My new book, with Natasha Tusikov, The New Knowledge: Information, Data and the Remaking of Global Power is out today, via Rowman & Littlefield. It’s available for free as an open-access download (thank you, Social Sciences and Humanities Research Council of Canada!), so please do check it out.
If you’re an instructor, we’d be more than happy to talk with your class about our book and the issues it discusses. I can be reached at bhaggart at brocku dot ca.
On that note, a bit on the book itself.
What it’s about
We’re both very happy with the way The New Knowledge turned out. It’s the culmination of work the two of us have been doing, together and apart, for the past seven years. These include ideas that we workshopped in our two co-edited volumes, Information, Technology and Control in a Changing World: Understanding Power Structures in the 21st Century(with Kathryn Henne, Palgrave Macmillan, 2019) and Power and Authority in Internet Governance: Return of the State? (with Jan Aart Scholte, Routledge, 2021). It also draws on our close reading of Sidewalk Labs and Waterfront Toronto’s failed attempt to build a smart neighbourhood in Toronto, which serves as the book’s leitmotif.
The New Knowledge unpacks the transformative implications of the rising centrality of the control of knowledge – particularly data and intellectual property – for the exercise of economic, social and political power. Put another way, no matter what field you’re working in, or what you do, pretty much every policy and activity has a data and IP – a knowledge – component. We’re negotiating trade agreements that are no longer about physical trade, but regulating knowledge flows (chapter 3) – agreements that are now more about regulating global production/value chains than international trade (also chapter 3). Companies that previously would have been seen as lowly IT providers are inserting themselves (and being welcomed) into all parts of society based on their ability to collect and analyze data (chapters 5 and 6). Property relations are being redefined, with the Internet of Things placing de facto control over connected devices with the supplier, not the nominal owner (chapter 7). The state, meanwhile, is as enamored of data and algorithms as everyone else, and is as ready to buy into its magical properties as anyone.
Understanding this transformation is a vital necessity for everyone. And while there are many who have been working on data and IP issues for a long time – many of whom we cite in this book – many people (including academics and policymakers) are coming across them for the first time. This book is for all those who want to break through the mysteriousness and opacity that often accompanies discussions of data and IP, which has only gotten worse in the current Artificial Intelligence mania.
And so we have two chapters devoted to demystifying knowledge (chapter 2) and data (chapter 4), written for those who would rather not wade through the deliberately murky musings of dense French philosophers. It’s our hope that these chapters will help to inoculate readers against those who ask us to trust in the data and the algorithm – to trust in AI, say – as if these were something magical and above human biases and fallibilities. The promise of AI, of data, of algorithms, is the promise of neutral knowledge unsullied by human prejudice, ignorance and bias. It’s a false promise, as scholars have known for decades, and as we highlight here. The reality is, you can never escape people. Once you have a firm grasp of what data is, that it’s people all the way down, it’s hard to take the claims of AI evangelists seriously.
The Power of Belief
The theme of belief in data, IP and algorithms recurs throughout the book. We argue that the defining characteristic of our current knowledge-driven society isn’t technological, but rather the belief that commodified knowledge – data and IP – is a superior form of knowledge. As we unpack in Chapter 5, this belief, and the privileging of commodified knowledge (and the people who claim mastery over its collection and analysis), is only partially related to global digitization and the internet. It is this belief that convinces urban-development experts, say, that a search engine and advertising company has the capacity to plan and build a city. Or that we should listen to computer scientists who claim to have invented god, and not just a glorified autocomplete machine.
Power and Control
The ability to control these socially valuable forms of knowledge (or, rather, forms of knowledge believed to be socially valuable) has allowed certain actors – namely the large (mostly) American “platforms” – to play dominant roles throughout the economy and society. Want to understand what Meta and Google are up to in the C-18 tug-of-war? Our book outlines how these companies are systematically working to make themselves central de facto governors over our lives, whether as two-sided markets or as standards-setters (chapter 6).
Nothing New
The belief that the challenges posed by AI, by algorithms, by platforms, by the internet are all new, and require novel responses, is the biggest challenge to addressing these challenges. Shoshana Zuboff put forward her book The Age of Surveillance Capitalism as an attempt to fill what she called a void, a “tabula rasa” that required novel responses to “unprecedented” challenges.
Not only does Zuboff do a disservice by effectively erasing the myriad scholars who have been working for decades on the specific issues she identifies, but she’s flat-out wrong in claiming that these are unprecedented challenges.
