on to win an award for our work with the International Red Cross and Red Crescent (Evans 2015) and received excellent press including from Harvard Business Review (Leberecht 2015; Methot et al. 2021), US News and World Report (Snider 2016), and Huffington Post (Gore and Soto 2017). In fact, we found out about the first Harvard Business Review article by accident. Someone suggested we look at the article because it seemed relevant to what we were doing, and only upon reading it did we see that Spark Collaboration was mentioned. And, a UK National Health Services (NHS) porter Yusuf Yousuf has given an inspiring talk at a TEDx about how he and a doctor started a Randomised Coffee Trial at the UK National Health Service (Yousuf 2019). Partly in response to the literature focusing on space and propinquity to promote serendipity, I wrote Serendipity as Strategy (Soto 2016), where I criticised the practice of firms using architectural design to create physical bottlenecks for people to accidentally bump into each other via casual collisions for serendipitous conversations. One of the most extreme examples is that, “Steve Jobs famously designed the Pixar headquarters with central bathrooms so that people from around the company would run into each other” (Silverman 2013). While I think such strategies pursue a laudable goal, I am convinced that it is far easier to set up a Randomised Coffee Trial initiative to facilitate people meeting unknown colleagues through scheduled meetings. This conviction is based on the belief that people want to connect and are looking for ways to do so better. Casual collisions cause bruises and limiting bathrooms puts unnecessary stress on bladders. But beyond the combination of accident and sagacity, social networks shaped the likelihood of experiencing serendipity, and the scale of the results. In the case of RCTs, Nesta is a very well-known organisation in the UK: if I had the same experience and published my blog post elsewhere it is unlikely that it would have received the attention that it did. Similarly, my own professional profile likely served to reinforce people’s willingness to engage and gave me the confidence to go out on limbs with unexpected returns, knowing that if I fell, I would have many opportunities to try something new. For those who believe in meritocracy, one’s professional position is attributed to one’s own merit. But through the process lens emphasised in this chapter, I see professional position as a form of accumulated privilege—attuned to how the settings in which one has been have contributed to one’s success. In the “Moral Implications of Serendipity” chapter of Robert Merton and Elinor Barber (2004) they state that the “Success and failure in a man’s life are to a considerable extent the products of his efforts, though they are also attributable to social and

11 Serendipity Across Contexts: From Offices to Post-conflict Settings

215

psychological forces beyond his control”. They go on to recognise that this question of individual merit is difficult to parse. As I now turn to examine serendipity in rural parts of Colombia, one thing that stands out to me is how the scale of impact of serendipity is shaped by networks and privilege. This is not to say that one needs a strong network or significant privilege to experience serendipity. We can all work towards experiencing more of it in our futures. As Pedro said to me, “Serendipity is a multiplier”; combined with the preceding sentences, the point is that we can all experience serendipity, but the multiplier will vary based on the skills, resources, and networks we are multiplying it with. As Busch says, “[I] have witnessed serendipity even in the most challenging environments. However, there are major constraints such as disability or structural poverty that can make it nearly impossible. It is a reality for many people around the globe to start out with an extremely low base serendipity potential” (2020, p. 229). Needless to say, these “major constraints” are not evenly distributed. So, while we can all experience serendipity, and we can adopt practices to increase the likelihood of experiencing more of it, the inequality in our networks is undeniable. People can improve their situation by maximising their serendipity potential, but so can everyone else. At times there have been incredible claims about the power of certain technologies to level the playing field. The idea of institutionalising serendipity can at times also evoke a similar sense of harnessing magic. My central aim in emphasising this point is to show that how serendipity is shaped by social networks helps reveal the inequities intertwined in the places we inhabit, whether professional settings, neighborhoods, or countries. From Corporate Offices to Peace Processes In 2016, I began a PhD program in Sociology at the University of Minnesota. I had spent three years full time on Spark Collaboration and wanted to return to topics which had a clearer social impact, and a stronger connection with Colombia, the country where my parents were born and with which I strongly identify. I felt there was a lot of value to the insights I had developed through my experiences with Spark. However, I needed a tangible case to explore the translation of these ideas beyond the office. I first began thinking about the reintegration of ex-combatants while watching a short-animated clip (ARN 2013) by the Colombian Agency for Reincorporation and Normalisation. The animated video shows a cartoon ex-combatant jumping from square to square in what appears to be a board game. In each square, the virtual cartoon figure accesses a new attribute, education, healthcare, job training, and in the end, he celebrates his full reintegration. It struck me that I found it inconceivable to speak of reintegration without a central focus on relationships. Disarmament, Demobilisation, and Reintegration (DDR) is the process by which individuals in armed non-state actor groups transition from combatants to civilians. As “luck” had it, my first year in the doctoral program my advisor, Joachim Savelsberg, invited me to collaborate with him on a project in Northern Ireland. The thirty-year period known as the Troubles involved the government military forces

216

M. Soto

and several armed groups that arose among unionists (UVF, UDA, UDF) and nationalists (IRA, INLA, PIRA), which resulted in bombings, murder, and other violence. We examined how community members from opposing sides were building relationships in the broader polarised setting in Belfast nearly twenty years after the signing of a peace agreement in 1998. We examined three initiatives that brought people together from across the nationalist and unionist divide that has shaped the conflict of this country since its inception. These three initiatives took different approaches to fostering relationships in cross-group activities. One initiative engaged individuals in a joint project where cooperation was goal oriented and focused on a shared objective. A second one developed a deep dialogue approach, where facilitators led group discussions over several weeks. A third one focused on arts and crafts, prohibiting discussion of the perceived sources of conflict: religion and politics. I initially classified these on a spectrum from goal oriented activity to undirected activity, hypothesising that the latter was more conducive to serendipitous relationship building. However, as we began to examine the cases more closely, it became clear that a more granular perspective was necessary. While organisations or initiatives frame interactions in ways that may be more conducive to serendipity, this sphere of influence is not absolute and individuals and organisations often (inadvertently) benefit from alternating between two modes: openness to discovery of the unknown and progress on the known. Even the more goal-directed activities were often based on and interwoven with modes of interaction that were more conducive to serendipity. Our interviewees discussed how the development of initiatives had been the result of both goal-oriented actions, and unexpected encounters that resulted in new relationships forming and revealing new opportunities. At times serendipity paved the way for a new project of collaboration, and vice versa. Casual encounters can develop into starting new projects, and goal-oriented work can be interspersed with practices such as structuring unstructured time (Haas and Mortensen 2016; Larson 2020), which are more conducive to serendipity. For my dissertation research, I conducted interviews and observation of the social reintegration of FARC ex-combatants in both urban and rural spaces in Colombia. The Revolutionary Armed Forces of Colombia—People’s Army (FARC-EP) was a Marxist-Leninist guerrilla group founded in 1964, which was enlisted on to the US’s Foreign Terrorist Organisations list in 1997 (U.S. Department of State n.d). The Colombian government signed a peace agreement with the FARC in 2016 and over 13,000 ex-combatants (ARN 2020) began their transition from being combatants to civilians; they formed a political party now known as Comunes, or the Common People’s party. Much of the media reported about formal settlement camps in rural areas known as Territorial Spaces for Training and Reincorporation (ETCRs2 ). However, the majority of ex-combatants had left those settlements and were living in cities across the country (ARN 2020). In the remaining part of this section, I will share three narratives which highlight a relationship building between ex-combatants and civilians akin to the 2

“Espacios Territoriales para Capacitación y Reincorporación”.

