Even Though the Device Looks Insanely Complicated
First published: September 14, 2026. Posted here for archival purposes.
“Everything in Between” is about the systems, institutions, and practices that people build, “things” of a sort that sit in between us, between groups of us, between “us” and “them,” and between us and other systems and institutions that seem terribly far away: “the market,” “the state,” the universe, and so on.
In addition to my posts here, I co-host a podcast titled “Your Leadership Podcast,” which is available on Spotify and wherever fine podcasts are available. I write about law and legal education at TaxProf Blog and for several years co-hosted a podcast about technology and law titled “Your Future Law Podcast.“ My older blog about Pittsburgh and renewing cities, Pittsblog, is still available online, as is my original blog about law, technology, and governance.
While The Hitchhiker’s Guide to the Galaxy (or H2G2 to fans) has acquired something of a sour taste lately owing to its mis-appropriation by certain Silicon Valley mega-moguls, in fact the series remains a highly useful guide to thinking through complex social problems. The key, as H2G2 itself indicates right on the cover (the guide, not the series), is recognizing that these look “insanely complicated” and (therefore) following its instruction: DON’T PANIC.
In that spirit I have begun collecting references to elements of a possibly massive mega-idea of institutional design, function, and change, something that I have been noodling on for more or less 20 years but without ever fully wrapping my head around all of its dimensions. To be honest, it is still a bit blurry, although my time with the knowledge commons world has helped me sharpen a lot of its edges. And the rise and pace of developments in Generative AI have made it urgent.
I will get to the recent references shortly, and then – in this essay and essays to come – to some preliminary implications and possibly even some next steps, but first: the mega-idea. Or perhaps meta-idea.
It is this:
“Knowledge,” in all of its origins, manifestations, purposes, and uses, is simply too much. It always has been. Ask Adam and Eve.
Even the word “abundance” and the cluster of concepts that it refers to is inadequate, really, to describe the sense that “everything that we know, as humans, and that we might know, and even if we limit those conditions to ‘true things,’ is simply too much.” Readers of this Substack will recognize that I am fond of movie analogies; at this point, I would insert two famous memes. From “The Princess Bride” (1987): “Let me explain. No, there is too much. Let me sum up.” And from “Bull Durham” (1988): “Don’t think. It can only hurt the ball club.” Inigo Montoya and Crash Davis. Who knew?
Too much for what? Too much for anything useful. Too much for living. Another meme by analogy, this one from the title of literature-turned-into-a-film: “The Unbearable Lightness of Being” (also 1988!). The existentialists knew.
Still, I am less interested in abstraction and more in pragmatism. How have we – humans – adapted? Does my starting point lead anywhere useful?
Over the millennia, “we” have collectively built a series of systems and processes for organizing and making sense of (waves hands at) everything.
Writing.
Libraries and archives.
Sacred texts.
Monasteries.
Universities.
Cities.
Classification systems and taxonomies.
The encyclopedia.
The universal museum.
Copyright law and the capitalist market (an aside: copyright exclusivity was developed to constrain the production of books; copyright openness (“ideas,” “fair use”) was developed to motivate or enable the production of books.)
Corporations (“the firm,” to economists).
The rule of law.
Disciplines of expertise.
Bureaucracies.
The Internet.
This is not an exhaustive list; other things belong here, too.
My point is simply that are all responses to problems of “too much information,” or “too much knowledge.” Knowledge is tractable to humans when it is organized. Domesticated. The phrases “information overload” and “cognitive (over)load” and their cousins populate one version of the related literature, a body of work that largely focuses on problems faced by individual human beings. How do I cope with the modern firehose? Ann Blair (Harvard), Too Much to Know: Managing Scholarly Information before the Modern Age (Yale, 2010), rightly and convincingly points out that these are not recent or even modern concerns.
