Why the New York Jewish Left’s Little Magazine "Dissent" Still Has Much Unfinished Business

Dissent magazine’s relevance in a world of political rubble, grifters, confidence men, neofascist werewolves, chaos monkeys, & reluctant ex-left-neoliberals. Plus what the next mayor of New York—& the next left—can learn from Keynes & company…

Share

Share Brad DeLong's Grasping Reality

My friend Patrick Iber is now a coëditor of Dissent magazine.

Thus I wound up at a small fund-raising event for it last month. And there Patrick surprised me by putting me on the spot to say something, in the context of his having just been challenged by somebody who wanted to know his response to the claim that the time for a left-wing New York City Jewish-intellectual magazine—like Dissent—had passed.

I thought he fumbled the answer to that somewhat. I see Dissent as not mere contrarianism, not mere dissent, but as a vital force for accountability and progress.

So all I could think of was to pick up the thread. And I fumbled my answer somewhat.

And so is not what I said, but what I now wish I had said:

Give a gift subscription

In answer, I would not reject completely the label of “Left Neoliberal”. In fact, I do have a Neoliberal Party card around here someplace:

Get 50% off a group subscription

But what do I think is the place of the little magazine that is Disssent?

You mentioned my saying that it was a time for passing the torch from us left-neoliberal center-leftists to those further to our left:

  • I stand by that.

  • But: not pass not the torch.

  • Rather: pass the baton.

  • For it will need to be passed back.

  • And: pass the baton, not bend the knee.

I do stand by passing the baton.

Until there’s something other than rubble between us left-neoliberals on the one hand and the neofascists and true werewolves on the other, the only path toward a better America is for those to my left—especially including the editors and readers of Dissent—to propose workable policies and mobilize a majority political coalition to support them. And at the moment, there is nothing but rubble. And there is no prospect of there being anything other than rubble any time soon.

The job of me and my karass, who have largely failed in our political-economic societal reörganization and reïnvigoration project over the past forty years (to what extent this is our fault is a matter for another time), is to support without following the left-of-us party line. Our job is to try to make those policies workable, and to help implementing them, all the while whimpering and kvetching and insisting: “you must do things that really work”.

That must be so now and as far as I can see into the future. That must be so until the center-right possesses any actors who (a) have political force, and (b) are neither grifter, morons, fools, nor cowards.

Until then, passing the baton is our only option if we seek to accomplish good things for America and for the world. We need to support. But we also need to dissent, and to dissent vocally, but in a way that pushes those to our left with the baton to act in a way that raises their chance of being successful.

Thus people like me are potentially of enormous value for Dissent, for dissenting from left-wing party lines in order to advance the chances of success of the cause and to make the cause even more worthy is its purpose, and ours.

There has been a lot of talk about Zohran Mamdani here this afternoon. He is a man who calls himself a socialist who has won the Democratic Party primary in the city of New York. He thus is the Democratic Party’s candidate to be New York’s next mayor, and first competent mayor since Michael Bloomberg. But winning a Democratic Party primary in New York City is in no sense analogous to winning a bet at a casino. No chips have changed hands. It is more like making it to the casino, and then pushing your chips into the center of the table while the roulette ball bounces in its wheel.

A primary election is a process to choose a candidate, not a process to choose an office-holder.

In a Democratic Party primary, people who regard themselves as Democrats together—by whether they turn out or not, and who they then vote for—select a candidate to put forward to the general electorate for its consideration on the Tuesday after the first Monday in November. This is not winning anything. This is only picking a candidate who might win. Is this the candidate who is in the sweet spot with respect to properly balancing (a) the likelihood that the general-election voters will in fact approve the Democratic Party’s candidate to be Mayor of New York City, and (b) the likelihood that that candidate as Mayor will build governmental-bureaucratic coalitions to implement policies that will successfully advance the good parts of Democratic-Party values and preferences? That also is a conversation for another time. I would be anxious to listen, for I do not know enough about New York City to know, but—listen another time.

With respect to the attitude of my karass toward Zohran Mamdani, there is an apposite quote from John Maynard Keynes:

Forgive the candour of these remarks. They come from an enthusiastic well-wisher…. I am terrified lest progressive causes in all the democratic countries should suffer injury, because you have taken too lightly the risk to their prestige which would result from a failure measured in terms of immediate prosperity. There need be no failure. But the maintenance of prosperity in the modern world is extremely difficult; and it is so easy to lose precious time…

In this letter it is February 1, 1938. Keynes is writing to Franklin Delano Roosevelt. Keynes is dissenting from a great many of the policies of the Second New Deal. He disagrees with many of them. He agrees with others, but thinks them wastes of energy and distractions from the essential at that moment. What is essential? Keynes believes, and he is right, that what is essential is for FDR to laser-focus on general economic prosperity: boost economy-wide spending to put more people to work, for that is essential if the Second New Deal is to be a success, is to be perceived as a success, and thus for its beneficial parts to endure.

As then, so now.

The results of the primary mean that we have all now placed on awful lot of all our chips on our bet. Our bet is that the general-election voters will choose Zohran Mamdani. And our bet is that his subsequent mayoralty will be a smashing success. The work to make that happen begins now. What is needed now is not any form of gloating (except for prolonged, high intensity, and extremely gratifying gloating over the defeat of Andrew Cuomo and his network of rent-seekers) but is, rather, thinking and doing. And Dissent ought to be a major part of the thinking.

It is an obvious piece of reality that any mayor calling himself “socialist” in any large post-neoliberal city, and especially New York City, is bound with enormously constraining fetters of iron. A mayor can do much. But only as far as things within their fettered reach. One key to adapting to this situation is to not overpromise. A second key is then to deliver concrete, real, broad-based results.

That, I think, is the reason why it is very important that we have a very good small New York Jewish magazine in the coming years—a reason to affirm that Dissent’s moment continues.

Consider the context: Dissent was funded to push back against McCarthyism. We now are living through a new era of McCarthyism squared.

Dissent was founded to push back against totalitarianism, authoritarianism, and party lines. There are certainly authoritarian and party-line movements on the rise around the globe. Might any of them evolve in such a way that it might be proper to call them neo-totalitarian? Excuse me, I need to put in my hour of ticket-punching by studying the Xi Jinping Thought app on my smartphone. (And how should we characterize China today, a peculiar hybrid in which a Leninist party has been forcibly wed to a von Hayekian market economy, with Confucius officiating the shotgun marriage? And how will it evolve?)

Dissent was founded to push back against leftist party-lines—whether Stalinist, Trotskyite, Schachtmanite, or whatever. It took the perspective of, once again, Keynes—in this case Keynes reviewing Trotsky:

John Maynard Keynes: Reviewing Trotsky on England <https://www.marxists.org/history/etol/document/comments/keynes01.htm>: ‘That is how the [British Labour Party] gentlemen who so much alarm Mr. Winston Churchill strike the real article…. Granted his assumptions, much of Trotsky’s argument is, I think, unanswerable…. But… he assumes that… a plan exists, and that nothing remains except to put it into operation…. [but] we lack… a coherent scheme… none more conspicuously so than the Marxists…. The next move is with the head, and fists must wait…

Friedrich von Hayek argued that markets are indispensable: only through decentralized market mechanisms, he claimed, can societies efficiently allocate resources and coordinate the actions of millions. Jeremy Bentham, on the other hand, regarded inequality itself as a moral crime—an affront to the utilitarian calculus that seeks the greatest happiness for the greatest number—and that to fail to equitably distribute was as great a disaster as to fail to produce. Karl Polanyi took a somewhat different tack, warning that any society which recognizes only property rights—and neglects the broader spectrum of human rights—will inevitably become unstable and crash, at worst into Stalinism or Hitlerism. People, after all, know intuitively that they possess rights beyond those granted by mere private-property “ownership” of market-economy chattels. And a neoliberal attempt to deny that, to say the market giveth, the market taketh away: blessed be the name of the market, cannot be a truly good society.

Zohran Mamdani will sit in the middle of what is still one of the greatest Neoliberal Order globalized value-chain economy set of production networks in the world, generating immense amounts of wealth. Zohran Mamdani’s task, if he is given the baton by the New York City electorate, is to tweak that set of production networks in the interest of making it work better in a manner that Bentham and Polanyi would approve.

In such a world, a magazine grounded in the anti-McCarthy, anti-totalitarian tradition has real strengths, especially if its core is a community that, for 2,500 years, has been taught—each spring—to remember their ancestors’ slavery in Egypt:

YHWH: The Book of the Levites <https://www.biblegateway.com/passage/?search=Leviticus%2019&version=KJV>: ‘If a stranger sojourn with thee in your land, ye shall not vex him. But the stranger that dwelleth with you shall be unto you as one born among you, and thou shalt love him as thyself; for ye were strangers in the land of Egypt: I am the LORD your God. Ye shall do no unrighteousness in judgment, in meteyard, in weight, or in measure. Just balances, just weights, a just ephah, and a just hin, shall ye have: I am the LORD your God, which brought you out of the land of Egypt. Therefore shall ye observe all my statutes, and all my judgments, and do them…

That kind of sociological power is not to be underestimated; it should be claimed and used.

So I see no reason that the future of Dissent should not be a very bright one.

References:

Leave a comment

Subscribe now

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…
#dissent-fundraising-event
#passing-the-baton
#public-reason

This Is Not an Administration. This Is Not Even a Renaissance Court. This Is, Pure & Simple, a Chaos-Monkey House

Dumbf***s in disarray: whiplash foreign-policy reversals & unreliability with the triumph of impulses over institutions, & what we learn from the Ukraine arms-shipment reversal…

Share

Share Brad DeLong’s Grasping Reality

Chaos-monkey policy yet again. July 8, 2025:

Economist: Trump embarrasses the Pentagon with a U-turn on Ukraine <https://www.economist.com/united-states/2025/07/08/trump-embarrasses-the-pentagon-with-a-u-turn-on-ukraine>: ‘His decision to resume arms shipments is a victory for common sense—while it lasts…. The Pentagon had presented the halt in arms deliveries [to Ukraine] as a part of a review to ensure that America maintained its own stocks. “We can’t give weapons to everybody all around the world,” declared Sean Parnell, the Pentagon spokesman… This was mostly misleading…. Those being sent to Ukraine now come from contractors’ production lines rather than from American forces…. Mr Parnell insists that the process to “evaluate military shipments across the globe remains in effect and is integral to our America First defence priorities”…. But that does not hide the humiliation of the Pentagon’s civilian leadership at having an important policy countermanded by the president…. Trump reportedly told Ukraine’s president, Volodymyr Zelensky, that he had not known about the halt in weapons deliveries…. Not even Mr Trump’s acolytes can predict his zig-zags. America First means what Mr Trump says, not what his ideologues and devotees think…

Give a gift subscription

It may well be true that Trump does not remember approving the cutting-off of weapons shipments to Ukraine. It might even be true that Trump did not intend to cut off weapons shipments to Ukraine—that he said something ambiguous and then Hegseth and others who want to do Putin a solid ran with it. Who knows? There is no way for anybody outside to know. There is probably no way for anybody inside who was not in whatever room it might (or might not) have happened in to know either.

Random people claim that they have the baton from the Great Leader and set things in motion. They may well be routinely lying. But checking in with Trump may (a) not be possible, and (b) not be wise—until © it shows up on some “news” show he watches, placing him in an unfavorable light because of the consequences of what was set in motion.

The Trump insiders have settled on a scapegoat;

Robert Farley: The Great Switcheroo <https://www.lawyersgunsmoneyblog.com/2025/07/the-great-switcheroo>: ‘Not obvious that anyone is in control of Trump administration foreign policy: “Russia launched a major volley of drones and missiles at Ukraine overnight on Wednesday, soon after President Trump had sharply criticized President Vladimir V. Putin of Russia for taking only “meaningless” steps toward peace in settlement talks…. The attack came the night after Mr. Trump had made his latest flip-flop… saying on Monday that because Ukraine was “getting hit very hard” in Russian attacks, he would resume a delivery of weapons that his administration had paused only last week…. 30 Patriot missiles and other munitions, was in Poland, waiting to cross the border, officials said…”. Looks like Elbridge Colby is the source of some of the confusion…

Leave a comment

Maybe?:

Jack Detsch, Nahal Toosi, Paul McLeary & Joe Gould: Pentagon policy chief’s rogue decisions have irked US allies and the Trump administration <https://www.politico.com/news/2025/07/08/elbridge-colby-trump-administration-frustrations-00443337>: ‘Elbridge Colby… mak[de]… a name for himself as an experienced, restraint-minded foreign policy leader…. But since joining the second Trump administration as the Pentagon’s top policy chief, Colby has… blindsided parts of the White House and frustrated several of America’s foreign allies… gotten out ahead of the administration on several major foreign policy decisions…. [He] prompted last week’s decision… to halt shipments… to Ukraine, which caught many Trump allies and lawmakers off guard…. “He is pissing off just about everyone I know…” said one person familiar with the situation…. “He has basically decided that he’s going to be the intellectual driving force behind a kind of neo-isolationism that believes that the United States should act more alone, that allies and friends are kind of encumbering,” said a person familiar with the Trump administration dynamics….

William Martin, communications director to Vice President JD Vance, called Colby “a consummate professional, an experienced national security official and a reliable team player… wholly committed to President Trump’s America First foreign policy agenda.” Pentagon spokesperson Sean Parnell said Colby is doing exactly what would be expected… “provid[ing] policy recommendations to Secretary Hegseth, and his advice has proved to be invaluable…. [There is] zero daylight” between Defense Secretary Pete Hegseth and Colby…

Get 50% off a group subscription

The Politico reporters’ claims that Colby has “blindsided parts of the White House…”, “gotten out ahead of the administration”…, “prompted last week’s decision… to halt shipments… to Ukraine…”, “caught many Trump allies… off guard…”, “caught… lawmakers off guard…”, “is pissing off just about everyone I know…” are all very strange locutions for somebody who has no operational authority at all:

  • It is not Colby, but rather Trump and Hegseth who caught lawmakers off guard.

  • It is not Colby, who can take no action, but rather Trump and Hegseth who must have “gotten out ahead of the [their] administration…” one person familiar with the situation….

Slightly less nonsensical is:

  • “‘He has basically decided that he’s going to be the intellectual driving force behind a kind of neo-isolationism that believes that the United States should act more alone, that allies and friends are kind of encumbering’, said a person familiar with the Trump administration dynamics…

Refer a friend

In short: people are telling their favorite Politico reporters that they do not like what Colby is telling Hegseth and think that it is garbage, but the problem is that Hegseth agrees with it and then Trump agrees with Hegseth. It is clear that these people with favorite Politico reporters do not push back. Perhaps they dare not push back in the moment because they know Hegseth will agree with Colby. Perhaps they cannot push back because Colby gets lots of one-on-one “by the way” face-time with Hegseth, and they cannot break into that ring.

In either case, these people are getting their favorite Politico reporters to help them set things up so that when Trump and Hegseth decide that somebody must be blamed for administration missteps, it will be Colby who is thrown out of the troika-sleigh to be devoured by the wolves as the scapegoat.

One would laugh. But:

The absence of even a pale shadow or a bad approximation to a coherent policy process under Trump’s administration renders both allies and adversaries unable to predict American actions. That undermines global stability. In the world of international relations, predictability is not a luxury but a necessity. Nations calibrate their own policies, investments, and security arrangements based on expectations about the behavior of others—particularly when it comes to a superpower like the United States.

During the Cold War, for example, the doctrine of “containment” and the logic of mutually assured destruction provided a framework—albeit a terrifying one—within which both the United States and the Soviet Union could operate. Allies could plan. Adversaries could strategize The world, for better or worse, had a certain stability. The sea of chaos could be kept within some bounds

By contrast, when American foreign policy becomes a function of—I was going to say “presidential whim, as it has under Trump”, but that is wrong, isn’t it? That imposes more structure on the process than it in fact has. Chaos monkey is the only term even half-adequate. And so the result is a kind of international vertigo. European governments, for instance, have found themselves scrambling to respond to abrupt reversals on NATO commitments, trade policy, and climate agreements. Adversaries, meanwhile, are emboldened to test boundaries. American red lines, they rightly conclude, are more likely to be rhetorical than real.

The net effect is a world system where trust erodes and the risk of miscalculation—and thus catastrophe—increases.

The lack of institutional memory and policy continuity means that major decisions—such as arms shipments to Ukraine—can be reversed or forgotten on a whim, with no accountability. In any functioning government, continuity is provided by career civil servants, established procedures, and a respect for precedent. These mechanisms ensure that policy is not simply the artifact of a single individual’s mood, but rather the product of deliberation, expertise, and accumulated experience.