11 Serendipity Across Contexts: From Offices to Post-conflict Settings

217

serendipity of Randomised Coffee Trials; (1) the initial development of the formal settlements, (2) a group of medical volunteers from an urban space that offer pro-bono services in remote rural areas, and (3) an ex-combatant in search of help navigating government systems. The Formal Settlements The initial development of formal settlement camps3 began with individuals from the government, military, and FARC all literally living in the same tents. Across the country, the Blue Tent Protocol involved bringing previously warring sides to live under one roof as they figured out the granular logistics of what peace meant locally. Maria,4 a military officer that oversaw the reincorporation process in the region, reminisced with me about the early days of the Blue Tent protocol. During the early period, she shared a tent with another woman who was from the FARC. The relationship was initially tense but developed into a strong friendship. She laughed to herself as she recalled the story. Maria had a workout routine every morning and while she had invited her tent companion to participate, she had always declined. In the evening, Maria watched movies in English to practice her language skills and one day she realised that her tent companion was surreptitiously trying to watch, too. Maria switched the language setting to Spanish and the ice broke soon thereafter as they got into discussions about the personalities and acts of those in the melodrama. Maria’s direct attempts to build the relationship had failed, and it was not until the end of the day watching movies that the relationship began to develop. This unique space permitted these sorts of interactions and Maria lamented how with the end of the Blue Tent Protocol, these interactions were no longer possible as the peace process advanced. She spent most of her time now in the state capital and only went to the formal settlements for brief visits for security inspections. Ex-combatants were being killed in large numbers and it was the military and police’s responsibility to ensure their safety. She wondered; how could she recreate a setting that had the integrative power of the Blue Tent? How could she help her officers build better rapport with the ex-combatants? It was clear that the security assessments would be more effective if there were better relationships between the military officers and the ex-combatants. From my perspective, it stood out that the breaking of the ice between these two former combatants did not take place during the (structured/formal) meetings, which they surely had throughout the day. The process was not initiated through a conscious effort to reconcile and bring them together. In fact, Maria’s own explicit attempt to build a bridge by inviting her tent partner to morning workouts failed, likely because the overt invitation generated a form of cognitive dissonance. It seemed, rather, that their connection had just happened spontaneously, unexpectedly, which might be the reason that Maria had such a fond memory. However, remaining in this awe meant 3

These were originally called “Zonas Veredales de Transición y Normalización” (ZVTNs), later became ETCRs, and are currently in a limbo state having lost their legal status but not yet incorporated into the normal urban and regional planning processes. 4 This is a pseudonym.

218

M. Soto

that Maria was not able to imagine how to recreate this type of context in other settings. I asked if the military officers approached their visits to the settlements as technical operations, which meant that a military officer arrived with a checklist of questions that they inquired about in sequence and left once dutifully completed. She confirmed. They were focused on effectively and efficiently conducting their task. This mode of goal oriented action, left little space for serendipity. I suggested that a simple change in protocol could create space for the unexpected. One could build in “structured unstructured time” (Larson 2020; Haas and Mortensen 2016). Whether before or after the questionnaire, the officer could be instructed to sit at the bakery for a period of time and just chat. This would create a space with serendipitous potential. It couldn’t be forced, nor would it be the same for everyone, but at the very least it would provide the preconditions for such developments. This was shortly before the start of the Coronavirus pandemic, so it was not implemented as far as I know. Medical Volunteers One of the challenges with the formal settlement camps is that they are in very remote areas, sometimes over an hour drive via unpaved dirt roads that flood easily when it rains to the closest small town. As such, medical services are hard to come by and usually in the form of a traveling one-stop ambulance rather than a standard medical office or hospital. Since the peace accord, a group of medical volunteers has begun making trips from the urban center where they reside to the remote areas, including the formal settlement camps. Considering the hours long trips, these typically take the form of one or two workdays with an overnight stay. Time and time again, the medical volunteers that I interviewed talked about the stories that came out over dinner at the end of the day. They worked hard during the day, seeing patient after patient. They knew that many had walked for hours to be seen and it would likely be a long time before they had another opportunity to see a doctor. A recurring theme from my interviews with the doctors who had participated in these pro bono visits was that they saw a larger number of patients than they usually do in their clinic. They were aware of the dire needs of these patients and the obstacles they had overcome to make it to see them. But their motivation and drive for one form of productivity (to treat the largest number of patients) shaped the way they engaged with their patients, there simply was not time for side stories. During the day, questions were focused on physical ailments and medical concerns. Similarly, as most of these doctors also spent the night at the settlement camp, they also emphasised that they began to see their past patients differently. It wasn’t just about acquiring more “information”, but about merging world views by better understanding what led combatants to pick up arms as well as set them aside. A focus on health provision as separate from these relationships may improve the health indicators of the target population, but it risks preserving the social divides which result in polarised sides, who cannot fathom what motivates those on the other side.

11 Serendipity Across Contexts: From Offices to Post-conflict Settings

219

Navigating Government Systems I attended a half day meeting of an ex-combatant association in hopes of getting in touch with prospective interviewees. I had already begun developing relationships with a handful, which authorised my attendance. During one of the session breaks, I was talking with one couple I met, when a man approached them to say hello. The couple introduced me, and we briefly had the opportunity to say hello. I told him that I was a PhD student and would appreciate an opportunity to meet with him later. He agreed and we exchanged numbers. What neither of us realised is that that day would be the last day before the regional government enacted strict quarantine and physical distancing restrictions due to the Covid-19 pandemic, which mandated staying at one’s residence four out of every five days. About three months later, I received a phone call from him. We hadn’t spoken since first meeting. Yet, he called me to ask for help because he didn’t know where else to turn. He assumed that I had powerful contacts in the country’s capital and was wondering if I could help. I laughed at the suggestion. I told him I would be happy to try but wasn’t sure how I could help. “What is it that you need?”, I asked. He told me he had submitted his application for government funds to support his entrepreneurial project months ago but had not heard anything back. Whenever he called his casemanager, they brushed him off. Was there anyone in Bogotá that I could reach out to? It then occurred to me that I knew the agency’s regional director. We had met the year before and had crossed paths a couple times since. A friend had once introduced me to the former national director of the agency, who in turn introduced me to the director of the region, where I conducted my research. We had a delightful conversation and would now greet each other warmly at various events but it never occurred to me how this relationship would have a material impact on someone’s livelihood. I sent the regional director a polite email indicating the plight of the ex-combatant that I had met. I was not sure if anything would come of it, but the ex-combatant called me back a couple hours later. His case manager had called to profusely apologise and promised to address the pending matters quickly. The ex-combatant was thrilled and reiterated his gratitude to me. He knew that I would have connections in the nation’s capital that could move this along. I chuckled again at what he thought my network in the capital would be like: Absurd as it initially seemed, in this case it was sufficient. The experience is likely not serendipity on his end, I can only guess that from his perspective it was a more straightforward turn of events. A recently demobilised ex-combatant has limited ties outside his peers. When encountering bureaucratic hurdles, they can do little besides remaining patient. This man met someone who he thought would have powerful connections and saved the contact for when he would need it. But, in my case, I was unaware of the significance of my own connections until this experience. I took for granted that I had this relationship with the regional director, because I had decided it was not significant for my research. It had a latent value for my research subjects which only became clear through this exchange.