My reframing focuses on collective practices rather than on individual decisions. The sociotechnical phenomena that I list above – and that’s what they are – are different ways in which decisionmaking practices of multiple sorts have been institutionalized.
Institutionalized. In slightly more formal language, these are all institutions. I use that word as I have implicitly throughout my writing here, simply to refer to sustained patterns of formal and informal social activity by which human beings coordinate their affairs, plan and build, and resolve conflicts. Institutions matter always; they matter especially in contexts where human histories, talents, capabilities, interests, and values are plural. All pie may be good pie, to borrow a silly line from a silly movie, but not all people like the same types of pie (and shockingly to me, not all people like pie). People find ways to get along. Often. Not always. And sometimes they get along … poorly.
Knowledge and information institutions are and always have been essential to institutional life, but in ways that often have not been recognized systematically. They may be are (also) institutions within institutions. And institutions about institutions. Institutions generating and supporting institutions. The diversity of knowledge institutions gives them power; makes them incredible research opportunities; and makes them insanely complicated.
My list of knowledge institutions above is ordered semi-sequentially not to suggest that they have been periodic, one institution taking the place of another, but instead as a reminder of how we have accumulated them over the centuries.
None of them is precisely alike. Their powers and limits, and biases and blinders, are both distinctive and almost entirely comparative. This institution works well in the following respects and poorly in others, but always by comparison to some other functioning alternatives (contemporary and/or historical) or some imagined version(s).
With my knowledge commons goggles on, I look at these institutions and observe this: each of them was (significantly, if not entirely) a “governance” response to social problems (dilemmas) associated with information and knowledge sharing at small and large scales, sometimes both at once. “Governance” in that usage simply means a collection of material and immaterial rules, norms, and systems for good, bad, and appropriate behaviors developed and administered in some collective context. My colleagues and I once labeled the collection of those dilemmas problems of “abundance,” which is not wrong but which is not entirely complete unless “abundance” is defined loosely, more or less as I have done above: there is too much.
Institutions have origins. Institutions experience change (which is not to say always that institutions can be changed, although sometimes they can be). Institutions exert influence. Institutions are often the ways in which individuals have agency in the world. Or do not.
Sometimes, institutions go away, or institutional change is so fundamental and deep that one institution appears to give way to another.
And cometh now Generative AI. How does it help to investigate and describe the many things that Generative AI “is” and that Generative AI “does” in these institutional terms. I wrote about an adjacent framing of this question earlier this year. If we put governance of Generative AI and governance by Generative AI into “conversation” with institutional analysis of related but different versions of the “too much knowledge” dilemma from other, earlier contexts, what might we learn? What might we take away?
We might add in “all of artificial intelligence” for good historical and sociotechnical measure. We might add in “Agentic AI” in order to give the question some buzzy immediacy.
When I wrote that essay six months ago, I had not yet come across other work bearing similar or related instincts. But now I have. If an important sociocultural project is “finding one’s people,” I am finding another group of mine.
So, and to highlight only the pieces of the argument that have migrated relatively recently across my virtual desk rather than those that might surface through a comprehensive search:
Woodrow Hartzog and Jessica M. Silbey, “How AI Destroys Institutions” (UC Law Journal, 2026)
From the abstract:
The affordances of AI systems erode expertise, short-circuit decision-making, and isolate people from each other. They are anathema to the kind of evolution, transparency, cooperation, and accountability that give vital institutions their purpose and sustainability. In short, current AI systems are a death sentence for civic institutions, and we should treat them as such.
Woody Hartzog and Jessica Silbey are law professors at Boston University, and this thoughtful article intrigued me when they circulated it in draft form. (In law professor-y ways, they are already my people. Or I am one of theirs.) Alas, despite their passionate defense of the civic sphere, there is no explicit mention of knowledge commons concerns, either by name or by suggestion. (Disciplinary connections between knowledge commons perspectives and legal scholarship are not consistently strong.) And their argument is more declarative than empirical. But the abstract makes the connection clear nevertheless: expertise and decision-making at scale are entirely in the center of knowledge commons investigations.