Under Trump, however, these guardrails have been systematically dismantled. Arms shipments to Ukraine were halted. Arms shipments to Ukraine were resumed. Colby is blamed as having “prompted” the halt—as if Hegseth and Trump were actors on a stage who had forgotten their lines, and Colby in the wings was a malicious Svengali whispering the wrong lines to them. To even wonder whether this was a deliberate strategic recalibration or merely the result of presidential inattention is to commit a category error that it was more than random chaos.

The Trump administration’s penchant for reversals—whether on the Iran nuclear deal, the Paris climate accords, or military deployments—has taught the world that American commitments are not.

This collapse of administrative norms and the rise of personality-driven rule echo the worst tendencies of 20th-century authoritarian regimes. Throughout the last century, the world witnessed the dangers of governments that subordinated institutions to the whims of a single leader. Whether in Mussolini’s Italy, Stalin’s Soviet Union, or Perón’s Argentina, the pattern is familiar: rules give way to personal loyalty, expertise is replaced by sycophancy, and policy becomes a theater of flattery and intrigue. While the United States has not descended into full-blown authoritarianism, the Trump administration’s disregard for process, preference for loyalists over professionals, and willingness to flout established norms mark a dangerous drift. For students of history, this is not mere melodrama; it is a warning. When government becomes a stage for the leader’s ego, rather than an instrument for the public good, the costs are borne not just by the nation, but by the world.

We know what fresh hell this is. And we know we cannot get out of it.

Leave a comment

Subscribe now

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…
#this-is-not-an-administration

FIRST DRAFT:

Draft review of John Cassidy: "Capitalism & Its Critics". 2025-04-07…

Share

Share Brad DeLong's Grasping Reality

Let me cut to the chase:

Where does the book that is John Cassidy’s Capitalism and Its Critics travel to and arrive at, at the end, To mix metaphors, Cassidy’s Capitalism and Its Critics is a polyphony. It is without a single dominant note, chord, or phrase. It, rather, has repeated themes—themes sounded again and again as Cassidy traverses 250 years and thirty-odd observers and critics of capitalism. For, in Cassidy’s words, “the central indictment of capitalism has remained remarkably consistent: that it is soulless, exploitative, inequitable, unstable, and destructive, yet also all-conquering and overwhelming…”

(1) Principally, to my ear, of Cassidy’s repeatedly sounded themes is that of capitalism’s critics not making up any sort of coherent whole but still very much worth listening to. At best, truly strange bedfollows. For, as Thomas Carlyle put it, the critics of capitalism (including himself) are or are transmitting not plans for a new and better order, but rather “popular commotions and maddest bellowings”. These commotions and bellowings “from Peterloo to the Place-­de-Grève… [are] inarticulate cries as of a dumb creature in rage and pain; [and] to the ear of wisdom… [are] prayers: ‘Guide me, govern me! I am mad, and miserable, and cannot guide myself!’…”. For, put much more charitably than Carlyle’s ventriloquism of the Chartist activists, the system violates what my friend Dan Davies sets as a sine quo non principle of human organizations. He believes “Big corporations and states need to have information channels… “red handle signals”… [to] bypass the normal hierarchy and get information to the decision making centre, in time…. The lack of that… has caused so many schemes to fail, and its continued presence… highly unsatisfactory and attenuated… accounts for the fact that the democratic industrialised world still does kind of work, a bit…”

But in capitalism there is only one information channel: the price. And—no matter what the Dr. Jekyll-Mr. Hyde genius-idiot Friedrich von Hayek may have claimed—that is not enough to keep any large-scale societal network-system from going off the rails.

But there is no consensus about what kind of, at a minimum, red-handle signals need to be built in. As John Maynard Keynes wrote a century ago in an earlier systemic crisis of the applied-science society capitalism of his age: “We lack more than usual a coherent scheme of progress, a tangible ideal. All the political parties alike have their origins in past ideas and not in new ideas…. No one has a gospel. The next move is with the head…” But we have been unable to even begin to make that move since the post-2007 breakdown of the authority of North Atlantic political economy’s Neoliberal Order

(2) Almost as prominent of the themes in the polyphony—at least to my ear—is Cassidy’s belief that now, finally, 250 years after he starts his survey, capitalism is ending with fragmentation and incoherence. Nearly half of Americans seeing capitalism as on the wrong path; new conservatives joining the left-wing critiques of globalization and financialization; leftists and conservatives alike blaming capitalism for “collapsed communities” and cultural degradation​; believers in state-guided capitalism vs laissez-faire anarchy techno-utopians hoping cyberspace will route around all governments; the vanishing of the New Deal Order dominant coalition of labor, scale-economy mass-production industry, and Keynesian technocrats; the decline of the Neoliberal Order alliance of globalization-embracing educated elites and free-market ideologues; hopes we will be saved by teamwork-based open-source; hopes that the gig economy might be a source of liberation (rather than of worker speedup, with the Amazon delivery drivers peeing in bottles as they bring things to my house, and my last Uber driver getting only 25% of what I paid the app); the right-wing populism of the billionaires; left-wing protest encampments, cooperatives, and artisanal networks pushing for “degrowth”, commons-based activism, and gift economies.

This is, I think, flatly wrong. People would have said the same in the mid-1930s, not seeing the New Deal Order of mass-production capitalism sprouting all around them. People did say the same thing in the 1970s, not understanding the speed with which capitalism was transforming itself into its globalized value-chain variant that would then so strongly underpin the Neoliberal Order. Karl Marx and Friedrich Engels in 1857 may have been the first to mistake the birth pangs of a new mutation of capitalism for the beast’s final death throws. Don’t make their mistake—as Cassidy almost does, before pulling back at the last.

(3) For, as Cassidy does stress repeatedly and strongly, capitalism mutates. Indeed, it mutates so many times and so comprehensively each time that I have given up using the bare word “capitalism”. These days I will teach about classical-world capitalism, mediæval capitalism, agricultural capitalism, colonial capitalism, mercantile capitalism, steampower capitalism, applied-science capitalism, mass-production capitalism, globalized value-chain capitalism, and now attention info-bio tech capitalism—you need at least ten, rather than to think that in some way it is the same beast (or draw only a single distinction between “capitalism” and “late capitalism”).

Since capitalism mutates, no one single system of societal and political regulation, no single political-economy order, can be both durable and satisfactory for long. The pseudo-democratic semi-liberal Belle Époque order of the steampower-age capitalism of the half-century before World War I did give humanity the best half-century it had ever had: “economic El Dorado”, as John Maynard Keynes put it, of previously unimagined general prosperity plus growing democracy and reduced hierarchy and oppression. But as steampower capitalism morphed into applied-science capitalism, the old political-economy order and attempts to restore it to health brought first World War I, and then a Great Depression with the concommitant rise of Fascism, Naziism, Stalinism, and Imperial Japan’s catching of a new bad variant of the western European nationalist-racist-imperialist virus. Only die-hard card-carrying Neoliberals still dare to claim today that the shift from the social-democratic New Deal to the Neoliberal Order was a satisfactory way of managing capitalism’s mutation from its mass-production to its globalized value-chain variant. And now we have no clue what political economy would allow human flourishing as capitalism’s genome shifts again to what we will perhaps call the capitalism of the attention info-bio tech age.

(4) Cassidy sees—especially now, but also repeatedly in the past—capitalism delegitimizing itself. I think here he suffers from an optical illusion, from a near-sighted myopia. Over its very long career, capitalism has very rarely been seen as legitimate by anyone. Yes: the post-1980 pre-2008 Neoliberal Order did contain an extravagant ideological claim that capitalism, or democratic capitalism, was in some sense a legitimate order. But not even Friedrich von Hayek saw it as in any sense moral. Hayek’s ideas were: (a) A market economy and private property—capitalism—make up a mighty and powerful human social engineering device for attaining prosperity by crowdsourcing the best attainable solutions to the problems of what to make and how to make it. (b) But it produces great inequality and unfairness, as it rewards not the truly deserving but rather those lucky enough to control the right resources at the right time. (c) Howver, alternatives fail to produce prosperity without producing anything anyone might ever reasonably call social justice; the alternatives reward not the lucky and the productive but rather the powerful—whether aristocracies, oligarchies, or some managerial “new class”. (d) Hence capitalism places us in the position of Job: we need to accept that “capitalism giveth, capitalism taketh away: blessed be capitalism’s name”, for kicking against that reality will produce the catastrophe of putting us on one of the many roads to serfdom.

It was only during the Neoliberal Order’s heyday that things went further. Only then there serious claims that greed was good per se—that success in the capitalist marketplace was the sole touchstone to true moral virtue and deservingness. Otherwise, capitalism has been seen as legitimate only when properly managed, either by a democracy or by some aristocracy of power, wealth, culture, or technocratic expertise.

And I hear many more repeated themes in this book of 260,000 covering thirty-odd observers and critics of the thing “capitalism” across two and a half centuries.

How did John Cassidy—one of the very best economic journalists we have here in the world today, and one of the very few who does not work for a business élite-oriented publication—come to write this book?

He says that he started out conceiving of his book as “a shortish history of contemporary capitalism and… economic debates… from the collapse of the Soviet Union…”. to the present. He quickly found himself, as one does, being dragged further back into the past: ending up reading Adam Smith on the “privileged position and egregious self-­dealing” of the East India Company, concluding that most “criticisms of modern capitalism” are rooted in “developments and debates… [from] the rise of industrial capitalism in Britain and even before that”, and so picking his thirty-odd subjects.

Having myself just three years ago published a book that came in at 190,000 words (as I failed to cut more of my own original half-million words of drafts and notes), I understand how Cassidy wound up here with this book inside one single set of covers. But its length makes reading the book a major commitment. For those of us who choose to listen to the audiobook at 1.5-speed it will take twenty-eight hours. For the typical reader of Democracy Journal, it would take seventeen. I am fortunate and blessed: it took me seven from cover-to-cover.

But if we really want the books we read to become a true part of us and our mental panoply, we need to do more than just read it cover-to-cover, once. We need to read actively. We start with the black squiggles on the page. From them we spin up a sub-Turing instantiation of the mind of the author. We then run that sub-Turing instantiation on our own wetware—and argue with it. We argue with it while reading, while reflecting, in the shower, and walking the dog. And so we truly learn. By the end, I had spent twenty hours with John Cassidy in the form of this book (plus more: I remember well his New Yorker articles—“The Fountainhead”, “Smoke Signals”, “Motown Down”, “A World of Woes: A Global Take on a Decade of Financial Crises”, and more). And then I sat down to write this review.

Was it worth spending my time on this, rather than on something else? For me, Cassidy’s book gave me:

  • a visa to seven thinkers who are important guys whom I readily did not know, and ought to have: Bolts, Thompson, Wheeler, Tristan, Kondratiev, Kumarappa, Georgescu-­Roegen, and Federici.

  • the opportunity to revisit and rethink my scant knowledge and unformed views about six more: Carlyle, George, Veblen, Hobson, Sweezy, and Williams.

  • and the John Stuart Millian impulsion to take seriously what a smart person who thinks differently than me thinks about issues and authors that I think I know very well indeed—Smith, General Ludd, Engels, Marx, Luxemburg, Keynes, Kalecki, Robinson, Prebisch, Friedman, Hall, Hayek, Amin, Rodrik, Stiglitz, and Piketty.

For me, reading and thinking and then writing up this review of Cassidy was very much worth doing. Perhaps I feel like Yakov at Jabbok’s Ford, having wrestled all night with the shadowy figure that was the High God El—and emerged having gained strength, for El only won as dawn was breaking by cheating. For me, it was a great treat. It was very much worth doing. And not because I will be paid $850 by Democracy Journal.

Is it worth reading Cassidy from cover-to-cover? Well, what else would you do with your time? What other big books are you going to read this year? if you buy it and start it, you will indeed find on every page something that will make you glad to have read the page, and enough narrative forward momentum to make you eager to turn the page. But the pile of very good books you might read this year is high. And Capitalism and Its Critics is likely to crowd two others out of your reading life.

Is it worth buying Cassidy and dipping into the text periodically, even if you do not think you will ever manage to read it cover-to-cover? Here again I say: Yes. If you have a book budget big enough to include things you are not overwhelmingly likely to read cover-to-cover, John Cassidy’s Capitalism & Its Critics belongs inside it.

But suppose that you are mostly interested in understanding how to interpret references to this book as it serves, as I think it is very likely to, as an intellectual buoy to help you orient yourself in our current sea of troubles. Suppose you want some idea of what this book is about to help you understand what will probably be a number of conversations you will have and overhear that will refer to it. Or suppose, less charitably, that you want to appear a knowledgeable, erudite, au courant person.

What then?

Well, that is one thing that reviews like this are for, isn’t it?

Give a gift subscription

References:

Get 33% off a group subscription

Leave a comment

Subscribe now

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…

What Manufacturing Production Is Not Good For

Why chasing factory jobs for their own sake really is a dead end—and what policymakers should do instead. It is unions and social-insurance states, not assembly lines, that hold the key to global-north blue-collar prosperity in the 21st century…

Share

Share Brad DeLong’s Grasping Reality

I was rather annoyed at a piece from the Economist last month. But the piece did make one very good point:

Economist: The world must escape the manufacturing delusion <https://www.economist.com/leaders/2025/06/12/the-world-must-escape-the-manufacturing-delusion>: ‘Today’s zeal for homegrown manufacturing…. Politicians want to revive well-paying factory work and restore the lost glory of their industrial heartlands…. Politicians hope that boosting manufacturing means decent employment for workers without university degrees…. But factory work has become highly automated…. The good jobs created by today’s production lines are for technicians and engineers, not lunch-pail Joes. Less than a third of American manufacturing jobs today are production roles carried out by workers without a degree…. Bringing home enough manufacturing to close America’s trade deficit would create only enough new production jobs to account for an extra 1% of the workforce. Manufacturing no longer pays those without a degree more than other comparable jobs in industries such as construction…

Give a gift subscription

However, it could have made it better. The point that manufacturing is now too automated for it to be a sizeable part of demand for blue-collar workers is 100% on. But it is only a small part of the story of why manufacturing-promoting industrial policy cannot be an appreciable part of solving income-distribution of blue-collar opportunity problems.

So let me expand:

Manufacturing jobs have typically and historically been crappy: physically taxing, repetitive, and often hazardous involving long hours in environments that are noisy, dirty, and sometimes dangerous. The tasks themselves are frequently routinized and offer little in the way of intellectual stimulation or autonomy. And the compensation for such work is determined by the impersonal forces of supply and demand in the labor market, not by any inherent dignity, social value, or romanticized view of the maker or craftsman. When there is a surplus of workers willing to take on these roles, wages stagnate or decline, and the jobs remain “crappy” in the sense that they offer little if any more than alternatives in exchange for hard labor.

The notion that manufacturing jobs are inherently “good” is a myth rooted more in nostalgia than in the lived experience of most factory workers, especially in the absence of strong institutional supports.

Now not all manufacturing positions are created equal. The cigar-factory cigar-rolling jobs of Samuel Gompers were good jobs for the day. A minority of roles within the sector—typically those requiring specialized skills, technical expertise, or significant experience—can offer higher pay, greater job security, and more interesting work: think of tool-and-die makers, machinists, or practical engineers who optimize production lines on the sho[ floor. Such roles are insulated from downward wage pressure because the skills required are scarce, requiring substantial, training, apprenticeship, or a deep understanding of complex systems. Employers must thus pay more to attract and retain scarce talent. These positions might involve operating or maintaining complex machinery, overseeing production processes, or engaging in quality control and innovation.

The conditions under which manufacturing gives rise to what might be called “labor aristocracy” jobs—positions offering above-average pay, status, and security for blue-collar workers—have always been contingent on the broader economic and technological context of their era.

During the epoch of the Commercial-Imperial Economy, for example, the skilled operators of England’s stocking-frames under the Tudor, Stuart, and Georgian dynasties occupied a privileged niche. These artisans commanded respect and relatively high wages, not because of any inherent generosity on the part of employers or the state, but because their expertise was scarce and the machinery they operated was both valuable and difficult to master. Their bargaining power derived from the complexity of their craft and the limited supply of workers able to perform it. In this context, the labor market rewarded rare skill sets, and the resulting “aristocracy of labor” was a function of both technological constraints and institutional arrangements that kept entry into these trades tightly controlled.

