Advancing Information Systems Theories, Volume II: Products and Digitalisation (Technology, Work and Globalization) [1st ed. 2023] 303138718X, 9783031387180

Citation preview

Edited by Leslie P. Willcocks Nik R. Hassan Suzanne Rivard

Advancing Information Systems Theories, Volume II Products and Digitalisation

Technology, Work and Globalization

Series Editors Leslie P. Willcocks Department of Management London School of Economics and Political Science London, UK Mary C. Lacity Sam M. Walton College of Business University of Arkansas Fayetteville, AR, USA

The Technology, Work and Globalization series was developed to provide policy makers, workers, managers, academics and students with a deeper understanding of the complex interlinks and influences between technological developments, including information and communication technologies, work organizations and patterns of globalization. The mission of the series is to disseminate rich knowledge based on deep research about relevant issues surrounding the globalization of work that is spawned by technology.

Leslie P. Willcocks Nik R. Hassan  •  Suzanne Rivard Editors

Advancing Information Systems Theories, Volume II Products and Digitalisation

Editors Leslie P. Willcocks Department of Management London School of Economics and Political Science London, UK

Nik R. Hassan Labovitz School of Business and Economics University of Minnesota Duluth Duluth, MN, USA

Suzanne Rivard HEC Montréal Montreal, QC, Canada

ISSN 2730-6623     ISSN 2730-6631 (electronic) Technology, Work and Globalization ISBN 978-3-031-38718-0    ISBN 978-3-031-38719-7 (eBook) https://doi.org/10.1007/978-3-031-38719-7 © 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 Palgrave Macmillan imprint is published by the registered company Springer Nature Switzerland AG. The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland Paper in this product is recyclable.

Acknowledgements

Many of the chapters in this volume found their gestation in the 9th Concurrent SIGPHIL@ICIS (Virtual Only) Workshop on Products of Theorizing: Towards Native Theories of Emerging Information Technology, Dec 12–13, Austin, Texas. This was convened by Nik Hassan, Suzanne Rivard, Ulrike Schultze, and Leslie Willcocks. We thank all participants in this and subsequent events that led to the first volume in this series—Hassan, N. and Wilcocks, L. (2021) Advancing Information Systems Theories: Rationale and Process (Palgrave Macmillan, London), and also a Special Issue on Products of Theorizing in the Journal of Information Technology, as well as Chaps. 4, 8, 9, 10, 11, and 12 of this volume. We would also like to acknowledge permissions granted to develop versions from or reproduce parts of the following publications: Chapter 2 originally appeared as Hassan, N., Lowry, P. and Mathiasson, L. (2021) “Useful Products in Information Systems Theorizing: A Discursive Formation Perspective,” Journal of the Association for Information Systems, 23, 2, 418–446. The case material and Tables in Chap. 3 are based on Lacity, M. Willcocks, L. & Gozman, D. (2021). “Influencing information systems practice: The action principles approach applied to robotic process and cognitive automation,” Journal of Information Technology, 36, 3, 216–240. v

vi Acknowledgements

Chapter 6 is a more developed and updated version of Mingers, J., and Willcocks, L. 2014. “An Integrative Semiotic Framework for Information Systems: The Social, Personal and Material Worlds,” Information and Organization (24:1), pp. 48–70. Chapter 7 is largely reprinted from Conboy, K., Morgan, L. and Gleasure, R. (2022) “When crowds play god: A Promethean perspective on crowdfunding,” European Journal of Information Systems, 31, 2, 207–226. Chapter 13 is reprinted from Mertens, W. and Recker, J. (2020) “New Guidelines for Null Hypothesis Significance Testing in Hypothetico-­ Deductive IS Research,” Journal of the Association for Information Systems, 21(4), .DOI: 10.17705/1jais.00629

Contents

1 Introduction:  On Types of Products of Theorizing  1 Suzanne Rivard, Leslie Willcocks, and Nik Rushdi Hassan 2 Useful  Products in Information Systems Theorizing: A Discursive Formation Perspective 17 Nik Rushdi Hassan, Paul Benjamin Lowry, and Lars Mathiassen 3 Debating  Genres and IS Research: The Case of Action Principles for Service Automation 79 Leslie Willcocks, Mary C. Lacity, and Daniel Gozman 4 A  Paradigm Shift in Understanding Digital Objects in IS: A Semiotic Perspective on Artificial Intelligence Technologies119 Julia Kotlarsky and Ilan Oshri 5 A  Cybernetic Theory of the Impact of Implementers’ Actions on User Resistance to Information Technology Implementation149 Suzanne Rivard and Liette Lapointe vii

viii Contents

6 Interrogating  Sociomateriality: An Integrative Semiotics Framework for Information Systems169 John Mingers and Leslie Willcocks 7 When  Crowds Play God: A Promethean Perspective on Crowdfunding211 Kieran Conboy, Rob Gleasure, and Lorraine Morgan 8 Routinization  of Digital Transformation of Work: A Discursive Practice Orientation Toward a Native IS Theory259 Daniel N. Treku, Joseph Manga, and Emmanuel W. Ayaburi 9 Patterns  for Visualizing the Aesthetic Qualities of Business Processes285 Monika Blattmeier 10 Information  Theory in IS309 Earl McKinney and Sebastian K. Boell 11 The  Primacy of Concepts and Implications for the IS Field339 Nik Rushdi Hassan 12 Propositions  for a Future Information Exchange Theory to Support Decision Making367 Ali Mohammed Bazarah and Yan Li 13 New  Guidelines for Null Hypothesis Significance Testing in Hypothetico-­Deductive IS Research385 Willem Mertens and Jan Recker I ndex439

Notes on Contributors

Emmanuel W. Ayaburi  is Assistant Professor of Information Systems in the College of Business Administration at Cleveland State University, Ohio. He received his Ph.D. in Information Systems from the University of Texas at San Antonio and his Master’s in Business Administration from Colorado Heights University. His research interests include behavioral information systems security and privacy, economics of information systems, and knowledge sharing. His work has been published in Information Systems Frontiers (ISF), International Journal of Information Management (IJIM), Computers in Human behavior, International Journal of Innovation and Technology Management (IJITM), Information Technology and People (ITP), and he has also published a number of IS conference proceedings. Ali Mohammed Bazarah  is an assistant professor at Stonehill College. He has gained a Ph.D. and MSc. in Information Systems from Claremont Graduate University (CGU). Before pursuing graduate studies at CGU, Dr. Ali received an MBA from King Fahd University, Saudi Arabia. His research focuses on various topics such as decision support, information exchange, data analytics, and location intelligence. More specifically, Dr. Ali’s research examines the theory and practice of information exchange with an emphasis on decision support in a multi-stakeholders’ environment. Dr. Ali has published in Association for Information Systems (AIS) ix

x 

Notes on Contributors

conferences, and his work was nominated for the top 25% best paper award at The Americas Conference on Information Systems (AMCIS) in 2022. Monika  Blattmeier teaches and researches as Professor of Business Processes in Mechanical Engineering at the University of Applied Sciences Emden/Leer in the Faculty of Engineering. Before her nomination as professor, she gathered various experiences in automotive design, purchasing, technology development, product engineering and project management at the BMW Group in Munich. Most recently, she worked as a process manager for the development of the powertrain in the 6 Series GT. Parallel to her work in industry, she was teaching quality management and technology management as a lecturer at various universities, for example at the University of Regensburg, at the HTW Dresden and at the OTH Regensburg. She received her PhD from the University of Duisburg-­ Essen and was specifically supervised by the Institute for Product Engineering. Monika wants to provide a different and broader view on business processes and uses not only the theories of engineering but also those of art and philosophy. Finally, her aim is to aesthetically design not only products and services for the customer but also organizational processes. Sebastian  K.  Boell  is a lecturer at the University of Sydney Business School in 2014. He investigates how organisations are able to create understanding about themselves and their environment using information and communication technologies (ICT). His research draws from his multi-disciplinary training in Information Systems, Information Science, Cognitive Psychology, Engineering and further expertise in different research methodologies including such as histography, qualitative research, interpretive work, scientometric and bibliometric approaches, practical information retrieval and literature analysis. He holds a PhD in Information Systems from the University of New South Wales and an MA in Information Science from Saarland University, Germany. His research is published in top-tier journals across the fields of Information Systems, Information Science, and Work and Organizational Studies.

  Notes on Contributors 

xi

Kieran Conboy  is a professor in the School of Business and Economics at the University of Galway and is a co-Principal Investigator in the Lero Irish Software research centre. He previously worked for Accenture Consulting and the University of New South Wales in Australia. He is also on the board of the Irish Research Council and has previously been Head of the School of Business as well as Dean of Business, Public Policy and Law. Kieran has published over 150 articles in leading international journals and conferences including Information Systems Research, the European Journal of Information Systems, Information Systems Journal, the Journal of the AIS, IEEE Software, the International Conference in Information Systems and the European Conference in Information Systems. He is an editor of the European Journal of Information Systems and has chaired international conferences in his field. Rob Gleasure  is an associate professor and Vice Head of the Department of Digitalization, Copenhagen Business School. He received his Ph.D. from University College Cork in 2013, and his research interests focus on the relationship between emotion, social structures, and online collaboration. His research has appeared in Information Systems Research, Journal for the Association of Information Systems, Journal of Management Information Systems, European Journal of Information Systems, MIT Sloan Management Review, and California Management Review, among other journals. Daniel Gozman  is the Director of Engaged Research and a senior lecturer at the University of Sydney Business School (AUS) and an Honorary Fellow at Henley Business School at the University of Reading (UK). Danny received his PhD from the London School of Economics (LSE). He is a member of the LSE’s Outsourcing Unit and a research fellow at University College London (UCL’s) Centre for Blockchain Technologies. Danny is a Freeman at the Worshipful Company of Information Technologists (Livery Company of the City of London). Currently, his work focuses on the intersection between policy, emergent technology (AI, Blockchain/DLT, Cloud, etc.) and innovation. He has acted as an academic adviser to international law firms and analyst groups and as an external examiner to higher education institutions. He is a senior editor for the Journal of Information Technology and an associate editor for the

xii 

Notes on Contributors

Journal of Electronic Commerce Research and Applications. Prior to academia, Danny worked for Accenture and Deloitte. Nik Rushdi Hassan  is Associate Professor of Information Systems at the Labovitz School of Business and Economics, University of Minnesota Duluth. He is currently a senior editor for Journal of Information Technology and ACM Data Base for Advances in Information Systems and an associate editor for the History and Philosophy Department of the Communications of the Association for Information Systems. He has served as the president of the AIS Special Interest Group on Philosophy in Information Systems and was one of the guest editors of the European Journal of Information Systems Special Issue on Philosophy and the Future of the IS Field. His research areas include the philosophical foundations of the IS field, theorizing and theory building, IS development, business analytics, social network analysis, and complexity science. Julia Kotlarsky  is Professor of Information Systems at the University of Auckland Business School (New Zealand). Julia’s research interests revolve around technology sourcing and innovation, digital transformation, digital sustainability and interface between Artificial Intelligence and humans. Her work was published in leading academic journals and books. Julia’s publications are based on research conducted in companies such as IBM, Tata Consultancy Services, SAP, Infosys, Cognizant, Pactera and many others. Mary C. Lacity  is the David D. Glass Chair and Distinguished Professor of Information Systems in the Sam M. Walton College of Business at the University of Arkansas. She was previously Curators’ Distinguished Professor at the University of Missouri and has held visiting positions at MIT, the London School of Economics, Washington University, and Oxford University. Liette  Lapointe  is Associate Professor of Information Systems at the Desautels Faculty of Management, McGill University. Her research focuses on behavioral aspects of information systems adoption and usage, including resistance to information technology implementation, enhanced use, ambivalence, collective use, and information technology

  Notes on Contributors 

xiii

addiction. She is also involved in the study of IT use in the context of healthcare, such as of IT in geriatrics, particularly on the issues related to Alzheimer’s disease and related disorders, on chronic care diseases, and on knowledge co-creation. Her work has been published in top journals, such as Information Systems Research, Journal of Information Technology, Journal of the Association for Information Systems, MIS Quarterly, Organization Science, and others. Yan Li  is an associate professor in the Center for Information Systems & Technology at Claremont Graduate University (CGU). Li’s research philosophy is grounded in design science and quantitative methods with a strong emphasis on relevance to the domain of interest. Her research focuses on data management and advanced analytics including machine learning, natural language processing, data warehousing, and health analytics. Her other research stream focuses on developing information and communication technology solutions for underserved populations, as well as improving social inclusion in healthcare. Paul  Benjamin  Lowry  PhD, is an Eminent Scholar and the Suzanne Parker Thornhill Chair Professor in Business Information Technology at the Pamplin College of Business at Virginia Tech, where he serves as the BIT PhD and Graduate Programs Director. He is a former tenured full professor at both City University of Hong Kong and the University of Hong Kong. He received his PhD in management information systems from the University of Arizona and an MBA from the Marriott School of Management. He has published 250+ publications, including 140+ journal articles in top IS journals such as MIS Quarterly, Information Systems Research, Journal of Management Information Systems (JMIS), Journal of the Association for Information Systems (JAIS), Journal of Information Technology and Information Systems Journal (ISJ). He is on the senior editorial board of JMIS and is a senior. Joseph  Manga recently joined the Mabee College of Business Administration faculty at Abilene Christian University in Abilene, Texas, after receiving his Ph.D. from University of Texas Rio Grande Valley (UTRGV). His research interests are in business intelligence and analytics, health information technology systems, and security policy compli-

xiv 

Notes on Contributors

ance. He has multiple publications and has presented in high-impact IS conferences and workshops. He has also reviewed for multiple IS journals and conferences. Lars Mathiassen  is a Georgia Research Alliance Eminent Scholar, professor in the Computer Information Systems Department, and cofounder of the Center for Digital Innovation at Georgia State University. His research focuses on digital innovation, on IT development and management, and on the use of IT for health services. He has published extensively in major information systems, management, and software engineering journals, and has co-authored several books, including Professional Systems Development, Computers in Context: The Philosophy and Practice of Systems Design, Object Oriented Analysis and Design, and Improving Software Organizations: From Principles to Practice. He has served as senior editor for MIS Quarterly, Information and Organization, and Journal of Information Technology. Earl  McKinney  is Professor of Management Information Systems at Bowling Green State University and previously at the United States Air Force Academy. He holds a Ph.D. from the University of Texas and a Master of Engineering from Cornell University. Earl’s research is in e-commerce, small team communication during a crisis, and theoretical work on the notion of information has been published in Behaviour and Information Technology, Human Factors, Information and Management, and MIS Quarterly. He consults for British Petroleum, the U.S. Forest Service, and several Air Force agencies on human factors and aviation communication issues. Willem Mertens  is a data scientist specializing on people analytics. He was People Reporting and Insights manager at Woolworths Group and previously a postdoctoral research fellow in information systems at Queensland University of Technology in Brisbane, Australia, and a research fellow of Vlerick Business School (Belgium). His primary topics of research interest include research methodology and the interaction between organizational processes and behavior. He has published in software engineering, information systems, retail, and sociology journals.

  Notes on Contributors 

xv

John  Mingers is an emeritus professor at Kent Business School, University of Kent, where he was previously Professor of Operational Research and Information Systems. John has an international reputation for his work on research metrics; the nature of information, meaning and knowledge; the use of systems methodologies in problem situations— multimethodology; and the philosophy of critical realism. He is an Academician of the Academy of the Social Sciences and has been an associate editor for MIS Quarterly. John has published over 200 academic papers. Book titles include: Systems Thinking, Critical Realism and Philosophy: A Confluence of Ideas (Routledge, 2015); Self-Producing Systems (Springer, 2013); and Multimethodology: Towards theory and practice and mixing and matching methodologies (editor, Wiley, 2020 edition). Lorraine  Morgan  is an associate professor and Funded Investigator with Lero (the Science Foundation Ireland Research Centre for Software) at the University of Galway. Her principal research interests include open innovation, open source software, inner source software, open business models and agile methods. Lorraine has published in leading journals including Information Systems Research, California Management Review, the European Journal of Information Systems, Journal of Strategic Information Systems and Information Systems Journal. Ilan Oshri  is Professor of Information Systems and the Director of the Centre of Digital Enterprise at the University of Auckland Business School (New Zealand). His research interests revolve around sourcing, digital transformation, digital sustainability and emerging technologies. Ilan conducted research and advisory with global firms such as IBM, Tata Consultancy Services, Boston Consulting Group, KPMG, Accenture and many others. His work was published in leading international journals. Ilan has published 22 books and dozens of industry reports and teaching cases on global sourcing, digital transformation and emerging technologies. Jan Recker  is an AIS Fellow, Alexander-von-Humboldt Fellow, Chaired Professor of Information Systems and Systems Development at the University of Cologne, and an adjunct professor at Queensland University

xvi 

Notes on Contributors

of Technology. His research focuses on systems analysis and design, digital innovation and entrepreneurship, and digital solutions for sustainable development. Suzanne  Rivard is Professor of Information Technology at HEC Montreal and is the HEC Montreal Endowed Chair in Strategic Management of Information Technology. She is a Fellow of the Royal Society of Canada, a Fellow of the AIS (Association for Information Systems), and an AIS Leo Award recipient. She received an honoris causa doctorate from Aix-Marseille Université and one from the Université de Lausanne. Her research interests are resistance to information systems implementation, outsourcing information systems services, software project management, and strategic alignment. Her work has been published in such journals as Communications of the ACM, Information and Management, Journal of Information Technology, Journal of Management Information Systems, Journal of Strategic Information Systems, MIS Quarterly, Organization Science, and others. Daniel N. Treku  recently joined the Business School faculty at Worcester Polytechnic Institute, Massachusetts, after receiving his Ph.D. from the UTRGV. His research focuses on blockchain technology, cryptocurrencies, and non-fungible token (NFT) price behaviors, data analytics, fintech, and conceptual and empirical studies of networks and digital platforms. He has multiple journal and book chapter publications and IS conference proceedings. He serves on the Big Data and Cognitive Computing review board and reviews for several IS and reference-­ discipline journals. Leslie  Willcocks is a professor emeritus at the London School of Economics and Political Science, Associate Fellow of Green Templeton College, Oxford, and co-editor of the Journal of Information Technology and JIT Teaching Cases. He has an international reputation for his work on automation and the future of work; ITO/BPO outsourcing; cloud computing; digital business; strategy; automation; IT and innovation; organisational change; and global business management. Leslie has published 70 books, and over 200 refereed papers in journals such as Harvard

  Notes on Contributors 

xvii

Business Review, California Management Review, Sloan Management Review, Journal of Management Studies, and MIS Quarterly. Recent books include: Global Business: Strategy in Context (SB Publishing, 2021); Global Business: Management (SB Publishing, 2021); and edited, with Nik Hassan, Advancing Theories in Information Systems Volume 1: Rationale and Processes (Palgrave Macmillan, 2021).

List of Figures

Fig. 2.1 Fig. 2.2 Fig. 2.3 Fig. 3.1 Fig. 4.1 Fig. 4.2 Fig. 4.3 Fig. 5.1 Fig. 5.2 Fig. 6.1 Fig. 6.2 Fig. 6.3 Fig. 6.4 Fig. 8.1 Fig. 9.1

Intermediate products in the discursive formation 23 Application of products of theorizing in media richness theory (MRT) 43 MRT framework adapted from Daft and Lengel (1986) 57 Service automation continuum (Lacity & Willcocks, 2017) 89 Peircean semiotics: core elements and their relationships (core elements of semiotics and their definitions are included in Appendix 2.) 123 Understanding digital objects through the lens of Peircean semiotics129 The foundations of a new paradigm 138 Cybernetic control 155 Cybernetic control by IT implementers 157 Peirce’s semiotic triangle 173 Jakobson’s six functions of semiotic systems 178 Stages in the interpretation of and response to signs 185 The relations between semiosis and the three worlds 187 Routinization of digital transformation of work view 263 Components of the concept of aesthetic business processes that want to observe aesthetic qualities, develop them as patterns to design aesthetic business processes, and finally think and learn together in companies 287

xix

xx 

Fig. 9.2

List of Figures

Patterns in business processes as a value chain that organizes the multidisciplinary communities Fig. 9.3 Aesthetic business processes in which an atmosphere can be experienced holistically, (a) in more or less stable networks (vertical direction) and (b) in more or less familiar activities (horizontal direction), inspired by communities of practice (Pyrko et al., 2019; Wenger et al., 2002) Fig. 9.4 The concept of aesthetic business processes is based on a macro-sequence that integrates the processes of knowledge creation, the four steps of visual thinking according to Dan Roam (2008) and life cycle of a pattern evolution Fig. 9.5 Visualizing business processes help to better understand the aesthetic qualities of business processes and their relationship to the core of a business process pattern Fig. 9.6 The concretization of the macro-sequence on methods and forms of visualizing by the students in the VisualBP project Fig. 12.1 Proposed theory of information exchange (ToIE) Fig. 13.1 Characteristics of and threats to the hypothetico-deductive research cycle

289

290

292 295 298 375 390

List of Tables

Table 1.1

Products of theorizing illustrated in this volume (adapted from Hassan et al., 2022) 4 Table 2.1 Summary of products of theorizing 25 Table 2.2 Implications from using products of theorizing 63 Table 3.1 Action principles after the first round of data collection 95 Table 3.2 Client adoption journeys 98 Table 3.3 Action principles after 22 adoption journeys data 102 Table 7.1 Thematic mapping of archetypes from literature on Prometheus218 Table 7.2 Characterisation of illustrative examples 224 Table 8.1 Comparing our routinization view with routine capability view (adapted from Swanson (2019)) 272 Table 8.2 Variations in performative and ostensive routines as practice components 274 Table 8.3 Capabilities supporting routine work content design transformation276 Table 8.4 Capabilities supporting routine work process transformation277 Table 8.5 Capabilities supporting routine sustainment of work transformation278 Table 11.1 The displacement of concepts and growth of knowledge 350

xxi

xxii 

List of Tables

Table 13.1 Main findings from the coding of 43 published IS papers between 2013 and 2016 that follow the hypothetico-­ deductive model of science Table 13.2 Change proposals by stage of the hypothetico-deductive model of science, differentiated by level of programming and implementation timeframe Table 13.3 New guidelines for hypothetico-deductive IS researchers Table 13.4 A diversified model of the peer review and publication process, by stage of the scientific process, with examples

403 410 414 419

1 Introduction: On Types of Products of Theorizing Suzanne Rivard , Leslie Willcocks and Nik Rushdi Hassan

,

The information systems (IS) field represents the exciting multidisciplinary area that links the rapidly changing technology of information (or communications and information technology, ICT) to the business and social environment. Lately, the discourse surrounding information and systems has leaped into the public’s consciousness in unprecedented ways through the rise of social media, the extensive use of artificial

S. Rivard (*) HEC Montréal, Montréal, QC, Canada e-mail: [email protected] L. Willcocks Department of Management, London School of Economics and Political Science, London, UK N. R. Hassan Labovitz School of Business and Economics, University of Minnesota Duluth, Duluth, MN, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. P. Willcocks et al. (eds.), Advancing Information Systems Theories, Volume II, Technology, Work and Globalization, https://doi.org/10.1007/978-3-031-38719-7_1

1

2 

S. Rivard et al.

intelligence (AI), cognitive automation, deepfake, the metaverse, and the weaponization of information, to name a few. Unfortunately, the concerns stemming from these IS are rapidly overtaking the ability of the field to address them, partly because the field lacks its native theories. The IS field is well-known to rely on theories from its “reference disciplines,” such as management, social psychology, economics, communication, and computer science. What is not happening to the extent that is exemplary of any mature discipline is the development of its native theories. The “Advancing Information Systems Theories” series aims to encourage and promote IS research that goes beyond borrowed legitimization and derivative research toward new and original research that naturally comes from its own theories—information system theories. The first volume of the series (Hassan & Willcocks, 2021b), subtitled “Rationale and Processes,” aimed at defining theory, understanding, and explaining why indigenous IS theories are needed, presenting theory development practices and processes, and proposing original approaches for the development of rich theories. In line with the objective of advancing IS theories and maintaining the long-term growth and vitality of the IS field, the editorial team for this second volume has espoused a conceptualization that departs from a view that portrays theory as a complete, detailed, flawless, deep, and exhaustive explanation of a phenomenon, an object that Weick (1995) refers to as “Theory That Sweeps Away All Others” (p. 386). Instead, this second volume is devoted to “products of theorizing” (Hassan, Mathiassen, & Lowry, 2019; Hassan & Willcocks, 2021b; King, 2021), which celebrates the outputs of the “interim struggles” (Weick, 1995) that characterize theory generation. It promises that interim conceptual artifacts of theorizing, such as metaphors, frameworks, and concepts (Lowry, Petter, & Leimeister, 2020), are valuable to the discipline as they can be leveraged and exploited further in subsequent research. In this way, a discipline might inch its way toward native theories over time (Rivard, 2021). We acknowledge that the process of theorizing can be long and arduous and, like all great things, will not be completed in a day, much less in an edited series. Nevertheless, the products of that work remain valuable

1  Introduction: On Types of Products of Theorizing 

3

because they bring us closer to theories with a capital “T.” Consistent with claims made in the first volume, limiting theory to verificational dimensions of scientific practice would limit the very potential of science itself and have an impoverishing impact on theorizing. Products of theorizing that are not verifications have their place on the epistemological continuum. Instead of theories, they may be called “questions,” “paradigms,” “concepts,” “propositions,” “frameworks,” or “models.” They all represent interim struggles, semantic forms of theory, or linguistic variations. What they are is useful. They are the products of theorizing and are precursors to strong theory, and if they are fresh and original, they go a long way in advancing IS theories (Hassan, Lowry, & Mathiassen, 2022). Following our call for contributions, we received several manuscripts. Interestingly, as synthesized in Table 1.1 and detailed below, some of the chapters resulting from our selection process mobilized or proposed more than one product of theorizing. The chapters we selected, however, did not cover all 12 types that we wished to illustrate; therefore, we drew on a few already published papers to present a complete set. Setting the stage for this volume, Hassan, Mathiassen, and Lowry (Chap. 2—Useful Products of Theorizing) introduce well-known but rarely analyzed and discussed products of theorizing. They propose three important roles, which they combine in a theorizing framework, for the 12 products of theorizing they introduce. In this framework, question, paradigm, law, and framework can play the role of theory frames that help define contextual and conceptual boundaries of theory (Rivard, 2021). Myth, analogy, metaphor, and model are theory generators as they help theorists develop their explanation of the phenomenon they aim to illuminate. “Models may, for example, serve as informal conceptions that operate as analogies or commentaries for a theory” (Hassan et al. 2022, p. 421). Finally, concept, construct, statement (proposition), and hypotheses are theory components. They are the building blocks that may emerge from the adoption of a given theory frame or the use of a theory generator. The authors further define and illustrate each product and illustrate the usefulness of their theorizing framework.

4 

S. Rivard et al.

Table 1.1  Products of theorizing illustrated in this volume (adapted from Hassan et al., 2022) Product role Product

Definition

Theory frame

A disciplinary question Chapter 3: “Debating Genres invites solving a and IS Research: The Case problem by of Action Principles for mobilizing the Service Automation”— unique discourse of Willcocks, Lacity, and a field. Gozman Primary product: Question Secondary product: Statements A paradigm is an Chapter 4: “A Paradigm Shift ensemble of in Understanding Digital scientific practices Objects in IS: A Semiotic that serve as Perspective on Artificial archetypical Intelligence problems and Technologies”—Kotlarsky solutions that are and Oshri agreed upon by a Primary product: Paradigm community of researchers. A law is a generalized Chapter 5: “A Cybernetic statement that Theory of the Impact of explains or predicts a Implementers’ Actions on phenomenon. User Resistance to Information Technology Implementation”—Rivard and Lapointe Primary product: Law Secondary products: Analogy, model, and statements A framework is map Chapter 6: “Interrogating of the phenomenon Sociomateriality: An under study and Integrative Semiotics includes the main Framework for Information concepts, constructs, Systems”—Mingers and variables, and their Willcocks relationships. Primary product: Framework Secondary product: Model

Question

Paradigm

Law

Framework

Chapter

(continued)

1  Introduction: On Types of Products of Theorizing 

5

Table 1.1­ (continued) Product role Product

Definition

Theory generator

A myth is “a dramatic Chapter 7: “When Crowds narrative of Play God: A Promethean imagined events … Perspective on used to explain Crowdfunding”—Conboy, origins or Gleasure, and Morgan transformations of Primary product: Myth something … an unquestioned belief about the practical benefits of certain techniques and behaviors that is not supported by demonstrated facts” (Trice & Beyer, 1984, p. 655, cited by Hassan et al., 2022). An analogy is a partial Chapter 8: “Routinization of similarity between Digital Transformation of two objects or Work: A Discursive Practice phenomena which is Orientation Towards a used to provide an Native IS Theory”—Treku, explanation. Manga, and Ayaburi Primary product: Analogy A metaphor “consists Chapter 9: “Patterns for in giving the things a Visualizing the Aesthetic name that belongs Qualities of Business to something else” Processes”—Blattmeier (McKeon, 1941, Primary process: Metaphor p. 1476, cited by Secondary product: Concept Hassan et al., 2022). A model is a Secondary product for Chaps. representation— 5 and 6. most often a simplification—of a phenomenon under study. It serves to represent the relationships among objects.

Myth

Analogy

Metaphor

Model

Chapter

(continued)

6 

S. Rivard et al.

Table 1.1­ (continued) Product role Product

Definition

Chapter

Theory Concept component

A concept is an abstract idea or the generalization of multiple instances of an object. A construct is a term for a concept that is neither directly nor indirectly observable and can be defined only in relation to those observables, whereas a variable is an observational term that changes for a construct (Kaplan, 1964, cited by Hassan et al., 2022). A statement is a sentence that invokes the authority of a discipline with which it is associated (Hassan et al., 2022).

Chapter 10: “Information Theory in IS”—McKinney and Boell Primary product: Concept

Construct

Statement

Hypothesis

A hypothesis is an operationalized proposition that takes the form of empirically testable conjectures or follows a procedural rule to infer other propositions (Kaplan, 1964, cited by Hassan et al., 2022).

Chapter 11: “The Primacy of Concepts and Implications for the IS Field”—Hassan Primary products: Concepts and constructs

Chapter 12: “Propositions for a Future Information Exchange Theory to Support Decision Making”—Bazarah and Li Primary product: Propositions Secondary product: Model Chapter 13: “New Guidelines for Null Hypothesis Significance Testing in Hypothetico-­Deductive IS Research”—Mertens and Recker Primary product: Hypothesis

1  Introduction: On Types of Products of Theorizing 

7

Theory Frames Chapter 3, entitled “Debating Genres and IS Research: The Case of Action Principles for Service Automation” by Willcocks, Lacity, and Gozman, exemplifies two products of theorizing. The first product is a disciplinary question, which plays the role of theory frame, as it “addresses an object of study as a problem requiring a solution based on the field’s rules of discourse and pattern of inquiry” (see Chap. 2). In Chap. 3, the authors challenge the traditional debates on whether IS is a discipline or a field, should focus on theorizing, or become much more applied, whether qualitative or quantitative methods are preferable and whether rigor or relevance should be the primary criteria for IS scholarship. In view of the emerging set of technologies related to robotic processes, cognitive automation, and artificial intelligence, the authors address the question of how well-positioned IS researchers are to study and provide timely, useful findings about these emerging technologies, and how can they improve on making our studies both rigorous and much more relevant? Following the helpful suggestions emerging from the genres in IS research debate, the chapter proposes, theorizes, and demonstrates a research-into-practice approach, designed to study emerging technologies and produce timely findings that are useful to both researchers and practitioners, as both try to get to grips with the development, deployment, and impacts of emerging technologies, in this case automation tools and platforms. The chapter demonstrates the research-into-practice approach drawing on 5 surveys and interviews on 22 adoption journeys conducted into service automation. Following their study of organizational adopters in different industries that deployed a range of automation tools and that reported a range of benefits, the authors propose a second product of theorizing, this one playing the role of theory components, in a set of 38 “action principles” or statements, which other organizations can enact and researchers learn from. Indeed, as data synthesizers these statements can also become useful components that can be mobilized for further theorizing. In Chap. 4 (“A Paradigm Shift in Understanding Digital Objects in IS: A Semiotic Perspective on Artificial Intelligence Technologies”), Kotlarsky

8 

S. Rivard et al.

and Oshri argue that the study of artificial intelligence (AI) solutions calls for a paradigm shift. As an intermediary product of theorizing, a paradigm plays the role of a theory frame, as it is described by Kuhn (1977) as a “shared exemplar for scientific practice in the form of concrete manifestations of analogies and metaphors on which communities of scientists and researchers agree in part or whole” (Hassan et  al., 2022, p.  422). Kotlarsky and Oshri show that when conceptualizing and theorizing digital objects with a semiotic paradigm, the IS field has bestowed stability in the relationship between the signifier, signified, and referent, which they coin a particular semiotic paradigm. They suggest that the nature of AI solutions, a type of digital object, challenges the existing paradigm as these digital objects are self-learning and self-programming, thus creating an ever-changing signifier and signified for which there will be a convention not obvious to a human actor. They, therefore, propose a pluralistic semiotic paradigm that accommodates the unique characteristics of artificial intelligence solutions and offers guidance to future studies to consider alternative approaches when studying the meaning of digital objects. To theorize the effects of implementers’ responses to user resistance, Rivard and Lapointe (Chap. 5—“A Cybernetic Theory of the Impact of Implementers’ Actions on User Resistance to Information Technology Implementation”) draw upon the law of requisite variety to theorize the effect of information technology implementers’ actions on user resistance. The law of requisite variety (Ashby, 1963) posits that to maintain the state of a system within a range of acceptable values, a control mechanism must possess as many types of responses as there are system states. The authors also mobilize another product of theorizing, analogy. Here they use the analogy of an implementer as a cybernetic system characterized by the notion of control, which has as its primary requirement the need to maintain the level and kind of output necessary to achieve the system’s objectives. These responses can be preprogrammed, provided by decision rules, or generated by the control mechanism’s ability to generate control responses. The authors conceptualize an IT implementation as a limited system made up of implementers, users, and an IT

1  Introduction: On Types of Products of Theorizing 

9

application. They propose a cybernetic model wherein the implementers are the control device of the cybernetic system, and their objective is to keep the intensity of user resistance at a level that is acceptable from an organizational point of view. To this end, implementers engage in various actions in response to user resistance behaviors. These actions constitute the feedback sent to the system by the control device. The authors make statements—propositions—that posit that some implementers’ actions have a negative feedback effect and maintain user resistance within an acceptable range. Other implementers’ actions have a positive feedback effect that will lead to significant organizational disruption, which may ultimately require the abandonment of the IT implementation. Frameworks also play the role of theory generator, as illustrated by Mingers and Willcocks in Chap. 6—“Interrogating Sociomateriality: An Integrative Semiotics Framework for Information.” In this chapter, the authors seek to contribute products of theorizing by interrogating the concept of sociomateriality for information systems studies using Peircean semiotics and critical realism as philosophies. The chapter contributes to the theme of the book by presenting an emergent integrative semiotic framework as a product of theorizing. The chapter explores key theories. Concepts of information, meaning, and embodiment are also utilized to help build an alternative integrative framework, consisting of three worlds—the personal, social, and material—in relationships of sociation, sociomateriality, and embodiment. Semiosis relates to the personal world through the generation and interpretation of signs and messages. It relates to the material world in that all signs must have some form of physical embodiment in order to be signs and must also be transmitted through some form of physical media. Semiosis relates to the social world in that the connotive aspects of sign systems are social rather than individual— they exist before and beyond the individual’s use of signs. The chapter examines critically the implications of this formulation for studying IS. It discusses commonalities with and departures from many sociomaterial studies, illustrates points with empirical examples, and details how the integrative framework can be utilized.

10 

S. Rivard et al.

Theory Generators As theory generators, myths serve to inspire and facilitate the explanation of a phenomenon. Chapter 7—“When Crowds Play God: A Promethean Perspective on Crowdfunding” by Conboy, Gleasure, and Morgan— illustrates the theorizing power of myths. The authors draw upon the myth of Prometheus—“the Greek god who tried to free humankind from the oppression of the gods by stealing fire from Zeus and giving it to humans, thus giving unprecedented control of their environment and destiny” (see Chap. 7) to theorize crowdfunding. They use this mythical figure associated with “defiant progress” and technological advancement—as a sensitizing device to develop a set of “competing, dialectic archetypes.” Crowdfunding projects have been the subject of contrasting narratives. To many, they are the antithesis of predatory bottom-line business ventures, and to others, they are an under-regulated vehicle for immature or unscrupulous project owners to exploit inexperienced and vulnerable investors. After introducing their six dialectical archetypes of crowdfunding, the authors mobilize Hegelian dialectic analysis of three high-profile crowdfunding campaigns. The overarching contribution of the study is that it provides a foundation for discussion of the positive and negative narratives surrounding crowdfunded project owners and explicates the limitations of crowdfunding as an enabler of positive systemic change. The dialectic approach provides a systematic means of identifying the essence of disagreement between narratives. While it may be too early to predict the outcomes for emerging technology-driven initiatives such as crowdfunding, the use of myth offers a sophisticated means to look for “rhyming” phenomena, where the phenomena at play are similar to the grand frailties of humankind throughout history. When using an analogy as a product of theorizing, authors refer to something familiar to explain or illustrate something more complex or less familiar (see Chap. 2). In Chap. 8, entitled “Routinization of Digital Transformation of Work: A Discursive Practice Orientation Towards a Native IS Theory,” Treku, Manga and Ayaburi leverage “analogizing” as a basis for developing a theory of routinization in the digital transformation of work. The authors observe that digital technology is repeatedly used to transform or improve how work is done via trials, experiments,

1  Introduction: On Types of Products of Theorizing 

11

and sound practices. They suggest that while much is known about factors that influence the adoption of technologies and their subsequent deployment in the workplace, there is no unifying theory for explaining the routinization of digital transformation of work, although the concept of organizational routines has been well-advanced for attaining meaningful work practices. The chapter accomplishes a double purpose: (1) it expounds routinization as an organizational capability and (2) it develops a native information systems theory of routinization of digital transformation of work. In Chap. 9—“Patterns for Visualizing the Aesthetic Qualities of Business Processes”—Monika Blattmeier introduces a new concept of business processes called the aesthetic business process by drawing on the metaphor of the atmosphere as a theory generator. The use of the atmosphere metaphor changes the experience of working with business processes from a rational and logical perspective to a more holistic and imaginative lived experience where members of the business process realize patterns of business processes that enable them to interact in more inventive and creative ways. The author also proposes a theory component, the concept of aesthetic business processes that aims to engender creative thinking in business processes. This concept complements the commonly viewed notion of business process improvement that is focused on effectiveness, efficiency, and cycle-time reduction. Inspired by the German philosopher Gernot Böhme, aesthetic business processes bring to bear the “atmosphere” of the business process, referring to what is given to the senses so that those involved in the process can holistically perceive a shared reality. The term atmosphere in business processes means what is given to the senses so that those involved in the process can holistically perceive a shared reality. The months of the Corona virus pandemic reminded us of the desire for presence, physicality, and three-­ dimensionality when collective work became increasingly digitalized. The aim of aesthetic business processes is to foster communities of practice, even in digital space, that can think artistically about what they do, use their imagination, experience their work as it unfolds and evaluate it not just according to rules but also according to what they feel. The chapter begins with a definition of aesthetic business processes, followed by how visualizations and patterns are used to abstract the shared reality in

12 

S. Rivard et al.

business processes. The macro sequence of the concept integrates processes of knowledge transformation, visual learning, and the life cycle of a pattern development. Patterns of aesthetic business processes emerge through careful observation and the explication of the aesthetic qualities of business processes. Finally, the application of the concept is demonstrated by students of a university visualizing business processes together in imaginative ways. We have not just a single but in fact several chapters that illustrate a model as a product of theorizing. For instance, in Chap. 6, Mingers and Willcocks discuss and theorize about several models—for example, Peircean, Saussurean, Morris’, Stamper’s, Jakobson’s—that could be used to represent the reality of how the personal, social, and material realities interrelate. They chose the Peircean model as their preferred choice and they anchored the framework they propose in this Peircean model. On the other hand, Rivard and Lapointe—Chap. 5—developed a cybernetic-­ inspired model of implementers’ responses to user resistance and their effect and used this model to develop their propositions.

Theory Components In Chap. 10—“Information Theory in IS”—McKinney and Boell argue that despite the term information being part of the IS field’s name, this concept is one of the least studied concepts in the field. The authors review the frameworks, taxonomies, and concepts related to information and how information is researched in the IS field. They adopt a broad conceptual view of information, intending to cover both the everyday, practical use of the term and how it is defined theoretically. Everyday conversations refer to the information age and the information society, to the strategic role of information that is a source of competitive advantage and the lifeblood of the organization. Analytical discussions refer to individuals providing or receiving information, security discussions emphasize the importance of protecting information, and decision-making is based on information. The authors set three goals for their chapter. First, organize and describe the existing landscape of theories of information. Second, suggest what a theory or theories of information could look like

1  Introduction: On Types of Products of Theorizing 

13

to address IS research interest by looking at concepts frequently related to information. And third, identify possible avenues for future research in IS, in particular by recognizing that information is not a static or singular concept or phenomenon but changes depending on situational, temporal, and cultural contexts, as well as epistemological assumptions and domain of application. They suggest that as a result, theories of information written by IS scholars will be more generally applicable for use beyond the IS domain and provide analytic distinctions that can make a difference in everyday practice. In Chap. 11—“The Primacy of Concepts and Implications for the IS Field”—Nik Hassan raises an important question: what are the concepts underlying constructs? The hypothetico-deductive approach that makes up the majority of research in IS depends on its constructs, which are nonobservables invented to provide solutions and measures to further research. Constructs, however, depend on their underlying concepts, which are rarely discussed explicitly in IS research, and often are conflated with constructs, making comparisons among research studies challenging. Historical examples of the dangers of not understanding the right concepts are illustrated to demonstrate the practical nature of concepts. And in order to take IS researchers out of their comfort zone with constructs, the chapter goes over how concepts, not constructs, form the foundation of the growth of established disciplines throughout history. This brief historical review is followed by a description of the relationship between concepts and constructs and implications such a relationship has for the IS field. The authors of Chap. 12—“Propositions for a Future Information Exchange Theory to Support Decision Making”—deplore that although information exchange (IE) theory is prevalently addressed in IS research, the field still lacks a theory that explains it. In this chapter, Bazarah and Li aim to understand information exchange behavior and its impact on facilitating the decision-making process in IS research. They argue that extant research generally relies on borrowed theories to explain IE and instead of developing theories native to the IS field. The authors deem that theories indigenous to IS are called for, especially because of the unique element in IE compared to other social exchanges such as goods or services. The primary product of theorizing in this chapter is a set of propositions that synthesize the theory of information exchange proposed by the authors.

14 

S. Rivard et al.

In Chap. 13—“New Guidelines for Null Hypothesis Significance Testing in Hypothetico-Deductive IS Research”—Mertens and Recker advocate a policy change among parties—for example, journals, scholars, and students—involved in hypothetico-deductive IS research. Indeed, the authors share their concerns about the design, analysis, reporting, and reviewing of quantitative IS studies that draw on null hypothesis significance testing (NHST). They note that although debates on misinterpretations, abuse, and issues with NHST have persisted for about half a century in many fields, they have remained essentially absent in IS. Mertens and Recker qualify this position as “untenable for a discipline with a proud quantitative tradition.” The chapter discusses threats associated with the application of NHST and illustrates how these threats manifest in recent IS scholarship. To encourage the development of new standards for NHST in hypothetico-deductive IS research, the authors propose “a balanced account of possible actions that are implementable short-term or longterm and that incentivize or penalize specific practices,” and they propose two sets of guidelines to support researchers in their work.

Concluding Remarks This volume illustrates the wide variety of products of theorizing that IS researchers can mobilize or develop when they endeavor to develop indigenous theories. It also illustrates that a given study can produce or mobilize several products of theorizing that complement each other in many ways. We hope those illustrations will inspire other researchers to explore and experiment the numerous products of theorizing.

References Ashby, W. R. (1963). Introduction to cybernetics. John Wiley & Sons. Hassan, N., Lowry, P. B., & Mathiassen, L. (2022). Editorial-useful products in information systems theorizing: A discursive formation perspective. Journal of the Association for Information Systems, 23(2), 418–446.

1  Introduction: On Types of Products of Theorizing 

15

Hassan, N., Mathiassen, L., & Lowry, P. (2019). The process of information systems theorizing as a discursive practice. Journal of Information Technology, 34(3), 198–220. Hassan, N., & Willcocks, L. (2021a). Introduction: Why theory? (Mis) understanding the context and rationale. In I.  N. R.  Hassan & L.  P. Willcocks (Eds.), Advancing information system theories: Rationale and processes (Vol. 1, pp. 1–52). Palgrave Macmillan. Hassan, N., & Willcocks, L. (Eds.). (2021b). Advancing information system theories: Rationale and processes. Palgrave Macmillan. Kaplan, A. (1964). The conduct of inquiry: Methodology for behavioral science. Chandler. King, J. (2021). Who needs theory? MIS Quarterly, 45(1), 495–498. Kuhn, T. S. (1977). The essential tension: Selected studies in scientific tradition and change. University of Chicago Press. Lowry, P. B., Petter, S., & Leimeister, J. M. (2020). Desperately seeking the artefacts and the foundations of native theory in gamification research: Why information systems researchers can play a legitimate role in this discourse and how they can better contribute. European Journal of Information Systems, 29(6), 609–620. McKeon, R. P. (1941). The basic works of Aristotle. Random House. Rivard, S. (2021). Theory building is neither an art nor a science. It is a craft. Journal of Information Technology, 38(3), 316–328. Trice, H. M., & Beyer, J. M. (1984). Studying organizational cultures through rites and ceremonials. Academy of Management Review, 9(4), 653–669. Weick, K. E. (1995). What theory is not, theorizing is. Administrative Science Quarterly, 40(3), 385–390.

2 Useful Products in Information Systems Theorizing: A Discursive Formation Perspective Nik Rushdi Hassan , Paul Benjamin Lowry, and Lars Mathiassen

Introduction Although major progress has been made in describing the nature of information systems (IS) theory (Gregor, 2006; Gregor & Jones, 2007) and in evaluating and refining existing theories (Grover et  al., 2008; Weber, 2012), the status of theories in IS has come under intense debate (Avison

N. R. Hassan (*) Labovitz School of Business and Economics, University of Minnesota Duluth, Duluth, MN, USA e-mail: [email protected] P. B. Lowry Business Information Technology, Pamplin College of Business, Virginia Tech, Blacksburg, VA, USA L. Mathiassen Computer Information Systems Department, The Center for Digital Innovation, Georgia State University, Atlanta, GA, USA © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. P. Willcocks et al. (eds.), Advancing Information Systems Theories, Volume II, Technology, Work and Globalization, https://doi.org/10.1007/978-3-031-38719-7_2

17

18 

N. R. Hassan et al.

& Malaurent, 2014; Gregor, 2014; Grover, 2012; King & Lyytinen, 2004; Straub, 2012; Weber, 2006). Avison and Malaurent’s (2014) and “theory fetish” critique suggests the emphasis on IS theory has produced less-thaninteresting research; Grover and Lyytinen (2015) claim that scripted research strategies that domesticate theories from other disciplines lead to a lack of boldness and originality in IS research; Markus (2014) suggests that lack of contribution may be due to narrow definitions or conflicting notions of IS theory as opposed to an overemphasis on IS theory; and Gregor (2014) argues that the discussion surrounding theory in IS may lead to questioning “theory” in itself and proposes that the IS field should strive to understand the theorizing process rather than debate “theory.” The status of theory in research cannot be placed in doubt. Philosopher and behavioral scientist Kaplan (1964, p. 294) regards theorizing as “the most important and distinctive” activity of human beings: “to engage in theorizing means not just to learn by experience but to take thought about what is there to be learned … lower animals grasp scientific laws but never rise to the level of scientific theory” (p. 295). Corley & Gioia (2011, p. 12) state that “theory is the currency of our scholarly realm,” and Alvesson and Sandberg (2011, p. 247) note that “as researchers, we all want to produce interesting and influential theories.” Academic priority is given first to those who can build original and interesting theories, second to those who can use them effectively, and third to those who understand them. Even for theory creators, only when scholars outside of their disciplines acknowledge and apply their theories can they say that their theories are useful (Corvellec, 2013). With the advent of new, unprecedented digital phenomena, theory development is becoming even more critical, and IS researchers are calling for the IS field to move beyond narrow definitions of knowledge and theory. Ågerfalk (2014) argued that contribution to knowledge is not limited to theoretical contributions but includes empirical contributions that challenge existing assumptions or reveal insights into a phenomenon without relying on any a priori theory. Hirschheim (2019) lamented the IS field’s obsession with positivist theories and, at a workshop panel (Willcocks et al., 2019), Dennis bemoaned the fixation of top IS journals with what Gregor (2006) calls Type IV theories—theories for explaining and predicting—a positivist position that limits theory to “a system of

2  Useful Products in Information Systems Theorizing… 

19

constructs and variables in which the constructs are related to each other by propositions and the variables are related to each other by hypotheses” (Bacharach, 1989, p. 498). Such a narrow view of theory excludes classical theories, such as Linnaean’s taxonomy (1735) in natural history, Mendeleev’s (1869) period table of elements in chemistry, and Weber’s (1930) Protestant ethic in sociology. Moreover, it excludes early IS theories such as Mason and Mitroff’s (1973) cognitive style theory and Gorry and Scott-Morton’s (1971) foundations of decision support systems, both of which have spawned decades of productive IS research. Moving beyond the dichotomy between what theory is and is not, Weick (1995) suggested viewing theory as a continuum of approximations. As interim struggles in the process of theorizing (Runkel & Runkel, 1984), these approximations hold the key to building exciting theories by opening up spaces for future thinking (Moore, 2004). Consequently, we consider all types of theories covered by Gregor (2006)—descriptive, explanatory, predictive, and prescriptive—as legitimate and suggest the IS field is best served at this stage of its conceptual development by unpacking the theorizing process with the help of what Weick (1995, pp. 385–389) called the “products” of that process: Products of the theorizing process … represent interim struggles in which people intentionally inch toward stronger theories … Those emergent products summarize progress, give direction, and serve as place markers. They have vestiges of theory but are not themselves theories.

Weick’s use of the term “products” emphasizes that the theorizing process produces approximations of theory but not necessarily theories themselves. As these approximations serve as the foundation for further theorization, they should not be dismissed just because they do not qualify as full-blown theories. Hence, we draw on Hassan et al.’s (2019) work on IS theorizing as a discursive practice and assemble 12 products—question, paradigm, law, framework, myth, analogy, metaphor, model, concept, construct, statement, and hypothesis—into one treatise to describe their primary roles as theory frames, theory generators, and theory components in the theorizing process. Whereas Hassan et al. (2019) focused on the theorizing processes, specifically the differences between the context of

20 

N. R. Hassan et al.

discovery and the context of justification, this chapter focuses on the products that are applied in those contexts. As such, we contribute to the current debates surrounding IS theories by addressing the question: “What coherent framework and theorizing toolset could IS researchers use to support their efforts in building original, interesting and influential theories?” Unfortunately for IS researchers, the toolset of products of theorizing is scattered among different disciplines, with no attempt to integrate in a meaningful way the many elements of theorizing. Also, the products of theorizing have also not been critically analyzed in the IS field, leaving IS researchers with little guidance on how to theorize. To fill this gap, we gather all 12 products of theorizing into one matrix and provide specific examples of such products in the chapters that follow. In this chapter we show how these 12 products are marshaled in theorizing, using the classic historical case of media richness theory (MRT) (Daft & Lengel, 1986; Daft et al., 1987). As a well-known, mature IS theory that has undergone fierce criticism, MRT allows for an in-depth analysis of how the theory maintained its validity and evolved in the face of criticism.

 heorizing and the Discursive T Formation Perspective Fundamentally, theorizing is about making claims in the form of theories, and the study of theory formation can be found in a recent development of discourse analysis (Schiffrin et al., 2001) called critical discourse analysis (Weiss & Wodak, 2003) that provides a rich framework for examining issues between theory and practice and between theory and methodology. An approach within this genre, which we apply in this chapter, is Foucault’s (1970, 1972) study of disciplinary activity, which describes how disciplines establish power relations to exert their authority. Using Foucault’s method of analyzing discourses, it is possible to answer questions about the validity of knowledge in certain social contexts, such as How did this knowledge emerge and gain influence? What are the components of that knowledge and theory that make up human consciousness? What can be claimed and what cannot be claimed?

2  Useful Products in Information Systems Theorizing… 

21

Understanding a field’s discursive formation is pivotal because it explains how the field of study emerges and gains influence as it attempts to legitimize the authority of its products of theorizing and its theories. The rules of this disciplinary activity, which Foucault (1972, p. 31) calls the discursive formation, establish relations that define the nature of the field. The operation of these rules makes possible the creation of new objects of study and makes claims about objects that belong to a specific discourse such that we can recognize economic discourse from psychological discourse, biological discourse from medical discourse, and computer science (CS) from IS discourse.1 Foucauldian discourse analysis enables us to break down theory into its products of theorizing to explain how they reinforce action and exert power.2 By drawing on what Foucault (1970, 1972) calls the “archeology of knowledge,” which describes the detailed historical development of disciplines and uncovers their ontological and epistemological assumptions, we elaborate on the products of theorizing using terms familiar to IS researchers. Recognizing the distinctiveness of the IS field’s discourse is especially consequential because of the field’s diversity and porous boundaries. The multidisciplinary nature of IS  Theorizing in biology thus takes a different form than theorizing in medicine because they are different discourses, even though statements about organs of the human body, tissues, and cells are found in both disciplines. The rules of discourse of biology concern the study of organic structures that support life. Conversely, the rules of discourse of medicine concern the observation of the human body to identify diseases that affect its health. Similarly, Revens (1972, p. 486) describes the discourse of CS as “computing techniques and appropriate languages for general information processing, for scientific computation, for the recognition, storage, retrieval, and processing of data … and … automatic control and simulation of processes,” which concerns the rules surrounding symbol processing (Denning et al., 1989) and differs from that of IS even though they share the same core concern: the computer. 2  When an IS researcher applies economic theory to study the use of computers using rules concerning value, prices, costs, and trade-offs, which are part of the discursive formation of economics, the power of the economic discourse influences the direction of the study and by extension the IS field. These cross-disciplinary activities present an interesting dilemma to IS researchers. The legitimacy already established by the recognized rules from these “reference disciplines” provides an effective career-building path for IS researchers but at the cost of not building a cumulative tradition within the IS discourse. Additionally, this phenomenon raises the key issue of which discourse rules one should follow: IS or economics. The related issue is whether the researcher is conducting economics research, IS research, or economics research in an IS context. The choice of applying specific rules of discourse has wide-ranging implications, especially in the ability of the IS field to invent its own native theories. If the field believes that the growth of its knowledge depends on inventing its own concepts, statements, and theories (Markus & Saunders, 2007), then leveraging the discourse of other disciplines is unlikely to support such a goal and the IS field will remain multimodal, unable to produce theories with a capital “T.” 1

22 

N. R. Hassan et al.

creates a confusing and precarious state where discourses from other disciplines spill into the IS field, causing IS researchers to continually vacillate between influential disciplines.3 As Foucault (1970, p. 64) explains, the discourse of a field or discipline is established before the field emerges, and theorizing takes place at the same time, giving “rise to certain organizations of concepts, certain regroupings of objects, certain types of enunciation, which form, according to their degree of coherence, rigor, and stability, themes or theories.” Similar to Abbott’s (2001) description of how disciplines coalesce around their axes of cohesion (Sarker et  al., 2019), some discourses eventually become fields of study, and because each field follows different rules and strategies in forming its discourse, each field builds unique theories concerning its phenomenon of interest. If any IS theorizing is to take place, it can only exist within IS discourse because that is where statements belonging to the field and its theories are situated and where the axis of cohesion resides (Sarker et al., 2019). That does not mean that IS discourse cannot exist within another discourse. For example, when a lawyer applies medical evidence to defend a client, medical discourse operates within legal discourse. Thus, when the discourse of computer science emerged in the late 1940s and early 1950s as a result of the invention of the computer, a different discourse called IS was emerging at roughly the same time and was later formalized in early IS textbooks (Hirschheim & Klein, 2012) and degree programs. The differences between the rules of these two discourses tell us who is speaking, the culture the person is part of, and on what authority or social institutions the person is involved. Studying the field’s discourse is pivotal for theorizing because the discursive formation provides answers to field-specific questions.4  During its formative stages, IS largely followed the rules laid down by the psychological discourse (cf. Mason & Mitroff, 1973) and, even today, social psychology continues to exert a strong influence (cf. Davis, 1989; Venkatesh et al., 2003). Later, the strategic management field exerted its influence (cf. Ives & Learmouth, 1984; Parsons, 1983) followed by other discourses such as CS, engineering, management, economics, and communication. 4  Field-specific questions determine one particular statement or proposition over that of another. Why was this theory formulated instead of another? Why were certain boundary conditions chosen? For example, medical questions will produce different answers related to suicide compared to, say, psychological or sociological questions even though the phenomenon is the same. 3

2  Useful Products in Information Systems Theorizing… 

23

Discursive Formation Theory Frames

Theory Generators

Question Paradigm Law Framework

Myth Analogy Metaphor Model

Theory Components Concept Construct Statement Hypothesis

Fig. 2.1  Intermediate products in the discursive formation

Through the interplay of rules that govern the formation of mutually exclusive objects of study, the discursive formation binds together a group of disparate concepts and statements while the discursive formation itself remains stable. Thus, according to Foucault, discourses are groups of statements in that they belong to the same discursive formation. We represent this discursive formation in Fig. 2.1 as the structure that contains all 12 products of theorizing within three roles: theory frames, theory generators, and theory components. The roles are not mutually exclusive but are depicted to give researchers a sense of where each product plays its primary role in theorizing. For example, the research question frames the research and may suggest analogies that could be applied, which would in turn help generate new concepts for new theories. This is exactly what Darwin (1859) did so eloquently, moving from his question

24 

N. R. Hassan et al.

about co-adaptation to drawing an analogy with selective breeding to inventing the concept of natural selection.5 While theories have been subjected to some form of verification or testing, the intermediate products have not necessarily undergone such a process. The intermediate products are pre-theoretical in the sense that they are typically created or emerge prior to verification or testing, within the context of discovery (Hassan et al., 2019). Models may, for example, serve as informal conceptions that operate as analogies or commentaries for a theory. Thus, models are independent of theories but can contribute to a theory (Kendler & Kendler, 1962; Lachman, 1960) and, just like concepts and statements, can become parts of the theory (Suppe, 1977; Torgerson, 1958). These pre-theoretical, conceptual artifacts inform the practice and imagination of academics and practitioners alike and have major implications for research and theory. As such, it is crucial to critically scrutinize them as expressions of interim struggles by researchers in theorizing. For example, when Burton-Jones and Straub (2006) scrutinized the pre-theoretical IS concept of use, they found that it had no accepted definition and that it had been operationalized in diverse ways according to the different nomological contexts to which it had been applied. Within these discursive practices, researchers invent, derive, examine, and apply various pre-theoretical products as scaffolding to continue the theorizing process. As these pre-theoretical structures are refined, they are eventually formalized into different types of theories (e.g., Gregor, 2006). In these discursive practices, there are no set discrete stages or linear paths, nor are theories ever finalized or “complete,” because taking that position would only limit the researcher’s creative thinking. Figure 2.1 depicts the process of how intermediate products may become inputs to each other, and Table  2.1 offers a brief summary of the 12 products. Information about these products can be found scattered across disciplines outside the IS field, which makes it difficult for IS researchers to build a clear mental  As illustration, Darwin, C.—(Darwin, 1859). On the Origin of Species. John Murray—asks what explains the “coadaptation of organic beings to each other and to their physical conditions of life” (p. 4) such that everything fits perfectly? This question, which reframed the discipline of biology, led Darwin to draw an analogy between the practice of selective breeding (artificial selection) that resulted in the change of the animal’s characteristics with the natural phenomenon of slow successive modifications. This analogy generated the concept of natural selection, which became a key component of the theory of evolution. 5

2  Useful Products in Information Systems Theorizing… 

25

Table 2.1  Summary of products of theorizing Product

Definition

Notes

Disciplinary questions A disciplinary question addresses an object of study as distinguish one discipline from a problem requiring a solution another and frame the theories of that discipline. based on the field’s rules of Asking questions includes discourse and pattern of problematizing the inquiry. phenomenon, which explicates and revises underlying theoretical assumptions (Gkeredakis & Constantinides, 2019). Largely maligned in the history Paradigm A paradigm is a shared exemplar for scientific practice of the IS field, paradigms have given birth to many new in the form of concrete disciplines. The social manifestations of analogies construction of technology is and metaphors on which an example of how paradigms communities of scientists and function as research heuristics researchers agree in part or to frame theory (Bijker, 1995; whole (Kuhn, 1977). Bijker et al., 1987). As part of theoretical reasoning, Law A law is a generalization that laws form the components of applies across space and time any theory and help define the and provides a framework for rules by which theories explain events that we use to plot and predict any phenomenon phenomena that may need (Schaller, 1997). explanation. It serves as the starting point from which we survey events looking for anomalies however they may be construed (Scriven, 1962). Framework A framework is the researcher’s A framework is broader and more inclusive than models or map of the territory being theories and can include both. studied, starting with its It is the total set of relations context and assumptions, and that unite the discursive consists of the main concepts, practices that give rise to constructs, variables, and their epistemological elements and related propositions. It can formalized systems (Gorry & take the form of a diagram or Scott Morton, 1971; Ives et al., a narrative; it may be “simple 1980; Mason & Mitroff, 1973). or elaborate, commonsensical or theory driven, descriptive or causal” (Miles & Huberman, 1994, p. 18). Question

(continued)

26 

N. R. Hassan et al.

Table 2.1 (continued) Product

Definition

Myth

A myth is “a dramatic narrative Myths have long been the source of creative theorizing in of imagined events … used to IS as researchers uncover explain origins or unquestioned assumptions transformations of something underlying IS phenomena, as … an unquestioned belief well as the means of studying about the practical benefits of qualitative elements of the certain techniques and phenomenon (Boland, 1982, behaviors that is not 1987; Boland & Pondy, 1983; supported by demonstrated Dearden, 1966; Hirschheim & facts” (Trice & Beyer, 1984, Newman, 1991). p. 655). An analogy, from Latin analogia Similitudes and resemblance played the most constructive for ratio or proportion, is a role as theory generators in rational argument using a the development of simplified, scaled-down knowledge in the history of reference to something mankind as well as in the IS familiar to explain or illustrate field (Angst et al., 2010; Keil, something more complex or 1995; Kuechler & Vaishnavi, less familiar (Bagnall, 2012; 2012; Sabherwal et al., 2001). Hesse, 1966). A metaphor “consists in giving Metaphors are the physical forms of entire networks of the things a name that analogies harnessed to clarify, belongs to something else” enrich, and enlighten and have (McKeon, 1941, p. 1476). historically been an active theorizing activity in the IS field (Kendall & Kendall, 1993; Mason, 1991) A model is an imperfect copy of Models and frameworks are often confused, as are models the phenomenon of interest and theories. Models are consisting of positive and abstractions and simplifications neutral analogies (Hesse, whereas frameworks are 1966). elaborations and networks of relations. That is why frameworks are composed of models (Davis, 1989; Delone & McLean, 1992).

Analogy

Metaphor

Model

Notes

(continued)

2  Useful Products in Information Systems Theorizing… 

27

Table 2.1 (continued) Product

Definition

Notes

In the IS field, concepts are A concept is a set of ideas rarely discussed in contrast to associated with the subject constructs, although the matter or elicited by a given generation of new concepts, word treated according to which is evidence of progress logical rules (Sartori, 1975). A in the field, is more conception is a concept that is fundamental than constructs, taken in a particular way. which are measures derived from concepts (Markus & Saunders, 2007). Whereas concepts are sets of Construct A construct is a term for a ideas from observables that concept that is neither directly allow for classification and nor indirectly observable and can be defined only in relation follow certain logical rules, constructs (or codes in to those observables, whereas interpretive research) are, in a variable is an observational essence, fictional entities term that changes for a invented to further research construct (Kaplan, 1964). (Furneaux & Wade, 2009; MacCorquodale & Meehl, 1948). The statement is the most Statement A statement is a mode of fundamental unit of any existence proper to a group of discourse and therefore any signs that describes a definite discipline, since it is the position for any subject (Foucault, 1972). A proposition statements (or claims) made by disciplines that justify their is the meaning of a logical existence. For example, causal declaration that bears truth value (Fawcett, 1998; Foucault, statements are central to most if not all theory (Markus & 1972). Rowe, 2018). In the IS field, hypotheses are Hypothesis A hypothesis is an commonly associated with operationalized proposition positivist research (Chen & that takes the form of Hirschheim, 2004), whereas empirically testable qualitative and grounded conjectures or follows a theorists argue that procedural rule to infer other hypotheses are natural propositions (Kaplan, 1964). components of their approaches (Flyvbjerg, 2006; Glaser & Strauss, 1967; Silverman, 2006). Concept

28 

N. R. Hassan et al.

model they can use to theorize in their research. Moreover, writings on these products often contradict each other and cause confusion among researchers. For example, few researchers can describe how a model differs from a framework. Hypotheses are often formed as if they were propositions and vice versa, and most IS authors use concepts and constructs synonymously. Because researchers are not on the same page regarding these products, it is not surprising that progress on theory development in the IS field has been slow. The last column in Table 2.1 contains notes with salient discussions within the IS field and other fields related to the products. Below, we expand on each product of theorizing, explaining why the product is foundational, and then describe specific aspects that require the attention of IS researchers.

Intermediate Products of Theorizing Question Asking questions is a major part of research and theorizing, and every field has its own peculiar set of questions. Following from the field’s discursive formation, evidence from linguistic and philosophical studies (Bal, 2002; Bromberger, 1992; Meyer, 1995) suggests that a discipline is defined by the set of questions it asks. Thus, not all research questions can be admitted into a discipline; the questions need to pertain to the discipline and become disciplinary questions. A disciplinary question addresses an object of study as a problem requiring a solution based on the field’s rules of discourse. An elegant example is how Durkheim (1951/1897, p.  324) posed the problem of suicide by asking the question: “Why in every society, a definite proportion of people commit suicide in any given period?” In doing so, he was not focusing on the state of mind (e.g., despair, neurosis, depression, or any psychological state), as one would expect in the case of suicide as framed from medical or psychological discourses; rather,

2  Useful Products in Information Systems Theorizing… 

29

he was linking suicide to his newly emerging discipline of sociology.6 As Foucault explains (1970, 1972), questions are the consequences of disciplines facing points of diffraction and contradictions in their findings that demand explanations and trigger programs of investigation that sometimes lead to the discovery of new disciplines. What has not been discussed in IS research circles is the nature of those research questions. Thus, within IS research, if the question being asked is an economic question, then economic principles, methods, tools, and resources can be expected to address it and the research will likely adhere to the rules of economic discourse rather than IS discourse. Conversely, a research question that concerns information, systems, or technology that the economic discipline itself does not have the principles, concepts, methods, or tools to address would open an opportunity for the IS field to create them and thereby contribute to the economics discourse. Historically, it would not be an understatement to say that the IS field emerged from this discursive practice of addressing questions that its reference disciplines had not satisfactorily addressed.7 Asking the right ­questions that interrogate and challenge the assumptions underlying existing literature is likely to produce interesting results (Alvesson & Sandberg, 2013; Davis, 1971; Slife & Williams, 1995). Asking the wrong questions will, at the very least, waste valuable research resources or lead research programs in a less productive or unintended direction. Theorizing within the IS discursive formation requires asking: “What makes the research question IS-specific?” By addressing this question, the  The questions that he was asking distinguished his unique discourse from that of medicine or psychology and framed his theories within sociology. Among the many novel concepts that Durkheim, É. (Durkheim, 1951/1897)—On Suicide: A Study in Sociology. Free Press—generated for sociology were the new concept of social cohesion along with sociological concepts of suicide, including altruistic, anomic, fatalistic, and egoistic forms of suicide. 7  For instance, Mason’s—Mason, R. O., & Mitroff, I. I. (Mason & Mitroff, 1973). A Program for Research on Management Information Systems. Management Science, 19(5), 475–487—early framework for IS began with answering the questions: “What is ‘knowledge,’ ‘effectiveness,’ ‘action;’ and further, who defines them and for what ‘purpose?’” (p. 475). Answering these questions created a framework connecting psychological types, problem types, and presentation modes. These questions did not fit exclusively into management, CS, or psychology alone. Similarly, Davis’s—Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 13(3), 318–340—TAM asks: “What qualities of systems increases its acceptance and the intensity of its use?”, a question seldom addressed in CS after a system is delivered. 6

30 

N. R. Hassan et al.

researcher is establishing what frames the research as IS research. Not asking IS questions exacerbates the field’s identity issues (Benbasat & Zmud, 2003) and prevents the field from demonstrating its value (Agarwal & Lucas Jr., 2005; Hassan, 2014b). The value of disciplines lies in their uncanny ability to ask questions about their areas of expertise and address questions whose answers we know that we do not know. Returning to creating a unique identity for the IS field and generating value in IS research, this means that effective theorizing in the IS field involves asking questions that are not being asked by other disciplines or asking questions that other disciplines are incapable of answering. The answers to these questions, and the IS theories emerging from them, define the IS field and its value to other fields.

Paradigm Partially as a result of criticisms of Kuhn’s (1970) paradigm concept and its varied interpretations (Banville & Landry, 1989; Popper, 1970), the role of the paradigm in theorizing has been largely neglected and misunderstood in the IS field (Hassan, 2014a; Hassan & Mingers, 2018). Although there are several notable exceptions (Chen & Hirschheim, 2004; Goles & Hirschheim, 2000; Iivari et al., 1998; Khazanchi & Munkvold, 2003; Mingers, 2004; Moody et  al., 2010; Richardson & Robinson, 2007), the IS field has abstained from actively debating about paradigms, at least in the concrete forms that Kuhn envisioned. Early attempts to theorize in IS by using paradigms were met with resistance due to, for example, the “disrepute into which this word has fallen” (Ein-­Dor & Segev, 1981, p. vii), and for much of the history of the IS field, the Kuhnian paradigm was made out to be a dubious undertaking (Adam & Fitzgerald, 2000; Banville & Landry, 1989). Some IS researchers have associated paradigms with the natural sciences and monism, stating that: The concept of paradigm, as Kuhn defines it, is derived from research in the physical sciences. This perspective may not serve well in the social sciences, where pluralistic models are more appropriate as the basis for understanding and analysis. (Larsen & Levine, 2008 p. 25)

2  Useful Products in Information Systems Theorizing… 

31

This trend of conflating paradigms with epistemology originated in the organizational sciences and the education field as part of their efforts to escape the hegemony of positivism (Hassan, 2014a; Hassan & Mingers, 2018). As a result, incommensurable research approaches developed into “paradigm wars” between positivists and interpretivists and realists and rationalists (Mingers, 2004), hampering research. Within these wars, paradigms were viewed as static, immovable structures that were to be defended at all costs, instead of as dynamically changing heuristics and frames for theorizing. Paradigms, as Kuhn (1977) describes in his response to critics, are really exemplars or concrete solutions to particular problems that can serve as the frame for solving other problems. This theoretical plight is unfortunate. Paradigms have been applied successfully in many fields, not just in the physical sciences.8 One notable example, the social construction of technology, which is often cited by IS researchers, is based on the Kuhnian paradigm. Explaining the basis of his concept of the “technological frame,” Bijker (1995) notes, “the analogy with Kuhn’s ‘paradigm,’ among other concepts, is obvious” (p. 123); he goes on to claim that the “technological frame is evidently one of the many children of Kuhn’s (1970) disciplinary matrix” (p.  126). Abbott  Minsky, M.—(1975). A Framework for Representing Knowledge. In J. Haugeland (Ed.), Mind Design II (pp. 111–142). MIT Press—a pioneer of artificial intelligence, acknowledges Kuhn, T.— (1970). The Structure of Scientific Revolutions (2nd ed.). University of Chicago Press—as inspiration for his frame theory: “the basic frame idea itself is not particularly original—it is in the tradition of the ‘schema’ of Bartlett and the ‘paradigms’ of Kuhn” (p. 113). Likewise, in the social sciences, Berger, P. L., & Luckmann, T.—(1966). The Social Construction of Reality. Anchor Books.—credit Kuhn, T. S.—(1957). The Copernican Revolution: Planetary Astronomy in the Development of Western Thought. Harvard University Press—for their understanding of the social construction of reality, and Ritzer’s Ritzer, G.—(1980). Sociology: A Multiple Paradigm Science. Allyn and Bacon, Inc. Sociology: A Multiple Paradigm Science—was based on the Kuhnian paradigm. The influence of Kuhn’s paradigms is particularly evident in science and technology studies, in which Kuhnian concepts of normal science, worldviews, and scientific revolutions forever changed the understanding of progress in science and technology. Other concepts influenced by the Kuhnian paradigm include but are not limited to: Collins and Pinch’s—Collins, H. M., & Pinch, T. J. (1982). Frames of Meaning: The Social Construction of Extraordinary Science. Routledge and Kegan Paul—frame of meaning; Constant’s Constant, E.  W.—(1980). The Origins of the Turbojet Revolution. Johns Hopkins University—technological tradition; Rosenberg’s—Rosenberg, N. (1976). Perspectives on Technology. Cambridge University Press—focusing devices; Gutting’s—Gutting, G. (Ed.). (1980). Paradigms and Revolutions: Applications and Appraisals of Thomas Kuhn’s Philosophy of Science. University of Notre Dame Press—technological paradigm; and Jenkins’s—Jenkins, R.  V. (1975). Images and Enterprise: Technology and the American Photographic Industry, 1839 to 1925. Johns Hopkins University Press—technological mind-set. 8

32 

N. R. Hassan et al.

(2001, 2004) supports this view, arguing that unified sets of premises, such as Kuhnian paradigms, can function as research heuristics and therefore help frame theory. Theory can remain either stagnant by staying within an isomorphic enclosure that blinds its adherents or progress from new paradigms that emerge from the discovery of contradictions and anomalies too glaring to be ignored (Foucault, 1970, 1972). The agnostic nature of the paradigm allows different quantitative and qualitative elements to work together to enhance creative theorizing. The generative metaphor (Schön, 1979) is an example of the use of paradigms to inspire perceptions of new features in the interest of generating novel views of problems.9

Law Law-like statements have always been part of inductive and deductive reasoning, going back to at least Francis Bacon.10 The hypothetico-­deductive method itself, which is the dominant approach in IS research (Chen & Hirschheim, 2004; Liu & Myers, 2011), requires an inquiry “into the causes as well as the laws of phenomena—that such an inquiry cannot be avoided; and that it has been the source of almost all the science we possess” (Whewell, 1840/1967, p. 322). As this process was refined, Hempel (1965) and colleagues added other methods, including deductive-nomological (covering-law explanation), deductive-statistical, and inductivestatistical explanations, which all revolve around universal and statistical laws. Given the centrality of laws as frames in theorizing, it is time for IS researchers to start including laws as part of their theorizing. Because a  Using the metaphor of the pump, Schön, D.  A.—(1983). The Reflective Practitioner: How Professionals Think in Action. Basic Books—describes how to generate new ideas for designing a paintbrush. Although the pump and the brush are two different products with two different delivery paradigms, they share developmental lines of thought in delivering paint such that the already familiar processes of one can be readily and creatively transferred to the other. 10  Francis Bacon once defined inductive reasoning as “nothing more than those laws and determinations of absolute actuality which govern and constitute any simple nature, as heat, light, weight, in every kind of matter and subject that is susceptible of them.” Spedding, J., Ellis, R. L., & Heath, D. D. (Eds.). (1901). The Works of Francis Bacon Vol IV. Houghton Mifflin. 9

2  Useful Products in Information Systems Theorizing… 

33

theory is essentially “a system of laws” and “theory explains the laws” (Kaplan, 1964, p. 297), laws explain facts and relate them to other facts. In many ways, laws frame theorizing processes but can, at the same time, be a part of theory. A law is established when the series of hypotheses is consistently tested to be true and is said to constitute fact when it is particular in content or a law when it is general. Scriven (1962) defines laws as generalizations that: provide a framework for events that may be used to plot phenomena in need of explanation and may serve as the starting point from which events may be surveyed, in search of nonconformists, not only as the rules under which we try to bring them.

Such a view of laws is seldom discussed within IS, partly because, to the field, even the possibility of social science laws is considered unlikely (Gregor, 2006). In a study of the top-two IS journals, Hovorka et  al. (2008) found no examples of studies that incorporated laws as part of explanatory methods. To say that laws are not amenable to the social sciences is counterproductive if not blatantly incorrect. For example, in economics, Say’s law (a powerful generalization that states that “products are paid for by products”) and the law of supply and demand (which determines exchanged quantities) work with other laws, such as the Walras’s law, to explain and constitute the quantity theory of money (Blaug, 1997). Laws are also not exclusive to deductive-nomological explanations or covering-law explanations. As Hempel (1965), Hempel and Oppenheim (1948), and Salmon (1984) explain, theorizing need not appeal only to universal laws, but it can rely on statistical laws or non-causal laws, or other means of explanation that do not require strict causality yet are still linked to law-like statements. Thus, the well-known Moore’s law in technology, which is not a causal law in the form of “A causes B,” not only explains the doubling of components on an integrated chip in 18 months, but it is so reliable that it has become part of predicting future trends, setting the pace of innovation and defining the rules and nature of competition related to digital innovations (Schaller, 1997). Instead of merely mentioning Moore’s law in

34 

N. R. Hassan et al.

passing in IS studies, it could become an essential part of policy making and implementation and of theories. Such efforts are already making strides, especially in developing countries (Brewer et  al., 2005). Taking Scriven’s (1962) definition of laws, there is nothing to stop IS researchers from, say, identifying non-conforming events (e.g., IS failures) or using established functional laws to describe rules of how systems are supposed to operate successfully without failures.

Framework The term framework—often called “conceptual framework,” “research framework,” or “theoretical framework”—is frequently conflated with the term model. Because frameworks act as an overall guide for and justify research, researchers need to critically examine existing frameworks related to the phenomenon being studied. Miles and Huberman (1994) describe the conceptual framework as the researcher’s map of the territory being studied, which consists of main concepts, constructs, and related statements. It can take the form of a diagram or a narrative; it may be “simple or elaborate, commonsensical or theory driven, descriptive or causal” (p. 18). Maxwell (2013, p. 39) considers the framework to be a theory and broadens its scope to include a “system of concepts, assumptions, expectations, beliefs, and theories that support and inform” the research. Ravitch and Riggan (2012, p. xiii) view the conceptual framework as a mechanism that resolves “why the topic one wishes to study matters, and why the means proposed to study it are appropriate and rigorous.” These definitions of the framework are captured by Foucault’s (1972, p. 191) notion of episteme, which he describes as “the total set of relations that unite, at a given period, the discursive practices that give rise to epistemological figures, sciences and possibly formalized systems.” In this sense, the framework is both a guide providing reasoned, defensible choices and a source of stability for theorizing and research. It is not surprising that frameworks have become some of the earliest products of theorizing used by IS researchers (Gorry & Scott Morton, 1971; Ives et  al., 1980; Mason & Mitroff, 1973) to build research programs that have continued for decades.

2  Useful Products in Information Systems Theorizing… 

35

Unfortunately, what often takes place within the IS field is the creation of so-called new research frameworks that borrow pieces of existing frameworks and add one or two additional components to create a veneer of novelty. This strategy is flawed because existing frameworks contain the key theories, concepts, and epistemological and ontological grounds for the phenomenon being studied. Arbitrarily adding components to a framework risks ignoring the structures of the existing framework (Hart, 1998) that contain underlying assumptions and perspectives, including answers to questions like: What is the theory? Whose theory is it and where does it come from? What are those intellectual traditions? What is the history of the development of intellectual traditions? What are the main arguments? Ignoring the history of the framework and its hidden assumptions jeopardizes research, especially if the assumptions and traditions of the borrowed framework contradict the current research. Additionally, when different perspectives are combined into the same framework, the researcher must explain how they fit into the same framework. For example, problems may arise when adoption research in IS (Venkatesh et al., 2003) combines concepts from social psychology, such as attitude (Fishbein & Ajzen, 1975), which is not temporally bounded, with concepts from communication studies, such as relative advantage, which assumes a change in perception over time (Rogers, 1983). Critically examining frameworks also means interrogating existing frameworks to uncover any weaknesses that might open opportunities for a myriad of theorizing strategies. The point of working with frameworks is not to borrow existing ones but to create new frameworks for the research. The new framework integrates all the products of theorizing and helps researchers assess and refine goals, develop questions, select appropriate methods, and identify potential validity threats. If there are existing theories, then the new framework provides a place for them with respect to the research. If there are no existing theories, then the framework becomes a nascent version of one. The framework helps design the research inductively, deductively, or using any other approach, and also assists in the ensuing data collection and data analysis (Ravitch & Riggan, 2012).

36 

N. R. Hassan et al.

Myth It is often difficult to comprehend how myths can contribute to theory building. It is easier to imagine such a process taking place when theory emerges from breaking down a myth. A classic case of such a process is the Copernican Revolution (Kuhn, 1957), which overthrew the myth of earth being the center of the universe. A myth is “a dramatic narrative of imagined events, usually used to explain origins or transformations of something … an unquestioned belief about the practical benefits of certain techniques and behaviors that is not supported by demonstrated facts” (Trice & Beyer, 1984, p. 655). Myths are theory generators because, although they are frequently used to refer to mistaken beliefs or popular misconceptions, they can help uncover unquestioned assumptions within existing belief systems and theories. Lévi-Strauss (1963, 1966) viewed myths as precursors to research, especially in theories of relations, whereas Cassirer developed a theory of symbolic forms inspired by his study of myths (Cassirer & Verene, 1979). Myths become useful inputs for theories when researchers apply them counterinductively. Because myths do not separate history from research, any myth, however ancient or absurd, has the potential to build, enrich, and even revise theories and knowledge. This was the case with the Pythagorean metaphysical beliefs of the earth’s movements and the development of traditional Chinese medicine (Feyerabend, 1978). As theory generators, myths perform multiple functions (Hirschheim & Newman, 1991; Mousavidin & Goel, 2007). They provide means of explanation; the language for studying symbols of value, solidarity, and social structure; and ways to manage conflict and contradictions (Cohen, 1969). Theorizing using nonrational myths has identified many factors with equal or greater influence on the effectiveness of system development strategies (Franz & Robey, 1984; Hirschheim & Newman, 1991). Early works by Boland (1982, 1987), Boland and Pondy (1983), Hirschheim and Newman (1991), and Robey and Markus (1984) are particularly impressive regarding the leveraging of myths, and we lament this lost art. For example, Hirschheim and Newman (1991) identified six common myths in IS development that have become common IS knowledge, such as the “overriding advantage of user involvement,” “the need to ameliorate

2  Useful Products in Information Systems Theorizing… 

37

user resistance,” and “the necessity of system integration.”11 Myths are closely related to rites, which depend on myths for validation (Cohen, 1969), and research on both myths and rituals can be found in several classic IS studies. For instance, Robey and Markus (1984) describe how system developers use rituals to maintain the appearance of rationality while working to achieve private interests—upholding the myth of rational decision-making in systems development.

Analogy Although research is a concrete activity requiring resources, methods, and tools, it also requires forms of abstract reasoning, including problem solving, analysis, and theorizing, all of which are supported by analogies. It is thus not surprising that analogies and similar structures, such as similitudes and resemblance, played the most constructive role as theory generators in the development of Western thought up to the Age of Enlightenment (Foucault, 1970, 1972). Analogy—from the Latin analogia for ratio or proportion—is a rational argument using a simplified, scaled-down reference to something familiar to explain or illustrate  Boland, R. J. (1982). Myth and technology in the American accounting profession. Journal of Management Studies, 19(1), 109–127, and Boland, R. J., & Pondy, L. R. (1983). Accounting in organizations: A union of natural and rational perspectives. Accounting, Organizations and Society, 8(2–3), 223–234, introduced the notion of rational and nonrational myths, highlighting the need for research that includes both types to understand the interaction of organizations and technology. Boland, R. J. (1987). The in-formation of information systems. In R. J. Boland & R. A. Hirschheim (Eds.), Critical Issues in Information Systems Research (pp. 363–379). John Wiley & Sons, for example, asserts that the “rational system” myth is noteworthy because users expect systems to meet developers’ costs and efficiency demands while simultaneously accomplishing mythical goals. Theorizing using nonrational myths identifies many factors with equal or greater influence on the effectiveness of system development strategies: Franz, C. R., & Robey, D. (1984). An investigation of user-led system design: rational and political perspectives. Communications of the ACM, 27(12), 1202–1209; Hirschheim, R.  A., & Newman, M. (1991). Symbolism and information systems development: myth, metaphor and magic. Information Systems Research, 2(1), 29–62. For example, early critics of management information systems (MIS) invoked the “myth of real-time systems”— Dearden, J. (1966). Myth of real-time management information. Harvard Business Review, 44(3), 123–132—to expose several fallacies regarding the assumed capabilities of computers to support management functions. Boland, R.  J. (1987). The in-formation of information systems. In R.  J. Boland & R.  A. Hirschheim (Eds.), Critical Issues in Information Systems Research (pp.  363–379). John Wiley & Sons, they described five pervasive myths, which he pejoratively called “fantasies,” about information that he believed obstruct progress in IS research. 11

38 

N. R. Hassan et al.

something more complex or less familiar (Bagnall, 2012; Hesse, 1967). Analogies are not merely literary devices, as they supply the raw materials for theorizing and, if suitably handled, can yield theories (Tsoukas, 1993). Within the context of discovery, analogies allow for demonstrative inferences that are difficult or impossible to achieve in purely positivist schemes of explication and justification. Scientific modeling using analogies was a major feature of early modern science, as illustrated by many examples. Analogies of electron flow and those of wave and corpuscular theories of light were foundational to theoretical development in physics.12 Darwin drew an analogy between artificial selection (i.e., the breeding of domesticated animals) and natural selection to argue for the plausibility of the latter. Einstein’s thought experiments and discoveries can be considered elaborate analogies (Geary, 2009) that were later concretized into propositions and hypotheses. Campbell (1920) highlighted the critical role of analogy in theorizing as follows: The value of the theory is derived largely, not from the formal constitution, but from an analogy displayed by the hypothesis. This analogy is essential to and inseparable from the theory and is not merely an aid to its formulation.

In the management field, Beer (1972, 1979) drew an analogy between the human body and the enterprise, and theorized that only five major subsystems are required to coordinate and control any organization. Although IS studies often apply analogies implicitly, the field seldom turns to explicit analogical reasoning. A design science study (Kuechler & Vaishnavi, 2012) applying analogical reasoning to translate theoretical domains into design domains suggests that the IS field is realizing the importance of explicit analogical reasoning. Implicit analogical reasoning can be found in many IS studies, but the reasoning and theorizing processes behind the analogy are typically left unexplored. For example, when  In using analogies, researchers select key similarities between domains rather than features of individual objects. For example, physics researchers draw an analogy between the flow of electrons in an electrical circuit and the flow of people in a crowded subway. The analogy depicting the flow of electrons via the flow of people emphasizes the movement of the objects, not the size or shape of the people compared to electrons. Gentner, D. (1983). Structure-Mapping: A theoretical framework for analogy. Cognitive Science, 1, 155–170; Gentner, D. (1989). Mechanisms of analogical reasoning. In S. Vosniadou & A. Ortony (Eds.), Similarity and Analogical Reasoning (pp. 199–241). Cambridge University Press. 12

2  Useful Products in Information Systems Theorizing… 

39

Keil (1995), Keil and Robey (1999), and Keil et al. (2000) apply the term escalation to the context of software project management, they use an analogy originally applied in military scenarios (Kahn, 1965) that draws from similarities between intensifying conflict and climbing the rungs of a ladder. Similarly, when IS scholars study punctuated equilibrium or systemic change (Street & Denford, 2012) or describe system adoption in terms of contagion (Angst et  al., 2010), they are leveraging analogies from other disciplines, such as geology and biology. Yet, within the IS field, such analogies are rarely analyzed.