At the same time, their work misses a key comparative dimension: compared to what? Andrew Perlman, Dean of the law school at Suffolk University, pointed out that omission in this short response.
Next:
Dan Williams (University of Sussex) at Conspicuous Cognition, “Most Questions About AI Aren’t About AI”:
For these and countless other questions, many of which we haven’t yet even anticipated, bodies of knowledge, concepts, and methods across existing fields of research will obviously be important. But it should be equally obvious that these questions throw up fundamentally novel kinds of puzzles and challenges that existing disciplines aren’t well-equipped to handle. The knowledge we have accumulated about individual and collective behaviour has targeted human agents (or, at most, organisms). Even when we have constructed more idealised models of rational agency, coordination, competition, conflict, and institutions, they have been rooted in our experiences of human agents.
Advanced AIs won’t just have a wide range of unprecedented capabilities in things like copying, parallelism, processing speed, high-bandwidth communication, and self-modification. They will also have new kinds of goals and motivations (if those terms are even applicable) shaped not by ruthless Darwinian evolution but by complex mixtures of intentional design, curated training data, human feedback, and economic and geopolitical pressures.
At present, we simply lack mature sciences of what might happen when millions or more artificial intelligences with superhuman capabilities and deeply inhuman goals enter our societies, cultures, economies, and institutions. To make progress on the many explanatory and normative questions this will raise, we will need new concepts, models, theories, and fields.
Playing in the same orchestra but not entirely from the same score:
Henry Farrell (Johns Hopkins) (“Programmable Mutter”), Alison Gopnik (UC Berkeley), and James Evans (University of Chicago), “The Research We Need to Understand A.I. Is Falling Apart”:
Nineteenth-century industrialization led to a series of huge social upheavals in which people left agrarian lives in the countryside for jobs in cities. Subjects were increasingly becoming citizens who could vote and influence policy to some degree and were educated in government-funded schools.
Not all of this was pretty. Workers endured terrible living conditions, while rapidly growing cities became cesspits of crime and disease. Politics was ravaged by conflict between those who wanted political and economic revolution and those who called for violent repression of these movements. Liberals, socialists and conservatives worried that these tensions would tear society apart.
The fear that machines were wrecking society created political demand for reliable information on what was happening on the ground and how people were living. In Britain, where the Industrial Revolution began, the journalist Henry Mayhew calculated that hundreds of Londoners made their living gathering bones, rags and cigar ends from the street, as well as dog excrement for tanneries. Commissions of inquiry and censuses followed and gathered information on working conditions and education. Politicians, lawyers and bureaucrats debated whether allowing poor people to vote would lead to disaster.
Social science research arose to study these issues in a more rigorous fashion. Economists studied markets, political scientists studied politics and the state, sociologists studied changing society, and psychologists studied minds. Though these investigations had flaws and were sometimes biased, together they rebuilt our public understanding of what was necessary and possible.
(Henry Farrell supplements that piece with this essay – “How should social science think about AI?” – making explicit how the argument relates to Charles Tilly’s Big Structures, Large Processes, Huge Comparisons (Russell Sage, 1984).)
Periodizing in a helpful if classic way, and focusing on the implications of “knowledge abundance” for teaching rather than for expertise-building or research:
David Deming (Harvard) at Forked Lightning, “The Three Eras of Undergraduate Education”:
When Charles William Eliot became president in 1869, the Harvard College curriculum was a narrow body of classical knowledge. Eliot would go on to create the modern elective system. President Lowell created concentrations and said in his inaugural address that “The best type of liberal education in our complex modern world aims at producing men who know a little of everything and something well.”
This second era called on Harvard to curate and develop expertise. A book contains knowledge. An expert organizes knowledge toward a purpose and helps students sift through what Lowell called the “unending stream” of information flowing past us.