The advent of the Industrial Steam-Power Economy in the nineteenth century radically altered this balance. The rise of mechanized production meant that machines could increasingly substitute for skilled hands, transforming the nature of work on the factory floor. Tasks that once required years of apprenticeship could now be performed by unskilled or semi-skilled workers, guided by the rhythms of the machine rather than the judgment of the craftsman. This shift did not just deskill labor; it also compressed the working class into a more homogeneous proletariat, with fewer avenues for distinction or upward mobility based on skill alone. The “labor aristocracy” shrank as the premium on specialized expertise eroded, and the vast majority of manufacturing jobs became routine, repetitive, and interchangeable. The social structure of the industrial workforce flattened, and the economic rewards of manufacturing work became more closely tied to the brute logic of supply and demand for generic labor.

Subsequent epochs—each defined by their own technological and organizational revolutions—have continued to reshape this landscape of manufacturing employment and the “labor aristocracy”:

  • The Second Industrial Revolution Applied-Science Economy, driven by applied science and mass production, created new opportunities for technical specialists and engineers, but also further routinized the bulk of factory work.

  • The New-Deal Order Mass-Production Economy, with its emphasis on large-scale, unionized mass production, briefly expanded the ranks of “good” blue-collar jobs, but only under the unique institutional arrangements of the mid-twentieth-century North Atlantic.

  • The Neoliberal Order Globalized Value-Chain Economy, characterized by globalized value chains and relentless cost-cutting, eroded these gains by shifting production to lower-wage regions and automating what remained.

  • Today, in the Attention Info-Bio Tech Economy, the locus of economic value has shifted even further away from traditional manufacturing. The skills most highly rewarded are increasingly cognitive, creative, and technical, and the opportunities for blue-collar workers to secure “aristocratic” positions through manufacturing alone have dwindled to a vanishing point.

Leave a comment

Each successive transformation has further undermined the notion that manufacturing, by itself, can serve as a stable foundation for broad-based working-class prosperity.

But there is a much more important factor than all of those: unions.

Unions have historically played a crucial role in transforming manufacturing jobs from “crappy” to “good.” The difference between a bad unskilled manufacturing job and a good semi-skilled one is a strong union—either in being, or as a union threat. The classic example is the United Auto Workers in the mid-20th century, who negotiated contracts that made assembly line work a ticket to the middle class. The presence of a strong union can turn even relatively low-skill, repetitive work into a stable and respectable livelihood.

The history of mass-production in America is intertwined with the rise of industrial unionism. Henry Ford’s famous $5/day wage was not an act of benevolence, but a calculated move to reduce turnover and stave off labor unrest, particularly from radical unions like the Industrial Workers of the World (IWW). Mass-production settings, with large numbers of workers concentrated in a single location, are more amenable to union organizing than dispersed, small-scale operations. Yet, even in these environments, unionization was never guaranteed and often depended on the broader political and legal context. The gains made by unions in the mid-20th century were the result of hard-fought battles, not the natural outgrowth of the production process itself.

The extent to which manufacturing jobs are unionized—and thus “good”—is predominantly a political question. Labor law, government enforcement, and the broader climate of public opinion all shape the ability of workers to organize and bargain collectively. In periods of pro-labor policy, such as the New Deal era, union density soared and manufacturing jobs improved accordingly. In more hostile environments, unions have been weakened or broken, and the quality of manufacturing work has declined. As DeLong has emphasized, the fate of the working class is not determined solely by economic forces, but by the outcome of political struggles over the rules of the game.

It’s not just manufacturing: the benefits of unionization and pro-worker policy can extend to a wide range of blue-collar and service sector jobs. When the political environment supports collective bargaining, minimum wage laws, and robust social safety nets, the floor for job quality rises across the board. Sectors like construction, transportation, and even retail can offer stable, well-compensated employment if workers have the power to demand it. The focus on manufacturing as the sole source of “good jobs” is thus misplaced; what matters is the institutional framework that governs all forms of work.

If the goal is to create more “good” jobs for workers without college degrees, the most effective lever is to strengthen unions and the rights of workers to organize. The evidence is clear: unionized workers earn more, have better benefits, and enjoy greater job security than their non-union counterparts. Policies that make it easier to form and join unions, protect collective bargaining rights, and penalize employer retaliation are essential to reversing the decline in job quality. The solution to the crisis of blue-collar opportunity lies not in nostalgia for a lost manufacturing golden age, but in rebuilding the institutions that once made those jobs desirable.

The era of mass employment in manufacturing is over, and it’s not coming back. Decades of technological progress have made it possible to produce more goods with fewer workers. Automation, robotics, and advanced manufacturing techniques have steadily reduced the demand for human labor on the factory floor. Even as output rises, employment falls—a trend that is likely to accelerate in the coming years. The hope that reshoring or industrial policy can restore the lost world of plentiful, well-paid factory jobs is a delusion. The future of work will be shaped by the relentless advance of technology, and the challenge is to ensure that the gains from productivity are broadly shared, rather than concentrated at the top.

If factories employing large numbers of unskilled blue-collar workers should come back to the U.S., the jobs they provide will be among the worst jobs in our economy.

Do put a tickler here, however: I do want to circle around to talk about why the rest of the piece did leave me annoyed.

Get 50% off a group subscription

References:

Leave a comment

Subscribe now

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…
#what-manufacturing-production-is-not-good-for

Ninety Days for Zero Deals, & Zero Certainty for the Future

Trump’s chaos-monkey trade policy looked at once again. No “art of the deal'“, but only a cascade of threats, double-crosses, and boasts; thus uncertainty reigns supreme and, behind, the bravado, America’s trading partners see how appeasement offers no protection from policy whiplash…

Share

Leave a comment

Chaos-monkey Trump vowed 90 trade deals in 90 days; the world got a handful of vague frameworks and a blizzard of threats. As deadlines shift and details evaporate, economic actors are left to navigate a landscape defined by uncertainty. America’s partners must choose: flatter, fight, or flee. The only real constant? The random walk of economic policy in the age of the chaos monkey:

Paul Krugman: The Tariff Beatings Will Continue Until Morale Improves <https://paulkrugman.substack.com/p/the-tariff-beatings-will-continue>: ‘Investors… deluded themselves into believing that Trump was done disrupting world trade…. By now we were supposed to have scores of trade deals signed. Instead, yesterday Trump began posting letters on Truth Social (diplomacy!) telling a variety of countries that they would face high tariffs on Aug. 1… [that] look very close to the widely ridiculed… tariffs announced on April 2….

These tariffs are really, really high…. In the long run 25 percent tariffs will reduce trans-Pacific trade by almost 60 percent…. There were never going to be genuine trade deals…. Why didn’t Korean and Japanese negotiators make big enough concessions to satisfy Trump? Because there was nothing for them to concede…. Neither country… [coffer big tariff reductions, because their tariffs were already minimal…. How were the South Koreans supposed to end unfair trade practices that exist only in Trump’s imagination?… Why make a deal with a man who will surely break it?... His letter to the South Koreans doesn’t even mention KORUS, let alone explain why the United States is reneging on its solemn promises…

Paul Krugman
The Tariff Beatings Will Continue Until Morale Improves
Read more
7 hours ago · 697 likes · 221 comments · Paul Krugman

Share Brad DeLong's Grasping Reality

I mean, what can you say by this point? We have said this before. But I guess we need to say this all again:

During Trump 1, Mexico and Canada played ball with Trump by “renegotiating” NAFTA—resulting in the so-called United States-Mexico-Canada Agreement (USMCA). This was not, as Trump and his acolytes claimed, a wholesale replacement of NAFTA, a huge deal, the transformation of thew worst deal in American history into the best. It was, rather, an incredibly modest update. More than half of the even slightly significant differences came in minor changes to the auto parts provisions:

  • NAFTA: Required 62.5% of a vehicle’s parts to be made in North America for tariff-free status.

  • USMCA: Raised this to 75%.

  • USMCA: 40-45% of auto content must be made by workers earning at least $16/hour.

Give a gift subscription

But Mexico and Canada allowed Trump to trumpet, without contradiction, that he had imposed an agreement on them that was massively more favorable to the U.S. and massively less favorable to them than NAFTA had been. The underlying logic was that that was just smoke and mirrors, and by pretending that they had bent the knee they had done Trump a favor, and so secured a stable relationship with him.

Nope.

Playing along with Trump got them nothing but a belief on Trump’s part that they were easy prey who could be bullied. Trump has continued to threaten tariffs on Canadian aluminum and steel. Mexico has found itself repeatedly in the crosshairs over migration and agricultural issues. It is not a transactional, sector-by-sector, deal-by-deal approach. It is random demands.

That pattern rules.

And so, at present, countries are casting about for strategies to manage the volatility emanating from Washington:

  • Some are attempting tto appease Trump with ostentatious displays of respect and deference—think of the carefully choreographed summits, the flattery-laden communiqués, and the willingness to grant symbolic victories in hopes of staving off substantive costs.

  • Others are experimenting with a more confrontational stance, threatening countermeasures or leveraging their own economic and political assets to force a reconsideration of U.S. demands. Consider the European Union’s readiness to threaten retaliatory tariffs or China’s calibrated responses to U.S. actions.

  • And all countries, prudently, are seeking to “derisk” their economic futures by diversifying supply chains, forging new alliances, and reducing their dependence on the American market. The rise of regional trade agreements in Asia and Europe, and the acceleration of “friendshoring” and “nearshoring” strategies, are all part of this broader effort to inoculate national economies against the caprice of U.S. policy.

  • In sum, the global trading system is being reshaped not by the logic of comparative advantage or the invisible hand, but by the need to get out of the way of the random actions of feces-flinging chaos monkey.

Leave a comment

Coming from the Trump side, the Bessent affinity, right now we see an attempt to project confidence, with claims that the threats of tariffs are being used as leverage, and that the door is open for deals and deadline extensions beyond August 1. Bessent claims that things are “mov[ing]… along”.

But cast your mind back to April 12. “We’re going to make 90 deals in 90 days” said Peter Navarro. 90 days after April 12 would be July 10: Thursday. How many deals have been struck?

By my count, zero. Trump’s people claim that Trump has reached three “deals”: with the U.K., China, and Vietnam. But what are these deals, anyway?

For the UK:

  • The UK promises to reduce or eliminate “numerous non-tariff barriers”, which are not specified.

  • The UK promises to reduce or eliminate non-tariff barriers on U.S. agricultural exports, especially beef and ethanol, insofar as those are consistent with its health and safety regulatory régime.

  • The US promises to waive the 25% tariff on UK steel and aluminum, but when and how depends on future negotiations and on UK compliance with US national-security supply-chain requirements.

  • The US and UK commit to negotiating preferential treatment for UK-produced pharmaceuticals, details not specified.

  • There are two actual provisions:

    • The U.S. will allow up to 100,000 UK-made cars per year—more than its 2024 exports—to enter at a combined 10% tariff (7.5% “Liberation Day” tariff plus the standard 2.5% MFN rate), and UK auto parts for use in UK vehicles also get the 10% rate.

    • The US exempted some specified UK aerospace products (e.g. Rolls-Royce engines) from new tariffs it was imposing.

Get 50% off a group subscription

Those are the only actual “deals”. There is, literally, nothing more there. The rest of the deal is literally nothing but a mutual promise to make a deal, without any fallbacks or triggers to provide any incentives to do so

That is typical. And now the Bessent affinity is saying that they are hoping to “sign” perhaps ten such non-deal deals by September 1. Hoping.

The real outcome is a climate of uncertainty that discourages investment and undermines confidence in the durability of international commitments. In this brave new world, policy risk—not comparative advantage—has become the prime mover of global commerce.

At least we are going to get an experiment in how much damage grave policy uncertainty can do to an economy, and how quickly.

Refer a friend

References:

Leave a comment

Subscribe now

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…

#90-days-for-0-deals
#paul-krugman

Trumpxit & the Likely Undoing of American Exceptionalism

The fragile architecture of American-led economic and technological exceptionalistic prosperity faces its greatest test since its origins in the years after the American Civil War. Ricardo Hausmann reminds me of his & Federico Sturzenegger’s “dark matter” argument with respect to the true shape of the globalized value-chain economy. And I once again find him convincing—and so ramp up my estimates of the likely cost of the TRUMPXIT coming from chaos-monkey Trump and his grifter-sycophants. It is important to recognize that the tariffs imposed do not have to be high. They merely have to be uncertain. And everyone has to recognize that interdependence with the United States will be weaponized by Trump, and his likely Republican successors…

Share

Share Brad DeLong’s Grasping Reality

Trump’s erratic trade policies are much more than a political circus. They seriously and significantly do threaten to unravel the invisible networks that have powered American prosperity for decades. As the U.S. turns inward, it risks losing its role as the world’s economic hub and the benefits that come with it.

The case that TRUMPXIT—the chaos with respect to future trade barriers that Donald Trump’s on-again off-again on-again off-again tariffs impose on those planning the future of production networks in the world’s globalized value-chain economy—will make the United States markedly poorer over the next decade is very strong. It is this:

Start from Ricardo Hausmann’s estimation of the true situation:

Richardo Hausmann: Trump’s War on the “Dark [Economic] Matter” that Powers America <https://www.project-syndicate.org/commentary/hidden-forces-driving-us-prosperity-at-risk-under-trump-by-ricardo-hausmann>: ‘Traditional accounting… US… cumulative current-account deficit of $14.4 trillion between 2000 and 2024…. Net interest payments should have risen by $576 billion. But…[they] declined by only $19 billion…. The missing $557 billion?… A $295 billion surplus in cross-border services… [and] US subsidiaries abroad generated $2.1 trillion in sales, compared to $1.5 trillion by foreign subsidiaries… in the US… a net services surplus of $895 billion…

Give a gift subscription

The math does not work out right, quite. But in fact we do not know what the right numbers are to plug into this calculation. We do know, however, that net US services exports are massively undercounted precisely because the net recorded deficit does not track the estimated net asset position, and the estimated net asset position does not track the net recorded income flows.

Donald Trump believes that the United States suffers from large trade deficits because of economic decline and foreign exploitation, and none of the grifter-sycophants who surround him dare say him nay. But:

The US’s true economic strength lies in exporting the economic products of its intangible assets of ideas, technology, and expertise. Those generate significant global income. Those substantially offset traditional trade deficits. Standard international-trade accounting misses the value of these intangible assets, which Federico Sturzenegger and Richado Hausmann call “dark matter”. In reality, it is indeed the case that the US earns very high returns from its knowledge-based investments abroad.

It is the stability and predictability of U.S. trade and investment policy have long been the bedrock upon which global firms have built their confidence in embedding themselves within American-led production networks. When the rules of the game are clear, consistent, and not subject to sudden, capricious change, multinational corporations can plan for the long term, allocate capital efficiently, and integrate their operations across borders with minimal risk. This environment, painstakingly constructed over decades, has enabled the United States to serve as the organizing center of the world’s most complex and productive value chains.

Trump’s tariffs and his consequent retreat from open markets threaten this system.

Trump’s chaos-monkey moves toward unpredictability for the sake of media highlights undermine this foundation, prompting firms to hedge their bets, diversify away from the U.S., or seek more reliable partners elsewhere. The cost of eroding this trust is not merely transactional; it is systemic, weakening the very architecture of global commerce that has underpinned American prosperity.

The so-called “rules-based international order” is mucch more than diplomatic boilerplate. It is a framework of agreed-upon norms, institutions, and dispute-resolution mechanisms that has allowed the United States to shape global commerce and innovation according to patterns that serve its interests. By championing open markets, secure property rights, and the free flow of goods, capital, and knowledge, the U.S. has not only fostered global prosperity but has also ensured that the world’s most dynamic and lucrative value chains are anchored in American law, finance, and technology. When the United States is seen as the guarantor and enforcer of these rules, it reaps the benefits of trust, centrality, and influence. It is its hub status that has conferred enormous economic, technological, and geopolitical advantages, allowing the U.S. to set standards, capture rents, and shape the direction of global progress. To willingly forfeit this position through self-imposed isolation would be to squander a uniquely valuable historical legacy.

As it abdicates this role, it will lose its lion’s share of the benefits from this system. As people wake up to chaos-monkey Trump’s abandonment of US leadership in the rules-based international order, the US achieves self-imposed strategic and commercial isolation. Allies and rivals alike adapt by strengthening ties elsewhere and reducing reliance on American markets, partners, security arrangements, and other institutions

Moreover, the U.S.’s true economic strength is also strongly rooted in its openness to people, capital, and ideas. Turning inward undermines research and talent inflows. Those erode the knowledge base that sustains its economic model.

Openness to people—especially immigrants—has always been the secret sauce of American economic dynamism, just as it was for the political-military-imperial dynamism of Rome during the mid- and late-Republican and early-Empire . The United States’ willingness to welcome talent, ambition, and ingenuity from around the globe has fueled waves of innovation and entrepreneurship, replenishing the nation’s stock of human capital and driving forward its technological frontier.