Metaphor Metaphors, the linguistic form of analogies, are products of theorizing that have been in use as theory generators at least since Aristotle’s time (Schön, 1963). Whereas analogies are abstractions of similarities, the metaphor selects tangible things that carry the meanings of those similarities. Notions like “my broken heart” or “a fishing expedition” help communicate abstract ideas, feelings, or even legal concepts; ergo, they represent powerful theory generators (Geary, 2009). Because knowledge is construed from some point of view, all knowledge is perspectival and thus metaphoric (Brown, 1976). Consequently, metaphors are not merely rhetorical devices, but essential products of and tools for theorizing. Metaphors are valuable at any stage of theorizing, including the preliminary stages of inductive and deductive theorizing and retroductive and abductive reasoning, as well as the later stages of extending existing theories. In his Poetics, Aristotle defines metaphor as a “carrying over” from one thing to another, with phor meaning “carrying” and meta meaning “beyond” (Kirby, 1997). Whereas an analogy finds similarities between two different things, a metaphor “consists in giving the things a name that belongs to something else” (McKeon, 1941, p.  1476). Aristotle asserts that crafting powerful metaphors depends on the ability to perceive likeness between things that are dissimilar or likeness that might not initially be obvious: The observation of likeness (homoiou theoria) is useful with a view both to inductive arguments and to hypothetical deductions, and also with a view to the production of definitions. (Aristotle, qtd. in Kirby, 1997, p. 536)

40 

N. R. Hassan et al.

Metaphors are valuable to theorizing for their ability not only to transfer meaning but also to impress, clarify, enrich, and enlighten (Ortony, 1979). The origin of the metaphor is usually elegant, beautiful, impressive, or respected in its own way and is often spoken spontaneously (Kirby, 1997). A metaphor harnesses an entire network of analogies to accomplish its task. For example, when early computer scientists used the metaphor of the brain to describe the computer’s central processing unit, they quickly transferred well-known functions of the brain to explain something unfamiliar at that time: computer processing. At the same time, this metaphor was intended to impress and enlighten the audience concerning computer technology, thereby serving as a source for the concept of “machine learning,” in turn inspiring awe for computers. Metaphors possess characteristics of good theorizing, such as originality, economy, consistency, elegance, and perspicuity. In IS research, metaphors are most often found in earlier studies and usually coupled with myths (Hirschheim & Newman, 1991; Kaarst-Brown & Robey, 1999).13  Early examples propose organismic, sports team, and city-state metaphors for IS strategic planning, offering alternatives to the war metaphor that dominated strategic thinking at the time. Mason, R.  M. (1991). Metaphors and strategic information systems planning. 24th Hawaii International Conference on System Sciences, Kauai, HI. Several IS articles explored the use of other metaphors to theorize about system development. Kendall, J. E., & Kendall, K. E.—(1993). Metaphors and methodologies: Living beyond the systems machine. MIS Quarterly, 17(2), 149–171, ibid.—emphasized the need for developers to understand the metaphors applied to system development to better communicate with users, whereas Oates, B.  J., & Fitzgerald, B.— (2007). Multi-metaphor method: organizational metaphors in information systems development. Information Systems Journal, 17(4), 421–449—later described how metaphors help developers theorize about organizations to tailor the methodology and process for specific IS development contexts. Some IS scholars have applied Schön’s—Schön, D. A. (1979). Generative Metaphor: A Perspective on Problem-Setting in Social Policy. In A.  Ortony (Ed.), Metaphor and Thought (pp. 254–283). Cambridge University Press—notion of a “generative metaphor” to the planning and development of systems to accommodate a multiplicity of interests and relationships. Atkinson, C. J. (2003). The Nature and Role of Generative Systemic Metaphor within Information Systems Planning and Development. In E. H. Wynn, E. A. Whitley, M. D. Myers, & J. I. DeGross (Eds.), Global and Organizational Discourse about Information Technology (Vol. 110, pp. 323–343). IFIP/ Springer. Using the metaphor of magic as it is applied to generally accepted rituals in IS development, Hirschheim, R. A., & Newman, M.—(1991). Symbolism and information systems development: myth, metaphor and magic. Information Systems Research, 2(1), 29–62—theorized about the social nature of IS development and how it affects a project’s probability of success. Brynjolfsson, E., Hofmann, P., & Jordan, J.—(2010). Cloud Computing and Electricity: Beyond the Utility Model. Communications of the ACM, 53(5), 32–34—applied the metaphor of electrical utilities to describe the types of services expected of cloud computing as a utility while also theorizing several dissimilarities between electrical utilities and cloud computing. 13

2  Useful Products in Information Systems Theorizing… 

41

Model Because no part of the complex universe can be understood without abstraction, simplified abstractions of the real world—or models—play the role of addressing the research question on a practical level of detail (Rosenblueth & Wiener, 1945).14 One of the earliest theorizing works on magnetism by William Gilbert (1893/1600) applied the model of the earth as a magnet to explain why compasses point north. Likewise, biologists and paleontologists have applied principles of modeling in their attempts to predict the role that biological structures in animals play in their habitat and to thus also predict (and speculate, in the case of, for example, dinosaurs) how they can be conserved (or why they went extinct).15 Models can take many forms, including mathematical, analogical, physical scale, computer, or conceptual, emphasizing a different aspect of the phenomenon of interest. One well-known mathematical model in the social sciences is the Black-Scholes option-pricing formula (Black & Scholes, 1973), which models the price of an asset following a log-­normal random  Using notions of positive analogies (i.e., common properties between two different objects), negative analogies (i.e., properties that differ between objects), and neutral analogies (i.e., uncertain as to whether positive or negative analogies exist), a model can be defined as an imperfect copy of the phenomenon of interest, consisting of positive and neutral analogies. Hesse, M. B. (1966). Models and Analogies in Science. University of Notre Dame Press. By analyzing the extent of positive, negative, and neutral analogies, researchers can draw out horizontal relations between model properties to the phenomenon of interest and speculate on vertical or causal relations stemming from those similarities. If both horizontal and vertical relations exist, Hesse would call those analogies material analogies, which enable predictions to be made from the model. 15  As Harré, R.—(1970). The Principles of Scientific Thinking. University of Chicago Press—explains, a model is no more than a putative analog for a real mechanism, modeled on things, materials, and processes that we already understand. Ibid. describes several types of models distinguished according to whether the subject of the model is also the source of the model. For instance, Weber’s— Weber, M. (1930). The Protestant Ethic and the Spirit of Capitalism (T. Parsons & R. H. Tawney, Trans.). G. Allen & Unwin, Ltd.—ideal types are models in which the subject of the model (e.g., the Protestant capitalist) is also the source of the model, just as a model airplane in a wind tunnel is constructed based on the original airplane. Harré terms these models homeomorphs, which can differ in terms of scale, purity, and level of detail. Models in which the subject is not the same as the model are termed paramorphs, which are used to model a process that is unknown or yet to be investigated. Economic models that demonstrate how the economy “expands” and “contracts” as a result of flows of activity are other examples of paramorphs. The subject of the model, the growth or shrinking of the economy, is not the same as its source, which is that of a balloon expanding or contracting. 14

42 

N. R. Hassan et al.

walk in just the right way to eliminate risks. An example of a conceptual model is the model for corporate social responsibility by Carroll (1979), which is based on the dimensions of categories of responsibility, social issues, and different ways of responding to those issues. As pre-­theoretical products, models can be created without any theoretical justification. A conceptual model is essentially a material analogy that is “wholly imaginary … not realized in any existing … physical system … modified and fitted ad hoc to the data” (Hesse, 1966, p. 67); nevertheless, because it may allude to known causal relations, it can be predictive, as Carroll’s seminal model from the management literature illustrates.16 In the IS field, models and frameworks are among the products of theorizing that are most mentioned, but they are also among the most ambiguous and problematic. IS researchers have difficulty distinguishing models from frameworks, and they are frequently conflated with “theories.” The Technology Acceptance Model (TAM) and the IS success model are cited as the two most applied IS theories (Moody et al., 2010; Straub, 2012), even though both are labeled and depicted as models. The first difference between them is depicted in Fig. 2.2, which shows frameworks as theory frames and models are theory generators. Second, because models are putative analogs for the phenomena of interest, they provide a means by which the researcher can theorize about those phenomena using things, materials, and processes that they already understand. Conversely, frameworks are detailed and elaborate maps that are related to the phenomena of interest and could include models. Hence, whereas models are images or representations of analogies, frameworks are detailed and elaborate maps that include models. Like frameworks, it is tempting to borrow models or cherry-pick elements of a theory or theories and integrate them into a new “IS theory”; to do so is relatively easy and often initially yields good results. For  Carroll’s—Carroll, A.  B. (1979). A Three-Dimensional Conceptual Model of Corporate Performance. Academy of Management Review, 4(4), 497–505—conceptual model theorizes the question of what social responsibility means for a corporation by building on three dimensions: (1) categories of social responsibility (i.e., ethical, legal, economic); (2) types of social issues that must be addressed (i.e., environmental, product safety, discrimination); and (3) the philosophy of the response (i.e., reactive, defensive, accommodative). Contrary to the typical theoretical demands of top IS journals, ibid. offers no theories to serve as the basis for this model of corporate social responsibility. Yet, it is a seminal work (with nearly 15,000 citations at the time of writing this article). 16

2  Useful Products in Information Systems Theorizing… 

43

Fig. 2.2  Application of products of theorizing in media richness theory (MRT)

example, two of the most popular theories in the social sciences, the diffusion of innovations theory (DIT) (Rogers, 1983), and the theory of reasoned action (TRA) (Fishbein & Ajzen, 1977) are also among the two most applied theories in assessing the influence of information technology (IT) on individuals (Lim et  al., 2009). However, these theories clash because they apply different models and seek incommensurate goals.17  These theories describe two different models of innovation. Diffusion of innovations theory (DIT) originates in the communication field and models innovation in terms of the flow of information. Consequently, flow-related analogies, such as channels that carry information, the time taken for the rate of adoption, and the social system engaging in the flow, provide a rich set of concepts and constructs to be researched. The theory of reasoned action (TRA) is a theory of behavior predicated on an individual’s behavioral intention, which in turn is affected by the individual’s attitude. Comparing DIT to TRA, because DIT includes a time element, it is able to describe the logistics curve of innovation, which is not possible when using TRA. Conversely, TRA’s focus on attitude is only tangentially addressed by DIT. 17

44 

N. R. Hassan et al.

Concept As shown in Fig. 2.1, the process of building a unique discourse revolves around working with theory frames and theory generators to give birth simultaneously and successively to theory components. Foucault (1970, 1972) explains how the rules of formation of the discourse form the basis of the regularity within which these theory components relate to or disperse from each other. The generation of new concepts in the field becomes evidence for the progress of that field of study. As Nobel laureate Sir George Thomson (1961) notes: Science depends on its concepts. These are the ideas that receive names. They determine the questions one asks, and the answer one gets. They are more fundamental than the theories which are stated in terms of them.

A concept is a set of ideas associated with or elicited by a given word, treated according to logical rules (Sartori, 1975).18 Such rules imply that concepts are discipline-specific and demarcate a field’s subject matter, as the field is made known to the world through those concepts. For example, no one doubts that respiration and circulation are biological concepts, as relativity and quantization are concepts belonging to physics. The question is: What concepts belong to the IS field? Concepts are not limited to physical characteristics and are particularly amenable to behavioral and social research, as argued by Dilthey (1883/1989).19 Unfortunately, the social sciences, the IS field included,  Sartori, G. (Ed.)—(1984). Social Science Concepts: A Systematic Analysis. Sage Publications—considers concepts as the basic unit of thinking in the same way that Dubin, R.—(1969). Building Theory. The Free Press—refers to concepts as “units” of theory. As Satori explains, “it can be said that we have a concept of A (or of A-ness) when we are able to distinguish A from whatever is not-­ A” (p.  74). Concepts are always associated with observable objects of study and are discipline-­ specific because they are superimposed on our experiences as a way for us to understand the world. Several concepts can be combined to form a gestalt that engenders certain expectations. 19  Providing an alternative to the positivistic approach of the natural sciences, Dilthey, W.— (1883/1989). Introduction to the Human Sciences. Princeton University Publishers—argues that positivist representational facts fail to capture the human experience and that “no real blood flows in the veins of the knowing subject constructed by Locke, Hume and Kant” (p. 50). He proposes that an emphatic understanding of human behavior (verstehen) is necessary to capture the “knowledge of the forces that rule society, of the causes that have produced its upheavals, and of society’s resources for promoting healthy progress [that] has become of vital concern to our civilization” (p.  56). This emphatic understanding opened the doors to a new category of disciplines of the human sciences. 18

2  Useful Products in Information Systems Theorizing… 

45

face challenges in clarifying their theoretical concepts. Blumer (1954), one of the earliest exponents of interpretivism, blames the vagueness of social science concepts for the counterproductive practices of reproducing abstruse research, borrowing heavily from the natural sciences and forcing our idiosyncratic categories onto a world far removed from our research. In IS, concepts are especially relevant to theorizing because the field lacks concepts of its own (Markus & Saunders, 2007). When concepts are illdefined, tautological (e.g., “performance is the perceived effect of the manager’s job performance”), or defined in conflict with everyday use or accepted research, they obscure rather than illuminate empirical reality (Czarniawska, 2013). Thus, whether IS concepts are invented or adapted, there is a dire need for conceptual development in the IS field. Although the search for new, unique concepts has preoccupied scholars, philosophers, and scientists since before the Age of Enlightenment, the means by which new concepts are created has remained a mystery. Schön (1963) suggests that the production of new concepts is closely related to understanding how to work with metaphors and analogies.20 Unfortunately, the IS field seldom explores the concepts it applies in research (Markus & Saunders, 2007), hindering the development of many of our core concerns (Orlikowski & Iacono, 2001; Weber, 2003; Zhang et al., 2011).

 He notes that “the new concept grows out of the making, elaboration, and correction of the metaphor” (p. 53). He calls this process the displacement of concepts, in which words undergo transposition (i.e., applying an old concept to a new situation), interpretation (i.e., assigning that concept to a specific aspect of the new situation), correction (i.e., an adjustment resulting from adaptation and modification), and spelling out (i.e., resolving commonalities and differences) as a way of addressing problems or improving understanding. Another way of creating concepts is by inductively deriving them from data using methods such as grounded theory. The process of coding in grounded theory is itself the process of conceptualizing data. Strauss, A., & Corbin, J. (1990). Basics of Qualitative Research: Grounded Theory Procedures and Techniques. Sage Publications. Philosophers like Foucault, M.—(1972). The Archaeology of Knowledge and the Discourse on Language (A. M. S. Smith, Trans.). Pantheon Books.—suggest creating new concepts by first observing the context from which the objects of study emerge, what kind of authorities delineate and acknowledge their existence, and how the objects of study can be classified and organized. Depending on these factors, concepts will exhibit different forms of ordering and demonstrate various justifications for their validity and ability to transfer their meaning to different domains. 20

46 

N. R. Hassan et al.

Construct Virtually, all research traditions distinguish between two groups of research terms: observables and nonobservables (Kaplan, 1964). Observables are either directly or indirectly observable. They form the empirical part of the research, circumscribe the locus of the problem, and help marshal the data. Nonobservables include constructs that are invented to provide solutions and measures to further research. Whereas concepts are sets of ideas from observables that allow for classification and follow certain logical rules, constructs (or codes in interpretive research) are, in essence, fictional entities (MacCorquodale & Meehl, 1948) that are abstracted from the concept, contrived to enable the use of some form of measurement or evaluation or to bridge several concepts in the study. In the IS field, concepts, constructs, and variables are often conflated and rarely distinguished, making comparison between research studies awkward.21 Additionally, constructs should follow concepts since concepts are derived from observations; however, in the IS field, concepts rarely take preference over constructs. This confusion leads to further confusion and the misspecification of reflective constructs and formative constructs (Petter et al., 2007).22  A variable is a term that varies for concepts whose applications rely on direct or indirect (inferred) observation. In situations where the concept cannot be observed directly or even inferred, it is called a construct, which is a concept that is neither directly nor indirectly observable and can be defined only in relation to observables. Kaplan, A.—(1964). The Conduct of Inquiry: Methodology for Behavioral Science. Chandler Pub. Co.—added that when the construct is hypothetical and its existence is dependent on the theory that creates it, it becomes a theoretical term. Keen, P. G. W.— (1980). MIS research: reference disciplines and a cumulative tradition. International Conference on Information Systems (ICIS 1980), Philadelphia, PA—was correct to criticize the IS field for not agreeing on a dependent variable; unfortunately, his analysis of the field’s use of constructs and indirect observables was lost in the confusion. Keen proposed that the IS field should abandon using observables and constructs such as usage and user satisfaction because they have little theoretical significance to the core concern of the field: information. For Keen, the IS field needed to agree on a definition of information before a theoretically sound and practice-relevant dependent variable could be established. Indeed, in the positivist vein, how could the usage or usefulness of information be measured when information itself had yet to be defined? Yet decades of research in IS are dedicated to such a pursuit. 22  These complex abstractions combine multiple concepts belonging to the field, making it difficult to unpack their actual content. Dubin, R.—(1969). Building Theory. The Free Press.—called these formative constructs or abstractions summative units, which is similar to Kaplan’s—Kaplan, A. (1964). The Conduct of Inquiry: Methodology for Behavioral Science. Chandler Pub. Co.—notion of collective terms or composite variables. 21

2  Useful Products in Information Systems Theorizing… 

47

In the positivist tradition, constructs help to make vague concepts such as intelligence more tangible and amenable to measurement, allowing the research to progress (Lewis et al., 2005). In layman’s language, constructs are attempts to make concepts less “abstract” even though constructs are abstractions of concepts. They help researchers make sense of hypothetical entities, act as heuristic devices, and form what is known as the nomological network, an “interlocking system of laws which constitute a theory” (Cronbach & Meehl, 1955, p. 290). Although the term nomological network explicitly includes laws, it is commonly understood in the IS field as an interlocking system of concepts and constructs typically represented by box-arrow diagrams. In nonpositivist research, constructs are typically applied differently than in positivist research. When interpretive researchers conceptualize data, they are not inferring the existence of certain entities or postulating people’s attributes; instead, they are creating “constructions of other people’s constructions of what they and their compatriots are up to” (Geertz, 1973, p. 9 in Walsham (2006)).23

Statement Once a field has defined its own concepts and constructs, it can start formulating the most fundamental unit of its discourse: statements or claims. A statement is a mode of existence proper to a group of signs that describes a definite position for any subject (Foucault, 1972). This cryptic Foucauldian definition of a statement contains many layers of meaning with key implications for theorizing. First, following discourse analysis, a  Concepts such as themes, meanings, and essences of human experiences are gathered using various means, such as (1) close involvement with the participants in the field, observing, listening, interviewing, and reflecting (e.g., case research, ethnography, grounded theory); (2) coming to an understanding of or interpreting texts and social action (i.e., hermeneutics); and (3) describing human experience (i.e., phenomenology). In the nonpositivist tradition, observations and interviews are primary research methods for accessing experiences, which are typically documented in textual, visual, or other formats like field notes, transcriptions, memos, narratives, or recordings. After some form of validation, these experiences undergo an interpretive process by the researcher that transforms them into abstract concepts that are indirectly observable or nonobservable. Analogous to positivist research, the interpretation of the researcher becomes the construction. Flick, U. (1998). An Introduction to Qualitative Research. SAGE Publications; Miles, M. B., Huberman, A. M., & Saldana, J. (2014). Qualitative Data Analysis: A Methods Sourcebook. Sage Publications; Silverman, D. (2006). Interpreting Qualitative Data. SAGE Publications. 23

48 

N. R. Hassan et al.

statement is not just any sentence, it is a mode of existence within a specific discourse that enables groups of signs to exist in a distinct way. Expressly, a statement is a claim that is subject to its discursive formation. For example, the statement “time is golden,” which contains the signs “time” and “gold,” when taken out of its discursive formation, bears little relation to the physical makeup of time or to its chemistry despite its use of the chemical element “gold,” but does makes sense within the discursive formation of English literature.24 Second, statements are made of signs grouped in a special manner. When a field makes a statement, it is not merely formulating a sentence, which is a series of linguistic signs following a grammatical rule (Foucault, 1972), nor is a statement the same as a proposition, which is the meaning of a logical declaration that bears truth value (Fawcett, 1998; Foucault, 1972).25 Rather, it is expressing what Foucault (1972) calls an enunciative function, which invokes the authority of the discipline that it is associated with. As a result, a researcher who views two sets of statements can clearly distinguish an IS statement from those of other disciplines. For example, it can be argued that the statement “user participation enhances the quality of a system” belongs to the IS field because its related concepts (participation and system quality) are IS-specific and have been theorized as such. The statement derives its authority from system developers, and its context and field of emergence is the system development process. Within the field-specific context, we can ask such questions as: “When is user participation best suited?” and “What form should user  Similarly, statements such as the “earth is round” and “organisms evolve” do not constitute the same statement before and after Copernicus (for the former), or before and after Darwin (for the latter), because those statements depend on the concepts, theories, and discursive formations of these scientists’ respective disciplines and thus exist in different modes in different times. Namely, these statements are closely related to the theories that they represent. 25  Several statements together can express a single proposition, and a single statement can give rise to different propositions. For instance, the table of elements in chemistry is composed of many signs but contains few sentences. Nevertheless, the grouping of signs, arranged in a special tabular manner, enunciates numerous statements about chemical elements. Likewise, a statement is not the same as a proposition. The sentences “no other element besides gold has the atomic number 79” and “it is true that gold has 79 protons in its atom” express the same logical proposition but are grammatically distinct sentences and modally distinct statements. In the field of accounting, for example, multiple different statements may make the same proposition regarding the financial health of a company. 24

2  Useful Products in Information Systems Theorizing… 

49

participation take?” A major part of theorizing, thus, is producing meaningful statements related to the discursive formation of IS—statements that carry truth value and may be useful to society. Simultaneously, these meaningful statements “connect the dots” and uncover relations that are not necessarily obvious to the layperson. As Gibbs (1972) noted, statements “assert order in the universe” (p. 93), which is one of the goals of theorizing. Statements that make assertions and form the building blocks of arguments are called propositions. Propositions tie together concepts using logical links to determine whether something is or is not the case (Copi & Cohen, 2001; Gibbs, 1972). Thus, the above statement about user participation, when reformulated using a logical form such as “the greater the level of user participation, the higher the probability of project success,” becomes a formal relational proposition (Fawcett, 1998) that hypothesizes a relationship between one or more concepts, as commonly found in the antecedent-consequent type of IS research (Furneaux & Wade, 2009). Because theory takes various forms (Gregor, 2006), the propositions that make up theory also differ in nature. Theory that is descriptive could be made up of existential and definitional propositions (Fawcett, 1998). Existential propositions exert the existence or level of existence of a concept.26 According to Doty and Glick (1994) and Gregor (2006), when these kinds of nonrelational propositions take the form of typologies and taxonomies, they already qualify as explanatory theories. Definitional propositions describe the characteristics of these practices in a constitutive definition, representational definition, or operational definition.27 Hambrick (2003) considered this typology to be among the most widely tested, validated, and enduring theory in management. Although representational and operational definitions are useful inputs into causal theories (Fiss,  Alavi, M., & Leidner, D.  E.—(2001). Knowledge management and knowledge management systems: Conceptual foundations and research issues. MIS Quarterly, 25(1), 107–136—keenly demonstrate these types of nonrelational propositions in their highly cited knowledge management research. Based on their review, they propose three common applications of knowledge management that can all be empirically tested: (1) the coding and sharing of best practices, (2) the creation of corporate knowledge directories, and (3) the creation of knowledge networks. 27  For example, the classic Miles and Snow typology of organizational strategy—Miles, R. E., Snow, C. C., Meyer, A. D., & Coleman, H. J., Jr. (1978). Organizational strategy, structure, and process. Academy of Management Review, 3(3), 546–562, ibid.—categorizes organizations into prospectors, analyzers, and defenders. 26

50 

N. R. Hassan et al.

2011), editors and reviewers of top IS journals rarely agree and often sweepingly declare taxonomies as atheoretical and exploratory and thus rarely publish them, further undermining IS theorizing. In addition to the identity of statements, how society values them is pertinent to the IS field. Foucault (1972, p. 118) introduced the concept of the law of rarity of statements, which states that it is not enough simply to make a statement—statements should have ramifications for the discursive formations they occupy and, because of their value, should build connections with other meaningful statements in the larger corpus of knowledge. Consequently, the challenge for the IS field is not just to formulate statements of interest to scholars but also to be of interest and use to society at large.28

Hypothesis When propositions take the form of empirically testable conjectures, they are called hypotheses, the product of theorizing with which many IS researchers are most familiar. Derived from propositions, usually by linking operationally defined concepts, hypotheses represent expectations about the way the world works, assuming the assertions of the model are empirically adequate. It is common to see IS researchers claim that hypothesis testing is only associated with positivist research (Orlikowski & Baroudi, 1991). However, that is an oversimplification. Flyvbjerg (2006)  Within the nonpositivist tradition, statements play an even more critical role in research because the crux of any interpretive, ethnographic, phenomenological, grounded, critical, or other nonpositivist tradition is statements made about the meanings and essence of human experience. Whereas positivist research creates statements by seeking out cause-effect relationships among its concepts and constructs, phenomenological research brackets out prejudgments, biases, and preconceptions to capture the essence and meaning of human experience and consciousness. Moustakas, C. (1994). Phenomenological Research Methods. SAGE Publications. Conversely, prejudgments and biases are foregrounded and highlighted in the way hermeneutical research forms its statements. Gadamer, H.  G. (1975). Truth and Method (2nd ed.). Continuum Publishing Group.. That is, the form of statements in nonpositivist research is determined less by the relationships between concepts and constructs (as can be seen in the typical box-arrow diagram in the IS field) than by how the researcher participates in the experiences of the research subjects (i.e., ethnography); induces, deduces, and verifies meaning from the data (i.e., grounded theory); understands and interprets text (i.e., hermeneutics); and perceives and reduces the quality of the experience to the things themselves (i.e., phenomenology). 28

2  Useful Products in Information Systems Theorizing… 

51

argues that qualitative case study research is as amenable to hypothesis testing as quantitative research. Grounded theorists (Glaser & Strauss, 1967; Strauss & Corbin, 1990) agree that, in addition to building hypotheses, the deductive processes of grounded theory require the testing of hypotheses. Hypotheses represent the most concrete form of the proposition and deal with rigorously defined concepts that undergird the claims of that proposition. Because of the focus on clarifying and sharpening concepts and relations, qualitative research is arguably better suited for both generating and testing hypotheses (Silverman, 2006).

 rafting Theory with the Products C of Theorizing Next, we demonstrate how a reflective and mindful application of the 12 products of theorizing supports novel and creative research. We illustrate the different roles that each product plays at various stages of research. Theorizing is a creative process of discursive practices; thus, no set rules or methods can be specified for these practices. Nevertheless, examples of how products are deployed in theorizing can be helpful to researchers. And, indeed, multiple examples are presented throughout this book. As any study proceeds, the researcher is saddled with decisions that involve the various products of theorizing, such as How do I address the research question? Do I have the right research question? What can help me build my thesis? Where do I look for inspiration to push the research forward? Using media richness theory (MRT) as an example, a classic theory originally crafted within management discourse but later reached its maturity in IS discourse, we reconstruct the authors’ discursive practices and applications of the products of theorizing. These discursive practices, shown in Fig. 2.2, represent one instantiation of the numerous paths that researchers can take with the products. The reconstruction is based on a combination of historical evidence, published documents, and personal communication with several of the researchers. The goal is to approach the unique logic-in-use of the researchers and, as we do so, to illustrate how they come together to support discursive practices. Each path is numbered

52 

N. R. Hassan et al.

and corresponds to the discursive practice described in the following subsections. The reconstruction does not imply that there exists a universal theorizing strategy involving the products. Each study and context produces its own unique set of products of theorizing and specific paths based on the creativity of the researchers.

Raising Questions from Myth (1) The first discursive practice in MRT began with debunking a myth. At that time, the management field entertained a myth that managers acted like orchestra leaders, performing the classic four “management activities” of planning, organizing, coordinating, and controlling. Using six different studies of managers and his own study of five American CEOs, Mintzberg (1973, 1975) found that managers are not reflective systematic planners; they spend more than 70% of their time engaging in informal verbal communication and acting spontaneously when informed. Inspired by the debunking of this myth, Robert Lengel (1983) and his supervisor Richard Daft formulated the problem statement that later triggered the research for his dissertation: Managers spend eighty percent of their time communicating, often working under intense time pressures. As a result, many errors or problems within an organization are caused by poor or inaccurate communication. The purpose of this study has been to explore techniques which managers can use to communicate effectively. (p. iii)

The challenge at this stage of theorizing was to decide which questions to address. The reliance of managers on verbal media, namely telephone calls and meetings, in the context of the increasingly sophisticated technologies of the time (email and video conferencing), raised tantalizing questions that needed answers, such as Which media should managers use to be effective in their roles as problem solver, negotiator, master of ceremonies and rituals, and mentor? Given the reliance of managers on verbal media and face-to-face meetings that Mintzberg found, how will managers react to the introduction of information technology designed to support these tasks?

2  Useful Products in Information Systems Theorizing… 

53

Drawing Analogies from Questions (2) Answers to questions posed can often be found in other fields of study. Because the questions were related to how professionals communicate, Lengel found an analogous situation in the dissertation on communication channels for research scientists and engineers by Bodensteiner (1970). Bodensteiner made the point that informal channels for communicating scientific research (e.g., face-to-face meetings, phone calls, memos) are just as important or even more important than formal channels (e.g., journal articles, official reports). The similarities between the two domains were close enough for Lengel to draw an analogy from the characteristics of communications among research scientists to the context of general managers. Whereas Bodensteiner’s (1970) study within the communication field asked about how the use of informal communication channels would change as a function of project uncertainties, Lengel’s own questions pertained to how media choices made by managers affected the richness of information, a concern more specific to the IS field. Following from the possibility that IS artifacts translate organizational messages at various levels of richness, Lengel asked if the richness of media is related to the translation richness of information. If media richness is related to the translation richness of information, then it is related to information processing needs, making it possible to predict what kind of media might be suited for managerial information processing needs. These questions, which pertain to how information processing takes place in organizations, situated this study within the field of IS, and they gave rise to other disciplinary questions that had to be answered before the first set of questions could be addressed (Lengel, 1983): “(1) what are the task characteristics that cause a need for rich information? and, (2) how do media differ in their ability to convey rich information?” (p. 11). These questions were foundational because they provided possible answers to larger questions that were being asked about the use of communications technology for managers (Daft et al., 1987; Lengel & Daft, 1984), such as Why do managers prefer face-to-face exchanges of information in lieu of expensive and extensive computer-based management aids or written media in general? Why does soft information often have more impact than hard data?

54 

N. R. Hassan et al.

When this study was published in Management Science (Daft & Lengel, 1986), the disciplinary IS question was the article’s first sentence: “Why do organizations process information?” It is this question that Daft and Lengel addressed in detail in their MRT studies. More importantly, in contrast to the scripted manner in which much of IS research is being undertaken (Grover & Lyytinen, 2015), Daft and Lengel not only borrowed from communication studies, but they drew analogies between similar phenomena in a different discipline with the questions they posed in IS research to create new concepts that applied directly to IS phenomena.

Building Metaphor from Analogy (3) MRT studies drew an analogy between complex managerial decision-­ making and the higher-level complex biological and social systems (Boulding, 1956), thereby allowing the richness concept to be applied also to information and to information processing (Daft & Lengel, 1986; Lengel, 1983). The analogies between organizations and complex biological systems offered two metaphors for the research: the image of the organization as an information processing machine and information as the lifeblood of human societies. The first metaphor implies that precision, clarity, logic, and rational behavior result in targeted optimal performance. The second metaphor, which compares managers to organs of the body that use information to interpret the external environment, emphasizes the intuitive, social, and nonlogical aspects of managing the organization.29 Consequently, a later MRT study (Daft et al., 1987) directly challenged the myth that more advanced communication technologies and telecommuting would replace face-to-face meetings and enhance managers’ performance.  The metaphor of the organization as a machine is exemplified by the notion of the “total information system” of the 1960s research (that use supposed “objective” data and formal reports to optimize decision-making processes and enable total systems management), supporting the prevailing myth of the total MIS. Mintzberg, H. (1972). The Myths of MIS. California Management Review, 15(1), 92–97. The biological metaphor of managers as intuitive and social elements of organizations is at odds with the machine metaphor, resulting in major implications regarding a manager’s information requirements. If managers’ behavior is predominantly intuitive, the information provided by formal, logical MIS will conflict with their needs. This conflict indicates why managers did not buy into newly introduced advanced communication technologies at the time—such as video-conferencing systems, supposedly capable of transmitting verbal and visual information. 29

2  Useful Products in Information Systems Theorizing… 

55

Adopting Paradigm from Analogy (4) Early theorizing for MRT was inspired by language and communication studies (Daft & Wiginton, 1979), which applied a linguistic paradigm to evaluate the quality of communications based on sound (phoneme), words used (morphene), and patterns applied (syntax) (Lengel, 1983). This paradigm suggests that because formal channels of communications limit coding, research scientists prefer verbal and face-to-face communication. Daft and Lengel (1983) argued that linguistics was “only one aspect of managerial communication” (p. 7). The linguistic paradigm ignored the role of media in conveying information and may have been too abstract to serve as a useful heuristic so Daft and Lengel adopted two other paradigms: Galbraith’s (1973) information processing paradigm and Weick’s (1979) sensemaking paradigm.

Answering Questions from Paradigms (5) Galbraith’s paradigm offered a way to link information processing and the notion of uncertainty to organization design. Similarly, Weick’s paradigm provided a means of explaining media richness using the concept of equivocality. By 1986, these two paradigms were integrated into MRT (Daft & Lengel, 1986) as complementary dimensions that explained why organizations process information—namely, to reduce task uncertainty and resolve equivocality. In this case, MRT researchers used paradigms to “see his problem as like a problem he has already encountered” (Kuhn, 1970, p. 189) and extend their theorizing into new discourses to better describe the phenomenon being studied.

Guiding Analogies from Paradigms (6) The two paradigms of information processing and sensemaking also provided guidance for MRT researchers to extend the analogies drawn from Bodensteiner’s (1970) work of ranking particular communication media in their capacity to process rich information. Concepts from the information processing and sensemaking paradigms such as structural

56 

N. R. Hassan et al.

mechanisms that reduce uncertainty, facilitate the transfer of richer information, and reduce equivocality were analogized to Bodensteiner’s (1970) concept of richer subjective and personal media. Both uncertainty reduction and equivocality reduction were integrated to create new concepts.

Applying Law from Questions (7) MRT studies did not establish any laws because many of their hypotheses faced challenges from follow-up empirical studies. However, MRT studies did implicitly apply the law of requisite variety (Ashby, 1968), which states that the number of states of the control mechanism must be greater than or equal to the number of states in the system being controlled. This implicit example of the use of laws in theorizing is an instance of where laws initiate the theorizing process rather than become the result of a theorizing process. MRT is one of the few research studies in IS that discusses general laws (Hovorka et al., 2008). Other than McLuhan’s (1988) proposed four laws of media, there appears to be no general law of media characteristics that could serve as part of the system for governing managers’ behaviors and media choices. The law of requisite variety (Ashby, 1968) fulfills this role in the case of MRT. Thus, to stabilize the organization, the control mechanism that addresses information needs requires multiple coding systems, cues, and rapid feedback, which are the concepts that MRT applies in its model.

Abstracting Models from Law (8) Following the law of requisite variety (Ashby, 1968) that requires the use of multiple coding systems, cues, and rapid feedback to stabilize the organization, the MRT model is based on the fit between information processing and effective media. This fit model abstracts the complex decision-making processes that managers undertake as reducing uncertainty and resolving equivocality through media, focusing on media choices made by managers. In this model, the richness of information reflects the amount of change in understanding from interpreting the information communicated. This fit model of MRT is visualized by

2  Useful Products in Information Systems Theorizing… 

57

several conceptual models. The first uses two dimensions—uncertainty and equivocality—to construct a two-dimensional conceptual model that categorizes four kinds of events and problems that managers address via their communication processes. The second model relates seven structural mechanisms (group meetings, integrators, direct contact, planning, special reports) on a continuum with respect to their capacity for reducing uncertainty and resolving equivocality. The third model defines two underlying task characteristics—task variety and task analyzability—that link the structural mechanisms and the richness of information required to accomplish different tasks. All these homeomorph models represent distinct aspects of complex managerial communication and decision-­ making processes.

Synthesizing Framework from Models (9) All these models are used to synthesize the MRT research framework (Fig. 2.3), which maps out the background of the research and the main concepts (elaborated below) to their associated propositions. Although this framework looks like a box-arrow diagram, it is not a causal model and includes the uncertainty-equivocality model; the structural mechanism model, which houses the different media and their capacities to process information; and the task characteristics model, which is related to information processing requirements. The original framework in Lengel’s dissertation included additional concepts not shown in the diagram, such as moderating variables of personality, organizational culture, and geometry.

Fig. 2.3  MRT framework adapted from Daft and Lengel (1986)

58 

N. R. Hassan et al.

In implementing this framework, Daft and Lengel recorded the manager’s preferred media choice rather than the manager’s actual use of media, which became a point of contention for MRT’s challengers. Thus, critics (e.g., Dennis & Kinney, 1998) maintain that the central proposition of MRT—that is, the use of richer media in equivocal situations results in higher performance—was never actually tested. The framework that was built from the communication-based models of the earlier MRT studies (Daft et al., 1987; Lengel, 1983) emphasized information flows and information processing concepts. In this framework, the translation of information, richness of information, analyzability of tasks, equivocality, and uncertainty all contributed to the main concept of media richness and the role of media in managerial communications and performance. MRT’s problem originated from the management discourse (i.e., Mintzberg, 1973), but the inspiration and the concepts for addressing the problem come from the communications field (i.e., Bodensteiner, 1970). As Daft and Lengel continued their study, the discourse shifted from purely management or communications concerns toward the IS discourse. The rules of the discourse, the discursive formation, shifted from how to get work done through others (i.e., management), and the format, content, and channels of human communication (i.e., communication studies) to the capacities of different structures and IT artifacts and implications on organization design and performance of the IS field (Daft & Lengel, 1986; Daft et  al., 1987). Working within the IS discursive formation, Daft and Lengel were able to explore more interesting questions concerning the relationship between information processing, media, and managerial tasks that would not have been asked had the theorizing process stayed with the management or communication studies disciplines.

 eriving Concepts from Paradigm, Analogy, Metaphor, D and Model (10) Guided primarily by multiple paradigms, MRT researchers used the pre-­ theoretical products of theorizing, including analogies, metaphors, and

2  Useful Products in Information Systems Theorizing… 

59

models, to derive many new concepts that would become part of the MRT research framework. MRT studies introduced the new IS concept of media richness by analogizing Bodensteiner’s (1970) concept of channel fitness to the concept of media richness. By doing so, MRT studies attributed to media the very concepts that Bodensteiner had applied to communication channels. MRT studies also borrowed from the information processing paradigm the concept of information richness, defined as the extent to which information changes understanding as a result of its interpretation. Lengel (1983) transposed concepts of channel richness and information richness from the discourse of communication in research organizations to the discourse of managerial communications. This transposition enabled MRT authors to invent new concepts such as translation, execution, and translation richness. MRT studies also suggested certain propositions about managers’ information needs. This transposition of concepts enabled MRT’s creators to interpret how managers acquire and disseminate information—what would in subsequent articles be transformed into the concept of sensemaking—and the role of media in meeting their information needs. By conceptualizing richness in terms of information needs and media capacity, the MRT authors were able to use IS concepts to theorize how managerial communication is linked to information processing.

Inventing Constructs from Concepts (11) Bodensteiner (1970) derived two constructs to define the concept of information channel richness: the channel’s capacity and its capability. Using these constructs and their related variables, Bodensteiner ranked communication channels according to their richness and found a positive relationship between the use of richer channels and periods of uncertainty in projects. Analogously, Lengel (1983) derived three constructs for the concept of information richness: the variety of information cues a medium can use, its feedback capability, and its personal or impersonal nature. Lengel’s set of constructs differs from Bodensteiner’s in that the former are characteristics of media rather than characteristics of the communication channel.

60 

N. R. Hassan et al.

Both Daft and Lengel and Bodensteiner characterize oral media as being more personal than written media. The way that the constructs of media and the concept of translation richness are defined as inherent and objective characteristics of media or situations reflects the positivist nature of the MRT study. As in any positivist study, the choice of concepts and constructs is critical. Indeed, both the concept of media richness and that of translation richness are summative or composite terms that often create complications in research and in theorizing.

Formulating Statements from Constructs (12) With the help of a pilot study, MRT researchers defined their core concepts and constructs, which were used to formulate statements that describe how those concepts relate to each other. The primary statement is that “managers will be more effective and efficient communicators if they chose rich media to do translation tasks and less rich media to do execution tasks” (Lengel, 1983, p. 55). Because tasks demonstrating high equivocality require more coordination and exchange of opinions, Daft and Lengel (1986) suggested that the use of more personal oral communication media would be related to higher levels of information processing. These statements do not imply that other factors (e.g., organizational mandates, personal preferences, culture) have no influence on the choice of media, only that a significant relationship exists between the preference for rich media and ambiguous, emotion-laden communication.

Testing Hypotheses from Statements (13) Based on the biological metaphor that complex information requires equally complex processing, a simple regression plot of the richness of media against the richness of information would show a strong positive relationship. This proffered relationship became the basis for constructing several hypotheses in MRT. These hypotheses were refined by the information processing and sensemaking paradigm by the time a portion of

2  Useful Products in Information Systems Theorizing… 

61

Lengel’s dissertation was published in MIS Quarterly (Daft et al., 1987). Chi-squared tests for the first hypothesis showed that media richness is not independent of translation richness, and results from testing the second hypothesis showed that managers prefer oral media over written media when communication situations are equivocal. The results also supported the third hypothesis, which states that managers who choose the most suitable media perform better. These results were challenged by follow-up studies reporting no such conclusive results (Dennis & Kinney, 1998; Markus, 1994; Rice, 1992). These propositions and hypotheses underwent changes during the theorizing process and these changes contributed to problems that would later be addressed by other IS researchers. Lengel’s (1983) dissertation, which was guided by the linguistic paradigm, focused on “the richness of information conveyed or the amount of convergence required to reach understanding” (p. 2), represented by the concept of translation richness (or information richness). Later, MRT studies (Daft et al., 1987; Lengel & Daft, 1984) corrected translation richness (which originally represented the amount of convergence required to reach understanding) into the concepts of uncertainty and equivocality. This change marked the first major source of contention in MRT-related studies. Follow-up studies showing managers using written media (email) to resolve equivocal situations contradicted MRT’s claim that oral media is better suited for such situations. These conceptual differences led other studies to challenge MRT. The frameworks that were built in studies that challenged MRT, including media synchronicity theory (MST) (Dennis & Kinney, 1998; Kinney & Watson, 1992; Rice, 1992; Valacich et al., 1994), emphasized the notion of new media, social presence, task activities in a group context, and group-related outcomes. Notably, the concepts applied in the MRT and MST research frameworks are starkly different. It is thus not surprising that studies challenging MRT found contradictory results. The goal of the MST study was not to refine or fix the existing MRT theory. MST offered a completely new theory that could explain the capabilities of new media. Consequently, the authors of MST built a fresh framework for their research while remaining in the same general area of study. As the case of MRT demonstrates, not only are the products of theorizing useful in themselves, but it is also reasonable to assume that if

62 

N. R. Hassan et al.

theorizing includes more such products, the quality of theorizing and therefore the quality of the research will be enhanced. As the researcher works on defining each product and its relationship with other products, the overall coherence of the research improves. This does not mean that all products require the attention of the researcher, because each research area is contextually unique; however, the consideration of more products in the theorizing process suggests that researchers, or members of a subfield, have expended an adequate level of effort in laying the foundation for the research.

Conclusion The status of theories in the IS field and the process of theorizing are being called into question. To address these issues, we assemble the intermediate products of the theorizing process that can be found scattered in theorizing resources from different disciplines and offer a coherent view of how theorizing can be enhanced in IS. As the examples in this chapter demonstrate, considerable theoretical progress is made when the products are identified, explicated, unpacked, synthesized, and transformed in the theorizing process, bringing the results closer to becoming a theory ready to be empirically tested. The processes of theorizing in which these products are involved are described in Hassan et al. (2019), and the wide-­ranging implications from using these products of theorizing for researchers, reviewers, and editors are summarized in Appendix Table 2.2. A focus on pre-theoretical products frees IS researchers to make bold conjectures and undertake innovative research that eschews the “incremental adding-to-the-literature contributions and a blinkered mindset” (Alvesson & Sandberg, 2014, p. 967) and allows them to progress forward unencumbered by the fear that their research will fail to make a theoretical contribution. Regardless of whether theorizing uses a positivist, interpretive, critical, critical-realist, or any other nonpositivist approach, the products of theorizing play a critical role in advancing research in a productive and innovative fashion. The beauty of the products lies in their ability to open up spaces for scholarly discussion within any IS subfield and between

2  Useful Products in Information Systems Theorizing… 

63

distinct subfields, encouraging more research in the context of discovery rather than in the context of justification (Hanson, 1958). Using a clear explication of the products and a pragmatic way forward, we trust that this manuscript will make the process of theorizing less of a mystery and more of a way to inspire innovative thinking in IS research.

Appendix Table 2.2  Implications from using products of theorizing Products

Implications

Discursive formation

Clarify within which discourse the research is undertaken and around which axis of cohesion the study is coalescing. Is it really extending IS research or is the research extending social psychology, computer science, economics, or other reference disciplines? Along which boundaries of various discourses are the cross-disciplinary studies being conducted? Problematize the research based on disciplinary questions. What makes the research question IS-specific? Are the same questions being asked by other disciplines or is the research asking questions that other disciplines are incapable of answering? Leverage exemplars in paradigms. Identify contradictions that could become the source of new paradigms. What paradigms exists out there that could provide concrete problem solutions for the research? What generative metaphors exist that could inspire novel views for the problems at hand? Identify generalizations that could serve as starting points for events to be examined. What social laws support IS theorizing? Recognize and marshal the total set of relations of epistemological elements that address why the topic matters and why the means proposed are appropriate and rigorous. What intellectual tradition, system of concepts, assumptions, beliefs, and theories support and inform the research? What weaknesses within existing frameworks open opportunities for theorizing?

Question

Paradigm

Law Framework

(continued)

64 

N. R. Hassan et al.

Table 2.2 (continued) Products

Implications

Myth

Interrogate unquestioned assumptions, common knowledge, or unproven beliefs. Has the researcher explored nonstandard tools or methods available at hand to investigate symbols of value, solidarity, and social structure that have held despite conflicts and contradictions? Draw analogies from similar structures in different domains in order to illustrate the phenomenon under investigation using a simplified reference to something more familiar. What scaled-down reference is capable of demonstrating and explaining something more complex? Harness the entire network of analogies in physical or linguistic objects that are different from the phenomenon being studied but are able to clarify, enrich, and enlighten. What metaphors exist that describe the phenomenon in an original, economic, consistent, and elegant manner? Abstract the phenomenon of interest using positive and neutral analogies in order to build a precise and economic representation of selected elements and relationships. What models can reveal the consequences of making certain assumptions or excluding certain elements? Invent IS-specific set of ideas that demarcate the IS field’s subject matter and declare to the world the identity of our field. What concepts belong to the IS field? Connect all the products of theorizing into meaningful IS statements that make claims, carry truth value, and are useful to society. Which IS statements connect the dots and reveal relations that are not obvious to the layperson? Which IS statements assert order in the universe?

Analogy

Metaphor

Model

Concept and Constructs Statement

References Abbott, A. D. (2001). Chaos of disciplines. University of Chicago Press. Abbott, A. D. (2004). Methods of discovery: Heuristics for the social sciences. Norton. Adam, F., & Fitzgerald, B. (2000). The status of the IS field: Historical perspective and practical orientation. Information Research, 5(4) Retrieved from http://www.informationr.net/ir/5-­4/paper81.html Agarwal, R., & Lucas, H. C., Jr. (2005). The information systems identity crisis: Focusing on high-visibility and high-impact research. MIS Quarterly, 29(3), 381–398.

2  Useful Products in Information Systems Theorizing… 

65

Ågerfalk, P. J. (2014). Insufficient theoretical contribution: A conclusive rationale for rejection? European Journal of Information Systems, 23(6), 593–599. Alavi, M., & Leidner, D. E. (2001). Knowledge management and knowledge management systems: Conceptual foundations and research issues. MIS Quarterly, 25(1), 107–136. Alvesson, M., & Sandberg, J. (2011). Generating research questions through problematization. Academy of Management Review, 36(2), 247–271. Alvesson, M., & Sandberg, J. (2013). Constructing research questions: Doing interesting research. SAGE. Alvesson, M., & Sandberg, J. (2014). Habitat and habitus: Boxed-in versus box-­ breaking research. Organization Studies, 35(7), 967–987. Angst, C. M., Agarwal, R., Sambamurthy, V., & Kelley, K. (2010). Social contagion and information technology diffusion: The adoption of electronic medical records in US hospitals. Management Science, 56(8), 1219–1241. Ashby, W. R. (1968). Principles of self-organizing systems. In W. Buckley (Ed.), Modern systems research for the behavioral scientist (pp. 108–118). Aldine. Atkinson, C.  J. (2003). The nature and role of generative systemic metaphor within information systems planning and development. In E.  H. Wynn, E. A. Whitley, M. D. Myers, & J. I. DeGross (Eds.), Global and organizational discourse about information technology (Vol. 110, pp.  323–343). IFIP/Springer. Avison, D., & Malaurent, J. (2014). Is theory king?: Questioning the theory fetish in information systems. Journal of Information Technology, 29(4), 327–336. Bacharach, S. B. (1989). Organizational theories: Some criteria for evaluation. Academy of Management Review, 14(4), 496–515. Bagnall, J. (2012). What’s the difference between analogy, metaphor and simile. Quora. https://www.quora.com/Whats-­the-­difference-­between-­analogy-­ metaphor-­and-­simile. Bal, M. (2002). Travelling concepts in the humanities: A rough guide. University of Toronto Press. Banville, C., & Landry, M. (1989). Can the field of MIS be disciplined? Communications of the ACM, 32(1), 48–60. Beer, S. (1972). Brain of the firm. Penguin. Beer, S. (1979). The heart of enterprise. Wiley. Benbasat, I., & Zmud, R. (2003). The identity crisis within the IS discipline: Defining and communicating the discipline’s core properties. MIS Quarterly, 27(2), 183–194.

66 

N. R. Hassan et al.

Berger, P.  L., & Luckmann, T. (1966). The social construction of reality. Anchor Books. Bijker, W. E. (1995). Of bicycles, bakelites, and bulbs: Toward a theory of sociotechnical change. MIT Press. Bijker, W.  E., Hughes, T.  P., & Pinch, T.  J. (1987). The social construction of technological systems. MIT Press. Black, F., & Scholes, M. (1973). The pricing of options and corporate liabilities. Journal of Political Economy, 81(3), 637–654. Blaug, M. (1997). Economic theory in retrospect. Cambridge University Press. Blumer, H. (1954). What is wrong with social theory? American Sociological Review, 11(1), 3–10. Bodensteiner, W.  D. (1970). Information channel utilization under varying research and development project conditions: An aspect of inter-organizational communication channel usages. [Unpublished PhD dissertation]. The University of Texas at Austin. Boland, R. J. (1982). Myth and technology in the American accounting profession. Journal of Management Studies, 19(1), 109–127. Boland, R. J. (1987). The in-formation of information systems. In R. J. Boland & R.  A. Hirschheim (Eds.), Critical issues in information systems research (pp. 363–379). Wiley. Boland, R. J., & Pondy, L. R. (1983). Accounting in organizations: A union of natural and rational perspectives. Accounting, Organizations and Society, 8(2–3), 223–234. Boulding, K.  E. (1956). General systems theory: The skeleton of science. Management Science, 2(3), 197–208. Brewer, E., Demmer, M., Du, B., Ho, M., Kam, M., Nedevschi, S., et al. (2005). The case for technology in developing regions. IEEE Computer, 38(6), 25–38. Bromberger, S. (1992). On what we know we don't know: Explanation, theory, linguistics, and how questions shape them. University of Chicago Press. Brown, R. H. (1976). Social theory as metaphor: On the logic of discovery for the sciences of conduct. Theory and Society, 3(2), 169–197. Brynjolfsson, E., Hofmann, P., & Jordan, J. (2010). Cloud computing and electricity: Beyond the utility model. Communications of the ACM, 53(5), 32–34. Burton-Jones, A., & Straub, D. W. (2006). Reconceptualizing system usage: An approach and empirical test. Information Systems Research, 17(3), 228–246. Campbell, N.  R. (1920). Foundations of science: The philosophy of theory and experiment. Dover.

2  Useful Products in Information Systems Theorizing… 

67

Carroll, A. B. (1979). A three-dimensional conceptual model of corporate performance. Academy of Management Review, 4(4), 497–505. Cassirer, E., & Verene, D. P. (1979). Symbol, myth, and culture: Essays and lectures of Ernst Cassirer, 1935–1945. Yale University Press. Chen, W., & Hirschheim, R. (2004). A paradigmatic and methodological examination of information systems research from 1991 to 2001. Information Systems Journal, 14(3), 197–235. Cohen, P. S. (1969). Theories of myth. Man, 4(3), 337–353. Collins, H. M., & Pinch, T. J. (1982). Frames of meaning: The social construction of extraordinary science. Routledge & Kegan Paul. Constant, E.  W. (1980). The origins of the turbojet revolution. Johns Hopkins University Press. Copi, I. M., & Cohen, C. (2001). Introduction to logic. Prentice-Hall. Corley, K. G., & Gioia, D. A. (2011). Building theory about theory building: What constitutes a theoretical contribution. Academy of Management Review, 36(1), 12–32. Corvellec, H. (2013). Why ask what theory is? In H. Corvellec (Ed.), What is theory? Answers from the social and cultural sciences (pp. 9–24). Liber CBS Press. Cronbach, L.  J., & Meehl, P.  E. (1955). Construct validity in psychological tests. Psychological Bulletin, 52(4), 281–302. Czarniawska, B. (2013). What social science theory is and what it is not. In H. Corvellec (Ed.), What is theory? Answers from the social and cultural sciences (pp. 99–118). Liber CBS Press. Daft, R. L. & Lengel, R. H. (1983. Last updated). Information richness: A new approach to managerial behavior and organization design (No. TR-ONR-DG-02): Department of Management, Texas A&M University. Retrieved from https://apps.dtic.mil/sti/citations/ADA128980. Daft, R. L., & Lengel, R. H. (1986). Organizational information requirements, media richness and structural design. Management Science, 32(5), 554–571. Daft, R. L., Lengel, R. H., & Trevino, L. K. (1987). Message equivocality, media selection, and manager performance: Implications for information systems. MIS Quarterly, 11(3), 355–366. Daft, R. L., & Wiginton, J. C. (1979). Language and organization. Academy of Management Review, 4(2), 179–191. Darwin, C. (1859). On the origin of species. John Murray. Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 13(3), 318–340.

68 

N. R. Hassan et al.

Davis, M. S. (1971). That’s interesting: Towards a phenomenology of sociology and a sociology of phenomenology. Philosophy of the Social Sciences, 1(4), 309–344. Dearden, J. (1966). Myth of real-time management information. Harvard Business Review, 44(3), 123–132. Delone, W. H., & McLean, E. R. (1992). Information system success: The quest for the dependent variable. Information Systems Research, 3(1), 60–95. Denning, P.  J., Comer, D.  E., Gries, D., Mulder, M.  C., Tucker, A., Turner, A. J., et al. (1989). Computing as a discipline. Communications of the ACM, 32(1), 9–23. Dennis, A. R., & Kinney, S. T. (1998). Testing media richness theory in the new media: The effects of cues, feedback, and task equivocality. Information Systems Research, 9(3), 256–274. Dilthey, W. (1883/1989). Introduction to the human sciences. Princeton University Publishers. Doty, D. H., & Glick, W. H. (1994). Typologies as a unique form of theory building: Toward improved understanding and modeling. Academy of Management Review, 19(2), 230–251. Dubin, R. (1969). Building theory. The Free Press. Durkheim, É. (1951/1897). On suicide: A study in sociology. Free Press. Ein-Dor, P., & Segev, E. (1981). A paradigm for management information systems. Praeger. Fawcett, J. (1998). The relationship of theory and research. F. A. Feyerabend, P. (1978). Against method. Verso. Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention and behavior. Addison-Wesley. Fishbein, M., & Ajzen, I. (1977). Belief, attitude, intention and behavior. Addison-Wesley. Fiss, P. C. (2011). Building better causal theories: A fuzzy set approach to typologies in organization research. Academy of Management Journal, 54(2), 393–420. Flick, U. (1998). An introduction to qualitative research. SAGE. Flyvbjerg, B. (2006). Five misunderstandings about case-study research. Qualitative Inquiry, 12(2), 219–245. Foucault, M. (1970). The order of things: An archeology of the human sciences. Pantheon Books. Foucault, M. (1972). The archaeology of knowledge and the discourse on language. Pantheon Books.

2  Useful Products in Information Systems Theorizing… 

69

Franz, C. R., & Robey, D. (1984). An investigation of user-led system design: Rational and political perspectives. Communications of the ACM, 27(12), 1202–1209. Furneaux, B., & Wade, M. (2009). Theoretical constructs and relationships in information systems research. In Y.  K. Dwivedi, B.  Lal, M.  D. Williams, S. L. Schneberger, & M. Wade (Eds.), Handbook of research on contemporary theoretical models in information systems (pp. 1–17). IGI Global. Gadamer, H.  G. (1975). Truth and method (2nd ed.). Continuum Publishing Group. Galbraith, J. R. (1973). Designing complex organizations. Addison-Wesley. Geary, J. (2009). Metaphorically speaking. TED. https://www.ted.com/talks/ james_geary_metaphorically_speaking?language=en Geertz, C. (1973). The interpretation of cultures. Basic Books. Gentner, D. (1983). Structure-mapping: A theoretical framework for analogy. Cognitive Science, 1, 155–170. Gentner, D. (1989). Mechanisms of analogical reasoning. In S. Vosniadou & A.  Ortony (Eds.), Similarity and analogical reasoning (pp.  199–241). Cambridge University Press. Gibbs, J. P. (1972). Sociological theory construction. . Gilbert, W. (1893/1600). On the loadstone and magnetic bodies and on the great magnet the earth. Wiley. Gkeredakis, M., & Constantinides, P. (2019). Phenomenon-based problematization: Coordinating in the digital era. Information and Organization, 29(3), Article 100254. Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory: Strategies for qualitative research. De Gruyter. Goles, T., & Hirschheim, R. (2000). The paradigm is dead, the paradigm is dead … Long live the paradigm: The legacy of Burrell and Morgan. Omega, 28(3), 249–268. Gorry, G. A., & Scott Morton, M. S. (1971). A framework for management information systems. Sloan Management Review, 13(1), 55–70. Gregor, S. (2006). The nature of theory in information systems. MIS Quarterly, 30(3), 611–642. Gregor, S. (2014). Theory: Still king but needing a revolution! Journal of Information Technology, 29(4), 337–340. Gregor, S., & Jones, D. (2007). The anatomy of a design theory. Journal of the Association for Information Systems, 8(5), 312–335.

70 

N. R. Hassan et al.

Grover, V. (2012). The information systems field: Making a case for maturity and contribution. Journal of the Association for Information Systems, 13(4), 254–272. Grover, V., & Lyytinen, K. (2015). New state of play in information systems research: The push to the edges. MIS Quarterly, 39(2), 271–296. Grover, V., Lyytinen, K., Srinivasan, A., & Tan, B. C. Y. (2008). Contributing to rigorous and forward thinking explanatory theory. Journal of the Association for Information Systems, 9(2), 40–47. Gutting, G. (Ed.). (1980). Paradigms and revolutions: Applications and appraisals of Thomas Kuhn’s philosophy of science. University of Notre Dame Press. Hambrick, D.  C. (2003). On the staying power of defenders, analyzers, and prospectors. Academy of Management Executive, 17(4), 115–118. Hanson, N. R. (1958). Patterns of discovery: An inquiry into the conceptual foundations of science. Cambridge University Press. Harré, R. (1970). The principles of scientific thinking. University of Chicago Press. Hart, C. (1998). Doing a literature review. SAGE. Hassan, N.  R. (2014a). Paradigm lost … paradigm gained: A hermeneutical rejoinder to Banville and Landry’s “Can the Field of MIS be Disciplined?”. European Journal of Information Systems, 23(6), 600–615. Hassan, N. R. (2014b). Value of IS research: Is there a crisis? Communications of the Association for Information Systems, 34, 801–816. Hassan, N. R., Mathiassen, L., & Lowry, P. B. (2019). The process of information systems theorizing as a discursive practice. Journal of Information Technology, 34(3), 198–220. Hassan, N. R., & Mingers, J. C. (2018). Reinterpreting the Kuhnian paradigm in information systems. Journal of the Association for Information Systems, 19(7), 568–599. Hempel, C.  G. (1965). Aspects of scientific explanation and other essays in the philosophy of science. Free Press. Hempel, C. G., & Oppenheim, P. (1948). Studies in the logic of explanation. Philosophy of Science, 15(2), 135–175. Hesse, M. (1967). Models and analogy in science. In P. Edwards (Ed.), The encyclopedia of philosophy (Vol. 5, pp. 354–359). The Macmillan & The Free Press. Hesse, M.  B. (1966). Models and analogies in science. University of Notre Dame Press. Hirschheim, R. (2019). Against theory: With apologies to Feyerabend. Journal of the Association for Information Systems, 20(9), 1340–1357.

2  Useful Products in Information Systems Theorizing… 

71

Hirschheim, R., & Klein, H. K. (2012). A glorious and not-so-short history of the information systems field. Journal of the Association for Information Systems, 13(4), 188–235. Hirschheim, R. A., & Newman, M. (1991). Symbolism and information systems development: Myth, metaphor and magic. Information Systems Research, 2(1), 29–62. Hovorka, D.  S., Germonprez, M., & Larsen, K.  R. (2008). Explanation in information systems. Information Systems Journal, 18(1), 23–43. Iivari, J., Hirschheim, R., & Klein, H. K. (1998). A paradigmatic analysis contrasting information systems development approaches and methodologies. Information Systems Research, 9(2), 164–193. Ives, B., Hamilton, S., & Davis, G.  B. (1980). A framework for research in computer-based management information systems. Management Science, 26(9), 910–934. Ives, B., & Learmouth, G. P. (1984). The information system as a competitive weapon. Communications of the ACM, 27(12), 1183–1201. Jenkins, R. V. (1975). Images and enterprise: Technology and the American photographic industry, 1839 to 1925. Johns Hopkins University Press. Kaarst-Brown, M. L., & Robey, D. (1999). More on myth, magic and metaphor: Cultural insights into the management of information technology in organizations. Information Technology & People, 12(2), 192–217. Kahn, H. (1965). On escalation: Metaphors and scenarios. Praeger. Kaplan, A. (1964). The conduct of inquiry: Methodology for behavioral science. . Keen, P. G. W. (1980). MIS research: Reference disciplines and a cumulative tradition. In Paper presented at the International Conference on Information Systems. Keil, M. (1995). Pulling the plug: Software project management and the problem of project escalation. MIS Quarterly, 19(4), 421–447. Keil, M., & Robey, D. (1999). Turning around troubled software projects: An exploratory study of the deescalation of commitment to failing courses of action. Journal of Management Information Systems, 15(4), 63–87. Keil, M., Tan, B. C. Y., Wei, K.-K., Saarinen, T., Tuunainen, V., & Wassenaar, A. (2000). A cross-cultural study on escalation of commitment behavior in software projects. MIS Quarterly, 24(2), 299–325. Kendall, J. E., & Kendall, K. E. (1993). Metaphors and methodologies: Living beyond the systems machine. MIS Quarterly, 17(2), 149–171.

72 

N. R. Hassan et al.

Kendler, H. H., & Kendler, T. S. (1962). Vertical and horizontal processes in problem solving. Psychological Review, 69(1), 1–16. Khazanchi, D. & Munkvold, B. E. (2003). On the rhetoric and relevance of IS research paradigms: A conceptual framework and some propositions. In Paper presented at the Hawaii International Conference on System Sciences. King, J.  L., & Lyytinen, K. (2004). Reach and grasp. MIS Quarterly, 28(4), 539–552. Kinney, S. T. & Watson, R. T. (1992). The effect of medium and task on dyadic communication. In Paper presented at the International Conference on Information Systems. Kirby, J.  T. (1997). Aristotle on metaphor. American Journal of Philology, 118(4), 517–554. Kuechler, W., & Vaishnavi, V. (2012). A framework for theory development in design science research: Multiple perspectives. Journal of the Association for Information Systems, 13(6), 395–423. Kuhn, T. (1970). The structure of scientific revolutions (2nd ed.). University of Chicago Press. Kuhn, T. S. (1957). The Copernican revolution: Planetary astronomy in the development of Western thought. Harvard University Press. Kuhn, T. S. (1977). The essential tension: Selected studies in scientific tradition and change. University of Chicago Press. Lachman, R. (1960). The model in theory construction. Psychological Review, 67(2), 113–129. Larsen, T.  J., & Levine, L. (2008). Citation patterns in MIS: An analysis of exemplar articles. In G.  León, A.  M. Bernardos, J.  R. Casar, K.  Kautz, & J. I. DeGross (Eds.), Open IT-based innovation: Moving towards cooperative it transfer and knowledge diffusion (pp. 23–38). Springer. Lengel, R.  H. (1983). Managerial information processing and communication media source selection behavior. [Unpublished PhD dissertation]. Texas A&M University. Lengel, R. H. & Daft, R. L. (1984. Last updated). An exploratory analysis of the relationship between media richness and managerial information processing (No. TR-ONR-DG-08). Department of Management, Texas A&M University. Retrieved from: https://apps.dtic.mil/sti/citations/ADA143503 Lévi-Strauss, C. (1963). Structural anthropology. Basic Books. Lévi-Strauss, C. (1966). The savage mind. Weidenfeld & Nicolson. Lewis, B. R., Templeton, G. F., & Byrd, T. A. (2005). A methodology for construct development in MIS research. European Journal of Information Systems, 14(4), 388–400.

2  Useful Products in Information Systems Theorizing… 

73

Lim, S., Saldanha, T., Malladi, S., & Melville, N.  P. (2009). Theories used in information systems research: Identifying theory networks in leading IS journals. In Paper presented at the International Conference on Information Systems. Linnaeus, C. (1735). Systema naturea. M. A. David. Liu, F., & Myers, M. D. (2011). An analysis of the AIS basket of top journals. Journal of Systems and Information Technology, 13(1), 5–24. MacCorquodale, K., & Meehl, P. E. (1948). On a distinction between hypothetical constructs and intervening variables. Psychological Review, 55(2), 95–107. Markus, M.  L. (1994). Electronic mail as the medium of managerial choice. Organization Science, 5(4), 502–527. Markus, M. L. (2014). Maybe not the king, but an invaluable subordinate: A commentary on Avison and Malaurent’s advocacy of ‘theory light’ IS research. Journal of Information Technology, 29(4), 341–345. Markus, M. L., & Rowe, F. (2018). Is IT changing the world? Conceptions of causality for information systems theorizing. MIS Quarterly, 42(4), 1255–1280. Markus, M. L., & Saunders, C. S. (2007). Editorial comments: Looking for a few good concepts … And theories … For the information systems field. MIS Quarterly, 31(1), iii–vi. Mason, R. M. (1991). Metaphors and strategic information systems planning. In Paper presented at the 24th Hawaii International Conference on System Sciences. Mason, R. O., & Mitroff, I. I. (1973). A program for research on management information systems. Management Science, 19(5), 475–487. Maxwell, J. A. (2013). Qualitative research design (3rd ed.). SAGE. McKeon, R. P. (1941). The basic works of Aristotle. Random House. McLuhan, M., & McLuhan, E. (1988). Laws of media: The new science. University of Toronto Press. Mendeleev, D. (1869). Über die beziehungen der eigenschaften zu den atomgewichten der elemente (On the relationship of properties to the atomic weights of elements). Zeitschrift für Chemie (Journal of Chemistry), 12, 405–406. Meyer, M. (1995). Of problematology: Philosophy, science, and language. University of Chicago Press. Miles, M. B., & Huberman, A. M. (1994). Qualitative data analysis: An expanded sourcebook. SAGE. Miles, M. B., Huberman, A. M., & Saldana, J. (2014). Qualitative data analysis: A methods sourcebook. SAGE.

74 

N. R. Hassan et al.

Miles, R.  E., Snow, C.  C., Meyer, A.  D., & Coleman, H.  J., Jr. (1978). Organizational strategy, structure, and process. Academy of Management Review, 3(3), 546–562. Mingers, J. (2004). Paradigm wars: Ceasefire announced who will set up the new administration. Journal of Information Technology, 19(3), 165–171. Minsky, M. (1975). A framework for representing knowledge. In J. Haugeland (Ed.), Mind design II (pp. 111–142). MIT Press. Mintzberg, H. (1972). The myths of MIS. California Management Review, 15(1), 92–97. Mintzberg, H. (1973). The nature of managerial work. Harper & Row. Mintzberg, H. (1975). The manager’s job: Folklore and fact. Harvard Business Review, 68(2), 163–176. Moody, D., Iacob, M.-E., & Amrit, C. (2010). In search of paradigms: Identifying the theoretical foundations of the IS field. In Paper presented at the European Conference on Information Systems. Moore, H. L. (2004). Global anxieties: Concept-metaphors and pre-theoretical commitments in anthropology. Anthropological Theory, 4(1), 71–88. Mousavidin, E. & Goel, L. (2007). Seeking dragons in IS research. In Paper presented at the Americas Conference on Information Systems. Moustakas, C. (1994). Phenomenological research methods. SAGE. Oates, B. J., & Fitzgerald, B. (2007). Multi-metaphor method: Organizational metaphors in information systems development. Information Systems Journal, 17(4), 421–449. Orlikowski, W. J., & Baroudi, J. J. (1991). Studying information technology in organizations: Research approaches and assumptions. Information Systems Research, 2(1), 1–28. Orlikowski, W. J., & Iacono, C. S. (2001). Research commentary: Desperately seeking the “IT” in IT research: A call to theorizing the IT artifact. Information Systems Research, 12(2), 121–134. Ortony, A. (Ed.). (1979). Metaphor and thought. Cambridge University Press. Parsons, G.  L. (1983). Information technology: A new competitive weapon. Sloan Management Review, 35(1), 3–14. Petter, S., Straub, D., & Rai, A. (2007). Specifying formative constructs in information systems research. MIS Quarterly, 31(4), 623–656. Popper, K.  R. (1970). Normal science and its dangers. In I.  Lakatos & A. Musgrave (Eds.), Criticism and the growth of knowledge: International colloquium in the philosophy of science (Bedford College 1965) (pp.  51–58). Cambridge University Press.

2  Useful Products in Information Systems Theorizing… 

75

Ravitch, S. M., & Riggan, M. (2012). Reason and rigor: How conceptual framework guides research. SAGE. Revens, L. (1972). The first twenty-five years: ACM 1947–1962. Communications of the ACM, 15(7), 485–490. Rice, R. E. (1992). Task analyzability, use of new media, and effectiveness: A multi-site exploration of media richness. Organization Science, 3(4), 475–500. Richardson, H., & Robinson, B. (2007). The mysterious case of the missing paradigm: A review of critical information systems research 1991-2001. Information Systems Journal, 17(3), 251–270. Ritzer, G. (1980). Sociology: A multiple paradigm science. Allyn & Bacon. Robey, D., & Markus, M. L. (1984). Rituals in information system design. MIS Quarterly, 8(1), 5–14. Rogers, E. M. (1983). Diffusion of innovations (3rd ed.). The Free Press. Rosenberg, N. (1976). Perspectives on technology. Cambridge University Press. Rosenblueth, A., & Wiener, N. (1945). The role of models in science. Philosophy of Science, 12(4), 316–321. Runkel, P. J., & Runkel, M. (1984). A guide to usage for writers and students in the social sciences. Rowman & Allanheld. Sabherwal, R., Hirschheim, R., & Goles, T. (2001). The dynamics of alignment: Insights from a punctuated equilibrium model. Organization Science, 12(2), 179–197. Salmon, W. C. (1984). Scientific explanation and the causal structure of the world. Princeton University Press. Sarker, S., Chatterjee, S., Xiao, X., & Elbanna, A. (2019). The sociotechnical axis of cohesion for the IS discipline: Its historical legacy and continued relevance. MIS Quarterly, 43(3), 695–719. Sartori, G. (1975). The tower of babel. In G. Sartori, F. W. Riggs, & H. Teune (Eds.), Tower of Babel: On the definition and analysis of concepts in the social sciences (Vol. No. 6, pp. 7–37). International Studies Association. Sartori, G. (Ed.). (1984). Social science concepts: A systematic analysis. SAGE. Schaller, R. R. (1997). Moore’s law: Past, present and future. IEEE Spectrum, 34(6), 52–59. Schiffrin, D., Tannen, D., & Hamilton, H. E. (Eds.). (2001). The handbook of discourse analysis. . Schön, D. A. (1963). The displacement of concepts. Tavistock. Schön, D. A. (1979). Generative metaphor: A perspective on problem-setting in social policy. In A.  Ortony (Ed.), Metaphor and thought (pp.  254–283). Cambridge University Press.

76 

N. R. Hassan et al.

Schön, D. A. (1983). The reflective practitioner: How professionals think in action. Basic Books. Scriven, M. (1962). Explanations, predictions, and laws. In H.  Feigl & G.  Maxwell (Eds.), Minnesota studies in the philosophy of science (Scientific explanation, space and time) (Vol. 3). University of Minnesota Press. Silverman, D. (2006). Interpreting qualitative data. SAGE. Slife, B. D., & Williams, R. N. (1995). What’s behind the research: Discovering hidden assumptions in the behavioral sciences. SAGE. Spedding, J., Ellis, R. L., & Heath, D. D. (Eds.). (1901). The works of Francis Bacon (Vol. 4). Houghton Mifflin. Straub, D. (2012). Editorial: Does MIS have native theories. MIS Quarterly, 36(2), iii–xii. Strauss, A., & Corbin, J. (1990). Basics of qualitative research: Grounded theory procedures and techniques. SAGE. Street, C.  T., & Denford, J.  S. (2012). Punctuated equilibrium theory in IS research. In Y.  K. Dwivedi, M.  R. Wade, & S.  L. Schneberger (Eds.), Information systems theory: Explaining and predicting our digital society (Vol. 1, pp. 335–354). Springer. Suppe, F. (1977). The structure of scientific theories. University of Illinois Press. Thomson, G. P. (1961). The inspiration of science. Oxford University Press. Torgerson, W. S. (1958). Theory and methods of scaling. Wiley. Trice, H. M., & Beyer, J. M. (1984). Studying organizational cultures through rites and ceremonials. Academy of Management Review, 9(4), 653–669. Tsoukas, H. (1993). Analogical reasoning and knowledge generation in organization theory. Organization Studies, 14(3), 323–346. Valacich, J.  S., Mennecke, B.  E., Wachter, R.  M., & Wheeler, B.  C. (1994). Extensions to media richness theory: A test of the task-media fit hypothesis. In Paper presented at the 27th Hawaiian International Conference on System Sciences. Venkatesh, V., Morris, M. G., Davis, G. B., & Davis, F. D. (2003). User acceptance of information technology: Toward a unified view. MIS Quarterly, 27(3), 425–478. Walsham, G. (2006). Doing interpretive research. European Journal of Information Systems, 15(3), 320–330. Weber, M. (1930). The Protestant ethic and the spirit of capitalism. G. Allen & Unwin, Ltd.. Weber, R. (2003). Editor’s comment: Still desperately seeking the IT artifact. MIS Quarterly, 27(2), iii–xi.

2  Useful Products in Information Systems Theorizing… 

77

Weber, R. (2006). Reach and grasp in the debate over the IS core: An empty hand? Journal of the Association for Information Systems, 7(10), 703–713. Weber, R. (2012). Evaluating and developing theories in the information systems discipline. Journal of the Association for Information Systems, 13(1), 1–30. Weick, K. E. (1979). The social psychology of organizing (2nd ed.). Random House. Weick, K. E. (1995). What theory is not, theorizing is. Administrative Science Quarterly, 40(3), 385–390. Weiss, G., & Wodak, R. (2003). Introduction: Theory, interdisciplinarity and critical discourse analysis. In G. Weiss & R. Wodak (Eds.), Critical discourse analysis: Theory and interdisciplinarity (pp. 1–32). Palgrave Macmillan. Whewell, W. (1840/1967). The philosophy of the inductive sciences, founded upon their history. Johnson Reprint Corp. Willcocks, L., Hirschheim, R. A., & Dennis, A. R. (2019). Panel presentation. In Paper presented at the SIGPHIL@ICIS Workshop on the Death of Theory in IS and Analytics, Munich, Germany. Zhang, P., Scialdone, M., & Ku, M.-C. (2011). IT artifacts and the state of IS research. In Paper presented at the International Conference on Information Systems.

3 Debating Genres and IS Research: The Case of Action Principles for Service Automation Leslie Willcocks , Mary C. Lacity , and Daniel Gozman

Introduction Information systems (IS) is a peculiar hybrid wanting often to define itself as a discipline, though it has assembled itself as something that could more persuasively be called a field. It is shot through with ‘discipline anxiety’— the kind of thing, for example, social sciences display toward natural

L. Willcocks (*) Department of Management, London School of Economics and Political Science, London, UK M. C. Lacity Information Systems Department, Walton College of Business, University of Arkansas, Fayetteville, AR, USA e-mail: [email protected] D. Gozman University of Sydney Business School, Darlington, NSW, Australia e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. P. Willcocks et al. (eds.), Advancing Information Systems Theories, Volume II, Technology, Work and Globalization, https://doi.org/10.1007/978-3-031-38719-7_3

79

80 

L. Willcocks et al.

sciences, economics toward mathematics, and management toward economics and quantitative methods. Part of this discipline anxiety has been to keep returning to a rigor versus relevance debate, as if both rigor AND relevance were not critical to the study of an applied, fast-changing study area, driven by emerging technologies, the exponential data explosion, and related phenomena. With this background, IS researchers often divide into many camps depending on topic focus, research methodology, and attitudes toward theorizing. Many privilege methodology, purity of data, and the rigor of analytical techniques over the value of the findings for practical purposes. Such scholars see themselves as practicing a social science heavily imbued with natural science techniques and values. Others expend their curiosity on the emerging phenomena and see IS as a more engaged and applied research area, somewhat more akin to law and medicine, for example, in how research can be conducted, what subjects to focus on, and the practical value of the findings (Wainwright et al., 2018). Within IS, several responses to this polarization have emerged. One has been to encourage different genres of research and academic output. A second is to take a contingency approach, advocating that methodologies and approaches be chosen that are suitable for wrestling with the phenomena under purview. A third has been to try to accelerate the speed of research and publication, in order to keep findings abreast with the fastchanging phenomena being investigated. In this chapter, we argue that both rigor and relevance are critical to any IS research project, that different research approaches will be needed depending on a range of factors, not least the embryonic nature or maturity of technologies and speed of adoption, and that, as an applied field, IS needs to position itself to investigate much more how emerging technologies are developed, what can be learned from and for practice, and what the impacts of these technologies are on individuals, organizations, and society. We are not, as a field, it is suggested, well-placed to do this. In what follows we discuss the practice-research gap in IS and assess the role of the genres debate in progressing the relevance of IS research. We make the case for a greater focus in IS on what we call

3  Debating Genres and IS Research… 

81

research-into-­practice. We then provide a demonstration of research-intopractice by detailing our study of automation in contemporary workplaces, spelling out the research methodology adopted, and the results achieved. We conclude by inviting the IS field to respond to the challenges we have presented. The motivation of this chapter is to question the suitability of the IS field’s pattern of inquiry for robotics and ‘AI’. The chapter argues how existing genres fail to capture the core concerns of robotics and ‘AI’. Theorizing using questions often lead to many other questions as the chapter does many times. Theorizing using questions also helped to formulate the research strategy. Emanating from the research approach and findings, the use of questions in theorizing leads to a series of ‘action principles’ as forms of statements. Thus, this chapter focuses on Question and Statement products of theorizing as listed in Chap. 2 (Table 2.1).

Genres and the Practice-Research Gap One purpose of this chapter is to address questions and concerns that have become widespread regarding the gap between IS theory/methodology and IS practice. This forms part of an even larger picture, with many suggesting that academic studies not just in IS but also in related areas such as economics, business, and management frequently have little relevance to practice (see, e.g., Beer, 2001; Behrman & Levin, 1984; Bennis & O’Toole, 2005; Beyer, 1982; Gibbons et al., 1994; Kieser et al., 2015; Koskela, 2017). It is not just that the owl of Minerva flies at dusk; all too frequently, it seems, the owl is reluctant to fly at all. There have been calls for more practice-based research approaches, in particular Action Research, but practice-­focused papers remain underrepresented in the AIS Basket of Eight Journals. (The MISQ 2004 Special Issue on Action Research is a notable exception—see Baskerville & Myers, 2004.) Despite the mandatory sections headed Implication For Practice that appear in leading journals, it is unusual for academics publishing there to provide deep insights and helpful and timely practices to address unfolding contemporary events.

82 

L. Willcocks et al.

 he Global Financial Crisis: The Owl That Did Not T Really Fly The problem is not restricted to IS. That theories published in IS, and related management and economics journals, do not provide timely and actionable insight into real-world problems is well illustrated by the criticisms (from and toward scholars) relating to the 2008 Global Financial Crisis. There have been concerns articulated over the absence of research published in top ranked management journals regarding the Global Financial Crisis (Tourish, 2015) and correspondingly why the crisis was largely ignored by organizational theorists until 2011 (Munir, 2011). Other related critiques focused on how scientifically developed economic models failed to predict the crisis (Colander et al., 2014) or even that business schools, and by intimation the teachings of their constituent scholars, produced individuals ethically and materially disconnected from the consequences of their actions, which directly contributed to the crisis (Ghoshal, 2005; Palin, 2016; Podolny et al., 2009). A ubiquitous challenge faced by academics working in professional schools (e.g., business, law, medicine, engineering) is how to imbue and strengthen the relationship between scientific and practice-based knowledge. This worthy aim is often used to justify the existence of such schools and yet studies which build knowledge that is simultaneously insightful to both the scientific and practitioner communities are cited as scarce across the different genres of management research (Augier & March, 2011; Hambrick, 1994; Khurana, 2007; McKelvey, 2006; Toffel, 2016; Van de Ven, 2007).

Genres in IS Research: A Solution? Some of this can be put down to scholarly insistence on focusing on rigor, data collection methods, data sets, and theorizing that consciously are not interested in impacting practice. In some academic journals an attempt is made to suggest that the products of such research have implications for practice, but many of these sections do read as an exercise in going through

3  Debating Genres and IS Research… 

83

the stipulated motions. In response, within IS, a diverse range of genres have been published (Galliers & Whitley, 2007; Stein et al., 2016). Established genres include literature reviews, research essays, theory development papers, issues and opinions, responses and empirical research (Avital et al., 2017: Te'eni et al., 2015; Rowe, 2012). Yet, despite the inclusion of an eclectic range of genres, the relevance of these outputs to practice and policy has often been questioned (Hirschheim & Klein, 2003). Within the IS community discussion continues regarding the future of the discipline and the role of philosophy, theory, and scientific research, and often their relationship to practice (Avison & Malaurent, 2014; Compeau & Olivera, 2014; Hirschheim & Klein, 2003; Lee, 1999; Robey & Markus, 1998; Rosemann & Vessey, 2008). For example, Thatcher et al. (2018 p. 192) suggest that: Scholars need to develop a reciprocal relationship with practice whereby they study problems in the discipline and contribute to practice. For example, we should not merely view the organizations that we collect data from as means to an end but a partnership that benefits research and the organization.

Meanwhile related works have focused on understanding how researchers and practitioners can effectively collaborate (Mathiassen & Sandberg, 2013). We sympathize with these positionings.

Roles for Philosophy and Science The application of philosophical concepts to IS studies has been proposed as one approach to guide the future direction of the field toward greater practical insight. It is suggested that through engagement with philosophical ideas, greater degrees of scientific rigor are possible, which allow for richer insight and practical relevance through critically identifying and developing actionable recommendations (Hassan et al., 2018; Lee, 2004; Rowe, 2018). Views on the importance of scientific approaches can be viewed on a continuum. One end of the spectrum advocates the traditional scientific approach which underlines the interdisciplinary,

84 

L. Willcocks et al.

replicable, and predictive value of some IS studies (Dennis et al., 2018; Thatcher et al., 2018) and the relationship between such studies and well-­ cited views of what constitutes a scientific approach (Alter, 2018; Gregor, 2018). At the other end of the spectrum, questions are raised as to whether IS should be considered a science at all, with a related school of thought suggesting that IS “has stagnated and requires nothing less than a remobilization”.

Differing Genres in IS Research: Another Way Forward? A further suggestion for how to address this perceived malaise is to introduce genres of IS studies outside of the scientific process which focus on story-telling (Daft, 1983), more varied modes of expression, new forms of conceptualization and methods, and crucially renewed engagement with practice. Newer genres of IS research have been proposed, which aim to offer new directions for the field and novel contributions beyond the academic article genre (Avital et al., 2017). Examples of new genres include those that emphasize felt experience over what is directly observable (Bødker, 2017), personal experiences recorded in the form of memoirs (Prasopoulou, 2017), fictional tales and conversations (Boland Jr & Lyytinen, 2017; Kaarst-Brown, 2017), new models of knowledge production through crowdsourcing (Love & Hirschheim, 2017), and offering open interpretation of cases through narratives (Avison et al., 2017). Indeed, ethnographic studies often seek to build narratives around what has been experienced or lived and so by their nature require closer engagement with practice with some researchers adopting the duel yet complementary roles of consultant and researcher (Rowe, 2012). Meanwhile, some have argued that theory development is the main purpose of academia and is what fundamentally distinguishes IS scholars from practitioners and consultants (Gregor, 2006). We do not argue against any of these positions, but we do propose that IS as a field needs to progress much further in its ability to carry out research work of, in, and applicable to practice.

3  Debating Genres and IS Research… 

85

The Need for More Engagement with Practice Why do we argue this? Because despite the inclusion of scientific methods, ethnographies, narratives, and philosophy in IS journals, many scholars continue to express concern around the stagnation of the field and propose developing closer engagement with practice (Galliers & Stein, 2018; Hirschheim & Klein, 2003; Rowe, 2012). Mathiassen (2017, p. 1) describes engaged scholarship as a “participatory form of research for studying complex real-world problems based on the different perspectives and understandings of key stakeholders”. The aim of such action-based research is twofold—to contribute practically to challenges faced by industry and policy makers while simultaneously providing new theoretical understandings (Mathiassen et al., 2012). The widening gap between theory and practice has been often discussed in terms of dissemination challenges with blame pointed in both the direction of managers, for not engaging with relevant theory and attempting to apply it, as well as at scholars, for not construing abstract knowledge into actionable forms of practice (Hodgkinson, 2001; Weick, 2001). However, focusing on dissemination problems may be ineffectual where the research itself holds no relevance or where “the wrong questions have been asked” (Pettigrew, 2001; pp.61–67). Then there is the different kinds of reasoning and knowledge argument. Thus, some philosophical perspectives on practice and theoretical reasoning characterize the role of practical and theoretical knowledge as being distinct. While practical reasoning is concerned with evaluating the value and attractiveness of actions, theoretical reasoning, in contrast, focuses on facts and explanation to understand why events have occurred and evaluate the truthfulness of related propositions. “Theoretical reasoning leads to modifications of our beliefs, whereas practical reasoning leads to modifications of our intentions” (Harman, 1986, p. 77). A related school of thought in management studies positions scientific research and practical knowledge as also quite distinct, with only occasional and limited areas of overlap. One is not necessarily derived from the other and so they often differ in the contributions offered. Consequently, the aim of academic research is, it is suggested, to advance scientific

86 

L. Willcocks et al.

knowledge through theoretical generalizations that may at some point in time become useful to practice, while practical knowledge is often contingent to discrete professional roles, experiences, situations, and is time limited (Van de Ven, 2007; Van de Ven & Johnson, 2006). A critique of this view is that it positions practitioner knowledge as largely independent and understates the multilateral flow of knowledge between academics and practitioners in a discipline where research and knowledge creation usually begin by studying practice (McKelvey, 2006).

The Theoretical Turn: What Kind of Theory? The interplay between theoretical and practical reasoning and the bilateral or unilateral relationship between the two forms of knowledge raise questions regarding the application of theory in IS studies. It has been observed how IS work often references ‘grand’ social theories from other disciplines such as economics, psychology, and sociology to create ‘mid-­range’ theories. As Grover and Lyytinen (2015 p. 272) suggest: During this process, highly stylized theoretical constructs from the reference theory typically get instantiated directly into the IS context without significant modification and extension (Colquitt & Zapata-Phelan, 2007). The key contribution of IS scholars in this process is to combine reference theory with some atheoretic, generic accounts of technology, represented as an investment or a perception or technology that is assumed to reside in the background.

Mid-range theories may then often fall between theory and practice as they do not offer high degrees of abstraction and generalizability, nor do they offer the detail and context-specific nuances required for a contribution to practice. Grover and Lyytinen (2015) further observe how the development of mid-range theory has become the dominant epistemic script in IS publishing, at a time when many scholars lament the gulf between theory and practice. One suggested solution is to expand upon the scripts or genres acceptable to IS journals and conferences and so allow for other pathways to knowledge production. Again, we have sympathy with this proposal.