220

M. Soto

Conclusion To summarise, I have established two propositions in the present paper, one promising, the other sobering. First, serendipity can be institutionalised. That is, it can be planned based on a better understanding of how networks incorporate information (both implicitly and explicitly) and how we are socialised to engage with the unknown and quickly categorise things as relevant and useful or not. Second, we ought to beware of the goal oriented mental model. Sometimes we can achieve more by being less focused. These two insights shed light on how to intervene in diverse social contexts ranging from Londoner office spaces to divided communities in post-conflict settings. The sobering conclusion is that all these relationships that have brought about serendipity actually, ultimately, rest upon the edifice built by the myths of meritocracy and individual agency. The interventions described above can have an impact, but only within a delimited scope. There is a risk of thinking that individuals alone have full control over this, a view I have called into question by highlighting the networks and processes needed to increase the likelihood of serendipity. I have not only shown how to plan for more of it, but also why it is insufficient to simply encourage everyone to be more sagacious without recognising the uneven distribution of necessary inputs. In the post-conflict vignettes from Colombia, individuals made unexpected discoveries by a combination of chance and sagacity. Maria could not have foreseen that Netflix would have been key to building a relationship with the FARC ex-combatant, but she was astute enough to notice her interest and to switch the language settings. The pro bono doctors made similar discoveries during down times, and the experience emphasises the way in which different ways of connecting shapes relationship building. In the final vignette, the ex-combatant reached out to me with a clear goal that he considered me capable of helping him with. From my end this was absurd, what could I possibly do in the face of his bureaucratic troubles? This experience helps visualise the disparate networks that he and I are a part of, often unknowingly, developed through our respective activities, and thus how our chance encounter made such a difference months later. Everyone can tap into hidden reservoirs of serendipity. We can be more open to conversations with strangers, and with people from far off fields and places. These behaviors promote the development of weak ties, and an exploring attitude to make the most out of them. However, while this can be practiced by everyone, the scale of effects are quite different across hierarchies, both because of the types of networks one has access to and because of one’s ability and access to approach situations, which is a scarce resource at times (Shafir and Mullainathan 2013). Serendipity scholars should focus more closely on networks and consider the ethical questions that relate to the structural differences between different agents that become apparent when comparing vastly different settings. For post conflict studies scholars, serendipity can be seen as a strategy for the development of relationships by building spaces (physical and temporal) that allow for people to connect in different ways. Some peace building endeavors inspired by Intergroup Contact Theory try to create spaces for reconciliation, but there is often a

11 Serendipity Across Contexts: From Offices to Post-conflict Settings

221

focus on particular events and intentional relationship building. The above examples highlight how institutionalising serendipity is central to transforming imaginaries of the other, not just in intentional spaces but also via structuring unstructured spaces. These interactions have the potential to transform world views. We need to find ways to foster more serendipity from offices to the post conflict settings.

References ARN. 2013. La Reintegración, paso a paso [Reintegration, step by step]. Agencia para la Reincorporación y Normalización (ARN) [Agency for Reincorporation and Normalisation]. Retrieved from https://youtu.be/-ntjjzWZ9-w. Accessed on 4 Nov 2021. ARN. 2020. ARN en cifras corte noviembre 2020 [ARN figures from November 2020]. Agencia para la Reincorporación y Normalización (ARN) [Agency for Reincorporation and Normalisation]. Retrieved from http://www.reincorporacion.gov.co/es/agencia/Documentos%20de%20ARN% 20en%20Cifras/ARN%20en%20Cifras%20corte%20Noviembre%202020.pdf. Accessed on 4 Nov 2021. Baym, N., J. Larson, and R. Martin. 2021. What a year of WFH has done to our relationships at work. Harvard Business Review. Retrieved from https://hbr.org/2021/03/what-a-year-of-wfhhas-done-to-our-relationships-at-work. Accessed on 4 Nov 2021. Burt, R. 2004. Structural holes and good ideas. American Journal of Sociology 110 (2): 349–399. Busch, C. 2020. Serendipity Mindset: the Art and Science of Creating Good Luck. Riverhead Books. Copeland, S. 2019. On serendipity in science: discovery at the intersection of chance and wisdom. Synthese 196: 2385–2406. https://doi.org/10.1007/s11229-017-1544-3. Emirbayer, M. 1997. Manifesto for a relational sociology. American Journal of Sociology 103 (2): 281–317. https://doi.org/10.1086/231209. Erard, M. 2004. THINK TANK; Where to get a good idea: steal it outside your group. The New York Times. Retrieved from https://www.nytimes.com/2004/05/22/arts/think-tank-where-to-geta-good-idea-steal-it-outside-your-group.html. Accessed on 4 Nov 2021. Erikson, E. 2013. Formalist and relationalist theory in social network analysis. Sociological Theory 31 (3): 219–242. Evans, M. 2015. Forging global unity with Randomised Coffee Trials (RCT) at the Red Cross. HRZone. Retrieved from https://www.hrzone.com/engage/employees/forging-globalunity-with-randomised-coffee-trials-rct-at-the-red-cross. Accessed on 4 Nov 2021. Gore, J., and M. Soto. 2017. Go FIKA yourself: a proven way to work better. Huffington Post. Retrieved from https://www.huffpost.com/entry/go-fika-yourself-a-proven_b_9289434. Accessed on 4 Nov 2021. Granovetter, M.S. 1973. The strength of weak ties. American Journal of Sociology 78 (6): 1360– 1380. Grant, A. 2013. Give and Take: A Revolutionary Approach to Success. Viking, Penguin Group. Haas, M., and M. Mortensen. 2016. The secrets of great teamwork. Harvard Business Review. Retrieved from https://hbr.org/2016/06/the-secrets-of-great-teamwork. Accessed on 4 Nov 2021. Johnson, S. 2010. Where Good Ideas Come from: The Natural History of Innovation. Riverhead Books. Kennedy, P. 2016. How to cultivate the art of serendipity. New York Times. Retrieved from https:// www.nytimes.com/2016/01/03/opinion/how-to-cultivate-the-art-of-serendipity.html. Accessed on 4 Nov 2021. Kingsbury, J. 2013. Here’s how Michael and I got RCT going at Nesta. Nesta. Retrieved from https://www.nesta.org.uk/blog/heres-how-michael-and-i-got-rct-going-at-nesta. Accessed on 4 Nov 2021.