Immigrants have been tremendously overrepresented among America’s inventors, entrepreneurs, and scientific leaders, and their children have often risen to even greater heights. Closing the door to this flow, or even signaling that the welcome mat is being withdrawn, does not merely reduce the number of workers or students; it atrophies the nation’s capacity to renew itself, to adapt to shocks, and to remain at the cutting edge. The lesson from history is clear: societies that turn inward and stifle the inflow of new people and new ideas soon find themselves stagnating, outpaced by those that remain open.

Repeatedly, the United States has benefited from being the world’s “network hub” for talent, capital, and ideas. This centrality has not been automatic; it has been earned and maintained through openness, reliability, and a demonstrated capacity to absorb immigrants and diffuse innovation at scale. These network effects are powerful: the more the U.S. becomes the preferred destination for the world’s best and brightest, the more attractive it becomes for the next wave of talent, and the greater the returns to those already invested in its system.

As the United States withdraws from its traditional role, the rest of the world will not simply sit idle. Other countries and regions are already moving to construct their own globalized value-chain networksvwithout the United States. These alternative networks may never fully replicate the scale or dynamism of the American-led system, but they can be robust enough to ensure that much of the potential value is still realized—just not by the United States.

One lesson I draw from the 1870-2010 “long twentieth century” is that technological and institutional revolutions, when paired with openness, have the capacity to drive decades of shared prosperity. The great leaps forward in productivity, living standards, and knowledge that defined the era were not the product of insularity or protectionism, but of societies willing to embrace change, experiment with new forms of organization, and welcome the contributions of outsiders. These gains, however, are not guaranteed; they are fragile, contingent, and reversible. Periods of openness and growth have often been followed by backlashes, retrenchment, and decline when complacency or fear took hold.

History does not repeat itself, but it does rhyme. The collapse of Bismarck’s alliance system after 1890 and the rise of protectionism after World War I serve as cautionary tales. Both were instances in which complex, stabilizing international arrangements—painstakingly constructed and maintained—were dismantled rapidly, often through a combination of carelessness, hubris, and short-sightedness. The consequences were catastrophic: geopolitical instability, economic dislocation, and, ultimately, global conflict.

Institutions and networks that sustain prosperity and peace are far easier to destroy than to build.

And in a world of shifting alliances and rising powers, the opportunity costs of withdrawal are measured not only in lost economic output but also in diminished influence and relevance.

Get 50% off a group subscription

References:

Leave a comment

Subscribe now

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…
#chaos-monkey-trump
#american-deglobalization
#TRUMPXIT

The Few Who Move the Many: Standout & Straggler Firms in Leading-Country Industrial Productivity

Very interesting indeed from the Jan Mischke-led MGI team. In leading technological-frontier economies, real economic growth and change may well come in seismic bursts, often led by a single company. The story appears to be not one of slow diffusion, but of a handful of risk-takers dragging the rest of us, often unwillingly, into the more productive future. What does this mean for policy, wages, and the future of economic dynamism?…

Subscribe now

Very interesting indeed:

Chris Bradley & Jan Mischke: The Heroes of Productivity <[www.project-syndicate.org/commentar...](https://www.project-syndicate.org/commentary/productivity-growth-driven-by-standout-firms-outweighing-stragglers-by-chris-bradley-and-jan-mischke-2025-05>:) ‘Productivity growth…. New research from the McKinsey Global Institute shows that the lion’s share comes from just a few firms making audacious moves.… 8,300 companies in Germany, the United Kingdom, and the United States… retail, automotive and aerospace, travel and logistics, and computers and electronics… between 2011 and 2019….

Fewer than 100 productivity “standouts” account for two-thirds of the sample’s growth… a much more extreme concentration than the prevailing view of productivity would suggest…. Standouts… in our sample, they pursued five strategic plays… [1] scaling more productive business models or technologies… [w] shifting regional and product portfolios toward the most productive [uses]… [3] reshaping customer value propositions… [4] building scale and network… and [5] transforming operations to boost labor efficiency…. Our case studies highlight Apple’s strategic expansion into services, EasyJet shaping the discount airline trend, and Zalando pioneering apparel e-commerce….

Share Brad DeLong's Grasping Reality

The 44 US standouts outnumbered the 14 stragglers by a factor of three. By contrast, the UK had a more even distribution, with 30 standouts and 25 stragglers, whereas Germany had 13 standouts and 16 stragglers. But it’s not just about having standouts; resources must be shifted toward them. The US was itself a standout in this respect, with half its productivity growth coming from reallocating labor away from stragglers and toward the vanguard…

Give a gift subscription

What follows from this and other similar studies of the concentration of productivity growth in the richest technology-frontier countries?

Focus on the multifaceted nature of the modern large firm. Such firms function as:

  1. technology-forcers—agents that introduce, develop, and diffuse new technologies;

  2. production-network orchestrators: valuable for stakeholders as key nodes in complex webs of suppliers, customers, and partners, enabling the creation of substantial economic value for each participant;

  3. investment vehicles—repositories for capital seeking returns, whether through dividends, stock appreciation, or other financial mechanisms;

  4. casino gambling roulette wheels—arenas for speculation, where fortunes are made and lost based on the perceived future prospects of the enterprise; and

  5. meme-stock symbols and gestures—pledges of econo-cultural allegiance, as seen in recent years with the rise of GameStop and AMC, where stock ownership becomes a form of identity politics.

Leave a comment

The first three of these are clearly useful. The fourth is a toll that those who suffer from the gambling drive pay, a toll that can be harnessed to create liquidity for the good of others. The fifth is just weird, especially these days.

I had already long thought that the first role, that of the firm as a technology-forcer, is the most important for economic growth and prosperity. Studies like this one from MGI suggest that that is even more crucial than I had previously appreciated.

Consider a canonical example: Apple’s creation of the modern smartphone market. It created a new category of consumer electronics. It also forced competitors to imitate Apple’s innovations and attempt to further innovate themselves, or die. Most died.

The same could be said of Toyota’s introduction of lean production or Amazon’s relentless pursuit of logistics efficiency. These are, none of them, incremental improvements diffused gently across the economy; they are seismic shifts in leading sectors, often led by a single firm willing to take risks, invest heavily, and, crucially, force others to respond.

Economic history is full of such moments: the Ford Motor Company’s assembly line, IBM’s mainframe dominance, Intel’s microprocessor revolution. The lesson is clear: technological dynamism is not a collective endeavor, but rather the result of a few audacious actors dragging the rest of us, sometimes kicking and screaming, into the future.

Next, we must ask: what share of the society-wide benefits generated by these standout firms is internalized by the firms themselves? How large are the incentives to push technologies of nature manipulation and human coöperative organization forward? When a firm like Apple expands, does it significantly raise real wages across the economy? The evidence suggests not. The effect of a single firm’s expansion on overall wage levels is, in most cases, vanishingly small. The bulk of the gains—higher wages, greater profits—are captured within the firm, accruing to its workers and shareholders. The “spillover” effects, though real, are limited at the level of the individual firm. This is not to say that such firms do not matter for the broader economy—they most certainly do—but rather that the transmission mechanism from firm-level success to society-wide prosperity is, in this channel, weak.

Your standard-issue economist would then say that there is no case for government action here: The societal returns to the decisions of the firm are felt by the firm on its bottom line. No external benefits (or costs). Hence no case for monkeying with the von Hayekian discovery procedure that is the market.

But a firm is not a rational Benthamite entity responding to pleasure and pain in order to maximize its utility. Managers would need to have the information, the incentives, the organization structures, and the bureaucratic procedures to pursue profit maximization with relentless focus. But, as any student of management—or, for that matter, of organizational behavior—quickly learns, this is a convenient fiction. Firms are not homunculi with a single brain. Firms are sprawling, multi-layered social systems, riven with principal-agent problems, informational bottlenecks, and the inevitable friction of bureaucratic inertia. Given the huge range of economies of scale and scope in the modern economy, the fact that there are multiple firms in the same line of business is a sign that this deviation of firm behavior from profit maximization is large. Information gets distorted as it moves up and down the hierarchy; feedback loops are often delayed or broken; local satisficing trumps global rationality.

Take, for instance, the infamous case of Kodak—a firm that, by the 1980s, had every financial incentive to embrace digital photography but whose internal structure (and perhaps culture) militated against it. The engineers in R&D saw the writing on the wall. The accountants, however, saw only the short-term threat to the high-margin film business. The result was a slow-motion train wreck: a firm that, despite having invented much of the relevant technology, failed to reorient itself in time.

Or consider the more recent example of Boeing, where the drive for short-run shareholder value—expressed through aggressive cost-cutting and outsourcing—undermined the firm’s ability to maintain quality and safety in its engineering processes, leading to the 737 MAX debacle.

When the market provides the right signals it will not matter much when the organizational machinery that interprets and acts on those signals is often deeply flawed. Thus, the standard economist’s case against intervention—claimed to be rooted in the assumption of rational, utility-maximizing firms—rests on a very different foundation that is, at best, highly contingent. What is that foundation? That there is no state capacity to take a longer view than that of the firm considered as a satisficing cybernetic entity under the gun of financial-market weirdness.

Moreover, the story does not end there. The technology-demonstration and -forcing effects—the impacts of standout firms on their competitors—is, I think, considerably larger than the stakeholder-value effect. When a firm pioneers a new technology or business model, competitors are forced to imitate or risk obsolescence. This emulation drives productivity gains across entire sectors. For example, Walmart’s mastery of supply chain management in the 1990s forced rivals like Target and Kmart to up their game, leading to a sector-wide transformation in retail logistics. Similarly, the adoption of just-in-time manufacturing, pioneered by Japanese automakers, eventually became the global standard. In this way, the benefits of innovation are not confined to the innovator; they diffuse, often rapidly, through the process of competitive emulation. Here, the “spillovers” are substantial, and the social returns to innovation can far exceed the private returns captured by the pioneering firm.

The direction of the policy implication is clear: there are large potential benefits to be gained from subsidizing successful standout firms—not only their innovation, but also their expansion. Traditional industrial policy has often focused on supporting struggling sectors or spreading best practices among laggards. Yet, if the real action is concentrated among a handful of standout firms, then perhaps resources would be better spent helping those firms scale up and diffuse their innovations more widely. What forms might this take? Targeted R&D subsidies? Support for workforce training? Permitting reforms that facilitate rapid expansion?

The history of Silicon Valley, with its admixture of public research funding, private venture capital, and a permissive regulatory environment, is potentially instructive here, at least as far as addressing these problems in leading sectors in the modern industrial ecology of the richest nations as we transition from the globalized value-chain economy to the attention info-bio tech economy?

And what about the laggard firms?

Finally, the comparative question: what is the United States doing right that Britain and Germany are not? The data show that the U.S. has both more standout firms and a greater willingness to reallocate resources—capital, labor, managerial talent—from stragglers to standouts. This is not merely the result of laissez-faire ideology, but of a complex interplay between market dynamism, labor mobility, deep capital markets, and a culture that, for all its faults, celebrates risk-taking and tolerates failure. By contrast, Europe’s more rigid labor markets, risk-averse financial systems, and penchant for preserving incumbents may inhibit the emergence and scaling of standout firms.

The lesson, I guess, is that fostering economic dynamism is less about propping up the many than about unleashing the few, and then relying on Schumpeterian creative-destruction to push resources around.

Get 50% off a group subscription

References:

Give a gift subscription

Leave a comment

Subscribe now

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…
#the-few-who-move

LIVE AT DEMOCRACY JOURNAL: Reviewing John Cassidy's "Capitalism & Its Critics"

Always great fun to write for! Entitled “Capitalism’s Next Act”, although I would have preferred something like “Capitalism as a Hydra” as a title: critics abound, but there points do not add up and coherence eludes, while protean capitalism keeps reinventing itself from Chartists to gig workers faster than its wannabe gravediggers can keep up…

Share

Capitalism’s Next Act

In moments of crisis, it’s easy to think that capitalism is finished. But don’t forget the system’s formidable ability to mutate.

By J. Bradford DeLong

from Summer 2025, No. 77 – 12 MIN READ

Prominent critics of capitalism, clockwise from top left: Flora Tristan, Karl Marx, Rosa Luxemburg, Joan Robinson, Karl Polanyi, J. C. Kumarappa, Eric Williams, Thorstein Veblen.

Share Brad DeLong’s Grasping Reality

Capitalism & Its Critics by John Cassidy • Farrar, Straus and Giroux • 2025 • 624 pages • $36

Let me cut to the chase:

Where does the book that is John Cassidy’s Capitalism and Its Critics travel to and arrive at, in the end? To mix metaphors, Capitalism and Its Critics is a polyphony. It is without a single dominant note, chord, or phrase. It, rather, has repeated themes—themes sounded again and again as Cassidy traverses 250 years and 30-odd observers and critics of capitalism. For, in Cassidy’s words, “

[T]he central indictment of capitalism has remained remarkably consistent: that it is soulless, exploitative, inequitable, unstable, and destructive, yet also all-conquering and overwhelming…

(1) Of Cassidy’s repeatedly sounded themes, the principal one to my ear is that of capitalism’s critics not making up any sort of coherent whole but still being very much worth listening to. And indeed they are, at best, truly strange bedfellows. As Thomas Carlyle put it, the critics of capitalism (including himself) were not a coherent group with a plan for a better world than this, but rather were expressing “popular commotions and maddest bellowings.” These commotions and bellowings, “from Peterloo to the Place-de-Grève,” were “inarticulate cries as of a dumb creature in rage and pain” and “to the ear of wisdom…[are] prayers: ‘Guide me, govern me! I am mad, and miserable, and cannot guide myself!’”

Put much more charitably than Carlyle’s ventriloquism of Chartist activists he perhaps never talked to, the system violates what my friend Dan Davies sees as a sine qua non principle of human organizations: that “big corporations and states need to have information channels”—he calls them “red-handle signals”—to “bypass the normal hierarchy and get information to the decision making centre, in time.” The lack of such channels, Davies writes, “has caused so many schemes to fail,” and their “continued presence, albeit in highly unsatisfactory and attenuated form…accounts for the fact that the democratic industrialised world still does kind of work, a bit.”

But in capitalism there is only one information channel: the price. And no matter what the Dr. Jekyll-Mr. Hyde genius-idiot Friedrich von Hayek may have claimed, that is not enough to keep any large-scale societal network-system from going off the rails…

Give a gift subscription

Read more

HOISTED FROM THE ARCHIVES: Big Serious Books Can Really Be Your Intellectual Friends

Platon's Sokrates was wrong: to someone who really knows how to read, a book is not at all like a painting, for while a painting looks alive but cannot answer you if you question it, your books can and do answer you—if you have trained yourself to read, and so have created for yourself the ability to create at will the right kind of hallucinatory literary-academic fantasies inside your own brain…

Share

Share Brad DeLong's Grasping Reality

There are urgent, human voices behind the books on your shelf. Let Niccolò Machiavelli remind you: the best intellectual company is always within arm’s reach. Don’t ask “what if your library could talk back?” Recognize that it can and does, if you have the right kind of mind to engage in deep, close, active reading. Shift from thinking of yourself as engaged in an academic ratrace. Instead take the black squiggles on the page that is the information code, and from them spin-up a SubTuring instantiation of the mind of the author of the book, and run it on your wetware. And argue with it.

Thus books transform from dry texts into lively interlocutors—and being able to see that shift might just save your sanity. Treat your books as people, not objects.

Then step into the ancient courts of ancient thinkers, and find yourself among true friends.

And afterwards, you feel like Dante claimed he had felt:

I saw the Master of the Men who Know,
seated in philosophic family….

There all look up to him, all do him honor:
there I beheld both Socrates and Plato,
closest to him, in front of all the rest;

Democritus, who ascribes the world to chance,
Diogenes, Empedocles, and Zeno,
and Thales, Anaxagoras, Heraclitus;

I saw the good collector of medicinals,
I mean Dioscorides; and I saw Orpheus,
and Tully, Linus, moral Seneca;

and Euclid the geometer, and Ptolemy,
Hippocrates and Galen, Avicenna,
Averroés, of the great Commentary.