3  Debating Genres and IS Research… 

87

Indeed, possible problems arising from the emphasis on theory include reverting to ideal types and scripts, distortion of the research setting, poor fit, and applicability of the reference theory leading to triviality (Avison & Malaurent, 2014). A suggested response is for journals to include a genre of research which may emphasize contributions other than theory, termed ‘theory light’ papers. Such work may be judged on criterion including timeliness, originality, methodological rigor, authenticity, and criticality. These papers “may contain some reference to established theories or theory construction, they would not pass the ‘acid test’ that it significantly uses or develops theory. It should make other significant contributions to IS research and practice” (Avison & Malaurent, 2014, p. 330). Correspondingly, papers which address “theories of the solution” have been advocated for where arguments are grounded in a well-­developed and articulated problems and the contribution lies in offering related solutions (Markus, 2014). Once again, this is consistent with our own positioning in this chapter. To summarize, if one acknowledges that practical and theoretical reasoning are distinct forms of knowledge, then it follows one might query the role of theory in studies aimed to contribute to practice-based knowledge, particularly where the phenomenon under consideration is emergent technologies and the changing business context in which they are employed. In this chapter, we respond to calls for ‘theory light’ research-into-practice papers and offer an example of this under-exposed genre of IS research. We take as the example our three-year on-going research into service automation. The object here is not to provide a fully worked paper but provide a demonstration of research approach and findings to feed into the debate on future directions for IS research and how it is presented in scholarly IS journals. We next explain the service automation landscape so that readers understand the subject of our research-into-practice example.

 nacting Research-into-Practice: The Case E of Service Automation Using software to automate tasks is not a new idea, but recent interest in service automation has certainly escalated in recent years. Provocative titles like Rise of the Robots: Technology and the Threat of a Jobless Future

88 

L. Willcocks et al.

(Ford, 2015), A World Without Work (Thompson, 2015), Rise of the Robots: How long do we have until they take our jobs? (Devlin, 2015), and The Globotics Upheaval (Baldwin, 2019) continue to fill the popular press. Although the term ‘robot’ conjures images of physical robots wandering around offices, the term as it relates to service automation really means the delivery by software of service tasks previously performed by humans (Willcocks & Lacity, 2016). We have been studying early organizational adopters of service automation technologies since 2014. The context has been complex, ambiguous, new, and uncertain—the type of contexts well suited for research-into-practice. At the beginning of our research investigation, we were bombarded by the proliferation of technical jargon, with names such as macros, scripting tools, robotic process automation (RPA), cognitive intelligence, machine intelligence, artificial intelligence, virtual agents, intelligence amplification, automatic content generators, cognitive learning technologies, autonomic platforms, cognitive computing, and business process management (BPM) systems as some common examples. One of our research outputs was to create a simple service automation continuum (Lacity & Willcocks, 2016b); it will assist readers to introduce it now.

Making Sense of the Robo-Babble We asked a very straightforward question to make sense of the proliferation of service automation jargon: What types of tasks are the tools designed to automate or augment? Looking at the data, process, and outcomes, we conceived of service automation comprising a continuum (see Fig. 3.1). Anchored on one side is the realm of RPA; the other, cognitive automation (CA). (We do not call CA technologies ‘Artificial Intelligence’ because the ‘AI’ label aggrandizes what these tools do in our opinion—see Lacity & Willcocks, 2018a, 2018b for a discussion.) The realm of RPA consists of tools that aim to automate tasks that have clearly defined rules to process structured data to produce deterministic outcomes. So what’s new here? A software ‘robot’ is configured to do computer tasks the way humans do, by giving it a logon ID, password, and

3  Debating Genres and IS Research… 

89

Fig. 3.1  Service automation continuum (Lacity & Willcocks, 2017)

playbook for executing processes. For example, an RPA tool can perform a credit check needed by an internal Enrerpriuse Resource Planning (ERP) system by logging on to a credit bureau website, entering the search terms, and recording results back to the ERP system. RPA has two distinctive features compared to other automation tools like BPM solutions and Software Development Kits (SDKs). First, RPA is easy to configure with drag-and-drop process design, so developers don’t need programming skills, just process, and subject matter expertise. Second, RPA software is non-invasive because it accesses other systems through the presentation layer—no underlying systems are touched (Willcocks & Lacity, 2016). Automation Anywhere, Blue Prism, and UiPath are the top RPA tools by market share as of 2019. The realm of CA consists of tools that aim to automate or augment tasks that do not have clearly defined rules. With CA technologies, inference-­based algorithms process data to produce probabilistic outcomes. The data is often unstructured, such as natural language, either written or spoken. For example, a CA tool may be used as a first-responder dispatcher to categorize a customer request. The tool will be instructed to route the request only if a certain confidence threshold has been met, such as the tool is 90 percent confident the call pertains to a new mortgage application. The data can also be highly structured, such as the pixels in an image. Whether the data is unstructured to structured, CA tools require a significant amount of training data (Lacity & Willcocks, 2018a). For

90 

L. Willcocks et al.

example, Andrew Ng (2016), Founder of Google’s Brain Deep Learning Project, wrote that it takes tens of thousands of labeled pictures to recognize people. Some of these tools also claim to have emotional intelligence capabilities, the ability to assess another human being’s sentiment or state of arousal. CA tools require a significant amount of data analytics, business process, and IT skills to train them to proficiency. IBM’s Watson suite, Google’s Machine Learning Kit, IPsoft’s Amelia, and Expert Systems’ Cogito are examples of CA tools. Given the newness of these service automation technologies, there are many unanswered questions. We conducted empirical research to answer two high-level questions: 1. Why are clients adopting service automation and what outcomes are they achieving? 2. What practices distinguish service automation outcomes? Our research results have been published in a number of books, practitioner journals, white papers, and working papers (see Hindle et al., 2018a, 2018b; Lacity et al., 2017a, 2017b, 2017c; Lacity & Willcocks, 2016a, 2016b, 2016c; Lacity & Willcocks, 2017; Lacity & Willcocks, 2018a, 2018b; Lacity et al., 2015; Scheepers et al., 2018; Willcocks & Lacity, 2015a, 2015b, 2015c, 2015d; Willcocks & Lacity, 2016). Our purpose here is to provide an example of our research-into-practice method. We reveal the process as emergent and naturalistic (Lincoln & Guba, 1985). Our account is purposefully confessional and self-reflective (Van Maanen, 1995). We show how the process of inquiry produced what we call ‘action principles’. Action principles are practices that explain the results found in real-world implementations. Action principles are cocreated with practitioners through the process of inquiry to articulate, understand, and provide meaning for associating actions with outcomes within a particular organizational context. They are lessons learned from practices enacted in the contexts studied. To illustrate the method, we first share why we became interested in the topic of service automation. Then we describe our data collection methods and how ‘action principles’, which are both rigorous and relevant, arise from these methods.

3  Debating Genres and IS Research… 

91

Topic Selection: “Is there any there there?” In late 2014, we were just completing a four-year research project on business process outsourcing (BPO), when the consulting firm HfS started talking about robotic process automation (RPA) as a new breed of tools designed to automate back-office services (HfS Research, 2014). Practitioners were also talking about IBM’s Watson use in healthcare1 and a plethora of other new tools designed to assist or replace human knowledge workers. Our curiosity was piqued, but we were not convinced that “there was any there there”—to paraphrase Gertrude Stein (1937). To investigate, one author went to the launch of the Institute for Robotic Process Automation (IRPA) in New York City in late 2014. She met some early client adopters of RPA, including A.J. Hanna of Ascension Shared Services and Lou Ferrara of the Associated Press. Ascension was using RPA to update HR records and to process payments. The Associated Press was making headlines for using a tool to generate news stories. She also met RPA tool providers, including Pat Geary, the Chief Marketing Officer (CMO) of Blue Prism. After that event, we decided to devote our time to study the space. There was no academic literature to search because none existed. Instead, we entered a messy world of provider hype with no compass.

Planning Field Work After the IRPA event in 2014, we formulated our research plan. We needed key participants at the front lines of service automation adoption to answer the research questions. Key participant interviews were an appropriate method because we sought answers to questions in which the subject matter was sensitive (like any form of automation) and because we were more concerned with the quality, not quantity, of responses  “WellPoint and IBM Announce Agreement to Put Watson to Work in Health Care,” Press release, 09/12/2011; “In year two, MD Anderson Moon Shots Program begins to spin off innovation,” MD Anderson News Release 10/30/14; “Memorial Sloan Kettering Trains IBM Watson to Help Doctors Make Better Cancer Treatment Choices,” April 11, 2014; “Cleveland Clinic uses IBM’s Watson in the cloud to fight cancer.” Computerworld UK, 10/29/2014. 1

92 

L. Willcocks et al.

(Elmendorf & Luloff, 2006; Fontana & Frey, 1994; Seidler, 1974). We planned to start our interviews with a very simple request: Tell us your automation story from your perspective. To fill in gaps in participants’ narratives, we intended to prompt them to discuss their service automation adoption journey in detail, including the aims of the automation, practices enacted, outcomes, and lessons learned. The specific questions planned were: • Client adoption journey: Why was automation considered? What was your initial business case? Who was involved? Did you do a proof-of-­ concept, and if so, when and on what process? How was the process selected? What tool did you pick and why? How did you manage the implementation? • Business value delivered: How has service automation delivered on the initial business case in terms of financial (i.e., cost savings, return on investment [ROI]), operational (i.e., improved quality, faster delivery, better compliance), and strategic value (i.e., strategy enablement, access to new customers, better customer retention)? • Lessons learned: What overall lessons did you learn? If you had to do your service automation implementation all over again, what three things would you change? Why? We also planned to interview the service automation providers at each client site, as well as any advisory firms that assisted with the client’s adoption journey.

Field Research Begins Beginning in January 2015, we started formally interviewing contacts made through the IRPA. Ascension Shared Services and the Associated Press were the first case studies, both based in the United States (US). Immediately, our interviews were yielding surprising findings. Neither organization used RPA to fire employees as suggested by the popular press. Instead, the RPA tools were performing mundane tasks so that employees could focus on more value-added work. A journalist, it turns out, would

3  Debating Genres and IS Research… 

93

rather interview a CEO than write a corporate earnings report. An accounts payable associate would rather reconcile records than copy and paste fields from a spreadsheet into an ERP system. Both organizations were under pressure to do more work without adding more headcount, and RPA was a way to achieve it (Willcocks & Lacity, 2016). Meanwhile, the CMO of Blue Prism provided introductions to the next two case studies at Telefónica O2 and Xchanging in the United Kingdom (UK). Based on just four cases, we were beginning to see patterns across the cases. So far, none of the four companies laid off employees as a consequence of automation and all reported multiple business benefits besides just cost savings. Interesting differences arose as well: The business sponsors at Ascension and Telefónica O2 excluded the IT department, whereas the business sponsors at the Associated Press and Xchanging included IT. There were also unique practices. For example, employees at Xchanging embraced RPA to the point where they anthropomorphized their software robots by giving them names, depicting them, inviting them to office parties, and holding contests for the ‘coolest robot’ (Willcocks & Lacity, 2015c, 2016). For each case, we (as researchers) flush out the action principles in the form of lessons learned and review these with interviewees until we reach a common understanding. Thus, action principles are co-produced among researchers and participants. We typically generate between five and ten action principles at each organization, sometimes based on a single interview with a key participant. For example, we initially identified the following action principles for the Telefónica O2 case (Lacity & Willcocks, 2016c; Willcocks & Lacity, 2016): 1. Test RPA capabilities with a controlled experiment by asking multiple RPA tool providers to compete on an identical proof-of-concept. 2. Develop criteria for determining which processes can be automated from a business and technical perspective. 3. Let business operations (BOs) lead. 4. Include multiple expected benefits in business cases for service automation. 5. Communicate the intended effect on jobs early in the process.

94 

L. Willcocks et al.

6. Do-it-yourself (Telefónica O2 insourced automation at the time because there were no consultants specializing in RPA back in 2010 when it first adopted RPA). 7. Continually improve the automated processes. With each subsequent case, we built a matrix of action principles across contexts. Table 3.1 provides an early version of the action principles extracted from the first four adoption journeys. The table comes from Lacity et al. (2021). One will note that Telefónica O2’s lesson “let business operations lead” has evolved to “let business operations lead, but bring IT onboard early”. Participants from Telefónica O2 and Ascension Shared Services finally reversed their decisions not to include IT. The business sponsor at Telefónica O2 who implemented RPA without informing the Chief Information Officer (CIO) was nearly fired for giving the software robots his personal ID and password. Ascension’s service automation projects were scaling so fast that they needed IT’s input. While interviews on client adoptions continued, we also collected survey data.

Survey 1 For the past two decades, we attend the International Association of Outsourcing Professionals (IAOP)’s Outsourcing World Summit (OWS). We collect data each year in the form of a brief survey during the client-­ only and provider/advisor-only networking sessions. In 2015, we made service automation the topic. The surveys assessed the maturity of service automation adoption, the drivers of service automation adoption, the perceived automatability of existing business services, and the preferred sourcing option. We collected 143 completed surveys, filled out by 63 clients, 64 providers, and 16 advisors (i.e., consultants). The client respondents were senior leaders in charge of sourcing strategy, governance, procurement, and provider management. Client respondents represented organizations from a variety of industries including financial services, software, technology, engineering services, manufacturing, aerospace, pharmaceuticals, life

3  Debating Genres and IS Research… 

95

Table 3.1  Action principles after the first round of data collection

Site: Number of interviews: 1. Use RPA to automate mundane tasks (action) to focus employees on more value-added work (outcome) 2. Gain C-suite support (action) to legitimate and support the RPA initiative (outcome) 3. Use a controlled experiment to assess tools (action) to select the tool that delivers the best financial value (outcome) 4. Include multiple expected benefits in business cases for service automation (action) to achieve the greatest value for the organization, employees, and customers (outcome) 5. Develop comprehensive criteria to select the best processes to automate (action) to achieve the greatest value for the organization, employees, and customers (outcome) 6. Let business operations lead and do not involve IT in the RPA initiative (action) to speed implementation (outcome) 7. Let business operations (BO) lead and involve IT early (action) to ensure long-term operational success (outcome)

Ascension Shared Services

Associated Telefónica Press O2 Xchanging

3 3

1 1

2 2

4 4

3

1

2

4

2

3

1

2

4

3

1

2

4

2

1

1

1

1

4

(continued)

96 

L. Willcocks et al.

Table 3.1 (continued)

Site:

Ascension Shared Services

8. Communicate the intended 3 effect on jobs early in the process (action) to obtain employee buy-in and to prevent panic and sabotage (outcomes) 2 9. Do-it-yourself rather than hire outsiders (action) to capture the most financial benefits (outcomes) 3 10. Continually improve the automated processes (action) to extract more business value (outcome) 11. Fix discoveries about process flaws before deploying RPA (action) to prevent merely performing a bad process more efficiently (outcome) 12. Reuse components (action) to scale quickly and to reduce development costs (outcomes) 13. Multi-skill the robots (action) to extract more financial value (outcome)

Associated Telefónica Press O2 Xchanging 1

2

4

1

2

2

4

2

2

2

sciences, healthcare, and other industries. Provider and advisor respondents represented organizations of varying sizes and geographic locations. Our first survey found that: • Client organizations were increasingly expecting their outsourcing providers to help automate services. The do-it-yourself model was not their preferred route, which differed from our first four case studies. • Client, provider, and advisor communities believed that service automation tools could decrease costs AND improve service quality. In

3  Debating Genres and IS Research… 

97

other words, multiple business benefits were expected, consistent with the first four cases. We published the results in the IAOP’s publication, Pulse Magazine (Lacity et al., 2015). From the survey, the action principles table was updated (not shown in this chapter) to reflect that ‘do-it-yourself ’ was not the only viable sourcing model. By now, readers see the iterative process of creating action principles. As more data is collected, the list of action principles is revisited, revised, and reviewed by the researchers.

Additional Field Research During 2015–2017, we conducted more interviews with clients, their service automation providers, and advisors to build a total of 22 detailed client adoption journeys (see Table 3.2). It is useful to establish that the 22 organizations were all ‘successful’ adopters, since this relates to the relevance of the action principles we derived. We have permission to name 11 of the client organizations. We assigned pseudonyms to the other client organizations (indicated by an asterisk in Table 3.2). Eight of the client organizations are headquartered in the UK, five are based in the US, two in Germany, and one client organization is based in each of the following countries: Australia, France, Netherlands, South Africa, Sweden, Switzerland, and Russia. The client organizations represent 14 industries, illustrating that service automation is not restricted to certain sectors. Among our client adoption journeys, 17 adoptions were led by business operations, 4 were led by people from IT departments, and 1 was led by an innovation center. Fifteen clients adopted automation technologies that fall within the realm of RPA, three fall within the realm of CA, and two used both. The 22 client organizations adopted service automation tools/platforms from Automation Anywhere (n=1), Automated Insights (n=1), Blue Prism (n=13), Celaton (n=1), Expert Systems (1), IBM Watson Services (n = 3), IPsoft Amelia (n=1), and Redwood (n=1). Five advisor organizations are represented in the study: The Everest Group, KPMG, HfS, Alsbridge (since bought by ISG), and Information Services Group (ISG). In addition to these empirical methods, several providers

Netherlands UK

Biotechnology Financial Services

Professional Services

Higher Education Natural Gas Financial Services Financial Services

Financial Services Healthcare Professional and IT services Insurance Services Professional Services

Consumer Goods Telecommunications

6. Consulting*

7. Deakin University 8. Energy* 9. Financial Services* 10. Financial services*

11. Financial services* 12. Healthcare* 13. IBM Division

16. Manufacturing* 17. Telefónica O2

14. Insurance* 15. KPMG

US US

Media Healthcare Insurance

Germany UK

Sweden UK UK/ Netherlands UK US

Australia Russia UK South Africa

France

US

IT Innovation Center BO BO

IT BO IT

BO BO BO BO

BO

BO BO

BO BO

BO

Adoption Head-quarters location

Healthcare

Industry

1. Ascension Shared Services 2. Associated Press 3. Blue Cross Blue Shield—North Carolina 4. Biotech* 5. Building Society*

Company name or pseudonym*

Table 3.2  Client adoption journeys

RPA RPA

CA RPA RPA RPA/ CA CA RPA RPA/ CA RPA CA

RPA

RPA RPA

RPA RPA

RPA

Pension enrollment Commercial loan grading; business development Back-office processing SIM swaps; pre-calculated credit

IT services desk Patient registration Help desk ticketing

Financial close Mortgage lending and savings Still considering pilot options Student engagement New customer registration Payroll verification Consumer credit

Corporate earnings reports Claims processing

Employee record updates

Realm First processes automated

98  L. Willcocks et al.

Electric and Gas

Healthcare

Transportation

BPO provider Insurance

18. Utility*

19. VHA

20. Virgin Trains

21. Xchanging 22. Zurich Insurance

BO Business Operations, IT IT Department

Industry

Company name or pseudonym*

UK Switzerland

UK

US

Germany

BO BO

BO

IT

BO

Adoption Head-quarters location

RPA CA

CA

RPA

RPA

Meter reading feasibility checks Web crawls for product descriptions Incoming customer correspondence Premium advice notices Personal injury claims

Realm First processes automated

3  Debating Genres and IS Research… 

99

100 

L. Willcocks et al.

gave product demonstrations and two authors completed an RPA foundations course to assess the claims about ease of use. In 2017 and 2018, we also conducted multiple interviews on failures, which were largely informed by the providers and advisors. (Few clients wished to discuss failures, but some clients shared stories of first failed attempts before a successful relaunch.) We asked providers and advisors to diagnose the practices which lead to failure. We documented these as ‘risks’ in Lacity and Willcocks (2017). We were now not only able to assess that successful outcomes were associated with applying action principles, but we could now show that failures often were associated with not applying action principles.

Additional Surveys After the first survey at the IAOP OWS in 2015, we collected surveys at the OWS in 2016, 2017, and 2018. The surveys are not very rich, but they do indicate trends, at least for the IAOP community. The surveys showed increasing rates of RPA and CA adoption in client and provider firms, increased evidence for multiple business benefits expected and delivered, and varied approaches to service automation sourcing (do-it-­yourself, rely on the provider, engage an advisor, buy as a service) (Lacity et al., 2015; Lacity & Willcocks, 2016a, 2016b, 2017, 2018b). In 2018, we decided to assess the action principles with a more thorough survey of client adopters. Partnering with Dr. John Hindle of Knowledge Capital Partners (KCP) and Dr. Shaji Khan from the University of Missouri, we developed a survey with 48 detailed questions based on our action principles. We received 764 responses, but most clients were not far enough in their adoption journeys to report outcomes and dropped out of the survey midway. We did get 112 completed surveys that reported on 238 RPA deployments. Once again, clients reported multiple business outcomes, including positive returns on investment, increased business agility and compliance, and enhanced customer experiences. Interestingly, this survey contradicted the IAOP survey in one regard: Relying on a traditional BPO provider to lead the service automation was ranked the

3  Debating Genres and IS Research… 

101

lowest in terms of business value delivered in the KCP survey; it was rated among the best sourcing approaches in the IAOP surveys. We divided the KCP results over five reports because we discovered so many complex relationships between action principles and outcomes (Hindle et al., 2018a, 2018b). In summary, we gathered 22 detailed successful client adoption journeys, interviews into dozens of failures, and five surveys.

Service Automation Action Principles Table 3.3 shows a more recent iteration of the 39 action principles gleaned from all the data. Compared to Table 3.1, Table 3.3 documents 29 additional principles. One will also note that a structure emerged; we organized the action principles along the adoption journey phases of Strategy, Sourcing Selection, Program Management, Tool selection, Design and Build, Run, and Maturity. We also have “General” action principles category (Lacity & Willcocks, 2018a). By comparing Tables 3.1 and 3.3, one can also see how action principles evolve with more data collection. For example, we initially identified the action principle ‘let business operations (BO) lead’ in Table 3.1. That action principle was changed to ‘let business operations (BO) lead, but bring in IT early’ based on subsequent data. With the next round of data, four new client adoption journeys (IBM, VHA, a financial services firm, and an insurance company) were successfully led by IT. Two of the IT-led cases were automating IT processes (help desk ticketing), not BO processes, which makes sense for IT to lead. Furthermore, the KCP survey did not find a statistically significant relationship between the RPA champion (IT, BO, or other) and business outcomes. This additional evidence led us to alter the action principle to ‘Decide who is best to “own” the automation program’ in Table 3.3. Our books discuss the nuances and trade-offs of each approach (Lacity & Willcocks, 2017, 2018a), while our Journal of Information Technology (JIT) article provides a more mature and definitive version of our methods, terminologies, and results (Lacity et al., 2021).

102 

L. Willcocks et al.

Table 3.3  Action principles after 22 adoption journeys data Relevant to: Strategy

Sourcing

1. You get what you pay for: Focus on long-term value rather than short-term ROI (action) to gain the most value from service automation tools (outcome) 2. Strategy drives investments: Include multiple expected benefits in the justification for service automation investments (action) to achieve the greatest value for the organization, employees, and customers (outcome) 3. Consider competitors’ reactions (action) to prevent mis-messaging to customers (outcome) 4. Use RPA as forward reconnaissance for CA (action) to gradually build automation skills and to use RPA savings to defer some CA costs (outcome) 5. Strategy envisions longer-term human workforce needs (action) and develops automation and HR plans to gradually meet that vision (outcome) 6. Select the best sourcing option (action) to ensure the success of the implementation (outcome). 7. If sourcing with a BPO or tool provider, incentivize the provider to share the benefits of automation (action) to prevent them from taking all the savings (outcome)

RPA

CA

Robust or Distinctive

√

√

R

√

√

R

√

D

√

√

D

√

√

D

√

√

R

√

√

R

(continued)

3  Debating Genres and IS Research… 

103

Table 3.3 (continued) Relevant to: Program management

Process selection

RPA 8. Manage RPA as a traditional business case (action) to increase returns on investment (outcome) 9. Manage cognitive automation as an innovation program (action) to increase experimentation and to rapidly shift direction (outcome) 10. Manage cognitive automation as a learning project (action) to adapt quickly to early lessons learned (outcome) 11. Select the organizational unit that is best suited to own the automation program (action) to achieve the desired priority (speed; longevity) (outcome) 12. Find the ‘Lewis and Clark’ program champions (action) who will overcome obstacles to ensure project implementation (outcome) 13. Take the robot out of the human: Use service automation tools to automate mundane tasks (action) to focus employees on more value-added work (outcome) 14. Aim for the triple win: Develop comprehensive criteria to identify the best processes to automate (action) to achieve the greatest value for the organization, employees, and customers (outcome) 15. Fix discoveries about process flaws before deploying service automation (action) to prevent merely performing a bad process more efficiently (outcome)

CA

√

Robust or Distinctive R

√

R

√

D

√

√

R

√

√

R

√

√

D

√

√

D

√

√

R

(continued)

104 

L. Willcocks et al.

Table 3.3 (continued) Relevant to: Tool selection

Stakeholder buy-in

RPA 16. Use a controlled experiment to assess tools (action) to select the tool that delivers the best financial value (outcome) 17. Don’t look for a Swiss Army Knife: Select a tool that does a few things well (action) to ensure technical success (outcome) 18. Negotiate the optimal level of client-provider transparency pertaining to machine learning algorithms (action) to ensure a good relationship (outcome) 19. Expect technical challenges as a first mover (action) to minimize disappointments (outcome) 20. Manage up: Gain C-suite support (action) to legitimate, support, and provide resources for the service automation initiative (outcome) 21. Manage down: Communicate the intended effect on jobs early in the process (action) to obtain employee buy-in and to prevent panic and sabotage (outcomes) 22. Report financial savings in terms of ‘hours back to the business’ (rather than Full Time Equivalent (FTE) savings) (action) to reinforce that automation is used to liberate employees from routine tasks (outcome) 23. Manage expectations out: Be transparent with customers that they are interacting with automation tools (action) to maintain high ethical standards and acceptance (outcome)

CA

√

√

Robust or Distinctive D

√

R

√

D

√

√

R

√

√

R

√

√

R

√

√

D

√

√

R

(continued)

3  Debating Genres and IS Research… 

105

Table 3.3 (continued) Relevant to: Design, build, and test

RPA √ 24. Pareto’s rule: Automate the smallest percentage of tasks that account for the greatest volume of transactions (action) to deliver the most business value (outcome) 25. Don’t underestimate the data challenge (action) required to get cognitive automation tools to perform competently (outcome) 26. Find new data sources if ‘dirty’ data cannot be cleaned (action) to get the cognitive automation tool to perform competently (outcome) 27. Compare training the automation √ tool to training a new employee (action) so that stakeholders expect the tools to be as competent as new employees and not as competent as experts (outcome)

CA

Robust or Distinctive

√

R

√

D

√

D

√

D

(continued)

Table 3.3 (continued) Relevant to: Run

Maturity

28. Redesign employee scorecards so that employees are credited with productivity gains contributed by their robot teammates (action) 29. Invite customers to experiment with the automated service but keep other channels open (action) to ensure good customer service (outcome) 30. Invite customers to provide feedback (action) to help improve the performance of a cognitive automation tool (outcome) 31. Keep subject matter experts continually engaged in data curation (action) to keep the automation relevant (outcome) 32. Robots need supervisors: Supervise the learning of the service automation tool (action) to prevent machines from making new decisions without human direction and approval (outcome) 33. Rethink human talent and skills (action) needed for long-term success (outcomes) 34. Assign clear roles of responsibility (action) to keep the automation operational and relevant (outcome) 35. Reuse components (action) to scale quickly and to reduce development costs (outcomes) 36. Multi-skill the robots (action) to extract more business value (outcome) 37. Create a Center of Excellence (CoE) (action) to disseminate the technology across the organization (outcome) 38. Integrate RPA and cognitive initiatives (action) to deliver end-toend service automation (action) 39. Continually innovate (action) to deliver value to customers, employees, and shareholders (outcome)

RPA

CA

Robust or Distinctive

√

√

D

√

D

√

R

√

D

√

R

√

√

R

√

√

R

√

√

R

√

D

√

√

R

√

√

D

√

√

R

3  Debating Genres and IS Research… 

107

Discussion Our research-into-practice approach produces action principles, which are practices that explain the results found in real-world implementations. Action principles are anchored in data, reviewed by participants, and offered for consideration by other practitioners for when they embark on their own adoption journeys. However, action principles are not ‘best practices’. Whereas ‘best practices’ imply that mimicry will always produce similar results—that ‘one-size-fits-all’—action principles recognize that context matters. We view practitioners as thoughtful agents that are best to decide whether an action principle would be effective in their own organization. In Lacity and Willcocks (2018a), we explained that the usefulness of a practice depends on: • the objectives the organization is trying to achieve; • the organization’s unique context; • whether the organization has the retained capability to implement the practice effectively; and • timing—there are good and less good moments to apply a specific practice. We have argued that action principles are both ‘rigorous’ and ‘relevant’. They also evolve in the face of new evidence.

Rigor of Action Principles We assess rigor in terms of the quality of the research process (Given, 2008). A unique action principle found within a single organizational setting can be as rigorous as an action principle found across many contexts. In this respect, we find it useful to distinguish between ‘robustness’ (frequency across contexts) and the ‘provocativeness’ of the finding. Robustness As evidence accumulates, action principles become more ‘robust’ if the practice holds up over multiple contexts. For example, the action principle, “Strategy drives SA investments: aim for multiple business benefits”, was evident in all cases in Table 3.3. Across all the successful

108 

L. Willcocks et al.

cases, we ended up documenting multiple sources of value delivered from service automation projects across shareholders, customers, and employees. We call it the ‘triple-win of value’, which was only achieved when client organizations conceived of service automation within a broad business strategy. Successful organizations aimed to redesign work to be more efficient and effective while delivering better customer services. From our ‘failures’ research, we found that clients who viewed service automation as a quick way to terminate headcount failed to deliver even that. Aggressive cost cuts through employee termination resulted in insufficient resourcing for the service automation project and non-­cooperation—even sabotage—by employees. However, robustness is not the only requirement for practitioner impact. Sometimes it’s the unusual practice that resonates with practice. Provocativeness  Daft (1983) argued that the quality of research should be measured by the intensity of the surprise in the findings. ‘Managing as a learning project’ was a unique practice found at SEB Bank that certainly meets the provocativeness test. SEB bank did not know what to expect from their deployment of its first cognitive virtual assistant. Rather than do a formal business case and expensive market testing on banking customers’ preferences, it just launched the agent as a live experiment to see how customers would react. (In contrast to SEB, the other cases in our study managed the service automation projects as traditional business projects with clear business cases that documented expected business outcomes.) Nicolas Moch, Head of Information, Strategy & Architecture at SEB, said, “We view it as a learning project where we learn together with the customers and the business. We need the latitude to switch directions as we learn together. If I am tied into a business case of delivering 50 FTE savings by year end, we’ll miss the strategic value” (Lacity & Willcocks, 2018a, p. 243). ‘Consider competitors’ reactions’ is another example of a unique action principle that resonates with practice. Participants from Deakin University thought about how its CA adoption might be perverted by outsiders. Professor Beverley Oliver, Deputy Vice-Chancellor for Education, said: “We were careful about the narrative. I did not want our competitors to twist what we are doing and launch their own campaign, ‘come study here

3  Debating Genres and IS Research… 

109

and talk to a real person because Deakin only lets you speak to a robot’” (Scheepers et al., 2018, p. 100). This action principle resonates with practitioners. Many organizations are skittish about advertising their automation capabilities. With the intense public focus on job creation, the optics of automation can be perceived as ‘job killers’. Demonstrating Rigor  The rigor of action principles generated from the research-into-practice method can be demonstrated through the careful documentation of mapping practices to context (such as Table 3.3) and through the craft of organizational story-telling (Daft, 1983), supported by participant quotations. In this chapter, we have focused on the former as evidence of rigor, but our books are the best source for illustrating rigor through the craft of story-telling (Lacity & Willcocks, 2017, 2018a; Willcocks et al., 2019; Willcocks & Lacity, 2016).  elevance and Limitations of Research-­Into-­Practice  We have demonR strated the emergent and iterative nature of practice-into-­research that produces ‘rigorous’ action principles. What about relevance? Publishing in practitioner journals such as Sloan Management Review (Lacity & Willcocks, 2016b), Harvard Business Review (Lacity & Willcocks, 2015), MISQ Executive (Lacity & Willcocks, 2016c; Scheepers et al., 2018), Pulse Magazine (Lacity et al., 2015, Lacity et al., 2017a, Lacity & Willcocks, 2018b), and Intelligent Sourcing Magazine (Lacity et al., 2017c) is one obvious demonstration. Multiple requests for consulting, speaking, and press interviews provide further evidence that practitioners are consuming one’s research. During 2015–2018, we spoke at over 200 major practitioner and university conferences in Australia, Japan, the United States, New Zealand, China, Europe, and the United Kingdom. While trying to sway readers to conduct more research-into-practice, we recognize that the approach has limitations. Relying on key participant interviews has one major drawback: Informant bias. Multiple interviews with different stakeholders are recommended to compensate for the weaknesses of bias and error (Seidler, 1974). For this research, we interviewed at least three key stakeholders (clients, providers, and advisors often at several levels within the organization); in practice we found their views to

110 

L. Willcocks et al.

be highly consistent or could be synthesized through researcher review relatively easily, but this may not be always the case. Beyond the service automation study, research-into-practice is suited for researchers who feel excited by contexts that are initially complex, ambiguous, and uncertain. Emerging technologies are difficult to study using other research methods, but, in our view, IS really misses out in terms of topicality, relevance, and influence by not finding a way to track and gain insight into digital and related technologies in their emergent rather than at their mature phases, as seems all too often to happen. We have spent many years honing the techniques we document here. Adopting these may speed the research process. But the approach still requires time in terms of attending practitioner events, interviewing participants, crafting qualitative data into action principles, developing relationships with practitioners so that their feedback is reflected in the work products and so that their contexts can be examined over time. For us, practice-intoresearch has to be a calling, and one that requires returning to research sites over time—often several years—to see how things turned out. Research-into-practice, as we present it here, needs also a rich understanding of context, history, process, and a longitudinal approach to studying phenomena. Furthermore, gaining access is not always easy. It requires credibility on the part of the researchers, generosity, and trust on confidentiality on the part of the respondents and also a pervading sense among all parties that the findings will be objective, relevant, fair, and insightful.

Conclusion On our informational purpose, we have demonstrated a workable research-into-practice approach that gains a great deal of contemporary information allowing us insight into how the technologies function, and how they are deployed, and with what results. We have also endeavored, through a synthesizing process, to arrive at findings on the practices that explain the positive outcomes. We will suggest that these findings provide rich insights for researchers and practitioners alike and form a strong foundation for those both doing follow-up research—as we are doing

3  Debating Genres and IS Research… 

111

ourselves—and for practitioners wrestling with real-world problems and automation deployments. We have made little theoretical contribution here, but that was never the intention and we cannot see how this obviates the approach and the value of the findings. While we have been mostly focused on the relevance of the findings to practitioners, we would invite comment also on the relevance to IS scholars of the approach taken and of the findings. On our polemical purpose, we hope we have demonstrated a way of doing relevant and rigorous research on the applied side of IS that is timely, does justice to the phenomena under investigation, and provides insights for multiple parties. Interestingly, a number of reflective practitioners have remarked on our research: In particular, how the findings make them think through again what they are doing and how to proceed. Maybe we also need more reflective researchers in IS to take the field forward, when we are endeavoring to study fast-moving technologies that are going to have massive impact on individuals, organizations, and society, and where we want to make a real contribution on how this is going to be rolled out.

References Alter, S. (2018). Applying seven images of science in exploring whether information systems is a science. Communications of the AIS, 43, 175. Augier, M., & March, J. (2011). The roots, rituals, and Rhetorics of change: North American business schools after the second world war. Stanford University Press. Avison, D., & Malaurent, J. (2014). Is theory king?: Questioning the theory fetish in information systems. Journal of Information Technology, 29(4), 327–336. Avison, D., Malaurent, J., & Eynaud, P. (2017). A narrative approach to publishing information systems research: Inspiration from the French new novel tradition. European Journal of Information Systems, 26(3), 260–273. Avital, M., Mathiassen, L., & Schultze, U. (2017). Alternative genres in information systems research. European Journal of Information Systems, 26(3), 240–247. Baldwin, R. (2019). The Globotics upheaval: Globalization, robotics and the future of work. Weidenfeld and Nicholson.

112 

L. Willcocks et al.

Baskerville, R., & Myers, M. (2004). Special issue on action Research in information systems: Making IS Research relevant to practice foreword. MIS Quarterly, 28(3), 329–335. Beer, M. (2001). Why management research findings are unimplementable: An action science perspective. Reflections, 2(3), 58–65. Behrman, J.N. & Levin, R.I. (1984, January). Are business schools doing their job? Harvard Business Review. Bennis, W.G. & O’Toole, J., (2005, May). How business schools lost their way. Harvard Business Review. Beyer, J. M. (1982). Introduction to the special issue on the utilization of organizational research. Administrative Science Quarterly, 27(4), 588–590. Bødker, M. (2017). “What else is there…?”: Reporting meditations in experiential computing. European Journal of Information Systems, 26(3), 274–286. Boland, R. J., Jr., & Lyytinen, K. (2017). The limits to language in doing systems design. European Journal of Information Systems, 26(3), 248–259. Colander, D., Follmer, H., Haas, A., Goldberg, M., Juselius, K., Kirman, A., & Sloth, B. (2014). 13 the financial crisis and the systemic failure of academic economics. The Economics of Economists: Institutional Setting, Individual Incentives, and Future Prospects, 21(2–3), 344. Colquitt, J. A., & Zapata-Phelan, C. P. (2007). Trends in theory building and theory testing: A five-decade study of the academy of management journal. Academy of Management Journal, 50(6), 1281–1303. Compeau, D. R., & Olivera, F. (2014). From ‘theory light’ to theorizing: A reaction to Avison and Malaurent. Journal of Information Technology, 29(4), 346–349. Daft, R. L. (1983). Learning the craft of organizational research. Academy of Management Review, 8(4), 539–546. Dennis, A. R., Valacich, J. S., & Brown, S. A. (2018). A comment on “is information systems a science?”. Communications of the Association for Information Systems, 43(1), 211–216. Devlin, H. (2015, February 4). Rise of the robots: How long do we have until they take our jobs? The Guardian. https://www.theguardian.com/technology/2015/feb/04/rise-­robots-­artificial-­intelligence-­computing-­jobs. Elmendorf, W., & Luloff, A. (2006). Using key informant interviews to better understand open space conversation in a developing watershed. Arboriculture & Urban Forestry, 32, 54–61.

3  Debating Genres and IS Research… 

113

Fontana, A., & Frey, J. (1994). Interviewing: The art of science. In N. K. Denzin & Y. S. Lincoln (Eds.), Handbook of qualitative research (pp. 361–376). Sage Publications. Ford, M. (2015). Rise of the robots: Technology and the threat of a jobless future. Basic Books. Galliers, R. D., & Stein, M. K. (2018). Information systems: To be, or not to be, a science? Is that the question? Communications of the Association for Information Systems, 43(1), 197–204. Galliers, R. D., & Whitley, E. A. (2007). Vive les differences? Developing a profile of European information systems research as a basis for international comparisons. European Journal of Information Systems, 16(1), 20–35. Ghoshal, S. (2005). Bad management theories are destroying good management practices. Academy of Management Learning and Education, 4(1), 75–91. Gibbons, M., Limoges, C., Nowotny, H., et al. (1994). The new production of knowledge: The dynamics of science and Research in contemporary societies. Sage. Given, L. (2008). Rigor in qualitative research. Sage Publications. Gregor, S. (2006). The nature of theory in information systems. MIS Quarterly, 30(3), 611–642. Gregor, S. (2018). The dangers of dance for the information systems. Communications of the Association for Information Systems, 43 MISSING PAGES, 205. Grover, V., & Lyytinen, K. (2015). New state of play in information systems research: The push to the edges. MIS Quarterly, 39(2), 271–296. Hambrick, D. C. (1994). 1993 presidential address: What if the academy actually mattered? Academy of Management Review, 19(1), 11–16. Harman, G. (1986). Change in view: Principles of reasoning. MIT Press. Hassan, N. R., Mingers, J., & Stahl, B. (2018). Philosophy and information systems: Where are we and where should we go? European Journal of Information Systems, 27(3), 263–277. HfS Research. (2014, April). The robotic premier league debuts. HfS Research Report, https://www.hfsresearch.com/pointsofview/robotic-­premier-­league-­ debuts. Hindle, J., Lacity, M., Willcocks, L., & Khan, S. (2018a). RPA: Benchmarking the client experience. Knowledge Capital Partners’ Executive Research Report. Hindle, J., Willcocks, L., & Lacity, M. (2018b). Keys to RPA success: Four Part Series. Available at https://www.knowledgecapitalpartners.com/publications:

114 

L. Willcocks et al.

Hirschheim, R., & Klein, H. K. (2003). Crisis in the IS field? A critical reflection on the state of the discipline. Journal of the Association for Information Systems, 4, 237–293. Hodgkinson, G. P. (Ed.). (2001). Facing the future: The nature and purpose of management research reassessed. British Journal of Management, 12(Special Issue), S1–S80. Kaarst-Brown, M. L. (2017). Once upon a time: Crafting allegories to analyze and share the cultural complexity of strategic alignment. European Journal of Information Systems, 26(3), 298–314. Khurana, R. (2007). From higher aims to hired hands: The social transformation of American business schools and the unfulfilled promise of management as a profession. Princeton University Press. Kieser, A., Nicolai, A., & Seidl, D. (2015). The practical relevance of management research: Turning the debate on relevance into a rigorous scientific research program. The Academy of Management Annals, 9(1), 143–233. Koskela, L. (2017). Why is management research irrelevant? Construction Management and Economics, 35(1–2), 4–23. Lacity, M. & Willcocks, L. (2015). What knowledge workers stand to gain from automation. Harvard Business Review (Online). https://hbr.org/2015/06/ what-­knowledge-­workers-­stand-­to-­gain-­from-­automation Lacity, M., & Willcocks, L. (2016a). Speed of automation adoption faster for providers than customers. Pulse Magazine, (May/June), 10–17. Lacity, M., & Willcocks, L. (2016b). A new approach to automating services. Sloan Management Review, 57(1), 41–49. Lacity, M., & Willcocks, L. (2016c). Robotic process automation at Telefónica O2. MIS Quarterly Executive, 15(1), 21–35. Lacity, M., & Willcocks, L. (2017). Robotic process automation and risk mitigation: The definitive guide. SB Publishing. Lacity, M., & Willcocks, L. (2018a). Robotic process and cognitive automation: The next phase. SB Publishing. Lacity, M. & Willcocks, L. (2018b). OWS18 findings revealed on client service automation deployments. Pulse Magazine. https://iaoppulse.net/2018/06/ research-­c orner-­ows18-­f indings-­revealed-­o n-­c lient-­s ervice-­a utomation-­ deployments-­what-­do-­they-­mean-­for-­your-­job-­and-­organization/ Lacity, M., Willcocks, L., & Yan, A. (2015). Are the robots really coming? Service automation survey findings. Pulse Magazine, Issue 17, 14–21. Lacity, M., Babin, R., & Willcocks, L. (2017a). Research center: Service automation trends survey. Pulse Magazine, Issue 28, 40–44.

3  Debating Genres and IS Research… 

115

Lacity, M., Willcocks, L. & Craig, A. (2017b). Service automation: Cognitive virtual agents at SEB bank. The LSE Outsourcing Unit Working Research Paper Series. Lacity, M., Willcocks, L., & Gunkel, G. (2017c). Smart sourcing: Cognitive automation at Zurich insurance. Intelligent Sourcing Magazine. Issue 4, 34–42. Lacity, M., Willcocks, L., & Gozman, D. (2021). Influencing information systems practice: The action principles approach applied to robotic process and cognitive automation. Journal of Information Technology, 36(3), 216–240. Lee, A. (1999). Rigor and relevance in MIS Research: Beyond the approach of positivism alone. MIS Quarterly, 23(1), 29–33. Lee, A. S. (2004). Thinking about social theory and philosophy for information systems. In J. Mingers & L. Willcocks (Eds.), Social theory and philosophy for information systems (pp. 1–26). Wiley. Lincoln, Y., & Guba, E. (1985). Naturalistic inquiry. Sage Publications. Love, J., & Hirschheim, R. (2017). Crowdsourcing of information systems research. European Journal of Information Systems, 26(3), 315–332. Markus, M. L. (2014). Maybe not the king, but an invaluable subordinate: A commentary on Avison and Malaurent’s advocacy of ‘theory light’ IS research. Journal of Information Technology, 29(4), 341–345. Mathiassen, L. (2017). Designing engaged scholarship: From real-world problems to research publications. Engaged Management Review, 1(1), 2. Mathiassen, L., & Sandberg, A. (2013). How a professionally qualified doctoral student bridged the practice-research gap: A confessional account of collaborative practice Research. European Journal of Information Systems, 22(4), 475–492. Mathiassen, L., Chiasson, M., & Germonprez, M. (2012). Style composition in action research publication. MIS Quarterly, 36, 347–363. McKelvey, B. (2006). Van De Ven and Johnson's “engaged scholarship”: Nice try, but…. Academy of Management Review, 31(4), 822–829. Munir, K. A. (2011). Financial crisis 2008-2009: What does the silence of institutional theorists tell us? Journal of Management Inquiry, 20(2), 114–117. Ng, A. (2016). What artificial intelligence can and can’t do right now, Harvard Business Review Blog. https://hbr.org/2016/11/what-­artificial-­intelligence-­ can-­and-­cant-­do-­right-­now. Palin, A. (2016, June 23). Financial crisis forced business schools to change curriculum. Financial Times. Pettigrew, A. M. (2001). Management research after modernism. British Journal of Management, 12, S61–S70.

116 

L. Willcocks et al.

Podolny, J. M., Kester, W. C., Kerr, S., Sutton, R. I., Kaplan, S., & Martin, R. (2009). Are business schools to blame. Harvard Business Review, 87(6), 106–108. Prasopoulou, E. (2017). A half-moon on my skin: A memoir on life with an activity tracker. European Journal of Information Systems, 26(3), 287–297. Robey, D., & Markus, M. L. (1998). Beyond rigor and relevance: Producing consumable research about information systems. Information Resources Management Journal, 11(1), 7–16. Rosemann, M., & Vessey, I. (2008). Toward improving the relevance of information systems Research to practice: The role of applicability checks. MIS Quarterly, 32(1), 1–22. Rowe, F. (2012). Toward a richer diversity of genres in information systems research: New categorization and guidelines. European Journal of Information Systems, 21(5), 469–478. Rowe, F. (2018). Being critical is good, but better with philosophy! From digital transformation and values to the future of IS research. European Journal of Information Systems, 24(6), 380–393. Scheepers, R., Lacity, M., & Willcocks, L. (2018). Realizing value from cognitive automation at Deakin University. MIS Quarterly Executive, 17, 89–107. Seidler, J. (1974). On using informants: A technique for collecting quantitative data and controlling for measurement error in organizational analysis. American Sociological Review, 39, 816–831. Stein, G. (1937). Everybody’s autobiography. Cooper’s Square. Stein, M. K., Galliers, R. D., & Whitley, E. A. (2016). Twenty years of the European information systems academy at ECIS: Emergent trends and research topics. European Journal of Information Systems, 25(1), 1–15. Te'eni, D., Rowe, F., Ågerfalk, P. J., & Lee, J. S. (2015). Publishing and getting published in EJIS: Marshaling contributions for a diversity of genres. European Journal of Information Systems, 24(6), 559–568. Thatcher, J., Pu, W., & Pienta, D. (2018). Information systems IS a (social) science. Communications of the AIS, 43(11), 189–196. Thompson, D. (2015). A world without work (pp. 50–61). The Atlantic. Toffel, M. W. (2016). Enhancing the practical relevance of research. Production and Operations Management, 25(9), 1493–1505. Tourish, D. (2015). It’s time to retire the ABS guide to journal quality. Times Higher Education, p. 32. Van de Ven, A. H. (2007). Engaged scholarship: A guide for organizational and social research. Oxford University Press.

3  Debating Genres and IS Research… 

117

Van de Ven, A. H., & Johnson, P. E. (2006). Knowledge for theory and practice. Academy of Management Review, 31(4), 802–821. Van Maanen, J. (1995). An end to innocence: The ethnography of ethnography. In J. Van Maanen (Ed.), Representation in ethnography (pp. 1–35). Sage. Wainwright, D., Oates, B., Edwards, H., & Childs, S. (2018). Evidence-based information systems: A new perspective and a road map for Research-­ informed practice. Journal of the Association for Information Systems, 19(11) Article 4. Weick, K. E. (2001). Gapping the relevance bridge: Fashions meet fundamentals in management research. British Journal of Management, 12. (Special Issue), S71–S75. Willcocks, L. & Lacity, M. (2015a, September 15). Businesses will increasingly use robots to deal with the explosion of data. LSE Business Review Online. Willcocks, L. & Lacity, M. (2015b, October 6). Will robots replace humans? LSE Business Review Online. Willcocks, L. & Lacity, M. (2015c). Poppy appeal: RPA at BPO service providers. Professional Outsourcing Magazine, Winter Issue, 12–18. Willcocks, L. & Lacity, M. (2015d). IT, phone home: Leslie Willcocks and Mary Lacity on Telefonica O2's RPA. Professional Outsourcing Magazine, Autumn Issue, 26–34. Willcocks, L., & Lacity, M. (2016). Service automation: Robots and the future of work. SB Publishing. Willcocks, L., Hindle, J., & Lacity, M. (2019). Becoming strategic with robotic process automation. SB Publishing.

4 A Paradigm Shift in Understanding Digital Objects in IS: A Semiotic Perspective on Artificial Intelligence Technologies Julia Kotlarsky

and Ilan Oshri

Introduction Traditional understandings of what digital objects are in the IS field center on their representation as either physical or digital components characterized by their ability to process information (Faulkner & Runde, 2019). Faulkner and Runde’s (2019) theorization of digital objects is part of a growing body of studies interested in understanding the nature of IT/IS artifacts (e.g., Akhlaghpour et al., 2013; Chatterjee et al., 2021; Iivari, 2017; Lee et al., 2015; Orlikowski & Iacono, 2001). We join this discourse by attempting to decode the meaning attributed to digital objects in the IS literature and consequently advance our understanding using the unique case of Artificial Intelligence (AI) as a digital object.

J. Kotlarsky (*) • I. Oshri University of Auckland, Auckland, New Zealand e-mail: [email protected]; [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. P. Willcocks et al. (eds.), Advancing Information Systems Theories, Volume II, Technology, Work and Globalization, https://doi.org/10.1007/978-3-031-38719-7_4

119

120 

J. Kotlarsky and I. Oshri

Building on the theory of digital objects (Faulkner & Runde, 2019), we apply the lens of semiotics (Peirce, 1932) to show that in the IS field, the meaning of digital objects is based on a stable and specific relationship between the signifier, signified, and the referent. We coin this understanding of digital objects as the particular semiotic paradigm. We then argue that Artificial Intelligence technologies are a distinct type of digital object—with self-learning and self-programming abilities. We use the term Artificial Technologies (AI) to refer to a broad range of intelligent technologies1 encompassing cognitive automation, machine learning,2 reasoning, hypothesis generation and analysis, natural language processing, and autonomic systems that self-manage their own operations and the processes they oversee. The key characteristic of such intelligent technologies is adapting to the environment and/or updating their own lines of inquiry while interacting with other systems and/or the environment. Traditional ways of understanding the meaning of AI objects with self-­ learning and self-programming abilities are currently not accommodated by the particular semiotics paradigm. Consider the following narrative: In July 2017, Facebook shut down an artificial intelligence (AI) system because the system had created its own language. Although the system was trained to communicate in English, engineers noticed that it had gradually diverged from English to use its own language when communicating with other AI agents. It was reported that “the resulting phrases appear to be nonsensical gibberish to humans but contain semantic meaning when interpreted by AI agents”. As such, Facebook’s AI system was able to communicate its new language to other AI agents; however, the new language was not understandable to humans.

 This definition is broader than those used by some academics and practitioners to describe Artificial Intelligence. It includes AI (see Appendix 1) and also other technologies, for example, autonomic systems. The definition is intended to capture the key characteristic of technologies designed to change as they interact with their environment and will allow IS researchers to easily relate to new, emerging technologies and accommodate changes in business terminology. Definitions of specific technologies, based on the IEEE Guide for Terms and Concepts in Intelligent Process Automation (2017), are included in Appendix 1 to illustrate the common element we have put forward as a key characteristic—adapting and updating while interacting with the environment. 2  At present in the IS field most common terms are Artificial Technology and Machine Learning (e.g., Asatiani et al., 2021; Lebovitz et al., 2021; Lou & Wu, 2021; van den Broek et al., 2021). 1

4  A Paradigm Shift in Understanding Digital Objects in IS… 

121

The report offered an insight into the exchanges between two AI agents: “Two negotiating bots, named Bob and Alice, used their own language to complete their exchange. Bob started by saying ‘I can i i everything else’, to which Alice responded, ‘Balls have zero to me to me to me…’. The rest of the conversation was formed from variations of these sentences.” 3 The report carried on to consider the implications for current and very likely future human-machine interactions, stating that such unexpected outcomes “[…] do make AI development more difficult [..] as humans cannot understand the overwhelmingly logical nature of the languages. While they appear nonsensical, the results observed by teams such as Google Translate indicate they actually represent the most efficient solution to major problems.

This narrative points to an unexpected and unintended outcome human actors are unable to make meaning of within the existing semiotics paradigm of digital objects, thus requiring a paradigm shift to accommodate the unique characteristics of AI technologies as self-learning and ever-changing systems. Consequently, in this chapter we establish the foundations of a pluralistic semiotic paradigm of digital objects, following a transformative paradigm approach (Hassan & Mingers, 2018) to offer guidance to human actors on how to understand self-learning and self-­ programming digital objects. Our work offers important contributions to the growing body of studies on products of theorizing (e.g., Hassan et al., 2022) and digital objects (Faulkner & Runde, 2019). In terms of products of theorizing, our work challenges the existing paradigm by examining the meaning attributed to digital objects through a semiotics lens. Through an examination of the characteristics of AI technologies, we advocate for a paradigm shift and establish the foundations of the pluralistic semiotics paradigm. As such we illustrate paradigm as a product of theorizing, consistent with Chap. 2 (Table 2.1). Our work also expands the IS literature stream on digital objects by offering an examination of the meaning of the digital objects, an area previously neglected in the IS literature.

 Text in italics is based on this article: https://www.independent.co.uk/voices/facebook-shuts-­ down-robots-ai-artificial-intelligence-develop-own-language-common-a7871341.html 3

122 

J. Kotlarsky and I. Oshri

We next introduce the theoretical foundation of the study as we examine the current paradigm of the meaning of digital objects in the IS field.

Theoretical Foundations Understanding Paradigms in IS Research Traditionally, the IS field has primarily referred to paradigms as the epistemological views that guide IS scholars with regard to their methodological choices. Accordingly, the IS field has distinguished between the positivist, interpretivist, critical, and mixed-method pluralist paradigms (Mingers, 2001, 2003; Venkatesh et al., 2013). As Hassan and Mingers (2018) point out, there has been a “misplaced focus on methodology” (p. 574) when paradigms are discussed in the IS context. Moreover, considering paradigms from a methodological perspective is limiting because “[the] reliance on the epistemological sense of the paradigm diverts the attention of the researcher from the ‘context of discovery’ and limits the research to the ‘context of justification’” (ibid. p. 574). They go on to suggest that “[t]here may be certain beliefs and practices in the IS field that have become endemic and require major paradigm shifts in the minds and practices of its researchers” (p.569). Based on the seminal work of Kuhn, Hassan and Mingers revisit the concept of paradigm as understood in IS, introducing a more vigorous, transformative understanding. According to Kuhn’s definition, the term paradigm can be considered from two different perspectives: On the one hand, it stands for the entire constellation of beliefs, values, techniques, and so on shared by the members of a given community. On the other, it denotes one sort of element in that constellation, the concrete puzzle-solutions which, employed as models or examples, can replace explicit rules as a basis for the solution of the remaining puzzles of normal science. (Kuhn, 1970, p. 175, cited in Hassan & Mingers, 2018 p. 570).

Elucidating Kuhn’s concept, Masterman (1970) proposed three categories of paradigm: (i) metaphysical paradigms—ways of seeing the

4  A Paradigm Shift in Understanding Digital Objects in IS… 

123

world, beliefs, and organizing principles; (ii) sociological paradigms— sociological rules accepted in a particular community and practices (e.g., political institutions, legal rules that inform judicial decisions and linguistics that inform grammatical usage); and (iii) conceptual or artifactual paradigms—standard (i.e., commonly accepted) tools, applications, technique widely used in a particular scientific community. Hassan and Mingers (2018) consider Masterman’s categories “to be the most useful for a multidisciplinary field like IS” (p. 577), arguing that the “Kuhnian paradigm when adopted in its transformative form, establishes the correct balance of metaphysical, sociological and artifactual components” (p. 583). It is this transformative aspect of the Kuhnian paradigm we adopt in this chapter as we (i) offer a conceptualization of how digital objects are currently understood by human actors in IS research, through the lens of semiotics, and (ii) argue for a paradigm shift in how we understand digital objects in the age of self-learning AI technologies. In order to introduce the paradigm that applies to current understanding of digital objects, we next provide our two theoretical foundations for the paradigm—fundamentals of semiotics (Mingers & Willcocks, 2014; Peirce, 1932) and the theory of digital objects (Faulkner & Runde, 2019).

Fig. 4.1  Peircean semiotics: core elements and their relationships (core elements of semiotics and their definitions are included in Appendix 2.)

124 

J. Kotlarsky and I. Oshri

The Fundamentals of Semiotics: A Peircean View According to semioticians, the world is represented via signs, and each sign denotes a meaning4 (de Saussure, 1916/2003; Peirce, 1932). A sign, according to Peirce (1932), “is something which stands to somebody for something in some respect or capacity.” The Peircean approach views signs in terms of the semiotic triangle: signifier,5 signified, and referent. In the physical world, the relationship between the three key Peircean elements is clear (see Fig. 4.1). A signifier connotes the signified and stands for the referent. The signifier can be a number, an image, a word, a sound, or a smell. The signified, which is something with a material or non-­ material presence, expresses the meaning of the signifier (Bailey et  al., 2012; Mingers & Willcocks, 2014). Peircean work distinguishes the relationship between the signifier and the signified in terms of three types of representations: iconic, indices, and symbolic. Iconic representations contain visual images of what they intend to signify (e.g., portraits, sketches, drawings) and signify because of the close resemblance to what they connote. Indices representations express physical and existential relationships between the signifier and the signified, as well as co-occurrence with what they signify. Well-known examples are thermometers as an index of temperature (Mingers & Willcocks, 2014) and smoke that signifies fire (Bailey et al., 2012). Symbolic representations are symbols that are only meaningful for those who understand its professional, social, or cultural convention6 because of the arbitrary link between the signifier and the signified (Bailey et  al., 2012; Peirce, 1932). For example, natural language or mathematical notations are based on symbols requiring relevant

 As noted in Oshri et al. (2018), scholars distinguish between Saussurian (de Saussure 1916/2003) and Peircean (Peirce, 1932) semiotics (also known as structural semiotics and social semiotics [Mingers 2014]). De Saussure views a sign as a dyadic relationship between two elements—the signifier and the signified. Peirce adds a third element—the object—thus forming a triadic relation between the elements of the sign (see Mingers and Willcocks (2014) and Mingers (2014) for a detailed discussion of the two approaches and why the Peircean approach may be considered more suitable for IS research). 5  Also referred to as representation (e.g., Bailey et al., 2012; Oshri et al., 2018). 6  The term “convention” refers to agreement to use the same rules in order to communicate (Crystal, 2003; Oshri et al., 2018). 4

4  A Paradigm Shift in Understanding Digital Objects in IS… 

125

conventions to decode them. Another example would be a traffic sign— white rectangle at the center of a red circle—that means “no entry.” To interpret the correct meaning of this sign, a driver needs to know the traffic conventions that give meaning to the sign. Therefore, to be meaningful in practice, the signifier needs to make sense to the community of actors using it to do their work (e.g., Nicolini, 2012). Semiotics, as an explanatory theoretical lens, has been used in management studies and in marketing in particular (e.g., Arnold et al., 2001; Lawes, 2002). To much lesser effect, there has been interest in using semiotics in the context of information systems, mainly to understand the human-computer interface (e.g., Stamper, 2001), to make sense of visual management systems (e.g., Beynon-Davies, 2018), and to use ICT as a communication tool (e.g., Warschauer & Grimes, 2007). In IS field semiotics have been offered as a methodology for studying information systems (e.g., Mingers & Willcocks, 2014, 2017) and, more recently, as a lens to capture representations in complex technology-enabled settings such as IT outsourcing (Oshri et al., 2018) and social media (Mikhaeil & Baskerville, 2019). In relation to the concept of representation, Bailey et  al. (2012) demonstrated that representations support different types of virtual work (virtual teams, remote control, and simulations) based on what it is that a technology makes virtual and whether the work is done with or on, through, or within representations. Visual representations have also been discussed on the continuum naturalistic to scientific, or concrete to abstract, with Diaz et  al. (2015) comparing their characteristics and the referents they stand for. Some IS studies have considered the use of technologies as a trigger to re-examine the meaning of signs through a semiotic lens. For example, Kallinikos (2011) describes how operators of an automated dairy production line developed stress and anxiety as their dependence on an abstract coding system that monitors the physical production line increased as a result of remoteness from the actual production process. In this case, while the relationship between the signifier and signified remained untouched, the representation of the physical object (referent) as a series of bulbs and lights had a negative effect on operators. Oshri et al. (2018) examined a case in which Mandarin-speaking teenagers were able to transcribe hand-written documents in 23 languages at high speed and accuracy. In this case, management manipulated the relationships between

126 

J. Kotlarsky and I. Oshri

the signifier and the signified in the outsourcing work so that operators ignored the meaning of the text when transcribing and instead treated text as a string of letters, denoting it a visual symbolic representation. These two studies provide insight into the opportunities offered by semiotics when the relationship between the signifier and signified changes, thus requiring individuals to decode new meaning and assume new conventions. Using a semiotics lens, we next examine the existing paradigm for understanding digital objects in the IS field.

 igital Objects Through a Semiotics Lens: D Current IS Paradigm The Theory of Digital Objects: Background Over the last 20 years there have been several attempts in the IS literature to theorize IT/IS and digital artifacts (e.g., Akhlaghpour et  al., 2013; Chatterjee et  al., 2021; Iivari, 2017; Lee et  al., 2015; Orlikowski & Iacono, 2001). At the heart of such interest is the persistent lack of engagement with the IT artifact in IS studies (e.g., Faulkner & Runde, 2019; Orlikowski & Iacono, 2001). Indeed, a recent study by Faulkner and Runde (2019) calls for a more thorough engagement with digital objects, which are the key pillars of information technologies nowadays. In their proposed theory of digital objects, they distinguish between material and non-material variants of objects, as well as hybrid objects that comprise material and non-material objects as component parts. Material objects exist in the physical mode of being (e.g., scanners, smartphones, and servers), while non-material objects have a non-physical mode of being (e.g., news articles, operating system, protocols, conceptual schemas). They further recognize a particular type of non-material objects—syntactic “objects that consist of symbols arranged into well-­ formed expressions, where well-formed means that these expressions adhere to the syntactical and semantic rules of the language in which they are couched” (p. 1284). Examples of syntactic objects include “natural language novels, manuals, contracts, textual entities in artificial languages such as musical notations, Morse code, or mathematics” (p.  1284). In

4  A Paradigm Shift in Understanding Digital Objects in IS… 

127

coming up with a definition of a digital object, Faulkner and Runde (2019) highlight that bitstring is one particular type of syntactic object— one made of bits (“the 0s and 1s employed in a binary numbering system, where these bits are structured according to an appropriate file format so as to be readable by computer hardware for which they are intended”; p. 1285). Bitstrings are divided into two categories—program files and data files. Based on these foundations, digital objects are thus defined as “objects whose component parts include one or more bitstrings. The set of digital objects, therefore, includes individual bitstrings as a limit case, but generally refers to a far broader category of objects, usually hybrids, in which bitstrings are combined with various types of material and non-­ material components” (ibid. p.  1285). To capture the relationship between the material and non-material components of different digital objects, Faulkner and Runde (2019) introduce the concept of a bearer, which can be material or non-material, to indicate how objects in the digital domain are placed in relation to one another. Specifically, they call “any material object on which a non-material object is so inscribed a material bearer of non-material object” (p. 1286), highlighting that such material bearer of a non-material object qualifies as a hybrid digital object. For example, a non-material (syntactic) object such as a newspaper article has to be inscribed on a suitable material bearer (e.g., smartphone, computer monitor, or a printed newspaper) that will allow a human actor to read the article displayed in a digital format on a screen or in a physical form. Non-material objects may have non-material bearers. For example, “a program file is the bearer of a set of instructions associated with a particular series of computations, while a data file is the bearer of an image, document, dataset, or some kind of data” (p. 1287). To conclude our overview of the theory of digital objects, it is important to bring to the fore two aspects that help understand how digital objects relate to each other. First, Faulkner and Runde (2019) highlight the role of computation7 in the conception, design, generation, and/or  Faulkner and Runde (2019) define computation as “the real-time processes performed by digital computers that involve the algorithmic manipulation of information borne by bitstrings. These processes, largely implicit in our account so far, are relevant here for their existential relationship with digital objects. This relationship is two-way, the existence of computation at once depending on and contributing to the existence of digital objects” (p. 1288). 7

128 

J. Kotlarsky and I. Oshri

production of digital objects of all kinds: bitstrings (digital objects of the lowest granularity) are borne through computation, as well as manufactured (i.e., hybrid) digital objects, including computers and the various equipment and devices involved with their use. Second, digital objects often have several layers of bearers: non-material bearers that can be borne (endlessly) through computational processes and which may be layered on a material bearer8: “This idea of the repeated layering of nonmaterial objects, facilitated by the capacity of bitstrings to act as bearers, seems to us one of the defining features of digital technology and a major factor in what makes contemporary digital objects unique” (ibid. p. 1288). These two aspects are specifically relevant to semiotics because they emphasize that (i) digital objects may have different levels of granularity, from bitstrings (lowest granularity) to any combination of material and/ or non-material objects of interest for a particular human actor (e.g., software developer or user of a particular device) and (ii) the computational processes and syntactic objects they rely on and produce are written following languages, logic, standards, procedures, and protocols that are known and understood by human actors. We now turn to semiotics to examine current understanding of digital objects in IS.

 pplying a Semiotics Lens to the Theory A of Digital Objects While de Saussure’s (1916/2003) and Peirce’s (1932) seminal work in semiotics considers the physical world, we can apply the same principles to the digital world by unpacking how we understand the meaning of digital objects. Thus, applying a semiotics lens to the theory of digital  Faulkner and Runde (2019) provide an example of layering: “this kind of layering occurs in relation to computer programming, where a given set of logical operations may be encoded in a variety of different higher-order programming languages, each giving rise to a syntactic object, the source code, that is a nonmaterial bearer of that instruction set. Each of these syntactic objects could then be encoded in binary as machine code for a variety of different processors, giving rise to multiple bitstring encodings of the same source code” (p. 1288). 8

4  A Paradigm Shift in Understanding Digital Objects in IS… 

129

Fig. 4.2  Understanding digital objects through the lens of Peircean semiotics

objects, any digital object can be depicted in a Peircean triangle of signifier, signified, and referent, through which the meaning of the digital object is understood by a human actor. We illustrate this in Fig. 4.2 and then discuss the depicted relationships using examples of digital objects from Faulkner and Runde (2019), as well as our own examples. Faulkner and Runde (2019) describe a material object—a typical flatbed scanner—that “comprises a light source, image sensor, glass panel, control circuitry, and various other components, arranged in a way that renders the object as a whole capable of converting an analog image into digital form” (p. 1284). Though such a scanner can be used by manually pressing a “start” button, many users nowadays will have a software application (driver) installed on their computer/device to activate the scanner from that device. From the semiotics viewpoint, an icon (signifier, in the form of an iconic representation) in the Applications folder of a user connotes the function of converting an analog into a digital image (signified) and stands for a flatbed scanner (referent, i.e., a hybrid digital object consisting of an integrated array of material and non-material components), as shown in Ex. 1 (Fig. 4.2). Another example (Ex. 2 in Fig. 4.2) is a news article displayed on a material bearer such as screen or paper (signifier, in the form of a symbolic representation) that connotes a

130 

J. Kotlarsky and I. Oshri

particular story (signified); it is interpreted by readers and stands for the news article (referent, a non-material object). Further examples in Fig. 4.2 of non-material and hybrid digital objects and how they are interpreted through signifiers are: Ex. 3—an image with name of a song on an iTunes library (signifier in the form of iconic and symbolic presentations) connotes the function of playing this song (signified) and stands for a music file with a particular song (referent, a non-material object). Ex. 4—an image of a virtual agent, with a chat window, displayed on a web page on Expedia (signifier—an iconic representation of the agent, with a box for natural-language symbolic input) connotes the function of inquiring and getting information from Expedia (signified) and stands for a virtual agent that impersonates a human agent of Expedia (referent, a non-material object). Ex. 5—an image of a clock on a smartphone that plays an alarm sound (signifier, an iconic representation that enacts indices representation— alarm sound) connotes a function of getting a sound warning in a chosen time (signified) and stands for digital alarm clock (referent, a hybrid object). These examples lend support to our contention, as manifested in the current theory of digital objects, that the relationships between the signifier, signified, and referent are understood as stable and particular. Digital objects are represented through a specific signifier that connotes a particular signified through (a set of ) conventions known9 to human actors.

 By “conventions that are known to human actors,” we are referring to the following range of conventions that human actors use consciously or unconsciously: (i) social and cultural conventions, many of which human actors obtain unconsciously through their upbringing in a particular cultural and social environment, and their own life experience. These include natural language conventions of the languages a human actor is familiar with; (ii) professional conventions that human actors learn throughout their education and practice in a particular professional domain; and (iii) conventions that human actors embed into digital objects or computational processes through which new digital objects are born, such as special-purpose languages (programming languages, math, scientific notations, design principles, schemas, protocols, and procedures, etc.). 9

4  A Paradigm Shift in Understanding Digital Objects in IS… 

131

We coin this as a particular semiotic paradigm of digital objects in which a digital object has a particular meaning manifested via a distinct set of relationships between the signifier and the signified. Importantly, understanding the meaning of digital objects depends on representations (primarily symbolic and iconic) that are built on various conventions designed by human actors. These conventions (i) are typically anchored in social, cultural, and professional conventions related to the context in which digital objects are used, and (ii) include syntactic and semantic rules of programming and special-purpose languages (in addition to natural languages), as well as (iii) logic, rules, and standards that are encoded in the functionality of digital objects and their bearers. Having established the current paradigm of digital objects in the IS field, we now examine AI technologies as an extension of digital objects.

 he Distinctive Case of Artificial Intelligence T in the Digital World In recent years we have witnessed the deployment of AI technologies across various sectors. Common examples of AI include speech recognition, self-driving cars, smart assistants, disease mapping, and virtual agents (e.g., chatbots). The term Artificial Intelligence broadly refers to the ability of a digital computer or computer-controlled robot to perform tasks commonly associated with intelligent beings10 and is widely used by practitioners in the popular and scientific media as well as in academic publications. The conceptual boundary of the term Artificial Intelligence has not yet been established, and it will most likely continue to evolve with development of new AI-enabled tools. However, for the purpose of our work, we use “Artificial Intelligence” (AI) to refer to a range of intelligent technologies that are characterized by adapting to the environment and/or updating their own lines of inquiry while interacting with other

10

 https://www.britannica.com/technology/artificial-intelligence

132 

J. Kotlarsky and I. Oshri

systems and/or the environment (see also footnote 1 and Appendix 1 for detailed explanations). Using terminology established in the theory of digital objects (Faulkner & Runde, 2019), all AI-enabled devices, solutions, and algorithms are digital objects in their own right (hybrid or non-material), with AI being one component (a more granular non-­ material digital object).11 Therefore, various AI-enabled entities, such as a chatbot (virtual agent) of Expedia, self-driving car, or satnav application (e.g., Waze), are all AI digital objects. An AI digital object either supports human agency or possesses agency in itself (e.g., Rai et al., 2019). In both cases, AI digital objects possess a self-learning capability. A statistical computer model capable of recognizing patterns in data is a classic example of the self-learning ability of an AI digital object (e.g., Balasubramanian et  al., 2020). The computer model, in this case, “fills in gaps” and learns from previous observations what is a desired (or not desired) outcome. In this regard, training the machine to seek a pattern is an important stage; however, AI algorithms continue to learn beyond the initial training period as they make additional observations through newly acquired data (Asatiani et al., 2021; Faraj et al., 2018; Lebovitz et al., 2021). Furthermore, some AI solutions possess a self-modifying code capability by which the algorithm can rewrite its own source code or programs (e.g., an autonomic system used in self-driving cars or AI art generator of some sort). We believe these new abilities have implications for the existing paradigm in IS about the meaning of digital objects. We present three narratives as examples of unintended outcome of the self-learning ability possessed by Artificial Intelligence that will later guide us in challenging the existing paradigm of digital objects:

 Sometimes instead of saying “AI digital object” we refer to an AI algorithm, AI agent, or AI solution, in particular when discussing a specific context or example (e.g., Facebook AI agent or Amazon AI hiring algorithm). 11

4  A Paradigm Shift in Understanding Digital Objects in IS… 

133

Narrative 1: Amazon Artificial Intelligence Recruiting System12 A recruiting tool used by Amazon made headlines when it was found to have been excluding women from its recommended applications for months. The system had learned from biased data and reflected those same biases. Even without gender being explicitly clear, the system associated some clubs, roles, and colleges with a certain gender and excluded any applicants that listed these on their resume. Amazon’s computer models were trained to vet applicants by observing patterns in resumes submitted to the company over a 10-year period. Most came from men, a reflection of male dominance across the tech industry. In effect, Amazon’s system taught itself that male candidates were preferable. It penalized resumes that included the word “women’s”, as in “women’s chess club captain”. And it downgraded graduates of two allwomen’s colleges, according to people familiar with the matter … Amazon edited the programs to make them neutral to these particular terms. But that was no guarantee that the machines would not devise other ways of sorting candidates that could prove discriminatory […] the technology favored candidates who described themselves using verbs more commonly found on male engineers’ resumes, such as “executed” and “captured” […] Gender bias was not the only issue. Problems with the data that underpinned the models’ judgments meant that unqualified candidates were often recommended for all manner of jobs […] with the technology returning results almost at random.

Narrative 2: Microsoft Chatbot “Gone Wrong”13 A chatbot created by Microsoft in 2016 was shut down after it delivered hate speech and sexist comments, all within 24 hours of being up and running. While some harmful comments were provoked by other online users asking the bot to repeat their words, some were unprovoked. These views had been learnt from both the training data and the self-learning capabilities the chatbot possessed as the algorithm ran through interaction with other users.

 Most of the text in italic is from this article: https://www.reuters.com/article/us-amazon-com-jobsautomation-insight-idUSKCN1MK08G 13  This narrative (including pictures) is based on this article: https://www.theverge.com/2016/ 3/24/11297050/tay-microsoft-chatbot-racist 12

134 

J. Kotlarsky and I. Oshri

Narrative 3: Google’s AI Creates Its Own “Universal Language”14 In November 2016, two months after its Neural Machine Translation (GNMT) system was switched on, Google found the AI computer had developed its own internal language to represent the concepts it uses to translate between other languages. Google created the GNMT system to help it automatically improve the fluency and accuracy of its translations. Rather than translating individual phrases or words, the machine learning system analyzes and makes sense of languages by looking at entire sentences. Following several months of testing, researchers behind the AI realized it was able to blindly translate languages even it had not studied one of the languages involved in the translation. In one instance, for example, the AI was taught Portuguese to English and English to Spanish translations. From this, the AI was able to make translations from Portuguese to Spanish.

 Paradigm Shift: From a Particular A to a Pluralistic Semiotic System of AI Digital Objects  nderstanding Artificial Intelligence as Digital Objects: U A Semiotics View The narratives present cases of unexpected and unintended outcomes generated by Artificial Intelligence. The unexpected and unintended15 outcome can be negative, as with Amazon’s hiring system, or positive, as in the Google’s Neural Machine Translation. All the narratives portray this type of digital object as capable of self-learning that could even be self-generating code (bitstrings). Extending beyond the question of AI explainability (e.g., Asatiani et al., 2020), the code generated by the algorithm raises issues concerning the users’ inability to understand the outcome and the even more surprising lack of control over the outcome produced by the algorithm.  https://www.thefuturescentre.org/signal/googles-ai-creates-its-own-universal-language/  For example, Mayer et al. (2020) use the term “unintended consequences” as a contrast to “intentions of senior managers” to discuss outcomes of introducing AI-based decision making that had positive and negative consequences for both employees and the organization. 14 15

4  A Paradigm Shift in Understanding Digital Objects in IS… 

135

From a semiotics viewpoint, observations of unexpected and unintended outcomes raise questions about the relationship between the signifier and the signified. Consider the case of the Facebook AI system presented earlier at the beginning of this chapter. Here, the AI algorithm had been constantly changing, self-creating new code and generating outcomes (phrases or sentences) to converse with other AI agents, but which are not understandable by its human operators. The resulting conversation between AI agents was based on a new language and a new convention, known to the AI developer, replacing the convention programmed by human actors. At the heart of our challenge to offer meaning to AI digital objects of this type are the following discrepancies within the existing semiotic paradigm of digital objects: 1. In digital environments that involve the enactment of AI, our understanding of a signifier (representation of an AI digital object) is challenged. According to the theory of digital objects, an AI tool is a hybrid digital object, composed of more granular digital objects connected by layers of bearers and bitstrings. As such, when AI digital object self-creates new lines of code (e.g., new bitstrings, program, and/or data files), it is changing the composition of the digital objects that form it.16 Therefore, the signifier (representation of the AI digital object and its components) is an ever-changing entity that is self-­ generated by its own source code. How are we to understand an ever-­ changing signifier? 2. A signifier that self-generates new computational outcomes (new languages, new language translations, new attitudes such as hatred or sexism) constantly creates multiple unexpected and unintended signifieds that possibly connote meaning not understood by humans (such as translation language, new agent conversable language). How are we to understand an ever-changing signifier and its multiple unintended signifieds?

 The theory of digital objects highlights the importance of computation in generating new digital objects; however, it does not discuss situations where components of a digital object (which are also digital objects, just more granular) are changing. 16

136 

J. Kotlarsky and I. Oshri

3. The relationships between the signifier and the signified are manifested through a convention of some sort. In a semiotics system where there is a stable signifier (e.g., mortgage application) and a signified (decision to approve or reject), the convention that translates a symbolic representation (signifier) into a computational outcome (signified) follows a decision process guided by a human-generated decision-making logic. The decision process the computation is based on does not change, regardless of changes in the content of the signifier (e.g., financial data, marital status, employment status). However, in the narratives we examined, as an AI digital object self-created new lines of code (bitstrings, program, and/or data files), it generated new logic that reflected the learning patterns but did not necessarily correspond with the logic the human actors were familiar with, nor the one they had bestowed on the system. This led to the creation of new conventions (e.g., the principle upon which a logic for hiring male candidates was established, as in the case of Amazon hiring AI), and in turn to the generation of computational outcomes (signifieds) with new meanings that may or may not be understood by human actors (e.g., Google’s “universal language” and the new language created by Facebook’s AI negotiating agents). Put simply, here we are observing the AI digital object as the generator of new conventions. How should human actors interpret, semiotically, a convention made of bitstrings generated by an AI digital object? What type of convention (e.g., symbolic, logical) would this be? Clearly, these characteristics of AI, as well as the occasional anomalies in AI behavior and outcomes, challenge the suitability of the particular semiotic system based on a stable signifier-signified relationship and a human-made convention. Instead, we observe an ever-changing signifier, an unexpected and unintended signified, and a convention that is hidden from the human eye.

4  A Paradigm Shift in Understanding Digital Objects in IS… 

137

Thus, we argue for a shift toward a new paradigm17 to accommodate the unique characteristics of AI as digital objects. Such a paradigm shift is timely and important because AI is increasingly becoming standard in numerous manufactured devices (hybrid digital objects) and software solutions (non-material digital objects) that are embedded in material and non-material bearers. Given that AI-enabled devices and solutions are hybrid digital objects that contain AI as one of their components (a more granular non-material digital object), it is imperative that understanding of digital objects in the IS field includes AI digital objects. To reiterate, Kuhn’s (1970) four characteristics of a paradigm shift have inspired and guided our quest for a new paradigm: (i) “all crises begin with a blurring of the paradigm and a consequent loosening of the rules for normal research” (p. 84); (ii) anomalies appear that do not fit the traditional view (Schneider, 1987); (iii) a new paradigm must provide the hope that it is possible to march forward (Schneider, 1987); and (iv) “during the transition period there will be a large but never complete overlap between the problems that can be solved by the old and the new paradigm” (Kuhn, 1970 p. 85).

 Several recent studies hint the need for a paradigm shift at the age of self-learning AI technologies. For example, van den Broek et al. (2021) discuss what they coin as “The Machine Learning Paradigm.” They argue that “the introduction of ML [Machine Learning] to organizations has come with a new set of assumptions regarding how knowledge should be produced in organizations (Elish & Boyd, 2018; Kitchin, 2014). By transforming data into knowledge, ML systems are positioned as superior means for creating insights, suggesting that machines can produce knowledge independent of domain experts to improve understanding and inform action (Domingos, 2015; Siegel, 2013). This new paradigm suggests that knowledge can be produced through a set of activities different from traditional IT systems to realize new promises for organizations” (p. 1558). In similar vein, Faraj et al. (2018) highlight that “learning algorithms will transform expertise in organizations, reshape work and occupational boundaries, and offer novel forms of coordination and control. […] their rapid deployment requires scholarly attention to societal issues such as the extent to which the algorithm is authorized to make decisions, the need to incorporate morality in the technology, and their digital iron-cage potential.” 17

138 

J. Kotlarsky and I. Oshri

Fig. 4.3  The foundations of a new paradigm

 oward a Pluralistic Semiotics Paradigm T of Digital Objects The manifestation of self-learning AI stretches our understanding of digital objects. As we argued earlier, digital objects have traditionally been understood as maintaining a particular meaning, as shown in Fig. 4.2, depicting a semiotics system in which a signifier of a digital object (referent) connotes a specific signified (Mingers & Willcocks, 2014; Oshri et al., 2018; Peirce, 1932). The conventions for such semiotic systems are enacted through a set of procedures designed by human actors, thus clearly defining the representation of the digital object (the signified) and the nature of the relationship between the signifier and the signified, which may or may not include pre-programmed computational processes of some sort. An icon of a scanner on the computer screen represents a digital object (scanner) with functionality to convert an analog to a digital image. An online mortgage application form is a representation (signifier) characterized by a one-to-one relationship with the signified (a function of approval or rejection of the application). However, AI digital objects have been observed self-creating an ever-changing signifier and convention, as shown in Fig. 4.3. Therefore, we argue that a new paradigm should offer a path to advance our understanding of self-learning and self-generating digital objects such as AI. We posit that the current paradigm of understanding digital objects, which is a particular semiotics system, is experiencing a transition. Taking

4  A Paradigm Shift in Understanding Digital Objects in IS… 

139

the assumption that the meaning of a digital object with self-learning and self-generating code abilities may not be understandable through conventions known to human actors, as depicted in Fig. 4.3, we propose a transformative paradigm approach (Hassan & Mingers, 2018) to guide the construction of a new paradigm. The foundations of the new paradigm are anchored in elements from metaphysical, sociological, and artifactual paradigms and offer guidance to human actors as they design, interact with, or manipulate AI digital objects. The new paradigm, termed a pluralistic semiotics paradigm, is anchored in the principle that human actors will accept the co-existence of multiple signifiers, signified, and unknown conventions generated by AI technologies. Yet, decoding the meaning of AI digital objects within a semiotics system will require human actors to re-examine the foundations upon which digital objects are currently understood. (In Appendix 2 we compare the particular and pluralistic semiotic paradigms.) By examining opportunities in metaphysical, social, and artifactual paradigms, we establish the foundations for the pluralistic semiotics paradigm.

The Metaphysical Paradigm Opportunity The concept of the metaphysical paradigm offers human actors the ability to reconsider what reality is and how it can be explained in the context of AI. At the heart of this idea is the competition between the world views of reality that human actors construct for themselves. Often constructed based on different beliefs and standards, such realities are thus accepted as not able to be integrated (Hassan & Mingers, 2018). The fundamental nature of AI as a digital object is that its meaning is bound to change without human actor intervention and often with little human actor ability to explain what has happened. This is counter to the current particular semiotics paradigm of digital objects. However, adopting a “gestalt switch,” a switch in our perspective on the relationships between the elements of a semiotics system (Hassan & Mingers, 2018), offers opportunities to develop a more accommodating semiotics paradigm but possibly an incomplete one. We envisage that a pluralistic semiotics paradigm will

140 

J. Kotlarsky and I. Oshri

assume new world views in which human actors accept a signified generated by an AI in an unknown manner (e.g., a new translation system created using conventions developed by AI through self-learning) and for which they are unable to detect the nature of the sign, so are oblivious to the convention at play. However, human actors can still perceive the series of elements as a whole, meaningful semiotics system by abstracting some parts (signifier or convention) and seeking the understandable (signified) rather than the unknown (convention). Human actors will need to fill in the blanks in this incomplete semiotics system; however, a pluralistic approach allows more than one explanation as to the nature of the signifier and the convention in play. Gestalt principles such as closure provide a way of imagining the completeness of the pluralistic semiotics paradigm, despite its incompleteness.

The Sociological Paradigm Opportunity The sociological paradigm offers an opportunity to revisit numerous facets of the social envelope of the existing paradigm, such as aspects concerning scientific progress or linguistic matters in the context of digital objects. From a linguistic perspective, the current use of digital object terminology (e.g., “material,” “nonmaterial,” “syntactic,” “bitstring,” and “bearer” [Faulkner & Runde, 2019]) re-institutionalizes the stable relationships between the building blocks of digital objects. Indeed, current categorizations of digital objects and their components do not accommodate a situation in which an object produces an unknown behavior or generates an unexpected outcome. There is an implied logic of causality in the structure of digital objects that is unable to accommodate the complexity and unpredictability AI can present. As a hybrid digital object, AI is likely to generate unexplainable behaviors and some of its components are likely to be evolving over time. A pluralistic semiotics paradigm offers opportunities for more fine-grained analysis of different types of digital objects (beyond syntactic objects, bitstrings, etc.) and for establishing relationships between these types of digital objects to account for instances that bring to the fore the changing nature of the key components. Using images and visual representations (Diaz et al., 2015) rather than words

4  A Paradigm Shift in Understanding Digital Objects in IS… 

141

can open up opportunities to address the rigidity of the particular semiotics paradigm. Images are deeply rooted in human minds, thus offering far more representational power than words. Conveying the meaning of digital objects using images is no less particular than words; it allows complex information to be encapsulated in a single visual representation. The theorization of IT/IS artifacts can thus benefit from visualizing digital objects of all sorts to expand the range of meaning attributed to the various types of digital objects, such as AI.

The Artifactual Paradigm Opportunity The quest to define and theorize the IT/IS artifact and digital object is in itself an opportunity to re-examine progress made in this area. While there have undoubtably been advances in recent years in understanding what a digital object is, current theorization of digital objects has also created a boundary around the phenomenon, thus potentially deterring IS scholars from considering a different approach to understand digital objects or seeking to extend current understandings of what digital objects are. Indeed, in our work, we have expanded the boundary around digital objects by including AI solutions and their unique characteristics. Furthermore, we have applied a semiotic view to reconsider the theory of digital objects and challenge the way the IS field understands such objects and the meanings attributed to them. To continue this exploration and further expand the boundaries around how we understand digital objects, additional analytical lenses and perspectives should be deployed when examining digital objects. Visual research and the principles of gestalt are among promising avenues for expanding understanding of digital objects.