222

M. Soto

Kniffin, K.M., B. Wansink, C.M. Devine, and J. Sobal. 2015. Eating together at the firehouse: how workplace commensality relates to the performance of firefighters. Human Performance 28 (4): 281–306. https://doi.org/10.1080/08959285.2015.1021049. Kuhn, T. 1962. The Structure of Scientific Revolutions. University of Chicago Press. Larson, B.Z. 2020. Give your remote team unstructured time for collaboration. Harvard Business Review. Retrieved from https://hbr.org/2020/10/give-your-remote-team-unstructured-time-forcollaboration. Accessed on 4 Nov 2021. Lederach, J.P. 2005. On serendipity: the gift of accidental sagacity. In The Moral Imagination: The Art and Soul of Building Peace. Oxford University Press. Leberecht, T. 2015. In the age of loneliness, connections at work matter. Harvard Business Review. Retrieved from https://hbr.org/2015/09/in-the-age-of-loneliness-connections-at-workmatter. Accessed on 4 Nov 2021. Lindsay, G. 2013. Engineering serendipity. New York Times. Retrieved from https://www.nytimes. com/2013/04/07/opinion/sunday/engineering-serendipity.html. Accessed on 4 Nov 2021. Lindsay, G. 2014. Engineering serendipity. The Aspen Ideas Festival. Retrieved from https://med ium.com/aspen-ideas/engineering-serendipity-941e601a9b65. Accessed on 4 Nov 2021. Maddox, A. 2015. Serendipity: Social Mobility Across Social Networks and Networked Digital Technologies. Retrieved from https://ssrn.com/abstract=2756287 (or) http://dx.doi.org/10.2139/ ssrn.2756287. Accessed on 4 Nov 2021. Medina, P. 2002. Eye contact in Harvard Square. The Harvard Crimson. Retrieved from https:// www.thecrimson.com/article/2002/11/22/eye-contact-in-harvard-square-chicago/. Accessed on 4 Nov 2021. Merton, R., and E. Barber. 2004. The Travels and Adventures of Serendipity: A Study in Sociological Semantics and the Sociology of Science. Princeton University Press. Methot, J.R., S. Allison, P.D. Gabriel, and E. Rosado-Solomon. 2021. Remote workers need small talk, too. Harvard Business Review. Retrieved from https://hbr.org/2021/03/remote-workersneed-small-talk-too. Accessed on 4 Nov 2021. Milgram, S. 1967. The Small-World Problem. Psychology Today 1(1): 61–67. Mische, A. 2011. Relational Sociology, Culture, and Agency. In The Sage Handbook of Social Network Analysis, ed. John Scott and Peter Carrington, 80–97. Washington DC: Sage. Olma, S. 2012. The Serendipity Machine: A Disruptive Business Model for Society 3.0. Lindonk & De Bres. Pachucki, M.A., and R.L. Breiger. 2010. Cultural holes: beyond relationality in social networks and culture. Annual Review of Sociology 36 (1): 205–224. https://doi.org/10.1146/annurev.soc.012 809.102615. Pentland, A. 2012. The new science of building great teams. Harvard Business Review. Retrieved from https://hbr.org/2012/04/the-new-science-of-building-great-teams. Accessed on 4 Nov 2021. Pentland, A. 2014. Social Physics: How Good Ideas Spread-the Lessons from a New Science. Penguin Random House. Sailer, K. 2015. Seeing is not interacting—thoughts on the new learning hub by heatherwick studio. Spaceandorganisation. Retrieved from https://spaceandorganisation.org/2015/03/13/seeing-isnot-interacting-thoughts-on-the-new-learning-hub-by-heatherwick-studio/. Accessed on 4 Nov 2021. Sand, M., and S. Copeland. 2020. Luck as a challenge for the responsible governance of science and technology. Journal of Responsible Innovation 7 (sup2): S1–S11. https://doi.org/10.1080/ 23299460.2020.1848848. Shafir, E., and S. Mullainathan. 2013. Scarcity: Why Having Too Little Means So Much. Henry Holt Company: Times Books. Silverman, R.E. 2013. The science of serendipity in the workplace—to encourage interaction and innovation, companies try smaller spaces, games; trivia helps break awkward silences. Wall Street Journal. Retrieved from https://www.wsj.com/articles/SB1000142412788732379810457 8455081218505870. Accessed on 4 Nov 2021.

11 Serendipity Across Contexts: From Offices to Post-conflict Settings

223

Snider, S. 2016. 3 Reasons to eat lunch with your co-workers. US News and World Report. Retrieved from https://money.usnews.com/money/careers/articles/2016-02-03/3-rea sons-to-eat-lunch-with-your-co-workers. Accessed on 4 Nov 2021. Soto, M. 2013. Institutionalising serendipity via productive coffee breaks. Nesta. Retrieved from https://www.nesta.org.uk/blog/institutionalising-serendipity-via-productive-coffee-breaks/. Accessed on 4 Nov 2021. Soto, M. 2016. Serendipity as strategy. UX Collective. Retrieved from https://uxdesign.cc/serend ipity-as-strategy-6dc62f252e4e. Accessed on 4 Nov 2021. Soto, M. 2020. The networks behind serendipity: how to increase the probabilities of unexpected discoveries. Serendipity Society. Retrieved from https://theserendipitysociety.wordpress.com/ 2020/03/18/serendipity-and-networks-how-to-incorporate-serendipity-into-your-strategy/. Accessed on 4 Nov 2021. U.S. Department of State. (n.d.). Foreign Terrorist Organisations. Retrieved from https://www.state. gov/foreign-terrorist-organisations/. Accessed on 4 Nov 2021. Uzzi, B., and J. Spiro. 2005. Collaboration and creativity: the small world problem. American Journal of Sociology 111 (2): 447–504. Waber, B., J. Magnolfi, and G. Lindsay. 2014. Workspaces that Move. Harvard Business Review. Retrieved from https://hbr.org/2014/10/workspaces-that-move-people. Accessed on 4 Nov 2021. Yousuf, Y. 2019. Finding inspiration in the NHS is easy, if you know where to look. TEDxNHS. Retrieved from https://www.ted.com/talks/yusuf_yousuf_finding_inspiration_in_the_nhs_is_ easy_if_you_know_where_to_look. Accessed on 4 Nov 2021.