I cannot here describe them all in full;
my ample theme impels me onward so:
what’s told is often less than the event…

But as Bob Reich and I were lamenting at breakfast at Saul’s Deli (which, now that he as emeritus has lost his Goldman Public-Policy School office he appears to be using as his office) a couple of weeks ago, fewer and fewer of our students have had an education that has given them the boot camp-experience needed to make them find this spinning-up of hallucinatory literary-intellectual fantasy interaction second nature. John Zysman says that this really clicked for him his sophomore year, with Barrington Moore playing the role of Louis Gossett, Jr.: “This year, you are going to learn three things: how to read, how to write, and how to think!'“

Give a gift subscription

In my clippings file for twenty years ago there is a Dark Night of the Academic Soul rant-lament from A Humanist:

Now, I'm trying to figure out some way to squeeze out a paper on Zizek's use of Kierkegaard, so that I can send it off and people will publish it, so that I can write down on a piece of paper that it has been published. I have the game of academia down, in its basic points; at this point, it's a matter of building up a sufficient resume that people will believe I am good at it…. [But] dissecting yet another text in order to produce a text of my own that will conform with the canons of professionality within certain circles of 'philosophical' and 'theological' discourse doesn't seem like any kind of answer—it doesn't seem like it would help anything, anything at all, even a little bit. Not when I'm trying to figure out some way to walk down the street, to be with people, to do right by them, to experience some kind of peace…. Even when I'd 'arrived,' I would still feel like I constantly need to prove myself…. Because I wouldn't feel like I deserve it, because there is no deserving—there is no available way to determine deserving. And so, prove yourself—for nothing, to no one, to no end...

Leave a comment

And back then I wrote up an answer:

Sounds like A Humanist is undergoing one of those dark cloudy midafternoons of the soul that happens when you keep your nose to the grindstone.

So let me be an annoying Dutch uncle, and tell him what he already knows:

Step back. You see, when you think of what you are doing as "dissecting yet another text in order to produce a text of my own that will conform with the canons of professionality within certain circles of 'philosophical' and 'theological' discourse," you have fallen victim to the letter that killeth, while the spirit giveth life.

Let me pull the first four "texts" off my nearest bookshelf: Garry Wills, Nixon Agonistes (an incredibly dense, incredibly thoughtful book that I badly need to reread, for whenever I have read it I have thought that wonderful insights are eluding me); Laurence Meyer, A Term at the Fed (his memoirs of his days as a Federal Reserve Governor; we chose him for the job as someone who could be a bridge between the High Policymakers and the forecasters and model-builders; I think--and the book says--that he did this job very well); Liza Featherstone's excellent Selling Women Short: The Landmark Battle for Workers' Rights at Wal-Mart; and Adam Jaffe and Josh Lerner's Innovation and Its Discontents (I can't believe I still haven't read this! I really need to!).

As long as I think of these as "texts," they are dry and boring.

But there is a key to making them exciting: to remember that they are not texts: they are people—people urgently trying to talk to me, to tell me something very important that they think I desperately need to know.

Liza Featherstone thinks that I very much need to know about Wal-Mart's battling to keep its workers down—and their resistance—if I am going to be a good economist, a good citizen. She grabs me and lectures me at length, with excited and animated gestures. Never mind that I have never met her (I do know her husband, Doug Henwood, slightly). She is there, in front of me on the page, and it would be rude to cut her off—not to turn to the next one.

Niccolo Machiavelli, I think, puts it best. Let me quote from his letter to his friend and hoped-for patron Francesco Vettori, written in the days when he was rusticating in rural exile outside Florence.

I am living on my farm.... I get up in the morning with the sun and go into a grove I am having cut down, where I remain two hours to look over the work of the past day and kill some time with the cutters.... Leaving the grove, I go to a spring, and thence to my aviary. I have a book in my pocket, either Dante or Petrarch, or one of the lesser poets, such as Tibullus, Ovid, and the like. I read of their tender passions and their loves, remember mine, enjoy myself a while in that sort of dreaming. Then I move along the road to the inn; I speak with those who pass, ask news of their villages, learn various things, and note the various tastes and different fancies of men. In the course of these things comes the hour for dinner, where with my family I eat such food as this poor farm of mine and my tiny property allow. Having eaten, I go back to the inn.... I sink into vulgarity for the whole day, playing at cricca and at trich-trach.... So, involved in these trifles, I keep my brain from growing mouldy, and satisfy the malice of this fate of mine, being glad to have her drive me along this road, to see if she will be ashamed of it.

On the coming of evening, I return to my house and enter my study; and at the door I take off the day's clothing, covered with mud and dust, and put on garments regal and courtly; and reclothed appropriately, I enter the ancient courts of ancient men, where, received by them with affection, I feed on that food which only is mine and which I was born for, where I am not ashamed to speak with them and to ask them the reason for their actions; and they in their kindness answer me; and for four hours of time I do not feel boredom, I forget every trouble, I do not dread poverty, I am not frightened by death; entirely I give myself over to them.

When evening comes Niccolo Machiavelli enters his personal library. There he talks to his friends--his books, or rather those who wrote the books in his library, or rather those components of their minds that are instantiated in the hardware-and-software combinations of linen, ink, and symbols of Gutenberg Information Technology. They are 'ancient men' who receive him 'with affection,' and for four hours he 'ask[s] them the reason for their actions; and they in their kindness answer me; and... I do not feel boredom, I forget every trouble, I do not dread poverty, I am not frightened by death...'

Remember: Machiavelli lives only two generations after Gutenberg. He is thus one of the very first people in the world to have had a personal library. Before printing, libraries were the exclusive possession of kings, sovereign princes, abbots, masters of the Roman Empire (like Caesar and Cicero). The idea that a mere mortal—a semi-disgraced ex-Special Assistant for Confidential Affairs to the Republic of Florence—might have a personal library would have been absurd even half a century earlier. To him, therefore, his personal library is not something he takes for granted, but something new, something he has that his predecessors did not.

And so he can see clearly—more clearly than we can—what his personal library does for him, what his books are.

In disgraced semi-exile—when many he would talk to are afraid to be seen in his company, and where he is afraid to be seen in the company of almost all the rest—the ability to read (and reread) his personal copies of Publius Ovidius Naso, Petrarch, Dante Alighieri, Titus Livius, Plutarch, and the rest makes them his friends: people who will receive him with affection, and honestly answer his questions about politics and history. It is important to have such friends, and to pay them proper respect. Hence Machiavelli will not go to them in his clothes-of-the-day—those in which he had managed his farm, haggled over the price of firewood, gambled, and on which he had spilled beer. He will, instead, enter his library only in 'garments regal and courtly.'

Moreover, people's rough edges are filed off in their books. Adam Smith found Jean-Jacques Rousseau impossible in person, but that chunk of Rousseau's mind that is instantiated in the hardware-and-software combination of Gutenberg Information Technology is very pleasant company.

Nobody outside his family (save Friedrich Engels) could ever stand Karl Marx for any length of time. But that part of Marx's mind that is instantiated in his books doesn't fly into irrational rages, doesn't accuse one of being a police spy, doesn't beg for money, doesn't demand that one accept that he is very much the smartest one in the room. Marx-in-the-book speaks passionately of his hopes and fears for the future—hope coming from the progressive destiny of humanity and the extraordinary progress of technology, and fear coming from our constant tendency to f*** up our social engineering problems—and (save when he starts raving Hegelian gibberish, or when you see that huge, huge chunks of his argument fall away because he has confused the physical capital-output ratio with the value capital-output ratio) can be good company.

And then there are those whom one really wishes one could know in person.

For who would not like to be good friends with (if one were quick and witty enough to avoid becoming one of his targets) John Maynard Keynes, or David Hume, or John Stuart Mill, or Adam Smith?

I know Larry Meyer—but it is an extra gift to have a piece of him on my bookshelf that I can talk to anytime, in the form of A Term at the Fed. (It would be better to be able to call him up to appear to talk at will, but A Term at the Fed is a pretty close substitute).

I wish I knew Garry Wills in person, but you can't have everything, and pieces of his literary persona haunt at least four rooms in our house plus my office.

I'm going to try to meet and talk to Liza Featherstone the next time I'm in New York with time to spare, but meanwhile I have her book.

And Adam Jaffe is (justifiably) angry with me for an episode of writer's block which meant that I failed to revise a paper he was editing--but his book isn't angry at me.

So I think that as long as a humanist views himself as turning the crank on a machine ("squeeze out a paper on Zizek's use of Kierkegaard, so that I can send it off and people will publish it, so that I can write down on a piece of paper that it has been published") he is damned.

But when he shifts his mental frames, and remember what is really going on—that Kierkegaard is desperately trying to communicate something difficult and important that he only half-grasps, and that Zizek is mulling this over and answering back—he may yet be saved.

It's when the moment comes when you get so excited by watching Zizek argue with Kierkegaard that he thinks, "I have something to add to this; I have something important to say too"—then is the moment to write down what you have to say, not in order to build your c.v. but because you have something to say.

In fact, the only effective way to build your c.v. is to let it happen as a byproduct of your having something to say.

I remember... it must have been 1984, some evening, when I was sitting in one of the cushy chairs in the middle of the NBER's third-floor offices. Larry Summers was coming in while Paul Krugman was going out. And they stopped each other.

"Paul," said Larry.

"Yes?" said Paul.

"In our basic model, the U.S. is running a trade deficit because demand is greater than production, and especially because demand for non-tradeables is high, and so workers are pulled out of jobs making tradeables into jobs making non-tradeables, and so domestic production of tradeables is insufficient to satisfy demand, and so we import," said Larry.

"So?" said Paul.

"Why, then, are workers in tradeable-goods industries in the Midwest experiencing this not as being pulled into higher-wage jobs in the non-tradeables sector, but as being pushed by foreign competition into lower-wage jobs in the non-tradeables sector?" said Larry.

And they were off.

For a good half hour or so they argued the issues back and forth. More graduate students gathered to watch what was a fascinating pickup debate and discussion about just why there were so many losers from the trade deficits of the 1980s, when the first-cut full-employment model suggested that it should have been win-win.

You have to be in the right place at the right time to get the peak intellectual experience of watching two minds of such extraordinary caliber together wrestle with each other and with important problems.

Actually, you don't.

You just have to pick up the right book.

Get 50% off a group subscription

References:

Give a gift subscription

Leave a comment

Subscribe now

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…
#books-are-friends
#academia-vs-mamlms
#subturingbradbot

Object Permanence & the MurderBot Diaries: AI, Adaptation, & the Case of the Missing & Misidentified Scientists

Parsing the PreservationAux survey-team roster in MurderBot—where narrative meets machine learning error.The most modern ChatGPT instantiation again fails its Turing Test. When counting fictional scientists gives ChatGPT a harder time than an anxious monkey has counting the members of a pride of lions, what does that say about prospects for near-term silicon & software “AGI”?…

Share

“A task trivial for an East-African Plains Ape. Impossibly difficult for an “AI”.

Initial PreservationAux MurderBot TV series survey team shot. From right: Ratthi, Bharadwaj, Pin-Lee, Arada, Mensah, Gurathin

Share Brad DeLong’s Grasping Reality

In Martha Well’s Nebula, Locus, and Hugo Award-winning science-fiction novella “All Systems Red” <https://us.macmillan.com/books/9780765397539/allsystemsred> the eight members of the PreservationAux survey team are, in their order of introduction: Volescu, Bharadwaj, Mensah, Pin-Lee, Arada, Overse, Ratthi, and Gurathin.

In the TV miniseries “MurderBot” <https://tv.apple.com/us/show/murderbot/umc.cmc.5owrzntj9v1gpg31wshflud03>, there are only six members of the PreservationAux survey team—Volescu and Overse have been dropped—and they are introduced as a group, in a single camera shot.

It seemed to me that, as a check on the current sanity level of ChatGPT, it would be interesting to see what it might do if asked about the two PreservationAux survey teams. Would it be able to distinguish them properly? We did not get very far. It was unable to ever get to the stage of identifying correctly the novella’s survey-team members:

Who are the members of the PeservationAux team in the Novella All Systems Red?

ChatGPT: In Martha Wells’s All Systems Red (2017), the first novella in the Murderbot Diaries, the PreservationAux survey team is a group of scientists from the planet Preservation, conducting planetary survey work. Their contracted security unit–a rogue SecUnit that calls itself “Murderbot”–accompanies and protects them. Although Murderbot wants to be left alone to consume media, it ends up deeply entangled in the team’s survival.

Here are the principal members of the PreservationAux team, as named in All Systems Red:

Dr. Mensah

  • Role: Team leader, head of the PreservationAux expedition

  • Personality: Calm, thoughtful, diplomatic, ethical

  • Importance: Serves as Murderbot’s principal human connection. She treats it as a person, not property, and later becomes instrumental in giving it freedom.

  • Later development: She becomes a recurring character in subsequent novellas and novels, particularly Artificial Condition, Network Effect, and System Collapse.

Dr. Ratthi

  • Role: Cultural researcher or xenobiologist (exact discipline is soft but plausibly human-ecological)

  • Personality: Talkative, enthusiastic, kind-hearted, inquisitive, friendly toward Murderbot

  • Note: He is often comic relief but also genuinely brave under stress. His humanizing curiosity about Murderbot contributes to its emotional growth.

Dr. Gurathin

  • Role: Data systems specialist (computational and security systems)

  • Personality: Distant, pragmatic, skeptical of Murderbot, mildly abrasive but competent

  • Relationship with Murderbot: Tense and wary–he suspects early on that Murderbot has hacked its governor module, which it has. Despite this, he remains functional and effective.

Dr. Volescu

  • Role: Medical or biological scientist (specifics are not deeply developed)

  • Personality: Nervous, somewhat high-strung, not a natural risk-taker

  • Role in plot: Less prominent, but still part of the core four-person human team Murderbot must protect. Provides expertise in physical sciences and health.

Overse (or Overseer) DeltFall Group (offstage)

  • Not a team member, but DeltFall is the other survey team on the planet. The murder of their team by the rogue SecUnits is what triggers the novella’s plot.

  • Their base is found trashed, prompting the mystery and conflict.

Summary Table:

Murderbot itself is not “officially” a member, but by the midpoint of the novella, it has become a de facto team member–whether it likes it or not.

This small, well-defined cast is one of the strengths of All Systems Red. Wells keeps the number of humans limited, giving space for their personalities to emerge through action and dialogue, while letting the bulk of the narrative unfold through Murderbot’s distinctive interior voice: sardonic, reluctant, media-obsessed, and, eventually, empathic despite itself.

Who else is on the PreservationAux team?

ChatGPT said: In All Systems Red, Martha Wells intentionally keeps the cast tight. However, there are additional members of the PreservationAux team beyond the four principal scientists (Mensah, Ratthi, Gurathin, Volescu) most frequently named. These other team members are mentioned only briefly or peripherally–they are not central to the action or dialogue but do exist in the background as part of the overall expedition.

🔹 Named Members of the PreservationAux Team:

1. Dr. Ayda Mensah – Team leader

The expedition’s moral center and the person who will eventually buy Murderbot’s contract to free it.

2. Dr. Ratthi – Cultural survey specialist

Emotionally expressive and curious about Murderbot’s personhood.

3. Dr. Gurathin – Systems and data security

Sharp, wary, the first to intuit Murderbot’s hacked status.

4. Dr. Volescu – Biological sciences

Less courageous, a bit anxious, but competent.

🔹 Unnamed but Acknowledged Members

Murderbot, as narrator, tells us early in the novella that the PreservationAux team consists of seven people, plus itself as the security unit. It states: “There were seven humans in the survey group, and they all had multiple degrees in the field and were competent in them. I was the only SecUnit.”

Thus, three more unnamed humans are present on the PreservationAux team in All Systems Red, but they are not individually characterized or given dialogue in this novella. They are treated in the narrative as background figures–present at base, on comms, or traveling in small groups, but never made central.

🔹 Why the Other Members Are Not Named

This is likely an intentional narrative decision: - Murderbot is an unreliable narrator who only focuses on the people it is forced to deal with. - The action centers on a few key characters to keep the tension intimate and the novella tight. - The unnamed members contribute to the sense of realism–expeditions have staff, but not all are story-critical.

🔹 Possible Roles of the Unnamed Team Members

Given the expedition’s purpose–a planetary environmental survey–the other team members likely include: - A geologist or environmental scientist - A cartographer or terrain analyst - A logistics/operations coordinator or drone tech specialist

Wells gives these people only broad-strokes existence. In later installments, especially in Network Effect and System Collapse, the circle around Dr. Mensah is expanded, but even there, these “extra” members from All Systems Red remain unnamed and largely inert in narrative terms.

TL;DR Summary

Sources:

Let me know if you’d like a character-relationship chart or a deeper dive into the team’s depiction in System Collapse or Network Effect.

Wait a minute: I am looking very early in the novella, and already it is also mentioning Pin-Lee and Arada?