Over the 20th century, the university became much more specialized, which greatly expanded intellectual life at Harvard. But it also fragmented us. Expertise makes you deep, but narrow. We see this outside the academy, where society requires ever-greater reliance on the expertise of others. Do you know how to fix your own car? How to repair your smartphone? How to give yourself a colonoscopy? Don’t try that one at home.
Here at Harvard, we’ve built a great university around the accumulation of specialized expertise. Most of us professors have spent our lives studying narrow subjects. That makes us experts, but it pulls us away from each other. Again, this isn’t unique to the academy, we see it everywhere in life. Specialization has great benefits, but also great costs, because it divides us into separate realities and inhibits the formation of common purpose.
I believe this second era is coming to a close. The internet has made expert content available to anyone around the world. You’ve grown up in an era where you have access to free expertise through YouTube. I’m sure many of you have used it to educate yourself, as you should. That content doesn’t replace your education here, because it isn’t personalized. Masterful lectures are part of a great education, but not the only part. They don’t answer your questions or meet your individualized learning needs.
My original go-to for this line of thinking was and remains Chad Wellmon (University of Virginia) and his 2016 book (Johns Hopkins), Organizing Enlightenment: Information Overload and the Invention of the Modern Research University. I read that book not long after I first encountered Joel Mokyr’s work on Enlightenment knowledge production and economic growth (culminating in A Culture of Growth (Princeton, 2016)), and in my head those two poles helped me align my original knowledge commons contribution to the field of higher education history and my more recent knowledge commons view of the Republic of Letters.
That brings me, finally, to Chad Wellmon’s recent book, After the University: Higher Education and the Future of Intellectual Work (Johns Hopkins, 2026).
The abstract:
When the pursuit of knowledge is eclipsed by money and power, what remains of higher learning?
What is a university for? Is it a sanctuary for disciplined study, or has it become something else entirely? In After the University, Chad Wellmon traces the long and often uneasy relationship between higher learning and the institutions that claim to protect it. Moving from the guilds of medieval Paris and the knowledge factories of Enlightenment-era Göttingen to the research empires of Berlin and Berkeley, Wellmon shows how the modern university has repeatedly reshaped itself to serve shifting social and political demands.
Across centuries, the goods of disciplined study—the joy of reading, the virtues of intellectual rigor, and the possibility of self-formation—have been overshadowed by the pursuit of external rewards such as money, prestige, and power. Part institutional history and part philosophical reflection, After the University examines how today’s institutions defend themselves not in the name of learning but in the language of productivity, innovation, and economic utility. Drawing on his experiences as a scholar, teacher, administrator, and witness to crises such as white supremacist marches and the COVID-19 pandemic, Wellmon illustrates how universities justify themselves through the outputs of graduates, research discoveries, and workforce training while leaving unmentioned the very practices that once defined them.
Despite this transformation, Wellmon argues that the university’s current state of turmoil exposes a new, enticing possibility: recognizing the practices of disciplined study as goods worth valuing in and of themselves rather than simply as means to other ends. With insight and urgency, After the University asks whether our institutions can still nurture intellectual desire—or whether we must find new homes for the life of the mind.
For fascinating reviews and comments on the book, read:
Nils Gilman at Small Precautions, “The Value of ‘Disciplined Study”:
Readers of this newsletter know I have argued that Kerr’s multiversity is finished — with the arrival of AI about to serve as the coup de grâce. Reading Wellmon has changed how I think about that argument. I had treated the contradictions as structural: too many missions bolted onto one balance sheet. He shows they are also conceptual. The multiversity could hold research, training, credentialing, and coming-of-age together only so long as human capital theory supplied a single currency in which all four could be denominated. Remove the currency — as an AI-flooded labor market is now doing to the credential’s signaling value — and nothing holds the conglomerate together.