Michael Soto is a data scientist with Hennepin County, Minnesota, a PhD Candidate in Sociology, and fellow at the Interdisciplinary Center for Global Change (ICGC) in Minnesota, US. His graduate research is on the role of networks and memory in reintegration in Colombia, with a comparative component on Northern Ireland. He has published in journals such as Demography and Population and Research Policy Review. His startup Spark Collaboration was highlighted in Harvard Business Review (twice), The New York Times, US News and World Report, Huffington Post, and HRZone; it automated the “Randomised Coffee Trials” process that he co-designed at Nesta, UK.

Chapter 12

Epilogue Sanda Erdelez

Abstract The concept of serendipity has captivated people’s attention for ages, from stories of serendipitous scientific discoveries to movies about lost and found loves to a renowned coffee shop in New York City. Serendipity can greatly influence the paths we take and the outcomes of our lives. The authors of the book Serendipity Science have also shown that serendipity is a thriving and expanding field of academic research. In this Epilogue, Sanda Erdelez, renowned serendipity researcher and early proponent of its rigorous study in Information Sciences, reflects on her experience as a researcher from her first work with information-encountering until today. In particular Erdelez cautions us to consider carefully about how serendipity can be used to promote surprise and good feelings without the accompanying value. While cultivating serendipity seems a priori a valuable thing to do, like other human tendencies, our desire to be pleasantly surprised can lead us astray.

The concept of serendipity has captivated people’s attention for ages, from stories of serendipitous scientific discoveries to movies about lost and found loves to a renowned coffee shop in New York City. Serendipity can greatly influence the paths we take and the outcomes of our lives. The authors of this book have also shown that serendipity is a thriving and expanding field of academic research. As a researcher in information science, I have dedicated my academic career to studying how people experience serendipity in their information behavior. In the field of library and information science (LIS), information behavior refers to how people find and interact with information in various contexts and environments. During the early 1990s, while conducting exploratory interviews with small business managers for my dissertation research, I discovered that many of them had stumbled upon crucial information for their businesses unexpectedly. This realization made me aware that the models and frameworks in LIS for researching information behavior mainly focused on goal-oriented searching for information. Mention of serendipitous discovery of information, when not actively searched for, was rare and usually S. Erdelez (B) Simmons University, Boston, MA, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 S. Copeland et al. (eds.), Serendipity Science, https://doi.org/10.1007/978-3-031-33529-7_12

225

226

S. Erdelez

presented as ‘interesting data’. This research gap intrigued me as I often experienced serendipity in my own information behavior. My curiosity led me to pursue research in the area of information encountering, serendipity in information behavior, which has defined my academic career. Over time, my understanding of information encountering has evolved and its scope and definition have been constantly redefined. When I mention my research interests to others, both academic and non-academic, it usually sparks interesting conversations. The questions I most often hear are: • Is serendipity really a subject of serious, scientific research? • Isn’t serendipity just pure luck? • Is the outcome of serendipity always positive? By now, after reading this book, you may find these questions simple to answer. However, my own answers have changed and become more complex over time. I believe that sharing the evolution of my thoughts on serendipity and information encountering serves as an appropriate conclusion to this book. I hope it will encourage you to reflect on the changing nature of this fascinating phenomenon and provide a glimpse into potential future directions for its perception in research and our personal experiences. Is Serendipity Really a Subject of Serious, Scientific Research? This book includes contributions from scholars who carry out both empirical and conceptual research across various disciplines within the social sciences and humanities. Thus, the answer to the question of whether one can study serendipity is a resounding yes! One can also look to the annual Golden Goose Awards, which celebrate serendipitous scientific research that had a significant impact on society. However, when I was asked the same question during my dissertation defense in the summer of 1995, I hesitated and provided a generic answer. I have often wished I could go back in time and answer that question with the knowledge I have now acquired from more than 35 years of research on information encountering and serendipity. So, why do researchers like myself in the field of Library and Information Science (LIS) study serendipity? Firstly, we follow a user-centric approach in studying how people interact with information. This approach aims to understand people’s experiences with information from their own perspective, rather than from the perspective of the information systems they interact with. Secondly, this user-centric approach allows us to see human information behavior in its natural complexity, including the dialectical interactions between external and internal processes, individual and social processes, and problem-driven and opportunistic processes. By focusing on the study of information encountering, we contribute to the understanding of the information dimensions of serendipity in people’s lives and its value for further study in organizational, social, political, and philosophical contexts. Critics often argue that studying information encountering is not serious because serendipitous moments are by definition unpredictable and difficult to study systematically in a controlled experimental setting, which is considered the gold standard of the scientific method. This criticism stems from a bias towards experimental

12 Epilogue

227

research and the natural sciences as the only “pure” sciences. However, the methods used by serendipity researchers in the social sciences have been well established. LIS and other serendipity researchers have developed a combination of methodologies, including surveys, in-depth interviews, diary studies, case studies, textual analyses, etc., that provide a deeper understanding of how people experience information encountering and serendipity. Tools such as web analytics and eye tracking are used to capture serendipity as it occurs during interaction with information systems, and quasi-experimental studies are designed to induce serendipity in a controlled research environment. This book shows that serendipity has attracted researchers and scholars from a wide range of disciplines, who each bring their own research tools and methodologies. This diversity also leads to cross-pollination of ideas and creates a fertile ground for “meta-serendipity” in serendipity research. Isn’t Serendipity Just Pure Luck? An expanded version of this question is as follows: Why study serendipity when some people just seem to experience it more frequently than others, and there may be little that can be done to influence the occurrence of these events? In response to this question, I often reference the famous quotation by Louis Pasteur that “chance favors the prepared mind”. This saying emphasizes the importance of having the skills of observation and the ability to apply knowledge gained through education and experience to interpret what is observed in order to make serendipitous discoveries in scientific research. Additionally, I highlight that Horace Walpole, who first used the term serendipity, included the idea of having sagacity or acuteness of mind as part of the definition. Through my interviews with individuals who frequently experience serendipitous information encounters, I have found that some people are more ‘serendipity-prone’ than others. These individuals not only regularly experience serendipity, but also expect and welcome these experiences as a regular part of their information behavior. Other researchers have also found differences in how people with various personality traits and professions experience serendipity. The common thread in these findings is that positive outcomes with serendipity can encourage people to be more open to it and, therefore, become ‘luckier’ than others. In addition to individual differences, research has shown that the environments where people work, live, and communicate can also increase their chances of experiencing serendipity. For example, in a work environment, the ‘water cooler effect’, where people come together to grab a drink, can create serendipitous opportunities. Libraries, bookstores, professional conferences, and trade shows have also been studied for their ability to contribute to unexpected discoveries and connections. The electronic information environment and its various technologies have been the most researched in terms of their impact on generating serendipity. Some early industry commentators even referred to Google as the ultimate serendipity engine. Many web applications use people’s natural attraction to serendipity by including features like a ‘surprise me’ button that connects users with randomly selected database content. The design of electronic information content, including its layout, presentation, and navigation, plays an important role in facilitating serendipity.