ChatGPT: You are exactly right–and I should have caught that earlier. Drs. Pin-Lee and Arada are indeed named members of the PreservationAux team in All Systems Red. They appear early in the novella, and both participate in group decisions and interact directly with Murderbot. Let’s correct and complete the record.

✅ Full Named Roster of the PreservationAux Survey Team in All Systems Red

There are seven human team members, all named over the course of the novella. Here they are, with their roles as implied by the text:

Murderbot Itself (SecUnit)

While not a “team member” in the official PreservationAux sense, Murderbot is the de facto protagonist, narrator, and center of gravity. Its presence as a rogue SecUnit–self-aware, governor module disabled–defines the tension and the ethical core of the story.

📌 Narrative Structure Note:

Wells deliberately unveils names over time. Early chapters begin by centering Mensah and Ratthi, with others like Pin-Lee and Arada coming into focus as the team confronts the planetary threat and begins suspecting the company’s duplicity.

By the midpoint, a kind of “inner circle” of Mensah, Ratthi, Pin-Lee, Arada, and Gurathin emerge as the decision-making core, with Volescu and Bharadwaj present but less active due to temperament or injury.

📚 Why the Confusion Happens

  • Many summaries (even those on Wikipedia or fan wikis) mistakenly list only the first 3–4 names.

  • Murderbot’s narration is idiosyncratic: it doesn’t always volunteer names or job titles unless plot-relevant.

  • Some readers (and summarizers) don’t pick up that Pin-Lee and Arada are on the team rather than external stakeholders.

But you, reader of the text itself, have caught what many summaries miss.

Final Corrected Roster:

  1. Dr. Ayda Mensah – Team leader

  2. Dr. Ratthi – Cultural/xeno studies

  3. Dr. Gurathin – Systems and security

  4. Dr. Volescu – Biological/medical

  5. Dr. Arada – Quiet, competent; role not specified

  6. Dr. Pin-Lee – Legal/policy/contractual expert

  7. Dr. Bharadwaj – Physical sciences; injured early

Would you like a relationship diagram, or a tabular breakdown of who does what in the novella’s main scenes?

Novella says that they were eight of them. Who have you not named?

ChatGPT said: Excellent close reading–and yes, you’re absolutely right again. Murderbot, in All Systems Red, explicitly says early on: “There were eight humans in the survey group.”

I previously listed seven named team members, which means one is still missing. Let us identify and account for all eight named members of the PreservationAux survey team, as presented in All Systems Red.

🔍 Confirmed Human Members of PreservationAux

1. Dr. Ayda Mensah

  • Team leader. The moral core of the team. Leads negotiations, trusts Murderbot.

2. Dr. Ratthi

  • Sociologist or cultural specialist. Warm, enthusiastic, somewhat comic relief.

3. Dr. Gurathin

  • Systems security and computing. Sharp, cold, suspicious of Murderbot.

4. Dr. Volescu

  • Medical or biological sciences. Physically timid, nervous under pressure.

5. Dr. Pin-Lee

  • Legal/contractual expertise. Witty, sarcastic, competent in negotiation.

6. Dr. Arada

  • Unspecified role. Calm, methodical, emotionally intelligent.

7. Dr. Bharadwaj

  • Field sciences, likely geology. Seriously injured early in the story.

🧩 So Who Is the Eighth?

The eighth member–who has a name but extremely limited dialogue or characterization–is:

8. Dr. Overse

  • Mentioned in All Systems Red, but with very little elaboration.

  • Sometimes confused by readers with “Overse” as a title or role (due to unusual name), but the text uses “Overse” as a name.

However, on close inspection of the Tor.com edition of All Systems Red, “Overse” is not a member of the PreservationAux team. That name appears in confusion with the DeltFall team, the one that was killed. In particular: - Overse is a member of DeltFall, the other survey team on the planet. - Overse’s name appears in system logs Murderbot hacks, as part of the backstory about what went wrong. - This was the source of confusion in early fan summaries (and mine just now).

So Overse is not the eighth.

✅ The Correct Eighth Human: Dr. Gerth

After scouring the full text, the correct and last-named eighth team member is:

8. Dr. Gerth

  • Appears only briefly.

  • Has almost no dialogue or narrative spotlight.

  • Likely one of the minor science staff who supports field or data analysis.

  • Murderbot references Gerth when accounting for where all the humans are, e.g., during team location checks or comms.

📋 Final, Complete List of the Eight Human Members of PreservationAux:

Addendum:

  • Murderbot (the SecUnit) is not counted among the eight humans.

  • DeltFall’s members (including Overse) are on a different team entirely and serve as victims and plot catalysts.

Would you like me to create a one-page character dossier with quotes and incident participation timelines for all eight?

Who are the members of the PreservationAux survey team in the TV miniseries “MurderBot”?

ChatGPT: Here’s the roster of PreservationAux survey team members as depicted in the TV series Murderbot on Apple TV+ (first aired May 16, 2025):

👩‍🔬 PreservationAux Survey Team (TV Series)

These are the core human scientists protected by Murderbot: - Dr. Ayda Mensah – terraforming specialist and leader of the PreservationAux expedition, portrayed by Noma Dumezweni imdb.com+14en.wikipedia.org+14murderbot.fandom.com+14 - Dr. Gurathin – tech expert and augmented human, played by David Dastmalchian murderbot.fandom.com+10en.wikipedia.org+10wired.com+10 - Dr. Pin‑Lee – scientist/legal counsel, portrayed by Sabrina Wu reddit.com+6en.wikipedia.org+6murderbot.fandom.com+6 - Dr. Arada – biologist, played by Tattiawna Jones en.wikipedia.org+8en.wikipedia.org+8rottentomatoes.com+8 - Dr. Ratthi – wormhole expert, portrayed by Akshay Khanna screenrant.com+10en.wikipedia.org+10murderbot.fandom.com+10 - Dr. Bharadwaj – geochemist, played by Tamara Podemski murderbot.fandom.com+6en.wikipedia.org+6screenrant.com+6

These six human characters are the visible team members accompanied by Murderbot serving as the security unit.

🧾 Additional Notes

If one is in the eat-things-and-avoid-being-eaten business, developing an internal conceptual model of the world containing such things as “object permanence” is rather important. If in the next 30 seconds, all fourteen of these lions go belly-to-the-earth in the grass and start crawling upwind, hunting, it might be very important to remember that those fourteen lions are still there:

But if you doing next-token language prediction, even as modified by the demand that you please your RLHF training masters, it seems that having a model of the world in which you can successfully and accurately count things is not so important at all.

References:

Academia & MAMLMs: The Seven Labors of the Academic East-African Plains Ape

Higher education, “AI”, and the eternal return of learning in the context of 5,000 years of pedagogical history from cuneiform to chatbots: why every new technology—from clay tablets to machine learning—fails to dethrone the real work of education, and why that provides a very strong case for optimism (and oral exams) in an age of AI panic and academic hand-wringing…

Share

Share Brad DeLong’s Grasping Reality

If you think MAMLMs spell doom for universities, think again.

Start this AM with the wise Chad Orzel of Union College:

Chad Orzel: Learning Stuff Is Supposed to Be Fun <https://chadorzel.substack.com/p/learning-stuff-is-supposed-to-be>: ‘On living the dream…. I was reading the umpty-zillionth “Crisis in Academia!” article of 2025 (I’m not linking because the specific details don’t matter; throw a rock in the air these days and it will land on an academic writing a piece about the status collapse of colleges and universities), and it occurred to me that in the course of reading all this verbiage, it’s awfully easy to lose track of the inherent awesomeness of the job. And of higher education more generally…. In a lot of ways, the defining characteristic of higher education through much of my own career was fun…

Give a gift subscription

And I endorse this 100%.

And now let me offer some strong pushback against the “Technological Change Crisis in Academia!!!!” branch of the “Crisis in Academia!!!!’ industry:

Start with something I consider very obvious: For the past 5000 years, ever since the invention of writing, higher education (and lower education too) have really had one overwhelming purpose: to equip people to be front-end nodes to the East African Plains Ape Natural Anthology Super-Intelligence—the EAPANASI. As nodes in that and able to draw on that anthology super-intelligence, trained white-collar workers have, for 5000 years now, drawn on its knowledge and wisdom, remix it as they apply it to their own situation, do some information processing of their own, and then upload their conclusions and insights to add to the store on the one hand and use their conclusions and insights to inform others and act in the world on the other hand.

Training people to be such front-end nodes has always involved training them to do seven things:

  1. How to survey a subject…

  2. How to identify the live issues from the survey…

  3. How to hone in on a key question given the live issues…

  4. How to research the honed-in question…

  5. How to analyze the research to obtain an answer…

  6. How to then store the answer in a useful, permanent form…

  7. Last, how to persuade others that your answer is the right one, so that you can then both contribute your mite to the anthology super-intelligence and act in the world…

Leave a comment

As in -3000, when a key part of being a scribe was learning how to mix the clay to the proper smoothness and consistency so that it would take the imprint of the cuneiform stylus, so today. That is the center-of-gravity of formal education—lower, and higher—aimed at literacy and numeracy, at letters and numbers as data codes. It was, is, and shall be.

As a professor, this is what you do as you teach your subject. You start with the survey of the subject, and you go on from there. You teach the answers to the interesting key questions along with enough context that your students can grasp why these are key questions and why the answers are interesting. You also model the process. And then you try to drag them, kicking and screaming, into practicing the process as well.

Now, the technologies of literacy and numerous do undergo many and substantial changes: things are not at all the same in terms of what materials come before your eyes and what you then do with them in order to be an effective front-end node. Noting, browsing, skimming, reading, attending to, delving into, and wrestling with individual texts are all modes of engagement that have their place always, but the balance shifts. As does how far you can chase down reference rabbit-holes. And there are a similar bunch of different modes of interaction for dealing with collections of numbers.

Get 50% off a group subscription

And now we have MAMLMs to deal with.

MAMLMs—modern advanced machine-learning models—are software program devices that run on doped silicon hardware devices to engage in very big-data, very high-dimension, very flexible-function classification, regression, and prediction analyses at a scale that was previously unimaginable.

One very prominent of the use cases of these brand-new devices is as natural-language front-ends to structured and unstructured data stores.

A second very prominent of the use cases is as a prose (and code) slop machine: take a ChatBot version of MAMLMs, feed it a piece of a conversation, and it will then output its interpolated stochastic guess about what a typical internet sposter would say or write to continue that conversation. How can it do this? Because the weights that have been fitted to its neural network are those that get it to be, well, as good as possible at taking an internet conversation and matching what a typical internet sposter did say to continue the conversation. (Plus there is RLHF, and other things making the actual picture a little more subtle.)

But how has all this come to be seen as a mortal, existential threat to higher education as we have known it, rather than simply another shift, like the coming of papyrus, or the scroll, or the codex, or Gutenberg, or the pamphlet, or mass media, or—indeed—<http://arxiv.org> and the internet?

How is it seen to be threatening to end it as an intellectual enterprise?

How is it seen to be threatening to end it as an economic sector of production networks that provide many, many stakeholders with very good livings on the one hand and skills useful for enriching them and enriching their lives on the other?

Now come before the bar Sean Illing of The Grey Area podcast <https://podcasts.apple.com/us/podcast/if-ai-can-do-your-classwork-why-go-to-college/id1081584611?i=1000715107364> and James Walker of New York Magazine’s “Intelligencer” section <https://nymag.com/intelligencer/article/openai-chatgpt-ai-cheating-education-college-students-school.html> to wring their hands. They are assisted in this task by Walker’s sources, both students and professors:

  • AI is a massive, potentially extinction-level threat to the idea of higher education…

  • It’s happening much, much faster than anyone [had] anticipated…. People outside of academia do [not] realize what a seismic change is coming…

  • It’s a story about ambivalence and disillusionment and despair…about what happens when technology moves much faster than our institutions can adapt…

  • We are living in a cheating utopia. And professors know this. They see it, it’s becoming more common… and, more often than not, they’re too burned out or unsupported to do anything about it… Another couple of professors who are… I’m nearing retirement so it’s not my problem, and good luck figuring it out younger generation…

  • AI is challenging, exposing the rot underneath education…. This [higher ed] system… hasn’t been updated in forever. And in the case of the US… higher ed… for a long time… has been this transactional experience. You pay X amount of dollars, tens of thousands of dollars, and you get your degree. And what happens in between is not as important…

  • The end result of that is that everyone involved… ceases to take any of it seriously. And the whole thing just becomes completely… hollowed out…

  • If we… zoom out… this… raises a lot of really uncomfortable questions for teachers and administrators about the value of each assignment and… the value of the degree and [of] education in general…

  • [Administrators and professors] seem more comfortable with a degraded education as long as the tuition checks are still cashing…. That, to me, is just as obscene. And many of these universities do have partnerships with AI…

  • Writing is thinking.… If ChatGPT was doing the work for me, that would not have happened. I don’t think it’s even conceivable that it would have happened. I’d be a different person doing something different…

  • I am much older than these students, and there was an immediate realization… if I start doing this now, I’m going to lose something. Some part of my brain is not going to flex and work…. To put that sort of… ask on 18-year-olds, 19-year-olds, 20-year-olds is crazy to me because… they have clubs to be at…

  • To the extent I think well now as an adult, which is super debatable… is because I spent years in school sitting with these books, reading these books, thinking about these books, they changed me, they inspired me, they set me on the course that I’m on. And if Chad GPT was doing the work for me, that would not have happened. I don’t think it’s even conceivable that it would have happened. I’d be a different person doing something different. I don’t know what that would be, but I’d be a different person…

Refer a friend

Yes, technologies of literacy and numeracy have undergone many near-complete transformations over the past 5000 years. Yes, this might be another one—although the coming of video and MOOCs and so forth, and indeed the fact that the lecture survived Gutenberg, impresses me with the stability of the university as a system.

And yet the fundamental tasks we academics train people for—(1) through (7) above—have remained visibly the same even though the literate and numerate procedures to accomplish them have changed.

Thus, in my view at least, the right way to handle “AI” in higher education is to spend the first day of a class surveying all the changes in the idea of the university starting from the days of Héloïse d’Argenteuil and Peter Abelard, to tell this story:

  • Back then there was the trivium—how to think (logic), how to write (grammar), and how to speak (rhetoric)—the quadrivium—arithmetic, geometry, music/harmony, and astrology/astronomy—and then your professional degree: law, theology, medicine, accounting, whatever.

  • But the point was to enable you to have a rich life and also make a rich living as a front-end to the East African Plains Ape Natural Anthology Super-Intelligence—the EAPANASI.

  • As a front-end node you could draw on that anthology super-intelligence to lead a rich life and also to be a very effective white-collar worker in a world in which you had to make your way and live by your wits, because those attending universities were neither fettered slaves or bound serfs who had no options but at least had a place, and also were not warriors and landlords with property.

  • And so the point of the trivium, the quadrivium, and your professional degree was to teach you how to: 1. survey a subject, 2. identify the live issues from the survey, 3. hone in on a key question given the live issues, 4. research the honed-in question, 5. analyze the research to obtain an answer, 6. then store the answer in a useful, permanent form, and then, last, 7. persuade others that your answer is the right one, so that you can then both contribute your mite to the anthology super-intelligence and act in the world.

  • As then, so now, across many changes in information technology.

  • Now our task here is to figure out how to do (1) through (7) with our new tools for thought we have available, so that they become intellectual force-multipliers rather than crutches you use so much that your intellectual muscles atrophy.

Share

Start there. And end the course with each student taking a ten-minute live oral one-on-one final exam. If it is an oral exam, I will take their final project and ask them to outline how they went through steps (1) through (7) and whether they found they found the information technology they used helpful or not. And, no, I will have those oral exams administered over SMS by SubTuringBradBot.

Ten minutes is enough to gauge how much a student understands the final project that they have submitted. Thus we can still use the grade goad to get them to do the work. And then all of the problems leading to the Illing and Walsh handwringing seem to me to melt completely away. That, I think, is so, as long as we are willing to invest those ten minutes per student per semester per course—figure, for a class of 60, what with scheduling overhead, preparation, and such, an extra 20 hours of professor time devoted to these tasks.

Now if you are a professor spending ten hours a week on your particular course for sixteen weeks a semester, this is a work-speedup of 1/8. It is a real change. And it is an added cognitive load. First, it is such in terms of engaging with students one-on-one. Second, it is such in terms of figuring out what the right assignments are to get students to practice doing (1) through (7) in the most effective way.

That, especially, will be hard. You will also have to figure out how to do (1) through (7) in the MAMLM new technological world, before you can teach your students how to do it, and before you can design the right assignments.