Conclusions This chapter has examined whether the current understanding of what digital objects are accommodates the recent introduction of Artificial Intelligence technologies. With the assistance of a semiotics lens, we have argued that in IS, the meaning of digital objects is perceived as representing

142 

J. Kotlarsky and I. Oshri

stable relationships between the signifier, signified, and referent. However, this paradigm is being challenged by the characteristics of Artificial Intelligence, thus requiring the paradigm shift proposed in this chapter. We promote the idea of pluralism in the new paradigm as the rigidity of the current theorization of digital objects falls short in accommodating the adaptive and constantly updating nature of Artificial Intelligence technologies. We offer the concept of the transformative paradigm to provide guidance for future studies on how to further expand our understanding of digital objects and continue challenging paradigms in the IS field.

 ppendix 1: Definitions Based on IEEE Guide A for Terms and Concepts in Intelligent Process Automation (2017) Artificial intelligence includes the capability for creating unique hypotheses, attributing data relevance, processing data relationships, and updating its own lines of inquiry to further the usefulness of its purpose. An autonomic system self-manages its own operation and the processes it oversees by adapting to its environment as it changes, by interacting with neighboring systems and establishing communication protocols. Cognitive automation refers to the identification, assessment, and application of available machine learning algorithms for the purpose of leveraging domain knowledge and reasoning to further automate the machine learning already present in a manner that may be thought of as cognitive. With cognitive automation, the system performs corrective actions driven by knowledge of the underlying analytics tool itself, iterates its own automation approaches and algorithms for more expansive or thorough analysis, and is thereby able to fulfill its purpose. The automation of the cognitive process refines itself and dynamically generates novel hypotheses that it can likewise assess against its existing corpus and other information resources. Cognitive computing is complex computational systems designed to (i) Sense (perceive the world and collect data); (ii) Comprehend (analyze and understand the information collected); (iii) Act (make informed decisions and provide guidance based on this analysis in an independent way); and (iv) Adapt (adapt capabilities based on experience) in ways comparable to the human brain. Machine learning refers to detection, correlation, and pattern recognition generated through machine-based observation of human operation of software systems, along with ongoing self-informing regression algorithms for machine-based determination of successful operation leading to useful predictive analytics or prescriptive analytics capability.

4  A Paradigm Shift in Understanding Digital Objects in IS… 

143

 ppendix 2: Applying Core Elements A of Semiotics to Digital Objects: Shifting from a Particular to a Pluralistic Semiotic Paradigm of Digital Objects Pluralistic semiotic Particular semiotic paradigm paradigm of digital of digital objects: applying objects: applying core core elements of semiotics to elements of semiotics the theory of digital objects to environments involving AI digital objects Digital object can be non-­ Referent Digital with or without Physical object or physical presence material (i.e., bitstring, abstract concept, program file, or data file) or The referent can exist that is, something as a digital object hybrid (i.e., bitstrings with material or (e.g., a bot) or combined with other types non-material physical object (e.g., of material or non-material presence that is an autonomous car) components) represented by the that is enacted and signifier controlled by intelligent technologies Ever-changing signifier Signifier can be a symbolic, Signifier (also Representation of AI iconic, or indexical referred to as digital object is representation of the object, “representation”) constantly changing, or a combination of Can be a word, an as the object representations (e.g., a image, a number, a possesses self-learning sound, or smell that digital alarm clock has iconic and self-generating and indexical stands for the code abilities (e.g., representations; a song in an object or a concept. iTune’s library will have iconic program files, data It connotes the files, bitstrings) and symbolic signified and stands representations) for the referent. A signifier can be iconic, indexical, or symbolic Core elements of semiotics and their definitions (based on Peircean semiotics)

(continued)

144 

J. Kotlarsky and I. Oshri

(continued) Signified The signified expresses the meaning of the signifier

Multiple signifieds Signified—meaning of the connoting the same digital object, as interpreted signifier may exist. by a human actor or another Meaning is generated digital object in communications The signified that is a between AI digital computational outcome of objects: while this one digital object (e.g., the results of an Excel macro) can meaning is serve as a signifier of another understood by the interacting AI objects, digital object (e.g., it may be interpreted scheduling optimization differently by a algorithm) human agent (or could be meaningless to the human agent) Given that the signifier is changing, the meaning of the signified will also change over time as AI digital objects interact with each other. Thus, there will be new (emerging) and/or non-intended signifieds for human actors to attempt to understand (e.g., data interpretation and learning by autonomous vehicles that could lead to a car accident18) (continued)

18  https://www.theguardian.com/technology/2018/mar/19/uber-self-driving-car-kills-womanarizona-tempe

4  A Paradigm Shift in Understanding Digital Objects in IS… 

145

(continued) Conventions used to interpret Ever-changing Conventions conventions the meaning of the signified Agreement on the Conventions used by AI rules to use in order can be based on (i) social, digital objects are cultural, and professional to communicate conventions, (ii) syntactic and constantly changing, (e.g., Crystal, 2003; as they are re-created semantic rules of natural, Oshri et al., 2018) through learning and/ programming, and special-­ or communications purpose languages, and (iii) between AI digital logic, rules, and standards objects that are encoded/ programmed in the functionality of digital objects and their bearers. The last two types of conventions (ii and iii) can be embedded in computational processes

References Akhlaghpour, S., Wu, J., Lapointe, L., & Pinsonneault, A. (2013). The ongoing quest for the IT artifact: Looking back, moving forward. Journal of Information Technology, 28(2), 150–166. Arnold, S. J., Kuzinets, R. V., & Handelman, J. M. (2001). Hometown ideology and retailer legitimation: The institutional semiotics of Wal-Mart flyers. Journal of Retailing, 77, 243. Asatiani, A., Malo, P., Nagbøl, P. R., Penttinen, E., Rinta-Kahila, T., & Salovaara, A. (2020). Challenges of explaining the behavior of black-box AI systems. MIS Quarterly Executive, 19(4), 259–278. Asatiani, A., Malo, P., Nagbøl, P. R., Penttinen, E., Rinta-Kahila, T., & Salovaara, A. (2021). Sociotechnical envelopment of artificial intelligence: An approach to organizational deployment of inscrutable artificial intelligence systems. Journal of the Association for Information Systems, 22(2), 8. Bailey, D.  E., Leonardi, P.  M., & Barley, S.  R. (2012). The lure of virtual. Organization Science, 23(5), 1485–1504. Balasubramanian, N., Ye, Y., & Xu, M. (2020). Substituting human decision-­ making with machine learning: Implications for organizational learning. Academy of Management Review, 47, 448. https://doi.org/10.5465/ amr.2019.0470

146 

J. Kotlarsky and I. Oshri

Beynon-Davies, P. (2018). What’s in a face? Making sense of tangible information systems in terms of Peircean semiotics. European Journal of Information Systems., 27, 295–314. Chatterjee, S., Sarker, S., Lee, M. J., Xiao, X., & Elbanna, A. (2021). A possible conceptualization of the information systems (IS) artifact: A general systems theory perspective. Information Systems Journal, 31(4), 550–578. Crystal, D. (2003). The Cambridge encyclopedia of the English language (2nd ed.). Cambridge University Press. de Saussure, F. (1916/2003). Course in general linguistics. Diaz, A. A., Urquhart, C., & Arthanari, T. S. (2015). Seeing for understanding: Unlocking the potential of visual research in information systems. Journal of the Association for Information Systems., 16(8), 646. Domingos, P. (2015). The master algorithm: How the quest for the ultimate learning machine will remake our world. Basic Books. Elish, M. C., & Boyd, D. (2018). Situating methods in the magic of big data and AI. Communication Monographs, 85(1), 57–80. Faraj, S., Pachidi, S., & Sayegh, K. (2018). Working and organizing in the age of the learning algorithm. Information and Organization, 28(1), 62–70. Faulkner, P., & Runde, J. (2019). Theorizing the digital object. MIS Quarterly, 43(4), 1279–1302. Hassan, N. R., Lowry, P. B., & Mathiassen, L. (2022). Editorial-useful products in information systems theorizing: A discursive formation perspective. Journal of the Association for Information Systems, 23(2), 418–446. https:// aisel.aisnet.org/jais_preprints/27/ Hassan, N. R., & Mingers, J. (2018). Reinterpreting the Kuhnian paradigm in information systems. Journal of the Association for Information Systems, 19(7), 568–599. IEEE Standards Association. (2017). IEEE guide for terms and concepts in intelligent process automation (pp. 2755–2017). IEEE Std. Iivari, J. (2017). Information system artefact or information system application: That is the question. Information Systems Journal, 27(6), 753–774. Kallinikos, J. (2011). Governing through technology: Information artefacts and social practice. Palgrave Macmillan. Kitchin, R. (2014). Big data, new epistemologies and paradigm shifts. Big Data & Society, 1(1), 1–12. Kuhn, T. (1970). The structure of scientific revolutions (2nd ed.). University of Chicago Press.

4  A Paradigm Shift in Understanding Digital Objects in IS… 

147

Lawes, R. (2002). Demystifying semiotics: Some key questions answered. International Journal of Market Research., 44, 251–264. Lebovitz, S., Levina, N., & Lifshitz-Assaf, H. (2021). Is AI ground truth really “true”? The dangers of training and evaluating AI tools based on experts’ know-what. MIS Quarterly, 45(3), 1501–1525. Lee, A. S., Thomas, M., & Baskerville, R. L. (2015). Going back to basics in design science: From the information technology artifact to the information systems artifact. Information Systems Journal, 25(1), 5–21. Lou, B., & Wu, L. (2021). AI on drugs: Can artificial intelligence accelerate drug development? Evidence from a large-scale examination of bio-pharma firms. MIS Quarterly, 45(3), 1451–1482. Masterman, M. (1970). The nature of a paradigm. In I. Lakatos & A. Musgrave (Eds.), Criticism and the growth of knowledge: International colloquium in the philosophy of science (Bedford College, 1965) (pp.  59–89). Cambridge University Press. Mayer, A.  S., Strich, F., & Fiedler, M. (2020). Unintended consequences of introducing AI Systems for Decision Making. MIS Quarterly Executive, 19(4), 239–257. Mikhaeil, C. A., & Baskerville, R. L. (2019). Using semiotics to analyze representational complexity in social media. Information and Organization, 29(4). Mingers, J. (2001). Combining IS research methods: Towards a pluralist methodology. Information Systems Research, 12(3), 240–259. Mingers, J. (2003). The paucity of multimethod research: A review of the information systems literature. Information Systems Journal, 13(3), 233–249. Mingers, J. (2014). “Guidelines for Conducting Semiotic Research in Information Systems.” Kent Business School, University of Kent. Mingers, J., & Willcocks, L. P. (2014). An integrative semiotic framework for information systems: The social, personal and material worlds. Information and Organization., 24, 48–70. Mingers, J., & Willcocks, L. P. (2017). An integrative semiotic methodology for IS research. Information and Organization, 27(1), 17–36. Nicolini, D. (2012). Practice theory, work and organization: An introduction. . Orlikowski, W. J., & Iacono, C. S. (2001). Research commentary: Desperately seeking the ‘IT’ in IT research: A call to theorizing the IT artifact. Information Systems Research, 12(2), 121–134. Oshri, I., Henfridsson, O., & Kotlarsky, J. (2018). Re-representation as work design in outsourcing: A semiotic view. MIS Quarterly, 42(1), 1–23. Peirce, C. S. (1932). Elements of logic. Harvard University Press.

148 

J. Kotlarsky and I. Oshri

Rai, A., Constantinides, P., & Sarker, S. (2019). Editor’s comments: Next-­ generation digital platforms: Toward human–AI hybrids. MIS Quarterly, 43(1), iii–x. Schneider, W., (1987). Connectionism: Is it a paradigm shift for psychology?. Behavior Research Methods, Instruments, & Computers, 19, 73–83. Siegel, E. (2013). Predictive analytics: The power to predict who will click, buy, lie, or die. John Wiley & Sons. Stamper, R. (2001). Organisational semiotics: Informatics without the computer? In K. Liu, R. Clarke, P. Andersen, & R. Stamper (Eds.), Information, organization and technology: Studies in Organisational semiotics (pp. 115–172). Kluwer. van den Broek, E., Sergeeva, A., & Huysman, M. (2021). When the machine meets the expert: An ethnography of developing ai for hiring. MIS Quarterly, 45(3), 1557–1580. Venkatesh, V., Brown, S., & Bala, H. (2013). Bridging the qualitative-­ quantitative divide: Guidelines for conducting mixed methods research in information systems. MIS Quarterly, 37(1), 21–54. Warschauer, M., & Grimes, D. (2007). Audience, authorship, and artifact: The emergent semiotics of web 2.0. Annual Review of Applied Linguistics, 27, 1–23.

5 A Cybernetic Theory of the Impact of Implementers’ Actions on User Resistance to Information Technology Implementation Suzanne Rivard

and Liette Lapointe

Introduction The management of information technology (IT) implementation projects is often depicted as a set of “management controls needed to impose discipline and coordinate action to ensure goals are met” (Nidumolu & Subramani, 2003, p. 159). Several researchers have adopted a control perspective to study the management of IT projects (Henderson & Lee, 1992; Kirsch, 2004; Nidumolu, 1996; Nidumolu & Subramani, 2003). This is a shortened, updated and revised version of a paper published originally as Rivard, S., Lapointe, L., “A Cybernetic Theory of the Impact of Implementers’ Actions on User Resistance to Information Technology Implementation,” Proceedings HICSS 43, Kauai, January 2010.

S. Rivard (*) HEC Montréal, Montréal, QC, Canada e-mail: [email protected] L. Lapointe Desautels Faculty of Management, McGill University, Montreal, QC, Canada e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. P. Willcocks et al. (eds.), Advancing Information Systems Theories, Volume II, Technology, Work and Globalization, https://doi.org/10.1007/978-3-031-38719-7_5

149

150 

S. Rivard and L. Lapointe

The term “control” is used here as it is defined in management research, which conceptualizes organizational control as a set of mechanisms to ensure that an organization moves toward its objectives (Ouchi, 1979). In an IT implementation project, the implementers–either managers of organizational units or IS professionals—are responsible for exercising control to ensure that the organization’s objectives for the project are met. In a review of this research stream, Kirsch (2004) comments that researchers have “typically targeted pre-defined controller-controlee dyads, emphasized control relationships within IS units, studied control modes needed to achieve preidentified project goals such as on-time and within-budget system delivery, examined modes of control singly rather than simultaneously, and taken a ‘static’ or ‘snapshot’ view of control” (Kirsch, 2004, p.374). She also deplores the fact that although these studies have produced important insights, they are limited in how well they capture control in a non-routine, complex, and dynamic setting. In this chapter, we conceptualize control in IT implementation projects in a way that departs from how it has traditionally been examined by IT researchers and addresses the concerns Kirsch raised. We (1) propose a theory of control relationships as they evolve outside IS units, (2) theorize how control modes influence resistance behaviors rather than the project timeline or its budget, and (3) acknowledge the dynamic nature of the environment in which control is exercised. More specifically, we propose a theory of the dynamics of implementers’ control on user resistance during an IT implementation. Our theory is based on General Systems Theory (GST); it conceptualizes implementers as a control mechanism whose objective is to maintain the level of user resistance within an acceptable range from an organizational point of view. This chapter contributes to the book by illustrating Law as a product of theorizing (see Table 2.1 in Chap. 2). Cybernetics is based in part on the Law of Requisite Variety which states that in order to control the system, the control mechanism has to have as many types of responses as there are states in the system. In order to control resistance, many types of responses (negative feedback—corrective feedback; positive feedback—reinforcing feedback) are required. This use of Law as a product of theorizing suggests several propositions.

5  A Cybernetic Theory of the Impact of Implementers’ Actions… 

151

In IT research, although many identified resistance as a critical concern during IT implementation, relatively few authors have specifically studied the phenomenon of IT implementers’ responses to user resistance and the effect of these responses. Indeed, the extant models proposed to explain user resistance are user-centric in nature in that they include antecedents closely related to the immediate user environment: either the users themselves or their immediate work system (Jasperson et al., 2005). Our study goes beyond the immediate user environment by focusing on the implementers’ actions as control mechanisms. It better explains the relationships between implementers’ actions and user resistance.

Theoretical Background Our theory is grounded in the literature on user resistance to IT implementation and mobilizes cybernetics as a theorizing lens.

User Resistance to IT Implementation Resistance to IT implementation is said to occur when users feel threatened by the system being implemented or its effects on their environment. Resistance materializes in a variety of behaviors that can be covert (e.g., being passively uncooperative (Marakas & Hornik, 1996) or overt (e.g., attacking the credibility of the implementers (Prasad & Prasad, 2000) or voluntarily committing errors when using the system (Ferneley & Sobreperez, 2006). Resistance behaviors can be classified into four categories that differ in terms of the intensity of the resistance: apathy (e.g., inaction and lack of interest toward the system being implemented), passive resistance (e.g., delay tactics, excuses, persistence of former behaviors), active resistance (e.g., voicing opposite points of view, forming coalitions), and aggressive resistance (e.g., engaging in sabotage, infighting, and making threats) (Lapointe & Rivard, 2005). Although user resistance is sometimes portrayed as a barrier that ought to be removed, we espouse the view, advocated in previous research, that

152 

S. Rivard and L. Lapointe

resistance is neither good nor bad (Hirschheim & Newman, 1988; Lapointe & Rivard, 2005; Marakas & Hornik, 1996; Markus, 1983; Martinko et al., 1996). Indeed, there are times when resistance is a means by which users communicate the fact that problems exist with the IT being implemented or with its effects; in such instances, resistance is functional. At other times, when it prevents the adoption of an IT that could benefit the organization, resistance becomes dysfunctional. In either case, implementers have to deal with this resistance. In previous studies, resistance to IT is deemed a complex phenomenon that cannot be reduced to a simple rejection of a new technology and has been conceptualized as the result of a complex interaction among a number of antecedents. Several authors have proposed models to explain how resistance to IT implementation develops. For instance, Markus (1983) portrays resistance as resulting from the interaction of system features with the intraorganizational distribution of power. Hirschheim and Newman (1988) state that the causes of resistance are multiple and diverse and that they occur “in a tangle of different threads” (Hirschheim & Newman, 1988, p. 400). Joshi (1991) proposes a model that posits that resistance stems from negative users’ assessments of the fairness of the exchange between their inputs and the outcomes of their interaction with a given technology. Marakas and Hornik (1996) argue that passive resistance misuse is situational, the result of the interaction between the uncertain conditions created by the introduction of a new system and individual traits. Martinko et al. (1996) suggest that resistance to IT depends on the interaction of several factors: internal and external influences, as well as the individual’s prior experience with the technology. Lapointe and Rivard (2005) conceptualize resistance to an IT implementation as behaviors that occur following perceptions of threats associated with the interaction between an object and initial conditions. Finally, Ferneley and Sobreperez (2006) suggest that resistance can be either positive or negative and that it often manifests itself in user workarounds or deviations from set procedures. They propose a dynamic model in which four antecedent conditions play a key role: enforced proceduralization, organizational and personnel issues, discipline, and non-engagement with the system. The authors suggest that any of the four conditions can

5  A Cybernetic Theory of the Impact of Implementers’ Actions… 

153

lead to resistance, be it positive or negative, which in turn may result in other kinds of workarounds. The literature is quite sparse in terms of identifying implementers’ actions in response to user resistance and of the effect of these actions on the level of intensity of the resistance. Indeed, in reviewing the literature, we did not find any study that focused on the implementers’ reactions to user resistance to IT. We nevertheless identified several such reactions that authors mentioned in discussions of other aspects of resistance to IT. Most IT implementers’ reactions to user resistance mentioned in the literature pertain to reactions intended to improve the situation in a supportive manner. We grouped these reactions under the label remedial reactions. They include actions such as unearthing the causes of resistance and determining which corrective actions can be undertaken (Joshi, 1991; Lauer & Rajagopalan, 2002). Remedial reactions also include efforts to divert users from exhibiting resistance behaviors. Some strategies are aimed at changing individuals’ perceptions of the system being implemented or its environment (Martinko et al., 1996) through training, communication, and fair procedures (Joshi, 1991). Strategies to influence users’ attitudes have also been considered Melone (1995). They include: reciprocity: granting a favor to users and expecting that corresponding advantages or privileges will be returned; commitment and consistency: binding users by making them assume a position that is aligned with the desired behavior; social proof: showing users that prominent others have already accepted the IS; liking: putting esteemed individuals in positions of responsibility in the implementation process; and scarcity: making information about the system scarce and granting users privileged access to it so as to induce positive feelings toward the IT. Other remedial actions are aimed at modifying: the system being implemented (e.g., IT redesign or restructuring [Martinko et al., 1996]), the user environment (e.g., adding temporary awards and job reclassification [Joshi, 1991]), or some characteristics of the users themselves (e.g., training to reduce learning effort and frustration [Joshi, 1991]). The literature suggests that IT implementers will sometimes exert pressure on users to force them to stop resisting. We grouped these reactions under the heading of antagonistic reactions. They include actions such as

154 

S. Rivard and L. Lapointe

isolating pockets of resistance to prevent resistance from spreading “into a full-blown mutiny or coup” and forcing implementation (Lauer & Rajagopalan, 2002, p. 1302) through authority—using formal power to ensure user compliance (Melone, 1995). We created a third category of IT implementers’ reactions, which we labeled lack of reaction. Although this type of reaction is not discussed much in the literature, Lauer and Rajagopalan (2002) hint at such inaction in their discussion of acceptance of and resistance to IT. Also, Lapointe and Rivard (2006) refer to management’s lack of response to user resistance to an IT implementation when they describe situations in which implementers explicitly chose to ignore users’ complaints about system features that they had deemed inappropriate.

Cybernetic Systems Cybernetics is part of General Systems Theory (GST), whose objective is twofold: “[GST] seeks to classify systems by the way their components are organized (interrelated) and to derive the ‘laws,’ or typical patterns of behavior, for the different classes of systems” (Buckley, 1968, p. xvi). Although GST dates to the 1940s (Von Bertalanffy, 1950), it is still considered an appropriate theoretical foundation for studying organizations in general (Scott, 2003) and IT in particular. Boulding’s (1967) typology of systems is widely acknowledged in GST (Scott, 2003). In this typology, systems vary according to the level of complexity of their parts and the nature of the relations among the parts. The typology organizes nine types of systems within a hierarchy of complexity. The level of least complexity is that of frameworks, which correspond to static structures, such as the anatomy of a human or animal or the arrangement of atoms in a crystal. The second level, clockworks, is the level of simple dynamic systems with predetermined motions, such as the lever and the pulley. The third level, cybernetics, represents systems “capable of self-regulation in terms of externally prescribed targets or criteria such as a thermostat” (Scott, 2003). Above this are then open systems, blue-printed growth systems, animal level, human level, social systems

5  A Cybernetic Theory of the Impact of Implementers’ Actions… 

155

and transcendental systems. Our theory is situated at level 3, the cybernetics level. The contribution of cybernetics has been to link control mechanisms as studied in natural systems to those engineered in manmade systems (Checkland, 1981). Cybernetic systems are characterized by the notion of control, which has as its primary requirement the need to maintain the level and kind of output necessary to achieve the system’s objectives (Johnson et al., 1973). More specifically, control consists in stabilizing the outputs produced by a system so that the latter reaches a steady state in which its outputs show only small, random variations around the desired value (Davis & Olson, 1985). As shown in Fig. 5.1, there are four basic elements of control. The first is a characteristic or condition to be controlled: that is, the variable in the system’s behavior that has been chosen to be monitored and controlled (Schoderbek et al., 1980). The second element is the sensing function, which involves measuring the value of the characteristic or condition to be controlled. The third element is comparing, which consists in weighing the value of the characteristic against the objective to determine whether the value falls within an acceptable range (Schoderbek et al., 1980). The fourth element is a corrective function, which consists in evaluating the significance of the variation, determining whether the system is under control or out of control, and evaluating alternative corrective inputs that can be fed back to the system to restore stability (Johnson

Fig. 5.1  Cybernetic control

156 

S. Rivard and L. Lapointe

et al., 1973). Although stability is the long-term goal of the control process, authors contend that short-term and periodic instability is necessary for system adaptation and learning (Schoderbek et al., 1980). The corrective inputs are referred to as feedback. There are two types of feedback. The first is negative feedback; its effect is to dampen or reduce fluctuations around the desired values of system outputs. The second type is positive feedback; its effect is to reinforce the direction in which the system is moving. Negative feedback leads the system to reach a steady state, while positive feedback may lead to system instability. Indeed, if the output of the system falls outside the range of acceptable values, positive feedback will reinforce the system’s behavior, which may lead to instability and even to the destruction of the system (Toates, 1981). Note that the terminology employed here may appear counterintuitive. Indeed, negative feedback tends to have a positive net effect in that it contributes to reaching the targeted values for the system outputs. Positive feedback tends to have a negative effect in that it contributes to moving away from the targeted values for the system output. The ability of the control mechanism to exercise appropriate control over the system is subject to the Law of Requisite Variety (Ashby, 1963). This Law states that to control the system, the control mechanism must have as many types of responses as there are states in the system. These responses can be preprogrammed, provided by decision rules, or generated by the control mechanism’s ability to generate control responses (Davis & Olson, 1985).

 Cybernetic Theory of the Impact A of Implementers’ Actions on User Resistance to IT Implementation As per Gregor’s (2006) taxonomy of theories in IS, we propose a theory for explaining and predicting the impact of implementers’ reactions to user resistance behaviors on the intensity of user resistance following the enactment of such reactions. The focal construct of our theory is user resistance: more specifically, the intensity of user resistance behaviors. We set the boundaries of our

5  A Cybernetic Theory of the Impact of Implementers’ Actions… 

157

theory as follows. Although resistance and acceptance have been said to be at either end of a continuum of IS adoption, an explanation of the antecedents of the acceptance end of the continuum is outside the boundaries of our theory. Indeed, while we recognize that some of the implementers’ actions may have the effect of increasing user acceptance of a system, the proposed theory does not pertain to those relationships. Also, our theory is limited to the effect of a single antecedent of user resistance: the implementers’ actions. Finally, the theory does not aim to explain the mechanism by which implementers’ actions influence user resistance. Rather, it focuses on the resulting level of user resistance. The theory is dynamic in nature. It conceptualizes an IT implementation as a limited system constituted of users interacting with an IT application—one that is either in the process of being implemented or has already been implemented. As shown in Fig. 5.2, the implementers are the system’s control device, and their objective is to keep the intensity of user resistance within a range that is acceptable for the organization, that is, at a level that does not create organizational disruptions. As in any cybernetic system, the control device comprises three basic functions: sensing, comparing, and correcting. As a control device, implementers have the ability to sense the level of intensity of user resistance behaviors. Whether or not the implementers will accurately assess the level of resistance depends on the acuteness of their sensing function. The implementers also have a comparing function, which assesses the

Fig. 5.2  Cybernetic control by IT implementers

158 

S. Rivard and L. Lapointe

intensity of resistance behaviors against values that are deemed organizationally acceptable. Once again, the accuracy of this assessment is likely to vary with the implementers’ competence or experience. After comparing, the implementers engage in a correcting function, which consists in assessing the scale of the gap between the intensity of resistance and an acceptable level and determining whether the system is in or out of control. In the latter case, implementers will evaluate alternative corrective actions, which will become inputs fed back into the system to restore stability. Our theory posits that some implementers’ responses to user resistance behavior have the effect of negative feedback; that is, they dampen or reduce fluctuations around acceptable levels of intensity of user resistance. Other implementers’ reactions to user resistance behavior, however, have the effect of positive feedback; they reinforce user resistance and intensify resistance behaviors, which may lead to organizational disruption and, eventually, to project abandonment. The relationships between implementers’ actions and the intensity of user resistance behaviors are synthesized in five propositions. In order to provide insights into the relationships between these constructs, we provide examples from the literature, with cases describing IT implementations. Although the main focus of these cases is not the management of resistance per se, the vignettes extracted nevertheless provide some evidence for our propositions. Proposition 5.1  Implementers’ remedial reactions to user resistance to an IT implementation have a negative feedback effect on user resistance. In Proposition 5.1, the implementers’ three control functions (sensing, comparing, and correcting) operate appropriately. Indeed, implementers effectively perceive the value of the level of resistance, properly determine whether it is within the range of acceptable values, and decide on the appropriate corrective inputs, which they then feed back into the system. Implementers’ responses are said to have a negative feedback effect when their impact is to reduce the level of user resistance behaviors. This is what is to be expected when implementers’ reactions to users’ resistance

5  A Cybernetic Theory of the Impact of Implementers’ Actions… 

159

behaviors are remedial in nature. In the literature, we found several case studies that illustrate this negative feedback effect. A first illustration is that of the case related by Massaro (1993) of the implementation of a medical information system at the University of Virginia Medical Center. Following the hospital-wide installation of a system to manage laboratory functions, the medical staff voiced negative responses. Concerned, the developers chose to deactivate the features that the users did not like and worked closely with the medical staff to accommodate their demands, which ultimately resulted in less user resistance. Another illustration is found in Joshi (2005), who reported the implementation of an order management system in a custom furnishing organization. In this case, the production personnel initially expressed reservations about the new system. Indeed, these employees saw the system as an additional burden, which led to unfavorable initial judgment and resistance. The implementers were able to successfully communicate the value of the system, explaining how crucial it was for the competitive survival of the organization. In the end, the benefits of the system became clear to the production staff, and this led to less resistance. Finally, Prasad and Prasad (2000) present a detailed account of the implementation of a new computer system in a Health Maintenance Organization. Following the introduction of the system, some employees—who feared that the system would be detrimental—tried to disrupt the implementation process by flooding the basement storage room, resulting in the destruction of some display terminals. In response, the managers—wanting to minimize resistant actions—made informal concessions, such as turning a blind eye toward unpunctuality, allowing some employees to leave the organization earlier than usual, and permitting more flexible work schedules. These compromises appeased the employees and helped reduce user resistance. Proposition 5.2  Implementers’ antagonistic reactions to user resistance to an IT implementation have a positive feedback effect on user resistance.  In Proposition 5.2, while the implementers’ sensing and comparing functions may be appropriate, the correcting function is deficient, and the implementers’ reactions do not operate appropriately. Indeed, the

160 

S. Rivard and L. Lapointe

antagonistic reactions of the implementers have a positive feedback effect in that they reinforce the users’ resistance behaviors. Such are the situations illustrated by the following examples from the extant literature. A first illustration is provided by the case of the implementation of TransLease, an inter-organizational information system designed to facilitate electronic commerce between motor vehicle leasing companies and repair agents (Allen et al., 2000). In this case, the target users were the repair agents. When the repair agents experienced problems with TransLease, their use of the system declined. In response, the motor vehicle leasing companies decided to employ coercive power and attempted to leverage their buying power to force the repair agents to use the system. As a result, organizational disruption was observed: the relationships between the lease companies and the repair agents degraded. There was a significant increase in the level of distrust between the two groups and increased user resistance. Another example is provided by Hirschheim and Newman (1988) in their Warwick case study, which describes the implementation of a system for policy processing (SPP) at the Cambridge branch of a medium-­ sized insurance company. With time, the underwriters, who were the target users, stopped using the computerized system and bypassed the changes by continuing to use printed documents. The project manager then decided to impose system use on her underwriters “by removing the top sheet of the underwriting document, forcing her staff to retrieve the information from SPP” (p. 403). This form of coercive strategy led to greater user resistance. Finally, Newman and Robey (1992) also provided an illustration of such a positive feedback effect in the case of the implementation of a student information system at Middleton State University. Here, the staff of the undergraduate program was criticizing the new system’s features while continuing to use the former system. In reaction, the implementers decided to turn off the old system, hoping that it would force the staff to use the new system. This led the staff to petition a formal complaint to the administration, an indication of greater resistance. Proposition 5.3  A lack of reaction on the part of implementers to user resistance to an IT implementation has a positive feedback effect on user resistance. 

5  A Cybernetic Theory of the Impact of Implementers’ Actions… 

161

In Proposition 5.3, any of the implementers’ control functions may be inappropriate. In some instances, the sensing function is deficient, and the implementers are not aware of the resistance. In other situations, the sensing function is appropriate, but the comparing function is not; the implementers are thus not sensitive enough to recognize that resistance does not fall within the range of acceptable values. Finally, there will be some instances in which the sensing and comparing f­unctions will be appropriate, but because the correcting function is deficient, the implementers cannot identify what are appropriate corrective inputs. A lack of reaction to user resistance has a positive feedback effect in that it reinforces users’ resistance behaviors. As illustrated in the following examples from the extant literature, implementers’ non-response to user resistance behaviors will be followed by increased resistance. This course of action is clearly illustrated by Silva and Backhouse (2003) in their case presenting the implementation of a new administrative information system for the management of research project contracts at the Center for the Study of Food and Nutrition Sciences of Central America. In this case, initial mild resistance from the researchers was ignored by the implementers. The case reports that some derisory comments were made by the researchers after the system was implemented but that no action whatsoever was taken in response. The case goes on to show that after a few months, the researchers were becoming increasingly frustrated with the system. Resistance grew and with time, the humor turned into criticisms, formal complaints, and, eventually, actual threats. Another example is provided by Wilson and Howcroft (2002) in a case relating the implementation of the Zenith Nurse Management System at the North England Eldersite Hospital. After the system became available in the nursing wards, even those who had been enthusiastic about the implementation became disappointed with Zenith. Indeed, it was considered a hindrance, distracting the nurses from their work rather than improving it. Despite a “catalog of negative opinions concerning the system” (Wilson & Howcroft, 2002, p. 243), the implementers did not formally recognize the nurses’ concerns. In the end, the level of resistance grew, and the nurses reverted to using the manual files; the system even went unused in many wards.

162 

S. Rivard and L. Lapointe

Yet, another illustration is provided in a case relating the implementation of a computerized stock control system in Barrington’s chocolate factory (Rowe, 1985). When the system was introduced, the operations staff was irritated by its deficiencies. With time, the operators became increasingly frustrated and claimed that they had lost trust in the system. The managers “were bewildered and felt impotent to rectify the situation” (Rowe, 1985, p. 6). As a result, the level of resistance grew. The operators reverted to their former manual methods, and some sabotaging of the system was even suspected. Proposition 5.4  Over the course of an IT implementation, a succession of implementers’ reactions with positive feedback effects will lead to significant organizational disruption and, ultimately, require the abandonment of the IT implementation.  This proposition reflects the dynamic nature of our theory. As per this proposition, a single action with a positive feedback effect is not likely to lead to a situation requiring project abandonment. Indeed, under systems theory, although stability is the long-term goal of the control process, short-term and periodic instability may promote system adaptation and learning. For instance, as per Proposition 5.3, a lack of reaction to user resistance is likely to lead to greater resistance. Yet if, following this increase, the implementers appropriately sense, compare, and correct the situation, the level of resistance may very well decrease. What Proposition 5.4 suggests, however, is that if a succession of responses increases resistance, the intensity of such resistance may very well reach a “point of no return,” at which it may be too late to correct the situation, no matter what means the implementers may take. The literature helps illustrate this situation. For instance, Lapointe and Rivard (2005) report the case of a clinical information systems (CIS) implementation in a hospital. When the functionalities related to patient admission and radiology or laboratory requests and results were implemented, some nurses and physicians manifested some reluctance to use the system because they found it difficult to learn and use. The implementers—the hospital administration—did not explicitly address this reluctance and continued to pursue the project with the implementation

5  A Cybernetic Theory of the Impact of Implementers’ Actions… 

163

of the Nursing Notes module (lack of reaction). One particular feature of this new module was not well received by physicians: only they were allowed to enter test and treatment prescriptions. The hospital’s accepted practice—before the CIS implementation—had physicians dictating prescriptions to the nurses. Although the nurses appreciated this new feature, numerous physicians refused to comply and wanted to continue dictating prescriptions to nurses. Conflicts between the two groups resulted. The administrators took the side of the nurses and tried to impose system use on the physicians (antagonistic reaction). As a result, all the physicians signed a petition to have it removed. Ultimately, the module was withdrawn, and the hospital director had to resign. Proposition 5.5  As per the Law of Requisite Variety, implementers who have a wider range of responses to user resistance behaviors are more apt at exerting negative feedback type of actions.  Simply put, the Law of Requisite Variety refers to the fact that for a control mechanism to achieve its objective of maintaining the value of the system’s output within the range of acceptable values, the mechanism must possess responses that correspond to all the potential values of the system’s output. In an IT implementation context, the Law of Requisite Variety suggests that for implementers to respond to user resistance behaviors appropriately, they need to have the means to respond to a wide array of resistance behaviors. Proposition 5.5 suggests that implementers who possess a broader range of responses are more likely to react better to user resistance. For instance, the literature suggests that it is sometimes necessary to react to user resistance by modifying the work environment. This may mean modifying the rewards system, undertaking job reclassification, or placing esteemed individuals in positions of responsibility. This suggests that in addition to adequate sensing and correcting functions, implementers ought to have the legitimacy for implementing the corrective actions. This is illustrated by Wagner and Newell’s (Wagner & Newell, 2007) narration of the implementation of an enterprise system (ES) in a university. In this case, the target users were faculty members, whose first manifestation of resistance was a lack of interest in the ES. The implementers

164 

S. Rivard and L. Lapointe

involved at this point in time were IS professionals. They did not appear to pay much heed to this lack of interest and continued deploying the ES. Once the implementation was complete, the resistance became active. The whole university community of users “refused to work with the financial management module” (Wagner & Newell, 2007, p. 515) and the faculty formed a coalition, demanding meetings with the project staff and the provost to express their discontent. Resistance only diminished after the provost intervened, demanding that the project team negotiate a workable solution with faculty members.

Conclusion In this chapter we have used cybernetics—a part of General Systems Theory—to propose a dynamic theory of the impact of implementers’ actions on user resistance to IT implementation. We advanced five theoretical propositions in which implementers are conceptualized as a control mechanism. Our propositions explicate how implementers’ responses to user resistance behavior can have either a positive or a negative feedback effect on levels of resistance, either reinforcing or tempering fluctuations around acceptable levels of intensity of user resistance. To advance these theoretical insights, we propose two potential future research strategies. One promising research strategy would be to look for rich longitudinal data. These data could first be used for a theory testing purpose. They could also be used to search for new and potentially contradictory or conflicting evidence and as such serve as a basis for developing the theory further (Grover et al., 2008). One research method that could detect such rich data is the case survey strategy. Sometimes labeled “case meta-­ analysis,” this research method requires the systematic selection, coding, and analysis of written case studies (Larsson, 1993). Given the availability of so many rich case studies of IT implementation in the extant literature, a case survey strategy would open a path for the interpretation of empirical data and for the close examination of resistance behaviors, implementers’ actions and reactions, and the feedback effect of these actions on resistance, as well as the potential identification of new constructs or relationships.

5  A Cybernetic Theory of the Impact of Implementers’ Actions… 

165

A second promising research avenue would be to use our theoretical propositions as a basis for formal hypotheses on the effect of implementers’ actions on user resistance behavior. Given the system-theoretical foundations of the model, simulations of the systems’ behaviors (Black et al., 2004) could be conducted to test the hypotheses. Recently, simulation has been recognized as a methodological approach that is making significant contributions to successful theory development and refinement (Davis et al., 2007). The theory that we propose has some limitations that we wish to acknowledge. First, it might be argued that portraying an organizational system as a closed system—“encompassing a set of stable and easily identifiable participants” (Scott, 2003, p.28)—lacks a certain realism. Indeed, we recognize that many actors, other than those we have identified in our theory, may play roles in IT implementation. We would argue, however, that a given IT implementation project is sufficiently circumscribed that we can indeed identify the key participants and that the identity of those participants remains rather stable over the course of the project. Second, it might be suggested that a closed system does not appropriately capture the fact that organizations are “open to and dependent on flows of personnel, resources, and information from outside” (Scott, 2003, p. 28). Although we acknowledge that multiple environmental conditions other than implementers’ actions may influence user resistance—many of which have been identified in extant models of user resistance—we contend that the aim of our theory is to explain and predict the effect of implementers’ actions on user resistance rather than to provide a complete explanation of how resistance emerges and evolves and its source. Notwithstanding these limitations, we believe that our theory possesses enough explanatory power to warrant further investigation.

References Allen, D. K., Colligan, D., Finnie, A., & Kern, T. (2000). Trust, power, and interorganizational information systems: The case of the electronic trading community TransLease. Information Systems Journal, 10(1), 21–40. Ashby, W. R. (1963). Introduction to cybernetics. John Wiley & Sons.

166 

S. Rivard and L. Lapointe

Black, L. J., Carlile, P. R., & Repenning, N. P. (2004). A dynamic theory of expertise and occupational boundaries in new technology implementation: Building on Barley’s study of Ct scanning. Administrative Science Quarterly, 49, 572–607. Boulding, K. E. (1967). General systems theory - the skeleton of science. Wiley. Buckley, W. (1968). Modern systems research for the Behavioral scientist. Aldine Publishing Company. Checkland, P. B. (1981). Science and the Systems Movement. Group, T. O. S. (Eds.), Systems Behaviour, 26–43. Davis, G. B., & Olson, M. H. (1985). 1985. Management Information Systems. Davis, J. P., Eisenhardt, K. M., & Bingham, C. B. (2007). Developing theory through simulation methods. Academy of Management Review, 32(2), 480–499. Ferneley, E. H., & Sobreperez, P. (2006). Resist, comply, or workaround? An examination of different facets of user engagement with information systems. European Journal of Information Systems, 15, 345–356. Gregor, S. (2006). The nature of theory in information systems. MIS Quarterly, 30(3), 611–642. Grover, V., Lyytinen, K., Srinivasan, A., & Tan, B. C. Y. (2008). Contributing to rigorous and forward-thinking explanatory theory. Journal of the Association for Information Systems, 9(2), 40–47. Henderson, J. C., & Lee, S. (1992). Managing I/S design teams: A control theories perspective. Management Science, 38(6), 757–777. Hirschheim, R., & Newman, M. (1988). Information systems and user resistance: Theory and practice. The Computer Journal, 31(5), 398–408. Jasperson, J. E., Carter, P. E., & Zmud, R. W. (2005). A comprehensive conceptualization of post-adoptive Behaviors associated with information technology work system. MIS Quarterly, 29(3), 525–557. Johnson, R. A., Kast, F. E., & Rosenzweigh, J. E. (1973). The theory and management of systems. McGraw-Hill Kogakusha. Joshi, K. (1991). A model of User’s perspective on change: The case of information systems technology implementation. MIS Quarterly, 15(2), 229–240. Joshi, K. (2005). Understanding user resistance and acceptance during the implementation of an order management system: A case study using the equity implementation model. Journal of Information Technology Cases and Application Research, 7(1), 6–20. Kirsch, L. J. (2004). Deploying common systems globally: The dynamics of control. Information Systems Research, 15(4), 374–396. Lapointe, L., & Rivard, S. (2005). A multilevel model of resistance to information technology implementation. MIS Quarterly, 29(3), 461–491.

5  A Cybernetic Theory of the Impact of Implementers’ Actions… 

167

Lapointe, L., & Rivard, S. (2006). Learning from Physician’s resistance to information technology implementation. Canadian Medical Association Journal, 174(11), 1573–1578. Larsson, R. (1993). Case survey methodology: Quantitative analysis of patterns across case studies. Academy of Management Journal, 36(6), 1515–1546. Lauer, T. & Rajagopalan, B. (2002). Examining the Relationship between Acceptance and Resistance in System Implementation. (Eds), AMCIS 2002, 1297–1303. Marakas, G. M., & Hornik, S. (1996). Passive resistance misuse: Overt support and covert recalcitrance in is implementation. European Journal of Information Systems, 5(3), 208–220. Markus, M. L. (1983). Power, politics, and MIS implementation. Communications of the ACM, 26(6), 430–444. Martinko, M. J., Henry, J. W., & Zmud, R. W. (1996). An attributional explanation of individual resistance to the introduction of information Technologies in the Workplace. Behaviour & Information Technology, 15(5), 313–330. Massaro, T. A. (1993). Introducing physician order entry at a major Academic Medical Center: I. impact on organizational culture and behavior. Academic Medicine, 68(1), 25. Melone, N. P. (1995). When people work scared: Understanding attitudes and gaining commitment in business process reengineering. In V. Grover & W. J. Kettinger (Eds.), Business process change (pp. 475–492). Newman, M., & Robey, D. (1992). A social process model of user-analyst relationships. MIS Quarterly, 16(2), 249–266. Nidumolu, S. R. (1996). A comparison of the structural contingency and risk-­ based perspectives on coordination in software-development projects. Journal of Management Information Systems, 13(2), 77. Nidumolu, S. R., & Subramani, M. R. (2003). The matrix of control: Combining process and structure approaches to managing software development. Journal of Management Information Systems, 20(3), 159–196. Ouchi, W. G. (1979). A conceptual framework for the design of organizational control mechanisms. Management Science, 25(9), 833–848. Prasad, P., & Prasad, A. (2000). Stretching the iron cage: The constitution and implications of routine workplace resistance. Organization Science, 11(4), 387–403. Rowe, C. J. (1985). Identifying causes of failure: A case study in computerized stock control. Behaviour & Information Technology, 4(1), 63–72. Schoderbek, C. G., Schoderbek, P. P., & Kefalas, A. G. (1980). Management systems: Conceptual considerations. Business Publications.

168 

S. Rivard and L. Lapointe

Scott, W. R. (2003). Organizations: Rational, natural, and open systems. Prentice Hall. Silva, L., & Backhouse, J. (2003). The circuits-of-power framework for studying power in institutionalization of information systems. Journal of the Association for Information Systems, 4(6), 294–336. Toates, F. M. (1981). Homeostasis. In Group, T. O. U. S. (Ed.), Regulation and control (pp. 208–213). Systems Behaviour. Von Bertalanffy, L. (1950). An outline of general system theory. British Journal for the Philosophy of Science, 1(2), 134–165. Wagner, E., & Newell, S. (2007). Exploring the importance of participation in the post-implementation period of an es project: A neglected area. Journal of the Association for Information Systems, 8(10), 508–524. Wilson, M., & Howcroft, D. (2002). Re-conceptualising failure: Social shaping meets is research. European Journal of Information Systems, 11(4), 236–250.

6 Interrogating Sociomateriality: An Integrative Semiotics Framework for Information Systems John Mingers

and Leslie Willcocks

Introduction This chapter contributes to the book by questioning one framework— that of sociomateriality—and developing and advancing another—an integrative semiotics framework—as a product of theorizing. The two most distinctive characteristics that distinguish human beings from other animals are their advanced abilities to use language to co-­ ordinate their actions (Maturana, 1978; Mead, 1934) and to develop and use tools to shape their environment (Habermas, 1978). Tool-making, and language and signification are also the basis of knowledge as Habermas points out in his theory of knowledge-constitutive interests (Habermas,

J. Mingers Kent Business School, Canterbury, UK e-mail: [email protected] L. Willcocks (*) Department of Management, London School of Economics and Political Science, London, UK © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. P. Willcocks et al. (eds.), Advancing Information Systems Theories, Volume II, Technology, Work and Globalization, https://doi.org/10.1007/978-3-031-38719-7_6

169

170 

J. Mingers and L. Willcocks

1978)—the empirical-analytic sciences which underlie technology, and the hermeneutic sciences which underlie human communication. Language and communication are based fundamentally on meaning and signification which is essentially cognitive, while tools and technology are primarily realized in a physical form. It is interesting, therefore, to consider the possible relationship between these two, very different, systems—the social and the material. Here we narrow technology down to the domain that is relevant to information systems, indicated by the common mnemonic ICT, that is information and communication technology. This immediately reveals that it is about technology applied to communication; in other words, it immediately cuts across the communication/signification divide. This is nothing new, but undoubtedly with mobile technologies and social networking it is more so than ever before. In the literature we can distinguish three primary positions regarding this relationship. First, are those who emphasize the dominance of one system over the other. For example, technological determinists such as Woodward (1958) and Perrow (1970) argue that the nature of the technology imposes major constraints on individuals and organizations. More recent work is highly varied in scope and level but generally treats technology as an independent variable within the research (Orlikowski & Scott, 2008, p. 439–446). Perhaps in reaction to this view, there developed a focus on the social aspects of technology—the ways in which people organized around or shaped technology(Howcroft et al., 2004)— which includes the social shaping of technology (SST) and the social construction of technology (SCOT) (Pinch & Bijker, 1984) perspectives. These could be seen as social determinists. Second, there are those who conceptualize two ontologically distinct systems that interact and mutually influence each other (Orlikowski & Scott, 2008, p.  446–454), for example the original socio-technical studies of Trist and colleagues (Trist & Murray, 1993), Zuboff’s (1988), and Zammuto et al.’s (Zammuto et al., 2007) work drawing on Gibson’s theory of ecological perception and affordances. Third, and most recent, we find theorists who argue that the two

6  Interrogating Sociomateriality: An Integrative Semiotics… 

171

systems are so inextricably inter-­twined that they cannot in fact be separated, for example actor-network theory (ANT) (Callon, 1991; Latour, 1987), object-centred sociality(Knorr-Cetina, 1997) and relational materiality (Law, 2004). Within IS, this position has for some time now been called “sociomateriality” to emphasize the inseparability of the social and the material (Leonardi & Barley, 2008; Orlikowski, 2000; Orlikowski, 2007; Orlikowski & Scott, 2008). This chapter seeks to interrogate the usefulness and limitations of sociomateriality approaches through developing a more explanatory framework informed by many theoretical perspectives. To return to the beginning, we are concerned with the interaction of communication and technology. Communication is underpinned by systems of meaning and signification. The discipline that most thoroughly deals with signification is semiotics—the science of signs and sign transmission. If semiotics conceptualizes the mechanisms of meaning and communication, and we are aiming to explore the relations between communication and technology, then we wish in this chapter to consider the relation between semiotics and sociomateriality. At first sight this may not look promising since, by and large, semiotics has restricted itself to only one side of the equation—the social. However, we will argue that semiotics can relate more directly to technology through the concept of embodiment. In the first section of the chapter we review briefly Peircean semiotics, taking into account more recent developments relevant to our purpose, especially with regard to business and ICTs. In the next section we develop out of Peircean semiotics our theoretical position with regard to sociomateriality. This will be from an underlying critical realist (CR) position and will relate particularly to theories concerning the nature of information and the role of embodied cognition. From this build, we then develop an integrative framework that locates semiosis as the founding set of operations at the centre of three worlds—the personal, social and material. In the final sections, we discuss critically the implications for sociomateriality as a founding concept for IS, illustrating our formulation and use of the integrative framework with empirical examples. This

172 

J. Mingers and L. Willcocks

chapter draws upon, updates and complements our previous work (see (Mingers & Willcocks, 2014) and (Mingers & Willcocks, 2017)). Ilan Oshri and Julia Kotlarsky provide further thinking on semiotics in IS in Chap. 4. There is some overlap between the two chapters in the account of semiotics, but this is necessary as we need to show how the integrative semiotics framework evolved out of several theoretical inputs.

Developing the Framework (1) Semiotics Semiotics or semiology is the study of signs and systems of signification where a sign is an event, an object, a symbol or a behaviour that represents something other than itself. Signs depend upon a shared set of meanings within a particular community and are the basis of all communication, whether linguistic or not. Semiotics studies the processes that lead signs to have particular meanings and the ways in which such meanings are communicated and have effects. In many ways, semiotics can be seen as the most fundamental of the social sciences since it underlies all communication and social action. In recent history, semiotics has two significantly different lines of development, one traceable to Ferdinand de Saussure (1960), a Swiss linguist, and the other to Charles Sanders Peirce, an American philosopher and scientist. From the point of view of this chapter, there is a major limitation of Saussurian semiotics—it only involves a dyadic relation between signifier and signified and does not include reference to the world outside the sign system, that is the world of objects and events to which signs can refer. Here we shall be concerned primarily with Peirce’s approach, though noting that Saussure’s work also informed parts of Giddens’s (1984) structuration theory—one of the approaches used in developing the sociomateriality concept in IS studies (Orlikowski, 2000).1

 The history of semiotics has been covered extensively and is not repeated here. Good general accounts include (Chandler, 2002; Krampen et al., 1987; Martin & Ringham, 2006; Noth, 1990). 1

6  Interrogating Sociomateriality: An Integrative Semiotics… 

173

Representamen icon, index, symbol

reference

sense

The sign Object • •

Immediate object as represented within the sign Dynamic object as implied by or generang the sign

Interpretant • • •

Immediate interpretant – direct meaning of the sign Dynamic interpretant – the effect of the meaning on an interpreter Final interpretant – the eventual effect aer unlimited semiosis

Fig. 6.1  Peirce’s semiotic triangle

Peircean Semiotics For Peirce (1958),2 a sign involves a triadic relation between a representamen (signifier), an interpretant (signified) and an object (see Fig. 6.1). A sign … {representamen} is something which stands to somebody for something in some respect or capacity. It addresses somebody, that is, creates in the mind of that person an equivalent sign, or perhaps a more developed sign. That sign which it creates I call the interpretant of the first sign. The sign stands for something, its object. It stands for that object, not in all respects but in reference to a sort of idea, which I have sometimes called the ground of the representamen. (Peirce, 1958, 2.228, original emphasis)3

Peirce was primarily interested in the process of semiosis, that is the way in which signs were continually interpreted and re-interpreted within the  We should note that Peirce wrote extensively about semiosis over many years, often developing or changing his terminology, so there is not a single model or theory. For example, in his late work there is the suggestion that semiosis may not be unlimited. 3  References to Peirce are to the volume and paragraph in the Collected Papers (Peirce, 1958). Other sources are Buchler (1940), Almeder (1980) and Greenlee (1973). 2

174 

J. Mingers and L. Willcocks

process of communication. Considering Fig. 6.1 in more detail, the representamen is the physical manifestation of the sign—its form as opposed to its content (properly speaking “sign” should only refer to the whole combination of the three aspects, but sometimes it is used to refer just to the representamen). The representamen brings with it two effects—the object that it represents and the idea that it generates in an interpreter. The interpretant is not the interpreter per se but does imply that there is some form of interpreter. The interpretant is seen by Peirce as another sign, thus leading to the idea of continual semiosis. These two effects can be seen as the same as Frege’s (1952) distinction between sense and reference. The “meaning” of a sign thus consists in both its sense and reference. The object is that which the sign stands in place of and can be physical, mental, imaginary or another sign. Peirce developed these basic categories in several ways (Noth, 1990). He produced complex typologies of different types of representamen. The main one was to distinguish between icons, indexes and symbols in terms of their relationship to the object. Icons are signs that resemble or imitate their objects in some way, for example a picture, a model, or a simulation. Indexes relate to their objects directly, either causally or temporally. For example, a thermometer is an index of the temperature; the sun setting is an index of nighttime coming. Symbols have no direct relationship to their object at all; the association is purely conventional as in language or mathematical notation. Symbols have a relationship purely through the habit of their association. Peirce was very concerned with the way in which signs came to be interpreted in practice (Almeder, 1980). For the interpretant, he distinguished between the immediate interpretant and the dynamical interpretant.4 The immediate interpretant is the “quality of the impression that the sign is fit to produce and does not consist in any actual reaction” (8.315). It is thus the intrinsic meaning or interpretability of the sign before anyone has actually interpreted it. The dynamical interpretant is the “direct effect actually produced by a sign upon the interpreter of it”  He also talked of a final interpretant but was somewhat unclear about what this meant. He also sometimes talked of the emotional/energetic/logical interpretants, but there is debate about the relationship between the two schemes (Atkin, 2006). 4

6  Interrogating Sociomateriality: An Integrative Semiotics… 

175

(CP 4.536). This is the idea or meaning that the sign generates in a particular person, related to the object represented by the sign. It is, therefore, in part dependent on the interpreter and may differ from one to another. It also generates some effect in the interpreter, whether a physical action or a mental one such as another sign, which may in turn be expressed in a representamen. Thus, we have the process of semiosis or signification. Peirce also distinguished two forms of the object of the sign, also called the immediate object and the dynamical (or mediate) object: We have to distinguish the immediate object, which is the object as the sign itself represents it, and whose being is thus dependent upon the representation of it in the sign, from the dynamical object which is the reality which by some means contrives to determine the sign to its representation (CP 4.536).

This very important distinction commits Peirce to at least some form of realism, although not a naive realism. The immediate object is that which is contained within the sign and picks out certain aspects or grounds of the “real” object—the dynamical one. The latter is the underlying, but not immediately present, trigger of the sign. To illustrate with an example, if someone asks “Where is the bathroom?” the immediate object is the concept of a bathroom as expressed in the sign. In this case it is just a bathroom in general, not a particular one. The dynamical object is the actual bathroom (assuming there is one), with all its particular characteristics, which exists outside the world of the sign. The immediate interpretant is the meaning of the question as a whole that any speaker would understand, and the dynamical interpretant is the effect the question has on an interpreter which may lead them to give directions or ask someone else. Semiotics was only a part of Peirce’s extensive philosophical thought. We should also note for later that he was one of the founders of American pragmatism (Buchler, 1940; Peirce, 1878),(CP 5.411, 5.197, 5.597) and thus his theory of meaning was built on his semiotics. Expressions gain their meaning through their conditions of use, that is the effects that they have on the world, which is precisely the dynamical and final interpretants of a sign or message. Less well known is that he was also a

176 

J. Mingers and L. Willcocks

phenomenologist (Buchler, 1940) (CP 1.284-7, 1.536-7, to some extent pre-dating Husserl [1973 {orig. 1913}]) and developed a theory of three modes of being: firstness—pure quality or actuality in itself; secondness— relations between one thing or quality and another; and thirdness—cognition, concepts and laws about firstness and secondness. The sign embodies all three modes. One area that Peirce did not develop was the physical or technological aspects of semiotics and communication, which is obviously of importance for this chapter, and so we consider here some later developments.

Further Peircean Developments Charles Morris (1938) was an American behaviourist who developed semiotics as the science of all signs to include non-linguistic and non-­ human sign processes. He used a similar triad to Peirce, calling them the sign vehicle (representamen), the designatum (object) and interpretant. He therefore characterized semiotics in terms of three dimensions: syntactic which studies the relations between signs and other signs; semantics which studies the relations between signs and their objects; and pragmatics which studies the uses of signs by their interpreters. Syntactics, or syntax, covers all the formal relations between signs, including the rules of language, or the sign system, and the syntagmatic and paradigmatic dimensions of sign relations as developed by Saussure. Semantics is a polysemous term that is closely related to meaning. Initially, Morris saw it as specifically the relations between the sign and its object, that is reference or denotation, but later included the sense, or immediate interpretant, of the sign as well. Pragmatics covers “the origin, the uses and the effects of signs” (Morris, 1938, p. 30) which would include the biological, psychological and social aspects of the intentional use of signs (Austin, 1962; Habermas, 1979b; Searle, 1969). Later work within information systems by Stamper (1997) extended Morris’s typology. Below the level of syntactics, Stamper added the material level of physical phenomena that allow the storage and transmission of signs, and the empiric level which concerns the effective and efficient transmission of messages as is dealt with by traditional information theory (Shannon & Weaver,

6  Interrogating Sociomateriality: An Integrative Semiotics… 

177

1949). Both of these are useful in considering sociomateriality. He also added an upper level—the social—that went beyond individual intentionality to the effects of signification in the social and organizational worlds (Stamper, 2001). A second major figure was Roman Jakobson(1956) who contributed to both Saussurian and Peircean semiotics. He saw that Saussure’s distinction between the syntagmatic and paradigmatic axes was essentially based on the difference between metaphor and metonymy or, more basically, similarity and contiguity.5 A metaphor is a figure of speech, or trope, in which one element (or signified) stands for another on the basis of some similarity or likeness (e.g. “the university of life”). Metaphor is incredibly common in language. Indeed, Lakoff and Johnson (1980) argue that the origin of almost all language is metaphorical in relation to our basic physical experience of the world. Metonymy is another trope in which one signified stands for another on the basis of a direct relation, for example part/whole, cause/effect or substance/form (e.g. “I spy the sails,” “The mercury’s through the roof,” “I’ll pay on plastic”). In this, it is very similar to Peirce’s indexical type of sign. So a sign gains meaning through a combination of likeness and contiguity. Jakobson also developed a model that applied to any speech or communicative act (Jakobson, 1960) comprising of six components or elements, each of which leads to a different function of language (Fig. 6.2). These are the addresser (emotive or expressive), the addressee (conative or volitional), the context (referential or denotative), the message (poetic), the code shared by addresser and addressee (metalingual), and the physical or psychological contact (phatic). The first three of these formed an earlier model of language functions developed by Bühler (1982, orig. 1934) and picked up by Habermas (1994) as underpinning for his theory of communicative action. In relation to the Morris/Stamper typology, we can see that the emotive and conative functions concern pragmatics, the context concerns semantics, the poetic and metalingual concern syntax, and the phatic concerns the empiric and physical dimension which again brings in materiality.  Freud used a similar distinction in his analysis of dreams which he saw as being based on displacement (metaphor) and condensation (metonymy) (Freud, 2004 [orig. 1900]). 5

178 

J. Mingers and L. Willcocks

Context referential

Addresser

expressive

Message

conative

poetic

Addressee

metalingual

Code phatic

Contact Fig. 6.2  Jakobson’s six functions of semiotic systems

Semiotics in Business and ICT Semiotics has been used in a range of business areas, particularly, as might be expected, in marketing, for example Mick (1986), Arnold et al. (2001), Harvey and Evans (2001), and Lawes (2002). In other domains, Barley (1983) used semiotics as a way of studying the systems of meaning within particular occupations; Fiol (1989) analysed CEO’s letters to shareholders to understand a company’s propensity to enter into joint ventures; Brannen (2004) studied the cultural differences that can undermine an organization’s transfer of policies and process abroad; and Cooper et al. (2001) used semiology to decode the reviews of regulated utility companies produced in the UK. Moving to IS and IT, a considerable body of work has developed around Stamper’s (1991) extension of Morris’s framework mentioned above. This work generally goes under the name “organizational semiotics” (Gazendam et  al., 2003; Liu et  al., 2001, 2002a, 2002b), but it mainly concerns information systems and systems analysis. Work within this tradition is generally based on Peircean semiotics and ranges from studies of instrumentation and the human-computer interface (HCI) (May & Andersen, 2001) through the development of information systems having regard to both their technical and human aspects (Stamper,

6  Interrogating Sociomateriality: An Integrative Semiotics… 

179

2001) to studies of IS within their organizational context (Clarke, 2002). There are other sources used; for example Robichaud (2002) employs Greimas’s narrative grammar to analyse a process of public consultation in a Canadian city, and Kryssanov et al. (2003) use Luhmann’s autopoietic social theory in user interface design. Moving away from Stamper’s framework, there are several semiotic analyses of ICT as a communicational tool. For example, Warschauer and Grimes (2007) analysed Web 2.0 type software such as blogs, wikis and social networking sites in terms of semiotic constructs such as authorship, audience and artefact. Tredinnick (2007) also used post-­structuralist semiotics (e.g. Barthes, Foucault, Derrida) to study the effects of hypertextuality in the WWW.  Menchik and Tian (2008) used Peirce’s and Morris’s semiotic frameworks to analyse the ways in e-mail users overcome the exclusion of non-linguistic cues and gestures in e-mail interactions. Mancini and Buckingham Shum (2006) discuss a discourse representation system, based on semiotics, specifically for domains of debate and contestation such as academic discourse. Researchers who have approached information from a semiotic perspective include Beynon-Davies (2009), Huang (2006), Brier (2001), (Oshri et al., 2018), Raber and Budd (2003), and Price and Shanks (2005). The final area we shall discuss is the human-computer interface (HCI) where signs and symbols obviously play a central role. Here, Andersen (1990) coined the term “computer semiotics” by which he meant adapting the semiotic theories that had primarily arisen in linguistics to the specific domain of computing. In particular, he drew on both the structuralist tradition of semiotics as represented by Barthes and Eco (Ramussen, 1986) and the phenomenological/speech acts approach as represented by Winograd and Flores (Winograd & Flores, 1987). However, for the purposes of this chapter we wish to highlight a more recent trend that goes beyond structuralism or phenomenology to encompass the idea of embodiment. De Souza (2005) has developed a theory of semiotic engineering which sees HCI as enabling an active communication process between the system user and (implicitly) the system designer, and O’Neill (2008) has built on these ideas. This strand of thought will form part of the theory developed in a later section.

180 

J. Mingers and L. Willcocks

Developing the Framework (2): Critical Realism As we have seen, sociomateriality covers two potentially distinct domains—the social and ideational, on the one hand, and the physical, on the other hand—so it is important that these can be reconciled philosophically. For this reason, we begin by locating our work within the critical realist (CR) paradigm which accepts the ontological reality of a variety of different entities, be they physical, social, cognitive or abstract (Archer et al., 1998; Bhaskar, 1978; Bhaskar, 1979; Bhaskar, 1993). Such entities do not need to be measurable, or even directly observable, so long as we can postulate that they have causal effects. CR has been advocated as a philosophy for IS (Dobson, 2001; Mingers, 2004; Mutch, 1999) and used in empirical research (Bygstad, 2010; Smith, 2006; Volkoff et al., 2007; Wikgren, 2005; Wynn & Williams, 2008) so we will only highlight the significant aspects. The first is the distinction between the transitive and the intransitive domains of science and knowledge. Science (and social science) is a human activity and therefore much of it is a social production—theories, experiments, papers, journals, debates, etc.—are all human-dependent and therefore transitive. However, the objects of knowledge, that which knowledge is about, are external and independent of our knowledge of them—they are intransitive. This is not only true for the physical world, where it seems uncontentious that physical laws would operate even if humans did not exist, but also for the social world, even though social laws and mechanisms can only operate in and through people in general. Even speech can become intransitive once it has been uttered and become detached from the circumstances of its production. Second, there is a distinction between the real, the actual and the empirical. The real, that is everything there is, consists of enduring structures and mechanisms that have particular tendencies and powers generating causal effects in the world. These structures, which may be unobservable and may not exercise their powers all the time, interact with each other and generate the actual events that do (and do not) occur. Some of these events are observed and experienced, and have the potential to become the empirical data of science. Both the actual and the empirical are part of the real and have causal effects of their own. CR also

6  Interrogating Sociomateriality: An Integrative Semiotics… 

181

emphasizes the idea of generative causality in opposition to the Humean version of a constant conjunction of events. And it is comfortable with the view that reasons can be causes, that is that the reasons an actor gives for their actions may be an adequate explanation although we must always be aware of unknown or perhaps unacknowledged conditions of action. With regard to semiotics, critical realism has already recognized its importance although it is not that well developed. Bhaskar himself says that “the centerpiece of any adequate theory of meaning must be the semiotic triangle” (Bhaskar, 1993, p. 222–223), where his triangle consists of signifier, signified and referent. In this, it is clearly invoking Peirce rather than Saussure with its inclusion of the referent or object of the sign. In fact, it can be seen as a simplified version of Peirce’s scheme (Nellhaus, 1998): the signified or interpretant belongs in the transitive dimension while the referent or object is part of the intransitive dimension. In fact, Nellhaus argues that Bhaskar’s ontological domain of the empirical—those events that we actually observe and experience—should be re-conceptualized as the domain of semiosis. Important for our theoretical approach is CR’s insistence that semiosis cannot be reduced either to the play of signifiers, as with Saussure or Derrida, or to a purely hermeneutic sphere (Fairclough et  al., 2004). Semiosis must always have external referents and extra-semiotic conditions and consequences: “semiosis presupposes embodied, intentional, practically-skilled social actors, social relations, material objects, and spatio-­ temporality” (Fairclough et al., 2004, p. 28), not to mention the technology that both enables and conditions communication. We could call this “material semiotics” to use a term from Haraway (1988) and Law (1995).

 eveloping the Framework (3): Information D and Meaning From the perspective of sociomateriality, we need to consider how signs and symbols get translated into action (embodied cognition) and how actions and information get transmitted (technology). As a first step we will consider the relation between meaning and information. Are they in

182 

J. Mingers and L. Willcocks

fact the same, so that the meaning of a message is the same as the information it conveys? Or are they distinct, in which case how do they relate to each other? A semiotic theory of information can be developed combining ideas from Dretske (1981), Habermas (1984), and Maturana and Varela (1980). Following Bateson (1973), the foundation of information, data and more generally signs, must be differences in the physical world, for without difference there is only uniformity. More particularly, differences that “make a difference,” that is generate an event or a sign. Events carry information because the occurrence of an event reduces the possibilities of what might happen to what actually does happen, as Shannon and Weaver (1949) argued. In particular, an event (which includes a sign or message) carries the information about what caused it or led to it. That is, what must be the case in the world for the event to have occurred? Such information exists independently of any observer; indeed, it might never actually be observed.6 Nevertheless, it carries with it the information concerning its own genesis. Information can also be transmitted provided that there are causal links between the sender and the receiver (not necessarily people). This occurs to the extent to which states of the sender are correlated or connected to states of the receiver. Independent events transmit no information; completely linked events transmit all information. Most situations are between the two extremes—the receiver can be affected by things other than the sender (noise), and not all the information from the source will affect the receiver (equivocation). We note also that, following Bhaskar (1993), absences can be causes and therefore can generate information. So, the gas bill that is not paid by the due date generates information to that effect for the company, which then triggers a reminder letter. Information is, then, clearly defined—semantic information7 is the propositional content of a sign, that is what is implied about states of affairs in the world given that the sign exists. This definition has several consequences:  In Bhaskar’s terms, differences and information exist in the domain of the actual, but if they are observed they becomes empirical. 7  In terms of Stamper’s typology, we are interested in the semantic, pragmatic and social levels. 6

6  Interrogating Sociomateriality: An Integrative Semiotics… 

183

• Information is an objective commodity—it is carried by events and signs whether or not it is observed or extracted, and information can be stored and transmitted by the environment, artefacts and people. • Information is distinct from its embodiment in a sign or message since the information itself can have causal events—a knock on the door leads us to open it not because of the physical knock but because it carries the information that someone is there. • Information must be true. We may misunderstand or misinterpret a sign, but the sign itself only carries true information. This approach allows us to define clearly concepts such as data, information and meaning: • Data is a collection of signs, usually brought together for some purpose, to store or transmit information. They are usually numeric, pictorial or linguistic. • Semantic information is the propositional content of data, typically in the form of a message but also in the form of a naturally occurring sign. • Meaning has two different usages. First, there is the system of meanings that are publically available within a sign system such as language. These can be drawn on by competent language users in their communications (Habermas, 1979a). It is that which allows an utterance to carry information, but it is not identical to that information. This is termed “connotation” above. The second usage is the “meaning” that the recipient gains from an utterance (“import”) and/or that which the sender intends (“intent”). Again, these are all different from the information itself (“signification”). This draws a clear distinction between information and meaning. Information is objective, in the sense of being independent of the sender or receiver, and must be true to be information. Meaning is intersubjective, in being at least partly dependent on human interpreters, and is generated from information. Thus, information systems, which store, process and transmit information, are only a part of wider systems of meaningful human communication.

184 

J. Mingers and L. Willcocks

 eveloping the Framework (4): Meaning D and Embodiment We now move on to the process by which messages come to be interpreted, reacted to and acted on to generate technologically based communicative interaction. In contrast to the traditional cognitivist, representationalist paradigm, we shall adopt the perspective of cognition as an active, embodied phenomenon. This draws on the phenomenology of Heidegger (1962) and Merleau-Ponty (1962, 1963), autopoiesis (Varela et al., 1991), and work within ICT such as Winograd and Flores (1987), Dourish (2001), O’Neill (2008), Schultze (2010), and Schultze and Orlikowski (2010). The essence of this position is to deny the Cartesian split between mind and body so fundamental in disciplines such as artificial intelligence, computing, information and cognitivist psychology, in favour of one that recognizes the essentially embodied nature of human cognition whether at the level of perception, thought, behaviour or language. This is also the position underscored by the work of Johnson (1987) and Lakoff and Johnson (1980), including their emphasis on reason shaped by the body, a cognitive unconscious to which we have no direct access, and metaphorical thought of which we are largely unaware. As autopoietic living systems, we have a nervous system that is organizationally closed and self-referring, but which is interactively open to the environment. The type and limitations of these interactions are shaped primarily by our own nervous system rather than by the environment (structure-determined). External events, for example messages with information, trigger responses, but the nature of the response is determined by the readiness of the nervous system at the time—indeed, the system determines what can be triggers for it. The transformation of information into meaning (digitalizing the analogue) is carried out largely unconsciously by the body presenting our conscious mind with pre-­ structured meanings.8 This is the process of embodied cognition.  Much of this was recognized by Peirce who saw that semiosis worked on the basis of “habits,” both mental and physical, which enables us to process and act on signs. 8

6  Interrogating Sociomateriality: An Integrative Semiotics… 

185

There is not thought and language … Expressive operations take place between thinking language and speaking thought; … It is not because they are parallel that we speak; it is because we speak that they are parallel … I do not speak of my thoughts; I speak them and what is between them. (Merleau-Ponty, 1964, p. 18, orig. emphasis)

This is not to say that the meaning triggered by signs and symbols is completely arbitrary or subject-dependent. The very fact that they can trigger anything in the nervous system reflects the way in which we are socialized to the wider social system within which connotative systems exist. We, as human beings, are “structurally coupled” with our immediate environment of people, signification systems and materials. We can say that signs act as affordances and constraints—they tend or afford to lead to particular interpretations and constrain against others—but this is always relative to the knowledge and intentions of the receiver. On one view, the interpretation of signs can be seen in terms of three stages, illustrated with an example in Fig. 6.3.

“What sort of day is it?” Immediate object: wanting to know how the weather is

Meaning 1: Understanding, Immediate interpretant

Dynamical object: the addressee has looked out of the window

Addresser

Meaning 2: Connotation, Dynamical interpretant

Addressee

Meaning 3: Intention, Final interpretant Meaning 2: Generation, Dynamical object

“It is rather stormy”

Meaning 1: Action, Immediate object Embodied cognition

Fig. 6.3  Stages in the interpretation of and response to signs

186 

J. Mingers and L. Willcocks

Suppose the addresser sees the addressee look out of the window at the weather and asks what it is like. In Peirce’s terms, the immediate object is wanting to know the current weather, the dynamical object is the fact that the addressee has looked at it. The first stage of interpretation (understanding or immediate interpretant) is simply appreciating the general meaning of the message, what any competent speaker of the language would be able to do. This will be done sub-consciously by the body and nervous system (what Heidegger would term “ready-at-hand”) unless there is some degree of ambiguity in which case the message may become more an object of conscious reflection (“present-at-hand”). The second stage (connotation or dynamical interpretant) brings in the individual knowledge and motivations of the addressee. It is the effect of the message or sign on that person; it is the process by which the semantic information carried by the message is transformed into meaning. This stage is not purely individual but socially structured in terms of the addressee’s forms of life (Wittgenstein, 1958). Finally, the third stage (intention or final interpretant) leads to some form of action or result which could be a reply, or an activity, or just a decision not to respond. In any event, the addressee’s state of readiness will be changed in some small way. Where some physical action such as a reply is involved, there will be a similar set of stages involved in the production of a response. The whole of this process is one of embodied cognition in that much of it happens beneath the level of consciousness, carried out through the structure of the body and nervous system. It is at this point that we can bring in technology, or perhaps in this context it is better to call it media—“the material of the world that affords the mediation of some form of content” (O’Neill, 2008, p. 138). Or, as Dourish (2001), who is concerned with tangible and social computing, has said: Tangible and social computing both capitalize upon our familiarity with the everyday world, a world of social and physical interactions. As physical beings, we are unavoidably enmeshed in a world of physical facts. … So, the social and the physical are inescapable aspects of our everyday experiences. (p. 100).

6  Interrogating Sociomateriality: An Integrative Semiotics… 

187

 Summary Integrative Framework: Semiosis A and Three Worlds We have now reached the point at which we can introduce the integrative semiotic framework (see Fig. 6.4) that draws on all the material developed above. In Fig.  6.4 semiotics is shown at the centre of a triangle formed by Habermas’s three worlds: the personal, the social and the physical worlds. But this framework goes beyond Habermas to consider both the relationships between the three worlds and semiotics’ mediating role within these relationships. Let us consider first the interior of the triangle. Semiosis relates to the personal world, the world of the subject, in the ways outlined above through the generation and interpretation of signs and messages. As human beings we exist in a world of meaning and communication and that is always a process of semiosis. Events and symbols have meaning for us (import) because of what they represent or stand for and our communicative intentions (conscious and unconscious) can only become operative when represented in some form of sign system. We would want to stress again that meaning is to be distinguished from information. Information is that content of a meaningful message that is, in fact, true whether or not it corresponds with the meaning that the recipient actually derives from the message.

Personal Intent, Import

Sociation

Embodiment

Semiosis

Social

Connotation, reproduction

Sociomateriality

Representation, transmission

Material

Fig. 6.4  The relations between semiosis and the three worlds

188 

J. Mingers and L. Willcocks

Semiosis also relates to the material world in that all signs have to have some form of physical embodiment in order to be signs, and they must also be transmitted through some form of physical media. By media we do not mean necessarily technology, although that is increasing our human mode of communication, but sight, sound, touch, etc., are all media for semiotic communication. Finally, semiotics relates to the social world in that the conative aspects of sign systems are social rather than individual—they exist before and beyond the individual’s use of signs. Some semiotic interactions may be non-social, for instance, taking hoof marks as a sign that a horse has passed by (an indexical sign), but the vast majority are symbolic and so rely on pre-existing agreements about the meanings of particular signs. Looking next at the outside, each side represents an ongoing form of relationship between the corners. Between the personal and social worlds, there is a relationship of what we might call sociation. This is essentially the relationship between structure and action that has been so heavily discussed and debated (reference withheld). From a critical realist position, the work by Bhaskar (1979), Archer (1988, 1995) and Mutch (2010) on morphogenesis addresses this area. The social and the individual are conceptualized as two real, independent but mutually interlocking systems. Society is an “ensemble” of structures, practices and conventions realized in the form of “position-practices”—role positions and social practices. It thus pre-exists the individuals who occupy these positions and conditions that activities they undertake. But, at the same time, society is reproduced or transformed by that individual activity. Whilst emphasizing the ontological reality of social structures, Bhaskar accepts that they only exist through the activities they govern. An alternative conceptualization would be through Giddens’s (1984) structuration theory which is already well known within IS (Jones & Karsten, 2008). Between the personal and the material worlds we have a relation of embodiment. This occurs in two ways—the first is embodied cognition which is to do with the physical human body and the manner in which this inextricably links thought and action, as has been outlined above. The second concerns technology, taken very broadly, and the ways in which it both enables and constrains human action (Dourish, 2001; O’Neill, 2008).

6  Interrogating Sociomateriality: An Integrative Semiotics… 

189

Finally, between the social and the physical we have a relationship of sociomateriality as envisaged in the introduction of this chapter (Leonardi & Barley, 2008; Orlikowski, 2000, 2007, 2010). As we recognized there, there are several different ways of conceptualizing this relationship: where one side or the other dominates; where there is a mutual interaction between different systems or where the two systems are inseparable. We shall discuss these three in more detail in the next section. In fact, what we can see from this diagram is that the outside relationships—sociation, embodiment and sociomateriality—are in fact all mediated through the process of semiosis. Since, phenomenologically, humans always already exist within a space constituted through meaning, and semiosis is the process of production and interpretation of meaning, it is not possible to conceptualize these forms of interaction without involving semiotics.

Interrogating Sociomateriality in Information Systems In this section we examine the implications of our framework for how sociomateriality has been conceptualized and used in the information systems (IS) field, focusing on leading work and critiques on the subject. Sociomateriality continues to be utilized, developed, used as a springboard and critiqued into the 2020s (see below). Within IS, sociomateriality is primarily associated with Orlikowski (2007, 2010) who in turn has drawn on writers such as Barad (2003), Suchman (2007) and Latour (2005). Orlikowski herself has changed and developed her position—the later papers adopt a significantly more “entangled” view than Orlikowski (2000). If we primarily address Orlikowski’s work, it is because she has been leading quite the most interesting and prolific work in the field to date. Two existing critiques of sociomateriality as a concept in IS are particularly pertinent to the present chapter and deserve comment. Firstly, Jones (2014) uses the eight criteria of Gerring (2001) to assess the “conceptual goodness” of sociomateriality. He finds the major criterion of coherence as

190 

J. Mingers and L. Willcocks

an area of particular weakness with sociomateriality, due to a number of terminological, definitional and theoretical inconsistencies. The case made for sociomateriality also exhibits some looseness of argumentation that makes it difficult to assess the scope of the concept, and hence the coherence of the ideas it involves.

Flowing from these weaknesses, Jones sees further difficulties in operationalization and validity of the concept.9 We concur and seek to locate and define the concept more precisely in relation to other concepts with which it regularly interacts but also to revise its realist assumptions and make transparent conceptual issues on meaning, information and semiosis that are only latent or unclear within the sociomateriality research literature. Secondly, Faulkner and Runde (2010) address how a focus on sociomateriality leads to the neglect of the non-material nature of many of the technological objects that populate the contemporary world (see also Chap. 4). Kallinikos (2011a) is correct in according modern information and communications technology and technological design (form and function) “a growing emancipation from the materials with which they are entangled.” Hyle (matter) increasingly gives way to eidos (function/ form), and software, for example, is to some degree technology without matter, reflecting a move towards a culture of the virtual or non-material (Kallinikos, 2011b). These developments underline the seriousness of the concern of Faulkner and Runde (2010) when they state with reference to Orlikowski’s work that “sociomateriality, at least as it has been developed so far, does not appear to have an explicit place for non-material technological objects.” Part of this, of course, has been because Orlikowski has been primarily addressing Law and Urry’s (2004) point that contemporary social science is ill-equipped to address emerging issues of ephemerality, multiplicity, dispersion and mobility, and that “some of these shortcomings arise from our conceptual difficulties in grappling with the intextricably material nature of sociality” (Orlikowski, 2007). But this has led to a radical under-conceptualization in the IS sociomateriality literature, not just of  Jones (2014) is also questioning of its field utility, and more positive on its resonance, contextual range and parsimony, while seeing sociomateriality as a concept having strong potential analytic/ empirical utility. 9

6  Interrogating Sociomateriality: An Integrative Semiotics… 

191

the non-material but of a range of issues and relationships that have always been imbedded in socio-technical interplay and that contemporary technological advances have pushed even further to prominence. In this, then, we concur with Faulkner and Runde (2010) but extend the analysis addressed but underplayed in their paper by suggesting that the integrative summary framework in Fig. 6.4 provides a more coherent and explicit description of how non-material technological objects vitally depend on semiosis for their ontological status. Indeed, we would suggest that one of the limitations of Faulkner and Runde’s otherwise excellent ground-clearing paper is that the relevance of semiotics to their arguments and position are heavily implicated but nowhere explicitly applied in a coherent manner. However, their work is more explicitly consistent with the Fig. 7.4 framework in their realist position, their characterization of non-material technological objects as “deeply social things” (p. 17), and in their qualified rejection (see page 28, note 15) of what they call the interpenetration thesis put forward by Orlikowski (2007). In terms of our summary integrative framework, we would first critique Orlikowski’s later work, where the relationship between the social and the material (primarily technology) is quite radical. Orlikowski and Scott (2008) neither have one dominating the other nor two independent but interacting domains, but rather the two are so inextricably interrelated that they cannot be separated, and their properties are defined only in relation to each other: In other words, entities (whether humans or technologies) have no inherent properties, but acquire form, attributes, and capabilities through their interpenetration. This is a relational ontology that presumes the social and the material are inherently inseparable. (Orlikowski & Scott, 2008, p. 455).

Interestingly, this quote would actually seem to be inherently self-­ contradictory since if we take it to be true then it would not be possible to parenthesize “humans or technologies” since the two would be indistinguishable. In terms of the framework articulated in this chapter, we would argue that sociomateriality as described in Orlikowski and Scott (2008) actually involves three reductions. Let us look at these in more detail.

192 

J. Mingers and L. Willcocks

Firstly, it reduces what should properly be two distinct but interacting structures to a duality that loses sight of both of its components. One reason for this reduction is the lack of what Elder-Vass (2008) calls a depth ontology, and this would seem to be an inherited influence from Giddens’s structuration theory10 and also from actor-network theory(ANT) (Elder-Vass, 2008). The distinctions between the empirical, the actual and the real provide a dimension of depth to critical realism’s ontology. Bhaskar’s ontology, as we use it, has a second dimension of depth in its recognition that reality is stratified into an ontological hierarchy of entities, in which higher level entities have emergent properties—properties not possessed by the lower-level entities that are their parts. These mechanisms in the domain of the real are responsible for emergent properties, synonymous with causal powers, that interact to produce actual events. For critical realists both social structures and human individuals are entities with emergent properties that arise from their ontological structure. Thus, a critical realist account of the social, material and personal can recognize that human individuals, social structures and indeed entities of other kinds have causal powers that are distinct from each other, and that both (or all) interact to produce events—for example social events—even though human individuals are parts of the social structure concerned. By contrast, Orlikowski and Scott (2008) subscribe to a relational ontology that presumes the social and the material are inherently inseparable, an ontological fusion signalled by the lack of a hyphen in “sociomaterial.” They also see ANT, sympathetically, as the most prominent part of the sociomateriality literature. But as Elder-Vass (2008) points out, ANT’s ontology has a multi-dimensional absence of depth ontology and tends to be limited to the empirical. Its ontological flatness and its assumption of symmetry between human and non-human actors mean that it cannot subscribe to critical realism’s assumption that particular causal powers (and hence the particular terminology appropriate to their description) vary depending on the underlying structure and mechanisms of each type of actor, be it material, social or personal.  Writing from a critical realist perspective on integrating institutional, relational, embodied and emergent attributes of structure, Elder-Vass (2008) rejects structuration theory’s conflationist ontology while seeing some value in Giddens’s theory of the embodied facet of social structure. 10

6  Interrogating Sociomateriality: An Integrative Semiotics… 

193

Secondly, sociomateriality reduces the role of active subjects without whom neither society nor technology would actually exist or be reproduced; semiosis itself only operates through individual subjects who must always be the ultimate efficient cause of any interaction. As critical realism argues, social structures (and technological structures to the extent that they are part of a social system as characterized in socio-technical systems and actor-network theories) do not exist independently of the activities and practices that they regulate, and only exist in their effects or occurrences. Moreover, social structures do not exist independently of peoples’ understanding of what they are doing. Thus, the social and the material cannot have a direct interaction, or indeed entanglement, without the mediation of people and semiosis. In the way sociomateriality is portrayed and operationalized in, for example Orlikowski, (2007) and Orlikowski and Scott (2008) people and semiosis are operational but, we would argue, insufficiently theorized. Thirdly, sociomateriality reduces the role of semiosis as the process and mechanism through which meaningful human activity occurs. Social relationships and structures are all embedded and represented in a semiotic fashion and physical media both represent and transmit signs and symbols but also only become objects of representation and interaction to the extent that they are or can be represented symbolically. In other words, technology/media are both a medium of semiosis, but also both a condition for and result of semiosis. Interestingly, the position represented in Fig. 6.4 is in fact quite compatible with Orlikowski (2000), which puts forward the idea of a technology-in-practice view, though we have of course provided the addition of semiotics. We also preserve the idea inherent in a critical realist position but not in that of Orlikowski (2000) that, notwithstanding the second point above, social structure is ontologically separate from the activities of individual people. It both pre-dates any particular individual and can in time be reproduced or transformed by their activities. Thereafter, our major differences would seem to be with the more recent papers on sociomateriality, but especially with Orlikowski (2007, 2010), Orlikowski and Scott (2008). The debate is ongoing, inevitably. Notably, Mutch (2013) provided his own critique to which Orlikowski & Scott

194 

J. Mingers and L. Willcocks

(2013) provided a robust response. More recent examples include (Orlikowski & Scott, 2021); (Scott & Orlikowski, 2022); (Pelizza, 2021); (Holz, 2021); (Fonseca et al., 2022) and (Hanseth et al., 2021).