228

S. Erdelez

Although serendipity is often seen as a result of good luck, research, including this book, has shown that it is not entirely a matter of chance. Christian Busch’s (2020) book, The Art and Science of Creating Good Luck and the Serendipity Mindset, emphasizes that designing serendipity involves both the skills and aptitude of the people experiencing it and their information environments. My early research showed that the more one experiences or is aware of experiencing serendipity, the more one anticipates it. This is because one becomes more relaxed in the way they find information, knowing that they will eventually come across what they are looking for. Although Busch and others have explored this topic, it remains challenging to explain how it can be adopted as a strategy or taught as a skill. Is the Outcome of Serendipity Always Positive? Serendipity punctuates the regular, often monotonous events of our daily life, whether at home, work, play, or simply going about our business. Its beauty lies in the mystery and surprise it brings, and people enjoy good surprises. Throughout my research career, I have had the opportunity to interview many people about their experiences with information encountering. Their excitement about these events was palpable, and their recall of the events before and after these serendipitous moments was surprisingly vivid. I specifically explored the emotional state of people before and after information encountering and discovered that for most of my respondents, there was a shift in the type of feelings experienced from negative to positive or neutral to positive. The happiness and excitement were both about the content of the information encountered and the unexpected manner in which it was uncovered. In most cases, the value of the information was immediately evident and had an immediate impact on the person. For example, one study participant found crucial information about a competitor’s new strategic moves, a task she was assigned at work, in a pile of discarded magazines while at a laundromat. However, the most profound feelings of happiness were about serendipity that resulted in positive outcomes over a longer period, such as a doctoral student who stumbled upon a book in the library stacks that changed the direction of his dissertation and ultimately shaped his research career. Just as in the examples from my conversations with everyday people, many scientific discoveries and business innovations attribute serendipity as the trigger for significant positive outcomes. The growing interest in serendipity and its ability to capture people’s attention and curiosity also raises concerns about its potential dark side. An early indication of the possibility of misusing serendipity was revealed in a study on information encountering within a news reading context by Yadamsuren (2010). The in-depth interviews with study participants showed that they often encountered serendipity in the form of bizzare news. The combination of serendipity’s intrinsic appeal with unusual and sensational information content may contribute to the spread of misinformation and disinformation. Recent events in the field of public health with COVID-19 and in extreme political activism can serve as the warning examples of this hypothesis. Another instance of how people’s positive perception of serendipity can be exploited is through instances of ‘fake serendipity’. Unlike genuine serendipity that occurs naturally, fake serendipity is artificially generated by various invisible

12 Epilogue

229

systems, ranging from simple web browser cookies to complex AI, that track our online activities and collect data on our interests. This information is then presented to us in customized online ads that seem serendipitous. Because the content that appears interesting or useful is combined with the additional “excitement” of serendipity, it is more likely to grab our attention. Such experiences fuel consumerism and are particularly common on various social media platforms. Despite the potential for both genuine and fake serendipity to lead to negative consequences, such as consumerism or the spread of misinformation, serendipity can also be a tool for breaking out of echo chambers and correcting misinformation. Echo chambers are often tied to specific information environments, but people encounter information across various environments. Serendipitous encounters may bring us into contact with new insights that differ from those circulated within an echo chamber. Thus, serendipity may provide a narrow, back door channel into an echo chamber where information that doesn’t conform to its echo can still enter. The recognition of how serendipity can be manipulated through faking and its potential to penetrate echo chambers could be a crucial aspect of the information literacy toolkit for combating misinformation and disinformation. Rather than accepting online serendipity at face value as genuine, it is necessary to take steps to prevent the collection of our online data by cleaning our device caches and opting out of data and location sharing. We can increase our awareness of fake serendipity by conducting ‘online search probes’, which aim to shed light on the workings of data collection algorithms. For instance, we could conduct a test search on a topic of interest, such as martial arts or exotic travel destinations, and then observe when and how this behavior is utilized to generate fake serendipity. While the idea of improving the quality control of the serendipity we encounter may seem unusual, I believe it is crucial for developing a heightened awareness of both the positive and negative aspects of information encounters. Conclusion After reading the book and reflecting briefly on the evolution of my views on serendipity and information encountering, I am led to reflect on how this relates to Walpole’s idealized definition of serendipity as an “unsought finding” (p. xi, this volume). If serendipity can be generated and personally experienced without the element of chance (as discussed in the chapters by Reviglio, Arfini, and Ross), is it still considered serendipity? The purpose of this question is not to raise doubts about serendipity, but to draw attention to how it arises and how it can impact our lives through information encountering, for better or for worse. However, we should not focus solely on researching serendipity out of mere curiosity. In the rapidly changing, non-linear online environment, early education and ongoing awareness about the value of serendipity and how it can be manipulated are becoming increasingly important. As such, it is crucial to translate our research into various practical fields, ranging from information literacy to ethical system design and development. Despite the efforts that I and others have made to promote serendipity and information encountering in the field of human information behavior research, much still needs to be done to connect across the many areas of research covered in this

230

S. Erdelez

volume. A comprehensive approach to serendipity is necessary, but we should also acknowledge that no one specific approach to studying serendipity is more important than another. Will these efforts lead to the creation of a new integrated field of serendipity studies? This depends on many contextual factors that are difficult to predict, including collective efforts such as this book, active serendipity-focused research agendas by scholars from various disciplines, and connections through the Serendipity Society. Let’s stack the odds in favor of this emerging field and not leave its existence to chance.

References Busch, C. 2020. The Serendipity Mindset: The Art and Science of Creating Good Luck. Penguin Random House. Yadamsuren, B. 2010. Incidental Exposure to Online News in Everyday Life Information Seeking Context: Mixed Method Study [Doctoral dissertation, University of Missouri]. MOspace. https:/ /hdl.handle.net/10355/10806.

Sanda Erdelez is the Interim Dean and Professor at the Simmons University in Boston, MA. She has been internationally recognized for her pioneering research in the areas of information encountering and opportunistic discovery of information. Her research has received funding from both corporate and government sources including Dell Inc., SBC Communication, Texas State, and NSF. She is an active member of the Association for Information Science & Technology (ASIS&T), where she has served in numerous leadership roles including the chair of Special Interest Group on Information Seeking and Use (SIG USE) and member of the ASIS&T Board of Directors.