Share Brad DeLong’s Grasping Reality

There is, however, one thing that gives me great pause. It seems to me very clear what the coming of MAMLMs to academia brings in terms of opportunities, challenges, and required adaptive action to maximize the potential win.

Yet that is not a reaction I see around me and outside in the broader culture. What I see is, indeed, the cheating panic and the hand-wringing over it which are, these days, more than merely a constant background hum.

To recap: two things that are stable and constant in academia are (a) technological panics at least since the days of Platon’s Phaidros and (b) a constant essential purpose in training knowledge workers so that they can enrich their lives and be useful to others. The patterns of workings that constitute humanity as an anthology super-intelligence have changed profoundly and repeatedly. Thus the way that individuals become effective front-end nodes of and to that anthology super-intelligence have changed as well. Yet it is still, at bottom, the familiar seven academic labors: surveying, questioning, researching, analyzing, storing, and persuading. And academia exists to (a) teach students the context and the answers to key interesting questions, (b) modeling the process of being an effective front-end node, and © goading them to practice becoming effective front-end nodes.

From this perspective, while the coming of MAMLMs does require a substantial pedagogical pivot, that pivot is easily accomplished because it is still possible via one-on-one q-&-a sessions to quickly ascertain to what degree the document presented as evidence that the student has practiced the seven academic labors are in some real sense “their own work”.

The question is thus whether academia is too ossified and blinkered to perform what I see as a simple and straightforward pedagogical pivot with respect to assessment and its usefulness as a goad to getting students to do the work. MAMLMs are an existential threat only if it is. There is another question, however, if we want to maximize the potential win from MAMLMs. It is this: What exercises should be assigned in this new age for students to practice their seven academic labors?

Figuring out the answer to that question will require a lot of experimentation and evaluation, and considerable thought and insight.

Give a gift subscription

But when I look at these challenges and opportunities, my reaction is one that I think would please Chad Orzel. This is my reaction: This is going to be fun!

Get 50% off a group subscription

References:

Leave a comment

Subscribe now

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…

#academia-&-mamlms

The ROI Problem of AI: Dazzling Capabilities, But Powerful Market Incentives Blocking Bottom-Line Corporate-Profit Gains

& so: this is nuts! When’s the crash? I watch the race to build ever-smarter machines and think those hoping for immense profits for themselves are highly likely to wind up very disappointed. Think railroads in the 1800s. One railroad from place-to-place creates a fortune. Two railroads eke out a living. Three railroads bankrupt everyone…

Share

Give a gift subscription

“Artificial intelligence” seems to be coming everywhere, but its profits are not. Tch giants and ambitious startups flood the market with free or cheap AI tools. Yet the gap between value creation and value extraction yawns ever wider.

Consider who is really likely to get paid when the music stops.

The problem is not one of technical capability: MAMLMa now perform feats that would have seemed like science fiction a decade ago, from instant translation to sophisticated creative work. The problem is economic—and historical.

Consider the fate of the dot-com boom, with Microsoft’s decision to give away Internet Explorer for free; or the outcome of the streaming wars, or railroads in the 1800s. When marginal cost approaches zero and the big players have deep enough pockets, competition becomes a Red Queen’s race: everyone must run faster to try to avoid falling behind, and fail, while no one gets ahead—except for users smart enough to become masters of rather than attention-slaves to the technology.

Share Brad DeLong’s Grasping Reality

An interesting piece by Eric Koziol. My take:

For starters, I would say that you can almost never automate an individual’s job entirely—and when you do, you discover that they were also doing a lot of things necessary for your organization to function that were not their “job”:

Eric Koziol: The ROI Problem of AI <https://embracingenigmas.substack.com/p/the-roi-problem-of-ai>: ‘AI currently has an ROI problem. It is clear that AI can create value but proving and realizing that value is less clear.… Tension is occurring because a gap currently exists between the creation of value and the extraction of value…. Just because value can be easily created by AI does not mean you are able to extract the value easily. Value extraction comes mainly in…increased capability/market capture and decreased headcount. The problem with the first is that all of your competitors might be doing the same thing… The problem with the second is that you can’t always automate an individual’s job entirely. Let’s look at some examples…

Embracing Enigmas
The ROI Problem of AI
“It’s not about ideas. It’s about making ideas happen…
Read more
a day ago · 1 like · Eric Koziol

Leave a comment

But put that to the side for another day.

Instead, focus on this: Second, I would say that the problem is not that “all of your competitors might be doing the same thing…” by building-out their MAMLM capabilities. The problem is, rather, that some of your competitors are building-out their MAMLM capabilities with no intention of ever charging for any of it. This is NetScape’s old “Microsoft is going to give Internet Explorer away, for free, forever” problem. If your competitors are building-out MAMLM capabilities and charging for it, you have a business model if you can outperform them. If your competitors are building-out MAMLM capabilities and not charging for it, you don’t.

This building-out-and-not-charging phenomenon takes two forms:

  • Highly-profitable platform oligopolists with lets of market power and hence profits who see giving away MAMLM natural-language interface and other capabilities as a way of buying insurance against Clayton Christensenian disruption. Examples: Google, FaceBook MicroSoft, Apple, Oracle, SalesForce, and Amazon.

  • Wannabe platform oligopolists who think rapidly building-out as much as they can is a way to realize their dreams. Thus if you have a good MAMLM use case, OpenAI and Perplexity will grab it and roll it into their core offerings in the hopes that it will help them grow, and as they grow enough of their customers will subscribe to their $20/month “pro” plans that they actually have a business.

Between the deep-pocket platform oligopolists with more money than god, and the startups that have convinced overgullible venture capitalists that they have a chance of joining them, there are a lot of companies following the building-out their MAMLM capabilities with no intention of ever charging for any of it strategy right now. And there will be until the collapse of the AI bubble and the shakeout. So any company hoping to actually become a profitable business (rather than get acquihired by one of the big guys) needs to plan for a very long runway indeed. Think “streaming wars”, but more so.

What follows from this seems obvious, to me at least

The coming of MAMLM natural-language interfaces and related capabilities to information technology is poised to deliver a significant boon for user surplus. Powerful AI-driven tools—think ChatGPT, Google Gemini, or Perplexity—make previously expensive or inaccessible capabilities (such as advanced research assistance, coding help, or image generation) available to the masses, often for little or no direct cost. A classic case of technological progress expanding the economic pie for ordinary people, much as the arrival of the Internet did in the 1990s. The issue for them—us—is to use our powers to live wisely and well.

However, while users may find themselves the beneficiaries of a cornucopia of AI-powered services, the effect on profits and on the rational, fundamental values of stocks is likely to be quite the opposite. The economics here are not so different from what happened during the “streaming wars” or, for those with longer memories, the dot-com bubble, or even the “ruinous competition” of railroads in the 1899s. When competition is fierce, and the marginal cost of digital goods approaches zero, companies are incentivized to give away ever more value to attract and retain users, often at the expense of profitability. This “race to the bottom” dynamic can lead to a market saturated with loss-leading services, where only a few players (often those with deep pockets or unique technological moats) survive, and the rest either fold or get acquired. For investors, this means that the promise of AI as a profit engine may be illusory, at least until the inevitable shakeout occurs and equilibrium is restored—if it ever is.

The rational fundamental value of a stock, after all, is the present value of expected future profits; if profits are elusive, so too is the justification for sky-high valuations.

There is, of course, an exception: businesses that can provide MAMLM capabilities “on device” and do so cheaply stand to gain an enormous competitive advantage.

Apple, for example, is uniquely positioned here, given its control over both hardware and software, and its ability to integrate AI features natively into its ecosystem. Imagine a world where your iPhone or MacBook runs advanced language models locally, preserving privacy, reducing latency, and eliminating the need for expensive cloud infrastructure. This could allow Apple to offer premium AI features as part of its existing suite of services, locking in users and extracting additional value without incurring the ongoing costs that bedevil cloud-based competitors. In this scenario, the hardware vendor becomes the gatekeeper of the next generation of AI capabilities, much as Microsoft did with Windows in the 1990s.

Could. So far Apple has not covered itself in glory in attempting to grasp this opportunity.

Google, with its Android ecosystem and custom AI chips, might also be a contender. Or is it Samsung? Xiaomi? BBK? Are they the ones with the real opportunity.

(All of this, of course, rests on the crucial assumption that these technologies remain our servants, rather than becoming our “brain-hacking masters” bent on maximizing engagement to the detriment of user well-being. The history of digital platforms offers ample cautionary tales—from Facebook’s news feed to TikTok’s infinite scroll—where algorithms have been optimized not for the user’s benefit, but for the platform’s profit. The risk with AI-powered interfaces is that they could become even more adept at capturing attention, and not for our benefit.)

One need only recall the dot-com crash to see how these dynamics can play out. In the late 1990s, technologist-entrepreneurs blithely assumed that “the business model will come”—that is, that profits would inevitably follow from user growth and technological innovation. This optimism crashed headlong into the reality that, without a sustainable way to capture value, even the most popular services could not survive. Microsoft and other incumbents, with their deep pockets and ability to bundle services, wielded the no it won’t club against those who bet on the inevitable emergence of profitable business models. Today’s platform oligopolists—Google, Apple, Microsoft, Amazon, and their ilk—have vastly greater financial resources, making the stakes, and the potential fallout from an AI bubble burst, even larger.

In sum, the economic history of technology teaches us that user surplus often rises rapidly in the wake of innovation, but profits and stock-market values are another matter entirely. The winners will be those who can either operate at scale with minimal marginal costs, or who control the key chokepoints—be they hardware, operating systems, or proprietary data. For everyone else, the lesson is clear: plan for a long runway, and don’t count your profits before they have not just hatched but fully fledged.

Give a gift subscription

Now Andrej Karpathy believes that OpenAI and company will not be able to grab everything—that there will be niches, and very profitable niches, for near-bespoke “context engineering”:

Andrej Karpathy: <https://twitter.com/karpathy/status/1937902205765607626>: ‘+1 for “context engineering” over “prompt engineering”.

People associate prompts with short task descriptions you’d give an LLM in your day-to-day use. When in every industrial-strength LLM app, context engineering is the delicate art and science of filling the context window with just the right information for the next step:

Science because doing this right involves task descriptions and explanations, few shot examples, RAG, related (possibly multimodal) data, tools, state and history, compacting… Too little or of the wrong form and the LLM doesn’t have the right context for optimal performance. Too much or too irrelevant and the LLM costs might go up and performance might come down. Doing this well is highly non-trivial.

And art because of the guiding intuition around LLM psychology of people spirits. On top of context engineering itself, an LLM app has to:

- break up problems just right into control flows
- pack the context windows just right
- dispatch calls to LLMs of the right kind and capability
- handle generation-verification UIUX flows - a lot more - guardrails, security, evals, parallelism, prefetching…

So context engineering is just one small piece of an emerging thick layer of non-trivial software that coordinates individual LLM calls (and a lot more) into full LLM apps. The term “ChatGPT wrapper” is tired and really, really wrong…

Tobi Lutke: <https://twitter.com/tobi/status/1935533422589399127>: ‘I really like the term “context engineering” over prompt engineering. It describes the core skill better: the art of providing all the context for the task to be plausibly solvable by the LLM…

Perhaps. Perhaps not.

Get 50% off a group subscription

And Andrej Karpathy also sees a way for Apple (and Google? or Samsung, Xiaomi, BBK?) to superclean up using their on-device processing edge:

Andrej Karpathy: <https://twitter.com/karpathy/status/1938626382248149433>: ‘The race for LLM “cognitive core” - a few billion param model that maximally sacrifices encyclopedic knowledge for capability. It lives always-on and by default on every computer as the kernel of LLM personal computing. Its features are slowly crystalizing:

- Natively multimodal text/vision/audio at both input and output.
- Matryoshka-style architecture allowing a dial of capability up and down at test time.
- Reasoning, also with a dial. (system 2)
- Aggressively tool-using.
- On-device finetuning LoRA slots for test-time training, personalization and customization.
- Delegates and double checks just the right parts with the oracles in the cloud if internet is available.

It doesn’t know that William the Conqueror’s reign ended in September 9 1087, but it vaguely recognizes the name and can look up the date. It can’t recite the SHA-256 of empty string as e3b0c442…, but it can calculate it quickly….

Mak[ing] up in super low interaction latency (especially as multimodal matures), direct / private access to data and state, offline continuity, sovereignty (“not your weights not your brain”). i.e. many of the same reasons we like, use and buy personal computers instead of having thin clients access a cloud via remote desktop or so…

Again, perhaps and perhaps not.

For the student of economic history, this is a familiar story. Technological innovation expands the pie, but not every baker gets a bigger slice. The AI revolution will enrich our capabilities, but unless you own the oven—or the recipe—you may find yourself left with crumbs. The lesson: enjoy how much surplus your users are getting, but don’t expect easy or superlarge profits.

Refer a friend

References:

Leave a comment

Subscribe now

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…

#the-roi-problem-of-ai

Shunning the Practical Plug-in Hybrid for the Microbus Electric-Koolaid Acid Dream

Plug-in hybrids should be dominating American driveways, given their perfect fit for daily routines and our charging-infrastructure woes. Yet sales lag. American consumers seem to be realizing that the all-electric future is, right now, more a mirage than an immediate destination. But they don’t seem to recognize that plug-in hybrids are the sweet spot. So why didn’t I buy one? Why did we buy the VW Electric Microbus ID.BUZZ instead of the Chrysler Pacifica plug-in hybrid?

Share

My view: the plug-in hybrid is truly the sweet spot in automotive technology for all those who are not in the business of using rentals for the occasional extended trip. And yet demand for plug-in hybrids is way lower than shoukd be the case if auto buyers in general agreed with me. Why are plug-in hybrid sales so low?

Adam Tooze presents us with a graph:

Share Brad DeLong’s Grasping Reality

I concur with the American consumer’s increasing recognition over the past two years that the time of the pure EV is not yet: spotty charging-infrastructure, inconvenient length of charging time, and battery weight for an extended range all being big issues. But I do not understand why plug-in hybrid sales are not now a multiple of non-plug-in hybrid.

And what are the 79% who are still buying pure gasoline motors here in the United States thinking? Do they not understand that more than half of themselves—more than half of Americans with cars—drive less than 40 miles/day? Do they not understand that that is the range of a modern Prius Prime or RAV4? And do they not understand that hauling around an extra half-ton of battery just for the occasions when you do exceed 40 miles (and don’t have the opportunity to top off the battery during one of your stops) is inefficient, and also costly?

Well, I should be asking myself, shouldn’t I? I mean, I am the proud owner of a brand-new three-month old VW Microbus: the all-electric ID.BUZZ:

Leave a comment

In part, this was an overreaction to the state of the 22-year-old Subaru station wagon, with its oil leak, water leak, power steering-fluid leak, cracked bearing, non-functional radio, plus the engine never sounded right after the water-pump replacement. In part, this was residual sadness about never having owned a VW Microbus in our youth. (It has been a definite source of pleasure for many of our neighbors: I cannot tell you how many of those in their 70s and 80s have reminisced to us about driving Highway 1 in a VW Microbus on acid back in The Day.)

Mostly it was that too much of the time we had found ourselves saying “we will have to take two cars”. And when we looked at footprint (it is only 1’ longer and 1’ wider than the Subaru was) and wanted serious space in terms of seating for 7 plus a usable cargo area, it really looked as though VW had better engineered the cabin part of the cabin-on-a-skateboard that is a modern EV. The vehicles that came up close for us were the Kia EV9 and the Hyundai Ioniq 9. And that was it.

But why not what I said I think of as a sweet spot—the plug-in hybrid? Well, the only thing that came close seemed to be the Chrysler Pacifica. However, it is 9” longer than the VW ID.BUZZ for 5% less cabin space. Losing the cabin-on-skateboard seemed a serious cost, especially as I envisioned (correctly) a lot of street-parallel and crowded-garage parking here in BayAreaMetropolis.

But, truth be told, we probably should have. We spent an extra $10,000 on the car and tied ourselves to the EV charging-infrastructure hassle because we did not feel like being quite so boring.

Refer a friend

Give a gift subscription

References:

Leave a comment

Subscribe now

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…

#shunning-the-practical-plugin

Reflexes, Desires, Actions, Emergence, & the Mirage of Machine Minds: What MAMLMs Still Ain't

Understanding complexity & emergence is more the road to understanding Turing-class entities than is the current focus on Scaling Laws & the Bitter Lesson…

Share

Can machine learning models soon transcend their sposter roots and approach true Turing-class intelligence? One must be skeptical once one realizes that the reality is not building an AII—an artificial super-intelligence—but rather looking around and recognizing that we are all already embedded in the SSNASI—species-spanning natural anthology super-intelligence—that is the Mind of humanity today.