Wellmon therefore earns his elegiac register. He notes that, despite everything that’s gone wrong with the university as a system, having or not having a BA still divides those doing well from those doing badly, even as “the estrangement of disciplined study and the university is crystallized in this social fact. People are driven toward the university not because they want to learn, but rather because they know that the road to a good life runs through the university. The university absorbs intellectual desire but redirects it into pursuits of status, credentials, and output.” But an exclusively output-oriented conception of higher learning, as he puts it, is “both diminished and pernicious.”
In my Persuasion piece, I argue that the pedagogy of higher ed must be remade around what humans uniquely provide — judgment, trust, taste, the constitution of goals, and so on — because AI is commodifying everything else. Wellmon’s book adds a crucial point: those capacities aren’t (or shouldn’t be) primarily valued for instrumental reasons. For in fact they are precisely what emerges from a commitment to disciplined study.
And if you have read this far, and know my prior arguments, you will be unsurprised by this closing reference to:
Hollis Robbins (Utah) at Anecdotal Value, “‘After the University’ (Review)”:
It took me nearly seven years as a dean at two universities to see the whole of what is causing the current crisis in higher education, which I most recently wrote about here. Wellmon’s After the University offers a lament for a lost fantasy (a term he returns to again and again): “a desire for intellectual community, excellence in thinking, and the intrinsic goods of learning, knowing, and thinking” (351). I am by this time too jaded by what I’ve seen to even think about using that word myself, but I am glad that Wellmon has it and that people still see the university as someday returning to something better than the wreck it is now.
She is a little blunter than I usually am.
But she is not wrong.
University professors across schools, colleges, departments, centers, and registers of seniority are, collectively, heavily invested in their tendency to construct a fantastic world – a “community of scholars” pursuing the “life of the mind” – and then to both imagine and to practice the principle that their highest calling is to steward the transmission of “critical thinking skills” in disciplinary registers (derived from the late 19th century problem solvers responding to the Industrial Revolution) via their teaching and via their scholarship.
If only they – collectively – knew fully the extent to which modern universities and colleges are deeply and irrevocably committed in other directions. The university as the institutional centerpiece of the “life of the mind” is Brigadoon, an imagined and largely imaginary idea even in one of its original U.S. constructions, the famous (and backward-looking) Yale Report of 1828 and its central attention to the university as training for “the discipline and the furniture of the mind.”
Chad Wellmon mourns. Hollis Robbins, and others, may believe that the proverbial reformist game is worth the candle.
I’m not so sure. I’m a fantasist at times as much as the next academic, and most of the time I would like nothing more than to steer “the university” and my own university back in the direction that Chad Wellmon describes.
My institutional angle suggests that my energy and attention should take different forms. I don’t have a shot to take at making universities better. (But never say never, a la Sean Connery!) The angle does not give me answers, or strategies. It does give me and others a lifetime’s worth of research topics. I do have questions, which have both a practical, reformist, or even entrepreneurial character as well as a descriptive, observational character:
What comes next, amid Generative AI and (everything else)? Amid yet another twist in the “there is too much” history of knowledge and information, what do our institutions look like and how do they operate? What will they look like and how will they operate? I don’t mean only “what do universities look like?,” or “what will they look like?” let alone “what should universities look like?” Maybe “the university” as an institutional form is so tethered to expectations originating in during the Enlightenment and continuing through the Industrial Revolution and the 20th (modern) century that “the university” as an institutional form is … over. Cooked. Done. Maybe we should be writing Shelley-esque valedictories to “the Ozymandian university”: “look on my Collegiate Gothic and Brutalist works, ye Professors, and despair!”
Too soon? I mean, seriously, “what do knowledge governance systems look like?” What will they look like? What should they look like? How do we break big questions like those into “nearly decomposable” and useful elements?
And how are “we,” collectively, getting from here to there, and when, and why? Processes – plural – matter as much as end-states.
DON’T PANIC.
Thanks for sticking with me.