Index

A Accident(al), 2–4, 11, 18, 19, 21–23, 35, 39, 45, 49, 52, 54, 56, 101–120, 128, 129, 131, 132, 134, 147, 149, 154, 168–183, 188, 192, 193, 198, 200, 206, 209, 214 Affordance, 38, 51, 59, 118–120, 146, 152, 157, 158, 161, 170, 178 Agency, 12, 45, 51, 52, 54, 58, 107, 111, 167–170, 173, 175, 177, 179–181, 210, 215, 219, 220 Aha, 2, 79, 111, 120, 137, 213 Algorithm, 35, 36, 39, 43, 50, 146, 147, 149, 151, 158–162, 229 Anomaly, 50, 54–57, 61, 105, 106, 118, 152 Archeolog(y)(ical), 25 AstraZeneca (Covid vaccine trial), 50 Attentive meandering, 46 Automation, 50

B Bavarsaikhan (Dr.), 24–28 Bisociation, 75, 78–80, 82, 85, 86, 111 Blog(ging), 3, 37, 39, 213, 214 Bolor toli (Crystal dictionary), 21 Bricolage, 51, 58, 59, 79 British Library, 35 Browsing, 32, 34, 35, 37, 38, 41–44, 79, 146, 149

C Capability, 34, 51, 53, 85, 86, 89, 111, 150, 153, 154, 160, 162, 171

Capacity, 9, 49, 50, 55–59, 74, 75, 77, 86, 87, 107, 117, 119, 126, 154, 156, 161, 195 Cause, 2, 5, 6, 51, 81, 108, 110, 120, 138, 146, 168, 169, 177, 178, 190, 211, 214 Chance, 3, 4, 7, 9–12, 20–22, 24, 36, 43, 45, 52, 56, 58, 69, 83, 85, 101–108, 110–118, 120, 126–129, 131, 134, 139, 147, 151, 152, 160, 171, 172, 176, 177, 181, 188, 192, 198, 220, 227–230 Classification, 34, 36, 39, 41 Cognition, Cognitive (-offloading) (-trajectory) (-probatonics), 10–13, 37, 73, 74, 79, 80, 110, 112, 113, 116–119, 125–127, 135–137, 139, 140, 162, 167–170, 172–183, 186, 207, 217 Collections, 3, 13, 15, 31–34, 36, 38, 40, 42, 43, 45, 103, 145, 229 Collective, 8, 13, 17, 28, 29, 58, 59, 62, 81, 84, 115, 187, 188, 192–194, 200, 201, 206, 211, 230 Colombia, 205, 206, 212, 215, 216, 220, 223 Conditions, 12, 28, 51, 53, 55, 57, 69, 72, 85, 86, 102, 105, 116, 126, 128–132, 134, 136, 137, 147, 152, 157, 167, 168, 174, 179, 180, 188, 189, 192 Constraints, 37, 50, 107, 160, 215 Contingency, 7, 13, 37, 86, 87, 102, 104, 107, 109, 115, 117, 120, 172, 177, 181 Convergence, 154, 161

© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 S. Copeland et al. (eds.), Serendipity Science, https://doi.org/10.1007/978-3-031-33529-7

231

232 Creativity, 74, 78, 79, 84, 110–112, 128, 147, 150, 153, 177, 182, 183, 186, 208 Cultivate, 9, 54, 57, 86, 119, 148, 149, 151, 154, 157, 160–162 Cybernetic Serendipity, 3, 149 Cyrillic, 20, 21 D Deer stones, 24, 25, 27 Design, 2, 5, 9, 12, 26, 31, 33, 35, 37, 38, 41, 51, 53, 58, 60, 77, 81, 86, 88, 114, 128, 145–154, 157–162, 180, 197, 198, 210, 214, 227, 229 Deviation, 50, 51, 54, 55, 61 Dewey Decimal Classification (DDC), 33, 34, 41, 44 Dictionary, 4, 17–22, 24, 119, 127 Discovery (also, Scientific Discovery), 2, 4–8, 10, 11, 13, 17–23, 26, 28, 29, 35–37, 39, 46, 50–52, 55, 56, 58–61, 70, 75, 101–120, 125–131, 134–137, 139, 140, 147, 148, 150, 155, 167, 170–179, 181, 187, 188, 192–195, 200, 201, 208, 209, 211, 216, 225 Disruption, disruptive, 3, 103, 105, 170, 173–178 Divergent, 78, 154, 161 Diversity, 15, 38, 74, 80, 81, 84, 88, 146, 150, 152, 154–158, 161, 197, 209, 227 Doubt (Generative), 9, 49, 56, 57, 76 E Echo chamber, 145, 146, 155, 156, 160, 161, 229 Effect, 3, 6, 32, 33, 35, 36, 39, 43, 71, 79, 85, 87, 106, 110, 116, 118, 137, 139, 153, 157, 168, 174, 177, 180, 188–192, 198, 200, 208, 220 Effort(ful), 5, 9, 32, 44, 49, 51–56, 60, 71, 76, 86, 107, 111, 126, 130, 131, 138–140, 188, 189, 192–194, 200, 201, 214, 217, 229, 230 Embodied (cognition), Embodiment, 34, 44, 120, 126, 128, 135–137, 139, 140, 179, 207 Emergence, 11, 14, 31, 41, 43, 54–56, 58, 59, 62, 72, 76, 84, 88, 109, 112, 116, 117, 126, 149, 153 Emotions, 75, 130 Entrepreneurship, 10, 69–72, 79, 86–88

Index Environment, 6, 7, 9, 12, 28, 31–33, 35, 37–40, 44, 46, 49, 50, 53–55, 57, 60, 73, 75–77, 81, 84, 85, 88, 111–113, 115, 116, 118–120, 126, 128, 136–138, 140, 145–147, 150–153, 159–162, 170, 174, 175, 177–180, 194, 211, 215, 225, 227–229 Error, 3, 51, 58, 59, 70, 73, 83, 103, 106, 110, 113, 117 Ethics, 2, 12, 150, 160, 161, 187, 201, 204 Experiment, 7, 50, 59, 61, 73, 80, 82, 88, 105, 118, 126, 128, 129, 132, 133, 135–137, 139, 175, 188–190, 195–200 Explorability, 38, 152 F Fairy Tale, 18, 23, 24, 192, 193 Filter bubble, 145, 146, 155, 156, 160, 161 Filtering, 35, 71, 84, 155 Findability, 152 Fleming, Alexander (Penicillin, Discovery of), 23, 80, 114, 129, 134, 193 Frustration, 173, 174 G Google, 23, 35, 36, 39, 41, 77, 114, 145, 155, 227 H Hash tags, 39, 146 Human Computer Interaction (HCI), 12, 145, 147, 148, 153 Humanities, 5, 32, 33, 40, 42, 44, 45, 226 Humanities Networked Initiative (HuNI), 32, 45 Huns, 26–29 I Ignorance (also Aching Ignorance), 11, 52, 61, 103, 125–127, 129–135, 137, 140, 174 Improvisation, 50, 51, 58, 61, 76, 173 Incidental (exposure), 22, 147, 148, 153 Indigenous, 34 Information retrieval, 34, 35, 37, 41, 45, 46, 145, 150 Innovation, 2, 6, 10, 13, 14, 17, 33, 39, 55, 58, 59, 69–74, 77–79, 81, 83, 86, 88, 104, 113, 114, 117, 119, 120, 147, 148, 153, 176, 187, 189, 191, 197,