Applying the Integrative Semiotic Framework The weight of our argument so far is that sociomateriality needs to be more precisely located within a broader integrative framework that makes more explicit a family of concepts and their relationships needed to study advanced information and communications technologies in contemporary organizations. We arrived at such a framework in Fig. 6.4, but how can this be applied in practice? We use several studies of diverse ICT contexts to address this question. See also Chap. 4 for relevant examples. Example 6.1 HCI Studies As established earlier, semiotics has been widely applied already in the study of ICTs and business though this has not always been through utilizing Peircean semiotics. Particularly interesting, in the light of our framework, is the work of Dourish (2001) and O’Neill (2008) who look at human-computer interaction by drawing on phenomenology, Heidegger, Merleau-Ponty’s work on embodiment, and semiotics, to develop the notion of embodied interaction. They are particularly interested in how interactive media can be studied and designed, taking into account the physical and social worlds in which they operate, and how media and technologies relate to the human beings interacting with them. Our framework plays directly into such work. As one example, O’Neill (2008), who at one point talks of the “semiotic screen,” refers to the Brazil-based SERG group that draws heavily on Peirce’s conception of a sign to develop their understanding of signification with interactive media in screen-based interfaces. Here they find particularly useful Peirce’s concept of how signification takes place through thirdness, where a representation is related to its object via an interpretant (Prates et al., 2000). If this example seems to focus on the cognitive semiotic dimension of the Fig. 6.4 framework, then these HCI researchers are also focusing on the embodied interaction of the personal and material, in a physical, human-designed space. Dourish

6  Interrogating Sociomateriality: An Integrative Semiotics… 

195

(2001) in particular also stresses the role of the social in HCI assessment and design processes, giving as an example the study of an air traffic control centre. Example 6.2 MPK20 and the Project Wonderland Rooms However, the relevance of the framework can hardly be restricted to HCI studies. Recall our position that signification processes must be located in relation to the personal world of minds, intentions and knowledge; the social world of power and normative practices; and the material world of space-time, technology and bodies (Habermas, 1984). Orlikowski (2007) provides a powerful example of these issues and the relevance of the Fig.  6.4 framework in her description of MPK20, and the Sun Microsystems Project Wonderland rooms, offices, screens and documents that form part of an online, three-dimensional, immersive environment for workplace collaboration. While she does not develop the analysis of this synthetic world, she asks interesting questions of how they can be researched. She chooses the perspective of entanglement and draws upon actor-network theory and Barad’s (2003) notion of apparatus to focus the research possibilities. However, given the centrality of humans and of meaning and communication in how MPK20 and the Project Wonderland Rooms operate, the sociomateriality perspective as described fundamentally lacks a coherent semiotic dimension in understanding the material, social and personal worlds being described and analysed. How does meaning arise? What sign systems are operating and how are they employed? What are the power dimensions of the control or lack of influence over meaning, communication and information? How do social structures and processes relate to personal understandings and influence action? What is the role of non-material technological objects in the generation of meaning, behaviour and performance? These are only illustrative research questions that follow from using our integrative summary framework but do not flow easily from applying a more limited “sociomateriality” perspective. Example 6.3 Virtual Worlds We can pursue this issue in the work of Schultze (2010) on embodiment and presence in virtual worlds. These are multi-modal platforms

196 

J. Mingers and L. Willcocks

featuring rich graphics, 3D rendering, high-fidelity audio and video, motion and interactivity. Examples include Second Life, Teleplace, EverQuest, consisting of technology designed to create and experience virtual spaces, objects and people with which a user can interact. We would suggest that a semiotic approach is much more attuned to studying such phenomena than the sociomateriality perspectives we are interrogating. For example, Schultze reveals a complex world where avatars re-embody the communicator who has been disembodied through computer mediation. Embodiment here means that the communicator can engage in practices of the body (e.g. smile, sit, move) and have a sense of presence whether in an actual or virtual environment. Semiosis pervades both the design process and features of the virtual world, but also how it is operationalized through social, personal and material interactions. As Schultze describes it, how the users construct an avatar with regard to personality, appearance and behaviour is imbedded in a system of meaning informed by the social norms and conventions shaped by both the actual and the virtual world. Indeed, Nowak and Rauh highlight some of the social norms of avatar appearance, with a humanoid-gendered avatar being the first choice of self-representation, having the more likeable and persuasive qualities necessary to be effective in social settings. Here we can see the personal and social worlds interacting through sociation and semiosis, mediated by technology (see Fig. 6.4). One of the avatar’s key affordances is embodiment, in the sense of giving participants a virtual body that enables them to engage in practices of the body and recapture the body’s non-discursive, semiotic capabilities. Here we see the personal and the material interacting through embodiment and semiosis. This interaction is also performed to give experiences of presence. Schultze (2010) helpfully lists six kinds of illusory presence—telepresence, social presence, co-presence, self-presence, hyper presence and eternal presence—made possible through personal-material interactions. And, as we saw with Project Wonderland, virtual worlds can also be social worlds where sociomaterial and semiotic interactions link the social, the material, and emergent meanings and performance.

6  Interrogating Sociomateriality: An Integrative Semiotics… 

197

Example 6.4 TripAdvisor Another example we will look at is the Orlikowski and Scott’s (2008) analysis of the social media TripAdvisor, which, they argue, as enacted social media, is a sociomaterial phenomenon. TripAdvisor, they record, is one of the largest online travel communities with over 20 million reviews and opinions stored on over one million hotels, restaurants and venues, contributed by some 30 million visitors per month. Orlikowski and Scott (2008) plot the transformation of how assessment knowledge of the industry is produced by studying two hotels. They find that TripAdvisor ranks them using the same algorithm, thus configuring them as rivals, though the hotels have different attributes, characteristics and, indeed, markets. There are also noticeable differences between the profiles of travellers posting their reviews of these two establishments. Postings can also be multi-media including images, for example photographs of rooms, level of cleanliness, window views. TripAdvisor’s ratings reflect individual users’ personalized and situated experiences in a hotel, and their decision to provide a review. Unlike the more traditional Automobile Association (AA) ratings that focus on operational issues and standardized assessments of facilities, TripAdvisor’s ratings are “temporally sensitive continually reconfigured, personal, and based upon relatively unregulated content” (Orlikowski and Scott, 2008). Not surprisingly hoteliers found the views and ratings often variable and subjective but also felt unable to do much to change the type of knowledge being produced through this form of social media. Thus, TripAdvisor, as a social media technology, gives the subjective reviews and ratings a determinacy and reach not otherwise achievable while also challenging the hoteliers’ and institutionalized hotel recognition schemes like AA’s primacy and control. This exploratory study brings out several points that lie within the sociomaterial dimension of the framework. But does it not thereby exclude many interesting questions from the other two dimensions—the personal/technological and the personal/social? For example, how do individuals themselves relate to the technology—in what circumstances do they use it? Why do they use it? Is it only very good or bad experiences that get recorded? What is their level of belief on the accuracy and reliability of the data provided? Is it the ease of access that determines use, that is convenience, rather than confidence in the information? Is the

198 

J. Mingers and L. Willcocks

description of “actual” experiences more convincing than statistical data? What is the role of semiosis in seeking answers to these questions? Then there is the personal/social dimension, that is sociation and semiosis— and how the personal influences the social “facts” being created, and how these are then influencing personal decisions. Semiotically, how does social media like TripAdvisor change meaning, the flow of information, and create ratings and assessment as social “facts”? These sorts of questions raise issues about the creation, interpretation and response to signs, and the role of signs in creating a continually remade and contested social and personal reality. Example 6.5 A Dairy Production Plant The study of a fully computerized dairy production plant by Kallinikos (2011b) is particularly alive to the role of signification in seeking to understand how ICTs interact and help change the material and cognitive foundations of work. He points out that as ICTs proliferate, they bring with them symbol schemes and codes that do not rely on the signifying conventions of similarity and proximity (Peirce’s indexical signification). This can create real difficulties in sense-making and explains to a degree the attempts in our previous examples to technologically create similarity and proximity for social and personal use, for example, in the Project Wonderland and avatar examples. Kallinikos points to computer technology bringing to organizations on a massivescale comprehensive systems of information tokens and codes that sustain software packages and also generating an immense output of data and information tokens. He follows Zuboff’s (1988) claim that ICTs alter the tangible, social and personal nature of work and transform it, literally to reading, that is “an encounter with symbol schemes and data items that are supposed to represent surrogate versions of physical and social items and relationships”. His study focuses on the milk refinement process at the heart of dairy production. Seven treatment lines produce over 50 products. Production is steered from a separate room that forms the production control centre of the plant. Planning, controlling and monitoring are highly computerized, with process operators following the status and progression of

6  Interrogating Sociomateriality: An Integrative Semiotics… 

199

production on panels of bulbs and monitors in the control centre. Several printouts also report incidents in the production process. The software package provides the means for controlling the quantity of received milk, channelling it into milk silos where it is preserved at the appropriate temperature before being refined. Quantity is reported both in numerical and analogue form. One form shows graphically the level of the milk in the silo tanks. The complex process of production is steered with the aid of three computers with human intervention at points to make commands and check progress. The mechanics of the various processes is steered through electronic sensors and more than 1000 electronic valves, which connect the seven treatment lines with the computers in the control rooms. Compared to other semi-automated production plants, high computerization changed the nature of work and personal-social-technological interactions. For example, in order to grasp how the software package worked and what it signified, it was necessary to reconstruct mentally the physical processes and flows regulated by the package. The cognitive burden and reliance on not easy-to-understand symbols and text were greatly increased through the use of the software, computers and related displays, while physical presence in the production process and information there from were greatly decreased, not least due to the operators’ location in the remote-control room. We see here fundamental changes in the way the personal, social and material interact through sociation, embodiment, sociomateriality and semiosis. Signification becomes increasingly abstract and representational. Thus, 16 panels with 700 colour-coded bulbs report the progression and status of the process. A process printout records failures and their location though in a highly coded form. A highly coded system printout, often requiring specialized staff for its interpretation, records disturbances to the adequacy of the conceptual and organizational logic of the software package itself and brings another series of codes, categories and definitions. In fact, the documentation of the software package and the installations to which it relates involve 50 manuals of symbols definitions, codes and relationships and functions. Not surprisingly, operators with experience in semi-automated plants decried the increased cognitive complexity and the constricted embodied and sociated interactions. Lack of physical experience with the installations and

200 

J. Mingers and L. Willcocks

lack of knowledge of the tangible reality of the plant were seen as severe limitations on operator capability, while at the same time the increased reliance on abstract coding systems increased their stress and anxiety. Kallinikos records that some moves towards improving referential attribution were made in the sign systems utilized. For example, the bulb system was geometrically organized to recapture the totality of the production process, through the decomposition and elaborate segmentation of its various steps. A sort of structural resemblance was thus established between the bulbs (symbol tokens) and the absent reality of the refinement process (reference). The geometric arrangement of the signalling system also contained an indication of temporal patterns. In a relatively simple and precise way, the process printouts came to complement the bulbs system by indicating through numerical description the installation item concerned and through binary coding (right-wrong, stop-go) its current state. However, operators seemed to suggest that the vicarious representations of the software “failed to restore the confidence that referential reality is capable of providing to people accustomed to context-­ embedded work based not just on the reasoning and distancing capacity of the eye, but on sensory-motor manipulation of tangible things”. His study is particularly rich in showing the central role semiosis needs to play in the study of ICTs in work organizations. ICTs interacting with people and organization in this case saw process operators seemingly needing to turn their back on the physical production process and devote themselves instead to the task of examining the very structure of signs, codes and symbol schemes, whereby physical relationships were mediated and regulated. The codifications of the software package did not represent a mirror image of the material and technological constitution of the work processes but produced a multi-layered fragmented systems of signs and codes that saw little relationship between token and referent, but influenced and was influenced by interaction through sociation, sociomateriality and embodiment as represented in our framework.

6  Interrogating Sociomateriality: An Integrative Semiotics… 

201

Summary In this section we have used case research in HCI, virtual worlds, social media and computerized production plants to illustrate the applicability of the integrative semiotics framework arrived at in this chapter. This, of course, does not exhaust the possibilities for the framework, especially in an IS field replete with technologies of information and communication highly dependent on codes and signification processes. Our own view of the three worlds framework in Fig. 6.4 is that there are different degrees of relationship between them in different contexts. Thus, social networking/communication would seem to be heavily technologically dependent, with each new generation bringing forth new possibilities, while religion, for example, is heavily based on social practices and symbolism, and little on technology. The three systems interact generating emergent and enactive phenomena within any particular context, and following our critical realist underpinning, these emergent phenomena can themselves affect the underlying systems in a process of downward causation.

Conclusion This chapter identifies and addresses limitations in the theoretical and operational coherence of the concept of sociomateriality as applied in many IS studies to date. By establishing philosophical foundations in a Peircean semiotics rendered consistent with critical realism, we were able to build in theories of information, meaning and embodiment to construct a consistent integrative framework of operationalizable concepts for studying information systems and the personal, social and material worlds they inhabit. In fact, we welcome the development of the discourse on sociomateriality in IS as a refreshing invitation to think through how we research contemporary advances in ICTs, particularly in terms of its aim to re-establish the importance of technology and to renew socio-­ technical studies, especially in the light of advances in ICT and their effects, such as social networking.

202 

J. Mingers and L. Willcocks

This chapter has responded by locating more precisely where sociomateriality, as a usable concept, fits with a family of concepts and with the relationships between them. The aim has been to provide a more philosophically grounded, more comprehensive theoretical framework for use in carrying out research. The framework is consistent with several aspects of previous conceptualizations of sociomateriality, not least the notion that sociomateriality is always being enacted, performed and in the making. But on our revised view, two other relationships—of sociation and embodiment—also need to be addressed on a more precise basis, and semiosis needs to play a central, explicit rather than implied part in the study of contemporary ICTs. We identified and attempted to address a number of limitations observed in the sociomateriality studies under review. These included a flat or contradictory ontology being employed; insufficient philosophical underpinning, with little philosophical or practical account of information and meaning; semiosis implied rather than raised to the level of a descriptive or explanatory framework; and insufficient attention, as a result, able to be given to non-material technological objects such as discussed by Faulkner and Runde (2010). In brief, we found that sociomateriality was not sufficiently located in the broader context and network of concepts and relationships needed to study contemporary ICTs. Sociomateriality as a concept has been asked to do too much work. Our integrative framework has sought to delimit its role and so make the concept more useful and valuable as a complement to other dimensions of human social activity.

References Almeder, R. (1980). The philosophy of Charles S. Peirce. Blackwell. Andersen, P. (1990). The theory of computer semiotics. Cambridge University Press. Archer, M. (1988). Culture and agency. Cambridge University Press. Archer, M. (1995). Realist social theory: The morphogenetic approach. Cambridge U. P. Archer, M., Bhaskar, R., Collier, A., Lawson, T., & Norrie, A. (Eds.). (1998). Critical realism: Essential readings. Routledge.

6  Interrogating Sociomateriality: An Integrative Semiotics… 

203

Arnold, S. J., Kuzinets, R. V., & Handelman, J. M. (2001). Hometown ideology and retailer legitimation: The institutional semiotics of Wal-Mart flyers. Journal of Retailing, 77(2), 243. Atkin, A. (2006). Peirce’s theory of signs. In Stanford encyclopedia of philosophy. Stanford University. Austin, J. (1962). How to do things with words. Oxford University Press. Barad, K. (2003). Posthumanist performativity: Towards an understanding of how matter comes to matter. Signs: Journal of Women in Culture and Society, 28(3), 801–831. Barley, S. R. (1983). Semiotics and the study of occupational and organizational cultures. Administrative Science Quarterly, 28(3), 393–413. Bateson, G. (1973). Form, substance and difference. In G. Bateson (Ed.), Steps to an Ecology of Mind (pp. 423–440). London: Paladin. Bateson (1973). Steps To An Ecology of Mind, Palidin, St.Albans. Beynon-Davies, P. (2009). Neolithic information: The nature of information. International Journal of Information Management, 29, 3–14. Bhaskar, R. (1978). A realist theory of science. Harvester. Bhaskar, R. (1979). The possibility of naturalism. Harvester. Bhaskar, R. (1993). Dialectic: The pulse of freedom. Verso. Brannen, M. (2004). When Mickey loses face: Recontextualization, semantic fit, and the semiotics of foreignness. Academy of Management Review, 29(4), 593–616. Brier, S. (2001). Cybersemiotics: A reconceptualization of the foundation for information science. Systems Research and Behavioral Science, 18(5), 421–427. Buchler, J. (Ed.). (1940). The philosophy of Peirce: Selected writings. Routledge and Kegan Paul. Bühler, K. (1982) (orig. 1934). Karl Bühler: Semiotic Foundations of Language Theory, New York: Plenum Press. Bygstad, B. (2010). Generative mechanisms for innovation in information infrastructures. Information and Organization, 20(3–4), 156–168. Callon, M. (1991). Techno-economic networks and irreversibility. In J.  Law (Ed.), A sociology of monsters: Essays on power, technology and domination. Routledge. Chandler, D. (2002). Semiotics: The basics. Routledge. Clarke, R. (2002). Intertextuality at work: Large scale organisation of Workpractices. In K.  Liu, R.  Clarke, P.  Andersen, & R.  Stamper (Eds.), Coordination and communication using signs: Studies in organizational semiotics 2 (pp. 59–83). Kluwer.

204 

J. Mingers and L. Willcocks

Cooper, S., Crowther, D., & Carter, C. (2001). Towards a semiology of the periodic review of UK regulated utilities. Financial Accountability & Management, 17(3), 291. de Saussure, F. (1960). Course in general linguistics. Peter Owen. de Souza, C. (2005). The semiotic engineering of human-computer interaction. MIT Press. Dobson, P. (2001). The philosophy of critical realism - an opportunity for information systems research. Information Systems Frontiers, 3(2), 199–210. Dourish, P. (2001). Where the action is: The foundations of embodied interaction. MIT Press. Dretske, F. (1981). Knowledge and the flow of information. Blackwell. Elder-Vass, D. (2008). Searching for realism, structure and agency in ActorNetwork theory. The British Journal of Sociology, 59(3), 455–473. Fairclough, N., Jessop, B., & Sayer, A. (2004). Critical realism and semiosis. In J.  Joseph & J.  M. Roberts (Eds.), Realism, discourse and deconstruction (pp. 24–42). Routledge. Faulkner, P., & Runde, J. (2010). The Social, the material and the ontology of nonmaterial technological objects. Cambridge University working paper, Cambridge. Fiol, C. M. (1989). A semiotic analysis of corporate language: Organizational boundaries and joint venturing. Administrative Science Quarterly, 34(2), 277–303. Fonseca, L., Silva, M., Silva, S., & Periera, G. (2022). Sociomateriality: An integrative review to understand the current status of scientific production. Research, Society and Development, 11(5), 1–13. Frege, G. (1952). Translations from the philosophical writings of Gottlob Frege. Blackwell. Freud, S. (2004 (orig. 1900)). The interpretation of dreams. Kessinger. Gazendam, H., Jorna, R., & Cijsouw, R. (Eds.). (2003). Dynamics and change in organizations: Studies in organizational semiotics. Kluwer. Gerring, J. (2001). Social science methodology: A criterial framework. Cambridge University Press, Cambridge. Giddens, A. (1984). The constitution of society. Polity Press. Greenlee, D. (1973). Peirce's concept of sign. Mouton. Habermas, J. (1978). Knowledge and human interests (2nd ed.). Heinemann. Habermas, J. (1979a). Communication and the evolution of society. Heinemann. Habermas, J. (1979b). What is universal pragmatics? In J.  Habermas (Ed.), Communication and the evolution of society (pp. 1–68). Heinemann.

6  Interrogating Sociomateriality: An Integrative Semiotics… 

205

Habermas, J. (1984). The theory of communicative action Vol. 1: Reason and the rationalization of society. Heinemann. Habermas, J. (1994). Towards a critique of the theory of meaning. In Postmetaphysical thinking (pp. 57–87). MIT Press. Hanseth, O., Masovic, J., & Mork, B. (2021). The Sociomateriality debate revisited: The contribution of assemblage theory. University of Oslo. Haraway, D. (1988). Situated knowledges: The science question in feminism and the privilege of partial perspective. Feminist Studies, 14(3), 575–599. Harvey, M., & Evans, M. (2001). Decoding competitive propositions: A semiotic alternative to traditional advertising research. International Journal of Market Research, 43(2), 171–187. Heidegger, M. (1962). Being and Time. Blackwell. Holz, E. (2021). Sociomateriality and organizational analysis: From rhetoric to relevance. Organizações & Sociedade, 28(97), 241–264. Howcroft, D., Mitev, N., & Wilson, M. (2004). What we May learn from the social shaping of technology approach. In J. Mingers & L. Willcocks (Eds.), Social theory and philosophical for information systems (pp. 328–371). Wiley. Huang, S. 2006. A semiotic view of information: Semiotics as a Foundation of Lis Research in information behavior. In ASIS&T '06 annual meeting: Information realities: Shaping the digital future for all. Austin, Texas. Husserl, E. (1973 (orig. 1913)). Logical investigations. Routledge. Jakobson, R. (1960). Linguistics and poetics. In D. Lodge & N. Wood (Eds.), Modern criticism and theory: A reader (pp. 32–56). Longman. Jakobson, R., & Halle, M. (1956). Fundamentals of language. Mouton. Johnson, M. (1987). The body in the mind: The bodily basis of imagination, reason and meaning. U. Chicago Press. Jones, M. (2014). A matter of life and death: Exploring conceptualizations of sociomateriality in the context of critical care. MIS Quarterly, 38(3), 895–926. Jones, M., & Karsten, H. (2008). Giddens’s structuration theory and information systems research. MIS Quarterly, 32(1), 127–157. Kallinikos, J. (2011a). Form Function and Matter: On The Matrix of Technological Objects. Paper in draft form from author, June. Kallinikos, J. (2011b). Governing Through Technology: Information Artefacts and Social Practice. Palgrave, London. Knorr-Cetina, K. (1997). Sociality with objects: Social relations in Postsocial knowledge societies. Theory, Culture and Society, 14(4), 1–30. Krampen, M., Oehler, K., Posner, R., Sebeok, T., & von Uexkull, T. (Eds.). (1987). Classics of semiotics. Plenum Press.

206 

J. Mingers and L. Willcocks

Kryssanov, V., Okabe, M., Kakusho, K., & Minoh, M. (2003). A theory of communication for user Interface design. In H.  Gazendam, R.  Jorna, & R. Cijsouw (Eds.), Dynamics and change in organizations: Studies in organizational semiotics. Kluwer. Lakoff, G., & Johnson, M. (1980). Metaphors we live by. University of Chicago Press. Latour, B. (1987). Science in action. OU Press. Latour, B. (2005). Reassembling the social: An introduction to actor network theory. Oxford University Press. Law, J. (1995). Notes on materiality and sociality. The Sociological Review, 43(2), 274–294. Law, J. (2004). After method: Mess in social science research. Routledge. Law, J. and Urry, J. (2004). Enacting The Social, Taylor and francis, London. Lawes, R. (2002). Demystifying semiotics: Some key questions answered. International Journal of Market Research, 44(3), 251–264. Leonardi, P. M., & Barley, S. R. (2008). Materiality and change: Challenges to building better theory about technology and organizing. Information and Organization, 18(3), 159–176. Liu, K., Clarke, R., Andersen, P., & Stamper, R. (Eds.). (2001). Information, organization and technology: Studies in organizational semiotics. Kluwer. Liu, K., Clarke, R., Andersen, P., & Stamper, R. (Eds.). (2002a). Coordination and communication using signs: Studies in organizational semiotics 2. Kluwer. Liu, K., Clarke, R., Andersen, P., & Stamper, R. (Eds.). (2002b). Organizational semiotics: Evolving a science of information systems. Kluwer. Mancini, C., & Buckingham Shum, S. J. (2006). Modelling discourse in contested domains: A semiotic and cognitive framework. International Journal of Human-Computer Studies, 64(11), 1154–1171. Martin, B., & Ringham, F. (2006). Key terms in semiotics. Continuum. Maturana, H. (1978). Biology of language: The epistemology of reality. In G.  Millar & E.  Lenneberg (Eds.), Psychology and biology of language and thought: Essays in honour of Eric Lenneberg (pp. 27–63). Academic Press. Maturana, H., & Varela, F. (1980). Autopoiesis and cognition: The realization of the living. Reidel. May, M., & Andersen, P. (2001). Instrument semiotics. In K. Liu, R. Clarke, P. Andersen, & R. Stamper (Eds.), Information, organization and technology: Studies in Organisational semiotics (pp. 271–298). Kluwer. Mead, G. (1934). Mind, self and society. University of Chicago Press.

6  Interrogating Sociomateriality: An Integrative Semiotics… 

207

Menchik, D., & Tian, X. (2008). Putting social context into text: The semiotics of E-mail interaction. American Journal of Sociology, 114(2), 332–370. Merleau-Ponty, M. (1962). Phenomenology of perception. Routledge. Merleau-Ponty, M. (1963). The structure of behaviour. Beacon Press. Merleau-Ponty, M. (1964). Signs. Northwestern University Press. Mick, D. G. (1986). Consumer research and semiotics: Exploring the morphology of signs, symbols, and significance. Journal of Consumer Research, 13(2), 196–213. Mingers, J. (1995). ‘Information and meaning: Foundations for an intersubjective account,’ Information Systems Journal, 5, 285–306. Mingers, J. (2004). Future directions in management science modeling: Critical realism and multimethodology. In S. Fleetwood & S. Ackroyd (Eds.), Critical realism in action in organizations and management studies. Routledge. Mingers, J., & Willcocks, L. (2014). An integrative semiotic framework for information systems: The social, personal and material worlds. Information and Organization, 24(1), 48–70. Mingers, J., & Willcocks, L. (2017). An integrative semiotic methodology for is research. Information and Organization, 27(1), 17–36. Morris, C. (1938). Foundations of the theory of signs. University of Chicago Press. Mutch, A. (1999). Critical realism, managers and information. British Journal of Management, 10, 323–333. Mutch, A. (2010). Technology, organization and structure - a morphogenetic approach. Organization Science, 21(2), 507–520. Mutch, A. (2013). Sociomateriality - taking the wrong turn? Information and Organization, 23, 28–40. Nellhaus, T. (1998). Signs, social ontology, and critical realism. Journal for the Theory of Social Behaviour, 28(1) pp. 1-+. Noth, W. (1990). Handbook of semiotics. Indiana University Press. O’Neill, S. (2008). Interactive media: The semiotics of embodied interaction. Springer-Verlag. Orlikowski, W. (2000). Using technology and constituting structures: A practice lens for studying Technology in Organizations. Organization Science, 11(4), 404–428. Orlikowski, W. (2007). Sociomaterial practices: Exploring technology at work. Organization Studies, 28(9), 1435–1448. Orlikowski, W. (2010). The sociomateriality of organisational life: Conisering technology in management research. Cambridge Journal of Economics, 34(1), 125–141.

208 

J. Mingers and L. Willcocks

Orlikowski, W., & Scott, S. (2008). Sociomateriality: Challenging the separation of technology, work and organization. Academy of Management Annals, 2(1), 433–474. Orlikowski, W., & Scott, S. (2013). Sociomateriality — Taking the wrong turning? A response to Mutch. Information and Organization, 23(2), 77–80. Orlikowski, W.  J., & Scott, S.  V. (2021). Liminal innovation in practice: Understanding the reconfiguration of digital work in crisis. Information and Organization, 31(1), 100336. Oshri, I., Henfridsson, O., & Kotlarsky, J. (2018). Re-representation as work Design in Outsourcing: A semiotic view. MIS Quarterly: Management Information Systems, 42(1), 1–23. Peirce, C.  S. (1878). How to make our ideas clear. Popular Science Monthly, 12(January), 286–302. Peirce, C.  S. (1958). Collected papers of Charles Sanders Peirce (8 volumes). Harvard University Press. Pelizza, A. (2021). Towards a Sociomaterial approach to inter-organizational boundaries: How information systems elicit relevant knowledge in government outsourcing. Journal of Information Technology, 36(2), 94–108. Perrow, C. (1970). Organizational analysis: A sociological view. Wadsworth. Pinch, T., & Bijker, W. (1984). The social construction of facts and artifacts: Or how the sociology of science and technology might benefit each other. Social Studies of Science, 14, 399–441. Prates, R., de Souza, C., & Barbosa, S. (2000). Methods and Tools: A method for evaluating the communicability of user interfaces. Interactions, 7(6), 31–38. Price, R., & Shanks, G. (2005). A semiotic information quality framework: Development and comparative analysis. Journal of Information Technology, 20(2), 88–102. Raber, D., & Budd, J. (2003). Information as sign: Semiotics and information science. Journal of Documentation, 59(5), 507–522. Ramussen, J. (1986). Information processing and human-machine interaction. North Holland. Robichaud, D. (2002). Greimas' semiotics and the analysis of Organisational action. In K. Liu, R. Clarke, P. Andersen, & R. Stamper (Eds.), Coordination and communication using signs: Studies in Organisational semiotics 2 (pp. 129–150). Kluwer. Schultze, U. (2010). Embodiment and presence in virtual worlds: A review. Journal of Information Technology, 25, 434–449.

6  Interrogating Sociomateriality: An Integrative Semiotics… 

209

Schultze, U., & Orlikowski, W. (2010). Virtual worlds: A performative perspective on globally distributed, immersive work. Information Systems Research, 21(4), 810–821. Scott, S., & Orlikowski, W. (2022). The digital undertow: How the corollary effects of digital transformation affect industry standards. Information Systems Research, 33(1), 311–336. Searle, J. (1969). Speech acts. Cambridge U. Press. Shannon, C., & Weaver, W. (1949). The mathematical theory of communication. University of Illinois Press. Smith, M. L. (2006). Overcoming theory-practice inconsistencies: Critical realism and information systems research. Information and Organization, 16(3), 191–211. Stamper, R. (1991). The semiotic framework for information systems research. In H.-E.  Nissen, H.  Klein, & R.  Hirscheim (Eds.), Information systems research: Contemporary approaches and emergent traditions (pp.  515–528). North Holland. Stamper, R. (1997). Organizational semiotics. In J. Mingers & F. Stowell (Eds.), Information systems: An emerging discipline? (pp. 267–284). McGraw-Hill. Stamper, R. (2001). Organisational semiotics: Informatics without the computer? In K. Liu, R. Clarke, P. Andersen, & R. Stamper (Eds.), Information, organization and technology: Studies in Organisational semiotics (pp. 115–172). Kluwer. Suchman, L. (2007). Human-machine reconfigurations. Cambridge University Press. Tredinnick, L. (2007). Post-structuralism, hypertext, and the world wide web. Aslib Proceedings: New Information Perspectives, 59(2), 169–186. Trist, E., & Murray, H. (1993). The social engagement of social science - a Tavistock anthology, Vol II: The socio-technical perspective. University of Pennsylvania Press. Varela, F., Thompson, E., & Rosch, E. (1991). The embodied mind. MIT Press. Volkoff, O., Strong, D., & Elmes, M. (2007). Technological embeddedness and organizational change. Organization Science, 18(5), 832–848. Warschauer, M., & Grimes, D. (2007). Audience, authorship, and artifact: The emergent semiotics of web 2.0. Annual Review of Applied Linguistics, 27, 1–23. Wikgren, M. (2005). Critical realism as a philosophy and social theory in information science? Journal of Documentation, 61(1), 11–22. Winograd, T., & Flores, F. (1987). Understanding computers and cognition. Addison Wesley. Wittgenstein, L. (1958). Philosophical investigations. Blackwell. Woodward, J. (1958). Management and technology. HMSO.

210 

J. Mingers and L. Willcocks

Wynn, D., and Williams, C. 2008. Critical realism-based explanatory case study research in information systems. In International conference on information systems (ICIS2008). AIS. Zammuto, R., Griffith, T., Majchrzak, A., Dougherty, D., & Faraj, S. (2007). Information technology and the changing fabric of organization. Organization Science, 18(5), 749–762. Zuboff, S. (1988). In the age of the smart machine: The future of work and power. Basic Books.

7 When Crowds Play God: A Promethean Perspective on Crowdfunding Kieran Conboy , Rob Gleasure and Lorraine Morgan

,

Introduction The collapse of global financial markets in 2007 caused an international crisis often attributed to a culture of opaque dealings and corporate greed (Porter & Kramer, 2011; Szmigin & Rutherford, 2013). This led to protests internationally, most notably the ‘Occupy Wall Street’ movement (Calhoun, 2013). This movement, characterised by the phrase ‘we are the 99%’, is based on the belief that radical inclusivity and democratisation are solutions to a socially irresponsible financial system (DeLuca et al.,

K. Conboy (*) • L. Morgan J.E. Cairnes School of Business and Economics, University of Galway, University Road, Galway, Ireland e-mail: [email protected]; [email protected] R. Gleasure Copenhagen Business School, Frederiksberg, Denmark e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. P. Willcocks et al. (eds.), Advancing Information Systems Theories, Volume II, Technology, Work and Globalization, https://doi.org/10.1007/978-3-031-38719-7_7

211

212 

K. Conboy et al.

2012; Maharawal, 2013). More extreme protesters felt the system was fundamentally broken, suggesting a need for sophisticated new models to engage community members and create shared value (Barton, 2011; Porter & Kramer, 2011). Coinciding with these developments is the emergence of crowdfunding. The amount raised by crowdfunding now exceeds that raised through traditional venture capital (Barnett, 2015). Crowdfunding platforms allow project owners to launch open, global calls for financial contributions (Belleflamme et  al., 2014; Beaulieu et  al., 2015) and to leverage backer participation to build awareness and public image (Gerber & Hui, 2013; Belleflamme et al., 2014). Despite this potential, public commentary seems mixed. On the one hand, project owners can use crowdfunding to bypass traditional financial intermediaries, democratise the distribution of capital, and decrease distance between them and their investors (Ordanini et al., 2011). This decreased distance can facilitate greater interpersonal connection between project owners and their supporters (Mollick & Robb, 2016). Thus, project owners using crowdfunding have been attributed an independent and inclusive ideology that moves away from morally agnostic bottom-line thinking in favour of socially responsible and sustainable business practices (Gerber & Hui, 2013; Mollick, 2014). Others have taken a more sceptical view. For example, legal onlookers argue that crowdfunding platforms create a regulatory blind spot where unscrupulous project owners can exploit amateur investors (e.g. Hazen, 2011; Bradford, 2012). Even where intentions are honourable, many project owners have mixed feelings about ongoing crowd participation, particularly during sensitive creative or strategic periods (Scott, 2015; Gleasure et al., 2017). This creates a contrasting narrative: one in which project owners and backers may temporarily align themselves with the perceived ideology, only to revert to more traditional practices after the fundraising period. This is supported by observations that project owners’ updates and communication often decrease significantly after initial fundraising (Mollick, 2014) and that backers can become increasingly hostile (Wortham, 2012) and divided (Kim et al., 2017) when rewards are delayed or compromised.

7  When Crowds Play God: A Promethean Perspective… 

213

The tension between these two narratives has implications for project owners that use crowdfunding platforms. These crowdfunding platforms encourage backers to engage with project owners based on their passion for particular causes or creative goals, meaning emotions are often high and different groups are often in competition for territorial ownership and decision-making power (Kim et  al., 2016; Gleasure & Morgan, 2018). Similarly, the disruptive potential of crowdfunding platforms threatens to undermine existing funding markets in a way that creates significant challenges for incumbents and beneficiaries of those markets (Morse, 2015). The resulting polarisation of narratives cannot be ignored, as each has significant potential to help and harm the project owner. Thus, the first objective of this study is: 1. To identify competing archetypes used to describe crowdfunded project owners, as well as the tensions within those competing archetypes. While narratives may diverge, they both observe (and contribute to) the same empirical events. This creates a paradoxical situation in which competing narratives are simultaneously interdependent and contradictory. Under these conditions, Hegel (1807) argued a ‘dialectic’ analysis should identify specific ‘truths’ within each version of events, such that some new, superior richer, complete understanding can emerge from the synthesis. This is difficult to achieve when looking at competing narratives holistically, particularly when those narratives diverge as sharply as they do in a crowdfunding context. However, deconstructing narratives into key tensions allows such dialectic analysis to focus on specific interdependencies and contradictions (see De Rond & Bouchikhi, 2004). The ultimate value of a dialectic is the new richer, complete understanding created. Thus, the second objective is: 2. To identify recommendations for project owners to help manage the narrative around specific projects. For reasons explained later in the chapter, we apply a Promethean lens to examine how this polarisation can be analysed. Specifically, this study frames each project owner using crowdfunding platforms as Promethean,

214 

K. Conboy et al.

or at least potentially Promethean, in the sense they take financing power (fire) from traditional financiers (the pantheon of Gods) and use the platform to give it to crowds of backers—those newly empowered humans who would not previously have had such an opportunity. The following section describes the emergence of crowdfunding and role of public narratives in generating value from these technologies. The next section explains the suitability of Prometheus as a sensitising lens. We then describe six pairs of contrasting Promethean archetypes that can be related to crowdfunding. Building on this, the method for the study is presented before findings from three illustrative example identify six emerging synthesis-based recommendations. These can collectively help project owners to manage public narratives around fundraising and the delivery of rewards. Finally, the contributions of the study are discussed.

Theoretical Background The Origins and Variations of Crowdfunding Crowdfunding platforms bypass traditional funding intermediaries by connecting fundseekers directly with crowds of potential backers (Belleflamme et al., 2014; Beaulieu et al., 2015). These platforms provide alternative financing models for a number of domains that attract public attention, from scientific research (Wheat et  al., 2013), to journalism (Aitamurto, 2011), to renewable energy (Vasileiadou et al., 2016), to a multitude of commercial ventures (Frydrych et al., 2014; Mollick, 2014). Multiple different paradigms of crowdfunding exist, the four most popular being (c.f. Belleflamme et  al., 2014) (i) lending, in which backers receive repayment with interest; (ii) equity, in which backers receive some ownership stake in a venture; (iii) rewards-based, in which backers receive some non-financial repayment, for example t-shirts, early versions of products; (iv) charity, in which backers donate for personal or social reasons. Rewards-based crowdfunding projects have arguably attracted the most intensive, sustained, and complex public discussion. This is partly because of the appeal to a general consumer audience, in addition to ideologically or financially motivated backers (Gerber & Hui, 2013). It is

7  When Crowds Play God: A Promethean Perspective… 

215

also because a series of ‘blockbuster’ projects have attracted disproportionate amounts of funding and generated large amounts of mainstream media attention (Liu et al., 2015). Such blockbuster projects have created strong public narratives around new products and services, with notable examples including the Pebble smartwatch, the Veronica Mars film, and the Exploding Kittens card game. The consumer-focus and sustaining public discussion of reward-based crowdfunding means it is especially important to understand how narrative is managed in this format. Hence, this study focuses on the rewards-based paradigm to maximise practical value and maintain a manageable theoretical scope.

Crowdfunding to Construct Public Narratives Narratives are important for most organisations and projects (Rantakari & Vaara, 2017; Vaara et al., 2016). However, public narratives are especially important for crowdfunding projects for three main reasons. The first concerns the ability of crowdfunding to generate awareness of new products and services. This is one of the major espoused benefits of crowdfunding, as large numbers of consumers can be reached early in the development process (Gerber & Hui, 2013; Kunz et  al., 2016; Beier et al., 2019). This early exposure can help relatively small ventures achieve widespread media attention, with notable examples including education projects such as Reading Rainbow, entertainment projects such as Exploding Kittens, and technological projects such as the Pebble smartwatch. Some have identified crowdfunding projects that don’t actually appear to need the funding and are instead motivated almost entirely by this opportunity for early consumer awareness, for example the ‘Pebble 2’ (see Brown et al., 2017). This public awareness can also become a double-­ edged sword where public narratives become negative, as openly visible backer frustration devalues the project and disincentivises project owners (Gleasure et al., 2019). The second reason why positive public narratives are important for crowdfunding projects is the ability to create a sense of shared ownership. The act of crowdfunding creates a shared commitment to a product or service that extends beyond the ordinary relationship between ventures

216 

K. Conboy et al.

and consumers (Zheng et al., 2018). This can increase mutual loyalty and encourage project owners and consumers to assume additional promotion activities in the pursuit of shared interests (c.f. Li et  al., 2019). Equally, where a narrative becomes negative it can also result in a form of ‘hijacking’, in which backers attempt to wrestle back control from project owners who are seen to neglect these shared interests (Wilson et al., 2017). The third reason why positive public narratives are important for crowdfunding projects is the ability to generate creative input, that is to use funding as a gateway for open innovation (Giudici et  al., 2018). Those backers who have become committed to a project are often keen to contribute ongoing creative input (Hui et al., 2014). However, this once again relies on a suitably positive narrative around a project as backers must buy into the ‘cause’ or social value of a project (Gerber et al., 2012; Gleasure et al., 2017).

The Myth of Prometheus as a Sense-Making Lens This story of new technologies being met with contrasting narratives is not new. Many technologies have been introduced for social good, leading to a battle for public opinion based on personal interests and mixed outcomes. Indeed, this is true of the Internet itself (e.g. DiMaggio et al., 2001; Warschauer, 2004). This phenomenon has been captured in one of the prevailing myths in western culture—the Greek myth of Prometheus. This story describes how Prometheus tried to free humankind from the oppression of the gods by stealing fire from Zeus and giving it to humans, thus giving unprecedented control of their environment and destiny. For three reasons, the myth of Prometheus provides a useful sense-­ making lens with which to frame crowdfunding narratives. First, while there are many types of paradox and methods for paradox resolution (Lewis, 2000; Schad et  al., 2016; Huber et  al., 2017;), they generally address collective approaches and outcomes, and pay less attention to actors and their relationships within and across paradoxes (Schad et al., 2016). Using Prometheus as a metaphor allows the tension to be viewed from the perspective of the main actors. Hatch et al. (2005) note that the

7  When Crowds Play God: A Promethean Perspective… 

217

archetypal nature of Greek gods makes their metaphorical, symbolic format highly transferable to current day situations. They are ‘messy’ and ‘capricious’ while retaining a clear sense of who they are and what they represent (Stacey, 2010). Such a paradox is common in contemporary, ‘liquid modern’ organisation environments (Doyle & Conboy, 2020), none more so than the open turbulent mix of personalities interacting on crowdfunding platforms. An exploration of the contradictory nature of the Greek gods and goddesses can provide insight into ways of resolving or accommodating such paradox (Schedlitzki et al., 2015; Stacey, 2010). Mythical Greek gods and their associated archetypes have been used to study issues in management literature; for example, leadership development (Schedlitzki et al., 2015), strategic resilience (Marshall & Ojiako, 2010), gender issues (Olsson, 2000) and the role of academia for informing practice (Conboy, 2019; Conboy, 2021; Greenberg et al., 2007) are often used to influence and reinforce behaviours (Jacobi, 1974; Mitroff, 1983). The Promethean perspective conceptualises the polar extremes of how a project owner’s actions may be interpreted by the stakeholders involved. Therefore, we argue it provides a unique analytical vocabulary with which to study the diverging narratives around crowdfunding. Second, the Promethean myth is about breaking from an embedded and hierarchical status quo. Crowdfunding communities act as a form of ‘resource mobilisation’, in which individuals with limited individual financial resources can pool those resources, so accessing investment options usually reserved for the wealthy (Gerber & Hui, 2013). It is this newfound access that ostensibly offers to bring about a democratisation-­ related Promethean leap in shared value-based capitalism. New god-like capabilities to select and direct complex projects and ventures are presented to the broader population; capabilities traditionally reserved for a small minority (a pantheon, even) of resource-rich individuals and organisations. Third, the Promethean myth is layered with perceptions of power and emancipation. Just as critics of Prometheus defended the existing balance of power and decision-making based on a sceptical view of humankind, so critics of crowdfunding doubt the ability of backers to make informed and responsible decisions (e.g. Hazen, 2011; Hanks et al., 2014).

218 

K. Conboy et al.

Promethean Archetypes Given the contrasting narratives around crowdfunding, we specifically focus on research that presents the myth of Prometheus in terms of contradictory, competing archetypes. A review of the literature identified 12 archetypes along 6 dimensions (see Table 7.1). Each are now presented along with the corresponding thesis and antithesis, operationalised in a crowdfunding context by analogising project owners as Prometheus, financing power as fire, traditional financiers as the pantheon of Gods and crowds of backers as newly empowered, democratised humans. Prometheus the Philanthropist versus Misanthropist  Benevolence and human betterment is the most common archetype associated with Prometheus (Dougherty, 2006; Morford et  al., 2015; Raggio, 1958). After stealing fire from Zeus, he gifted it to mankind. This gift of fire translates into the gift of technology and human advancement (Dougherty, 2006; Morford et al., 2015). This gift also extends beyond the fire (technology) itself to the gift of democratisation—to give humans the power Table 7.1  Thematic mapping of archetypes from literature on Prometheus Archetypes of Prometheus

Thesis

Philanth ropist

Enlightener Revolutionary and forethinker Liberator Empowerer Altruist

Antithesis

Misanth ropist

God of reckless ambition

Thief

Creator of work

Dougherty (2006) Grene (1940) Lloyd-­Jones (2003) Morford et al. (2015) Raggio (1958) Witzel (2012) Ziolkowski (2000)

X

X

X

X

Acclaimed sufferer

Self-­ Maligned interested, and cunning ignored trickster by humanity X X X X

X

X

X

X X

X

X

X

X

X X

X

7  When Crowds Play God: A Promethean Perspective… 

219

to choose what parts of that gift to use and not (Dougherty, 2006). Prometheus is referred to as the ‘vehement and weariless champion’ of humans against oppression (Morford et al., 2015). Yet, the downside of this gift is that fire also represents the ‘historic tools of devastation and destruction’ (Dougherty, 2006, p.  18) and so Prometheus has been referred to as ‘the initiator of technological evil’ (Ziolkowski, 2000), unleashing humankind’s powerful, all-consuming and sometimes unethical desire for knowledge, technology (Ziolkowski, 2000) and ultimately power (Dougherty, 2006). In a crowdfunding context, the philanthropist interpretation would suggest that while product innovation and delivery are usually a key criterion in a rewards-based project, a crowdfunding initiative would also ensure that the product is designed to create ‘good’ in the world, even if an opportunity for extra profit or reward is missed and reduced as a result. A misanthropic view would indicate that if it means increased reward, a crowd project allows or even encourages potentially ‘bad’ uses with negative outcomes. 

Thesis: Project owner uses crowdfunding platform to demonstrate and create opportunities for philanthropy. Antithesis: Project owner uses crowdfunding platform to demonstrate and create opportunities for misanthropy.

Prometheus the Revolutionary ‘Forethinker’ Versus Prometheus the God of Reckless Ambition  The origins of the name Prometheus lie in the compound of pro meaning ‘before’ and metis meaning ‘clever intelligence’ (Dougherty, 2006). Promethean fire is a ‘symbol of defiant progress’ and foresight (Morford et al., 2015). The distinction between regular and Promethean foresight is the latter represents something game-­ changing and previously incomprehensible (Dougherty, 2006)—imagine those who discovered fire trying to envisage or articulate it beforehand. The opposing archetype is one cited by researchers as Prometheus’ enablement of ‘reckless ambition’, where extreme innovation, even if well intended, can bring negative unforeseen consequences (Raggio, 1958; Dougherty, 2006; Morford et al., 2015). Researchers of the Promethean

220 

K. Conboy et al.

myth have shown that such reckless ambition is not just a possibility but an inevitability. In fact, this archetype is further heightened by Prometheus who intentionally and proudly ‘stopped mortals from seeing their fate’, and so by giving them fire ‘planted in them blind hopes’, blissfully unaware of the potential downfall to come (Morford et al., 2015).  Thesis: Project owner uses crowdfunding platform to demonstrate and create opportunities for revolutionary forethought. Antithesis: Project owner uses crowdfunding platform to demonstrate and create opportunities for reckless ambition.

Prometheus the Liberator Versus Thief On the one hand, previous research portrays the theft of fire as an offence, a crime, taking from the gods one of the key features that differentiates them from mortals. While in the classics it was clear that the name Prometheus referred to forethought, recent etymological research links the term meth of Pro-meth-­ eus to a Sanskrit root math- which means to steal (Dougherty, 2006; Witzel, 2012). A counter-argument views the theft as liberation: of taking back something that did not belong to the gods but something to which humans were divinely entitled. In a crowdfunding context, one could view the platform technology as being liberated and used as a mechanism for the global population of backers to achieve liberation, to contribute to the building of what they want and for their good, rather than the building of a product that may exploit them and their needs and ultimately not fulfil what they want.  Thesis: Project owner uses crowdfunding platform to demonstrate and create opportunities for liberation. Antithesis: Project owner uses crowdfunding platform to demonstrate and create opportunities for theft.

Prometheus the Enlightener and Empowerer Versus Prometheus the Creator of Work  Prometheus is often portrayed as the god of education and enlightenment, who provided humans with knowledge of science,

7  When Crowds Play God: A Promethean Perspective… 

221

engineering, maths and art (Dougherty, 2006; Raggio, 1958; Morford et al., 2015). While Athena and Prometheus are both perceived as intelligent gods of education, Athena ensured a continual dependence on her for wisdom imparted whereas Promethean education places emphasis on empowerment and self-improvement (Greenberg et al., 2007). However, the opposing argument is that prior to Prometheus’ theft of fire, the earth ‘gave its fruit freely’ (Dougherty, 2006) and humans did not have to work. Zeus’ punishment was to force humans to then become self-reliant, and as Dougherty (2006) notes, this aspect of the myth is often used to portray the back-breaking work of agriculture and the soulless toil of industrial revolution. In a crowdfunding context, the empowering aspect is one where backers are empowered: to contribute to the ideas or the tasks that go towards building the product they will use, or indeed not contribute at all, and simply reap the benefits of what finally emerges. The negative connotation is that the product owner gets the backers to toil and engage in all the back-breaking work that goes into designing, building and testing the final product, with the owner observing in a god-like manner, watching the backers toll, choosing the features he or she wants even if not those that are in the best interests of the backer population, while still owning all the product rights and intellectual property.  Thesis: Project owner uses crowdfunding platform to demonstrate and create opportunities for enlightenment and consumer empowerment. Antithesis: Project owner uses crowdfunding platform to demonstrate and create opportunities to increase consumer work.

Prometheus the Altruistic Versus Prometheus the Self-Interested, Cunning Trickster  As with benevolence the common assumption in the literature is that the theft of fire was for selfless reasons. The counter-­ perspective describes Prometheus as a cunning trickster (Dougherty, 2006; Witzel, 2012), (Raggio, 1958) and a thief (Witzel, 2012, p.357, 366). Rather than altruism, some argue the theft was to validate his god status while simultaneously diminishing that of Zeus, with humans merely the ‘pawn’ in a ‘gigantic battle of divine wills’ (Morford et al., 2015). 

222 

K. Conboy et al.

Thesis: Project owner uses crowdfunding platform to demonstrate and create opportunities for altruism. Antithesis: Project owner uses crowdfunding platform to demonstrate and create opportunities for self-interest and trickery.

Prometheus the Acclaimed Sufferer Versus Prometheus the Maligned and Ignored by Humanity  Having given fire to humans, Prometheus had to witness the subsequent misanthropy; thus, he is the symbol of the suffering creator (Grene, 1940; Dougherty, 2006). However, despite Prometheus being credited with empowerment of humanity, the lack of appreciation and even awareness of his existence is notable (Raggio, 1958). Lloyd-Jones (2003) refers to the irony that the Olympic flame— what is left of the ancient torch race—is iconic, but its original purpose of honouring Prometheus’ fire is known to few. However, while few know Prometheus, the masses are familiar with Pandora, despite her raison d’etre was to avenge the deeds of Prometheus. Further, when Prometheus is referenced, it is sometimes to apportion blame, for example the burning of Athens in 500 BC and the bombing of Dresden after World War 2 (Dougherty, 2006).  Thesis: Project owner that uses a crowdfunding platform is acclaimed by and suffers for backers. Antithesis: Project owner that uses a crowdfunding platform is maligned or ignored by backers.

Method A Hegelian Dialectic The extraction of knowledge from competing narratives is typically performed through a ‘dialectic’ analysis which seeks to identify and reconcile competing versions of events (Benson, 1973; Myers, 1995). Under such polarisation, Hegel (1807) proposed that ‘true’ knowledge is obtained using a triadic relationship later summarised as thesis, antithesis and

7  When Crowds Play God: A Promethean Perspective… 

223

synthesis (Carlo et al., 2012; Wijnhoven, 2012). This process begins with a proposed acceptable description of reality (the thesis). Perceived limitations or contradictions in this thesis lead to the emergence of an alternate description (the antithesis). However, there is no possibility to simply choose between the opposing poles or to locate them in different spheres. Instead, according to the dialectical logic of Hegel, there is a simultaneous presence of conflicting ideas. As a result, a third point of view must arise to supersede flaws and reconcile the thesis and antithesis (a synthesis). It then becomes the new thesis which in turn generates another antithesis, a subsequent new synthesis and so on. A review of the literature reveals three key themes underpinning Hegel’s anatomy of dialectics. First, a ‘thing’ is argued to become more fully developed and better through successive dialectic or self-criticism and reconstruction. Second, Hegel (1975) regarded dialectics as ‘an inevitable and inherent component of worldly existence’, necessary to the creation of all things and specifically the transition of things, from potentiality or abstraction to actuality and content. Third, Hegel believed in the acquisition of an overarching universal—each new thesis represents a process of continuous advance over the previous thesis, until a ‘true’ endpoint is reached. The Hegelian dialectic has been applied to a number of situations in which competing perspectives are entangled with complex and emotive relationships of varying power-distance, for example the emotional and physical struggle and potential rebuttal of underprivileged people to their social, economic and political environment (Howard, 2010) and Hegel’s own application of dialectics to the master-slave relationship (Sinnerbrink, 2007, p. 20). This suggests the Hegelian dialectic is also appropriate for engaging with nuances in the relationship between backers and project owners that use crowdfunding platforms.

Selection of Illustrative Examples Three Kickstarter projects were purposefully selected as illustrations. Kickstarter declared itself a public benefit corporation (PBC) in 2015, meaning shared value and social benefits are part of the platform’s

224 

K. Conboy et al.

regulatory self-identification and public narrative. Kickstarter also operates a rewards-based crowdfunding model that has arguably best captured public attention (Gerber & Hui, 2013; Frydrych et al., 2014). Such high levels of fame/infamy increase the symbolic power of associated narratives and adds to their importance for broader collective sense-making (Trice & Beyer, 1984). Further, Kickstarter is the largest rewards-based crowdfunding platform at the time of writing. Thus, the narratives around project owners on Kickstarter are arguably most likely to divide public perception and provide rich insights into the tensions therein. Five selection criteria were used (Table 7.2). First, the projects achieved a sufficient level of media attention and collective awareness to take on some symbolic or expressive qualities. Without this, the generative power of the narrative around the project owner may be limited. Second, projects were selected that attracted controversy and divisive narratives to provide the dialectic richness to generate meaningful insights. Third, projects were selected that span different target industries, to look for industry-related moderators that may impact the interpretation of findings. Fourth, projects were selected with varying salience of shared value considerations, to facilitate comparative analyses of such motivations on other behaviours and outcomes. Fifth, projects were selected with different levels of commercial success. This allows the analysis of narratives to be balanced against indicators of tangible outcomes and their ability to meet backers’ expectations. Oculus VR proposed to develop a head-mounted virtual reality (VR) headset called the ‘Oculus Rift’ capable of revolutionising the video game and digital entertainment industries. Oculus VR (Prometheus, according Table 7.2  Characterisation of illustrative examples Media attention Controversy Industry Amanda High Palmer Oculus High VR Torquing High

High High High

Shared value Commercial motivations success

Music and High performance Digital Medium entertainment Lifestyle and Low technology

Medium High Low

7  When Crowds Play God: A Promethean Perspective… 

225

to the lens adopted here) presented the project to backers (humans) as both utilitarian and revolutionary, arguing the arrival of VR would represent a significant change in the landscape of digital interaction. Following lengthy discussion on different VR-specific websites and fora, Palmer Luckey (founder of Oculus VR) launched a Kickstarter campaign in August 2012 to fund development of the Oculus Rift, arguing this would allow greater development freedom than if they used traditional investors or existing platform owners (the gods). That campaign raised $2.4 million, almost ten times the initial goal of $250,000. The second illustrative example is Amanda Palmer (Prometheus) and her campaign entitled ‘Theatre Is Evil: the album, art book and tour’, launched in 2011, which aimed to pioneer new musician-consumer practices. These practices would bypass record labels (the gods) to allow musicians to interact with fans (humans) more directly and earn income from voluntary contributions. Amanda Palmer is an American singer-­ songwriter and her Kickstarter campaign surpassed its goal of $100,000 in September 2012, generating more than $1 m in pre-orders—the largest and most successful crowdfunding campaign ever undertaken by a musician at that time. The third illustrative example is the ZANO drone, proposed by the Torquing Group (Prometheus) as an intelligent, autonomous nano-drone that would allow individuals to capture photo and video ‘selfies’ from aerial positions. The group sought backers (humans) rather than traditional investors (the gods) because ‘to make the ZANO truly accessible to everyone we need to get the volumes up so the price goes down – it’s that simple’. This meant shared value was least salient in narratives around this project, as most backers focused on the practical, hedonic value of the technology. The founders sought £125,000 funding in November 2014. Over £2 m was raised, the most by any European crowdfunded project to date. However, the company only managed to deliver 600 of 15,000 pre-ordered drones by October 2015, of which only four were delivered to Kickstarter backers, and in December 2015 they declared bankruptcy.

226 

K. Conboy et al.

Gathering Data Crowdfunding projects take place in a digital environment in which public discourse is openly visible to all. Hence, for descriptive validity, data collection focused on discourse and behavioural data left behind from interactions and events (c.f. Urquhart & Vaast, 2012; Levina & Vaast, 2015; Whelan et al., 2016). The Kickstarter platform was the first and most formative data source for each illustrative example, providing the fundraising pitch, backers’ comments on the campaign (during and after fundraising), and public updates provided to backers since the campaign launched (including additional backers’ comments specific to those updates). Data from Kickstarter were compared and contrasted with data from other ‘less central texts’ (Urquhart & Vaast, 2012) on other sites. This additional data reflects the need for qualitative studies to make use of diverse data and secondary sources to improve internal validity via triangulation (Miles & Huberman, 1994) and increase contextual sensitivity (Ragin, 1992). Two sources of ‘less central texts’ were selected. First, we selected external platforms where project owners and backers could interact outside of Kickstarter, for example project owners’ personal websites, blogs, social media accounts, videos and other public communications. Second, we selected discussion among non-participants and third-party media coverage/commentary, for example news reports and critiques. A complete list of sources is provided as appendices.

Analysing and Identifying Themes Analysis of the narrative provided support for all six pairs of contrasting archetypes, though some featured more than others in each illustrative example. Pairs were characterised according to emerging latent themes (c.f. Braun & Clarke, 2006) describing different events and instances of discourse. These themes were related to the relevant archetypes as part of iterative coding, until each set of contrasting themes could be characterised by a thesis and antithesis that (i) accurately described the data across all three illustrative examples, (ii) distilled the narratives around discrete

7  When Crowds Play God: A Promethean Perspective… 

227

loci of disagreement, (iii) related those opposing positions in suitable terms to the overarching Promethean narrative. Finally, dialectic analysis synthesised these contrasting themes into six corresponding and complementary prescriptive recommendations that aimed to capture a more complete, richer unifying understanding of what is present in both the both positive and negative narratives. Note the prescriptive nature of these syntheses reflects Hegel’s emphasis on practical philosophy and social agency (Pippin, 2000). Hence, syntheses were created to inform action rather than passive understanding. Cross-­coder comparisons were made throughout the coding and theory building process to test the reliability and validity of findings (Benbasat & Zmud, 2003). Further, a temporal element was noted in that while each theme was continuously evolving, some were more prevalent in the early envisioning stage of a project, some during the operationalisation and some during post-project reflection.

Findings Philanthropists or Misanthropists? Most discussion describing project owners as philanthropists or misanthropists occurred early in the narratives, particularly those at the outset of the project. This makes sense, given this narrative has a big picture focus that may provide a useful foundation for subsequent diverging discussion. The first theme describing project owners as misanthropists suggested they were introducing a new product to make consumers’ lives more enjoyable. This theme had relatively little discussion for Amanda Palmer, presumably because the actual product in question was not necessarily innovative in the same way as the ZANO or Oculus Rift (i.e. was not a new medium or platform). For Torquing, most instances of narrative focusing on this impact involved backers expressing their excitement at the idea of ‘swarms’ of ZANO drones capturing events from every angle. More intensive discussion was observed for Oculus VR, which many backers heralded as the gatekeeper to a new VR-based

228 

K. Conboy et al.

techno-­utopian fantasy, for example a number of individuals on Kickstarter referred to ‘childhood dreams’. The second theme described project owners as introducing a new practice that will encourage businesses to act responsibly towards consumers and society in general. This theme had little or no discussion for Torquing and only became salient for Oculus VR after consumer headsets were given as gifts, with some further extending gratitude to Facebook (despite that company’s lack of popularity among other backers). However, this theme was prominent for Amanda Palmer, who openly despaired at the overly commercialised nature of the music industry, which she described as failing ‘on a cosmic level’. Many backers hailed the crowdfunding approach as visionary and claimed her campaign should set the precedent for many creative artists to follow. The antithesis characterised project owners as misanthropists, who were launching products that would make the world a worse place to live for humankind. The first antithetical theme argued project owners were introducing a new product that will ultimately make consumers’ everyday lives miserable. This was not observed for Amanda Palmer; however, Oculus VR and Torquing did receive some external coverage to this effect. Critics of Oculus VR questioned whether VR could have serious negative consequences, for example several online media articles pointed to issues such as physical decline, increasing desensitivity to violence, pornography and identity theft, irresponsible marketing and subliminal messaging, and coping with bereavement. Critics of Torquing pointed to the privacy implications for individuals recorded without their consent. What is interesting about these accounts is not their existence (each is intuitive) but their infrequency, which was almost entirely absent among collaborating actors. On the one hand, these actors have already decided to engage with a project, so are presumably satisfied with the vision overall. Nonetheless, concerns of this type would appear natural candidates for mindful and explicit discussion, consensus and management. The second antithetical theme concerned the precedent for new practices, suggesting project owners were introducing a new practice that will cause businesses to act selfishly and create a dysfunctional industry. As with the corresponding thesis, this theme had little or no discussion in the Torquing example. However, for Oculus VR, external commentators

7  When Crowds Play God: A Promethean Perspective… 

229

argued that the trend towards ‘early access’ was actually damaging the reputation of independent companies involved in game development. This was partly due to the tendency to create unrealistic hype (leading to eventual disappointment) for the purposes of fundraising. It also diluted interest in the actual experience, as playable finished products were often years away when their media coverage peaked. Similarly, Amanda Palmer critics lamented the burden of crowdfunding on introverted artists.

Revolutionary Forethinkers or Recklessly Ambitious? Early narratives also focused on long-term goals of the projects. At a dialectical-­level this acts as a future-looking negotiation reflecting the ‘stakes’ for the project, including those laid out at the outset and those emerging as a project unfolds. Thus, the thesis from positive commentators characterised project owners as demonstrating revolutionary forethought. The first theme suggested project owners could dominate emerging markets by leveraging crowdfunding. This theme was salient across all three illustrative examples. Oculus VR highlighted the value of Kickstarter as a means to ‘get these dev kits into the hands of developers as fast as possible’ to generate standards and feedback. Backers went on to celebrate the relationship for a variety of other reasons, including the ability to maintain independence from large platform owners who could stifle its reach and impact. The ZANO was described by both backers and Torquing as ‘revolutionary’, ‘cutting-edge’ and ‘a game-changer’ for the drone industry. Backers of Amanda Palmer were perhaps most ambitious, suggesting her exemplary use of Kickstarter would not only allow new types of musician to flourish, but it would herald an entirely new age for the music industry. The second theme focused on project owners’ ability to attract new consumers in related markets. This theme was most salient for Torquing, which actively marketed the ZANO in this vein. The project owners argued that the small size, minimal cost and ease of use meant the drone had the potential to open a previously niche set of capabilities to the wider public. This potential was the reason Endgadget (a tech news site) shortlisted the ZANO in the top 49 of 20,000 projects at the 2013

230 

K. Conboy et al.

Consumer Electronics Show (the world’s largest trade show). Similarly, backers of Amanda Palmer suggested her project was enabling artistic projects to engage with mostly neglected less-mainstream audiences, for example ‘Those of you with artistic visions of your own, look to the example she has set and get going. Make your art. Improve your craft. Connect with people and share. The bar hasn’t been raised, the whole damn floor has been lifted up. I can’t think of a more exciting time for the creative soul.’ Backers of Oculus VR were initially least vocal around this theme, as many were part of a pre-existing sub-community of developers. However, over the course of the project the excitement towards the headset from non-developers fed into a collective optimism among positive commentators concerning the product’s reach as a ‘paradigm shifter’. The antithesis characterised project owners as demonstrating reckless ambition. Unlike the aforementioned optimism that project owners could dominate emerging markets, the first antithetical theme predicted these companies were doomed to fail because they were being encouraged to be naive and unrealistic. This became salient across all three projects when delays in production began to emerge. Oculus VR had even foreseen this and, mid-way through fundraising, asked that non-developers cease from making further contributions. Yet this was largely forgotten by backers when these delays became a reality, at which point many of those backers began to question the project owners’ professionalism. The same criticisms were levelled at Amanda Palmer; however, it was Torquing for whom this theme was most salient. Numerous additional features were added as part of stretch goals to attract funding, including extras like facial recognition and 360-degree panoramas. Several backers warned about the threat this feature bloat presented for achieving basic functionalities. An interviewee for the subsequent Kickstarter investigation attributed this to ‘a pervasive culture of overconfidence’, suggesting the project owner felt they had already done the hardest part of development in securing funding. The second antithetical theme suggested that rather than attracting new consumers, project owners would disappoint and alienate them by presenting specific products or services before they were consumer-ready. This was particularly common in external media coverage of Oculus VR, with several different sites reporting on aesthetic and commercial

7  When Crowds Play God: A Promethean Perspective… 

231

limitations. Some of these pointed to the price of the Rift, others were more concerned with fundamental issues, such as the volatility of nascent markets and the need for perfect functionality. Criticisms of Amanda Palmer suggested she lacked the financial management skills to sustain a career, with many media sources comparing her fundraising to a form of begging. For Torquing, this theme only emerged when production issues and other technical problems created delays. Backers began to query the feasibility of individual functions while external media sources queried whether the promotional video created a misleading impression of the drone’s capabilities and readiness.

Liberators or Thieves? The contrasting narratives positioning project owners as liberators or thieves typically focused on the earlier stages of projects. The thesis from positive commentators characterised those project owners as liberators for consumers. The first theme was that backers were being given the capability to have creative input they would otherwise have been denied. For both Oculus VR and Torquing, this meant backers saw themselves as active participants in development, either because of early access to development resources or the chance to offer recommendations and take part as beta testers. For Amanda Palmer, this theme manifested slightly differently, as instead of testing or developing content directly, backers sought to encourage additional tour locations and to ensure their success by actively promoting them. The second theme concerned the sharing of creative outputs with the crowd that are not widely available. For Amanda Palmer, this included a range of exclusive options, including ‘art-sittings’, private dinner parties, concerts, paintings and other works. Backers also received 30 pages of song-inspired album art and thank-you cards inside the Kickstarter-­ backer editions, and depending on the backer package, received full art books with new material from other artists and musicians, including Palmer’s author husband Neil Gaiman. For Torquing and Oculus VR, the intention was to make most or all the outputs commercially available in the future. Thus, this theme focused on the earliness with which backers would have access to these outputs.

232 

K. Conboy et al.

The antithesis characterised project owners as thieves. This interpretation applies the concept somewhat figuratively, as traditional investors and potential partners did not lose material or financial resources directly. Instead, they were denied opportunities that could have acted as the catalyst for longer-term value creation. The first antithetical theme described how the rush to mainstream markets meant potential long-term partners were being denied the opportunity for creative input. This theme was most obvious for Oculus VR, where potential partners included game developers such as Epic and Valve, and more divisively, companies such as Microsoft and Sony that owned key gaming platforms. This theme was also relevant when Amanda Palmer asked musicians to come play on her tour in exchange for only beers and hugs. This prompted a backlash among many external commentators, with many arguing she should more readily share her good fortune with other artists that may not have benefitted from the same financial support. The second antithetical theme argued the rush to mainstream markets also meant project owners stole potential partners’ opportunity to expand on creative outputs with complementary products and services. Sceptical commentators argued that emerging markets required new products or services be presented to consumers as part of a product/service ecosystem. This theme manifested early for Amanda Palmer, as external media sources defended the role of music labels and suggested her project may mislead other artists without her experience, background and established connections. In contrast, this theme was not initially evident for either Torquing or Oculus VR.  However, that changed for the latter after Oculus VR’s acquisition by Facebook. While many backers were outraged, others admitted they thought the long-term feasibility of VR was increased once power was returned to some such existing large businesses.

Enlighteners and Empowerers or Creators of Work? The next loci of disagreement concerned the greater involvement of consumers. The thesis for this redistribution of responsibilities was one of empowerment. The first theme reflected backers’ ability to decide whether a project should go ahead. This was predictably common when

7  When Crowds Play God: A Promethean Perspective… 

233

fundraising was underway, during which many backers commented on this sense of empowerment as a collective motivation for assuming additional financial responsibility. This theme also manifested after fundraising was completed, as part of a proud reflective narrative pre-empting the long-­term positive impact of projects and the landmark it could represent. However, this theme was strongest for both Amanda Palmer and Oculus VR, both of whom were praised for embracing their consumers. This idea was mentioned notably less in the narrative around the Zano, though many backers celebrated the level of fundraising early on. The second theme concerned backer’s ability to ‘lift the hood’ on development to observe and contribute to activities that would otherwise be hidden from them. This theme was most explicit for Torquing, whom several backers heralded as strong positive exemplars of good communication early on in the project. Amanda Palmer also received significant praise on this front early in her campaign, as backers and external commentators celebrated her ongoing philosophy of honest disclosure. This theme was least prominent for Oculus VR; however, it became increasingly explicit as soon as backers felt their expectations for ongoing updates were not being met. Indeed, this tension was observed across all three illustrative examples when production was delayed or unforeseen issues appeared to have arisen that prevented project completion. The antithesis framed each project owner as creating additional work for consumers. Unlike the thesis, which suggested backers were privileged to assume additional responsibilities for projects, this antithesis suggested backers were being exploited under the facade of collaboration. The first antithetical theme focused on the additional risk assumed by backers by giving money to products/services that may never actually be delivered. This theme was evident across all three illustrative examples, particularly when the delivery of rewards became compromised. However, the idea that backers had been duped into assuming financial risk was by far most salient for the Torquing and Oculus. Many backers refused to accept they had assumed additional risk when much of the eventual benefit would go elsewhere, with some going so far as to accuse project owners of each project of intentional fraud. The second antithetical theme was that backers are forced to assume additional communication and development practices, not only in terms

234 

K. Conboy et al.

of funding but more fundamentally from a business development perspective. Backers of Amanda Palmer and Torquing were tasked with ‘spreading the word’ to ensure the respective projects reached as large an audience as possible. Backers of Oculus VR also assumed these promotional responsibilities, not only among consumers but among other developers. Oculus VR further encouraged these backers and external developers to begin developing the applications that would make the device appealing to mainstream audiences.

Altruists or Cunning/Self-interested Tricksters? Competing narratives also formed around the new boundaries of involvement and distributions of responsibilities. The thesis from positive commentators characterised each project owner by altruism. This meant that collaborating actors were willing to compromise on certain selfish interests if it means more likelihood of the project having a broader impact. The first theme for this sacrifice related to compromise from the project owners. This was evident across all three projects. For Oculus VR and Amanda Palmer, several backers suggested that the passion of the founders meant they were willing to sacrifice financial gain for the sake of high-­ quality outputs, that is better headsets and artistic fidelity, respectively. The narrative for Torquing celebrated their attempts to stick with local manufacturers, with some backers noted how unusual it was not to outsource component production to cheaper countries. The second altruistic theme was the corresponding willingness for backers to sacrifice their selfish interests. This theme was most prominent across all three projects when delays began to challenge the early access nature of backers’ rewards. As criticisms began to mount from disenfranchised backers, many backers encouraged patience and suggested these types of delays were common. This unselfishness became more pronounced for Oculus VR after the Facebook acquisition, when many clearly disappointed backers expressed their best wishes for Oculus in the future, even though they were no longer personally ‘along for the ride’. The antithesis characterised project owners according to cunningly veiled self-interest. The first antithetical theme was that project owners

7  When Crowds Play God: A Promethean Perspective… 

235

were only committed to their relationship with backers as long as it added direct practical value to their business. Predictably, this was most salient for Torquing. Disillusioned backers made allegations that ranged from ineptitude, to criminal neglect, to outright and intentional fraud. Similar allegations of selfishness and false pretence also surrounded Oculus VR after its acquisition, with disillusioned backers claiming Oculus VR was a ‘masquerading golddigger’. For Amanda Palmer, this theme emerged when she described her strategy of using Kickstarter as a ‘loss leader’ in order to make the fans trust her, so she could subsequently crowdfund more reliably on Patreon, a rival fundraising platform. This prompted some concern as to the integrity of this strategy, given many backers had intended their contribution to support both Amanda Palmer and the Kickstarter platform itself. The second antithetical theme suggested backers’ cunning/self-interest mirrored that of the project owner, that is backers were only committed to the project as long as it served their individual selfish interests. This was evident during the delayed delivery of rewards across all three projects, as disillusioned backers became increasingly restless and vocal. This was particularly strong for Oculus VR, as several backers requested their money back as soon as it became obvious development was delayed. It was also evident when some backers who were entitled to free consumer headsets failed to respond in time to take advantage of the offer, at which point several abandoned any admiration for the nature of the gesture. However, it was again Torquing where this theme was most pronounced, as a significant number of backers demanded refunds when delivery of the drones began to look unlikely.

Acclaimed, Suffering Heroes or Maligned Failures? As the projects progressed, outcomes were increasingly evaluated against different actors’ expectations. Outcomes across the three illustrative examples disappointed at least some of the collaborating actors in terms of contextual impact and/or the delivery of rewards. The thesis characterised project owners as suffering heroes who had struggled to overcome obstacles presented by unsupportive actors inside or outside the project. Within this thesis, the first theme linked project owners’ limited contextual impact

236 

K. Conboy et al.

to interference from other businesses active in those markets. Once again, key triggers were Oculus VR’s acquisition by Facebook and Amanda Palmer’s call for other musicians to perform with her without payment. Some backers also referred to Torquing’s loyalty to local manufacturing, though this was infrequent. Sympathetic backers discussed the real-world obstacles presented to smaller ventures, for whom the costs of rapid growth can quickly become unmanageable. Many backers promoting this narrative highlighted the attempts at transparency when Amanda Palmer and Torquing provided financial breakdowns. The second theme was that unreasonable backers had got in the way of the project owners. This featured only mildly for both Amanda Palmer and Oculus VR, for whom some backers and onlookers suggested the need to accommodate backers’ suggestions had created a distraction. However, this narrative featured strongly for Torquing, for which many backers and onlookers felt loud and public discontent from an increasing proportion of backers had contributed to issues throughout the project. The project owners attributed this narrative to the significant jump in scale required to meet the higher-than-expected demand, explaining in an interview with the BBC that ‘We’ve had to go from a small research and development company to volume manufacturing. From expectations of building 800 to 1,000 Zanos to gearing up to building 13, 14 or 15,000.’ This meant the efforts to be responsive became decreasingly realistic, with several backers suggesting they should relieve themselves of those duties in the interests of keeping the project moving. The antithesis characterised project owners as maligned by those who naïvely trusted them. The first antithetical theme suggested project’s limited impact was inevitable given the project owners’ unrealistic views of market conditions and consumer demand. Amanda Palmer was criticised by onlookers for not being a ‘true independent’ but rather a ‘refugee from a major-label system’, suggesting her success was an anomaly rather than an exemplar. Oculus VR was maligned for overstating the potential of VR, as well as underestimating the loss of ideological and community impact resulting from the Facebook acquisition. All the projects were also maligned for eroding the impact of Kickstarter, the value of which was commonly seen as diminished by an underwhelming impact (practical or ideological) across the three projects.

7  When Crowds Play God: A Promethean Perspective… 

237

The second antithetical theme suggested that project owners promised rewards to backers they were ill-equipped to deliver. For Amanda Palmer, this theme focused on perceptions of poor financial management and decision-making. Criticisms of Torquing went deeper, following the failure of the project to deliver key features promised in the original crowdfunding campaign; for example the drones were only capable of staying airborne for several minutes, collided with walls, video quality was questionable and showed no signs of avoiding obstacles during autonomous flight. Lastly, criticisms of Oculus VR questioned whether the company had even intended on maintaining the independence that had attracted many backers to them.

Discussion A number of existing studies have alluded to the potential for crowdfunding to create a strong public image for new businesses (e.g. Ordanini et al., 2011; Gerber & Hui, 2013; Brown et al., 2017). Yet the ability of crowdfunded project owners to leverage this potential into long-term positive outcomes is less clear. One of the reasons for this lack of clarity is the novelty of the phenomenon under study. This lack of clarity is problematic, partly because it makes it challenging to gauge the potential of crowdfunding to create positive image but also because it is not clear which communication practices will prove to be most (or least) effective. This study offers an in-depth exploration of this issue by analysing the narratives around three of crowdfunding’s famous ‘index’ cases (Patton, 2014). The next sections focus on synthesising and analysing findings, as well as identifying the key contributions to research from the study.

Synthesis and Practical Recommendations The synthesis of contrasting perspectives is the essential mechanism by which new ‘truths’ can be not only identified but also realised (De Rond & Bouchikhi, 2004; Carlo et al., 2012; Wijnhoven, 2012). Hegel (1975) described synthesis as a means of transformation, based on the discovery

238 

K. Conboy et al.

of new potentialities and the actualisation of new assemblages of things from existing components. This is intuitively relatable to crowdfunding narratives, where projects are often emerging in dialogue with backers, meaning uncertainty is high and goals and group participation may be dynamic (Gleasure & Morgan, 2018). Under such conditions, constructed narratives typically act as a means of both sense-making and sense-giving for strategic change (Gioia & Chittipeddi, 1991). We therefore present the syntheses as recommendations for practice, constructed to help project owners transition to more appropriate public narratives and less polarisation of commentators. Note we do not intend these recommendations to help with the practicalities of projects, such as production or fundraising. Instead, the focus is on the public narrative as a related but distinct component of a crowdfunding project. Further, these recommendations do not represent a return to the data—the interpretation remains the same—rather, they are built by abstracting an additional level into existing literature to engage with this larger problem of managing a community around a project. Recommendation #1: Project owners should explicitly describe any selfless, positive aspects of their use of a crowdfunding platform, as well as steps to mitigate or eradicate negative impacts. Concerning the dialectic between the philanthropist and misanthropist narratives, it is notable themes were divided within the thesis and antithesis according to their micro or macro focus, leading to some overlap within each narrative. Nonetheless, two distinct opportunities for synthesis are observed from the analysis. The first is the observation that claims of philanthropy/misanthropy were not only made by project owners; they were also made by backers and external commentators. Torquing in particular made no claims of this type, yet the narrative emerged regardless. This suggests project owners may need to be proactive and explicit in managing this tension in the narrative. The second opportunity for synthesis is the observation that positive and negative narratives were interdependent, that is the presence of a positive narrative theme tended to accompany the presence of the corresponding negative theme. The narratives around Oculus VR contained philanthropy/misanthropy-focused discussion at both a product and a practice-level. Amanda Palmer was discussed extensively at a

7  When Crowds Play God: A Promethean Perspective… 

239

practice-­ level (positively and negatively) but not at a product-level. Torquing was discussed extensively at a product-level (positively and negatively) but not at a practice-level. This suggests product owners should address potential negative commentary, as well as the positive, in order to avoid other commentators filling in this gap. Recommendation #2: Project owners should explicitly describe how using a crowdfunding platform fits within the context of their larger growth strategy. Concerning the dialectic between the revolutionary forethinker and recklessly ambitious narratives, once again each of the opposing themes shares some overlap and some contradictions. The departure between revolutionary foresight and reckless ambition was ultimately a function of people’s perceptions whether or not related industries were ripe for widespread disruption. However, both perspectives acknowledged that project owners were presenting at least some useful characteristics or innovations that may have struggled to attract conventional funding. This problem was well-publicised for Amanda Palmer, who had come into conflict with previous record labels regarding the limited size of her listener base. Similar observations are made for Oculus VR and Torquing, for whom viral marketing and the early availability of capital allowed them to surge beyond competitors. The second area of convergence between thesis and antithesis was that once each project owner was funded, those project owners were expected to demonstrate a level of ‘professionalism’ for planning and management activities comparable to traditional-style businesses. This was difficult for all three project owners, for whom funding was predicated on an accelerated development model that made it increasingly difficult to satisfy competing expectations among backers for quality and timeliness. Thus, while crowdfunding lowered the barriers to entry in emerging markets for project owners, it appears those project owners have to adhere to traditional standards to stay competitive over time as those markets grow. In general, backers appeared surprised and frustrated by the diminishing interactions, tipping the tone of the later narratives towards the negative. Recommendation #3: Project owners should explicitly justify the use of crowdfunding platforms as a form of positive disintermediation. Concerning the dialectic between the liberator and thief narratives, the contradictions between the thesis and antithesis centred upon the relative

240 

K. Conboy et al.

value of involving consumers and potential partners at a particular point in a market’s lifecycle. Thus, the disagreement concerned the value of disintermediating the supply chain and consequently the value of potential partners. Yet while the relative value of this trade-off was contested, the nature of the trade-off was not. The advantage of traditional models is potential partners are able to integrate and leverage their specialised knowledge into the early stages of development. This could create supply chain-level synergies and compatibilities with that product/service, which could reduce costs and build sustainable inter-firm relationships. The use of crowdfunding platforms sacrifices these benefits in exchange for larger volumes of diverse, consumer-facing and less-specialised creative input. This increases consumers’ sense of ownership, as development activities represent a shared achievement. It also generates more market feedback at an earlier stage in development than would otherwise be possible. The second set of contrasting themes disagree on the relative value of involving potential partners and consumers during product/service delivery. Traditional models that involve potential partners have the advantage of building products/services within a complementary ecosystem, for example VR-ready games and applications for the Rift, tour venues and promotors for Amanda Palmer. The crowdfunding model assumes consumers can either take enough value from the product/service without such an ecosystem or create it themselves in the future. This allows a leaner product/service to be created and launched into a more collaborative environment. However, it also means the initial uses of a product/ service are limited by comparison. Recommendation #4: Project owners should identify specific periods and tasks suitable for shared responsibility when they use crowdfunding platforms. Concerning the dialectic between the enlightener and empowerer and creator of work narratives, both thesis and antithesis agreed that these projects required more participation from backers and that such participation held potential value for the project owner. The disagreement between opposing themes concerned the extent to which backers actually benefitted from their additional participation in crowdfunding projects. For financing, the antithesis suggested that backers were misled or misguided into assuming high-risk financial responsibilities, while the thesis celebrated their opportunity to influence the types of projects

7  When Crowds Play God: A Promethean Perspective… 

241

undertaken. Whatever the intent, backers who treated their participation as a traditional pre-purchase transaction were disappointed across all three illustrative examples. This suggests backers must have a particular interest in the project in question (beyond simply obtaining rewards) if they are to assume these financial responsibilities in a sustainable fashion. Similar observations were made regarding backers’ willingness to assume additional communication and development responsibilities. For the most part, project owners wanted backers to contribute to tasks that would leverage the scale and reach of the crowd for business development, for example viral marketing and creating contact points to supportive networks. Conversely, backers expressed most excitement around the opportunities for creative input, which makes sense given these activities were traditionally less accessible to them. This created a common pattern where backers would participate in surges, offering large volumes of creative recommendations and spreading the word about the project. Project owners would struggle to integrate these creative suggestions, meaning backers’ enthusiasm and engagement faded, as did their contribution to viral marketing and business development. Hence, backers may be willing to assume communication and development responsibilities, but only periodically during periods of high engagement. Recommendation #5: Crowdfunding projects should explicitly identify new opportunities for shared value as part of their justification for using crowdfunding platforms. Concerning the dialectic between the altruist or self-interested trickster narratives, the source of disagreement between the thesis and antithesis centred upon the willingness of project owners and backers to bear individual losses. The altruistic perspective suggested that each actor viewed these losses as an acceptable concession to achieve broader shared value. The cunning/self-interest perspective implied each actor was pushing to minimise personal losses and maximise personal gain within the boundaries of shared value. Despite these conflicting views on the nature of individual motives and attitudes towards shared value, both perspectives agreed this shared value was key to balancing those motives and attitudes (whether altruistic or self-interested). This means that project owners only make sacrifices as long as the selfish benefits from shared value outweigh the cost of those sacrifices. Once the costs become

242 

K. Conboy et al.

unmanageable, the project owner must protect themselves to prevent the project/company from failing, in which case the shared value is also harmed. In a similar vein, the limited commitment to shared value means backers will also make sacrifices as long as the selfish benefits from shared value outweigh the cost of those sacrifices. Those backers most invested in the ideology behind projects were most patient when the need for compromise arose. This became particularly problematic for Torquing, the least shared value-driven project, as a critical mass of backers turned on the project owners almost immediately once delays were announced. Thus: Recommendation #6: Project owners should explicitly identify high/low risk rewards and priority goals when using crowdfunding platforms. Concerning the dialectic between the acclaimed suffering heroes or maligned failures narratives, the essence of the disagreement between corresponding themes in the thesis and antithesis concerned the allocation of blame. Proponents of crowdfunding typically suggest members of the public possess a heightened social focus that compensates for the loss in practical project-selection competence, while detractors suggest such loss of competence is more problematic. The thesis, purported by proponents of crowdfunding, therefore attributed disappointments to environmental interference around the project, while the antithesis, purported by detractors, attributed them to the project owners themselves. The extent to which either view is accurate is beyond the scope of this study, where the focus is on narrative. However, despite these differences, both perspectives agreed (i) some or all the goals of the projects had not been achieved and (iii) the value placed upon specific goals varied among backers. This was most evident in each project when specific goals were threatened, at which point the heterogeneity of interests and motivations among backers became obvious through the relative scale of their disappointment. For both contextual impact and the delivery of rewards, this suggests a need to make the range of environmental and reward-based goals explicit and identify which goals are priority and which are high/low risk. This is already done to some extent via stretch goals. However, there is no evidence that these stretch goals are viewed with any less expectation of fulfilment among backers. An intuitive explanation for this is that any

7  When Crowds Play God: A Promethean Perspective… 

243

loss of favourable ambiguity might threaten overall funding levels, as each backer is currently free to assume their environmental or development goals are central. Nonetheless, the reputational cost of failing to achieve ambitious but implicitly guaranteed goals is demonstrable for each of the crowdfunding projects selected. This suggests the proposed dynamic may be favourable for all involved, assuming there is still sufficient value in the revised description of goals. Many of these opportunities are already discussed in terms of the social benefits of crowdfunding (c.f. Giudici et  al., 2018, Burtch & Chan, 2019). Yet the need to explicate them as part of project-level narrative is novel. The recommendations thus provide clear guidance based on lucid theoretical paths that link the ideology of crowdfunding to public discourse. Part of this theoretical path is the recognition of power as a force for dialectic divergence. The myth of Prometheus is characterised by a power-struggle between gods and humans, as the former attempt to maintain the status quo in the face of growing dissent from the latter. It appears the same is true of crowdfunding projects, for whom scepticism among established businesses and financial services professionals is common (e.g. Harrison, 2013; Hanks et al., 2014). Previous IS research has established the usefulness of power-based and dialectical perspectives in exploring the impact of IT (e.g. Jasperson et al., 2002; Doolin, 2004). Yet explicit discussion of power, negotiation and competing dialectical perspectives features only rarely in existing crowdfunding research. Notable exceptions include Aitamurto’s (2011) case study of crowdfunded journalism on Spot.us, as well as studies by Scott (2015) and Booth (2015) that analysed fan consensus in crowdfunded movies and graphic novels. This research offers an entry point for further dialectical and power-based crowdfunding research.

Contributions to Research The first and foremost contribution of this study is that it provides a foundation for more deliberate and focused discussion of the positive and negative narratives surrounding crowdfunded project owners. Positive narratives are an important influence on fundraising activities (Thies

244 

K. Conboy et al.

et al., 2016; Anglin et al., 2018; Majumdar & Bose, 2018). Yet the value of a positive narrative extends beyond the ability to raise money. The value of a positive narrative for creating future markets for products and services is one of the key espoused benefits of crowdfunding (Brown et  al., 2017; Zhao et  al., 2019). This has been described as a form of ‘affective economics’, in which backers provide projects with evolving forms of value over time as they deliver on different outcomes (Hills, 2015). The themes and recommendations identified provide, for the first time, a vocabulary with which to study these narratives in a holistic, longitudinal and Promethean manner. It further provides an overarching framework capable of relating studies of specific projects according to the accompanying narratives. Since the Association for Information Systems (AIS) launched its ‘grand vision project’ for an ICT-enabled society, one of the key goals has been to coordinate research activities to amplify their effects (Fedorowicz et al., 2015; Pennarola et al., 2015). Thus, the recommendations presented in this study act as an entry point for researchers looking to consider public perceptions of crowdfunding and the impact of public discourse. This addition complements other ongoing efforts focused on privacy, security, addiction, conflict and harassment (c.f. Lee, 2015) by integrating such public discourse into ongoing discussion of shared value and responsible capitalism. The second contribution is the explication of predictable limitations for crowdfunding as an enabler of positive systemic change. Crowdfunding is often touted as a means of revolutionising markets and changing business trends (Brem et al., 2019; Stevenson et al., 2019). It is also seen as a way to help female entrepreneurs (Greenberg & Mollick, 2017; Pergelova et al., 2019) and drive regional growth (Yu et al., 2017; Kim & Hann, 2019) and promote sustainability (Calic & Mosakowski, 2016). Others have suggested it provides new opportunities to help people on the margins on society, for example because they have new opportunities for self-­ employment (Burtch et  al., 2018) or because of spiralling medical expenses (Burtch & Chan, 2019). Just as the narratives around Prometheus and his gift of fire to humankind were shaped by imperfect human nature and powerful individuals’ desire to maintain the status quo; so the narratives around crowdfunded project owners are limited by (i) the selfish interests of backers and (ii) the negative sense-making story put forward

7  When Crowds Play God: A Promethean Perspective… 

245

by disenfranchised individuals and businesses. This doesn’t mean that project owners cannot achieve a positive narrative, nor that the narratives around different projects are equivalent—clearly some projects benefit from a more positive narrative than others. However, the synthesised conceptualisation of each thesis/antithesis encourages future research to focus on gradients of positivity/negativity and the manner in which narratives are formed. This is also important for the growth of crowdfunding more broadly. Analysis of the millennium development goals and sustainable development goals suggests that progress is rarely binary; rather, improvements occur gradually and unevenly (Sachs & McArthur, 2005; Chen & Ravallion, 2012). Public perceptions will play a key role for crowdfunding and the shared value ideology which it aspires/claims to accommodate. In a similar light, we also highlight the role of shared value and shared responsibility in crowdfunding. These topics receive little direct attention in existing crowdfunding research, yet they appear centrally important to the long-term social impact of the phenomena. The third contribution surrounds the usefulness of myth and archetypes to make sense of narratives for emerging phenomena. Despite perceptions that crowdfunding is important and the future evolution of the paradigm is uncertain, there are varying views on how the platform will develop (e.g. Assenova et al., 2016; Younkin & Kashkooli, 2016; Testa et al., 2019). The same uncertainty is also present for other alternative finance technologies such as initial coin offerings (ICOs) (Block et al., 2020) an invoice trading (Dorfleitner et al., 2017). Myths exist because of repeating patterns in human history. In the case of Prometheus, these patterns were the tendency of society to achieve only mixed benefits from new technologies, even when those technologies possess extraordinary capabilities for good. This is important, as while it may be too early to predict the outcomes for emerging technology-driven initiatives, these mythical narratives offer a sophisticated means to look for ‘rhyming’ phenomena, lending to comparisons and ergo, predictions. In this study, the Promethean myth not only assisted in the analysis of individual narratives, but it also assisted in the identification of commonalities across each narrative studied. Thus, this study demonstrates the value of using myth as a theoretical lens, as well as providing a sophisticated methodological exemplar to inform other IS topics amenable to a myth-based lens.