Theory is the scaffolding of an argument. In The New Knowledge, our scaffolding was provided by three International Political Economy scholars in particular: Susan Strange, Robert Cox and Karl Polanyi (chapters 2 and 3). I won’t go into the details here (read the book!), but importantly, all three of them made their primary theoretical contribution before the mainstreaming of the internet (1980s for Strange and Cox, 1940s(!) for Polanyi), and were not primarily focused on data, intellectual property or related issues. The mark of a good theory is its applicability beyond the situation or era in which it was first proposed, and these three more than fit the mark. Most importantly, our successful application of these three theorists to what we call the knowledge-driven society suggests that if existing theories can be used to understand our current moment, existing policy solutions are also available to us. No need to reinvent the wheel.
(This isn’t to say that we simply apply their work unchanged. We do propose reinterpretations of all three, but in a way that solves puzzles posed by their own formulations while maintaining the essence of their own theories. That is our theoretical contribution to our understanding of International Political Economy.)
That said, we also draw on the work of two other, more contemporary, theorists, José van Dijck (for her concept of dataism) and Evgeny Morozov (for his concept of technological solutionism), who better than most understand the ideological orientation of our current age.
Knowledge Feudalism, Digital Economic Nationalism and … a third option
As interested as we are in theory, and as important as it is to describe and understand our current moment, we also have a strong, pragmatic interest in policy, a byproduct of our pre-academia years spent working in various capacities for the Canadian federal government. We identify two dominant policy approaches to the knowledge-driven society. The first is knowledge feudalism. This is the dominant player’s strategy: if you already control significant amounts of economically and socially valuable knowledge (and the mans to disseminate and analyze it), then you’ll want to ensure that others pay to play. This is the strategy of the US and its companies (e.g., Meta, Google).
Challengers, seeking access to such knowledge, will be more open to cooperation, more-open data and IP flows, and state intervention (the state being the only actor capable of going toe-to-toe with US and Chinese corporate champions). We call this Digital Economic Nationalism. And while it’s seen as more benign than US Knowledge Feudalism, its end goal remains domination of others. This is one reason why, in the book, we are relatively critical of the European Union’s General Data Protection Regulation (GDPR): it may have some good points to it, but it’s designed around European interests and European values, not those of other countries, to whom the EU presents itself as a “regulatory superpower” (chapter 8).
While we acknowledge that Digital Economic Nationalism is a logical response to knowledge feudalism (and shouldn’t be seen as akin to mercantilist protectionism (chapter 3)), it fails to get rid of the oppressive power dynamics created by the knowledge/data/IP-driven society. And so, in Chapter 9 and the Conclusion, we propose a policy of decommodification. Drawing on concepts of data justice, group (as opposed to individual) privacy, Indigenous data sovereignty, the practical work on Barcelona’s smart city and Karl Polanyi’s analysis of fictitious commodities (data and IP being fictitious commodities), we argue that data and knowledge commodified by intellectual property laws must be seen first and foremost not as commodities to be repurposed at the whims of others. This approach to data mirrors how other fictitious commodities are treated. Labour and environmental regulations, and bankruptcy laws, are all designed to limit the marketization of things that are essential to human and social functioning. That’s the direction we need to head with data and IP laws. (And AI laws, which are functionally equivalent to data laws.)
Instead, the context within which data and knowledge is generated must be respected, with the people who serve as the source of this knowledge, and who will be acted upon using this knowledge, given control over how their data and knowledge will be used. Almost all problems in the data- and knowledge-driven economy are the result of knowledge appropriated away from these individuals and groups: it’s not a problem when a user sends their data to Google Maps to get from Point A to Point B; it is a problem when that data is sold to insurance companies and ends up being used to increase an identified group’s insurance rates, or deny them insurance altogether.
A note on ChatGPT and clarity of thought
Readers will note that, in contrast to the avalanche of words that has accompanied the November 2022 release of ChatGPT, the book is somewhat light on mentions of ChatGPT, large language models and “artificial intelligence.” We would invite the reader to see this as a feature, not a bug. As we note in the book, “artificial intelligence” is a term with many different sets of meanings. It’s also a term that current overuse has degraded to the point to meaninglessness. When writing the book, we made a point to avoid using the term, preferring instead to focus on precise concepts, such as “predictive algorithms” or “algorithmic regulation.” We intentionally burrowed down to the fundamental concepts at play, namely data, knowledge, algorithms, and the perceptions of these. It is our hope that focusing clearly on these underlying concepts – which are equally applicable to the AI debate in 2023 as they were in 2022 – will help readers cut through the AI froth, to focus on the ideas and concepts that truly matter.
With all that out of the way, here’s the Run the Jewels song that served as the inspiration for this post’s title.
You must be logged in to post a comment.