Index 199, 200, 204, 206, 209, 211–213, 228 Interaction, Interactive, 5, 7, 10–12, 53, 70, 74, 75, 77, 80, 81, 84, 85, 101, 104, 109, 111, 114–116, 118, 126, 134, 138, 145, 146, 148, 153, 168, 170, 172, 173, 179, 180, 182, 205, 207–211, 216, 217, 221, 226, 227 Interdisciplinary, 2, 37, 133, 134, 138, 139, 145, 150, 154, 160, 162, 223 Interface, 38, 41, 42, 44, 82, 152, 160 Internet public library, 40 Intuitive, Intuition, 23, 80, 104, 109, 113, 125, 129, 131, 152

J Jenner, Edward, 13, 187, 188, 190

233 Mechanistic, 49–53, 55 Merton, Robert, 4, 103, 214 Micro, 12, 15, 128, 171, 172, 182 Mongolia, Mongols, 6, 8, 17, 20, 21, 24–29 Museums, 25, 31, 36, 42, 43, 45

N Negative capability, 9, 49–51, 56, 60, 62 Network (social), 77, 152, 156, 206–208, 210, 214, 215

O Operationalize, 87, 146, 152 Organism(ic), 49, 51–53 Outsight, 180

L Leadership, 2, 57, 58, 230 Library, 2, 9, 21, 31–38, 40, 41, 44, 45, 82, 149, 161, 225–228 Loanword, 18 Long tail content, 152 Luck (also, Epistemic Luck, Moral Luck), 2, 4, 6, 7, 9, 11, 13, 19–22, 24, 49, 51, 52, 54, 60, 61, 72, 88, 102, 104, 105, 110, 113, 116, 117, 131, 134, 150, 172, 187–189, 193, 195, 198–201, 204, 206, 215, 227, 228

P Paradox, 11, 51, 55, 60, 61, 125, 126, 129, 131, 135, 136, 139, 198 Peace, 206, 215–218, 220 Performativity, 60, 61 Personalized, Personalization, 145–147, 149–152, 154, 155, 158–162 Pluralist, 44, 181 Post-conflict studies, 206 Power, 7, 35, 49, 55, 60, 61, 83, 88, 158, 205, 207, 211, 215, 217 Prepared Mind (/Pasteur), The, 102, 110–113, 115, 132, 151, 171, 172, 192, 227 Process, 6, 8–10, 23, 32, 35–38, 41, 44, 49–62, 69–76, 78, 79, 81–88, 101–107, 109, 111–116, 118, 120, 126, 128, 129, 131, 132, 134–137, 139, 140, 147, 148, 154, 158, 160, 162, 167–169, 175, 177–179, 181–183, 186–189, 193, 194, 201, 205, 206, 208, 213–217, 220, 226 Programming, 148, 151, 161 Pseudo (Serendipity) (Personalization), 12, 71, 145, 154, 155, 157, 158, 160–162

M Macro, 15, 178, 183 Management and Organizational Studies (MOS), 9, 49, 52, 57, 60, 62 Manipulation, Manipulative, 12, 118, 145, 146, 148, 157–161, 168, 179, 181

R Randomized Coffee Trials, 14, 78, 81, 205, 206, 212–214, 217, 223 Randomness, 32, 33, 36, 39, 50, 149–151 Recommender systems, 9, 12, 145, 147, 150, 161

K Kairos, 159 Keyword search, 35, 36 Khorig Uul, 25, 26, 28 Knowledge, 2, 7, 11, 17, 19, 25, 26, 31–35, 37, 38, 40–46, 53, 56–61, 80, 84, 101–104, 107–112, 115–117, 120, 125–127, 129, 130, 132, 133, 135, 139, 140, 145–147, 150, 151, 153, 155, 156, 170, 173, 175, 178, 190, 193, 194, 200, 226, 227 Kuhn, Thomas, 208

234 Relational, Relations, Relationships, 6, 7, 32, 33, 37, 45, 46, 55, 57, 71, 80, 102, 104–106, 110, 113, 115, 117–120, 126, 130, 132–134, 138, 139, 149, 158, 161, 168, 172, 175, 177, 178, 182, 187, 193, 194, 200, 204, 206–211, 213, 215–220 Relational sociology, 210 Relevance, 32, 35, 40, 74, 80, 85, 87, 149, 150, 152, 156 Resource discovery, 35 Responsibility, 13, 83, 115, 147, 187, 195, 198, 200, 204, 217 Retrospective, 6, 50, 52, 106, 168, 183

S Sagacity, 2, 4, 6, 9, 11, 18, 19, 24, 31, 45, 70, 79, 85, 103, 104, 107, 109–118, 120, 127, 153, 170–173, 178, 192, 193, 199–201, 206, 207, 214, 220, 227 Search Engines, 31, 32, 35, 40, 41, 146, 147, 149 Serdevtiim, 23 Serendipity Pattern (/Robert Merton), 103, 109, 119 Serenget, 23 Social media, 5, 39, 128, 146, 149, 150, 154, 158–160, 229 Social networks, 77, 152, 156, 206–210, 214, 215 Strategy, 3, 9, 10, 35, 49, 50, 60–62, 69–72, 76, 80, 83, 86–88, 103, 110, 114, 120, 133, 150, 182, 197, 214, 220, 228 Structuring unstructured spaces, 221 Super-encounterer, 37, 111 Surfing (the internet), 149 Surprise, 14, 50, 54, 55, 57, 62, 106, 107, 109–111, 117–119, 130, 133–137, 147, 152, 161, 173, 175, 178, 188, 192, 209, 213, 225, 227, 228

Index T Three Princes of Serendip, The, 9, 18, 31, 45, 192, 206, 207 Trace, 2–4, 6, 14, 20, 108, 172, 180, 181 Transliteration, 17, 18 Twitter, 39, 40

U Uncertainty, 9, 49–51, 53, 70, 76, 77, 87, 114, 151–153, 158, 174, 179, 189, 196 Unexpected, 14, 20, 22, 24, 31–37, 42, 45, 54, 55, 57–59, 62, 71–80, 82–86, 88, 103, 106, 107, 109–111, 114, 118–120, 127, 128, 130, 132–134, 146, 150–152, 155, 170, 173, 195, 205, 206, 208, 209, 211, 212, 214, 216, 218, 220, 227, 228

V Value, 9, 14, 31, 37, 46, 53, 58, 77, 82–84, 88, 101, 107, 109, 110, 112–116, 119, 120, 146, 150, 151, 153, 157, 160, 161, 177, 187, 192, 198, 199, 201, 209, 215, 219, 225, 226, 228, 229

W Walpole, 2, 4, 9, 18–20, 23, 24, 31, 42, 45, 103, 108, 112, 173, 174, 192, 193, 206, 207, 227, 229 Water cooler effect, 227 Weak connections, 71, 152 Web, The, 35, 36, 39–41, 149, 160, 227 Wisdom, 4, 11, 55, 73, 101, 102, 104, 110, 112, 115, 118, 120, 127, 160, 188, 192, 211

Z Zemblanity, 51, 60, 88