246 

K. Conboy et al.

Finally, the fourth key contribution concerns the narrative-based Hegelian dialectic performed and the value demonstrated by that approach. Not only did the Hegelian dialectic allow a systematic means of producing prescriptive recommendations, it also provided a means of identifying the essence of disagreement between those narratives. Crowdfunding represents a departure from typical venture funding practices (Younkin & Kashkooli, 2016). Established business practices are rarely subject to the same level of scrutiny as newer alternatives, even where those established practices are dysfunctional (Abrahamson & Fombrun, 1994; Anand et al., 2004). As a result, crowdfunded projects are not only required to manage more uncertainty, but they must also manage the potential for imitation (Cowden & Young, 2020; Lin & Boh, 2020). This adds further weight to the competing narratives around a project, as they become generative both internally and externally. Further, the persuasive element of each narrative meant the inflation of one perspective appeared to result in corresponding inflation elsewhere. This occurred symmetrically around theses/antitheses; for example projects demonstrating greater discussion of revolutionary foresight also demonstrated greater discussion of reckless ambition. It also occurred longitudinally, for example projects where the theses/antitheses were inflated early on tended towards more pronounced and subjective stories as a project progressed. This is important, as the tendency to describe events using a shared vocabulary is key to group membership, meaning movement between groups becomes increasingly difficult the more differently those groups describe events (Blumer, 1986; Ashforth & Mael, 1989). Without movement between groups, competition and discrimination may become embedded (Tajfel & Turner, 1979). Thus, other things being equal, crowdfunding projects suffering from polarised narratives may be subject to increasingly committed and divided populations of onlookers.

Conclusions: Limitations and Further Research There are a number of limitations of this study that merit attention. First, we treat the construction of narrative as separate from the actual performance of projects and the delivery of promised outcomes. This was

7  When Crowds Play God: A Promethean Perspective… 

247

consistent with our research objectives, as it allowed us to theorise about narrative construction in a more focused manner. However, the timeliness and quality of outcomes from a project also predictably impacts its narrative (see Rantakari & Vaara, 2017; Vaara et al., 2016). Thus, future research may wish to differentiate how narratives are constructed differently in successful and unsuccessful projects, by experienced and inexperienced entrepreneurs, and by those with varying domain backgrounds, for example. Second, there is the use of narrative and in particular myth as a lens in scientific research. There is a fundamental divide between mythos—a story or set of stories having a significant meaning for a particular group— and logos—where that significant meaning is based on objective and verifiable evidence. Habermas, one of the key detractors of myth, dismisses it as ‘fuzzy’ and subjective (Habermas, 1984). Such subjectivity is often inescapable, particularly where the Promethean myth is concerned, as its nature requires a reader to decide what is ‘good’. The researchers addressed this by presenting opposing views in a dialectical, thereby reducing the influence of personal bias. The use of myth also requires some conceptual ‘tidying’, as not all ideas translate equally precisely. For example, while Prometheus gave fire to humans to use as they see fit, a project owner who gives financing power to backers via crowdfunding must then work closely with those backers to achieve desired outcomes. The researchers managed this by continuously challenging comparisons during theorising to ensure any such differences would not undermine conclusions. Further, this study only focused on the application of Promethean thinking to crowdfunding and, more specifically, rewards-based crowdfunding campaigns. This paradigm was selected as public narratives around specific projects were most likely to be far-reaching and sustained. Yet public narratives are also important in each of the other paradigms (lending, equity and charity) and it is not clear to what extent the findings of this study can be generalised to those other paradigms or indeed other rewards-based platforms. For example, the idea of ‘thieves’ may be less applicable in charity crowdfunding, where additional parallel investment may also be welcomed from traditional funders and potential partners. Also, while all the project owners studied did use the platform for

248 

K. Conboy et al.

co-production, this may not always be the case, and so some Promethean archetypes such as liberators and thieves may not be as important or even applicable in all other studies. Conversely, some of the Promethean archetypes discussed here may not be as relevant in reward-based campaigns versus those that are not based on rewards. For example, while we still found the notion of philanthropy to be of value in this study, one would logically assume that such an archetype would be more commonly found in campaigns where philanthropy is not just present but is the fundamental goal of the campaign. We thus call for further research to test the validity and utility of the proposed recommendations in future studies across different contexts. Also, while the Promethean lens revealed many tensions in the context of this crowdfunding study, this is only one lens that could have been used, and so there are likely to be other tensions that could be uncovered through other theoretical lens. Future research could draw on more generic frameworks of tension (Huxham & Beech, 2003; Langley & Tsoukas, 2016) or paradox (De Rond & Bouchikhi, 2004; Smith & Lewis, 2011). We also acknowledge that our theoretical application of the Promethean myth focused on the Promethean and human actors, with less attention paid to the ‘gods’ of traditional funders and potential partners. This was consistent with the overarching aims of the study, in which project owners and backers play a prominent role. However, future research could add significant value by providing an alternative framing for narrative construction from the perspective of these traditional funders and potential partners. A particularly interesting yet limiting characteristic of this study is a temporal one, in that it focused on a time-bounded, finite campaign. Many crowdfunding campaigns are shorter, longer, raise funding periodically or sometimes have an infinite life. Further research is required to see if these recommendations hold over these varying time frames. A further argument for temporal-oriented research is that the concept of Promethean thinking and Promethean leaps is likely to be dependent on temporal context. One’s expectations of what could be considered Promethean would depend on the context of work and whether that was over a shortor long-term horizon.

7  When Crowds Play God: A Promethean Perspective… 

249

The dialectic synthesis was conducted in this study by the authors with limited focus on the perceptions of the competing actors constructing the narrative. However, a very interesting study could examine how owners and backers themselves conduct and learn from dialectic synthesis, how they individually and collectively state and reconcile different perspectives and how crowdfunding platform design could aid such a process. It may also explore how those actors struggle with other individuals and groups for control of the narrative, as well as how they reconcile inconsistencies in their own interpretations under such conditions. We also call for research that applies the Promethean lens to other aspects of IS, given that the accuracy or suitability of the recommendations may not be upheld in these contexts. The myth is highly flexible, having been applied in numerous forms across a range of disciplines. Further, and as stated earlier, the Promethean lens is highly relevant to the IS field at the current time given (i) advancements in technological capabilities that enable ‘godlike’ achievements and manipulate the very axes of nature: space, time, energy, matter and life; (ii) an insatiable demand-side technological appetite from a tech-savvy digital native society; (iii) a pervasive prevalence of negativity, malevolence and misanthropy surrounding technology and (iv) a rapidly increasing diffusion of technology across society resulting in increased technological power of the masses.

References Abrahamson, E., & Fombrun, C. J. (1994). Macrocultures: Determinants and consequences. Academy of Management Review, 19(4), 728–755. Aitamurto, T. (2011). The impact of crowdfunding on journalism: Case study of spot. Us, a platform for community-funded reporting. Journalism Practice, 5(4), 429–445. Anand, V., Ashforth, B. E., & Joshi, M. (2004). Business as usual: The acceptance and perpetuation of corruption in organizations. The Academy of Management Executive, 18(2), 39–53. Anglin, A. H., Short, J. C., Drover, W., Stevenson, R. M., McKenny, A. F., & Allison, T. H. (2018). The power of positivity? The influence of positive psy-

250 

K. Conboy et al.

chological capital language on crowdfunding performance. Journal of Business Venturing, 33(4), 470–492. Ashforth, B. E., & Mael, F. (1989). Social identity theory and the organization. Academy of Management Review, 14(1), 20–39. Assenova, V., Best, J., Cagney, M., Ellenoff, D., Karas, K., Moon, J., et  al. (2016). The present and future of crowdfunding. California Management Review, 58(2), 125–135. Barnett, C. (2015). Trends show crowdfunding to surpass VC in 2016. Forbes Online. Retrieved April 4, 2016, from http://www.forbes.com/sites/chancebarnett/2015/06/09/trends-­show-­crowdfunding-­to-­surpass-­vc-­in-­2016/ #2dae5eca444b. Barton, D. (2011). Capitalism for the long term. Harvard Business Review, 89(3), 84–91. Beaulieu, T., Sarker, S., & Sarker, S. (2015). A conceptual framework for understanding crowdfunding. Communications of the Association for Information Systems, 37(1), 1–31. Beier, M., Früh, S., & Jäger, C. (2019). Reward-based crowdfunding as a marketing tool for established SMES: A multi case study. Available at SSRN 3338084. Belleflamme, P., Lambert, T., & Schwienbacher, A. (2014). Crowdfunding: Tapping the right crowd. Journal of Business Venturing, 29(5), 585–609. Benbasat, I., & Zmud, R. W. (2003). The identity crisis within the IS discipline: Defining and communicating the discipline’s core properties. MIS Quarterly, 27(2), 183–194. Benson, J.  K. (1973). The analysis of bureaucratic-professional conflict: Functional versus dialectical approaches. Sociological Quarterly, 14(3), 376–394. Block, J. H., Groh, A., Hornuf, L., Vanacker, T., & Vismara, S. (2020). The entrepreneurial finance markets of the future: A comparison of crowdfunding and initial coin offerings. Small Business Economics, 57, 865. https://doi. org/10.1007/s11187-­020-­00330-­2 Blumer, H. (1986). Symbolic interactionism: Perspective and method. University of California Press. Booth, P. (2015). Crowdfunding: A Spimatic application of digital fandom. New Media & Society, 17(2), 149–166. Bradford, C.  S. (2012). The new federal crowdfunding exemption: Promise unfulfilled. Securities Regulation Law Journal, 40(3), 195–200. Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77–101.

7  When Crowds Play God: A Promethean Perspective… 

251

Brem, A., Bilgram, V., & Marchuk, A. (2019). How crowdfunding platforms change the nature of user innovation–from problem solving to entrepreneurship. Technological Forecasting and Social Change, 144, 348–360. Brown, T. E., Boon, E., & Pitt, L. F. (2017). Seeking funding in order to sell: Crowdfunding as a marketing tool. Business Horizons, 60(2), 189–195. Burtch, G., Carnahan, S., & Greenwood, B.  N. (2018). Can you gig it? An empirical examination of the gig economy and entrepreneurial activity. Management Science, 64(12), 5497–5520. Burtch, G., & Chan, J. (2019). Investigating the relationship between medical crowdfunding and personal bankruptcy in the United States: Evidence of a digital divide. MIS Quarterly, 43(1), 237–262. Calhoun, C. (2013). Occupy wall street in perspective. The British Journal of Sociology, 64(1), 26–38. Calic, G., & Mosakowski, E. (2016). Kicking off social entrepreneurship: How a sustainability orientation influences crowdfunding success. Journal of Management Studies, 53(5), 738–767. Carlo, J. L., Lyytinen, K., & Boland, R. J., Jr. (2012). Dialectics of collective minding: Contradictory appropriations of information technology in a high-­ risk project. Management Information Systems Quarterly, 36(4), 1081–1108. Chen, S., & Ravallion, M. (2012). An update to the World Bank’s estimates of consumption poverty in the developing world. World Bank Report, Washington, DC, USA. Conboy, K. (2019). Being promethean. European Journal of Information Systems, 28(2), 119–125. Conboy, K. (2021). Enabling promethean leaps: An examination of storytelling techniques in information systems development. Communications of the AIS, 48, 333. Cowden, B. J., & Young, S. L. (2020). The copycat conundrum: The double-­ edged sword of crowdfunding. Business Horizons, 63(4), 541–551. De Rond, M., & Bouchikhi, H. (2004). On the dialectics of strategic alliances. Organization Science, 15(1), 56–69. DeLuca, K. M., Lawson, S., & Sun, Y. (2012). Occupy Wall street on the public screens of social media: The many framings of the birth of a protest movement. Communication, Culture & Critique, 5(4), 483–509. DiMaggio, P., Hargittai, E., Neuman, W. R., & Robinson, J. P. (2001). Social implications of the internet. Annual Review of Sociology, 27(2001), 307–336. Doolin, B. (2004). Power and resistance in the implementation of a medical management information system. Information Systems Journal, 14(4), 343–362.

252 

K. Conboy et al.

Dorfleitner, G., Rad, J., & Weber, M. (2017). Pricing in the online invoice trading market: First empirical evidence. Economics Letters, 161, 56–61. Dougherty, C. (2006). Prometheus. Routledge Press. Doyle, R., & Conboy, K. (2020). The role of IS in the covid-19 pandemic: A liquid-modern perspective. International Journal of Information Management., 55, 102184. https://doi.org/10.1016/j.ijinfomgt.2020.102184 Fedorowicz, J., Agarwal, R., Lee, G., Lee, J. K., Watson, R., & Zhang, P. (2015). The AIS grand vision project: What, why, and how. In International Conference on Information Systems, Puerto Rico. Frydrych, D., Bock, A. J., Kinder, T., & Koeck, B. (2014). Exploring entrepreneurial legitimacy in reward-based crowdfunding. Venture Capital, 16(3), 247–269. Gerber, E. M., & Hui, J. (2013). Crowdfunding: Motivations and deterrents for participation. ACM Transactions on Computer-Human Interaction (TOCHI), 20(6), 34. Gerber, E. M., Hui, J. S., & Kuo, P. Y. (2012). Crowdfunding: Why People Are Motivated To Post And Fund Projects On Crowdfunding Platforms. In Proceedings of the International Workshop on Design, Influence, and Social Technologies: Techniques, Impacts and Ethics, Vol 2, p. 11. Gioia, D. A., & Chittipeddi, K. (1991). Sensemaking and sensegiving in strategic change initiation. Strategic Management Journal, 12(6), 433–448. Giudici, G., Guerini, M., & Rossi-Lamastra, C. (2018). Reward-based crowdfunding of entrepreneurial projects: The effect of local altruism and localized social capital on proponents’ success. Small Business Economics, 50(2), 307–324. Gleasure, R., Conboy, K., & Morgan, L. (2019). Talking up a storm: How backers use public discourse to exert control in crowdfunded systems development projects. Information Systems Research, 30(2), 447–465. Gleasure, R., & Morgan, L. (2018). The pastoral crowd: Exploring self-hosted crowdfunding using activity theory and social capital. Information Systems Journal, 28(3), 489–515. Gleasure, R., O’Reilly, P., & Cahalane, M. (2017). Inclusive technologies, selective traditions: A socio-material case study of crowdfunded book publishing. Journal of Information Technology, 32, 326. Greenberg, D., Clair, J., & Maclean, T. (2007). Enacting the role of the management professor: Lessons from Athena, Prometheus and Aclepius. Academy of Management Learning & Education, 6(4), 439–457. Greenberg, J., & Mollick, E. (2017). Activist choice homophily and the crowdfunding of female founders. Administrative Science Quarterly, 62(2), 341–374.

7  When Crowds Play God: A Promethean Perspective… 

253

Grene, D. (1940). Prometheus bound. Classical Philology, 35(1), 22–38. Habermas, J. (1984). Reason and the Rationalization of Society, Volume 1 of The Theory of Communicative Action, English translation by Thomas McCarthy. Boston: Beacon Press. Hanks, S., Romano, G., & Tonelli, E. (2014). Madness of crowds or regulatory preconception?: The weak Foundation of Financial Crowdfunding Regulation in the US and Italy. European Company Law, 11(5), 237–252. Harrison, R. (2013). Crowdfunding and the revitalisation of the early stage risk capital market: Catalyst or chimera? Venture Capital, 15(4), 283–287. Hatch, M., Kostera, M., & Kozminski, A. (2005). The three faces of leadership: Manager, artist, priest. Blackwell Publishing. Hazen, T. L. (2011). Crowdfunding or fraudfunding-social networks and the securities laws-why the specially tailored exemption must be conditioned on meaningful disclosure. North Carolina Law Review, 90, 1735–1770. Hegel, G.  W. F. (1807) Phenomenology of spirit. (A.  V. Miller, Trans.). Clarendon Press. Hegel, G. W. F (1975) Aesthetics. Lectures on fine art (T. M. Knox, Trans. vol. 2). Clarendon Press. Hills, M. (2015). Veronica Mars, fandom, and the ‘affective economics’ of crowdfunding poachers. New Media & Society, 17(2), 183–197. Howard, P.  N. (2010). The digital origins of dictatorship and democracy: Information technology and political Islam. Oxford University Press. Huber, T. L., Kude, T., & Dibbern, J. (2017). Governance practices in platform ecosystems: Navigating tensions between co-created value and governance costs. Information Systems Research (forthcoming), Hui, J. S., Greenberg, M. D., & Gerber, E. M. (2014, February). Understanding the role of community in crowdfunding work. In 17th ACM Conference on Computer Supported Cooperative Work & Social Computing, Baltimore, USA. Huxham, C., & Beech, N. (2003). Contrary prescriptions: Recognizing good practice tensions in management. Organization Studies, 24(1), 69–93. Jacobi, J. (1974). Complex/archetype/symbol. Princeton University Press. Jasperson, J.  S., Carte, T.  A., Saunders, C.  S., Butler, B.  S., Croes, H.  J., & Zheng, W. (2002). Review: Power and information technology research: A metatriangulation review. MIS Quarterly, 26(4), 397–459. Kim, J., Cho, D., & Lee, B. (2016). The mind behind crowdfunding: An empirical study of speech emotion in fundraising success. In International Conference on Information Systems, Dublin.

254 

K. Conboy et al.

Kim, K., & Hann, I.  H. (2019). Crowdfunding and the democratization of access to capital—An illusion? Evidence from housing prices. Information Systems Research, 30(1), 276–290. Kim, Y., Shaw, A., Zhang, H., & Gerber, E. (2017). Understanding trust amid delays in crowdfunding. In Proceedings of the 2017 ACM Conference on Computer Supported Cooperative Work and Social Computing (pp. 1982–1996). Kunz, M.  M., Bretschneider, U., & Leimeister, J.  M. (2016). Woohoo TinkerBots! The marketing effect of Crowdfunding. In Proceedings of the 37th International Conference on Information Systems (ICIS), Dublin, Ireland. Langley, A., & Tsoukas, H. (2016). The SAGE handbook of process organization studies. Sage. Lee, J. K. (2015). Guest editorial: Research framework for AIS grand vision of the bright ICT initiative. MIS Quarterly, 39(2), iii–xii. Levina, N., & Vaast, E. (2015). Leveraging archival data from online communities for grounded process theorizing. In K.  D. Elsbach & R.  M. Kramer (Eds.), Handbook of qualitative organizational research: Innovative pathways and methods (pp. 215–224). Routledge. Lewis, M. W. (2000). Exploring paradox: Toward a more comprehensive guide. Academy of Management Review, 25(4), 760–776. Li, Y., Zhang, Z., Wang, R., & Chen, Y. (2019). Consumer purchase intention toward crowdfunding products/services: A cost–benefit perspective. Sustainability, 11(13), 3579. Lin, Y., & Boh, W.  F. (2020). How different are Crowdfunders? Examining archetypes of Crowdfunders. Journal of the Association for Information Science and Technology, 71, 1357. https://doi.org/10.1002/asi.24332. forthcoming. Liu, J., Yang, L., Wang, Z., & Hahn, J. (2015). Winner takes all? The “blockbuster effect” In crowdfunding platforms. In International Conference for Information Systems. Lloyd-Jones, H. (2003). Zeus, Prometheus, and Greek ethics. Harvard Studies in Classical Philology, 101, 49–72. Maharawal, M. M. (2013). Occupy Wall street and a radical politics of inclusion. The Sociological Quarterly, 54(2), 177–181. Majumdar, A., & Bose, I. (2018). My words for your pizza: An analysis of persuasive narratives in online crowdfunding. Information & Management, 55(6), 781–794. Marshall, A., & Ojiako, U. (2010). From the myth of Prometheus to strategic resilience: Two cognitive paradigms linking risk and innovation. Prometheus, 28(4), 343–360.

7  When Crowds Play God: A Promethean Perspective… 

255

Miles, M. B., & Huberman, A. M. (1994). Qualitative data analysis: An expanded sourcebook (2nd ed.). Sage Publications, Inc. Mitroff, I. (1983). Archetypal social systems analysis: On the deeper structure of human systems. Academy of Management Review., 8(2), 387–397. Mollick, E. (2014). The dynamics of crowdfunding: An exploratory study. Journal of Business Venturing, 29(1), 1–16. Mollick, E., & Robb, A. (2016). Democratizing innovation and capital access. California Management Review, 58(2), 72–87. Morford, M., Lanardon, R., & Sham, M. (2015). Classical mythology. Oxford University Press. Morse, A. (2015). Peer-to-peer crowdfunding: Information and the potential for disruption in consumer lending. Annual Review of Financial Economics, 7, 463–482. Myers, M. D. (1995). Dialectical hermeneutics: A theoretical framework for the implementation of information systems. Information Systems Journal, 5(1), 51–70. Olsson, S. (2000). Acknowledging the female archetype: Women managers’ narratives of gender. Women in Management Review, 15(5/6), 296–302. Ordanini, A., Miceli, L., Pizzetti, M., & Parasuraman, A. (2011). Crowd-­ funding: Transforming customers into investors through innovative service platforms. Journal of Service Management, 22(4), 443–470. Patton, M.  Q. (2014). Qualitative research & evaluation methods: Integrating theory and practice. Sage Publications. Pennarola, F., Lee, J. K., Krcmar, H., Sol, H. G., & Bjorn-Anderson, N. (2015). AIS grand vision project for bright ICT. In European Conference on Information Systems, Münster, Germany. Pergelova, A., Manolova, T., Simeonova-Ganeva, R., & Yordanova, D. (2019). Democratizing entrepreneurship? Digital technologies and the internationalization of female-led SMEs. Journal of Small Business Management, 57(1), 14–39. Pippin, R. (2000). Hegel’s practical philosophy: Rational agency as ethical life. Cambridge University Press. Porter, M. E., & Kramer, M. R. (2011). Creating shared value. Harvard Business Review, 89(1/2), 62–77. Raggio, O. (1958). The myth of Prometheus: Its survival and metamorphoses up to the eighteenth century. : Warburg and Courtauld Institutes 21(1/2):44–62.

256 

K. Conboy et al.

Ragin, C. C. (1992). “Casing” and the process of social inquiry. In C. C. Ragin & H.  S. Becker (Eds.), What is a case? Exploring the foundations of social inquiry (pp. 217–226). Cambridge University Press. Rantakari, A., & Vaara, E. (2017). Narratives and processuality. In A. Langley & H.  Tsoukas (Eds.), The Sage handbook of process organization studies (pp. 271–285). Sage. Sachs, J. D., & McArthur, J. W. (2005). The millennium project: A plan for meeting the millennium development goals. The Lancet, 365(9456), 347–353. Schad, J., Lewis, M. W., Raisch, S., & Smith, W. K. (2016). Paradox research in management science: Looking Back to move forward. Academy of Management Annals, 10(1), 5–64. Schedlitzki, D., Jarvis, C., & MacInnes, J. (2015). Leadership development: A place for storytelling and Greek mythology? Management Learning, 46(4), 412–426. Scott, S. (2015). The moral economy of crowdfunding and the transformative capacity of fan-ancing. New Media & Society, 17(2), 167–182. Sinnerbrink, R. (2007). Understanding Hegelianism. Acumen Publishing. Smith, W. K., & Lewis, M. W. (2011). Toward a theory of paradox: A dynamic equilibrium model of organizing. Academy of Management Review, 36(2), 381–403. Stacey, R. (2010). Complexity and organizational reality: Uncertainty and the need to rethink management after the collapse of investment capitalism. Routledge. Stevenson, R. M., Kuratko, D. F., & Eutsler, J. (2019). Unleashing main street entrepreneurship: Crowdfunding, venture capital, and the democratization of new venture investments. Small Business Economics, 52(2), 375–393. Szmigin, I., & Rutherford, R. (2013). Shared value and the impartial spectator test. Journal of Business Ethics, 114(1), 171–182. Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. The Social Psychology of Intergroup Relations, 33(47), 74. Testa, S., Nielsen, K. R., Bogers, M., & Cincotti, S. (2019). The role of crowdfunding in moving towards a sustainable society. Technological Forecasting and Social Change, 141, 66–73. Thies, F., Wessel, M., & Benlian, A. (2016). Effects of social interaction dynamics on platforms. Journal of Management Information Systems, 33(3), 843–873. Trice, H. M., & Beyer, J. M. (1984). Studying organizational cultures through rites and ceremonials. Academy of Management Review, 9(4), 653–669. Urquhart, C., & Vaast, E. (2012). Building social media theory from case studies: A new frontier for IS research. In International Conference on Information Systems, Orlando, USA.

7  When Crowds Play God: A Promethean Perspective… 

257

Vaara, E., Sonenshein, S., & Boje, D. (2016). Narratives as sources of stability and change in organizations: Approaches and directions for future research. Academy of Management Annals, 10(1), 495–560. Vasileiadou, E., Huijben, J. C., & Raven, R. P. J. M. (2016). Three is a crowd? Exploring the potential of crowdfunding for renewable energy in The Netherlands. Journal of Cleaner Production, 128, 142–155. Warschauer, M. (2004). Technology and social inclusion: Rethinking the digital divide. MIT Press. Wheat, R. E., Wang, Y., Byrnes, J. E., & Ranganathan, J. (2013). Raising money for scientific research through crowdfunding. Trends in Ecology & Evolution, 28(2), 71–72. Gony, Oxford: Oxford University Press. Whelan, E., Teigland, R., Vaast, E., & Butler, B. (2016). Expanding the horizons of digital social networks: Mixing big trace datasets with qualitative approaches. Information and Organization, 26(1), 1–12. Wijnhoven, F. (2012). The Hegelian inquiring system and a critical triangulation tool for the internet information slave: A design science study. Journal of the American Society for Information Science and Technology, 63(6), 1168–1182. Wilson, M., Robson, K., & Botha, E. (2017). Crowdsourcing in a time of empowered stakeholders: Lessons from crowdsourcing campaigns. Business Horizons, 60(2), 247–253. Witzel, E. (2012). The origins of the World’s mythologies. Oxford University Press. Wortham, J. (2012, September 17). Success of crowdfunding puts pressure on entrepreneurs. New York Times. http://www.nytimes.com/2012/09/18/technology/success-­of-­crowdfunding-­puts-­pressure-­on-­entrepreneurs.html Younkin, P., & Kashkooli, K. (2016). What problems does crowdfunding solve? California Management Review, 58(2), 20–43. Yu, S., Johnson, S., Lai, C., Cricelli, A., & Fleming, L. (2017). Crowdfunding and regional entrepreneurial investment: An application of the CrowdBerkeley database. Research Policy, 46(10), 1723–1737. Zhao, J., Li, Y., Ding, Y., & Liu, C. (2019). The value of leading customers in a crowdfunding-based marketing pattern. PLoS One, 14(4), e0215323. Zheng, H., Xu, B., Zhang, M., & Wang, T. (2018). Sponsor’s cocreation and psychological ownership in reward-based crowdfunding. Information Systems Journal, 28(6), 1213–1238. Ziolkowski, T. (2000). The sin of knowledge: Ancient themes and modern variations. Princeton University Press.

8 Routinization of Digital Transformation of Work: A Discursive Practice Orientation Toward a Native IS Theory Daniel N. Treku , Joseph Manga and Emmanuel W. Ayaburi

,

Introduction Ever-complex digitization and heightened interest in exploring new information technologies (IT) have called for more dynamic ways to understand the nature of work, particularly if scholars and organizational strategists want to get a handle on the future of work (Morton et  al., 2020). Digital transformation, which embodies IT innovation and D. N. Treku (*) Worcester Polytechnic Institute, Worcester, MA, USA e-mail: [email protected] J. Manga Abilene Christian University, Abilene, TX, USA e-mail: [email protected] E. W. Ayaburi Monte Ahuja College of Business, Cleveland State University, Cleveland, OH, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. P. Willcocks et al. (eds.), Advancing Information Systems Theories, Volume II, Technology, Work and Globalization, https://doi.org/10.1007/978-3-031-38719-7_8

259

260 

D. Treku et al.

digitization, has become a central tenet of IS discipline discourses (Chanias et al., 2019; Rowe, 2018; Vial, 2019) and IS conferences (e.g., the annual International Conference in Information Systems—ICIS). It has been advanced as a lens for management information systems (MIS) teaching and research to balance rigor and practice appropriately, and to delineate the IS field as a major disciplinary area (Fichman et al., 2014; Yoo et al., 2010). Indeed, several IS special issue calls, such as Organization and Management’s issue on Data Governance and Digital Innovation and the European Journal of Information Systems’ issue on Managing and Sustaining Digital Transformations, highlight the relevance of digital transformation with emerging technologies (Davidson et al., 2021). Digital transformation also involves muddling through the generativity of emerging technologies (Baiyere et  al., 2020; Legner et  al., 2017), by which ostensive and performative routines (Rossi et al., 2020) are produced to achieve meaningful organizational work. Therefore, digital transformation of work (DTW) weaves past, present, and future IT advancements, rallies centralized and legacy IT systems with decentralized technologies (Bhatti et al., 2021), and engenders discipline- and industry-focused technologies (e.g., Fintech) (Alt et al., 2018; Boratyńska, 2019). As such, theorizing digital transformation, particularly in work redesign, holds much promise for building native IS theories (Hassan & Willcocks, 2021), increasing the legitimacy of the IS discipline (Lyytinen & King, 2004; Orlikowski & Iacono, 2001), and communicating digital transformation as a core property of the IS discipline (Benbasat & Zmud, 2003). Digital transformation promises more than a fashion wave in IS research (Baskerville & Myers, 2009), and enduring concepts, frameworks, models, or theories could be advanced from a focus on digital transformation of work. For many years, the concept of routinizations has been at the center of organizational and economic transformation discourses (Becker, 2004). Routinization discourses—whether viewed as embodying structure and agency (Barley, 1986; Feldman, 2004; Feldman & Pentland, 2003) or otherwise—have been at the heart of past and contemporary IS concepts such as structuration, socio-technical conceptions, socio-materiality, and IS affordances, and have informed various paradigms and epistemes. Succinctly, routinization presents a unique context to understand the

8  Routinization of Digital Transformation of Work: A Discursive… 

261

digital transformation of work and develop a native IS theory for the following reasons. First, the concept of organizational routines and, more recently, “technology as a routine capability” (Swanson, 2019) is a fundamental and multidisciplinary notion for understanding transformative work practices (see Becker, 2004, for an interdisciplinary review of ‘routines’ as a key organizational capability). Second, it is anchored within the IS discipline due to the use of IT for work redesigns which are critical for continuous performance outcomes (Cohen & Bacdayan, 1994; Hoffer Gittell, 2002). Thus, this routinization commonality allows for a more efficient comparison or contrast of disciplinary work practices as the discussion leads to properly situating a native IS model of the digital transformation of work (DTW). In this light, an enduring analogy of DTW routinization is helpful in developing a native IS theory. This study aims to analogize DTW routinization, in the best way possible, from related literature and examples. As such the chapter illustrates analogy, poses questions, and makes statements as products of theorizing (see Table 2.1 in Chap. 2). In particular, it focuses on controlled genetic mutation as an analogy for routinization of DTW. As a transformative phenomenon, routinizing digital transformation is analogous to genetic mutation as new variants (emergent) of technologies are innovated to alter how work is done. Generally, work operations or systems (Alter, 2008) may be redesigned using emerging technologies to fit specific job and system requirements to improve productivity and experience (Hackman, 1980). This is achieved by altering core job dimensions of work, that is, the DNA of how work is done, to influence employees’ psychological states and individual attributes (Hackman & Oldham, 1976). For example, redesigning operating theaters over time to keep infections low in hospitals has changed patient-health workers’ interactions. Such interactions lead to high employee motivation, performance, job satisfaction, and less employee absenteeism and turnover (Hackman & Oldham, 1976), notably when redesigned work becomes accepted routines in the organization. Therefore, digital transformation of work, which is ever evolving in the redesign of work practices using digital innovations, with an overall expectation of sustained pattern, that is, routinization (irrespective of the myriad of trials and experiments involved),

262 

D. Treku et al.

can be likened to a controlled genetic mutation. And this process can lead to meaningful work outcomes and sound investments by organizational decision-makers. The chapter proceeds as follows. The next sections present IS-disciplined imaginations and discourses (Weick, 1989) in explaining the routinization of DTW following “the process of information systems theorizing as a discursive practice” (see Chap. 2; also Hassan et al., 2019). We then discuss why this discursive practice is critical to the evolution and survivability of the IS field and arrive at some conclusions. In these respects, the chapter leans toward developing a native IS theory, namely, DTW routinization.

F oundational Theorizing: Practices of Routinization We approach the study of DTW routinization as “an imperfect copy of the phenomenon of interest consisting of positive and neutral analogies” (Hassan et  al., 2022, p.  5; Hesse, 1966). Drawing on analogies to the routinization of DTW (see Fig. 8.1) will afford the simplification of the abstractions that each concept conveys. First, we explore the foundational practices of routines and then we operationalize DTW to elicit dimensions for inscribing these analogies. “Organizational routines are at the foundation of any work process that involves coordination among multiple actors” (Pentland & Feldman, 2008, p. 236). Routines are repetitive, recognizable, interdependent, and patterned actions performed by multiple actors across time and space that can be adaptive rather than seen as only static and unchanging objects (Feldman & Pentland, 2003, 2005). This definition challenges routinization ontologies that posit routines as structures that create inertia, inflexibility, and mindlessness, which are generally represented by three static metaphors: abstractions of programs, habits, and genes (Feldman & Pentland, 2003). Such a static conception of routines embodies technology determinism (Pentland & Feldman, 2008), which, in part, sees changes to organizational routines as responses to external

External turbulence

Performative routines

Feedback Loop: Implications of power, ideology and institution that directly affect technology artifacts beyond design and use such as with intelligent systems, emergent IT, and portfolio or infrastructural technologies

Ostensive routines

Within internal variation turbulence: mutually constitutive routine duality creating more IT malleability and informing routine capability:

Variation (consists of routine dynamics) Power

Ideology

Institution

Variation effects on the implications of power, ideology, and institution

Tangents of power, ideology and institutions implications that may later affect design, implementation, and use of malleable artifacts in the “sphere of worlds” (i.e., indirect effects of the implications of power, ideology, and institution)

External and Internal turbulences

Focal artifact: Inherent dynamic capability or core properties of malleable IT (design, implementation, and use of malleable artifacts) – technology as routine capability

External and Internal turbulences

Fig. 8.1  Routinization of digital transformation of work view

External turbulence

Controlled genetic mutation of focal IT via design, execution, diffusion, and shift: Specific-task-mutated property Generic-task-mutated property The odd problem of external turbulence such as COVID-19 (not frequent but hugely affects work redesign and workspaces)

8  Routinization of Digital Transformation of Work: A Discursive… 

263

264 

D. Treku et al.

pressures and not the result of inherent capabilities of the routinized processes, human actors, and associated tools to adapt (Feldman & Pentland, 2003). In addition, these conceptions result in designing artifacts for work processes as a “simple matter of creating new checklists, rules, procedures, and software” to determine patterns of action, creating the disconnect between work goals and results (Pentland & Feldman, 2008). In their routines as a duality—theory of stability and change in organizational routines—Feldman and Pentland (2003) harmonize the two ontologies of routines via a mutual relationship between ostensive and performative aspects: where ostensive routines constrain and enable performative (or fluid) routines while performative routines create and recreate more ostensive (or cemented) routines (Pentland & Feldman, 2008; Rossi et al., 2020). The ostensive aspect is the “abstract, generalized idea of the routine,” and the performative aspect consists of “specific actions, by specific people, in specific places and times” (Pentland & Feldman, 2008). In this vein, the performative routines align with the resultant actions of actors while their goals align with the inscribed expectations in the patterns of work a designed or adopted artifact enables. Owing to the problem of disconnect between actors’ goals and the results of work, “interaction design” frameworks have been mooted (Cooper et al., 2007). These frameworks only focus on human-technology artifacts’ engagement to the neglect of how organizational actors work around artifacts’ ostensive constraints to attain acceptable work results different from what static or inscribed routines could afford. Through the lens of enterprise resource planning systems (ERP) artifacts, Rossi et al. (2020) show that when performing routines in a digital workplace, users utilize different IT modules with varying degrees of malleability to balance attainment of fluid routines and simultaneously generate more stable or cemented routines. Rossi et al. (2020) addressed Feldman and Pentland’s (2003, p.114) expectation of future studies to explain “why routines sometimes display great inertia and sometimes do not” and to provide explanatory factors that are “at the level of routine rather than the level of the organization.” Open-ended IT modules allow for performative or fluid routines and closed infrastructural IT modules for carrying out cemented routines and inscribing new ostensive routines (Rossi et al., 2020). The level of fluidity and stability of routines to attain local goals is therefore enacted on a

8  Routinization of Digital Transformation of Work: A Discursive… 

265

digital transformative continuum. In this chapter, we interchangeably refer to this enactment phenomenon as digital or IT malleability or digital turbulence. Consequently, we emphasize that digital turbulence or IT malleability partly results from the need to address the conflict between technology modules and the goals of organizational work 1 (Robey, 1987). Each end of the continuum is capable of generating and inscribing performative and ostensive routines. The current study gives a rich understanding of the relationship between ostensive and performative in the routine dynamic as rigid routines inscribed in technology conflict with the routines needed to accomplish operational goals out of the need for endogenous change—as described in Feldman and Pentland (2003). The study allows for the theoretical shift on the digital transformation strategy, as far as routines are concerned, by focusing on the malleability of the digital modules (on an open-ended to closed IT modules continuum) which actors engage in to get work done via mutually constitutive routinization process (Rossi et al., 2020). The preceding discussion elicits questions on what happens to routinization processes and digital work transformation when external turbulence—for example, the COVID pandemic—demands work redesigns. Thus, how can we explain digital malleability and routine dynamics when it is not a result of conflict between local goals and inscribed routines on prevailing IT infrastructure? What will the explanation be when such external turbulence presents the need for new global process templates for redesigning or recrafting digital work and workspaces? How can we explain digital malleability and routinization processes when local work conflicts and external turbulences are both prevalent? Goh et  al.’s (2011) process model of routinization advances a co-evolutionary adaptive process of health information technology (HIT) implementation, which concurs with the flexible, agentic routinization view. Such adaptive routinized frameworks de-emphasize the important role of ostensive routines in the co-­evolutionary processes of performative routines. Moreover, as a discursive endeavor for creating IS discipline-specific products of theorizing  Except for referenced concepts, most italicized texts in this chapter are ‘statements’ and ‘questions’ that form part of the products of theorizing (Hassan et al., 2022) and help elicit further enriching discussions—see Table 2.1 in Chap. 2. 1

266 

D. Treku et al.

(Hassan et al., 2022), the scope of such contextual framing is challenging considering the “implementation window.” The implementation context embodies a process of ensuring that designed artifacts for work reflect the goals of the focal organization and the expected actions of users. Thus, this window represents one aspect of the IT adoption continuum. It is not clear what becomes of the efficacy of such an explanatory model for adaptive routines post-artifact implementation. Further, when adaptive artifacts are designed and implemented, what becomes of the performative routines for patterned actions that did not foresee constraints of the adapted, yet inscribed, routines during implementation? Hassan et al.’s (2019) exemplary “diffusion of innovation (DOI)” (Rogers, 1995) analogy reinforces the importance of the discipline’s (IS) discursive formation as a starting point in the theorizing processes. DOI “is a more comprehensive model that includes pre-adoption stages, the innovation-decision process, continued adoption and discontinuance” (Hassan et al., 2019, p. 204). In problematizing routinization, we consider how best to craft an analogy of routinization of DTW given that internal and external pressures and conflict between implemented artifacts and future performative actions may stimulate changes to ostensive routines, albeit not exclusively. Technological changes and economic decisions produce affordances that generate organizational, institutional, and contextual changes (Orlikowski & Robey, 1991; Zammuto et al., 2007). Bailey and Barley (2020), arguing that new technology implications can no longer be viewed as situated, contextual, and emergent, provide a unified approach to studying workplace technology. They outline four issues that align with Robey and Boudreau’s (1999) logic of opposition. These issues, namely, variation, power, ideology, and institutions, are relevant to our discussions here and critical to understanding the digital transformation of work. Scholars (e.g., Bygballe et al., 2021; Feldman et al., 2016) have posited the concept of “routine dynamics” to explain how best the different underpinnings of routines, artifact designs, and work transformation can be conflated for meaningful scholarly work on routines and organizational practices. Routine dynamics, to a large extent, addresses the issue of variation, which is a premise of the IS discipline (Mingers & Willcocks, 2004); “that the implications of a technology vary significantly by its

8  Routinization of Digital Transformation of Work: A Discursive… 

267

context of use,” although the problem of variation obscurity 2 remains (Bailey & Barley, 2020, p. 2). Thus, to what extent do they affect or are they affected by the implications of power, ideology, and institutions in the face of digital turbulence—that is, conceptions of digital malleability in routinization, the design and use of intelligent systems—and other internal and external pressures for effective digital work redesign and transformation? Bailey and Barley’s unified approach which calls for scholars to look beyond documentation of how the use of technologies shapes work and work practices, therefore, presents a promising direction in this regard. The issue of power implication is encapsulated in technological politics (Jasperson et al., 2002; Markus, 1983) in that: Regardless of the real reasons that groups wanted to adopt a new technology, they could only justify their case by appealing to greater efficiency and lower labor costs. In other words, the design and spread of technologies, as well as the technologies’ expected outcomes, were shaped by the interests, goals, and perspectives of those who had the authority to authorize the purchase. (Bailey & Barley, 2020, p. 2)

In addition, the technology oligarchy and powerful stakeholders affect the purpose and emergence of certain technologies and how they get adopted (Bailey & Barley, 2020). We conceive that the power issue is thus internal and external in nature (i.e., generates internal and external turbulences). Therefore, how do internal power play and external technological power influences impact our study of organizational routines? On the issue of ideology, specific technological communities approach design with a set culture and a design ideology different from what would-be adopters may have (Bailey & Barley, 2020). For instance, the ERP design community may be driven by ideologies different from the artificial intelligence (AI) design community. Forsythe (2001) pointed out that the AI design community typifies an ideology that favors the  Referring to how skills, tasks, and work practices among actors vary across organizations with the same technology under consideration or the absence of the how and the why a technology implemented a technology. An example of the first variation type is how “an analysis that suggests that intelligent technologies can automate a particular job may be true for jobholders in some organizations and locations, but not to others” (Bailey & Barley, 2020, p. 2). 2

268 

D. Treku et al.

technical over the social. Therefore, the idea of work redesign is contingent on applicable technologies and associated community ideologies and the dominant ideology in an organizational technology infrastructure if more than one technology suite is in use. In other words, cultural influences and design ideologies, mostly external influences, will affect the work transformation processes. We ask: to what extent do these external ideological influences shape our understanding of organizational routines in anticipated digital work redesign and transformation? Regarding the institutional implication issue, institutions of education, legal, family life, transportation, and religion affect policy directions, urban planning, skills training, and technology research and work redesign in response to issues of technological underemployment and unemployment for social justice and fair distribution of resources and access, and organizing workspaces (Bailey & Barley, 2020). Successful organizational routinization in this regard will therefore be impacted by the affinity with which certain technologies have been adopted to address these external institutional pressures. The implication of these external pressures comprises the impact of global process templates in the mutually constitutive routinization model posited by Rossi et al. (2020). Again, regarding issues of variation, power, ideology, and institution implications on technological designs, the influential work of bodies such as Data and Society, the United Nations’ AI for Good, the Future of Life Institute, and Data Science for Social Good programs at universities with funding and links to technology firms and design communities embodies the criticality of the considerations discussed thus far (Bailey & Barley, 2020). To reiterate, Bailey and Barley’s unified approach to inculcating these implications is a welcome approach. Nonetheless, we have less knowledge on the effect of the mix of these four issues on malleable artifactual redesigns and routine dualities in ensuring successful digital work transformation processes. In addition to artifact malleability in contextual settings, intelligent systems—AI, machine learning, big data, robotics, smart sensors, internet of things (IoT), blockchain for intelligent systems, and analytics— pose dynamic challenges to the work transformation agenda. Moreso, in the face of unexpected widespread technological unemployment which exacerbates boundary conditions, requiring a rethinking of how we study the occasioning of organizational routines for successful work outcomes

8  Routinization of Digital Transformation of Work: A Discursive… 

269

(Bailey & Barley, 2020; Tsai, 2021). We conceive, therefore, that the inherent capability of digital transformation of malleable artifacts as an organizational resource—not only objects of inertia and conflicts—will instigate a continuous change in routinized work processes, thereby predominating organizational work practices that are result-actioned rather than just goal-­ oriented. We argue that the resultant inherent capability of any technological artifact lined up for work transformation is a function of the implications of these four issues and vice versa: dynamic in nature. The dynamism comes from the myriad of external and internal influences, as well as different actor goals on the digital work transformation agenda across organizations. Thus, how do the IS discipline make sense of these dynamics for the routinization of DTW for research, teaching, and practice? In other words, how do we coherently present what we know now to form a thesis on the routinization of DTW?

 ositioning Routinization: The Constituted P Sphere of Worlds Swanson (2019) posits that technology manifests itself in four constitutive spheres: the sphere of worlds, practices, routines, and devices. Mainly, in a world view, humans are always seeking some form of high-level order or bureaucratic arrangement to understand or define how things should work around them. Routinizing DTW is thus seeking the appropriate lens to understand how constituted practices, routines, and devices underscore the digital transformation agenda or work redesign at the sphere of worlds. Ours is a broader view that draws upon what is known of routine as a practice component, although maybe a narrower starting lens. Such a broader world view is not far-fetched in seeking a native IS theory in a discursive manner. After all, it is an endeavor that will undergo several ‘chicken and egg’ discourses until disciplined and plausible meanings are ascribed. Our positioning will be tautological if the reference is only to the routines as constitutive of devices but nested within spheres of practices already nested in spheres of world. 3  See the explanation of the four spheres of technology in Swanson (2019).

3

270 

D. Treku et al.

 hy Routinization of DTW and Not the DTW W of Organizational Routines? Routines are integral to technological change (Swanson, 2019), but the underlying assumption refers to the aspect of technology put into use within organizations. These integral routines are “constructed as components of broader practices subject to transformative change” (Swanson, p.  1009). In as much as our work alludes to such constructions and reconstructions within organizational IT use, we must consider the different sets of implications of these routines of which future practices may or may not be directly linked to the work practices that ensue within organizations as users appropriate the technology following Bailey and Barley’s (2020) conceptions. These considerations are challenged and eliminate the pervasiveness of emergent IT systems if the ever-evolving DTW is constricted to provide an enduring basis for routinization. For instance, although turbulent working conditions, such as those brought about by COVID-19, may be seen as external pressures (i.e., force majeure from the environment), other external pressures from power, ideological, and institutional influences that present ongoing turbulent conditions may impact what we know of routines as practice components. Aside from turbulent conditions from external pressures, DTW portends ongoing internal and systematic turbulences that we may well agree inspires adaptations to new work processes (i.e., innovations). Thus, even without COVID conditions, increased IT virtualization processes have challenged routinized IT practices in organizational work. Every major IT advancement could also add to this digital turbulence. With COVID-19, virtual technology use also increased with increasing IT budgets for holding virtual events and remodeling workplace spaces. Therefore, routinization of DTW provides the broader lens of routines as sources of dynamic agencies that bring variations to ongoing work transformations fueled by environmental and technological turbulence for better sustainability benefits. In this routinization application, we could broaden the scope of how ostensive routines come to be imbricated. An imbrication that does not equate to the technology but strongly informs the perception of users and how such “routines instigate the generation of

8  Routinization of Digital Transformation of Work: A Discursive… 

271

broader execution of routines serving human practices” (Swanson, 2019) in consonance with routines as duality within malleable technologies (Feldman & Pentland, 2003; Rossi et al., 2020).

 ontrolled Genetic Mutation as an Analogy C for Routinization of DTW Genetic mutations are changes in the DNA sequence that results in diversity among living organisms (Kirkwood, 1989). Although naturally occurring, several examples also exist of induced genetic mutations that keep vital properties of the parent DNA sequence (Campbell & Eichler, 2013). Nelson and Winter (1982), therefore, posited that innovation in capabilities (i.e., diversity) occurs through gene-like recombination of existing routines. Similarly, digital transformation is due to innovation in technology capabilities that result in a plethora of new IT usage, establishing the new technology4 as a routine capability (Swanson, 2019) (see Table  8.1 for comparison). The concept of genetic mutation can be applied to understanding the design, adoption, and use of IT artifacts and their historicity in work redesign practices to maintain core scripts of organizational routines, modify old scripts, and create new ones to satisfy a digital transformative agenda (Barley & Tolbert, 1997). For instance, since the launch of Six Degree, possibly the first social media networking platform, there are now several diverse ranges of social networking sites (SNS) that have resulted from digital innovations. These new SNS have transformed several facets of life, including how work is done. An example is how an organization’s traditional product marketing has evolved and become part of the digital marketing transformation that relies heavily on several SNS features. For mutations to be sustained, their properties must be heritable. Similarly, for investments in digital innovations that lead to digital transformation, the core technology properties must be passed from one variant of technology to the next. This will ensure continuity in the use and benefits of emergent or new technology. Just as the interaction between inherited mutation and the  Swanson (p. 1020) defines a new technology as a capability forged from introducing new devices and routines to a human population and its practices. 4

272 

D. Treku et al.

Table 8.1  Comparing our routinization view with routine capability view (adapted from Swanson (2019)) Central concepts

Routine capability view

Devices

Device as routine component Built-in routine performance

Affordances

Routines Capabilities

Practices

The sphere of worlds

Routine as practice component Technology as routine capability built-in practice

Routinization of DTW Malleable digital infrastructure or IT portfolio as a routine component Built-in performative routines and new ostensive routines separate from the imbricated ones Performative routines and new ostensive routines as practice components Malleable digital infrastructure as controlled routine capability in practice constitutive of specific or generic mutation that informs future core properties or inherent capabilities of new technology Practices as constituted from routines, capabilities, and turbulences

Practices as constituted from routines and capabilities Suggested as a Defines the scope of our routinization research direction view as constitutive of power, ideology, in Swanson (2019) and institutional influences

operating environment leads to diversity, the relationship between the core properties of new technology (i.e., the inherent capabilities) that drive digital change and boundary conditions under the influence of internal and external turbulences leads to diversity in DTW. Thus, when an organization’s market strategy (i.e., work design) is developed around using Facebook, for instance, the emergence of TikTok as a new digital marketing technology may create a portfolio of malleable digital artifacts (i.e., flexible technology architecture) for an organization’s digital marketing strategy (an example of digital transformation of work) which will require changes (i.e., controlled change) that perpetuate the organizations marketing goals. In this regard, routinization of the organization’s digital marketing strategy is possible when the core properties of the new driving technology (TikTok) are heritable from the properties of Facebook. This is usually the case with technology versioning and new product improvements. This design consideration or adoption ideal

8  Routinization of Digital Transformation of Work: A Discursive… 

273

allows for the level of digital malleability needed to make the necessary adjustments for future ostensive and performative routines in work redesign as adopted artifacts are put to result-oriented use. Flexible technology infrastructure (another for the portfolio of malleable digital artifacts) is a result of imbrications of previous humans and agencies (Leonardi, 2011) that formulate their core properties giving life to inscribed routines that either constrain or afford work. The imbrication metaphor on routine (Leonardi, 2011) is steeped in routines as things (Feldman et  al., 2016) and, although fundamental to our discussion, does not account for the controlled changes needed to address the diversity issue and the ‘variation’ implication issue espoused in Bailey and Barley. Beyond routines as things are routine dynamics (Feldman et al., 2016) which account for the needed controlled changes owing to the implications of variations (Bailey & Barley, 2020) in the core properties of the technology infrastructure. Hence, just as some genetic mutations affect one carrier while others affect all offspring of the parent (Kirkwood, 1989), changes to the core properties of the underlying technology may lead to innovations of a specific technological redesign. In contrast, changes to core properties that affect all properties of dependent technology would lead to innovations regarding individual tasks and the entire work systems via the inscribed (or imbricated) patterned actions of the new artifact designs. Referring to the earlier example, when an organization’s marketing strategy includes an SNS such as Facebook and other offline modules such as email as a malleable technology portfolio, the work process effects of any change to one technology depends on the degree of coupling among their properties. Hence, routinization (i.e., controlled change) of the organization’s digital marketing strategy could be specific, that is, task-driven (either SNS use or email use only) or generic, that is, the entire communication routine is a function of the interdependence of the core properties of the SNS and the email communication technologies. Based on our discussion and analogy, we establish two propositions that inform our conception of routinization of digital transformation of work: (1) Routinization of DTW is an organizational capability when the core properties of the driving technologies are heritable. (2) Routinization as organizational capability is a function of the interdependent abilities of core properties of the driving technologies.

274 

D. Treku et al.

Digital Transformation and Work Redesign Swanson offers four modes of technology as routine capability change: change by design, change by execution, change by diffusion, and change by shift. We agree to these modes as they appeal to the broader technological sphere of worlds as well as technology as routine capability. Particularly, these modes suggest how we can understand the genetic mutations of core properties of malleable technological infrastructure (i.e., their inherent capabilities as routine capabilities) and the variations among interdependencies to inform our routinization view. To explain the variations in performative and ostensive routines as practice components of mutated technologies, we must explicate the dimensions of digital transformation or organizational work which they target. Our thinking here is consistent with the collection of routines as a specific organizational capability: “An organizational capability is a high-level routine (or collection of routines) that, together with its implementing input flows, confers upon an organization’s management a set of decision options for producing significant outputs of a particular type” (Winter, 2003, p. 991) (Table 8.2). With emerging technologies, the core job dimensions of work redesign-­ task significance, task identity, skills variety, autonomy, and feedback process (Hackman & Lawler, 1971) define the human-centered organization’s capabilities. Thus, the inherent ability to effect changes to these dimensions of work affects the psychological state of employees (Hackman & Oldham, 1976), and this consideration is the foundation of human-­ centered work redesign. We discuss three digital work transforming capabilities that allow for routine practices around the core job dimensions: design as transforming capability, process as transforming capability, and sustainment as transforming capability. Table 8.2 Variations in performative and ostensive routines as practice components World implications

Power

Nature of turbulence

Internal External External External

Modes of genetic mutation Design Execution X Diffusion X Shift

Ideology Institution X

X X

X

8  Routinization of Digital Transformation of Work: A Discursive… 

275

Design as Transforming Capability Organizations continuously use a combination of capabilities, such as netsourcing, to foster innovations by transferring their work/tasks sourcing to digital platforms (Rai et al., 2019). Some of the platforms undertake tasks through artificial intelligence (AI) applications. Tasks on digital platforms can be performed by either humans or AI agents, but domination by one party may lead to ineffective results. Hence, a human-AI hybrid built-in routine performance is required for human rationalization and maintenance of error-free solutions (Rai et al., 2019). Developing the skills set as organizational practices to manage task substitution (AI replaces humans), task augmentation (AI and humans are complementary), and task assemblage (AI and humans integrate to function) will lead to the efficient design of work content that will enhance task significance for employees (Rai et al., 2019). Human-AI hybrid’s efficiency depends on the quality of data used in the content design. Some organizations invest in resources such as blockchain technologies to design contents and infrastructures that ensure or guarantee guardianship over the organization’s data assets (Rai et al., 2019). Therefore, embedding IT-enabled functions in features as components of the work design that permit choices or task identification (known as platform repertoire) allows employees to own work practices and organizations to be differentiated from their competitors (Li et al., 2019). Table 8.3 summarizes key IS analogies that serve as capabilities essential for routine work content redesign.

Process as Transforming Capability Organizational processes can sometimes be complex, and leaning on traditional procedures may not produce desired outcomes. Such processes may be improved to capture value through technological capabilities that involve embedding applications that support these organizational processes (Tarafdar et al., 2019). Competencies such as data science, business domain proficiency, enterprise architecture expertise, operational IT backbone, and digital inquisitiveness are now relevant in transforming

276 

D. Treku et al.

Table 8.3  Capabilities supporting routine work content design transformation Core job dimension

Routine capability

Task Built-in routine significance performance

Technological properties Human-AI hybrids management

Skills variety

Practices as Data constituted guardianship from routines ability and capabilities

Task identity

Device as routine Platform component repertoire availability

Objectives To implement task substation, task augmentation, and task assemblage that help in the digital transformation of work content To assist the organization in achieving a greater competitive advantage and improve experiences for the employees through an effective decision from quality work data. To help the organization to benefit from empowering employees by granting the choice to decide the content of work

how work is executed (Tarafdar et al., 2019). This is because transforming the work process with the preceding competencies radically enhances business operations and reduces costs. Additionally, some work types require real-time data for service delivery. Routine practices that use data analytics technologies to create data lakes in the work process and routines facilitate and enhance sensemaking and coordinative actions and governance of the business processes. In the context of DTW, the use of the data lakes in the work process may rely on human-AI hybrid capabilities and allows for efficient resource utilization by workers (Rai et  al., 2019). For instance, data lakes, technology as routine technology built-in capability, in the work process aid in effective data storage and visualization applications as the seamless transition of work experiences (Lehrer et al., 2018). Table 8.4 shows a summary of analogy capabilities that support routine work process redesign.

8  Routinization of Digital Transformation of Work: A Discursive… 

277

Table 8.4  Capabilities supporting routine work process transformation Core job dimension

Routine capability

Process capabilities

Task identity

Routine as practice component

Task modularization

Skills variety

Routine as practice component

Task Technology as significance routine capability built-in practice

Objectives

To enable flexibility in task allocation and coordination, thereby leading to task process transformation Big data analytics To govern the data analytic utilization process and visualization applications for data sensemaking that will facilitate an organization’s process of achieving digital and service innovativeness Operational IT To drive the organization backbones toward IT success, improving business processes, reducing costs, and generating revenues

Sustainment as a Transforming Capability The use of disruptive technologies such as human-AI hybrid capabilities in digital platforms can provide organizations with some degree of dominance, which makes them innovative and competitive through collective collaborations among employees (Rai et  al., 2019). The collaborative environment empowers employees to support organizational sustenance through the institutionalization of changes to work due to disruptive technologies. Organizational skills view technology as routine built-in capabilities that should aim at attracting, organizing, coordinating, controlling, and monitoring work through standards, policies, and rules to govern and sustain everyone on the platform (Gol et  al., 2019). For example, the use of enterprise resource planning capabilities supports data sharing, facilitates collaboration and communication among organizational units, and affords the achievement of sustained work redesign (Dery et  al., 2017). As a built-in routine performance, the ability to

278 

D. Treku et al.

Table 8.5  Capabilities supporting routine sustainment of work transformation Core job dimension

Routine capability

Sustainment capabilities

Autonomy

Built-in routine performance

Human-AI hybrids management

Skill variety

Technology as routine capability built-in practice

Task Routine as significance practice component

Objectives

To enhance platform governance and organizational sustainment through individual collaboration and empowerment Enterprise To support data sharing and resource communication among planning skills employees, facilitate collaboration between management and employees, and propel a sustained environment to achieve work meaningfulness Human To enhance and empower resources workforce management by analytics scanning employee usage experiences for speedbumps, structuring work through access management and control, leading to a digitally sustained work environment

manage and control processes using disruptive technologies supports the sustainment of work (Dery et al., 2017). Table 8.5 shows capabilities of analogies that support the sustainment of redesigned work.

Conclusion There are increasing investments in emerging technologies aimed at altering work practices and promoting the productivity of organizations. Despite improved knowledge on strategies to harness the potential of new and existing technologies for effective work practices even under

8  Routinization of Digital Transformation of Work: A Discursive… 

279

turbulent conditions, the efficacy of these strategies may be less impactful if present routinization concepts are not re-examined to explicate DTW. We expounded on technology as a routine capability to present a robust routinization view for explicating DTW.  By using the genetic mutation analogy, we explored the development of a native theory-driven explanation of routinization of DTW. Figure 8.15 summarizes the routinization of digital transformation of work view. Effective digital transformation comes with skills requirements that allow for managing new forms (mutants) of the emerging technology that promotes the overall objective of the work achieved with the technology. The variations of performative/ostensive routines—design, execution, diffusion, and shift—become organizational capabilities based on tradeoffs between power, ideology, and institution. While digital transformation of work is essential and sometimes unavoidable, particularly when confronted with turbulence, it is the ability of the organization to understand the core properties of the technology that would result in the routinization of any consequent work. The current research lays the foundation through the postulation of analogies for the future development of unifying IS native theory of digital transformation. Future theorizing discourses and studies may focus on exemplifying our routinization of DTW view via emergent technologies such as blockchain and existing technologies such as ERPs and 3D technology. Such critical exemplifications would help establish propositions and/or hypotheses for positing a native IS theory of DTW routinization.

 Note on Fig. 8.1: We adapt the Bailey and Barley (2020) technology trajectory as a cyclical representation and conflate it with our routinization diagram, conceiving that the left and right extensions in their issue implications may be either tangential or informing further design, implementation, and use trajectory. We adapt the Bailey and Barley technology trajectory as a cyclical representation and conflate it with our routinization diagram, conceiving that the left and right extensions in their issue implications may either be tangential or in a feedback loop (Blakçori & Aroles, 2021) to inform further design, implementation, and use. Our cyclical representation is consistent with the argument that envisioning the kind of future work wanted and feeding the visions back into the artifactual design for better work outcomes. Tangents of power, ideology, and institutional implications may affect future designs indirectly, while feedback implications of power, ideology, and institutional influences are the direct effect on the inherent capability of the technology inscribed in the design for better and agile-patterned actions. 5

280 

D. Treku et al.

References Alt, R., Beck, R., & Smits, M. T. (2018). FinTech and the transformation of the financial industry. Electronic Markets, 28(3), 235–243. https://doi. org/10.1007/s12525-­018-­0310-­9 Alter, S. (2008). Defining information systems as work systems: Implications for the IS field. European Journal of Information Systems, 17(5), 448–469. https:// doi.org/10.1057/ejis.2008.37 Bailey, D.  E., & Barley, S.  R. (2020). Beyond design and use: How scholars should study intelligent technologies. Information and Organization, 30(2), 100286. https://doi.org/10.1016/j.infoandorg.2019.100286 Baiyere, A., Salmela, H., & Tapanainen, T. (2020). Digital transformation and the new logics of business process management. European Journal of Information Systems, 29(3), 238–259. https://doi.org/10.1080/0960085X. 2020.1718007 Barley, S. R. (1986). Technology as an occasion for structuring: Evidence from observations of CT scanners and the social order of radiology departments. Administrative Science Quarterly, 31, 78–108. Barley, S.  R., & Tolbert, P.  S. (1997). Institutionalization and structuration: Studying the links between action and institution. Organization Studies, 18(1), 93–117. Baskerville, R. L., & Myers, M. D. (2009). Fashion waves in information systems research and practice. MIS Quarterly, 33(4), 647–662. https://doi. org/10.2307/20650319 Becker, M.  C. (2004). Organizational routines: A review of the literature. Industrial and Corporate Change, 13(4), 643–678. https://doi.org/10.1093/ icc/dth026 Benbasat, I., & Zmud, R. W. (2003). The identity crisis within the IS discipline: Defining and communicating the Discipline’s Core properties. MIS Quarterly, 27(2), 183–194. Bhatti, A., Malik, H., Kamal, A. Z., Aamir, A., Alaali, L. A., & Ullah, Z. (2021). Much- needed business digital transformation through big data, internet of things and blockchain capabilities: Implications for strategic performance in telecommunication sector. Business Process Management Journal, 27(6), 1854–1873. https://doi.org/10.1108/BPMJ-­12-­2020-­0553 Blakçori, F., & Aroles, J. (2021). Assessing the role of managerial feedback in changing routines in small and medium enterprises. Journal of Organizational Change Management, 34(3), 570–589. https://doi.org/10.1108/JOCM-­08­2020-­0235

8  Routinization of Digital Transformation of Work: A Discursive… 

281

Boratyńska, K. (2019). Impact of digital transformation on value creation in Fintech services: An innovative approach. Journal of Promotion Management, 25(5), 631–639. https://doi.org/10.1080/10496491.2019.1585543 Bygballe, L. E., Swärd, A., & Vaagaasar, A. L. (2021). A routine dynamics lens on the stability-change dilemma in project-based organizations. Project Management Journal, 52(3), 278–286. https://doi.org/10.1177/875697 2820986660 Campbell, C.  D., & Eichler, E.  E. (2013). Properties and rates of germline mutations in humans. Trends in Genetics, 29(10), 575–584. Chanias, S., Myers, M. D., & Hess, T. (2019). Digital transformation strategy making in pre-digital organizations: The case of a financial services provider. The Journal of Strategic Information Systems, 28(1), 17–33. https://doi. org/10.1016/j.jsis.2018.11.003 Cohen, M.  D., & Bacdayan, P. (1994). Organizational routines are stored as procedural memory: Evidence from a laboratory study. Organization Science, 5(4), 554–568. Cooper, A., Reimann, R., & Cronin, D. (2007). About face 3: The essentials of interaction design. Wiley. Davidson, E., Winter, J., Wessel, L., & Winter, S. (2021). Call for papers—Data governance, digital innovation and grand challenges. Information and Organization. https://www.journals.elsevier.com/information-­a nd-­ organization/call-­f or-­p apers/journals.elsevier.com/information-­a nd-­ organization/call-­for-­papers/undefined Dery, K., Sebastian, I. M., & van der Meulen, N. (2017). The digital workplace is key to digital innovation. Mis Quarterly The Executive, 16(2), 135–152. Feldman, M.  S. (2004). Resources in emerging structures and processes of change. Organization Science, 15(3), 295–309. Feldman, M.  S., & Pentland, B.  T. (2003). Reconceptualizing organizational routines as a source of flexibility and change. Administrative Science Quarterly, 48(1), 94–118. https://doi.org/10.2307/3556620 Feldman, M. S., & Pentland, B. (2005). Organizational routines and the macro-­ actor. In Actor-network theory and organizing (pp. 91–111). Liber Feldman, M. S., Pentland, B. T., D’Adderio, L., & Lazaric, N. (2016). Beyond routines as things: Introduction to the special issue on routine dynamics. Organization Science, 27(3), 505–513. https://doi.org/10.1287/orsc.2016.1070 Fichman, R. G., Dos Santos, B. L., & Zheng, Z. (Eric). (2014). Digital innovation as a fundamental and powerful concept in the information systems curriculum. MIS Quarterly, 38(2), 329–353.

282 

D. Treku et al.

Forsythe, D.  E. (2001). Studying those who study us: An anthropologist in the world of artificial intelligence. Standard University Press. Goh, J. M., Gao, G. (Gordon), & Agarwal, R. (2011). Evolving work routines: Adaptive routinization of information Technology in Healthcare. Information Systems Research, 22(3), 565–585. https://doi.org/10.1287/isre.1110.0365 Gol, E. S., Stein, M.-K., & Avital, M. (2019). Crowdwork platform governance toward organizational value creation. Journal of Strategic Information Systems, 28(1), 175–195. Hackman, J. R. (1980). Work redesign and motivation. Professional Psychology, 11(3), 445–455. Hackman, J. R., & Lawler, E. E. (1971). Employee reactions to job characteristics. Journal of Applied Psychology, 55(3), 259–286. https://doi.org/10. 1037/h0031152 Hackman, J. R., & Oldham, G. R. (1976). Motivation through the design of work: Test of a theory. Organizational Behavior and Human Performance, 16(2), 250–279. https://doi.org/10.1016/0030-­5073(76)90016-­7 Hassan, N. R., Lowry, P. B., & Mathiassen, L. (2022). Useful products in information systems theorizing: A discursive formation perspective. Journal of the Association for Information Systems (JAIS), 23(2), 418–446. https://doi. org/10.17705/1jais.00730 Hassan, N. R., Mathiassen, L., & Lowry, P. B. (2019). The process of information systems theorizing as a discursive practice. Journal of Information Technology, 34(3), 198–220. https://doi.org/10.1177/0268396219832004 Hassan, N.  R., & Willcocks, L.  P. (2021). Introduction: Why theory?(Mis) understanding the context and rationale. In Advancing Information Systems Theories: Rationale and Processes (pp. 1–52). Palgrave Macmillan. Hesse, M. (1966). Models and analogies in science. University Press, Notre Dame. Hoffer Gittell, J. (2002). Coordinating mechanisms in care provider groups: Relational coordination as a mediator and input uncertainty as a moderator of performance effects. Management Science, 48(11), 1408–1426. https:// doi.org/10.1287/mnsc.48.11.1408.268 Jasperson, J. (Sean), Carte, T. A., Saunders, C. S., Butler, B. S., Croes, H. J. P., & Zheng, W. (2002). Review: Power and information technology research: A Metatriangulation review. MIS Quarterly, 26(4), 397. https://doi. org/10.2307/4132315 Kirkwood, T.  B. L. (1989). DNA, mutations and aging. Mutation Research/ DNAging, 219(1), 1–7.

8  Routinization of Digital Transformation of Work: A Discursive… 

283

Legner, C., Eymann, T., Hess, T., Matt, C., Böhmann, T., Drews, P., Mädche, A., Urbach, N., & Ahlemann, F. (2017). Digitalization: Opportunity and challenge for the business and information systems engineering community. Business & Information Systems Engineering, 59(4), 301–308. https://doi. org/10.1007/s12599-­017-­0484-­2 Lehrer, C., Wieneke, A., vom Brocke, J., Jung, R., & Seidel, S. (2018). How big data analytics enables service innovation: Materiality, affordance, and the individualization of service. Journal of Management Information Systems, 35(2), 424–460. https://doi.org/10.1080/07421222.2018.1451953 Leonardi, P.  M. (2011). When flexible routines meet flexible technologies: Affordance, constraint, and the imbrication of human and material agencies. MIS Quarterly, 35(1), 147–168. Li, H., Fang, Y., Lim, K. H., & Wang, Y. (2019). Platform-based function repertoire, reputation, and sales performance of E-marketplace sellers. MIS Quarterly, 43(1), 207–236. https://doi.org/10.25300/MISQ/2019/14201 Lyytinen, K., & King, J. L. (2004). Nothing at the center?: Academic legitimacy in the information systems field. Journal of the Association for Information Systems, 5(6), 8. Markus, M. L. (1983). Power, politics, and MIS implementation. Communications of ACM, 26(6), 430–444. Mingers, J., & Willcocks, L. (2004). Social theory and philosophy for information systems. John Wiley & Sons, Ltd. https://www.wiley.com/en-­us/Social+Theor y+and+Philosophy+for+Information+Systems-­p-­9780470851173. Morton, J., Wilson, A. D., & Cooke, L. (2020). The digital work of strategists: Using open strategy for organizational transformation. The Journal of Strategic Information Systems, 29(2), 101613. https://doi.org/10.1016/j.jsis.2020.101613 Nelson, R. R., & Winter, S. G. (1982). An evolutionary theory of economic change. Harvard University Press. https://www.hup.harvard.edu/catalog.php?isbn= 9780674272286. Orlikowski, W. J., & Iacono, C. S. (2001). Research commentary: Desperately seeking the “IT” in IT research—A call to theorizing the IT artifact. Information Systems Research, 12(2), 121–134. Orlikowski, W. J., & Robey, D. (1991). Information technology and the structuring of organizations. Information Systems Research, 2(2), 143–169. Pentland, B. T., & Feldman, M. S. (2008). Designing routines: On the folly of designing artifacts, while hoping for patterns of action. Information and Organization, 18(4), 235–250. https://doi.org/10.1016/j.infoandorg. 2008.08.001

284 

D. Treku et al.

Rai, A., Constantinides, P., & Sarker, S. (2019). Next generation digital platforms toward human AI hybrids. MIS Quarterly, 43(1), iii–ix. Robey, D. (1987). Implementation and the organizational impacts of information systems. Interfaces, 17(3), 72–84. https://doi.org/10.1287/inte.17.3.72 Robey, D., & Boudreau, M.-C. (1999). Accounting for the contradictory organizational consequences of information technology: Theoretical directions and methodological implications. Information Systems Research, 10(2), 167–186. Rogers, E. M. (1995). Diffusion of innovation. The Free Press. Rossi, M., Nandhakumar, J., & Mattila, M. (2020). Balancing fluid and cemented routines in a digital workplace. The Journal of Strategic Information Systems, 29(2), 101616. Rowe, F. (2018). Being critical is good, but better with philosophy! From digital transformation and values to the future of is research. European Journal of Information Systems, 27(3), 380–393. https://doi.org/10.1080/0960085X. 2018.1471789 Swanson, E. B. (2019). Technology as routine capability. MIS Quarterly, 43(3), 1007–1024. https://doi.org/10.25300/MISQ/2019/14653 Tarafdar, M., Beath, C. M., & Ross, J. W. (2019). Using AI to enhance business operations. MIT Sloan Management Review, Summer, 10. Tsai, C.-W. (2021). Toward blockchain for intelligent systems. IEEE Consumer Electronics Magazine. Vial, G. (2019). Understanding digital transformation: A review and a research agenda. The Journal of Strategic Information Systems, 28(2), 118–144. https:// doi.org/10.1016/j.jsis.2019.01.003 Weick, K. E. (1989). Theory construction as disciplined imagination. Academy of Management Review, 14(4), 516–531. Winter, S. G. (2003). Understanding dynamic capabilities. Strategic Management Journal, 24(10), 991–995. https://doi.org/10.1002/smj.318 Yoo, Y., Henfridsson, O., & Lyytinen, K. (2010). Research commentary—The new organizing logic of digital innovation: An agenda for information ­systems research. Information Systems Research, 21(4), 724–735. https://doi. org/10.1287/isre.1100.0322 Zammuto, R.  F., Griffith, T.  L., Majchrzak, A., Dougherty, D.  J., & Faraj, S. (2007). Information technology and the changing fabric of organization. Organization Science, 18(5), 749–762.

9 Patterns for Visualizing the Aesthetic Qualities of Business Processes Monika Blattmeier

Introduction This chapter focuses on how to bring knowledge to life in business processes. It provides an example of metaphor as a product of theorizing— see Table 2.1 in Chap. 2. The chapter applies the metaphor of the atmosphere to describe a different type of business process. The traditional business process is a series of tasks performed in a sequence that has a beginning and an ending. The aesthetic business process is made up of experiences—much like how one experiences the atmosphere—as a result of the elements and conditions present in the business process. Where the traditional business process improves outcomes by linking services or tasks in parallel (e.g., horizontally across different functional

M. Blattmeier (*) Department of Mechanical Engineering, Hochschule Emden/Leer Fachbereich Technik - Abt. Maschinenbau, University of Applied Sciences, Emden, Germany e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. P. Willcocks et al. (eds.), Advancing Information Systems Theories, Volume II, Technology, Work and Globalization, https://doi.org/10.1007/978-3-031-38719-7_9

285

286 

M. Blattmeier

departments), focusing on the aesthetic business process improves the overall outcome in different ways, by constructing various “atmospheres,” in particular working on the “thinking” atmosphere (engendering independent thinking in constructing business processes) and the “sensing” atmosphere (feeling what works in a business process). The concept of aesthetic business processes aims to stimulate aesthetic satisfaction in digitalized business processes so that new thinking necessary for business process design becomes possible. Climate change, globalization and digitalization are challenging companies to set not only economic goals but also social and ecologically friendly goals. To accomplish these different goals, organizations need to align their organizational culture, the company’s “personality” (Flamholtz & Randle, 2011, p. 6; Schein, 2016) that includes explicit and implicit factors related to social and ecology. Groysberg et al. (2018) explain culture as “the tacit social order of an organization: It shapes attitudes and behaviors in wide-ranging and durable ways” (p. 44). Sophisticated structures and strategies can fail if the culture is not aligned with it in business processes. “Culture eats strategy for breakfast”: the statement attributed to Peter Drucker says that while culture is not everything, without a culture there is not much left (Wörwag & Cloots, 2021, p. 325). Probably the strongest trend of our time seems to be digitalization, boosted by the Corona pandemic. The sudden necessity for home office work has increased the need for new forms of work, which can be summarized under the term New Work (Bergmann, 1977; Foelsing & Schmitz, 2021). “If you want to get anything done form a group” explains the artist Mary Beth Edelson (Mieville, 2022, p.  130). Collectives as places of collective thinking and creation seem to offer an answer to the isolation of these times. But the months of the Corona pandemic have also shown that “as practical as coordination, presentations and the concrete exchange in the digital tile are, it is difficult to develop creativity or build relationships in front of the screen” (Fischer, 2021, p. 3). The personal closeness that can be built up in a conversation cannot be achieved by digital means: “body language, moods, all that is largely absent in a video conference,” explains the manager and consultant Christof Horn (Horn et al., 2022, p. 74). So how do you stimulate curiosity and fantasy in working communities when their thoughts and actions increasingly take place in digitalized business processes?

9  Patterns for Visualizing the Aesthetic Qualities of Business… 

287

It is against this background that the concept of aesthetic business processes offers solutions. The aim of aesthetic business processes is to integrate the various forms of human experience of those involved in the process into a holistic experience. At the center of the concept is the ability to generate new imaginative structures and mental images from experiences in a variety of ways (Schlaeger & Tenorth, 2020). The current modelling of business processes serves primarily to make the cross-­ functional sequence of value-adding activities transparent (Schmelzer & Sesselmann, 2020; Normann, 2001). But as the writer C. S. Lewis noted, in creating transparency, which he compares to “seeing through,” we also pay a price (Roß, 2020): “If you see through everything then everything is transparent and a wholly transparent world is an invisible world. To ‘see through’ all things is the same as not to see” (p. 187). In this sense, the concept of aesthetic business processes offers in Component 1 (see Fig. 9.1) a repertoire of different aesthetic strategies to perceive aesthetic qualities in the world of business processes, to make them visible in the format of a pattern and thus as an image. The concept uses a digital atlas as an electronic knowledge repository (Kankanhalli et al., 2005; Durcikova et al., 2018) to collect and classify these images and to make them available for the design of aesthetic business processes in Component 2 of the concept (see Fig. 9.1). Thus, Component 3 can succeed in integrating

Fig. 9.1  Components of the concept of aesthetic business processes that want to observe aesthetic qualities, develop them as patterns to design aesthetic business processes, and finally think and learn together in companies

288 

M. Blattmeier

knowledge management, organizational learning and innovation management (Fig. 9.1). This lack of integration between knowledge management, organizational learning and innovation management was criticized by Franken and Franken (2020) as not being consistent enough in typical business process improvement efforts.

 esthetic Business Processes: A A Working Definition In his sketch of the aesthetic economy, Gernot Böhme (2016) uses the term “aesthetic work” to describe human activities “that aim to give things and people, cities and landscapes an appearance, to give them a charisma, to provide them with an atmosphere” (p.  26). Applying his atmosphere metaphor, we refer to aesthetic business processes as accessing what is primarily given to the senses in two different ways: (1) through the experiences that people have in business processes, such as bodily senses, whether pleasant, unpleasant, or neutral and (2) the creation of this atmosphere as a result of the elements and conditions present in business processes. The opposite of aesthetic business processes would be, for example, business processes in which people do one thing and think about many other things in multitasking mode. Before going into the atmosphere of business processes in more detail, the meaning of business processes in organizations should first be explained. Business process patterns mean, with regard to the aesthetics of business processes, that in good and proven practices the aesthetic qualities are perceived. Based on the uniqueness of each company, actions of the business design can be characterized by (a) specialization and modularization in the vertical direction, (b) linking of services in the horizontal direction and (c) optimization of the value creation structures. The structure of an organization follows the business process (see Fig. 9.2). According to the Bauhaus architects’ principle “structure follows function,” business processes abstract a cross-functional sequence of valuecreating activities that are oriented toward the customer (Suter et  al., 2019; Davenport, 1993). For business process reengineering that seeks to fundamentally redesign a company’s business, two things are crucial

9  Patterns for Visualizing the Aesthetic Qualities of Business… 

289

Fig. 9.2  Patterns in business processes as a value chain that organizes the multidisciplinary communities

(Normann, 2001; Spender, 1996): (1) How should operations be designed to meet customer needs in terms of a company’s strategic goals? (2) How can information and communication technologies support to bring together the work of several people? Due to the different functional areas involved in the production of a product or service, which are organized using division of labor, communication is critical and there is a risk that no one feels ultimately responsible for the product or service. It is particularly important that interdisciplinary teams coordinate the processes for the production of services together across the hierarchy and in its process chain horizontally and continuously (Bourne et al., 2018). Let us look at two perspectives in this issue. Perspective 1 for the Atmosphere as a Mood Produced in Aesthetic Business Processes (Production-Aesthetic)  As in a stage set, an atmosphere is created in business processes when the people “present” perceive, because of certain factors, a certain mood from the processes. Nancy Kline (Miketta, 2018) noted that quality of

290 

M. Blattmeier

everything we do depends on the quality of our thinking which she calls the “Thinking Environment.” This approach of accentuating our thinking allows the members working in the business process to think independently and listen attentively so that a culture of deep, sincere respect for each other emerges in this “Thinking Environment.” She refers to Momo, the heroine in the children’s book by Michael Ende: “Momo could listen in such a way that stupid people suddenly had very clever thoughts. Not because she said or asked something that gave the other person such thoughts, no, she just sat there and listened, with all attention and all sympathy” (Miketta, 2018, p. 35). In aesthetic business processes, this kind of attention flows into the various forms of interaction (see Fig. 9.3), which can be described, for example, as a result of relying less on skilled individual workers and more on people’s abilities to cooperate in complex highly interactive situations (Blackler & McDonald, 2000). 

Fig. 9.3  Aesthetic business processes in which an atmosphere can be experienced holistically, (a) in more or less stable networks (vertical direction) and (b) in more or less familiar activities (horizontal direction), inspired by communities of practice (Pyrko et al., 2019; Wenger et al., 2002)

9  Patterns for Visualizing the Aesthetic Qualities of Business… 

291

Perspective 2 for the Atmosphere as a Bodily Sensing of Emotional Impressions ­(Reception-­Aesthetic). A second perspective that creates the atmosphere in aesthetic business processes analyzes the experiences of the person integrated physically in the business process. It is important to distinguish between experiencing through action and experiencing of feelings (see Fig. 9.2). Experiencing leaves physical traces such as emotions of joy that inspire or that of fear that crushes. “If we want to learn something new and gain fresh insights, we have to go into the experience, because there is a wealth of unconscious knowledge waiting to be kissed awake,” explains Judith Papadopoulos (2021, p.  49). The communication model Embodied Design, which she developed for innovation and communication management, emphasizes the role of the body in experience-based thinking and combines this with rational-analytical thinking in reflection and action. With reference to the Greek concept of aesthesis as sensual perception, aesthetic business processes aim at what engages the senses. Based on the aesthetics of nature or the environment, aesthetic business processes are characterized by an atmosphere that fills people with awareness (Böhme, 2016).  The Concept of Aesthetic Business Processes: The Macro-sequence In order to activate a holistic experience in business processes, the concept of aesthetic business processes wants to offer different ways to develop a felt sense (Papadopoulos, 2021, p. 149; Gendlin, 1997; Juchli & Wiltschko, 1982). In a felt sense, implicit knowledge is expressed through the actions of speaking, drawing or making music, facial expressions and gestures or movements. As a result, an atmosphere can be created that invites members of the business process to perceive an inner resonance as a basis for free-spirited thinking. The concept of aesthetic business processes (see Fig.  9.4) is based on the following processes of knowledge creation: Interactions Between Implicit and Explicit Knowledge  The theory of the “The Knowledge-Creating Company” by the Japanese scientists I. Nonaka and H. Takeuchi is based on the difference

292 

M. Blattmeier

Fig. 9.4  The concept of aesthetic business processes is based on a macro-sequence that integrates the processes of knowledge creation, the four steps of visual thinking according to Dan Roam (2008) and life cycle of a pattern evolution

between implicit and explicit knowledge in companies. While explicit knowledge can be conveyed in words and numbers, implicit knowledge escapes formal language. Implicit knowledge means the experience-based ­knowledge that contains technical elements, such as manual skills, and cognitive elements, such as beliefs. Nonaka and Takeuchi refer to M.  Polanyi and see knowledge creation as something that empathizes with the object, that is, in self-involvement. In order to understand something as a meaningful whole, we need to integrate our body into the details. The interactions of implicit and explicit knowledge result from social processes in four different forms: (1) In socialization, tacit knowledge is exchanged through observation and practice, for example, imitation. (2) In externalization, implicit knowledge is communicated in the form of metaphors, analogies, models or hypotheses. (3) Combination results from combining newly created and existing knowledge to fuse them into a new product. (4) In a form of “learning by doing,” internalization involves the incorporation of explicit knowledge into the implicit knowledge of individuals and/or the organization. 

9  Patterns for Visualizing the Aesthetic Qualities of Business… 

293

Visualizing in Research-Based Learning  If it is true that “imagination is more important than knowledge, because knowledge is limited” as Albert Einstein described it, then the concept of aesthetic business processes wants to nurture and nourish imagination in addition to curiosity. Imagination describes the ability to shape the experiential knowledge acquired through curiosity into images in a variety of ways (Schlaeger & Tenorth, 2020; Turner, 1998). Learning through research means that the learners independently discover something new for them, reflect on it and present the results appropriately (Healey & Jenkins, 2008; Huber, 2009). Dewey (1934) made inquiry accessible to learning processes, explaining inquiry as structuring “so that the elements of the original situation are transformed into a unified whole” (p. 131). In order to illustrate this wholeness that the learner experiences, it is necessary to reduce the experience to its characteristic. In research, this reduction is noticeable in the cognitive performance (Zeller, 2016). The concept of aesthetic business processes specifically uses the four steps of visual thinking according to Dan Roam, meaning looking, seeing, imagining and showing (see symbols in Fig. 9.4) to support learners in their holistic understanding. In the interplay of thinking, seeing and feeling, the forms of experience are to be integrated into a holistic experience so that the learners realize what is of value.  Reconstructing Patterns as Proven Solutions in an Evolutionary Way The concept of aesthetic business processes uses the format of design patterns in the sense of Christopher Alexander, to capture the essence of successful, well-proven practices in business processes. Business process patterns help in two ways: on the one hand, patterns structure a complex reality of business processes with their format in order to make the essence of business process patterns recognizable. Alexander (1979) sees a pattern as a three-part rule, which expresses a relation between a certain context, a problem, and a solution. On the other hand, an exchange of knowledge takes place through the explicit description and naming of these structures. Therefore, the processes of pattern visualization, pattern development and pattern application follow the knowledge transformation according to Nonaka and Takeuchi. In this process, it is important to harmonize different ideas. Strictly speaking, the visualized

294 

M. Blattmeier

patterns do not project real structures but rather individual or—in the case of pattern mining in teams—inter-individual knowledge structures (Kohls & Wedekind, 2008, 2009). Alexander et al. (1977) point to the challenge of finding the true invariant and hope that this will take place in the continuous development of pattern descriptions. Each individual pattern description represents the current best assumption about which arrangement of the physical environment will help solve the given problem (Kohls & Wedekind, 2008, 2009). The decisive factor here is that by explicating the designated form, a consensus on the meaning is established within a group.  isualization: A  Method to  Experience the  Aesthetic Qualities V of Business Processes The following section explains how visualization helps to experience the aesthetic qualities of business processes and thus to perceive patterns as good and proven practices (see Fig. 9.5). In this context, visualization of business process patterns complements the description in the format of a design pattern. A business process pattern is to be used in Alexander’s sense to understand, analogously to the architecture of buildings, the geometry of processes and their most important relationship between elements. Visual Learning as Learning with and from Visualizations While business process patterns focus on the outcome of visualizing, the actions of visualizing are about experiencing the aesthetic qualities of business processes in advance. Gareth Morgan (1986) uses images to view and understand organizations in metaphorically different ways. He is interested in using visualization (images) to bring certain interpretations to the foreground. His images of organizations often follow familiar lines of thought, while others provide new insights and perspectives. Such thinking forms the basis for metaphors. Skills in dealing with visualizations, which have become known as visual literacy (Debes, 1970; Seels, 1994), encompass internal and external components: in particular, the continuum of visual literacy (Seels, 1994; Wafi & Wirtz, 2015) extends from visual thinking as a mental process to visual communication as an inter-

9  Patterns for Visualizing the Aesthetic Qualities of Business… 

295

Fig. 9.5  Visualizing business processes help to better understand the aesthetic qualities of business processes and their relationship to the core of a business process pattern

action process. The steps of looking, seeing, imagining and showing of visual representations are based on a preparation phase (Blattmeier, 2020, 2021) of the project presented in this chapter. In it, students devised strategies for visual learning when they used visualization in research-­ based learning. Sensory perceptions as well as sensory skills form the prerequisite for visual learning. How Business Process Patterns Are Identified As Bauer and Baumgartner (2012) summarize, patterns are not invented but found. Pattern mining is about identifying best or good practices in business processes. Takashi (2016) points to three different pattern mining meth-

296 

M. Blattmeier

ods that can also be used in identifying business process patterns: collaborative introspection, pattern mining interviews and pattern mining workshops. A prerequisite for the creation of knowledge in pattern mining is the exchange of implicit knowledge between different disciplines and with the appropriate perspective and motivation. But implicit knowledge is not easily communicated because it is primarily acquired through experience and hardly ever articulated in words. In particular, looking and seeing help to collect visual information, to select it and finally to recognize the essentials as a pattern.  How Business Process Patterns Are Visualized and Further Developed  The use of visualization proves to be particularly valuable in transforming implicit knowledge into explicit knowledge, especially in a collaborative dialogue. The visual thinking steps of seeing and imagining help to rethink existing assumptions in business processes from the ground up. Diversity in the visualizing teams is particularly helpful because it brings together different ways of looking at things, but also fluctuation and chaos in the visualizing teams and their environment. The emerging visualizations of business process patterns need to be explained at some point to decide what value they represent for the company and the community in the processes. Showing in visual thinking makes it possible to capture the character of business processes and present them comprehensibly in the format of predefined frameworks, which are also used in strategic process management. Dan Roam (2008) explains that showing frameworks simplify the process of visualizing because the question “Oh, boy, which picture could I possibly use to solve this problem?” with the question “Which of the six frameworks maps to the problem I see?” (p. 135). If all patterns are unfinished, it is also necessary to continuously develop the good practice collected in business process patterns and also their visualization. In both cases, the basis for pattern development is the existing visualization of the business process patterns (see Fig. 9.5). Since new knowledge can be created by combining explicit knowledge through a new looking, seeing, imagining and showing, pattern development is close to the combination phase in Nonaka and Takeuchi’s dynamic model of knowledge transformation. 

9  Patterns for Visualizing the Aesthetic Qualities of Business… 

297

In application, visualizations of business process patterns as an outcome and as a process empower and develop good practice in business processes and can also help to mobilize new practices in business processes. However, the focus of the project presented below will be on pattern identification and visualization.

Illustration: Visualizing Aesthetic Qualities of Business Processes in Companies After defining aesthetic business processes, showing their goal and conceptualizing the way to achieve it, the effect of the concept of aesthetic business processes will be briefly shown here using the example of a qualitative study. The study refers to the VisualBP project co-financed by INTERREG, in which students of the University of Applied Sciences Emden/Leer visualized business process patterns of small and medium enterprises (SMEs) in the German-­Dutch border region during the winter semester 2021/22. The aim of VisualBP was to express the aesthetics of business processes in visualizations. This involved approx. 60 students of the Bachelor’s degree program Mechanical Engineering and Design in the module Quality Management, approx. 25 students of the international Bachelor’s degree program Industrial Engineering in the module Quality Management and Quality Assurance, and approx. 15 students of the Master’s degree program Mechanical Engineering in the module Production Systems The students organized themselves in teams of up to five students, visited patterns as good solutions in a particular business context and described these textually and visually with the process participants of the companies. Specific Processes in the Concept of Aesthetic Business Processes (see Fig. 9.6)  Four units corresponding to Dan Roam’s stages of visual thinking (2008) describe a macro-sequence that can be concretized depending on the context of application. The VisualBP project initially focused on the socialization and externalization of the implicit knowledge to be visualized.

298 

M. Blattmeier

Fig. 9.6  The concretization of the macro-sequence on methods and forms of visualizing by the students in the VisualBP project

Each of the four units required a period of about four weeks. Different methods and forms of visualization (see Fig. 9.6) enabled the learners to exemplify the experiential knowledge of business process patterns:  Listening together in the start-up phase: As being part of a greater whole, the students gathered first impressions of the culture lived in the business processes in Unit 1. They visited different sub-processes of business processes for a short time, analyzed documents and conducted qualitative interviews with process participants. The students were able to draw on a guideline that provided various questions. Specifically, these were questions (a) about the role and task of the process participants, (b) about the analysis of typical situations in which problems occur and (c) about the analysis and solution of problems in the respective business processes. At the end of Unit 1, the pattern identification took place within a group discussion with various stakeholders of the companies, the students and the lecturers. In particular, the external perspective of the students made it possible to identify the good practices characteristic of the specific company. In the case of the obw (Ostfriesische Beschäftigungs- und

9  Patterns for Visualizing the Aesthetic Qualities of Business… 

299

Wohnstätten GmbH), an institution for the integration of people with disabilities in the labor market, the patterns were as follows: breaking down into sub-processes, using a pool of diverse resources, valuing and creating work, subsidiary communication, enabling for the general labor market, listening and switching off. • Visualizing together in research-based learning: While looking in Unit 1 was about listening to others and one’s own inner world, students consciously selected impressions in Unit 2 of seeing to understand and appreciate the spectrum of business processes. In Unit 3, students practiced collectively suspending shared memories, visualizing them in such a way that everyone could reflect on them and develop ideas. Finally, in Unit 4, the aim was to show the identified business process patterns. Above all, it was about listening to the story that developed in the visualization, which is more than what each individual can articulate. • Supporting teamwork: The documentation of the preliminary and final pattern visualizations as well as the research design required for this was done via the web-based Pattern-Pool-Tool Visual-Bp. This enabled students to comment on the interim results together and to further develop them as a result of reflection in a seminar led by experienced students. The visualization of the processes was based on (a) a pattern description with the elements problem, solution, effect, context and field of tension and (b) the basics of Business Process Modelling Notation (BPMN). In addition to the objective dependencies, it was also necessary to visualize a variety of other factors that decisively influence a business process pattern as a socio-­technical system. For example, rhythms can be based both on technical time measurement and on the internal clock of the process participants. On the Atmosphere Produced in Business Process Patterns  Many of the patterns discovered in business processes could be perceived by the students as a common good of the respective corporate culture. The following experiences are based on the application of the concept of aesthetic business processes in a business context provided by Ostfriesische Beschäftigungs- und Wohnstätten GmbH (obw).  Perceptible culture in

300 

M. Blattmeier

the business process: With various offers, the obw wants to support people with and without limitations so that they can again “participate and develop as much as possible in all areas of life.” Increasing stress or occupational crises can lead to workers not being available to the general labor market. The obw sees itself as a pioneer of an inclusive society and pursues the mission of living a diverse coexistence. Students visited various locations as part of organizational culture-­related observations and found: “The obw’s mission statement is omnipresent in every department and at every management level.” • Perceptible culture in business process patterns: A team of students has participated in the everyday life of the EMLO (Emder Lohnfertigung) as part of the obw. Depending on their needs and abilities, employees there take on various activities for which they are very grateful. A team of students observed the process participants in the assembly of steel bushings in plastic fuel tanks with the help of a mechanical manual punch for a local automotive supplier. They also focused on verbal data and concluded from the quote of an employee “I like working here because I enjoy my work. There is nothing like that in other companies” the special importance of the working climate in the EMLO for the participants. • Listening and switching off as a business process pattern: Together with the process participants, a student team chose the business process pattern listening and switching off. Due to the high level of self-­responsibility that is transferred to individuals in working life, cooperation in business processes should be actively shaped by all those involved in the process. Teams are generally made up of a whole cosmos of individual differences that can make teamwork exciting but also exhausting (Isaacs, 1999; Kricke & Reich, 2016). In a team, autonomy and freedom are limited because each subject as an actor is always a participant in a team and in an inclusive understanding. Therefore, in the pattern of listening and switching off, all participants take responsibility for the communal. The prerequisite for this is that those involved in the inclusive working environment basically adjust to a heterogeneous group and expect a diversity of prior knowledge, prerequisites, attitudes, attitudes, etc. (Kricke & Reich, 2016).

9  Patterns for Visualizing the Aesthetic Qualities of Business… 

301

• About the Experiences Learners Have When Visualizing  During the visualization, the students created a self-reflective portfolio in which they described their subjective experiences, among other things. The following excerpts are intended (a) to provide an insight into the effect of the concept of aesthetic business processes and (b) to observe the development of imagination as the ability to generate new structures and mental images from remembered experience and already existing imprints.  Perspective 1: Having worked so intensively on the topic of process visualization, we were also able to capture, examine and ­evaluate these processes in companies and finally visualize them. When visualizing these processes, we also combined several of the visualization models, which resulted in new models and patterns. With the help of patterns, we were able to further develop these models. • Perspective 2: The pattern approach was initially a new way of approaching topics for the majority of the group. Especially to understand processes and to visualize them, the pattern approach offers a very good entry into the topic. In addition, the methodology of the pattern approach helps to promote the creativity of the team. In our team, for example, the possibility of visualizing patterns via the editor yED Live also generated creativity. The lecture is less oriented toward a conventional lecture; rather, the methodology of research-based learning is applied. We initially found this type of learning difficult, as there can be different results, especially when solving tasks. And not knowing what is right or wrong is a challenge for the interpretation of results. From this we learned that different perspectives (especially between teams) can also lead to different results, through a different interpretation of the given task. At this point we have learned that there can be several justified outcomes. • Perspective 3: We have learned the most in terms of self-competence. The greatest difficulty here was to engage in abstract and free thinking. In addition to abstract and free thinking, we were able to expand our attention to detail, creativity and our ability to adapt to new circumstances. In the lectures, we received regular feedback on the tasks we had completed, which influenced our critical faculties, our ability to reflect and our self-confidence in our own abilities.

302 

M. Blattmeier

Conclusion The concept of aesthetic business processes aims to stimulate a kind of aesthetic satisfaction in digitalized business processes so that new thinking necessary for business process design becomes possible. It is about fostering communities of workers who can think artfully about what they are doing, who can use their imagination, who can experience their work as it progresses, who can take advantage of the unexpected, who can form judgments about the direction the work is taking based on rules but also on feelings. In order to explore the impact of the outlined concept of aesthetic business processes, the atmosphere of aesthetic business processes was visualized using the example of a concrete company by students in research-based learning. While the previous section focused on the description of the experiences from a reception-aesthetic and production-­aesthetic perspective of the learners, a reflection of the teaching person will take place at this point. Critical Appraisal of the Concept of Aesthetic Business Processes The application of the concept of aesthetic business processes in the entrepreneurial context of Ostfriesische Beschäftigungs- und Wohnstätten GmbH made it clear that visualizations offer a means through which meanings can be conveyed that defy textual description. Secondly, visualizations offered individuals the opportunity to use and develop their mental capacities in their own unique ways. Thirdly, visualization enabled a certain quality of experience, which is to be called aesthetic and stands for a holistic experience. The concept of aesthetic business processes thus enables knowledge management in a broader sense, in which competences and creativity of the process participants are developed and feelings are related to information. So far, the phases of socialization and externalization have been illustrated, in which implicit knowledge has been identified and articulated. The next stage is to experience the phases of combination and internalization in business processes. The knowledge in the visualized business process patterns is to be combined with new knowledge for the design of future business processes. 

9  Patterns for Visualizing the Aesthetic Qualities of Business… 

303

Instructions for Managing Complexity  Visualizing guided by the concept of aesthetic business processes made students curious, activated them to learn together, partly on the basis of “productive confusion.” In her book The spark of learning, Sarah Rose Cavanagh explains that “the highest levels of interest, flow, and productive confusion will occur if you manage to navigate your students along a tight channel between the highly complex and the challenging” (p. 141). Therefore, in order to concretize the macro-sequence of the concept of aesthetic business processes outlined so far, a better understanding of the interactions between learners’ emotions and their performance is necessary. Finally, confusion should motivate learners to resolve the cognitive incongruence in confusion. Confusion should not be allowed to turn into frustration, boredom and resignation and hinder learning progress (Pekrun, 2018). Aesthetic business processes require emotions, which nevertheless have a profound influence on human thought and action. Emotions control attention, shape motivation, and influence memory, self-regulation and the use of problem-solving strategies (Pekrun, 2018). The learning with the students has further shown that the complexity of business processes must be manageable in visualization. Formal modelling languages use different levels of abstraction for this purpose. For example, Alistair Cockburn in his book Creating Use Cases Effectively suggests at least three levels to describe process chains as use cases from the perspective of different users (Castela et al., 2001; Cockburn, 2000).  Derivation of Aesthetic Strategies for the Application of the Concept The aesthetic attitude required for the concept of aesthetic business processes means that knowledge is processed more intensively in business processes. The quote from the Russian formalist Sklovskij reflects this attitude: “Automatization eats away at things, at clothes, at furniture, at our wives, and at our fear of war […]. And art exists that one may recover the sensation of life; it exists to make one feel things, to make the stone stony” (Rosebrock, 2018, p.  19). Especially in mechanical engineering, it may be irritating if not only a cognitive-analytical but also an emotional or subjective understanding is demanded. Therefore, the concept of aesthetic business processes wants to provide strategies to stimulate an aesthetic attitude for a differentiated processing of information in

304 

M. Blattmeier

business processes and, with appropriate attention, also synesthetic experiences, so that, for example, a stone can be experienced as “stony.” The reception theorist Sven Strasen, who is oriented toward literary science, has designed a kind of process model for literary reading (Rosebrock, 2018). In it, he emphasizes the special importance of cultural knowledge in establishing coherence even in the case of uncertain references of literary texts. The surface structure and textual basis are searched more intensively for meaning-bearing elements. Non-understanding or uncertain understanding is expected and accepted and is not an obstacle. Therefore, as part of a further development of the concept of aesthetic business processes, a repertoire of strategic activities is to be built up in order to ­convey an aesthetic attitude.  The further development of the concept of aesthetic business processes, which thus aims to manage the complexity of business processes in visualization via different perspectives and to formulate strategies for an aesthetic understanding, ultimately requires a knowledge management system that no longer stores knowledge but allows it to become fluid and alive. The Pattern-Pool-Tool Visual-Bp, designed as a digital atlas, should therefore allow patterns of aesthetic business processes to continuously evolve, be made available for inspiration and decision making (Cooper, 2000; Arnold et al., 2006), and thus provide “insights into the interpersonal nature of knowledge in the practice of organizational life” (Strati, 2003). Acknowledgments  I would like to thank the members of the workshop group at EuroPLoP 2022, the organizer Tiago Sousa, and especially my shepherd Alberto Silva for his technical expertise and motivational guidance in my writing process. In addition, I am very grateful to the students who worked with me on the project and to my assistants Naeimeh Ramezanpoor and Lisa Artemiev for their collaboration on this chapter. Finally, my special thanks go to Andrew Burton-Jones, Leslie Willcocks and Nik Rushdi Hassan for introducing me into the world of information systems and developing my philosophical thinking and theorizing. Sometimes I wonder whether they understood my chapters better than I did.

9  Patterns for Visualizing the Aesthetic Qualities of Business… 

305

FundingThe project was co-financed by the European Regional Development Fund (ERDF) within the framework of the INTERREG program Germany–Netherlands.

References Alexander, C. (1979). The timeless way of building. Oxford University Press. Alexander, C., Ishikawa, S., Silverstein, M., Jacobson, M., Fiksdahl-King, I., & Shlomo, A. (1977). A pattern language: Towns, buildings, construction. Oxford University Press. Arnold, V., Clark, N., Collier, P. A., Leech, S. A., & Sutton, S. G. (2006). The differential use and effect of knowledge-based system explanations in novice and expert judgment decisions. MIS Quarterly, 30(1), 79. https://doi. org/10.2307/25148718 Bauer, R., & Baumgartner, P. (2012). Schaufenster des Lernens. Eine Sammlung von Mustern zur Arbeit mit E-Portfolios. Waxmann Verlag. Bergmann, F. (1977). On being free. University of Notre Dame Press. Blackler, F., & McDonald, S. (2000). Power, mastery and organizational learning. Journal of Management Studies, 37(6), 833–852. https://doi. org/10.1111/1467-­6486.00206 Blattmeier, M. (2020). Visuelles Rekonstruieren von Geschäftsprozessen. In L.  Nadig, & U.  Egle (Eds.), Controlling. Accounting & Audit. Risk & Compliance. Finanzen (pp. 401–411). Zug, Verlag IFZ-Hochschule Luzern. Available at: www.hslu.ch/-­/media/campus/common/files/dokumente/w/ifz/ seminare-­k onferenzen/carf/konferenzbeitraege-­2 020/lehre/visuelles-­ rekonstruieren-­von-­geschftsprozessen-­blattmeier.pdf?la=de-­ch. Blattmeier, M. (2021). Towards a structure of a pattern language for visualising in business processes. In Hillside Europe e.V. (Ed.), Proceedings of the 26th European Conference on Pattern Languages of Programs. European Conference on Pattern Languages of Programs 2021. https://dl.acm.org/ doi/10.1145/3489449.3489983. Böhme, G. (2016). Ästhetischer Kapitalismus. Suhrkamp Verlag. Bourne, M., Melnyk, S., & Bititci, U. S. (2018). Performance measurement and management: Theory and practice. International Journal of Operations & Production Management, 38(11), 2010–2021. https://doi.org/10.1108/ ijopm-­11-­2018-­784

306 

M. Blattmeier

Castela, N., Tribolet J., Silva A., & Guerra, A. (2001) Business process modelling with UML. In Proceedings of the International Conference on Enterprise Information Systems ICEIS (pp. 679–685). Cockburn, A. (2000). Writing effective use cases. Addison-Wesley Professional. Cooper, R. B. (2000). Information technology development creativity: A case study of attempted radical change. MIS Quarterly, 24(2), 245. https://doi. org/10.2307/3250938 Davenport, T. H. (1993). Process innovation: Reengineering work through information technology. Harvard Business Press. Debes, J. (1970). The loom of visual literacy. In C. M. Williams & J. L. Debes (Eds.), Proceedings of the first national conference on visual literacy (pp. 1–16). Pitman Publishing Corporation. Dewey, J. (1934). Art as experience. Penguin. Durcikova, A., Lee, A.  S., & Brown, S.  A. (2018). Making rigorous research relevant: Innovating statistical action research. MIS Quarterly, 42(1), 241–263. https://doi.org/10.25300/misq/2018/14146 Fischer, G. (2021, October). Zu mir, zu dir oder ins Büro. brandeins, 3. Flamholtz, E., & Randle, Y. (2011). Corporate culture: The ultimate strategic asset. Stanford University Press. Foelsing, J., & Schmitz, A. (2021). New Work braucht New Learning: Eine Perspektivreise durch die Transformation unserer Organisations- und Lernwelten. Springer Gabler. Franken, R., & Franken, S. (2020). Wissen, Lernen und innovation im digitalen Unternehmen: mit Fallstudien und Praxisbeispielen. Springer-Verlag. Gendlin, E. T. (1997). Experiencing and the creation of meaning: A philosophical and psychological approach to the subjective. Northwestern University Press. Groysberg, B., Lee, L., Price, J., & Yo-Jud Cheng, J. (2018, January–February). The Leader’s guide to corporate culture. Harvard Business Review, 96, 44–52. Healey, M. & Jenkins, A. (2008, October). Developing students as researchers. UC Magazine, 17–19. Horn, C., Molitor, A., & Ziegler, A. (2022, April). Das beste Instrument zur Vernetzung ist ein Kühlschrank mit Bier. brandeins, 74. Huber, L. (2009). Warum Forschendes Lernen möglich und nötig ist. In L. Huber, J. Hellmer, & F. Schneider (Eds.), Forschendes Lernen im Studium (pp. 9–35). UniversitätsVerlagWebler. Isaacs, W. (1999). Dialogue: The art of thinking together. National Geographic Books. Juchli, E., & Wiltschko, J. (1982). Die Beziehung zwischen dem Impliziten und dem Expliziten. Available at: https://www.focusing-­netzwerk.de/fileadmin/

9  Patterns for Visualizing the Aesthetic Qualities of Business… 

307

daten/focusing/PDFs/Literatur/3_Theorie_und_Praxis_von_Focusing_ und_Psychotherapie.pdf. Kankanhalli, A., Tan, B. C. Y., & Wie, K.-K. (2005). Contributing knowledge to electronic knowledge repositories: An empirical investigation. MIS Quarterly, 29(1), 113. https://doi.org/10.2307/25148670 Kohls, C., & Wedekind, J. (2008). Die Dokumentation erfolgreicher E-Learning-Lehr−/Lernarrangements mit didaktischen Patterns. In S. Zauchner, P. Baumgartner, E. Blaschitz, & A. Weissenbäck (Eds.), Offener Bildungsraum Hochschule (pp. 217–222). Waxmann Verlag. Kohls, C., & Wedekind, J. (2009). E-Learning-Patterns. Nutzen und Hürden des Entwurfsmusteransatzes. In N.  Apostolopoulos, H.  Hoffmann, V.  Mansmann, & A.  Schwill (Eds.), E-Learning 2009. Lernen im digitalen Zeitalter (pp. 61–72). Waxmann Verlag. Kricke, M., & Reich, K. (2016). Teamteaching: Eine neue kultur des Lehrens und Lernens. Beltz Verlagsgruppe. Mieville, P. (2022, March). Zu mir, zu dir oder ins Büro. form, 295, 130–135. Miketta, M. (2018). Thinking environment: Denkräume schaffen in coaching und Beratung. Junfermann Verlag GmbH. Morgan, G. (1986). Images of organization. SAGE Publications. Normann, R. (2001). Reframing business: When the map changes the landscape. John Wiley & Sons. Papadopoulos, J. (2021). Embodied design in Innovation und Kommunikation: Durch erlebensbezogenes Denken neue Handlungsräume entdecken  – Theorie und praxis. Springer Gabler. Pekrun, R. (2018). Emotion, lernen und leistung. In M. Huber & S. Krause (Eds.), Bildung und emotion (pp. 215–230). Springer-Verlag. Pyrko, I., Dörfler, V., & Eden, C. (2019). Communities of practice in landscapes of practice. Management Learning, 50(4), 482–499. https://doi. org/10.1177/1350507619860854 Roam, D. (2008). The back of the napkin: Solving problems and selling ideas with pictures. Penguin. Rosebrock, C. (2018). Strategien des ästhetischen Lesens. Literarisches Lernen in rezeptionsästhetischer Perspektive. In D. Scherf & A. Bertschi-Kaufmann (Eds.), Ästhetische Rezeptionsprozesse in didaktischer Perspektive (pp. 14–26). Beltz Juventa. Roß, J. (2020). Bildung – Eine Anleitung. Rowohlt Verlag GmbH. Schein, E. H. (2016). Organizational culture and leadership. John Wiley & Sons. Schlaeger, J., & Tenorth, H. (2020). Bildung durch Wissenschaft: Vom Nutzen forschenden Lernens. Berliner Wissenschafts-Verlag.

308 

M. Blattmeier

Schmelzer, H. J., & Sesselmann, W. (2020). Geschäftsprozessmanagement in der Praxis: Kunden zufrieden stellen - Produktivität steigern - Wert erhöhen. Carl Hanser Verlag GmbH Co KG. Seels, B. A. (1994). Visual literacy: The definition problem. In D. M. Moore & F. M. Dwyer (Eds.), Visual literacy. A spectrum of visual learning (pp. 97–112). Educational Technology Publications. Spender, J. (1996). Organizational knowledge, learning and memory: Three concepts in search of a theory. Journal of Organizational Change Management, 9(1), 63–78. https://doi.org/10.1108/09534819610156813 Strati, A. (2003). Knowing in practice: Aesthetic understanding and tacit knowledge. In D. Nicolini, S. Gherardi, & D. Yanow (Eds.), Knowing in organizations: A practice-based approach (pp. 53–75). Sharpe. Suter, A., Vorbach, S., & Wild-Weitlaner, D. (2019). Die Wertschöpfungsmaschine Prozesse und Organisation aus der Strategie ableiten. Carl Hanser Verlag GmbH Co KG. Takashi, I. (2016). Pattern language 3.0 and fundamental behavioral properties. In P. Baumgartner, T. Gruber-Muecke, & R. Sickinger (Eds.), Pursuit of pattern languages for societal change (pp. 200–233). epubli. Turner, M. (1998). The literary mind: The origins of thought and language. Oxford University Press. Wafi, S., & Wirtz, M.  A. (2015). Visualisierungskompetenz in didaktischen Kontexten: Eine Einführung. Springer-Verlag. Wenger, E., McDermott, R. A., & Snyder, W. (2002). Cultivating communities of practice: A guide to managing knowledge. Harvard Business Press. Wörwag, S., & Cloots, A. (2021). Arbeitskulturen im Wandel: Der Mensch in der New Work culture. Springer Gabler. Zeller, J. (2016). Visualisierung als methode des forschenden lernens. In D. Kergel & B. Heidkamp (Eds.), Forschendes lernen 2.0 (pp. 129–145). Springer.

10 Information Theory in IS Earl McKinney

and Sebastian K. Boell

Introduction This chapter organizes and describes the existing landscape of theories of information. We then suggest what a theory or theories of information could look like to address IS research interest. We do this by looking at concepts frequently related to information. Finally, the chapter identifies possible avenues for future research in IS, in particular by recognizing that information is not a static or singular concept or phenomenon but changing depending on situational, temporal, and cultural contexts, as well as epistemological assumptions and domain of application. As a

E. McKinney (*) Allen W. and Carol M. Schmidthorst College of Business, Bowling Green State University, Bowling Green, OH, USA e-mail: [email protected] S. K. Boell University of Sydney Business School, Darlington, NSW, Australia e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. P. Willcocks et al. (eds.), Advancing Information Systems Theories, Volume II, Technology, Work and Globalization, https://doi.org/10.1007/978-3-031-38719-7_10

309

310 

E. McKinney and S. K. Boell

result, we believe theories of information written by IS scholars will be more generally applicable for use beyond the IS domain and provide analytic distinctions that can make a difference in everyday practice. While essential to many everyday practical business domains, information is also the sine qua non of information systems. Information can be a theoretical glue to help branches of IS relate to each other to form a whole (Boell, 2017). A clear understanding of information benefits both IS research and practice and the wedding of the two (McKinney & Yoos, 2019). Absent a clear theoretical foundation, information has become an impossibly broad term that covers a multitude of ideas; the term is sloppy, disjointed, and without definition (McKinney & Yoos, 2010) a semantic chameleon (Thom, 1975) and taken for granted (Beynon-Davies, 2009). It has various underlying assumptions (Boland, 1987; Checkland & Holwell, 2006) that when made clear are incommensurable (Boell, 2017). While much research over the past 40  years has identified the need for a theory of information home to IS, little research has engaged the topic (Baskerville & Myers, 2002; Boell, 2017; Kohli & Grover, 2008; Lee, 2010). Thus, while foundational to future IS research and practice, information is under theorized in IS and has become disjointed with no common theory or definition. The term has become reified (McKinney & Yoos, 2010)—it is frequently used but almost always unspecified, its assumptions rarely addressed, and it is applied to an unbounded variety of problems. One reason information is theoretically underdeveloped is that it is so common in the practice of everyday use. It has become an umbrella covering any conceivable content (Travica, 2011). Over the past decades only a few definitions have been offered, and the assumptions underlying these definitions have not been well vetted (Boell & Cecez-Kecmanovic, 2015). This wide variety in everyday use has made development of applicable theories of information a challenge. For another view of information, see also Chaps. 4 and 6. This chapter contributes to the book by elaborating the concept of information as a product of theorizing.

10  Information Theory in IS 

311

Etymology and History of “Information” Information is a foundational term for IS and an essential concept in many related domains. Information’s etymology offers a useful starting point to any debates about theories of information. While IS is still relatively a new field, information has a long history. Information has been an essential characteristic of all human communities over the past 50,000 years (Beynon-Davies, 2009; Chandler & Cortada, 2000; Gleick, 2011; Wright, 2007). The idea behind the label information can be ultimately traced to the Greek term forma and Latin informatio and informare, meaning reinforcement or detection: to make or produce both technically and biologically, to describe or define, or to unfold or illuminate, to put into form (Applegate et al., 1999; Capurrom & Hjorland, 2003; Haken, 1992; Marcus, 2000). According to Virgil (70–19 B.C.), Zeus lightning bolts were hammered out (informatum) by Vulcan and the Cyclops. Cicero (106–43 B.C.) used informare to describe the process of instructing or improving something, in particular the mind by imposing a form on it (Borgmann, 1999). Later, Tertullian (160–22 A.D.) called Moses the populi informatory: the people’s educator or molder (Capurrom & Hjorland, 2003). Augustine described information as a molding process like the representation or impression of a ring on wax and called the process of visual perception informatiosensus (Capurrom & Hjorland, 2003). In medieval times information continued to have an active, constructive meaning as something which gives form or character to matter or to the mind (Campbell, 1982). This interpretive and constructive flavor that emphasizes meaning and understanding continues to this day. However, during the industrial age, the demands to create communication networks rewarded theories that could objectively measure communication. The interpretive communication perspective of information was overtaken by the need for an information perspective that would be more suitable for measurement. Thus, Claude Shannon’s definition of information (Shannon & Weaver, 1949) as the measurable reduction in uncertainty gained prominence. Information was seen from an objective communication perspective as a selection of signals (Capurrom & Hjorland, 2003). This definition

312 

E. McKinney and S. K. Boell

divorces information from the intended meaning of communication and the meaning given by a receiver. Instead, Shannon was interested in the transmission of signals along a channel. Simply stated, according to Shannon, if a set of signals expressing absolute nonsense could be reproduced by a receiver, successful transmission of information is achieved. This objective, reduction in uncertainty definition of information became popular in engineering, mathematics, and philosophy. In this view, information is related to novelty: a less probable event is more informative as it reduces more uncertainty than a more probable event (Bar-Hillel & Carnap, 1953; Dretske, 1981). The classic example is the difference between a fair and weighted coin. Flipping a fair coin has a less predictable outcome and reduces more uncertainty; flipping a weighted coin informs us of nothing we do not already know—there is no uncertainty and thus, no information. Similarly, Karl Popper linked information to logical probability by asserting that hypotheses that are more improbable, if true, are more informative, and they communicate more knowledge. At about the same time popular use of the term information also referred to the tokens produced and processed by computing devices, information and data began to be used interchangeably.

Definitions of Information After our brief investigation of the history of the term information we continue now by considering how information is defined in common, every day, practical use. According to Merriam-Webster information is: 1a (1) : knowledge obtained from investigation, study, or instruction; (2) : intelligence, news; (3) : facts, data; b : the attribute inherent in and communicated by one of two or more alternative sequences or arrangements of something (such as nucleotides in DNA or binary digits in a computer program) that produce specific effects; c (1) : a signal or character (as in a communication system or computer) representing data; (2) : something (such as a message, experimental data, or a picture) which justifies change in a construct (such as a plan or theory) that represents physical or mental experience or another construct; d : a quantitative measure of the content

10  Information Theory in IS 

313

of information specifically : a numerical quantity that measures the uncertainty in the outcome of an experiment to be performed; 2 : the communication or reception of knowledge or intelligence. (Merriam-Webster, 2020)

As can be seen, information has a range of meanings from the transmission of data, the exchange of messages, the choice among alternatives to a subset of knowledge. Beyond Webster a number of existing definitions view information as data that has been organized and meaningful, the assembly of data in comprehensive form, patterns of organization of matter and energy (Bates, 2005), and data that has been processed. Others argue information transforms data into insight as information is organized and meaningful data (Chaffey & Wood, 2005; Hislop, 2005), or that information leads to meaning that influences decisions (Davis & Olson, 1985). More broadly, information is also defined as self-organized complexity based on physical, biological, and human domains (Bawden, 2007). In IS textbooks information is defined in a variety of ways. Information is knowledge derived from data, data that has been shaped into form (Laudon & Laudon, 2019), facts that are organized and processed, and data in context (McNurlin et al., 2009). Not only is information inconsistently used, but other related terms are also treated similarly. For example, data, knowledge, meaning, and message are frequently used when information is defined, but these are also thinly described terms used in many different ways (Floridi, 2019). From an IS perspective a common everyday definition of information should help shed light on similarities and differences in how people, organizations, societies, and technology use information. Such a definition could be the foundation when interacting with practitioners who could apply it productively to gain understanding in common contexts as part of their day-to-day activities. A definition would be acceptable if it helps resolve applied issues and helps improve an individual’s understanding or prediction of events. For example, a definition should specifically address whether information is the same phenomenon for individuals, organizations, and machines.

314 

E. McKinney and S. K. Boell

Questions arising from the issues discussed in this section are: Do both people and machines have information, data, or both? When we talk about information how could common terms such as knowledge, shaping, and context that make sense with people be applied to technology? Are both data and information amenable to mathematical optimization and logical manipulation? Is the location of information the same for people and machines—is information outside of people and machines or does it exist also inside?

 xisting Frameworks and Taxonomies E of Theories of Information The body of literature outside of IS on information is immense. A study by Boell (2017) helps organize this landscape. Mingers and Standing (2018) as well as McKinney and Yoos (2019) also offer descriptions of many of the same theories from the critical realist and nominalist perspectives, respectively. In library and information sciences (LIS), a field like IS where information is central, the term is also being debated (e.g., Bates, 2005; Capurrom & Hjorland, 2003; Ma, 2012; Travica, 2011). Further, there are excellent reviews of the various theories of information in domains outside of IS or LIS such as Ayres (1994) or Ibekwe-Sanjuan and Dousa (2014). Finally, there are several special issues in journals: Triple (2013), two special issues of Synthese (2009, 2010), and Information Research (2010). A theory of information for IS would need to address the many ways information comes to bear in different organizational contexts. It should particularly address the key elements that make IS a unique discipline that is interested simultaneously in both social aspects and technical aspects. This interest is described variously as sociotechnical systems, soft systems (Checkland & Holwell, 1998), or the intersection of technical systems with organizational systems (Lee, 2010). One way to think about this is to describe IS as operating at the intersection of technical artifacts, social artifacts, and information artifacts (Lee et al., 2015). Another way is to describe information as a phenomenon that links the physical world to the social world (Stamper, 1992).

10  Information Theory in IS 

315

A concept of information needs to be appropriate for the ambiguous and often subjective experience of people interacting with IT to create and communicate meaning, the organizational use of information, and the often mathematical, finite, and objective domain of technology. Furthermore, given IS interest in organizations, a theory of information needs to work not only for individuals but also for collections of individuals operating in a social setting under different professional orientations and within and across different organizational settings.

More than One Theory? To begin our review, we start by asking if we are in search of one theory of information or if indeed IS may be better served by having several theories of information that allow us to highlight and investigate different phenomena and issues of interest to the field. While McKinney and Yoos (2010), Boell (2017), and Emamjome et al. (2018) propose frameworks that encompass various theories of information for different needs, Mingers and Standing (2018), and Floridi favor a single theory. Further, while McKinney and Yoos (2010) first highlighted a taxonomy of four views on information, their recent theory of information (2019) suggests a single best subjective theory within their original adaptation view. Some might argue that the reason information has become reified is that it has been applied too widely to too many different phenomena (McKinney & Yoos, 2010; Boland, 1987; Checkland & Holwell, 2006). It is hard to imagine another term as common in as many disciplines as the term information. From physics to psychology to communication to computer science, the term is widely used in both theoretical and practical domains for many different issues. This is related to theory scope. Should IS seek a grand theory of information or a mid-range theory. A grand theory seeks to explain how broadly used social structures work; grand theories typically have sweeping generalizations and are unbounded in space and time (Gregor, 2006). A mid-range theory typically addresses specific aspects of human behavior applied to specific problems or topics. Motivation, adaptation, culture, and attitude are examples of concepts often evoked by grand

316 

E. McKinney and S. K. Boell

theories, in contrast to the Technology Acceptance Model and Task-­ Technology Fit are mid-range theories widely referred to in IS.  Weber suggests IS is in need of grand theories (Weber, 2003). Another important issue affecting the breadth of a theory of information is its domain. While several of the theories presented below support a single theory of information, it is not clear if an IS theory of information that addresses our issues will be satisfactory for information issues outside of IS such as in physics or philosophy. Perhaps an IS theory of information should constrain its application to organizational and technical domains and not to natural scientific domains. Another argument for theories of information is that in IS there has been more work on taxonomies of information than on a general theory of information, indicating that there are already a number of potentially incommensurable approaches to information that are nonetheless potentially productive when thinking about IS phenomena (Boell, 2017; McKinney & Yoos, 2010). Boell (2017) thus argues for a variety of acceptable theories of information. He considers it premature and potentially stultifying to attempt to identify a singular view. Perhaps mid-range theories of information can be developed for different domains such as IT security, user behavior, organizational culture, etc., or different theories of information can be developed for different levels of application— individuals, systems, pragmatic, societal. Boell (2017) developed a consequential framework of information based on a review of information, not just within IS but based also on information use in related disciplines such as library and information science, psychology, natural sciences, philosophy, and computer science: physical information relates information to physical concepts or the physical world. Objective information relates information to human understanding and signs where information is independent of the recipient. Subjective-information grounds the concept of information in the appropriation of information by a subject. Sociocultural information dictates that wider sociocultural aspects are fundamental. In an alternative framework, McKinney and Yoos (2010) nominated four types of information which they then assessed against existing IS research. The view currently dominating IS, the token view, equates information with data. In the syntax view information is used to address

10  Information Theory in IS 

317

communication issues within an objective paradigm. The third is the representation framework which allows for both external (to the mind or machine) and internal representations. This framework employs objectivist assumptions to model systems and processes concerning how systems change state. Finally, the adaptation approach employs subjectivist assumptions and the concept of information to explain how systems react to changing environments. Questions arising from the issues discussed in this section are: How should IS best scope and frame different theories of information? How can different theories of information be productively used for IS research, education, and practice? If and how can theories of information be useful to convey insights from IS research to other disciplines?

Concepts Commonly Related to Information Even cursory reviews of information theories such as the one offered in this chapter reveal a number of common terms related to information. Most theories provide an account of data, signs, message, meaning, communication, knowledge, news, facts, change, or truth (Beynon-Davies, 2009). A theory of information should address a number of these elements by indicating their relationship to information (e.g., see Mingers & Standing, 2018; McKinney & Yoos, 2019). While there are many common terms related to information that are of relevance to IS, the two most common ones are data and signs. We now discuss at more length how these concepts are related to information. See also Chaps. 4 and 6.

Information and Data The most frequently cited framework about information in IS textbooks is the data-information-knowledge-wisdom or short DIKW based on the work of Langefors (1973) and Ackoff (1989). DIKW offers specific definitions of data, information, and knowledge forming them into a simple to use hierarchy where information is based on data, knowledge is based on information, and wisdom is based on knowledge.

318 

E. McKinney and S. K. Boell

Information is the meaning that results from data processing; in other words, information is processed data. More specifically, “Information is the meaning produced from data based on a knowledge framework that is associated with the selection of the state of conditional readiness for goal directed activities” (Kettinger & Li, 2010, p. 415). Information rests on data and is derived from data. Both information and data are objectively true measures or descriptions of objects or events, as they correspond to real or abstract objects. Data are true and all information and knowledge are built on the foundation of these truths. Information resolves questions about who, what, where, and when (Ackoff, 1989). Knowledge is the long-term organization or structure of information. Others such as Checkland and Scholes (1990) hold that data and meaning create information. In DIKW, information is not formally defined nor made mutually exclusive from data and knowledge (Kettinger & Li, 2010; Travica, 2011). Kettinger and Li recognize that knowledge will vary from person to person and as a result so must information, but they claim both knowledge and information are objective or at least intersubjective. Boell (2017) also suggests DIKW is simplistic, as there are many situations in which it is knowledge, not data that leads to information. Further, DIKW theory offers few explanations about the key terms used to describe information—meaning and context. This failure contributes to ambiguity about what information is, how meaning is judged (Travica, 2011), and if information can have different meaning to different individuals (Kettinger & Li, 2010). Floridi (2011) has examined the philosophy of information, and while not an IS view of information it is valuable to highlight. Floridi first specifies that information is declarative, objective, and semantic and then enumerates three requirements for what he calls a broad general definition of information (GDI). Information comes from data if the data are truthful, well-formed, and meaningful. Well-formed data are clustered together correctly, according to the rules (syntax) that govern the chosen system, code, or language being analyzed. Floridi provides an example of the Rosetta Stone containing Egyptian hieroglyphics, and symbols were well-formed in that they had their own rules and syntax. Syntax should be viewed inclusively (not just linguistically), as “what determines the form, construction, composition or structuring of something (engineers,

10  Information Theory in IS 

319

film directors, painters, chess players and gardeners speak of syntax in this broad sense)” (2019, p. 16–17). Floridi allows for a special case of environmental information based on data that is truthful and well-formed but not meaningful. While not meaningful, this data is patterned or correlated. A plant, animal, or mechanism can use environmental information (patterns of correlated data) that are not meaningful. Floridi argues the issue about meaning is not how meaning is created but whether the data can be described as being meaningful independently of the informee (Boell, 2017). To Floridi the hieroglyphics on the Rosetta Stone were meaningful even before they could actually be translated; they “comply with the meanings (semantics) of the chosen system, code or language in question” (Floridi, 2019 p.  17) and therefore were meaningful to the original artists and their audiences. Floridi’s work is highly influential in computer science and philosophy. However, from a systems perspective it is important to point out that for data to be considered truthful and well-formed (two basic assumptions for Floridi), we need to rely on criteria that allow us to judge truthfulness or correctness according to syntax (well-formed). As Floridi assumes that these criteria can be established when we assess data, they must come from a position where truth and syntax are given. However, once we allow for truthfulness and syntax to be part of the system of human experience, their legitimacy and correctness may be challenged and hence, they no longer offer an unshakeable foundation for defining information. McKinney and Yoos (2010) and Lee (2010) argue that IS would benefit from making a vivid distinction between data and information, referring to data as the tokens manipulated by machines or data as outside the boundary of a system. More specifically, Demetis and Lee (2019) suggest that the key to understanding information requires defining data and the boundary between information and data. Data and information are only possible to define as they relate to each other. Grounded in cybernetics and systems theory, their essay posits that data is in the environment and only becomes information to the system observing the data when the system internalizes the data. Data always entails a variety of information potentialities which are selectively reduced by the observing system.

320 

E. McKinney and S. K. Boell

Information and Signs Like data, many theories of information also relate to signs (see also Chap. 6). According to Peirce (1955) a “sign […] is something which stands to somebody for something in some respect or capacity” (p. 99). Dretske (1981) extended Pierce’s theory to information. He held that information is the dependence of the sign on the object, and this dependence provides the opportunity for learning. For example, smoke (a sign) means fire (object) because smoke is caused by (depends on) fire. The smoke is highly informative of a fire, whereas clouds are less informative about rain as the degree of dependence is less. Dependence in these examples is natural, but dependence can also be engineered as in a doorbell sound (sign) means (depends on) the button has been pressed (object) (Dretske, 1981; McKinney & Yoos, 2010). Signs are immanent in the environment (Floridi, 2011). Information to Dretske (1981) is a piece of knowledge about events or situations. It is what one learns, or can learn, from a sign where signs are events, conditions, or states of affairs. Information is based on probability values that refer to the observed states of affairs—the less likely a sign the more informative (Floridi, 2019). Information is objective. Dretske’s concept of information is largely concerned with perceptional knowledge for particular observers (Mingers & Standing, 2018). Thus, if a sign is not ever observed by an individual it is not considered information. One such example of signs and levels of information is the recent work of Mingers and Standing (2018) who offer a semiotic level taxonomy based on earlier work by Mingers. Here information has a single definition— information is the true propositional content of a sign or message, but the signs themselves can be arranged in a taxonomy. Environmental signs are events in the material world that leave traces in the form of physical differences. Syntactic level signs have a convention or coding like a map. Semantic level signs have no direct link from their referents such as resemblance or causality, like natural languages. Pragmatic information is about more than the propositional content of the message and includes emotionality of the speaker and rightness of the speech act.

10  Information Theory in IS 

321

The relationship of information and signs often introduces the idea of a semantic or higher-order concept of information; that signs are one thing, but higher-order meaning conveyed by the signs is quite another (Beynon-Davies, 2009). Much like the higher-level relationship of information to data in the last section, many recent theories of information portray levels of information.

Other Concepts Related to Information In addition to data and signs, information theories are also frequently related to knowledge, news, facts, production of effect, meaning, objectivity, and truth. Other concepts that were once common—communication, noise, quantity of information, representation, perception, and wisdom—are less central to current theories than theories a generation ago. Clearly, this list of common terms is not exhaustive as new theories of information may improve our understanding of other previously unidentified, but related, concepts. Perhaps other common terms will need to be included and a method proposed for the inclusion of new relevant common terms. Along with data and signs, these terms play important roles in the story of information. Truth  For instance, Mingers and Standing (2018) relate information to objectivity and truth, proposing and defending a semantic theory of information that holds that information is both veridical and objective. Veridical implies correctness or truth. By true they specify that information is truth-constituted, that the meaning of the data must actually be true to be information (Mingers & Standing, 2018). Objective implies that information is carried by signs and messages which are independent of observers. Information is therefore the relationship between a sign and the event(s) that caused it. It is what could be inferred about states of affairs given the sign has occurred that could not be inferred without the sign. For example, information is the true state of a company’s finances rather than those presented by deceptive accounting practices. This idea of objectivity as observer independent signs is shared with Floridi’s (Floridi, 2019) GDI but extends it to environmental and pragmatic set-

322 

E. McKinney and S. K. Boell

tings. Signs carry information even if they are never observed. Signs must be well-formed (syntax) and meaningful within a linguistic system.  Knowledge  Another concept frequently related to information is knowledge. Earlier, we discussed the DIKW hierarchy where information rests on the foundation of data and results into knowledge. Tuomi (1999) argues for a theory where the relationship is reversed. For Tuomi knowledge is shared and interpersonal and the foundation for the creation of information which is knowledge that is stripped of a specific context. In turn it is information that gives rise to data. So “data emerge last—only after knowledge and information are available” (Tuomi, 1999, p. 107). A similar argument for the importance of knowledge for the conception of information is made by Kettinger and Li (2010). In their Knowledge-­ Based Theory of Information (KBI), Kettinger and Li argue that knowledge provides the framework that underlies the process of creating information from data. Here too is knowledge foundational to the conception of information.  To be sure, other, other common terms, relevant to the discussion of information are not recounted here due to space. Interested readers are encouraged to refer to the work of Bates (2005), Boell (2017), Capurrom and Hjorland (2003), Machlup (1983), Mingers (1996), and Zins (2007). Questions arising from the issues discussed in this section are: What is the family of key concepts related to information that are of particular relevance to IS? How should proposed theories of information relate to these terms? What are productive alternative framings of information in relation to data, signs, and other concepts for IS research, education, and practice?

Information, Systems, and Change In the previous section we described terms related to information that a theory of information should address. It is our view the most effective and productive theories of information exist within a framework relating information to systems and change. In this general framework information

10  Information Theory in IS 

323

plays the role of enabling systems to change. We will not further constrain the concepts of information, system, or change, but rather suggest that a theory of information most suitable to IS, IT, business, and organizational contexts advances how systems use information to change. Businesses and society are increasingly transformed and disrupted by technology encouraging a future research agenda where IS is investigating the role of IT as part of changing sociotechnical systems (Sarker et al., 2019; Yoo, 2010). Theories of information have a long history with systems theories (e.g., Chatterjee et  al., 2017; Checkland & Holwell, 1998; Demetis & Lee, 2016, 2019). Older theories of systems and information, cybernetic theories, often addressed mechanical systems such as missiles and water containment systems emphasizing the concept of feedback for understanding how systems operate and can be regulated over time (Wiener, 1961). Later as theories of systems theories were applied more broadly to biology, individuals, and human activity, information addressed how systems learn, inform, and adapt themselves and each other (Maturana & Varela, 1980). Some theories of systems and information use a wider framework seeking to demonstrate how information affects systems at a societal level (BeynonDavies, 2009; Borgmann, 1999; Gleick, 2011; Wright, 2007). It is therefore beneficial to look more closely how the relationship between information, systems, and change can be theorized. Some initial theories for this may be cybernetic theories, difference theories, and theories of organizational change.

Cybernetics Cybernetics is the study of self-regulating systems first described by Norbert Wiener (1961). In a cybernetic system, information is what is exchanged with the environment as a system changes and makes it change felt upon the environment. A change made by the system leads to a change in the system’s environment. This change is then fed back to the system as information which the system then again actions upon. Hence an essential, if not the defining, concept of early cybernetic thinking is that of feedback. Information in cybernetics is feedback to the system that precipitates change. For more on cybernetic theory and its application see Chap. 5.

324 

E. McKinney and S. K. Boell

Cybernetic systems strive to remain viable by regulating their changes within tolerance limits in the face of perturbations from the environment impinging on the system. This principle is universal and applies to systems operating at different levels and in different domains, including physical, technical, biological, and so on. For example, the thermostatic system in a building strives to keep room temperatures at a constant level by initiating heating or cooling, based on the sensed temperature. Moreover, the concept of nesting of systems in super-systems is another essential idea in cybernetic thinking. Every cybernetic system is contained within a super-system, controlling the subsystem by regulating the set of possible actions of the subsystem. Second-order cybernetics emphasizes recursion by looking at adaptation of observing systems, not the observed system. The first-order observed systems, a guidance system for an aircraft and a student, receive feedback from the environment and change their actions. Second-order systems, pilots and teachers, observing first-order systems, also receive feedback and change. In the example of the building thermostatic system, the second-order cybernetic system is the human building occupant who regulates the reference room temperature, warmer when present for comfort, and cooler when absent for energy savings. Checkland’s Soft Systems Methodology (Checkland & Holwell, 1998) continued the effort to place the idea of information within systems theory first initiated by cybernetics. Soft systems inquire about the world and create their own subjective (soft) information. This hierarchy of systems is evident in Checkland’s self-referencing systems: “we think about ourselves thinking about the world and we change” (Checkland & Holwell, 1998, p.  99). This is the same second-order system as Soren Kierkegaard’s description of the self as a being relating itself to itself (Kierkegaard, 2004). Brier’s cybersemiotics (2005) uses information to address communication among adapting systems. Semiotics is the study of signs and human communication. Self-cognition, the reality that individuals perceive and create, is what each person communicates to others with the signs of language. Brier argues that such interpersonal communication is best described as “a dance of mutual structured couplings” (p. 379) that develops between autopoietic systems, and suggests that this dance is important in distinguishing human social systems from more primitive ones (e.g., cells). Through mutual dancing each system perceives its partner’s

10  Information Theory in IS 

325

dance steps and adapts. Communication is an ensemble of perceivable signs around which the autopoietic systems have danced in concert long enough to enable them to create information and adapt.

Difference McKinney and Yoos (2019) offer a subjective theory of information based on differences. Their theory of information holds that differences perceived by individuals lead to a change in mind. More specifically, perceptions of difference in the environment are conceived as information where information is a changing mind. Based on a nominalist, anti-­positivist foundation, the difference theory explains how an individual creates information in uncertainty using criteria, criteria meta levels, and recursion. The difference theory makes a sharp distinction between data and information by reserving the term data to mean tokens. This difference theory draws from the work of Bateson who proposed an ecological philosophy based on a cybernetic epistemology of self-­organization and self-regulation. In this work Bateson defined information as any difference that makes a difference. Similar to cybernetics Bateson’s epistemology understands the cycle between a system and its environment, producing an ongoing feedback loop of information that changes the system over time. Anything that a system is capable of noticing in the environment is a difference. If this difference becomes relevant to the system to act upon, perhaps the difference is between a goal and a current state, the system will take action, hence the difference “makes a difference.” While Bateson applied this epistemology across a wide spectrum of domains, McKinney and Yoos’ (2019) difference theory only addresses a specific act—how an individual’s mind is changed in an environment of ambiguity.

Organizational Change One area of interest to IS is IT-related organizational change (e.g., Kumar et al., 2016; Orlikowski, 1996; Volkoff et al., 2007). IT plays a major role in how organizations are operating, and one may even argue that nowadays IT is involved in all aspects of organizational change and in many

326 

E. McKinney and S. K. Boell

cases, IT is the driving factor for such change. Early theorizing of IT on organizations either saw IT as determining organizational change or regarded social forces as determining how IT is adopted (Orlikowski, 2007). However, over the last two decades IS research has increasingly realized that neither idea is capturing the mutual shaping or entanglement of IT and the social (Orlikowski & Scott, 2008). There is currently a reorientation in IS happening that argues that IS has much to offer understanding technological change by taking a sociotechnical perspective (Sarker et al., 2019). At the same time, there is also an emerging realization in IS that information offers an important middle ground stretching on the one hand to the physical world of technological devices and on the other hand to the social world of socially constructed and shared meaning (Boell & Cecez-­ Kecmanovic, 2015; Stamper, 1992). Theories of information that predominantly consider technological aspects, such as conceptions building on Shannon and Weaver (1949), fall short on the side of IS’s interest in considering social aspects associated with technology. A similar shortfall in considering social aspects can also be made about many conceptions that focus on the relationship of information to data or cognitive conceptions where information exists in the mind of an observer. Most current theories thus omit the wider social context within which information is captured, created, shared, and consumed. From a social-constructivist point of view, when theorizing what information is, however, social context cannot be treated as merely a background or context as it plays an integrative role in shaping what is and indeed can even be perceived and considered to be(come) information. Conceptions of organizational change could therefore draw from conceptions of information that take into consideration social as well as technological aspects. Development of such conceptions of information will require research approaches that can capture both technological as well as social aspects and how these are changing or shaping each other over time. To unpack the sociotechnical nature of information thus requires historical or longitudinal methodological approaches providing empirical evidence that enables such analysis. So far historical research methods (Porra et al., 2014) are not widely used in IS research and when employed they are mainly used for capturing the adoption of technology into

10  Information Theory in IS 

327

specific organizations (McKenney et al., 1997; Porra et al., 2005), industries (McKenney et al., 1995; Reimers et al., 2013), or to describe how IT is adopted by organizations over longer time frames (Nolan, 2000). There is therefore an opportunity and need for IS to engage with historical research methods to better understand the sociotechnical nature of information. As IS seeks to guide organizations in their use of IT, the long-term consequences of sociotechnical systems on information have much to offer. Theoretical lenses that may be useful for advancing such undertakings may, for example, draw from process philosophical accounts (Tsoukas & Chia, 2002) or infrastructure studies (Star & Ruhleder, 1996). In particular, there is already an existing research stream in IS looking at information infrastructures often from a sociotechnical perspective (Hanseth & Lyytinen, 2016). A first study unpacking sociotechnical organizational change from an infrastructure perspective is Boell and Hoof (2020), where the authors demonstrate that organizational structures, standards, and information technology over time shape what information is within a particular organizational setting. This study therefore furthers a conception of information as sociocultural (Boell, 2017), where information is shaped and made possible on the context of an organizational setting, standardizations of data, and emerging technological possibilities. Questions arising from the issues discussed in this section are: How is information situational, by which we mean changing or varying across time and space? How is information related to systems, recursion, and feedback? How and what types of differences can be considered to be information? How does a sociotechnical conception of information relate to IT? How can information be conceptualized and reveal important insights in relation to wider consideration of societal and organizational changes?

 xemplifying Variations in Existing E Information Theories As highlighted in McKinney and Yoos (2019), one way to better grasp differences in information theories is to use a common example. A timetable listing the arrival times of trains can be used to contrast how

328 

E. McKinney and S. K. Boell

theories of information view concepts related to information and how such concepts may be regarded as irrelevant or important. In DIKW (Ackoff, 1989), arrival times are data, and information results from processing that data as well as from answers to “who,” “what,” “how many,” and “when” types of questions. To Floridi, information is data that is true, well-formed, and meaningful. A timetable constitutes information if the data contained is true, and it is well-formed (the timetable appears in an appropriate format) and is about something, in this case train arrivals (rendering it meaningful). For cybernetic theories the timetable is information as it provides a desired state of a system, deviations from which the system of railroads would attempt to correct via feedback to conductors and planners. The timetable and a current clock would be information to conductors and passengers to use to adjust their behavior to meet the train or the scheduled arrival on time. To McKinney and Yoos (2019), the timetable is data. Observers of the timetable conceive their own information from differences they perceive in the timetable. One observer creates information that the train he is waiting on is late, another person that no trains arrive from Lancaster after 10:00 pm, and another that there are more arrivals than she previously thought. To Demetis and Lee (2019) the timetable is also data until a traveler selectively reduces the information potential of the timetable to create his/her own information. For sign theories such as Dretske, the timetable is containing signs that if corresponding to reality contain information about trains to an observer. Likewise, in Mingers and Standings’ veridical theory (2018), information is carried in the timetable about the arrival of trains only if the time provided in the timetable reflects a true statement about the world. Therefore, a misspelled town name or a misprint in the arrival time will disqualify these entries from being considered as information. For Kettinger and Li the timetable is also data; it can only become information if knowledge is applied to the timetable, for example, the knowledge how timetables are organized in columns and rows, or that numbers represent a particular time of the day. A sociocultural or sociotechnical (Boell, 2017) understanding of the timetable points wider social context or infrastructures that are necessary for a timetable to be considered information. The timetable only can contain information because of

10  Information Theory in IS 

329

socially shared meanings involving standardized time and adherence to accepted way of presenting the flow of time, places, and trains organized in rows and columns. Questions arising from the issues discussed in this section are: How can we present, compare, and contrast different theories of information with each other? What will be disclosed, made visible but also disregarded when different theories of information are applied to a single case study? What are means to choose an appropriate theory of information for particular research interests?

Philosophical Assumptions and Validity Philosophical assumptions have a significant impact on any theorizing, information theories are no exception. Differences in philosophy impact the questions a theory of information will address and the types of research conducted. Over the last few decades IS has developed and started to appreciate a variety of dominant philosophical paradigms contributing to the diversity and vibrancy of the IS research community. On the flip side, as the discipline does not share a single philosophical starting point, parts of the IS community will necessarily judge, evaluate, and appreciate information theories differently. It is therefore important that researchers are aware of these differences when they develop and present their theories and findings to the IS community. Making this issue more challenging is that authors often do not specify and consistently describe the philosophy that underpins their conception of information. Also, individual authors and theories may not neatly fall into categories described by different philosophical taxonomies or frameworks used to describe information. For instance, Mingers’ (2004) use of critical realism does not have an easy classification on McKinney and Yoos objective/subjective dimension. Likewise, Floridi’s theory also does not directly express how it should be classified as his particular notion of meaning as being objectively contained in physical inscriptions can itself be challenged on philosophical grounds, as others argue that the existence of meaning necessarily requires a subject.

330 

E. McKinney and S. K. Boell

Differences in philosophical assumptions become immediately apparent when information theories describe the environment. One fundamental question is whether the environment conveys information through messages and signs that are objective, or if an individual subjectively creates information from perceptions of the environment. Is information therefore objective, existing independently of observers or receivers, or is information subjective, created in the mind of observers on receipt of a message, or the perception of a difference? These varying philosophical viewpoints also impact our debate on terms related to information like knowledge, as answers to the question, what can be known for sure, vary based on epistemological stance (Boell, 2017). Can some knowledge, or information, be known to be true, or does all knowledge include an aspect of ambiguity or uncertainty? As with discussions about the environment, discussions about truth highlight different assumptions. While some argue that in contrast to misinformation, information necessarily has to be true (e.g., Dretske, 1981; Floridi, 2019), others point out the inherent difficulty of establishing what ultimately can be considered to true. These differences in philosophical foundations lead not only to differences in theories but also to different criticisms of the various theories. To a subjectivist, the objectivist’s assumptions about truth and meaning are hard to accept. Subjectivists argue that in practical experience we notice that information end users get from applications is different and largely dependent on the situation, goals, motivations, and attention of individual users. Similarly, an objectivist finds it difficult to understand why a subjectivist is willing to allow obviously untrue statements to be regarded as information and why unperceived environmental information such as the rings of a tree should not be regarded as information. One may argue that it is simplistic to use the philosophical taxonomy of objective and subjective. While many other philosophical distinctions exist as well, we consider this difference to be representative of the distinctions that lead to the variety of information theories and how these differences impact a theory of information. While philosophical differences of information theories will always be profound, future research will benefit from clear expression of philosophical assumptions. Absent an explicit philosophical commitment, it is difficult to understand

10  Information Theory in IS 

331

similarities and differences in theories. Extant theories of information are committed to various epistemological assumptions. Some of these assumptions are clearly stated, while others are not. Research that systematically identifies assumptions of existing and novel information theories will be helpful in better understanding differences in assumptions and their implications for IS research. In addition, future theories should also specify their ontology and epistemology, and the implication of a theory in relation to truth, environmental information, intersubjective agreement, and meaning.

Validation As with all theory development, a key issue is how information theories are validated. Gregor (2006) as well as others in this volume have presented criteria for IS theory validation. Existing theories of information vary in their validation, and several do not specify a method of validation. McKinney and Yoos (2010) explicitly suggest criteria for validation of information theories. They suggest that criteria for judging if an information theory is valid could be: Is the conception of information related to common use (dictionary) and therefore congruent with intuitive use in practice? Does a theory advance the conception and use of existing theory and thereby generates subsequent application for theory? Is a theory parsimonious and therefore as succinct as possible in terms or constructs or concepts involved that are necessary to express the theory? Can a theory of information clearly discriminate by allowing noninformation phenomena to be clearly rejected? And finally, can a theory account for events and thereby help us understand a variety of phenomena in our environment? Again, criteria for theory validation depend on assumptions about the scope and applicability of the theory, philosophical commitments, and the nature of the phenomena. As with epistemological assumptions, there seems no accepted method for identifying which criteria are most salient for validation. Although the means of validation vary, it is incumbent on future authors of information theory to identify a means of its own validation.

332 

E. McKinney and S. K. Boell

Finally, a validated theory of information for IS should be applicable to practice by offering important insights or observable constructs for the workplace. While some theories may be philosophically or conceptually driven, a modified version of the theory could be created that readily addresses issues apparent to practitioners. “Translating” proposed theories of information into versions that are readily applicable in practice should also be considered as a worthy effort by IS scholars. Questions arising from the issues discussed in this section are: What is the ontological status of information—for example, is it a thing, a concept, or a relation, and what are the implications arising from this? Is it desirable to have common criteria to evaluate, assess, or corroborate theories of information? If so, what could be such criteria and under what circumstances are they appropriate to be applied?

Conclusions and Ways Forward Information is a topic relevant to many academic domains in IS such as decision making, networking, security, analytics, and database. Information is also an essential concept for the practice of IS. Theories of information would help IS better understand what information is, who can have it, if it can be transmitted or stored, how it can be secured, and whether it can be false or misleading. Most of the current theories in information both within and outside of IS were developed prior to the advent of recent key developments in IS, as argued in Chap. 4. These developments—cloud, AI, security, privacy, and analytics—are rarely discussed as examples of information use and were beyond the scope of this chapter. Instead, the bulk of applications of information theory are about decision making. Future work might compare existing theories of information using examples of information from these new developing areas. Finally, some theories are limited by design to certain applications while others attempt to be more universal. If the future is several theories, perhaps they should be more clearly named such as subjective-­information, true information, or information as difference. Other examples of specific naming might include information-perception or external-

10  Information Theory in IS 

333

i­nformation to imply that an element of perception is being considered; or information-meaning or internal-information to imply the aspect of information that is unique to different individuals. We could also assign a prefix like pragmatic information or social information to identify what level of information we are describing. Other domains employ such as labelling scheme—intrinsic and extrinsic motivation, self-control, and non-verbal communication. For many years the term and ideas behind information were poorly theorized within IS. However, the tide may be changing as researchers have identified the need to better our understanding of information (Boell, 2017; Boell & Cecez-Kecmanovic, 2015; Kettinger & Li, 2010; McKinney & Yoos, 2010). Recently, several definitions and theories for the term have emerged in the IS literature (e.g., Demetis & Lee, 2019; Lee, 2010; McKinney & Yoos, 2019; Mingers & Standing, 2018). The intent of this chapter was to organize and describe the landscape of information theory and aspects covered by these theories to advance theorizing of information in information systems.

References Ackoff, R. (1989). From data to wisdom. Journal of Applied Systems Analysis, 16, 3–9. Applegate, L., McFarlan, F. W., & McKenney, J. (1999). Corporate information systems management. Irwin McGraw-Hill. Ayres, R. (1994). Information, entropy, and progress. AIP Press. Bar-Hillel, Y., & Carnap, R. (1953). Semantic information. British Journal for the Philosophy of Science, 14(4), 147–162. Baskerville, R., & Myers, M. (2002). Information systems as a reference discipline. MIS Quarterly, 26, 1–14. Bates, M. (2005). Information and knowledge: An evolutionary framework for information science. Information Research: An International Electronic Journal, 10(4), n4. Bawden, D. (2007). Information as self-organized complexity: A unifying viewpoint. Information Research, 12(4), colis31. Beynon-Davies, P. (2009). Neolithic informatics: The nature of information. International Journal of Information Management, 29, 3–14.

334 

E. McKinney and S. K. Boell

Boell, S. (2017). Information: Fundamental positions and their implications for information systems research, education and practice. Information and Organization, 27(1), 1–16. Boell, S., & Cecez-Kecmanovic, D. (2015). What is ‘information’ beyond a definition? ICIS 2015 Proceedings, paper 1363. Retrieved from http://aisel. aisnet.org/icis2015/proceedings/ConferenceTheme/4/. Boell, S., & Hoof, F. (2020). Accounting for information infrastructure as medium for organisational change. Accounting History Review, 30(1), 45–68. Boland, R. (1987). The in-formation of information systems. In K. Lyytinen, R.  Boland, & R.  Hirschheim (Eds.), Critical issues in information systems Research. John Wiley and Sons. Borgmann, A. (1999). Holding on to reality: The nature of information at the turn of the millennium. University of Chicago Press. Brier, S. (2005). The construction of information and communication. Semiotica, 154, 355–399. Campbell, J. (1982). Grammatical man, information, entropy, language, and life. Simon and Schuster. Capurrom, R., & Hjorland, B. (2003). The concept of information. Annual Review of Information Science and Technology, 37, 343–411. Chaffey, D., & Wood, S. (2005). Business information management: Improving performance using information systems. Pearson Education Limited. Chandler, A., & Cortada, J. (2000). A nation transformed by information: How information has shaped the United States from colonial times to the present (in Chandler A.D. and J. W. Cortada, eds.). : Oxford University Press. Chatterjee, S., Xiao, X., Elbanna, A., & Sarker, S. (2017). The information systems artifact: A conceptualization based on general systems theory. Proceedings of the 50th Hawaii International Conference on Systems Sciences, 5717–5726. Checkland, P., & Holwell, S. (1998). Information, systems, and information systems. John Wiley and Sons. Checkland, P., & Holwell, S. (2006). Data, capta, information and knowledge. In M.  Hinton (Ed.), Introducing information management the business approach. Elsevier. Checkland, P., & Scholes, J. (1990). Soft systems methodology in action. Wiley. Davis, G., & Olson, M. (1985). Management information systems: Conceptual foundations, structure, and development. McGraw-Hill. Demetis, D., & Lee, A. (2016). Crafting theory to satisfy the requirements of systems science. Information and Organization, 26(4), 116–126.

10  Information Theory in IS 

335

Demetis, D., & Lee, A. (2019). An observer-relative systems approach to information. Proceedings of the 52nd Hawaii International Conference on System Sciences, 6300–6309. Dretske, F. (1981). Knowledge and the flow of information. MIT Press. Emamjome, F., Gable, A., Bandara, W., & Gable, G. (2018). Re-thinking the ontology of information. In International Conference on Information Systems (ICIS 2018), 13–16 December 2018, San Francisco, CA. Floridi, L. (2011). The philosophy of information. Oxford University Press. Floridi, L. (2019). Semantic conceptions of information, the Stanford encyclopedia of philosophy, Edward N.  Zalta (ed.). https://plato.stanford.edu/ archives/sum2019/entries/information-­semantic/ Gleick, J. (2011). The information: A history, a theory, a flood. Pantheon Books. Gregor, S. (2006). The nature of theory in information systems. MIS Quarterly, 30(3), 611–642. Haken, H. (1992). The concept of information seen from the point of view of physics and synergetics. In K. Haefner (Ed.), Evolution of information processing systems: An interdisciplinary approach for a new understanding of nature and society (pp. 153–168). Springer. Hanseth, O., & Lyytinen, K. (2016). Design theory for dynamic complexity in information infrastructures: The case of building internet. Enacting Research Methods in Information Systems, 3, 104–142. Hislop, D. (2005). Knowledge management: A critical introduction. Oxford University Press. Ibekwe-Sanjuan, F., & Dousa, T. (2014). Theories of information, communication and knowledge. Springer. Information Research (2010). Special Issue. Volume 15(4), December. Kettinger, W., & Li, Y. (2010). The infological equation extended: Towards conceptual clarity in the relationship between data, information, and knowledge. European Journal of Information Systems, 19, 400–421. Kierkegaard, S. (2004). The sickness unto death: A Christian psychological exposition of edification and awakening by anti-Climacus. Penguin. Kohli, R., & Grover, V. (2008). Business value of IT: An essay on expanding research directions to keep up with the times. Journal of the Association for Information Systems, 9(1), 1–10. Kumar, V., Loonam, J., Allen, J., & Sawyer, S. (2016). Exploring enterprise social systems & organisational change: Implementation in a digital age. Journal of Information Technology, 31(2), 97–100.

336 

E. McKinney and S. K. Boell

Langefors, B. (1973). Theoretical analysis of information systems (4th ed.). Auerbach Publishers. Laudon, L., & Laudon, J. (2019). Management information systems: Managing the digital firm, global edition. Pearson Education Limited. Lee, A. (2010). Retrospect and prospect: Information systems research in the last and next 25 years. Journal of Information Technology, 25, 336–348. Lee, A., Thomas, M., & Baskerville, R. (2015). Going back to basics in design science: From the information technology artifact to the information systems artifact. Information Systems Journal, 25(1), 5–21. Ma, L. (2012). Meanings of information: The assumptions and Research consequences of three foundational LIS theories. Journal of the American Society for Information Science, 63(4), 716–723. Machlup, F. (1983). Semantic quirks in studies of information. In The study of information: Interdisciplinary messages (pp. 641–671). John Wiley & Sons. Marcus, S. (2000). The evolution of the concepts of information and communication. http://ir2.ici.ro/ictpa/ictpa2000/marcus.html. Merriam-Webster. (2020). Information. [online] https://www.merriam-­webster. com/dictionary/information. Maturana, H., & Varela, F. (1980). Autopoiesis and cognition: The realization of the living. Riedel. McKenney, J., Copeland, D., & Mason, R. (1995). Waves of change: Business evolution through information technology. Harvard Business School Press. McKenney, J., Mason, R., & Copeland, D. (1997). Bank of America: The crest and trough of technological leadership. MIS Quarterly, 21(3), 321–353. McKinney, E., & Yoos, C. (2010). Information about information: A taxonomy of views. MIS Quarterly, 34(2), 329–344. McKinney, E., & Yoos, C. (2019). Information as a difference: Toward a subjective theory of information. European Journal of Information Systems, 29, 1–15. McNurlin, B., Sprague, R., & Bui, T. (2009). Information systems management in practice (8th ed.). Pearson. Mingers, J. (1996). An evaluation of theories of information with regard to the semantic and pragmatic aspects of information systems. Systems Practice, 9(3), 187–209. Mingers, J. (2004). Real-izing information systems: Critical realism as an underpinning philosophy for information systems. Information and Organization, 14(2), 87–103. Mingers, J., & Standing, C. (2018). What is information? Toward a theory of information as objective and veridical. Journal of Information Technology, 32, 1–20.

10  Information Theory in IS 

337

Nolan, R. (2000). Information technology management. In In a nation transformed by information (pp. 217–256). Oxford University Press. Orlikowski, W. (1996). Improvising organizational transformation over time: A situated change perspective. Information Systems Research, 7(1), 63–92. Orlikowski, W. (2007). Sociomaterial practices: Exploring technology at work. Organization Studies, 28(9), 1435–1448. Orlikowski, W., & Scott, S. V. (2008). Sociomateriality: Challenging the separation of technology, work and organization. The Academy of Management Annals, 2(1), 433–474. Peirce, C. (1955). In J.  Bachler (Ed.), Philosophical writings of Peirce. Dover Publications. Porra, J., Hirschheim, R., & Parks, M. (2005). The history of Texaco’s corporate information technology function: A general systems theoretical interpretation. MIS Quarterly, 29(4), 721–746. Porra, J., Hirschheim, R., & Parks, M. (2014). The historical Research method and information systems Research. Journal of the Association for Information Systems, 15(9), 536–576. Reimers, K., Johnston, R., & Klein, S. (2013). An empirical evaluation of existing IS change theories for the Case of IOIS evolution. European Journal of Information Systems, 23(4), 373–399. Sarker, S., Chatterjee, S., Xiao, X., & Elbanna, A. (2019). The sociotechnical axis of cohesion for the IS discipline: Its historical legacy and its continued relevance. MIS Quarterly, 43(3), 695–719. Shannon, C., & Weaver, W. (1949). The mathematical theory of communication. University of Illinois Press. Stamper, R. (1992). Signs, organizations, norms and information systems. ISOP 92 Proceedings Third Australian Conference on Information Systems, Wollongong, Australia, 5–8 October, 21–65. Star, S., & Ruhleder, K. (1996). Steps toward an ecology of infrastructure: Design and access for large information spaces. Information Systems Research, 7(1), 111–134. Syntheses. (2009). Special Issue, 167(2), March. Syntheses. (2010, August). Special issue, the nature and scope of information, 175(Supplement-1). Thom, R. (1975). Structural stability and morphogenesis. Benjamin. Travica, B. (2011). Back to basics of informing: The INIS principle. Interdisciplinary journal of information. Knowledge and Management, 6, 177–196.

338 

E. McKinney and S. K. Boell

Triple, C. (2013). Special Issue. The difference that makes a difference, 11(1), 93–106. Tsoukas, H., & Chia, R. (2002). On organizational becoming: Rethinking organizational change. Organization Science, 13(5), 567–582. Tuomi, I. (1999). Data is more than knowledge: Implications of the reversed knowledge hierarchy for knowledge management and organizational memory. Journal of Management Information Systems, 16(3), 103–117. Volkoff, O., Strong, D., & Elmes, M. (2007). Technological embeddedness and organizational change. Organization Science, 18(5), 832–848. Weber, R. (2003). Theoretically speaking. MIS Quarterly, 27(3), iii–xii. Wiener, N. (1961). Cybernetics: Or control and communication in the animal and the machine. The MIT Press. Wright, A. (2007). Glut: Mastering information through the ages. Cornell University Press. Yoo, Y. (2010). Computing in everyday life: A call for Research on experiential computing. MIS Quarterly, 34(2), 213–231. Zins, C. (2007). Conceptual approaches for defining data, information, and knowledge. Journal of the American Society for Information Science and Technology, 58(4), 479–493.

11 The Primacy of Concepts and Implications for the IS Field Nik Rushdi Hassan

Introduction Nobel laureate Sir George Thomson said (1961, p. 4): Science depends on its concepts. These are the ideas that receive names. They determine the questions one asks, and the answer one gets. They are more fundamental than the theories which are stated in terms of them.

Despite the importance of concepts, the information systems (IS) field seldom explores concepts, especially native IS concepts in the context of research (Markus & Saunders, 2007). The goal of this chapter is to highlight the primacy of concepts in research and the importance of generating novel concepts to address unprecedented phenomena overtaking the field. Concepts are contrasted to the term that most IS researchers are N. R. Hassan (*) Labovitz School of Business and Economics, University of Minnesota Duluth, Duluth, MN, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. P. Willcocks et al. (eds.), Advancing Information Systems Theories, Volume II, Technology, Work and Globalization, https://doi.org/10.1007/978-3-031-38719-7_11

339

340 

N. R. Hassan

more familiar with—constructs—and the relationship between concepts and constructs is described. The chapter begins with an argument from Toulmin about how concepts form the very foundation of human understanding and that the ability for disciplines to generate new concepts becomes evidence the discipline is progressing. Historical examples of the dangers of not understanding the right concepts are illustrated to demonstrate the practical nature of concepts. And in order to take IS researchers out of their comfort zone with constructs, the chapter goes over how concepts, not constructs, form the foundation of the growth of established disciplines throughout history. This brief historical review is followed by a description of the relationship between concepts and constructs and implications such as relationships have for the IS field. Many information systems (IS) researchers find the philosophy of science a daunting and challenging part of doing research. In part, such apprehensions can be blamed on how inaccessible the discourse of philosophy, metaphysics, and epistemology has become of late, and for many, irrelevant for their immediate research needs. Toulmin (1972) had warned against this separation of philosophy from practical knowledge. He considers philosophers who pursue analytical debates autonomously from empirical discoveries of science as misguided. This separation began when mankind discovered in antiquity that not only can they acquire and make use of knowledge, but they are also aware of their activities as knowers and searchers of knowledge. Since discovering these two intellectual activities, the dangers of scientists becoming unreflective and philosophers losing their relevance have only increased. That’s why Toulmin sought to unify the two and narrow the gap between human knowledge and human knowing by describing mankind’s epistemic self-portrait— how they acquire and validate their knowledge—the problem of human understanding itself. And Toulmin concludes that human understanding, and the accompanying growth in human knowledge, rests in a word—on the concept and its growth, for it is through concepts that human understanding is achieved and expressed. It is through the processes, procedures and mechanisms by which concepts are developed, acquired, used, and improved that particular sciences and disciplines emerge and become established as authorities. With every discovery, with every new phenomenon and unprecedented events that overwhelm the

11  The Primacy of Concepts and Implications for the IS Field 

341

human experience, we add to the list new concepts to allow us to understand and grasp those experiences. He was not alone in this endeavor. Schön (1963) notes that the discovery of new concepts had always been the preoccupation of the scholars for millennia. “People have been trying to explain the emergence of new concepts for over two thousand years” (p. 3). For Schön what underlies the process of the emergence of new discoveries is the “displacement of concepts,” which involves the process of bringing the familiar to bear on the unfamiliar, treating the new in terms of the old, that is, the use of metaphors. Schön (1963) notes that Dewey (1938), C. I. Lewis (1946), and Wittgenstein (1953) view concepts as tools for coping with the world and for solving problems, and we use language as a way of thinking about and characterizing concepts, ultimately to discover how the world works. Whether it concerns Marx’s dream for a “classless society” or Bohr’s idea of the quantum leap, concepts have always been at the center of these endeavors. The continuing need for new concepts cannot be understated, especially in this troubling postmodern, post-factual era. In times when people are more likely to accept accounts based on their emotions and beliefs rather than on actual facts, the role of concepts cannot be more critical. When someone understands something, it is done through an appreciation of a series of related concepts. Knowledge grows as concepts related to that body of knowledge evolve and grow. Similarly, one’s understanding of any phenomenon can be corrupted through the use of ill-conceived or illegitimate concepts based on lies and misinformation. When a series of concepts is based on lies and misinformation they shape the understanding that person toward a misconception—an idea that is incorrect. The Iraqi defector who applied for asylum in Germany became the US chief informant code-named “Curveball,” fed the US intelligence community via German authorities with false accusations that Baghdad had biological weapons. Supported by Iraq’s history of chemical weapons in the past, it was easy to conceptualize the notion of weapons of mass destruction (WMD) to frame the situation post 9/11. The Bush Administration acted on the false information even before they interviewed him. Conflicting reports from the unreliable informant, many coming from his poorly translated information that was intended to ensure that his services would

342 

N. R. Hassan

be retained and visa for asylum granted, were used to justify the preemptive invasion of Iraq (Drogin & Goetz, 2005). According to estimates, over 600,000 Iraqi civilians died from the Iraq war justified by WMD (Chulov & Pidd, 2011). Our ability to process and handle these kinds of events dwindles rapidly when we do not possess the conceptual apparatus to properly categorize and conceptualize them. Similarly, despite clear signs and warnings about the likelihood of violence at the US Capitol on January 6th, 2021, authorities never acted on those cues and were woefully unprepared for what ensued (Montgomery, 2021). The Office of Intelligence and Analysis (I&A), part of the US Department of Homeland Security, is tasked with providing intelligence to state, local, and other non-federal officials. Three offices under the I&A were responsible for work related to January 6th: the Current and Emerging Threats Center (CETC), Counterterrorism Mission Center (CTMC), and Field Operations Division (FOD). The Office of Inspector General (OIG) reported that leading up to January 6th, the I&A offices from all three offices focused their attention on American journalists reporting on the unrest and on non-violent protests taking place in Portland, Oregon, rather than reporting on potential threat information and domestic terrorists. The OIG blamed inexperienced information collectors who did not follow I&A guidelines on reporting those threats and failed to produce the required open source intelligence report (OSIR) that met the threshold for reporting credible threats. Despite reporting from five CETC collectors who documented clear threats referring to the use of weapons to target law enforcement, an individual threatening to kill 50 people, storming Congress, hanging politicians with ropes and that “there would not be enough law enforcement officers to stop the number of armed people arriving in the area,” no OSIRs were produced. In a word, they could not conceptualize the possibility of people storming Congress (Cuffari, 2022, pp. 10–18). One explanation for the inability of the intelligence community to grasp the gravity of the situation is, at least in the short memories of the analysts, that storming Congress was unprecedented. And they were technically correct because the last time the Capitol was attacked was 1814, when British soldiers, a foreign element, attempted to burn the Capitol building that was still under construction, some say in retaliation

11  The Primacy of Concepts and Implications for the IS Field 

343

for the Americans burning Upper Canada’s capital of York (NCC Staff, 2022). However, never has the Capitol been stormed by the country’s own citizens intending on overturning the US elections. Domestic terrorists were not something that needed the attention of the I&A. As Zuboff (2019) explains, “this is how the unprecedented reliably confounds understanding” (p. 12). The same happened to the indigenous peoples of the Caribbean islands who could not comprehend the sight of the Spanish conquerors who they thought were gods and welcomed their own destruction. Zuboff (2019) describes a similar unprecedented phenomenon in our economic life (pp. 8–12): Surveillance capitalists discovered that the most-predictive behavioral data come from intervening in the state of play in order to nudge, coax, tune, and herd behavior towards profitable outcomes … [to] shape our behavior at scale … the goal is to automate us … an unprecedented market operation into the unmapped spaces of the internet, where it faced few impediments from law or competitors … protected by the illegibility of the automated processes that they rule, the ignorance that these processes breed and the sense of inevitability that they foster … [that] we now pay for our own domination. Surveillance capitalism’s products and services are not the objects of a value exchange … they are “hooks” that lure users into their extractive operations in which our personal experiences are scraped and packaged as the means to others’ ends … [it] operates through unprecedented asymmetries in knowledge … [they] know everything about us, whereas their operations are designed to be unknowable to us. They accumulate vast domains of new knowledge from us, but not for us. They predict our futures for the sake of others’ gain, not ours … [and] becomes the dominant form of information capitalism in our time (original emphases).

Just like the doomed indigenous Caribbeans, we, as a field of study, suffer from the unrecognizability of the unprecedented. The unprecedented nature of what Zuboff calls “surveillance capitalism” has allowed it to elude adequate oversight because our existing concepts cannot adequately grasp it. Traditional economics can label it as a “monopoly,” but it’s not exactly that. Supporters of individual privacy can flag it as a violation of individual rights, but what if we willingly accent to its use on us? Many would like us to think that it’s the inevitable expressions of new technology, whereas the surveillance capitalist program is meticulously

344 

N. R. Hassan

calculated and lavishly funded, which makes technology actually just the means to economic ends. As Zuboff (2019) argues, new naming, a new set of concepts is required if we are to effectively tame it. Concepts are, as Hassan et al. (2022) explain, sets of ideas associated with the subject matter being studied, elicited by a given word treated according to logical rules. Following Foucault (1970, 1972), we call the unprecedented objects that need to be studied. Objects of study provide the justification for any discipline’s existence. Various human sciences emerged as disciplines because certain objects of study made them necessary. Coinage and wealth justified the need for the analysis of wealth, the precursor to modern economics. The problem of prices and the nature of money preoccupied the attention of scholars before the eighteenth century, at least until the unprecedented event of production and its accompanying industrial revolution. The human condition of madness did not become a field of study until people who were suffering from motor disturbances, hallucinations, and speech disorders were organized and placed in institutions that later gave birth in the nineteenth century to the discipline of psychiatry or more specifically, psychopathology. As a field of study that has its core concern of information and technology, the IS field is awash with the unprecedented and yet, we continue to hold on or only deal with concepts that we are already familiar with or worse, uncritically borrow from others (Markus & Saunders, 2007). For this reason, we need to start a new program of concept analysis that is capable of generating those new concepts we desperately need to deal with existing experiences and the unprecedented and to facilitate communication among ourselves as scholars as a precursor to growing our body of knowledge.

 he Historical Emergence of New Objects T of Study and Their Concepts To provide further evidence for the need for new concepts, a brief journey of the emergence of new concepts throughout the history of science will be useful. Concepts are how we demystify the miracles of nature and the world around us. For example, in physics, when we describe a “field,”

11  The Primacy of Concepts and Implications for the IS Field 

345

that concept was developed by the scholars of physics to describe a specific phenomenon, electromagnetism, that they can agree on and further the goals of their disciplines. Similarly, when physics explains that light is both wave and particle, the characteristics of both concepts use existing meanings ascribed to them but add knowledge that perhaps only the students of physics can appreciate. The reason for this is because concepts are essentially inventions (Nersessian, 1984), such that it is appropriate to say that what scholars do is to invent concepts to grasp the workings of the world. Although some positivist-leaning scholars may disagree to the use of the verb invent since to them discoveries are activities of finding what’s already out there which cannot be invented, that has been what the great scholars have done for millennia. In other words, scientific concept formation is a process to define, articulate, and describe what we mean when we say such and such a thing corresponds to a concept; that something, the object of study, is a “field,” “wave,” or “particle.” We begin with how modern biology became possible as a discipline. Biology and its concepts did not exist until it emerged in the late eighteenth century, when scholars such as Cuvier (1800–1805), Bonnet (1769), Diderot (1713–1784/1964), and Lamarck (1809/1960) began proposing new concepts and started writing in a way that contradicted but at the same time extended the scholars of Natural History such as Jonston (1657), Linnaeus (1735), Buffon (1749–1788), and Tournefort (1694). Even Natural History and its concepts did not exist before the seventeenth century because in that era, the discourse about nature was concerned more with history of living beings rather than how they are categorized into various genus and species. Books were written on the History of the Nature of Birds (Belon, 1555), The Admirable History of Plants & Herbs (Duret, 1605), The History of Serpents and Dragons (U. Aldrovandi, 1608), which contained anecdotal evidence (possibly fables) of the plant or animal usefulness, legends, and stories associated with them, where they might be found, medications that could be concocted from them, foods they provide, and what various travelers might have said about them. Obviously, with so many that could be written about these plants and animals, the objects being studied were the plants and animals themselves, while the concepts were whatever that were visible or could be said about them, including their anatomy, methods of

346 

N. R. Hassan

collecting or capturing them, various uses, different habitats, concepts for their consumption, and methods of medication using their substances, all in their glorious colors and fantastic descriptions. When Jonston (1657) wrote his Natural History of Quadrupeds, the title itself alluded to the different method by which those same animals would be described. Those concepts were not arbitrarily chosen; they were based on a set of rules and a coherent schema that hitherto did not exist. A new object of study—the quadrupeds emerged as a result. This new object is studied using a new concept of structure was invented to categorize living beings not by mere observation or collection of fables but by a specific method of analysis, an extra level of abstraction above that of Aldrovandi (1640) decades earlier. For example, the structure of a plant is described by its roots, stem, leaves, flowers, and fruit, which in various combinations determine its species. The object of the reproductive organs of a plant is described by its stamens and pistil, the form they assume, and how they are geometrically distributed in the flower. The new concept of character follows from the concept of structure and establishes the identities and differences between each species. In other words, the paradigm of representation of the diversity of forms became the means by which Natural History claimed its authority. The material form of this paradigm is classically demonstrated by Ray (1682) and Linnaeus (1737), as they analyzed a given animal by their name, theory, kind, species, attributes, and use. Ray (1682) was the first to divide the vegetable kingdom into monocotyledons and dicotyledons based on these rules of classification. Linnaeus invented, as a result of his analysis, the first scientific naming system known as binomial nomenclature, producing a taxonomy that gave each species a unique name based on the Latin combination of genus and species. In order for these new concepts to emerge, what was originally the historical study of living beings had to be become natural, hence the term Natural History. For biology to emerge as a discipline, the discourse of Natural History had to undergo its own transformation. While the brilliant taxonomy produced by Natural History organized the natural world in its beauty and complexity, it ascribed to the rule of fixism, which is that the diversity of species could be organized in a fixed grid of structures and character in such a way that it was possible to locate a particular species within

11  The Primacy of Concepts and Implications for the IS Field 

347

that permanent tabulation even if that species was not yet discovered. In the late eighteenth century, Bonnet (1769) theorized that earth periodically suffers global catastrophes that each time destroyed most life but enabled the survivors to evolve. He was first to introduce the concept of evolution in the context of living beings, which challenged the existing rule that the diversity of life was fixed. Although the idea of evolution at the time was very different from the notion of evolution proposed by Darwin (1859) as it passed through Diderot (1713–1784/1964) and Lamarck (1809/1960), such a transformation demonstrates how disciplines grow with the addition of new concepts describing different kinds of evolution. These new concepts described the new object of study of what would be called biology that differed from Natural History. The objects that Natural History studied were living beings while the object being studied by biology was life itself or specifically the forces that sustain life. In order to study life itself, the paradigm of representation could no longer carry the new biological discourse. The study of life had given birth to other new objects of study that look past the relationship between the character of an organ and its function, something that was never possible with Natural History. The new concept introduced by Cuvier (1800–1805) and others to the world that made this possible was the concept of organism, the novel idea at the time that freed living beings from its taxonomy to being able to be part of life-giving functions regardless of the visible form of the organ. The new concept of organic structure gave preeminence to the new concept of organic function and made it possible for biologists to conceptualize and study functions that sustain life, including respiration, digestion, circulation, and locomotion. The focus no longer became the character of the organ but the life-giving functions that may comprise numerous organs that do not resemble each other in terms of structure or character. Gills in the fish may not resemble lungs in mammals, but they coexist and depend on other organs to serve the function of respiration. And thus, modern biology was born. In the social sciences, the same process of displacement of concepts took place to establish the precursors to modern economics. Modern economics did not exist until the eighteenth century because before that production did not exist. But the notions of value, price, trade, circulation, income, and interest did exist and were the concepts that merchants and

348 

N. R. Hassan

politicians regularly consumed to manage the then thriving object of exchange. During this period, money vacillated between becoming a standard medium for exchange and a commodity, which itself increased or decreased in value, depending on the quantity of precious metals available. Davanzatti (1588/1934) noted that providence prepared enough gold for all the things that exist or are done on earth. Nonetheless, Gresham’s Law (Macleod, 1858) stipulated that “bad money will drive out good money,” so coinage value will vary based on its supply. In part, the concept and practice of mercantilism in the sixteenth to eighteenth century was based on this static understanding of wealth. Countries had no choice but to limit the imports of gold (and other commodities) in order to maintain the value of their own commodities and therefore sustain their economies. Instead of raw metal, trade and exchange became the source of wealth. The concept of value became a priority as part of the process of exchange. Why are some things worth more than others? Throughout the sixteenth to eighteenth century, the concept of value underwent several displacements, including viewing value based on the concepts of utility and psychology (used by the Physiocrats) which defines the purpose of exchange as distributing whatever that is in excess to those who need it. However, there was no standard for valuing commodities until Adam Smith (1776) proposed and described how the concept of labor could be used to establish a constant measure for the value of things, since any value has its origins in labor. Later, Ricardo (1817) would extend such an analysis because of the unprecedented event of new forms of production that were invented in the nineteenth century. Value, price, and everything economic would depend on different forms of production, which Marx (1866) would use to conceptualize the promise of communism. This brief journey through the historical development of concepts illustrates how each discipline grows as a result of their scholars standing on the shoulders of previous scholars, inventing new concepts as unprecedented events find their way into the world creating new objects that needed to be addressed. Sometimes the object studied is discovered as a result of the work of researchers and scientists working in their laboratories. More often, new objects of study emerge as the result of the events occurring in the context of that age. For example, before Pinel, the father of modern psychiatry, categorized insanity as a disease which required

11  The Primacy of Concepts and Implications for the IS Field 

349

humane treatment, people who were insane were left to their families to cope, locked up and chained in prisons without any treatment, or just left to the elements (Weiner, 1992). Pinel supported the efforts of Esquirol to establish an asylum for the insane in the eighteenth century and handled various mental ailments that became new objects of study in medicine. The discipline of psychopathology emerged as a result of interactions of families, institutions, economic and social processes, norms and authorities, medical and legal that delimit what could or could not be done. Similarly, in the physics laboratories of Faraday and Maxwell, electromagnetism as an object of study wase being experimented upon and as a result, new concepts that described what a field meant were speculated on. Criticizing Coulomb’s and Newton’s theory of “action at a distance,” Faraday theorized that electromagnetism was not acting with the absence of any medium, and he introduced the new concept of a field that he considered to be a physical object. This new insight was leveraged by Maxwell to combine electricity and magnetism into one theory (Nersessian, 1984). The stock of knowledge of the world depended on the ability of the discipline’s scholars to invent new concepts within the context of discovery (Nik Rushdi Hassan et al., 2019). The elements of this process are depicted in Table 11.1 for the two major disciplines of biology and economics and how their prior disciplines and concepts are displaced by new concepts and the new indigenous theories that they produced using those concepts.

The Growth of Concepts in the IS Field The concern for concepts has also captured the attention of the IS field. As early as the 1980s, the International Federation for Information Processing (IFIP), the leading multinational organization in Information and Communications Technologies and Sciences with over 3500 members, and its Working Group 8.1 Design and Evaluation of Information Systems formed the Framework of Information System Concepts (FRISCO) task group to address the most fundamental concepts in the IS field. The FRISCO first manifesto noted that (Falkenberg et  al., 2000, p. 2):

New theories

Cosmological theories of creation Kant’s theory of the heavens Buffon’s geological history

New The nature of living beings objects of study New Structure of living beings concepts Character of living beings Taxonomy of living beings Fixed structures and characters

Natural History

Gresham’s Law Bullionist theory Physiocrat theories Theory of exchange Cantillon effect

Theory of diminishing returns and comparative advantage Marxian economic theories Keynesian theory Monetary theory

Value and price based Value and price on utility and based on labor and psychology production Mercantilism Supply and demand Circulation of money

Organic structure Organic functions— respiration, digestion, circulation, locomotion Evolution—evolving structures and characters Vitalist theory Cell theory Evolutionary theory Theory of heredity

Money, trade-offs

Economics

Coinage, wealth, exchange

Analysis of Wealth

Life

Biology

Disciplines as they progressed from the sixteenth to eighteenth centuries

Table 11.1  The displacement of concepts and growth of knowledge

350  N. R. Hassan

11  The Primacy of Concepts and Implications for the IS Field 

351

There is a growing concern within IFIP WG 8.1 about the present situation, where too many fuzzy or ill-defined concepts are used in the information system area. Scientific as well as practice-related communication is severely distorted and hampered, due to this fuzziness and due to the frequent situation that different communication partners associate different meanings with one and the same term. There is no commonly accepted conceptual reference and terminology, to be applied for defining or explaining existing or new concepts for information systems.

The FRISCO task group admitted that because too little communication takes place between the different “cultures” or subgroups within the field that organizational, cognitive, and social aspects of IS development have been neglected, resulting in partial or total failures of development projects. Even the FRISCO task group underestimated such problems delaying the completion of the report to nearly 15 years until its publication. The authors hoped that the report would provide the necessary reference background for these different subgroups to build a more coherent system of concepts and suitable terminology that can serve as a theoretical basis for practical work. To their credit, the task force drew from philosophical resources such as ontology, semiotics, linguistics, cognitive science, organization science, sociology, and system science to provide critical insights into the notions of information, knowledge, and communications on which the whole field hinges. They acknowledge that because of the numerous number of interdisciplinary interest groups, conflicting philosophical positions, and lack of communication, an integrated and unified conceptual foundation is challenging to say the least. The goals of the FRISCO report, if compared with the history of the development of objects of study and their associated concepts of other established disciplines in the previous section, did not view conceptual formation as part of the context of discovery. Instead, the report formed the framework around existing information systems and did not open up opportunities to describe or discuss new and novel IS. For example, the structure of the proposed conceptual framework is based on understanding IS that “exist exclusively within organisations, to support their work, and to fulfil their information and communication requirements” (p. 12). As early as 1999, Markus (1999) had already prognosticated that the IS

352 

N. R. Hassan

that was known at the time was likely to disappear. Today, individual and personalized IS in forms such as social media, that is not circumscribed by organizational boundaries, has become a major concern. Second, the FRISCO framework depends on the understanding of “systems” as “specific conceptions (in the minds of people), and can be represented in some (modelling) languages. Thus, in order to understand systems, we have to investigate the issues of conceptions, models and languages” (p. 12). This focus on modeling is the reason why most works citing the FRISCO report are concerned with conceptual modeling, an important but limited subgroup within the IS field. The method in which the modeling is undertaken takes on a formalized tone, with ontological definitions, constructivist approaches that may not resonate with the larger community of IS researches and certainly did not support the task of identifying emerging objects of study or unprecedented phenomena. The report spends a large portion (more than 100 pages) on detailed definitions and formalizations, albeit to provide the intended level of coherence, but ultimately lost the attention of the IS community who were more concerned with the explosion with ecommerce, the internet, and so many other unprecedented events at the time. At about the same time the FRISCO report was being published, the IS community worked on developing its first keyword classification scheme (Barki et  al., 1988). By that time, the IS community and its resources had grown so quickly from a handful of journals into nearly 40 outlets that it was impossible to keep abreast of research and organize the IS body of knowledge. The computerized retrieval systems of the computing organizations did not cater to the needs of the IS research community, especially the lack of an agreed-upon list of keywords to describe IS content. The IS keyword classification alleviated some of the problems mentioned, but like the Linnaean classification of the animal kingdom in Natural History, the growth of knowledge in IS requires a transformation that even adding categories to the list of keywords would not support. Both the FRISCO report and the IS Keyword Classification Scheme performed an excellent job of cleaning up house. But the house itself was in dire need of an upgrade. Others made efforts to develop a coherent conceptual foundation for their own subfields, but there were few and undertaken in disparate

11  The Primacy of Concepts and Implications for the IS Field 

353

research programs. IS researchers were convinced that taxonomies provide the structure and organization that would enable the study of relationships among concepts (McKnight & Chervany, 2001) (Williams et  al., 2008). Glass and Vessey (1992, 1995) proposed a taxonomy of application domains. Nickerson et al. (2013) refined the process of developing a taxonomy in IS and applied their method to the mobile ICT device industry. As the Linnaean classification example accomplished, various categories of information technology could be analyzed using the concepts these taxonomies provided; however, work still needs to be undertaken to elaborate and connect theoretically coherent concepts. Going back to IS-related objects and its unprecedented events, new objects of study—such as that of surveillance capitalism—require new concepts in order to stem its nefarious consequences. Taxonomies provide a preliminary foundation toward developing those concepts. However, tabulating those concepts is not enough. Furneaux and Wade (2009) distilled nearly 700 constructs from Information Systems Research and MIS Quarterly during the nine-year period from the beginning of 1999 to the end of 2007. These constructs were applied in nomological networks of the research published by these two top IS journals. Again, the constructs represent conceptual artifacts that are mostly part of research performed within the context of justification, not within the context of discovery. More importantly, they represent antecedents and consequents within existing research models, investigating phenomena that exist in the past, and not necessarily investigating new objects of study that require the IS field’s attention. In the end as Lee (2010, pp. 336–337) summarized in his retrospective of the IS field: To its detriment, past research in information systems (IS) has taken for granted many of its own key concepts, including ‘information,’ ‘theory,’ ‘system,’ ‘organization,’ and ‘relevance.’ …. [and that] the theories-in-use are taken for granted and not noticed by the people holding them … ‘technology’ remains an ever-­ present but unreflected-upon idea that IS researchers take for granted for when they do their research, much like the water that fish take for granted even as it surrounds them when they swim.

354 

N. R. Hassan

Instead of focusing on the most important objects of study listed above, the focus and energy within IS research is expended on existing concepts borrowed from other fields of study whether they take form of usefulness and technology acceptance (from social psychology) (Davis, 1989), adoption of innovation (from communication) (Moore & Benbasat, 1991), self-efficacy (from psychology) (Compeau & Higgins, 1995), resource-based view (from management) (Bharadwaj, 2000), or productivity-effectiveness (from economics) (Brynjolfsson, 1993), all of which are the most cited and therefore consume the most effort and energy from the IS community.

The Primacy of Concepts over Constructs One reason why so many important objects of study in IS have been taken for granted can be traced to how research is undertaken in IS. It is well known that the dominant epistemological paradigm in IS research from its inception until only recently has been positivistic (Orlikowski & Baroudi, 1991) (Chen & Hirschheim, 2004) (Davison & Martinsons, 2011). Using this approach, research starts with proposing hypotheses and then proceeds to the empirical stage during which data are collected and analyzed to test those hypotheses. Reichenbach (1938) and Popper (1934) coined this process as the “context of justification” to prioritize it from what precedes it, which they call the “context of discovery.” Thus, the context of justification is the stage of research, in which the idealized logic of science, a reconstruction of the actual steps and thinking that took place, is presented in its perfected and refined form. Positivistic or quantitative-type research and qualitative research differ in how they treat concepts and measurements (Goertz & Mahoney, 2012). Quantitative researchers rely more on their data, measurement, and statistical models to derive conclusions. Accordingly, they focus attention on the nature of the data and quality of quantitative measures. They spend less time on the concept and more focus is placed on operationalizing and analyzing datasets. Following this approach, research definitions for concepts are either borrowed or quickly assumed, whereas qualitative researchers delve into the meaning of their concepts and make

11  The Primacy of Concepts and Implications for the IS Field 

355

sense of their research from the interpretation of those meanings. Thus, the constructs and concepts tend to be taken for granted in positivistic research, whereas they are a major concern in qualitative research. One major consequence of such an approach can be seen in the use of constructs as opposed to concepts. This consequence was inherited from the social sciences and made its way into the organizational sciences and into the IS field to develop research models or frameworks. The use of constructs or specifically mental constructs is made popular by Schutz (1954), who, in the efforts to distinguish between the methods of the natural sciences and the methods of the social sciences, made clear that any knowledge in the world involves “mental constructs, syntheses, generalization, formalizations, idealizations specific to the respective level of though organization” (pp. 265–266). So elements of personal life and all objects of culture are abstracted in the process of studying them. Schutz applied the term “construct” to refer to any conceptual artifact, but the constructs formed by the social sciences and those formed by the natural sciences are structurally different. Nature as studied by the natural scientists does not “mean” anything to the molecules and atoms therein, whereas the social world studied by the social scientists has specific meaning to the human beings living in it. While the facts and events in the natural sciences are not pre-selected or pre-interpreted, the humans acting in their world have pre-selected and pre-interpreted their realities, and it is these pre-interpretations that the social scientists are interested in capturing because they determine the behaviors of those humans being observed. The reason why the term “construct” is used is because what the social scientists are investigating are the common-sense thinking and thought objects that the human beings being observed constructed. So while those constructs are directly constructed by the observed humans, the social scientists themselves need to build their own constructions of those constructs—hence Schutz’s expression “constructs of the second degree” (p. 267). These constructs help researchers make sense of observations of human behavior by acting as heuristic devices and, together with other observables and constructs, form what is known as the “nomological network,” defined as the “the interlocking system of laws which constitute a theory” (Cronbach & Meehl, 1955, p.  290). They are nothing more than

356 

N. R. Hassan

“postulated attribute of people, assumed to be reflected in test performance” (Cronbach & Meehl, 1955, p.  283) or abstractions used to explain more observable phenomenon (Morgeson & Hofmann, 1999). In other words, constructs are concepts that are invented because researchers struggle to find the connection between two observable concepts especially in psychological contexts. The theory behind the notion of “construct validity” provides the justification that the construct imagined follows the observable concepts. MacCorquodale and Meehl (1948) distinguish between two kinds of constructs, those that merely abstract the empirical relationships (often called intervening variables) and those that are hypothetical, which involves supposing some non-observables. Both are concepts with different characteristics and both require different approaches to research. The concern that researchers are committing fallacies of reification and personification remains when using constructs. For example, when we consider that groups of people possess “abilities,” “personalities,” or “memories” other than that of the personal kind, do those terms really mean anything? Or when we say that we can manage an organization’s “knowledge” (Alavi & Leidner, 2001) or organizations are capable of “learning” (Senge, 1990), how do we distinguish between the knowledge that exists in a person versus the organization’s knowledge, or does organizational learning work the same way as individual learning? Before such answers can be adequately addressed, it is the concepts surrounding those phenomena that should be studied because the construct emerges from the concept. Unfortunately, the term construct, which started off as merely a means of fixing a problem in psychological theorizing, has turned into not only something theoretical or conceptual but has become something objectively real and the primary focus of the inquiry (Slaney & Garcia, 2015). This understanding of the construct has taken up the attention of researchers in the organizational sciences and, by extension, the IS field. Management researchers discuss about the need to clarify its constructs, the scope conditions under which it applies, conceptual distinctions, and semantic relationships to other constructs (Suddaby, 2010). Constructs are considered synonymous with variables in research models with no mention of the concept that it is derived from (Sutton & Staw, 1995). Concepts, the mainstay of science, are lost in the mix.

11  The Primacy of Concepts and Implications for the IS Field 

357

Implications for the IS Field If we agree with Nobel laureate Sir George Thomson (1961) that “science depends on its concepts” not constructs, and if we agree with Blumer (1954) that what the social sciences needs is to clarify its concepts, it behooves the IS field to do the same. Addressing these fundamental theorizing issues provides the direction for the IS field to become a more socially relevant and intellectually cogent field.

The Distinctive IS Discourse Before any new field of study emerges and gains influence, its discourse has to first take shape (Foucault, 1970, 1972) based on its own specific rules of formation. And since a concept is defined as a set of ideas associated with or elicited by a given word, treated according to logical rules (Sartori, 1975), it is these same rules that characterize any object or study that the field addresses or concepts that that field might apply. It is this unique set of rules, called the field’s discursive formation, that distinguishes that field of study from other fields of study or disciplines. It is these set of rules that determines what can or cannot be said, what can or cannot be claimed in that field of study. The operation of these rules makes possible the creation of new objects of study and makes claims about objects that belong to a specific discourse such that we can recognize economic discourse from psychological discourse, biological discourse from medical discourse, and computer science (CS) from IS discourse. Thus, theorizing in biology takes a different form than theorizing in medicine because they are different discourses, even though statements about organs of the human body, tissues, and cells are found in both disciplines. The rules of discourse of biology concern the study of organic structures that support life. Conversely, the rules of discourse of medicine concern the observation of the human body to identify diseases that affect its health. Similarly, the Association for Computing Machinery (ACM), the professional body that represents the field of CS, describes the discourse of

358 

N. R. Hassan

CS as “computing techniques and appropriate languages for general information processing, for scientific computation, for the recognition, storage, retrieval, and processing of data … and … automatic control and simulation of processes,” which concerns the rules surrounding symbol processing (Denning et al., 1989) and differs from that of IS even though they share the same core concern: the computer. The IS discourse, unlike CS, is sociotechnical. It is neither only social as in the social and behavioral sciences, nor is it purely technical as in CS. That does not mean that IS discourse cannot exist within another discourse or vice versa. For example, when a lawyer applies medical evidence to defend a client, medical discourse operates within legal discourse. The boundaries between discourses are by their nature, porous (Klein, 1993). By extension, the objects of study and its related concepts of IS are sociotechnical. And because the objects studied and the related concepts declare to the world who we are, they need to be consistent with the field discursive formation. When an IS researcher applies economic theory and concepts to study the use of computers using rules concerning value, prices, costs, and trade-offs, which are part of the discursive formation of economics, the power of the economic discourse influences the direction of the study and by extension the IS field. These cross-disciplinary activities present an interesting dilemma to IS researchers. The legitimacy already established by the recognized rules from these “reference disciplines” provides an effective career-building path for IS researchers but at the cost of not building a cumulative tradition within the IS discourse. Additionally, this phenomenon raises the key issue of which discourse rules one should follow: IS or economics. The related issue is whether the researcher is conducting economics research, IS research, or economics research in an IS context. The choice of applying specific rules of discourse has wide-­ ranging implications, especially in the ability of the IS field to invent its own native theories. If the field believes that the growth of its knowledge depends on inventing its own concepts, statements, and theories (Markus & Saunders, 2007), then leveraging the discourse of other disciplines is unlikely to support such a goal and the IS field will remain multimodal, unable to produce theories with a capital “T.”

11  The Primacy of Concepts and Implications for the IS Field 

359

The IS Objects of Study Focus on the unprecedented provides a suitable strategy for the IS field to be socially relevant and intellectually cogent. These unprecedented events that are transforming the world are inherently sociotechnical. The spread of misinformation or fake news is not inherently a new phenomenon. After Caesar’s assassination, Octavian, Caesar’s adopted son, began a vicious program of disinformation and fake news against Mark Antony who claimed succession, which eventually helped Octavian defeat Mark Antony and transform him into Augustus, the first Emperor of Rome (Kaminska, 2018). What is unprecedented is the speed by which misinformation travels to influence its target and the contagion that proceeds in people’s persona, understanding, and culture (Aral, 2020). To study such phenomena requires delving into both the social and technical, which is perfectly natural for the IS scholar. However, such endeavors present a challenge to them because they need to master to a significant extent, the concepts from the social and technical disciplines, to invent and craft their own IS concepts. Zuboff’s (2019) study of surveillance capitalism provides an example of how the unprecedented can be identified, analyzed, and possibly addressed. The identification of an object of study starts with a search for its origins, that is, its date of birth and the circumstances surrounding it. According to Zuboff, surveillance capitalism made its debut in August 2011 when three events that occurred thousands of miles between each other took place: (1) the Apple hack of the music industry by inverting the relationship between the music industry and its customers, (2) a fatal shooting in London that sparked violent protests, and (3) Spain’s challenge of Google to not keep their search histories. The three events represent socio-economic changes that laid the foundations for the birth of surveillance capitalism, (1) the yearning for individualization and the need to exercise control, (2) facing the challenge of battling economic and political forces that treat every individual as invisible, resulting in economic and social inequalities, and (3) how those first two contradictions were exploited by tech companies who saw an opportunity to capture those individualized choices as each struggling segment of society thought they were getting “free” internet services from those companies.

360 

N. R. Hassan

These surfaces of emergence exist within specific authorities of delimitation (Foucault, 1972) that designated, named (typically with friendly sounding terms and policies), and established certain practices. Legal authorities lagged several decades behind the activities of tech companies, allowing them to violate privacy with impunity. Marketing units within each business or marketing companies were more than happy to encourage such practices in anticipation of the individualized eyeballs that would fatten their revenues. The perfect storm of those authorities made possible the expansion of surveillance capitalism. By naming and studying their grids of specifications (Foucault, 1972), the secondary objects of study, the systems by which different kinds of surveillance methods are implemented using technology, and ultimately the concepts that characterize them, can this phenomenon be tamed. The object of study of surveillance capitalism is not purely a social nor technical phenomenon, it is a sociotechnical phenomenon that requires the expertise of the IS scholar.

IS Concepts The final, but definitely not the least of the implications from this analysis of the primacy of concepts, is the need to invent concepts that we can claim to be IS concepts if we are to add to the knowledge of how to address the phenomenon. In addition to all the implications described above, the invention of new concepts supports the next level of the discursive activity for the IS field—the crafting of new native theories. The IS field is struggling with native theories because the field lacks concepts of its own (Markus & Saunders, 2007). Concepts are always associated with observable objects of study of that field and are discipline-specific because they are superimposed on our experiences as a way for us to understand the world. Several concepts can be combined to form a gestalt that engenders certain expectations as the field is made known to the world through those concepts. For example, no one doubts that respiration and circulation are biological concepts, as relativity and quantization are concepts belonging to physics. The question is: What concepts belong to the IS field?

11  The Primacy of Concepts and Implications for the IS Field 

361

By definition, concepts imply a specific locus of the problem and slice up different aspects of the world depending on the discipline’s discursive formation. A concept is significant because it classifies and divides the subject matter so that it can enter into different propositions about the subject matter other than the classification itself. The better our concepts, the better the theory we formulate with them, and in turn, the better the concepts available for the next, improved theory (Kaplan, 1964). Zuboff (2019) introduced several new concepts that would not have been possible without the sociotechnical characters the constitute them. Analogous to the Adam Smith’s division of labor, she introduced the new concept of “division of leaning” (p. pp. 176–195) as the critical axis of social order in the twenty-first century. While Adam Smith’s division of labor was about giving access to more products (total productivity) by specialization, Zuboff’s division of learning is about the access (or lack thereof ) to more knowledge as a result of the differing ability to access, understand, and benefit from data and datafication. The second new concept Zuboff introduced is “behavioral surplus” (pp. 63–97), which is used to describe how Google repurposed the behavioral data they collected to not only improve their search or other services, the behavioral data was used to improve the profitability of ads for both Google and its advertisers. It is through this behavioral surplus that the company was able to sustain and grow exponential profits. This concept was also invented by analogizing how the Ford Motor Company and General Motors were able to maintain levels of production and quality at scale to how Google extracts data at scale through the use of advanced algorithms with the help of “new contractual forms” and “personalization and customization” (pp. 63–64). New concepts have always been invented through the use of metaphors and analogies and such processes of theorizing is something the IS field can learn from.

Conclusion This new century’s unprecedented phenomena are still in its pre-­ conceptual and pre-theoretical stages of being understood. What is required is a full theorizing of these unprecedented phenomena by

362 

N. R. Hassan

naming them as new objects of study and described using new concepts that are sociotechnical in nature, and not just purely social or purely technical as how some of the IS field’s “reference disciplines” operate. The process starts by acknowledging the primacy of concepts, the mainstay of science, which are more fundamental than the theories they constitute. When concepts are ill-defined, tautological (e.g., “performance is the perceived effect of the manager’s job performance”), or defined in conflict with everyday use or accepted research, they obscure rather than illuminate empirical reality (Czarniawska, 2013). Thus, whether IS concepts are invented or adapted, there is a dire need for conceptual development in the IS field. When Leavitt and Whisler (1958) defined IT for the first time, they used concepts such as “intellectual technology” versus “social technology.” They differentiated “information technology” from “machine technology.” They were the first to suggest major sociological, political, and cultural changes to management and organizations stemming from this new technology, with the help of their new concepts. That history of IS can be repeated in a positive and productive manner with the unprecedented of this new century.

References Alavi, M., & Leidner, D. E. (2001). Knowledge management and knowledge management systems: Conceptual foundations and research issues. MIS Quarterly, 25(1), 107–136. Aldrovandi, U. (1608). The Historie of serpents and dragons. Ferronium. Aldrovandi, U. (1640). The Historie of serpents and dragons. Ferronium. Aral, S. (2020). The hype machine. Random House. Barki, H., Rivard, S., & Talbot, J. (1988). An information systems keyword classification scheme. MIS Quarterly, 12(2), 298–322. Belon, P. d. M. (1555). L’histoire de la nature des oyseaux (the history and nature of birds). G. Cauellat. Bharadwaj, A. (2000). A resource-based perspective on information technology capability and firm performance: An empirical investigation. MIS Quarterly, 24(1), 169–196.

11  The Primacy of Concepts and Implications for the IS Field 

363

Blumer, H. (1954). What is wrong with social theory? American Sociological Review, 18, 3–10. Bonnet, C. (1769). La Palingénésie philosophique (the philosophical revival). C. Philibert. Brynjolfsson, E. (1993). The productivity paradox of information technology. Communications of the ACM, 36(12), 66–77. Buffon, G. (1749–1788). Buffon's histoire naturelle, générale et particulière (natural history: General and specific). : National Printing. Chen, W., & Hirschheim, R. (2004). A paradigmatic and methodological examination of information systems research from 1991 to 2001. Information Systems Journal, 14(3), 197–235. Chulov, Martin and Helen Pidd. (2011, 15 Feb). Defector admits to WMD lies that triggered Iraq war. The Guardian. Compeau, D. R., & Higgins, C. A. (1995). Computer self-efficacy: Development of a measure and initial test. MIS Quarterly, 19(2), 189–211. Cronbach, L.  J., & Meehl, P.  E. (1955). Construct validity in psychological tests. Psychological Bulletin, 52(4), 281–302. Cuffari, J. V. (2022). I&A identified threats prior to January 6, 2021, but did not issue any intelligence products before the U.S. Capitol breach. Cuvier, G. (1800-1805). Leçons d'anatomie comparée (lessons of comparative anatomy). Baudouin. Czarniawska, B. (2013). What social science theory is and what it is not. In H. Corvellec (Ed.), What is Theory? Answers from the Social and Cultural Sciences (pp. 99–118). Liber CBS Press. Darwin, C. (1859). On the origin of species. John Murray. Davanzatti, B. (1588/1934). Leçon Sur la monnaie (lesson on coinage). In J. Y. Le Branchu (Ed.), Écrits notables Sur la monnaie (notable writings on coinage) (pp. 217–241). : Alcan. Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 13(3), 318–340. Davison, R.  M., & Martinsons, M.  G. (2011). Methodological practice and policy for organisationally and socially relevant IS research: An inclusive– exclusive perspective. Journal of Information Technology, 26(4), 288–293. Denning, P.  J., Comer, D.  E., Gries, D., Mulder, M.  C., Tucker, A., Turner, A. J., et al. (1989). Computing as a discipline. Communications of the ACM, 32(1), 9–23. Dewey, J. (1938). Logic: Theory of inquiry. H. Holt and Co. Diderot, D. (1713–1784/1964). Eléments de Physiologie. Librairie M. Didier.

364 

N. R. Hassan

Bob Drogin & John Goetz (2005, Nov 20). How U.S. fell under the spell of ‘Curveball’. Los Angeles Times. Duret, C. (1605). Histoire admirable des plantes et herbes esmerueillables & miraculeuses en nature (the admirable history of Plants & Herbs). N. Bvon. Falkenberg, E.  D., Lyytinen, K., & Verrijn-Stuart, A.  A. (Eds.). (2000). Information system concepts: An integrated discipline emerging, IFIP TC8/ WG8.1 international conference on information system concepts: An integrated discipline emerging (ISCO-4), September 20–22, University of Leiden, The Netherlands. Kluwer Academic. Foucault, M. (1970). The order of things: An archeology of the human sciences. Pantheon Books. Foucault, M. (1972). The archaeology of knowledge and the discourse on language. (A. M. S. Smith, Trans.). Pantheon Books. Furneaux, B., & Wade, M. (2009). Theoretical constructs and relationships in information systems research. In Y.  K. Dwivedi, B.  Lal, M.  D. Williams, S. L. Schneberger, & M. Wade (Eds.), Handbook of research on contemporary theoretical models in information systems (pp. 1–17). IGI Global. Glass, R. L., & Vessey, I. (1992). Toward a taxonomy of software application domains: History. Journal of Systems and Software, 17(2), 189–199. Glass, R. L., & Vessey, I. (1995). Contemporary application-domain taxonomies. IEEE Software, 12(4), 63–76. Goertz, G., & Mahoney, J. (2012). Concepts and measurement: Ontology and epistemology. Social Science Information, 51(2), 205–216. Hassan, N. R., Lowry, P. B., & Mathiassen, L. (2022). Editorial-useful products in information systems theorizing: A discursive formation perspective. Journal of the Association for Information Systems, 23(2), 418–446. Hassan, N. R., Mathiassen, L., & Lowry, P. B. (2019). The process of information systems theorizing as a discursive practice. Journal of Information Technology, 34(3), 198–220. Jonston, J. (1657). Historia naturalis de quadripedibus (natural history of quadrupeds). Johannes Jacobus Fil. Schipper. Kaminska, I. (2018, January 17). A lesson in fake news from the info-wars of ancient Rome. Financial Times. Kaplan, A. (1964). The conduct of inquiry: Methodology for behavioral science. Chandler Pub. Co. Klein, J. T. (1993). Blurring, cracking, and crossing: Permeation and the fracturing of discipline. In E. Messer-Davidow, D. R. Shumway, & D. J. Sylvan (Eds.), Knowledges: Historical and critical studies in Disciplinarity (pp. 185–211). University Press of Virginia.

11  The Primacy of Concepts and Implications for the IS Field 

365

Lamarck, J.  B. (1809/1960). Philosophie Zoologique (zoological philosophy). Hafner. Leavitt, H., & Whisler, T. (1958). Management in the 1980s. Harvard Business Review, 36(6), 41–48. Lee, A. S. (2010). Retrospect and prospect: Information systems research in the last and next 25 years. Journal of Information Technology, 25(4), 336–348. Lewis, C. I. (1946). An analysis of knowledge and valuation. Open Court. Linnaeus, C. (1735). Systema Naturea. M.A. David. Linnaeus, C. (1737). Genera Plantarum. M.A. David. MacCorquodale, K., & Meehl, P. E. (1948). On a distinction between hypothetical constructs and intervening variables. Psychological Review, 55(2), 95–107. Macleod, H.  D. (1858). The elements of political economy. Longman, Brown, Green, Longmans, and Roberts. Markus, M. L. (1999). Thinking the unthinkable: What happens if the IS field as we know it goes away? In W. Currie & B. Galliers (Eds.), Rethinking management information systems. Oxford University Press. Markus, M. L., & Saunders, C. S. (2007). Editorial comments: Looking for a few good concepts … And theories … For the information systems field. MIS Quarterly, 31(1), iii–vi. Marx, K. (1866). Capital (Vol. 1). Otto Meissner. McKnight, D. H., & Chervany, N. L. (2001). What trust means in E-commerce customer relationships: An interdisciplinary conceptual taxonomy. International Journal of Electronic Commerce, 6(2), 35–59. Montgomery, D. (2021). 24 warning signs of an insurrection that should have been obvious. Washington Post Magazine, p. A21. Moore, G. C., & Benbasat, I. (1991). Development of an instrument to measure the perceptions of adopting an information technology innovation. Information Systems Research, 2(3), 192–222. Morgeson, F. P., & Hofmann, D. A. (1999). The structure and function of collective constructs: Implications for multilevel research and theory development. Academy of Management Review, 24(2), 249–265. NCC Staff. (2022). On this day, the British set fire to Washington, DC. https:// constitutioncenter.org/blog/on-­this-­day-­the-­british-­set-­fire-­to-­washington-­ d-­c. Accessed Jan 1 2023. Nersessian, N. J. (1984). Faraday to Einstein: Constructing meaning in scientific theories. Martinus Nijhoff Publishers. Nickerson, R. C., Varshney, U., & Muntermann, J. (2013). A method for taxonomy development and its application in information systems. European Journal of Information Systems, 22(3), 336–359.

366 

N. R. Hassan

Orlikowski, W. J., & Baroudi, J. J. (1991). Studying information technology in organizations: Research approaches and assumptions. Information Systems Research, 2(1), 1–28. Popper, K. R. (1934). Logik Der Forschung (the logic of scientific discovery). Mohr. Ray, J. (1682). Methodus Plantarum Nova. Faithorne & Kersey. Reichenbach, H. (1938). Experience and prediction: An analysis of the foundations and the structure of knowledge. University of Chicago Press. Ricardo, D. (1817). On the principles of political economy and taxation. John Murray. Sartori, G. (1975). The tower of babel. In G. Sartori, F. W. Riggs, & H. Teune (Eds.), Tower of babel: On the definition and analysis of concepts in the social sciences (Vol. No. 6, pp. 7–37). International Studies Association. Schön, D. A. (1963). The displacement of concepts. Tavistock Publications. Schutz, A. (1954). Concept and theory formation in the social sciences. The Journal of Philosophy, 51(9), 257–273. Senge, P. M. (1990). The fifth discipline: The art and practice of the learning organization. Doubleday/Currency. Slaney, K. L., & Garcia, D. A. (2015). Constructing psychological objects: The rhetoric of constructs. Journal of Theoretical and Philosophical Psychology, 35(4), 244–259. Smith, A. (1776). An inquiry into the nature and causes of the wealth of nations. Edwin Cannan. Suddaby, R. (2010). Editor's comments: Construct clarity in theories of management and organization. Academy of Management Review, 35(3), 346–357. Sutton, R. I., & Staw, B. M. (1995). What theory is not. Administrative Science Quarterly, 40(3), 371–384. Thomson, G. P. (1961). The inspiration of science. Oxford University Press. Toulmin, S. (1972). Human understanding. Princeton University Press. Tournefort, J. P. d. (1694). Élémens de Botanique. L'Imprimerie royale. Weiner, D. B. (1992). Philippe Pinel's “memoir on madness” of December 11, 1794: A fundamental text of modern psychiatry. The American Journal of Psychiatry, 149(6), 725–732. Williams, K., Chatterjee, S., & Rossi, M. (2008). Design of emerging digital services: A taxonomy. European Journal of Information Systems, 17(5), 505–517. Wittgenstein, L. (1953). Philosophical investigations. (G.  E. M.  Anscombe, Trans.). B. Blackwell. Zuboff, S. (2019). The age of surveillance capitalism: The fight for a human future at the new frontier of power. Public Affairs.

12 Propositions for a Future Information Exchange Theory to Support Decision Making Ali Mohammed Bazarah and Yan Li

Introduction Information exchange (IE) has received a lot of attention in the literature not only from the business side to improve the communication within the organization but also from the social side to build a better community. In this chapter, information exchange is defined as the acquisition and sharing of information among two or more stakeholders to improve the decision quality. Stakeholders are defined as “anything influencing or influenced by the firm” to include suppliers, employees, customers, governments, and competitors [Freeman as cited by Sharp et  al. (1999, p. 387); Pouloudi and Whitley (1997)].

A. M. Bazarah (*) Meehan School of Business, Stonehill University, Easton, MA, USA e-mail: [email protected] Y. Li Center for Information Systems and Technology, Claremont, CA, USA © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. P. Willcocks et al. (eds.), Advancing Information Systems Theories, Volume II, Technology, Work and Globalization, https://doi.org/10.1007/978-3-031-38719-7_12

367

368 

A. M. Bazarah and Y. Li

Existing IS research in IE focuses on the impact of IE on the direct relevant stakeholders in a business domain, such as supply chain (Moberg et  al., 2002; Naslund & Williamson, 2010; Nicolaou et  al., 2013) or healthcare (Lin & Chang, 2018; Rivard et al., 2019; Ko & Chang, 2018). There is limited research about IE through an information system among multiple not directly related stakeholders, especially when some of the stakeholders are not business entities or even individuals. The challenge arises when exchanging information among different stakeholders, such as how to identify different information needs (Dennis, 1996; Cress et  al., 2006) for different stakeholders, analyze different motivations toward information exchange (Nicolaou et  al., 2013; Matschke et  al., 2014; Zheng et al., 2013; Lin & Chang, 2018), and design a communication channel for information flow among stakeholders (Rivard et al., 2019; Ko & Chang, 2018). Additionally, IE literature in IS mostly adopts theories from other fields, such as the social exchange theory, theories from organizational economy and organizational policy, or social cognitive theory, to understand and explain IE from the business perspective, such as the impact of IE on the business performance and the cost and benefits of IE within an organization. There is an absence of native IS theory to understand the implications of IE with respect to the stakeholders’ decision-making process, especially since what is being exchanged is information—the unique element in IE compared to other social exchanges. Therefore, the objective of this research is to propose a theory for information exchange among different stakeholders to support their decision making. The proposed theory, which is the expected contribution of this research, would represent significant contributions to the IE research in IS by bridging the gap of IE in information research and by having a practical implication on organizations to facilitate the decision-making process in a multiple stakeholder environment. The rest of the chapter is organized as follows. The next section reviews different research streams of IE in IS. We then review existing theories used in IE research. The following section presents the research methodology. We then propose a theory of information exchange (ToIE). Finally, we discuss the findings and offer conclusions. For the purposes of this volume, this chapter illustrates Statement as a product of theorizing as discussed in Chap. 2 (see Table 2.1).

12  Propositions for a Future Information Exchange Theory… 

369

IE Research in IS IE has been addressed in many research areas, and numerous studies have examined its effectiveness in improving business performance or decision making. Two main research streams are studied in the literature. The first research stream attempts to identify characteristics of strong relationships between parties involved in IE, such as trust or risk perception. The second research stream seeks to develop frameworks for maintaining these strong relationships (Moberg et  al., 2002). For example, Naslund and Williamson (2010) created an IE framework in supply chain that promotes collaboration and integration among participated businesses, and Serenko and Bontis (2016) proposed a knowledge sharing framework that helps managers to enhance IE among employees. To better understand the IE research, we reviewed different research areas of IE in IS and identified four main research areas of IE: knowledge sharing (Serenko & Bontis, 2016; Xiao et al., 2012; Camp & Sexton, 1992), healthcare (Ko & Chang, 2018; Rivard et al., 2019; Walker et al., 2016), supply chain (Moberg et al., 2002; Nicolaou et al., 2013; Naslund & Williamson, 2010; Homburg, 2000), and social media or virtual communities (Hajli & Lin, 2016; Hall et al., 2010; Zheng et al., 2013; Xiao et al., 2012; Lin & Chang, 2018). In these identified research areas, some studies focus on the impact of IE on improving business performance, such as in healthcare (Ko & Chang, 2018; Rivard et al., 2019) and in supply chain management (Moberg et al., 2002). Other studies focus on the antecedent factors that motivate participants to share or exchange their information such as in the case of knowledge sharing (Serenko & Bontis, 2016; Xiao et al., 2012) and social media (Walker et al., 2016; Lin & Chang, 2018; Hajli & Lin, 2016; Hall et al., 2010; Zheng et al., 2013). None of these prior studies examined the consequences of IE on improving the individual decision quality of multi-stakeholders. The importance of IE is evident in the literature and its different research streams and domains. It also highlights the need for the continuous improvement of IE effectiveness. The focus of this research is to address IE among different stakeholders for the purpose of improving each stakeholder’s decision quality. This is different from prior IE studies, where the purpose is to reach a consensus among multi-stakeholders. To

370 

A. M. Bazarah and Y. Li

achieve this research objective, we further reviewed existing theories for IE in IS, which we describe in the next section. When reviewing these existing theories, the researchers also reviewed factors that are related to the IE effectiveness. These factors would be considered in proposing a theory of information exchange (ToIE).

Existing Theories for IE in IS As discussed earlier, existing IE studies have adopted different theories from other fields, such as social exchange theory, social cognitive theory, political organizational theory, and economic organizational theory. However, to the best of our knowledge, there is no existing IS native theory of information exchange that explains the phenomena of IE and addresses its implications among multi-stakeholders for decision making. Among these theories used in the IE literature, political and economic organizational theories do not help much in addressing the phenomenon of information exchange between multi-stakeholders for the purpose of improving the decision making. Political organization theory is that “each organization strives to optimize its self-interest by (1) minimizing their dependence on other organizations and (2) maximizing the dependence of other organizations on themselves” (Homburg, 2000). It basically strives to understand the organizations’ self-interests and explains how to optimize their outcomes while dealing with other organizations. Economic organizational theory mainly focuses on standardizing data structures and definitions to reduce the overall costs. It is used to address issues associated with reducing coordination costs and increase the overall benefits among the participated organizations (Nicolaou et  al., 2013; Homburg, 2000). Social cognitive theory focuses more on personality traits or individual characteristics and examines their willingness to participate in IE.  For example, Conati et al. (2014) used social cognitive theory to address the impact of users’ characteristics on the use of value charts for data visualization. Similarly, Lin and Chang (2018) used the same theory to address the interaction effect of health information on social media platforms.

12  Propositions for a Future Information Exchange Theory… 

371

Many studies of IE in the IS field used the theory of social exchange as a theoretical foundation (Serenko & Bontis, 2016; Zheng et al., 2013; Lin & Chang, 2018; Hall et  al., 2010; Liao et  al., 2017; Xiao et  al., 2012). Social exchange theory aims to explain individual social exchange behaviors from cost and benefit perspectives. It suggests that people engage in certain social behavior to maximize benefits and reduce costs (Lin & Chang, 2018) and to get some kind of rewards (Zheng et  al., 2013). Serenko and Bontis (2016) define social exchange as “a joint activity of two or more players when each actor possesses and may offer something valuable from the other actors’ perspective.” When applied in the context of information exchange, Lin and Chang (2018) define it as “people sharing their information may perceive it to be fair when others do the same thing, and such fairness perception will lead to greater social interactions.” According to social exchange theory, an individual who produces benefits for the receiver in social exchange may prompt certain forms of reciprocal benefits or rewarding expectations (Serenko & Bontis, 2016). Depending on the rewarding expectations, social exchanges can be classified into four distinct modes: negotiated mode, reciprocated mode, generalized mode, and productive mode (Liao et al., 2017; Serenko & Bontis, 2016). In the “negotiated mode,” individuals clearly establish rewarding or reciprocal conditions before the exchange takes place. In the “reciprocated mode,” the producer believes that the recipient will eventually share something in return without an established reciprocal condition. In the “generalized mode,” the individuals assume that if they share their knowledge, others will share theirs later on. In the “productive mode,” all individuals collaboratively share their knowledge to produce a common good and to help others unconditionally (Serenko & Bontis, 2016). The “generalized mode” and the “productive mode” are found to have a more positive impact on knowledge sharing (Serenko & Bontis, 2016; Liao et al., 2017). The definition of social exchange theory makes it most relevant to the IE research. For example, when the theory of social exchange is used to explain the online IE behavior (Hall et al., 2010) where if a person does a favor to others, others would do the same. However, many IE studies that use the theory of social exchange have focused on the economic side

372 

A. M. Bazarah and Y. Li

of social exchange such as reducing the cost or increasing the benefits. Additionally, the theory of social exchange is built upon the exchange of goods or services. There are three unique characteristics of information exchange that the social exchange theory does not consider. First, the information exchange process includes three unique steps: information recall, information exchange, and information processing (Dennis et al., 1998). This process is different from other forms of social exchange, such as goods or services. Second, the unique usage of exchanged information would have a different impact on different receivers, depending on how the information being evaluated and used (Dennis, 1996). Third, unlike goods or services that are exchanged for one-time consumption, the information exchanged can be re-shared and has a long-lasting impact. These characteristics promote the need to develop a theory of information exchange.

Methodology This research seeks to develop a theory of information exchange that would help to understand and explain the phenomena of information exchange among multi-stakeholders with a focus on the decision-making process. The multi-stakeholders are individuals or organizations who would participate in the IE through an information system for the purpose of having access to comprehensive information to improve their own decision quality. In this scenario, every stakeholder will have access to large pool of information which can be utilized to improve the individual decisions of every stakeholder. This is different from prior studies in IE where the purpose is to reach a consensus or an optimal decision that fits all stakeholders as in the case of group support systems as highlighted by Dennis (1996). There are different methods for theory building. Among these methods, Grounded Theory is a popular method that uses a structured approach to develop theories inductively from data (Myers, 2013, p. 104). Another method of theory building uses an analytical conceptual approach to develop propositions by logically building relationships between carefully predefined factors (Wacker, 1998). For example,

12  Propositions for a Future Information Exchange Theory… 

373

Walker et al. (2016) developed propositions to address the relationships between health information exchange systems (HIE) and the participating provider organizations. Similarly, Waldman (1994) used propositions to propose a theory of work performance. This research uses the later method to propose a theory of information exchange because existing IE literature identifies different factors and their relationships to IE.  The theory of social exchange is used as a theoretical foundation to guide the ToIE development. To apply the analytical conceptual approach, we followed the steps explained by Wacker (1998). We first reviewed different research streams of IE in IS.  Then, we performed a comprehensive literature review on existing theories of IE and identified the deficiencies in information exchange theory. Furthermore, we reviewed the factors that would affect and motivate IE behavior. Based on the findings from the previous steps, we then developed propositions that explain the factors that would influence information exchange among multiple stakeholders and how it would impact their decision-making process. The proposed theory of information exchange, ToIE, will address the previously identified factors with the focus of information exchange through an information system among multi-stakeholders.

 Multi-stakeholder Theory of Information A Exchange (ToIE) Developing an effective ToIE for multiple stakeholders requires a deep understanding of what motivates IE.  Not all people are motivated to share their information with others. In the IE literature, many factors that motivate IE have been identified. Among these factors, information quality is a significant factor that affects the continuous use of IE (Moberg et al., 2002; Hilmer & Dennis, 2000; Zheng et al., 2013; Nicolaou et al., 2013). Information quality can be defined as the information format, accuracy, and timeline (Moberg et  al., 2002). Nicolaou et  al. (2013) found that information quality directly affects trust and risk perception of users which ultimately affects how they exchange information. Similarly, Hilmer and Dennis (2000) discovered that information quality

374 

A. M. Bazarah and Y. Li

allows users to analyze the exchanged information and incorporate the information into their knowledge during their decision-making process. System quality is another factor that is identified in the literature for information exchange (Zheng et al., 2013; Hilmer & Dennis, 2000). System quality means the ease of use or the time and effort required to use the platform and engage with the content (Hilmer & Dennis, 2000). Zheng et al. (2013) found that information and system quality directly affect the individual benefits and user satisfaction, which ultimately determine the user’s intention to exchange their information. Interactive visualization of information is one aspect of system quality that reduces the required effort to reach particular information and allows users to reach the needed information more easily (Hilmer & Dennis, 2000). Perceived benefits and clear outcome expectations are major determinants of continuance use of IE (Lin & Chang, 2018; Homburg, 2000; Hall et al., 2010; Xiao et al., 2012; Zheng et al., 2013; Cress et al., 2006). Perceived benefits are the expected gain that individuals hope to get from a specific interaction (Zheng et al., 2013). Perceived benefits can be categorized as monetary or non-monetary benefits. Examples of non-­ monetary benefits include information and social benefits, such as access to more information or recognition (Zheng et al., 2013). Examples of monetary perceived benefits include increases in return on investment, increases in sales, decreases in inventory, improved flexibility, and better-­ utilizing resources (Naslund & Williamson, 2010). Outcome expectation is the anticipated value or the expected consequences of individual behaviors (Lin & Chang, 2018; Xiao et  al., 2012). Individuals would exchange information in virtual communities with the expectations of enriching their personal knowledge, seeking social support, making friends, and so on (Xiao et al., 2012). Lin and Chang (2018) found that outcome expectation is a significant factor that mediates the influence of human and information interaction on the use of IE. Homburg (2000) and Hilmer and Dennis (2000) found that incentives would increase the quality of exchanged information. Similarly, perceived information benefits (e.g., access to new information) and social benefits are key motivators for people to participate in information exchange activities (Zheng et al., 2013).

375

12  Propositions for a Future Information Exchange Theory… 

The proposed theory of information exchange (ToIE) for multiple stakeholders is stated as “when some stakeholders share their unique information, other stakeholders will also share their information, which will build common information that would improve the decision-­making process of every stakeholder involved.” The idea is that the more access to comprehensive and relevant information from all stakeholders, the better the decision-making quality. As shown in Fig.  12.1, four motivational factors (i.e., information quality, system quality, perceived benefits, and outcome expectations) encourage stakeholders to exchange their unique information. After stakeholders share their unique information, the unique information from different stakeholders is aggregated as “common information.” The common information allows every stakeholder to have a comprehensive view of decision context and select relevant information for decision making. Thus, access to common information will lead to better decision quality. To increase the effectiveness of exchanging unique information, participated stakeholders need to share information that is not prevalent or already known to others. In other words, the stakeholders need to know what common information other stakeholders would need and Information Quality: Format, Accuracy, Timeline

System Quality: Ease of use, Effort, Interactive Visualization

Exchange Unique Information

Common Information

Perceived Benefits: Monetary, Nonmonetary Outcome Expectations: Enrich knowledge, Seeking social support, Making friends

Fig. 12.1  Proposed theory of information exchange (ToIE)

Decision Quality

376 

A. M. Bazarah and Y. Li

then share their “unique information” that could contribute to the “common information.” In the next section, we describe the propositions based on the proposed theory of information exchange (ToIE).

Propositions The quality of decision outcomes depends on to what extent the information is shared among group members (Dennis et al., 1998). Based on the information shared status, information can be classified into three categories: common information, unique information, and partially shared information. The common information is shared and accessed among all group members (Dennis et al., 1998). The unique information is known to only one member but not to others (Dennis et al., 1998). The partially shared information is known to some members but not all (Dennis et al., 1998). In the context of decision making (i.e., choosing the best alternative or evaluating different decisions), the quality of the decision is better when group members have access to the common information than the two other types. In an experiment, Dennis (1996) found that groups who exchanged only a small portion of information made poor decisions compared to those who shared at least 50% of the information. Information exchange is found to positively influence the quality of the decision-­ making process (Hajli & Lin, 2016; Nicolaou et al., 2013). When stakeholders exchange their unique information, the aggregated unique information will create common information. Therefore, the impact of exchanging unique information on improving the decision making is mediated by common information. This will lead to proposition 1: When stakeholders have access to the common information and integrate them into their unique information during the process of decision-making, the quality of the decision will be much better than when only unique information is considered.

As some stakeholders share their unique information, due to the antecedent motivational factors, and become active in IE, other stakeholders who benefit from this information will do the same and share theirs as

12  Propositions for a Future Information Exchange Theory… 

377

well. Therefore, each stakeholder needs to share unique but not sensitive information. Building on social exchange theory that “people sharing their information may perceive it to be fair when others do the same thing”(Lin & Chang, 2018), stakeholders who benefit from the shared information will feel the responsibility to share their information with others. This allows the building of an information exchange platform that would help stakeholders to rationalize their decisions, which leads to proposition 2: As some stakeholders share their unique information, there will be an increasing tendency of other stakeholders who perceive the benefits to share their unique information as well, as a result, the overall Information exchange will increase.

The common information will be presented in the system in an interactive way where stakeholders can easily engage with the information, filter it, and categorize it based on their own needs. Three system quality factors are identified to promote stakeholders to engage in information exchange behavior through information systems. They are ease of use, the effort required, and visualization. “Ease of use” and the “required effort” are proven in the literature such as these explained by the technology acceptance model (Davis, 1989); visualization tool in IE is proved to be effective because it increases the possibility that important information will not be overlooked by individuals (Hilmer & Dennis, 2000). Interactive visualization provides easy channels where people can obtain useful insights from information and use them in their decision process (Ko & Chang, 2018; Lin & Chang, 2018; Rivard et al., 2019). Higher system quality allows stakeholders to engage with the system and effectively participate in information exchange, which will lead to better decisions. Thus, we state proposition 3 as: Ease of use, effort required, and Interactive visualization will lead to more engagement in information exchange.

The quality of the exchanged information, such as information format, accuracy, and sequence (Moberg et al., 2002), is found as a significant factor that motivates users to exchange their information (Moberg et  al.,

378 

A. M. Bazarah and Y. Li

2002; Hilmer & Dennis, 2000; Zheng et al., 2013; Nicolaou et al., 2013). Additionally, information quality allows participants to analyze the exchanged information and incorporate them into their knowledge during the decision-making process (Hilmer & Dennis, 2000). Both system and information quality directly affect the individual benefits and user satisfaction, which ultimately influence the user’s intention to exchange their information (Zheng et al., 2013). According to the channel disposition model (Swanson as cited by Davis, 1989), information quality and system quality are the main components that users use to tradeoff between using a particular system or not. When the shared information is accurate and up to date, and the system is easy to use and no extra effort is required, stakeholders will not only use the system to incorporate exchanged information into their decision but also be motivated to share their information with similar quality. Therefore, the impact of information quality and system quality on decision making is mediated by the exchange of unique information. Moreover, the impact of information quality and system quality on common information is mediated by the exchange of unique information. This explains proposition 4: Information quality and system quality will yield greater value of information access to participating stakeholders which will motivate other stakeholders to exchange their information.

The other two determinants of continuance use of IE are perceived benefits and outcome expectations (Lin & Chang, 2018; Homburg, 2000; Hall et al., 2010; Xiao et al., 2012; Zheng et al., 2013; Cress et al., 2006). While perceived benefits can be depicted as the access to more information or recognition (Zheng et  al., 2013), outcome expectations may include improving the personal knowledge, seeking social support, and making friends (Xiao et al., 2012). Perceived benefits and outcome expectations are important factors that influence information exchange in the online community (Xiao et al., 2012). Furthermore, information benefits, such as access to new information, as well as social benefits that people will receive by being voluntary participants in the information exchange, would encourage and motivate them to continue participating in information exchange (Zheng et al., 2013). Therefore, the impact of perceived

12  Propositions for a Future Information Exchange Theory… 

379

benefits and outcome expectations on decision making is mediated by the exchange of unique information. Moreover, the impact of perceived benefits and outcome expectations on common information is mediated by the exchange of unique information. This leads to proposition 5: Perceived benefits and outcome expectations will motivate stakeholders to exchange their information.

As these identified factors motivate different stakeholders to participate in information exchange, a large amount of high-quality information will be shared across stakeholders. As a result, each stakeholder will have access to a large pool of information that can be used during the decision-making process. This is similar to the group decision support systems that facilitate different people in an organization to make decisions by providing them with access to a large pool of information rather than any individual would have (Dennis, 1996). When group members have access to more information, they would be able to reach the optimal decision (Hilmer & Dennis, 2000). Different from the traditional group decision support systems, the proposed IE theory would motivate different stakeholders, not limited to the organizational boundaries, to share quality information and improve their decision quality. Thus, we represent proposition 6 as: Stakeholders will have a greater potential to improve their decision-making process when they have access to a large amount of accurate information that is shared among themselves.

Discussion and Conclusion This research aims to understand information exchange (IE) behavior within IS research. It highlights the gaps in existing IE research in IS, investigates the commonly used theories in IE, and identifies factors that motivate IE behavior. Based on existing literature, this research proposes a theory of information exchange (ToIE) that aims to explain and understand the phenomena of information exchange and its implications among

380 

A. M. Bazarah and Y. Li

multiple stakeholders with a focus on supporting the decision-­making process and improving an individual’s decision quality. Based on an analytical conceptual approach, we developed the ToIE into six propositions. The ToIE is based on the assumption that when people have access to relevant and comprehensive information, the resulted decision will have better quality (Dennis, 1996). The focus of ToIE is to improve the decision quality of every stakeholder involved in IE. However, better decision quality is not a decision efficacy. This is not to say that decision efficacy is not important. Instead, we believe that the decision-making process with comprehensive information from different perspectives would have better quality than the one with limited information. A decision that considers comprehensive and relevant information can be characterized as rational, at least at the time when the decision is made, and is not random or arbitrary. The ToIE can be used to understand the IE through different types of information systems where their main purpose is to improve the decision quality of multi-stakeholders. The theory specifies the characteristics of such a system to facilitate the IE effectively. That is, the information system designed for IE should focus on both the quality of the exchanged information, such as information format, accuracy, and sequence (Moberg et al., 2002), and the quality of the system, such as the ease of use, the time and effort required (Hilmer & Dennis, 2000), and the interactive visualization (Hilmer & Dennis, 2000). A theory should have clearly defined boundaries to provide specifications about where, when, and how the theory may be applied. This is important in theory testing to reduce the possible refutation of inapplicable boundaries. The boundaries of the ToIE include: • First, ToIE is only applicable in the context of information exchange and is different from the exchange of goods or services. • Second, ToIE assumes that the IE behavior involves two or more stakeholders who have common interests in activities that they are engaging in. That is, each stakeholder would have some perceived benefits or expected outcomes to maximize their benefits and get some reward in the IE process.

12  Propositions for a Future Information Exchange Theory… 

381

• Third, ToIE assumes that the information exchange is through an information system that facilitates the IE activities. The selected system for IE has to maintain information quality and system quality factors that are explained above and highlighted by the ToIE. • Lastly, ToIE assumes that the objective of stakeholders participating in the IE activities is to improve their own decision quality, instead of achieving a consensus. The main contribution of this research is a theory of information exchange (ToIE) that explains the IE behavior among multiple stakeholders with the goal of improving their decision-making process. While the antecedent motivational factors are similar to these in the existing literature (DeLone & McLean, 2004), ToIE includes two new constructs, unique information and common information, that are unique in the context of information exchange among multi-stakeholders. Additionally, ToIE provides new insights into relationships between the antecedent motivation factors and different types of information, and how they would impact the quality of the decision. The propositions developed in this research may be further operationalized as hypotheses to evaluate the validity of the ToIE. This research also makes practical contributions that provide insights on how to design information systems to facilitate information exchange and the decision-making process in a multiple stakeholder environment. The scope of this research focuses only on information quality, system quality, perceived benefits, and outcome expectations as motivations for IE among multiple stakeholders. This is because these factors can be empirically measured and have a significant influence in IE as stated above. Other factors such as environmental, personal, interpersonal, and cultural factors (Matschke et al., 2014) are left for future research.

References Camp, S. M., & Sexton, D. L. (1992). Technology transfer and value creation: Extending the theory beyond information exchange. The Journal of Technology Transfer, 17(2–3), 68–76. https://doi.org/10.1007/BF02199480

382 

A. M. Bazarah and Y. Li

Conati, C., Carenini, G., Hoque, E., Steichen, B., & Toker, D. (2014). Evaluating the impact of user characteristics and different layouts on an interactive visualization for decision making: Evaluating the impact of user characteristics and different layouts on an interactive visualization for decision making. Computer Graphics Forum, 33(3), 371–380. https://doi. org/10.1111/cgf.12393 Cress, U., Kimmerle, J., & Hesse, F.  W. (2006). Information exchange with shared databases as a social dilemma: The effect of Metaknowledge, bonus systems, and costs. Communication Research, 33(5), 370–390. https://doi. org/10.1177/0093650206291481 Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 13(3), 319. https://doi. org/10.2307/249008 DeLone, W.  H., & McLean, E.  R. (2004). Measuring e-commerce success: Applying the DeLone & McLean information systems success model. International Journal of Electronic Commerce, 9(1), 31–47. https://doi.org/1 0.1080/10864415.2004.11044317 Dennis, A. R. (1996). Information exchange and use in group decision making: You can Lead a group to information, but you Can’t make it think. MIS Quarterly, 20, 433–457. Dennis, A. R., Hilmer, K. M., & Taylor, N. J. (1998). Information exchange and use in GSS and verbal group decision making: Effects of minority influence. Journal of Management Information Systems, 14(3), 61–88. Hajli, N., & Lin, X. (2016). Exploring the security of information sharing on social networking sites: The role of perceived control of information. Journal of Business Ethics, 133(1), 111–123. https://doi.org/10.1007/s10551-­ 014-­2346-­x Hall, H., Widén, G., & Paterson, L. (2010). Not what you know, nor who you know, but who you know already: Examining online information sharing Behaviours in a blogging environment through the lens of social exchange theory. Libri, 60(2), 117–128. https://doi.org/10.1515/libr.2010.011 Hilmer, K., & Dennis, A. (2000). Stimulating thinking: Cultivating better decisions with groupware through categorization. Journal of Management Information Systems, 17(3), 93–114. https://doi.org/10.1080/0742122 2.2000.11045649 Homburg, V.  M. F. (2000). The political economy of information exchange politics and property rights in the development and use of interorganizational information systems. Knowledge, Technology & Policy, 13(3), 49–66. https://doi.org/10.1007/s12130-­000-­1020-­z

12  Propositions for a Future Information Exchange Theory… 

383

Ko, I., & Chang, H. (2018). Interactive data visualization based on conventional statistical findings for antihypertensive prescriptions using National Health Insurance claims data. International Journal of Medical Informatics, 116, 1–8. https://doi.org/10.1016/j.ijmedinf.2018.05.003 Liao, W., McComas, K. A., & Connie Yuan, Y. (2017). The influence of unrestricted information exchange on willingness to share information with outsiders: Unrestricted exchange and external information sharing. Human Communication Research, 43(2), 256–275. https://doi.org/10.1111/ hcre.12104 Lin, H.-C., & Chang, C.-M. (2018). What motivates health information exchange in social media? The roles of the social cognitive theory and perceived interactivity. Information & Management, 55(6), 771–780. https:// doi.org/10.1016/j.im.2018.03.006 Matschke, C., Moskaliuk, J., Bokhorst, F., Schümmer, T., & Cress, U. (2014). Motivational factors of information exchange in social information spaces. Computers in Human Behavior, 36, 549–558. https://doi.org/10.1016/j. chb.2014.04.044 Moberg, C.  R., Cutler, B.  D., Gross, A., & Speh, T.  W. (2002). Identifying antecedents of information exchange within supply chains. International Journal of Physical Distribution & Logistics Management, 32(9), 755–770. https://doi.org/10.1108/09600030210452431 Myers, M.  D. (2013). Qualitative research in business & management (2nd ed.). SAGE. Naslund, D., & Williamson, S. (2010). What is Management in Supply Chain Management? - a critical review of definitions, Frameworks and Terminology. Journal of Management Policy and Practice, 11(4), 11–28. Nicolaou, A. I., Ibrahim, M., & van Heck, E. (2013). Information quality, trust, and risk perceptions in electronic data exchanges. Decision Support Systems, 54(2), 986–996. https://doi.org/10.1016/j.dss.2012.10.024 Pouloudi, A., & Whitley, E. (1997). Stakeholder identification in inter-­ organizational systems: Gaining insights for drug use management systems. European Journal of Information Systems, 6, 1–14. Rivard, A., Gentili, M., & Koizumi, N. (2019). Interactive maps for UNOS data visualization. The Journal of Heart and Lung Transplantation, 38(4), S398. https://doi.org/10.1016/j.healun.2019.01.1013 Serenko, A., & Bontis, N. (2016). Negotiate, reciprocate, or cooperate? The impact of exchange modes on inter-employee knowledge sharing. Journal of Knowledge Management, 20(4), 687–712. https://doi.org/10.1108/ JKM-­10-­2015-­0394

384 

A. M. Bazarah and Y. Li

Sharp, H., Finkelstein, A., & Galal, G. (1999). Stakeholder identification in the requirements engineering process. In Proceedings. Tenth International Workshop on Database and Expert Systems Applications. DEXA 99 (pp. 387–391). IEEE. https://doi.org/10.1109/DEXA.1999.795198 Wacker, J. G. (1998). A definition of theory: Research guidelines for different theory-building research methods in operations management. Journal of Operations Management, 16(4), 361–385. https://doi.org/10.1016/ S0272-­6963(98)00019-­9 Waldman, D. A. (1994). Designing performance management systems for total quality implementation. Journal of Organizational Change Management, 7(2), 31–44. https://doi.org/10.1108/09534819410056113 Walker, D. M., Huerta, T. R., & Diana, M. L. (2016). Running head: Value from health information exchange. International Journal of Organization Theory & Behavior, 19(2), 233–259. https://doi.org/10.1108/IJOTB-­19-­ 02-­2016-­B005 Xiao, H., Li, W., Cao, X., & Tang, Z. (2012). The online social networks on knowledge exchange: Online social identity, social tie and culture orientation. Journal of Global Information Technology Management, 15(2), 4–24. https://doi.org/10.1080/1097198X.2012.11082753 Zheng, Y., Zhao, K., & Stylianou, A. (2013). The impacts of information quality and system quality on users’ continuance intention in information-­ exchange virtual communities: An empirical investigation. Decision Support Systems, 56, 513–524. https://doi.org/10.1016/j.dss.2012.11.008

13 New Guidelines for Null Hypothesis Significance Testing in Hypothetico-­Deductive IS Research Willem Mertens

and Jan Recker

Introduction Statistical techniques for testing hypotheses—have more flaws than Facebook’s privacy policies. —Siegfried (2014)

Our chapter extends a conversation occurring across several top information systems (IS) journals (e.g., Burton-Jones & Lee, 2017; Gregor & Klein, 2014; Grover & Lyytinen, 2015) that focuses on pushing a prominent information systems (IS) research tradition toward “a new state of

W. Mertens Colruyt Group, Halle, Belgium J. Recker (*) Universität Hamburg, Faculty of Business Administration, Information Systems and Digital Innovation, Hamburg, Germany e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 L. P. Willcocks et al. (eds.), Advancing Information Systems Theories, Volume II, Technology, Work and Globalization, https://doi.org/10.1007/978-3-031-38719-7_13

385

386 

W. Mertens and J. Recker

play” (Grover & Lyytinen, 2015)—namely, positivist, quantitative research based on the hypothetico-deductive model of science (GodfreySmith, 2003, p. 236). This conversation is bound to theory-­based, quantitative empirical studies that seek to explain and predict IS phenomena (Gregor, 2006). While the conversation relates to a large majority of IS research (Gregor, 2006; Grover & Lyytinen, 2015), including survey and experimental research traditions, it excludes several important traditions such as interpretive and qualitative research, design science research, and certain quantitative traditions like purely data-driven predictive methods and analytical modeling. The chapter illustrates hypothesis as a product of theorizing, as discussed in Chap. 2. As our colleagues before us, we find it necessary to constantly assess and revisit all aspects of our scholarship to ensure that we as a community constantly perform and improve on our fundamental mission of understanding how information systems can be effectively developed and deployed in the human enterprise. Moreover, like the previous contributions of our colleagues in this conversation, we have a specific focus: the way the IS community1 applies null hypothesis significance testing (NHST) within the hypothetico-­deductive tradition. NHST is a method of statistical inference by which a hypothesized factor is tested against a hypothesis of no effect or relationship based on empirical observations (Pernet, 2016). NHST is the dominant statistical approach in scientific use today (Gigerenzer, 2004) and broadly permeates through society. For example, the concept p-value—a key component of the NHST lexicon—has featured in statistics and algebra courses in schools in many countries since the 1930s and has been used as part of the Scholastic Assessment Test (SAT) testing in the United States since at least the 1990s. Our proposal details changes to the way that NHST in hypothetico-­ deductive research is applied in IS. We argue that this proposal is important because it affects research practices employed by large parts of the IS community. The issue, we argue, is not necessarily vested in NHST but in ourselves.2 The way NHST is used in the research practices employed in our  That is, the entire IS scholarly ecosystem of authors, reviewers, editors/publishers, and educators/ supervisors. 2  We will also discuss some of the problems inherent to NHST, but our clear focus is on our own fallibilities and how they could be mitigated. 1

13  New Guidelines for Null Hypothesis Significance Testing… 

387

ecosystem of authors, reviewers, editors/publishers, and educators has become so deeply rooted and ritualized that it has formed normed habits that are difficult to break. This presents a potential threat to IS research on two counts: first, some applications of NHST (such as the use and interpretation of the p-value) have always been susceptible to misunderstanding and misuse (see, e.g., Cohen, 1994; Dixon, 2003; Fisher, 1955; Lang et al., 1998; Neyman & Pearson, 1928). Second, changes to the phenomena and research settings in which IS scholarship is situated (such as the advent of digital population data or the emergence of computational advances to data analysis—e.g., Berente et al., 2019; Freelon, 2014; Lazer et al., 2009) have begun to challenge incumbent practices; some of these changes have led to the emergence of questionable research practices that skirt the line between the ethical and the unethical rather than appearing as blatant misconduct (O’Boyle et al., 2017). We also argue that our proposal is timely. Conversations around the correct application of NHST in the sciences date back to its origin in the proposals for significance testing by Fisher (1935b) and for acceptance testing based on critical rejection regions by Neyman and Pearson (1928, 1933). Several recent developments have reinvigorated this debate, which has paradoxically remained both rampant and dormant for decades. First, the movement to quantify academic productivity and outcomes through journal rankings and citation analysis since the early 2000s as part of the now well-established “publish or perish” mantra has led to the emergence of several questionable research practices such as HARKing or p-hacking (Kerr, 1998; O’Boyle et al., 2017; Simonsohn et al., 2014; Starbuck, 2016). Second, although the open science movement—that is, the idea that all scientific knowledge elements (including publications, data, physical samples, and software) should be openly shared as early as is practical in the discovery process (Nielsen, 2011)—dates back hundreds of years (David, 2004), it has gained momentum especially over the past ten years because digital technologies increasingly provide a range of novel services including data sharing platforms, computationally intensive data analytics, crowdsourcing for project funding, open access publishing, data and publication archiving, and others.

388 

W. Mertens and J. Recker

Third, the increasing availability of large-scale volumes of digital trace data (Freelon, 2014; Howison et al., 2011) through the increasingly ubiquitous digitalization of everyday life (Vodanovich et al., 2010; Yoo, 2010) has led to a vast increase in opportunities to conduct studies with extremely large organic sample sizes, which draws into doubt statistical practices historically used to draw inferences from small-sample populations (Lin et al., 2013; Starbuck, 2016; Xu et al., 2019). Fourth, advances in computational approaches to data analytics and statistical software packages with respect to interfaces, computational power, and usability have led to an increase in their popularity and application (e.g., Hair et al., 2012; Ringle et al., 2012), allowing researchers to easily sift repeatedly through data in search of patterns (Bettis, 2012). Some argue that the increase in the application of such methods has not been met with similar attention paid to methodological details (e.g., Rönkkö & Evermann, 2013; Rönkkö et al., 2016). Fifth, the replication crisis (Open Science Collaboration, 2015; The Economist, 2013; Yong, 2012) has led to renewed and heightened skepticism about commonly used statistical procedures, as well as confirmation, positivity, and publication bias, which has traversed from psychology to virtually all disciplines in the social sciences. In the IS field, the replication crisis has led to the establishment of a dedicated journal on this topic, the AIS Transactions on Replication Research (Dennis & Valacich, 2015; Saunders et al., 2017).3 Finally, we argue that our proposal is relevant to the IS field. While some of the above developments (e.g., the publish or perish movement, the replication crisis) are not restricted to the IS field alone, several others, in particular, the advent of digital trace data, the rise of computational approaches to data analytics, and the continued emergence of

 Remarkably, contrary to several fields, the experiences at the AIS Transactions on Replication Research after three years of publishing replication research indicate that a meaningful proportion of research replications have produced results that are essentially the same as the original study (Dennis et al., 2018). 3

13  New Guidelines for Null Hypothesis Significance Testing… 

389

technologically enabled open science initiatives, speak fundamentally to the core phenomena of our field.4 We develop our proposal as follows. We first review NHST and its role in the hypothetico-deductive model of science. We review historic and emergent threats that relate to how NHST is applied in this scientific model. We then analyze the 100 most impactful recent papers in top IS journals to identify whether NHST is commonly applied in leading IS scholarship and whether there are indicators that suggest that the discussed threats also occur in IS. We then make suggestions for how the IS field should move forward with the application of NHST in order to stimulate reflection and change. We detail proposals for theorizing statistical testing, using statistics for analysis, reporting results, and publishing. We offer two concrete sets of guidelines that our field can adopt immediately.

 HST and its Role in the Traditional N Hypothetico-Deductive Research Cycle The point of this chapter is neither to describe the origins and development of the hypothetico-deductive research cycle and its use of NHST in detail nor to focus on the perceived or actual weaknesses of NHST as a technique in isolation. There are several accounts of the origin and evolution of NHST as a heuristic method of inference (e.g., Pernet, 2016; Szucs & Ioannidis, 2017) and a multitude of analyses of properties of the technique itself (e.g., Amrhein et  al., 2019; Branch, 2014; Wasserstein & Lazar, 2016). We use an idealized account of a typical research process so that we can identify where potentially problematic practices involving NHST have always existed or recently emerged, which is important because such practices can threaten the efficiency, validity, and robustness of the hypothetico-­ deductive research cycle. Figure  13.1 shows a stylized version of the hypothetico-deductive research cycle.  This trend is evidenced, for example, in the emergent number of IS research articles on these topics in our own journals (e.g., Berente et al., 2019; Howison et al., 2011; Levy & Germonprez, 2017; Lukyanenko et al., 2019). 4

390 

W. Mertens and J. Recker

Mid-range script reliance

Theroretical imprecision

Publication bias

Hypotheticodeductive scientific process

Low statistical power

HARKing Sample inflation in digital trace data p-Hacking

Fig. 13.1 Characteristics research cycle

of

and

threats

to

the

hypothetico-deductive

Studies based on the hypothetico-deductive model of science typically proceed in six stages: 1. Researchers posit a new theory in the form of one or more hypotheses (e.g., people with small hands type faster). 2. They then design an empirical study to obtain data (e.g., measures of typing speed and hand size). 3. Next, they collect the data from a sample (e.g., a group of students).

13  New Guidelines for Null Hypothesis Significance Testing… 

391

4. Then, they attempt to corroborate the hypotheses by analyzing the gathered data and calculating some test statistic (e.g., a t-test comparing the typing speed of people with large hands to that of people with small hands). The researchers calculate a probability, the p-value, based on the specified statistical model, that a particular test statistic (e.g., the average typing speed) will be equal to or more extreme than its observed value, while assuming that some logical rival hypothesis is true in the population (e.g., people with small or large hands type at the same speed). This rival hypothesis is referred to as the null hypothesis because it typically assumes the absence of an effect (e.g., no difference in typing speed). The p-value—the probability of finding the difference in typing speed that we found in our sample, or a larger difference, if we assume that there is no difference in the population—is then usually compared to certain thresholds (typically 0.05 or 0.01). 5. The researchers then interpret the results based on the statistical tests. If the null hypothesis is rejected, they typically construe this result as denoting “acceptance” or “support” for the hypothesis stated earlier (e.g., people with small hands indeed type faster). 6. Finally, they submit a report detailing theory, study design, and outcomes to a scientific peer-reviewed journal for publication. The use of practices associated with NHST is deeply engrained in this scientific model. Not only is NHST the dominant approach to statistical data analysis, as described above (Gigerenzer, 2004; Hubbard, 2004; Lin et al., 2013), NHST also forms the logical basis for most hypothesis development (Edwards & Berry, 2010; Lee & Hubona, 2009). Identifying samples that yield sufficient statistical power for NHST is a key component of study design (Baroudi & Orlikowski, 1989; Faul et al., 2007; Goodhue et al., 2007), and data collection procedures involve several techniques for increasing statistical properties relevant for NHST such as sample size (Sivo et al., 2006). Finally, the interpretation and reporting of results also commonly follow recommendations that relate to NHST, either in the form of validation guidelines (Gefen et al., 2011; Straub, 1989; Straub et al., 2004) or in the form of entire scripts, that is, institutionalized patterns for knowledge creation and dissemination (Grover & Lyytinen, 2015; Tams & Straub, 2010).

392 

W. Mertens and J. Recker

The story goes that using NHST within the hypothetico-deductive process in this way is based on an intellectual debate, a misunderstanding of that debate, and a matter of convenience (Branch, 2014; Gigerenzer, 2004; Greenland et al., 2016; Lehmann, 1993). The debate mainly took place in the first half of the twentieth century between Fisher (e.g., 1935a, 1935b, 1955), on the one hand, and Neyman and Pearson (e.g., 1928, 1933), on the other hand. Fisher introduced the idea of significance testing involving the probability p to quantify the chance of a certain event or state occurring, while Neyman and Pearson introduced the idea of accepting a hypothesis based on critical rejection regions. Fisher’s idea is essentially an approach based on proof by contradiction (Christensen, 2005; Pernet, 2016): we pose a null model and test whether our data conform to it. This computation yields the probability of observing a result that is at least as extreme as a test statistic (e.g., a t-value), assuming that the null hypothesis of the null model (no effect) is true. This probability reflects the conditional, cumulative probability of achieving the observed outcome or a larger effect: p(Obs≥t|H0). Neyman and Pearson’s idea comprises a framework of two hypotheses: the null hypothesis of no effect and the alternative hypothesis of an effect, together with controls for the probabilities of making errors. This idea introduced the notions of control of error rates and critical intervals. Together, these notions allow for distinguishing Type 1 (rejecting H0 when there is no effect) and Type 2 errors (not rejecting H0 when there is an effect). While both parties disagreed with each other’s approach, a blend between both approaches emerged as the now dominant approach to testing hypotheses (Lehmann, 1993). It has been argued that this occurred because scientists were in need of clear heuristics, were likely confused by the ongoing debate, and thus created a usable “blend” (Field, 2013; Reinhart, 2015). It is this blend of practices that emerged in the application of NHST, more so than properties of NHST itself, that is at the core of active concerns in several disciplines; thus, this blend of practices should also be critically reflected upon in IS. It is important here to note that we do not mean to discredit the hypothetico-­deductive model per se. In fact, like many of our colleagues, we have ourselves followed this model many times and benefited from the

13  New Guidelines for Null Hypothesis Significance Testing… 

393

advantages it provides: namely, (1) a strong foundation for building a cumulative knowledge tradition; (2) a means for both novel theory generation and incremental theoretical advance through intension and extension (Burton-Jones et  al., 2017; Kaplan, 1998/1964); (3) a means for comparison and reproduction of study results across different settings and samples; (4) a shared language that is common to scientists across many fields; and (5) cognitive advantages for both authors and readers in creating and assessing knowledge creation and the scripts that are produced. Yet, it is healthy to constantly revisit our scholarship procedures and ask whether normed habits and practices remain effective and efficient vehicles in light of new theory, empirics, and ongoing changes to knowledge transfer mechanisms. Therefore, the analysis that follows focuses on the practices that exist in terms of using NHST in this model, as well as the threats for knowledge creation efficiency, validity, and robustness that flow from these practices.

 hreats Emerging from the Application T of NHST in the Hypothetico-Deductive Research Cycle NHST has been controversial since its inception (e.g., Branch, 2014; Gigerenzer, 2004; Greenland et al., 2016), but recent developments have amplified some of the traditional concerns and given rise to the emergence of new concerns. We first review traditional threats to research, stemming from the application of NHST, that have persisted over time. We then discuss emergent threats that have come to the forefront only or particularly in recent years. We discuss both types of threats and the potential risks associated with them in some detail, noting that even broader accounts of these threats are available in the literature (Amrhein et  al., 2019; Baker, 2016; Branch, 2014; Christensen, 2005; Dixon, 2003; Gelman & Stern, 2006; Gigerenzer, 2004; Greenland et  al., 2016; McShane & Gal, 2017; Meehl, 1978; Munafò et al., 2017; Nickerson,

394 

W. Mertens and J. Recker

2000; Reinhart, 2015; Schwab et  al., 2011; Szucs & Ioannidis, 2017; Wasserstein & Lazar, 2016).5

 raditional Threat 1: NHST Is Difficult to Understand T and Often Misinterpreted NHST builds on the p-value measure, which is arguably a sophisticated statistic because it provides an approach to summarizing the incompatibility between a particular set of data and a proposed model for the data. The most common context for applying NHST is in a model describing hypotheses constructed under a set of assumptions in combination with the null hypothesis. However, applying NHST in this way typically involves construing double negatives and null hypotheses that are, by design, meant to be obviously false. Key terms such as “statistical significance” and “p-value” are demonstrably often misconstrued (Amrhein et al., 2019; Cohen, 1994; Greenland et al., 2016; Haller & Kraus, 2002; McShane & Gal, 2017; Reinhart, 2015). Several misinterpretations are particularly common: for example, the p-value is not an indication of the strength or magnitude of an effect (Haller & Kraus, 2002). Any interpretation of the p-value in relation to the effect under study (strength, reliability, probability) is wrong because p-values refer only to the null hypothesis. In addition, while p-values are randomly distributed (if all the assumptions of the test are met), when there is no effect, their distribution depends on both the population effect size and the number of participants, making it impossible to infer the strength of an effect from them. Similarly, 1-p is not the probability of replicating an effect (Cohen, 1994). Often, a small p-value is taken to indicate a strong likelihood of getting the same results on another try, but, again, this cannot be validated because  To illustrate the magnitude of the conversation, in June 2019, The American Statistician published a special issue on null hypothesis significance testing that contains 43 articles on the topic (Wasserstein et al., 2019). 5

13  New Guidelines for Null Hypothesis Significance Testing… 

395

the p-value does not offer information about the effect itself (Miller, 2009). Because the p-value depends on the number of subjects, it can only be used in high-powered studies to interpret results. In low-­powered studies, the p-value has a large variance across repeated samples. A p-value also is not an indication favoring a given hypothesis (Szucs & Ioannidis, 2017). Because a low p-value only indicates a misfit between the null hypothesis and the data, it cannot be taken as evidence that supports one specific alternative hypothesis more than any other possible alternatives such as measurement error and selection bias (Gelman, 2013). In fact, it is likely that the proportion of false positive findings in NHSTbased studies is much greater than assumed (Nuzzo, 2014; Szucs & Ioannidis, 2017). The p-value also does not describe the probability of the null hypothesis p(H0) being true (Schwab et al., 2011). This common misconception arises from a confusion between the probability of an observation given the null p(Obs≥t|H0) and the probability of the null hypothesis given an observation p(H0|Obs≥t), which is then taken as an indication for p(H0). The only correct interpretation is that the p-value indicates the probability of obtaining the observed result or anything more extreme than that actually observed in the available sample data, assuming that (1) the null hypothesis holds true in the population (by design, largely an invalid assumption) and (2) all underlying model and test assumptions are met (e.g., random sampling, independence of sampled units, normality of distributions) (McShane & Gal, 2017). The possible risk associated with incorrectly interpreting NHST is that researchers may either disregard evidence that fails to attain statistical significance or undervalue it relative to evidence that purportedly attains it, in turn leading to ill-informed judgments based on the evaluation of evidence (McShane & Gal, 2017). Interventions or treatments designed based on incorrectly interpreted evidence can lack effectiveness or even be harmful. Also, spurious findings may be published, leading to the diffusion of unsubstantiated theoretical claims.

396 

W. Mertens and J. Recker

 raditional Threat 2: NHST Is Sensitive to Sampling T Strategy and Sample Size The logic of NHST requires an appropriate sampling strategy. NHST logic demands random sampling because results from statistical analyses conducted on a sample are used to draw conclusions about the population. If samples are not drawn independently from measured variables and either selected randomly or selected to represent the population precisely, the conclusions drawn from NHST are not valid because it is impossible to correct for sampling bias, which statistical significance testing assumes is nonexistent (Leahey, 2005). Nevertheless, it is common practice to forego this requirement (Leahey, 2005; Starbuck, 2013). With large enough sample sizes, a statistically significant rejection of a null hypothesis can be highly probable even if the underlying discrepancy in the examined statistics (e.g., the differences in means) is substantively trivial (Smith et al., 2014). Sample size sensitivity occurs in NHST with so-called point-null hypotheses (Edwards & Berry, 2010), that is, predictions expressed as point values. While such hypotheses types are desirable in the natural sciences (Szucs & Ioannidis, 2017, pp. 10–11), in social sciences such as management, psychology, and information systems, they lead to the paradox of stronger research designs yielding weaker tests because most hypotheses are specified as directional statements (such as positive or negative relationships between two variables), whereas the point-null hypothesis describes the absence of a correlation, mean, or variance difference (Schwab et al., 2011). Researchers who gather large enough samples can then basically reject any point-null hypotheses because the confidence interval around the null effect becomes smaller (Lin et al., 2013). The potential risk is that applications of NHST using large sample sizes may lead to worse inferences (Meehl, 1967). Depending on the type of sampling strategy, especially in observational studies, it can be nearly impossible to control for the relationships of all irrelevant variables that are correlated with the variables of interest. This can lead to the identification of many correlations that can be mistaken as revealing true relationships (Bruns & Ioannidis, 2016), which can lead to the computation of biased and inconsistent estimations of effects.

13  New Guidelines for Null Hypothesis Significance Testing… 

397

Traditional Threat 3: NHST Logic Is Incomplete NHST rests on the formulation of a null hypothesis and its test against a particular set of data. This tactic relies on the so-called modus tollens (denying the consequence) (Cohen, 1994), a form of logic frequently used in both positivist and interpretive IS research (Lee & Hubona, 2009). While modus tollens is logically correct, problems arise when it neglects predata probabilities. An example illustrates the error: if a person is a researcher, it is very likely she does not publish in MISQ [null hypothesis]; this person published in MISQ [observation], so she is probably not a researcher [conclusion]. This logic is, evidently, flawed.6 The logic that allows for the falsification of a theory loses its validity when uncertainty and/or predata probabilities are included in the premises, yet both uncertainty (e.g., about true population parameters) and predata probabilities (preexistent correlations between any set of variables) are at the core of null hypothesis significance testing as applied in the social sciences, especially when used in single research designs (such as single-survey or -experiment designs) (Falk & Greenbaum, 1995). In social reality, no two variables are ever perfectly unrelated (Meehl, 1967). A second manifestation of incomplete logic is that NHST neglects predictions under H1 (Szucs & Ioannidis, 2017). A widespread misconception is that rejecting H0 allows for accepting a specific H1 (Nickerson, 2000). But NHST does not require a specification of the data that H1 would predict, it only computes probabilities conditional on H0. Rejection of H0 thus offers no insight into how well the data might fit a general or specific H1. The possible risk associated with incomplete NHST logic, beyond conceptual confusion and generation of misleading inferences, is that it may entice researchers to judge theories as better or worse, even in the absence of direct comparisons to alternative theories. It also favors vaguely defined hypotheses because they are harder to definitely assess against credible alternatives. It makes it difficult and unlikely that theories can ever be conclusively falsified (Edwards & Berry, 2010).  An analogous, more detailed example using the relationship between mammograms and the likelihood of breast cancer is provided by Gigerenzer et al. (2008). 6

398 

W. Mertens and J. Recker

 raditional Threat 4: NHST Fosters Selective T Threshold-Based Reporting P-value thresholds such as