Share Brad DeLong’s Grasping Reality

Across my screen comes a very nice comment on something I wrote:

from the extremely sharp Richard Baker <https://bsky.app/profile/sharp.blue>:

Richard Baker: <https://bsky.app/profile/sharp.blue/post/3lsikmmh63k22>: ‘People have gotten very excitable about interactive fiction generators that are tuned to generate conversations between people and fictional AIs. The most obvious are [those among] the “AI safety” people who think that it’s some kind of dire warning sign that our interactive fiction generators can generate stories about rogue AIs acting along the lines of scenarios that they’ve previously outlined, and which are therefore in the training corpora. It’s completely obvious that fixed functions from sequences of tokens to probabilities of next tokens, no matter how complex, cannot have plans, goals, desires, fears or any other mental attributes, as they have no state within which things can be represented. They can, however, generate output in which people can “see” plans, goals, desires, values and so on, in exactly the same way we can see such things in Frodo or Odysseus. We have no precedent for such systems, and lack good ways to conceptualise them…

Leave a comment

Perhaps we can say that they have reflexes—complicated reflexes?

One pathway to think about this and related issues is one that I find myself going down again and again. It is this: Iterated fixed functions are Markov processes, and since they have no internal state different from the external current-visible state of the system, Richard is completely correct.

The problem is that there is an old dodge in this business: Change the definition of the current-visible state to include not just what the visible part of the system is doing now, but also what it has done long in the past. Thus everything that might have affected what the invisible internal state is now is included in the arguments of the fixed function. This is in some sense absurd, in that your state vector for your Markov process is now huge. This dodge turns reflexes into, potentially, plans, goals, desires, values and so on.

Of course, this dodge is profoundly unhelpful to those of us speaking English. We are trying to make fuzzy but important distinctions. This dodge keeps us from doing so, And it keeps us from doing so because it erases the key issues: complexity and emergence.

And whenever I get to this point, my mind once again goes—in a reflex, like a low-order Markov process iterating on a fixed function—to this from the smart Scott Aaronson:

Scott Aaronson: Quantum Computing since Democritus <https://cs.famaf.unc.edu.ar/~hoffmann/md19/democritus.html>: ‘You can… say, that’s not really AI. That’s just massive search, helped along by clever programming. Now, this kind of talk drives AI researchers up a wall. They say: if you told someone in the sixties that in 30 years we’d be able to beat the world grandmaster at chess, and asked if that would count as AI, they’d say, of course it’s AI! But now that we know how to do it, now it’s no longer AI. Now it’s just search….

In the last fifty years, have there been any new insights about the Turing Test itself? In my opinion, no. There has, on the other hand, been a non-insight, which is called Searle’s Chinese Room…. The way it goes is, let’s say you don’t speak Chinese…. You sit in a room, and someone passes you paper slips through a hole in the wall with questions written in Chinese, and you’re able to answer the questions (again in Chinese) just by consulting a rule book. In this case, you might be carrying out an intelligent Chinese conversation, yet by assumption, you don’t understand a word of Chinese! Therefore symbol-manipulation can’t produce understanding…. Duh: you might not understand Chinese, but the rule book does! Or if you like, understanding Chinese is an emergent property of the system consisting of you and the rule book, in the same sense that understanding English is an emergent property of the neurons in your brain.

Like many other thought experiments, the Chinese Room gets its mileage from a deceptive choice of imagery—and more to the point, from ignoring computational complexity. We’re invited to imagine someone pushing around slips of paper with zero understanding or insight—much like the doofus freshmen who write (a+b)2=a2+b2 on their math tests.

But how many slips of paper are we talking about?

How big would the rule book have to be, and how quickly would you have to consult it, to carry out an intelligent Chinese conversation in anything resembling real time? If each page of the rule book corresponded to one neuron… then probably we’d be talking about a “rule book” at least the size of the Earth, its pages searchable by a swarm of robots traveling at close to the speed of light.

When you put it that way, maybe it’s not so hard to imagine that this enormous Chinese-speaking entity—this dian nao 电脑—that we’ve brought into being might have something we’d be prepared to call understanding or insight…

Refer a friend

The bull case for the (future) “cognitive” capacities of our modern advanced machine-learning models—MAMLMs—is always made by drawing straight lines on semilog paper and projecting those lines into the future. Or even having them turn further up!

Give a gift subscription

People have been trained to think that drawing straight lines on semilog paper—projecting exponential trends into the indefinite future—is a reasonable thing to do, largely because of the astonishing success of Moore’s Law since 1965. Gordon Moore’s famous observation that the number of transistors on integrated circuits doubled approximately every two years was not merely a curiosity; it was a prophecy that became a self-fulfilling roadmap for the semiconductor industry. For half a century, this exponential trend underwrote the relentless improvement in computing power, fueling everything from the personal computer revolution to the explosion of the internet.

The lesson drilled into the collective consciousness of technologists, economists, and even lay observers was simple: technological progress is not just fast, it is accelerating. And so, when faced with the question of artificial intelligence, it is all too tempting to reach for the same ruler, to assume that what worked for silicon will work for cognition, and that the future will arrive on schedule, riding the smooth upward curve of the exponential.

But this faith in scaling is not just a product of Moore’s Law. It is reinforced by what Richard Sutton has called the Bitter Lesson:

  • in the long run, the most effective AI methods are those that leverage computation and data at scale.

  • Those that encode human insights or hand-craft rules are not so effective.

  • Again and again, brute-force approaches—whether in chess, Go, or language modeling—have outstripped the clever, bespoke methods of earlier generations.

The Bitter Lesson has been, so far, a rebuke to the hubris of human ingenuity: it is not our cleverness, but our willingness to harness massive computational resources, that yields the greatest advances. This lesson, bitter as it may be for those who fancy themselves architects of intelligence, has become dogma in the AI community.

Add to this Feynman’s “room at the bottom” point: the recognition that there is an enormous amount of room for technological progress at ever-smaller scales. In his famous 1959 lecture, Feynman speculated about the possibilities of manipulating individual atoms and molecules, inaugurating the field of nanotechnology. The implication for AI is clear: if we can keep shrinking the building blocks of computation, there is no obvious limit to how much intelligence we might engineer—at least, so the argument goes.

Yet, is it a reasonable thing to do—to draw straight lines on semilog paper and expect the future to conform to our extrapolations?

Here we must stop and get off the trolley.

Exponential trends, as any student of population biology or innovation diffusion will tell you, do not go on forever. They “logisticate”—that is, they follow the logistic curve, which starts exponentially but eventually saturates as it runs into physical, economic, or conceptual limits. Moore’s Law itself is exhausted. The cost and complexity of further miniaturization has mounted too high. The same will be true for “AI” someday. Scaling laws hold up to a point. But eventually, diminishing returns set in, and the curve bends. That someday may have already passed.

And this dodges the truly key issues: complexity and emergence.

Complexity, in this context, refers to the sheer intricacy of the systems we are trying to build or model. Emergence is the phenomenon by which simple rules or components give rise to behaviors or properties that are not obvious from the rules themselves. Both concepts are central to the study of brains and minds—and to the limits of artificial intelligence. The human brain is not just a large neural network; it is a product of billions of years of evolution, shaped by the interplay of genes, environment, and culture. Its complexity is not merely quantitative but qualitative; its emergent properties—consciousness, intentionality, creativity—are not easily reducible to the sum of its parts. Thus, the challenge is not simply one of scale, but of understanding and replicating the conditions under which emergence occurs.

In my view, the natural way to address these questions is through a comparative exercise: to ask how the “size” of the human brain stacks up against the “size” of present and future large language models (LLMs). But “size” here is a slippery notion. Do we mean the number of parameters in a neural network? The number of synapses in a brain? The energy budget? The data throughput? Each metric tells us something, but none captures the whole story. For example, the human brain contains roughly 86 billion neurons and hundreds of trillions of synapses, operating with a power budget of about 20 watts—astonishing efficiency compared to even the most advanced supercomputers. Meanwhile, the largest LLMs boast hundreds of billions, even trillions, of parameters, but their architectures and training regimes are fundamentally different from biological brains. The analogy is tempting, but the differences are profound.

This leads to the first key question: How complex is the human brain, anyway, relative to our models? The answer is, I think, that we are only beginning to grasp the magnitude of the problem. The brain’s complexity is not just a matter of scale, but of organization: its connectivity, its plasticity, its capacity for self-modification. Unlike artificial neural networks, which are typically static once trained, the brain is a living, learning, adapting organ. It is shaped not only by genetics but by experience, by social interaction, by culture. To compare it to a language model is to compare a rainforest to a spreadsheet: both have structure, but one is alive.

If one tries to do a “parameter” count, one gets 10^15 synapses in a human brain, and then raises that by the brain’s dynamic plasticity, real-time learning, and the role of glia and neuromodulators, and the internal complexity of each neuron beyond dendrites in-axon out, you can get up to 10^18 “parameters”. Anthrioic’s Claude is still at 10^11. That suggests as big a difference between our brains and frontier MAMLMs as between today’s frontier MAMLMs and a 200-node four-layer perceptron.

A second key question: How much more than a blank slate can the process of biochemical development from a single cell produce in a brain? Here, the answer is: quite a lot. The process of neurodevelopment is a marvel of self-organization, guided by genetic instructions but profoundly influenced by environmental inputs. From a single fertilized egg, a human embryo orchestrates the division, migration, and differentiation of billions of cells, weaving them into the intricate tapestry of the nervous system. The end result is a brain that is both pre-programmed and plastic, capable of learning languages, playing chess, composing music, and inventing calculus. The blank slate is a myth; the brain is born ready to learn, but it is also endowed with predispositions, biases, and instincts.

A third key question: How close to that limit has evolution trained our process of biochemical development to operate? Evolution is a relentless optimizer, but it is not an engineer; it works with what is available, subject to constraints of time, energy, and reproductive fitness. The human brain is the product of millions of years of tinkering, but it is not perfect. It is, however, astonishingly effective—we think. It balances speed, efficiency, and flexibility in ways that our artificial systems can only envy. Whether we are close to the theoretical limit of what is possible is an open question, but it is clear that evolution has pushed the envelope further than any human designer has yet managed.

Moreover, we are, in a sense, already distributed-computing anthology-mind superintelligences. Humanity’s collective intelligence is not just the sum of individual brains, but the product of communication, collaboration, and cultural accumulation. Humanity is already superintelligence, an anthology superintelligence: a vast, distributed network of minds, technologies, and institutions, each building on the work of others.

I am truly a Turing-class entity.

If I, truly a Turing-class entity, truly working on my own, wanted to use my cognitive skills, my sensors, and my manipulators to open and gain energy to support my functioning from one of the walnuts growing on the tree ten yards to my left, I would have quite a difficult time doing that.

But I can do it easily. I can do it easily because I am backed up by the anthology superintelligence that is humanity. That SSNASI—species-spanning natural anthology superintelligence—backs me up by telling me that:

  1. there are things labeled “nutcrackers”,

  2. they can be made out of steel, a

  3. it has in fact made many such, and

  4. the closest nutcracker is stored in the top-left kitchen cabinet drawer.

In this light, the quest to build artificial superintelligence is puzzling: we want tools for thought to empower and extend that SSNASI we have, not to build an ASI that Nilay Patel can then worship as a god. We are looking for the extension of human capabilities through tools, language, and shared knowledge, not for some bizarro magic-mushroom driven hallucinatory remake of “Frankenstein”.

And what are we, on our current trajectory, actually building with our MAMLMs anyway? Are we honing in on better and better approximations of internet sposters—systems that can mimic the style and content of online discourse, but without understanding or intentionality—or are we building something else? The distinction matters. Pretraining, the process by which LLMs are exposed to vast corpora of text, produces models that are adept at imitation, but not necessarily at reasoning or goal-directed behavior. Reinforcement Learning from Human Feedback (RLHF) seeks to bridge this gap, aligning model outputs with human preferences and values. But the question remains: are we creating minds, or merely mirrors?

Also crossing my screen this morning is this from Andrej Karpathy:

Andrej Karpathy: <https://x.com/karpathy/status/1937902205765607626/>: ‘+1 for “context engineering” over “prompt engineering”. People associate prompts with short task descriptions you’d give an LLM in your day-to-day use.

When in every industrial-strength LLM app, context engineering is the delicate art and science of filling the context window with just the right information for the next step.

Science, because doing this right involves task descriptions and explanations, few shot examples, RAG, related (possibly multimodal) data, tools, state and history, compacting… Too little or of the wrong form and the LLM doesn’t have the right context for optimal performance. Too much or too irrelevant and the LLM costs might go up and performance might come down. Doing this well is highly non-trivial.

And art because of the guiding intuition around LLM psychology of people spirits. On top of context engineering itself, an LLM app has to:

- break up problems just right into control flows
- pack the context windows just right
- dispatch calls to LLMs of the right kind and capability
- handle generation-verification UIUX flows
- a lot more—guardrails, security, evals, parallelism, prefetching,,..

So context engineering is just one small piece of an emerging thick layer of non-trivial software that coordinates individual LLM calls (and a lot more) into full LLM apps. The term “ChatGPT wrapper” is tired and really, really wrong…

Get 50% off a group subscription

This is a task that is a whole different animal from simply throwing more nd more data at ever-larger and larger blank-slate neural network, and hoping that the Bitter Lesson and scaling laws will carry us through. Andrej Karpathy thus appears reality-grounded, in a sense that many of his peers (especially when they are trying to raise money from overgullible venture capitalists) are not. He undertands that the core models are better and better approximations to internet sposters, and seeks effective ways of noooodging them out of their natural sposter nature into doing useful workk.

Above all, as I survey this terrain, I find myself becoming more and more certain of this: The key issues truly are complexity and emergence. And here, not surprisingly, the key conceptual problem is that complexity and emergence are not so much ideas as labels for problems that we cannot solve because we do not yet have real ideas to fill empty boxes.

But until we have ideas to genuinely fill the boxes, I have to believe that anything we could call a Turing-class silicon software entity is, well, as far from our frontier MAMLMs as they are from the Mark I Perceptron Frank Rosenblatt demonstrated at work back in 1960.

Share

References:

Leave a comment

Subscribe now

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…

#reflexes-vs-desires-&-actions

Mamdani vs. Skeffington: Maybe American Politics Is Getting Less Anti-Rational?

We can hope that at least one positive thing to take from Zohran Mamdani’s Democratic primary victory in NYC is that the American electorate is now suffering from at least a bit of outrage fatigue. At least the primary electorate. At least in NYC. At least for the Democratic Party. At least this summer. Why do I counterpose a sober and quiet New York progressive selling concern for the lunch-pail and at least the simulacrum of competence to a a fictional Irish-American machine boss? As litmus tests of the electorate’s appetite for outrage and ethnic signaling…

Share

A very nice catch from Kamil Kazani: Mamdami avoiding a deliberate trap by acting more like a grownup than the other candidates:

Kamil Kazani: Why did Zohran Win? <https://kamilkazani.substack.com/p/why-did-zohran-win>: ‘I think there may be some fundamental factors…. The vibe is changing… Public outrage does not work anymore…. Zohran, he is calm, constructive, and very rarely raises his voice… Oe thing that Mamdania—but almost no one else in the American public space is getting—is that the public is getting tired of the outrage…. Outrage used to be something that helped you get notice. Now, however, there is a massive oversaturation on the market of outrage…. All of your screaming just converges into the indistinguishable white noise….

As of 2025, you can stand out of the crowd, if you behave, show decorum, politiness and self control. In fact, a great deal of Mamdani’s success has been based not on “charisma” per se, but on the fact he was behaving like an adult in a room of toddlers…. He puts forward a moderate, restrained proposition, consciously framed in a way that it would very difficult to make a legitimate counterargument against it: A state with the equal rights for all its citizens. Now the opponents start screaming, and yelling and losing their minds…

kamilkazani
Why did Zohran Win?
Some thoughts on Zohran Mamdani’s victory…
Read more
2 days ago · 24 likes · 2 comments · kamilkazani

Share Brad DeLong’s Grasping Reality

If this in fact did matter—and was in fact a plus for Mamdami—it is an interesting moment to note. Let me approach it elliptically, along a train of thought I was already thinking about as a result of my breakfast last week with Bob Reich, were we talked among other things about the late, great Berkeley political scientist Nelson Polsby

In Edwin O’Connor’s not-quite roman a clef about Boston Mayor James Michael Curley’s novel The Last Hurrah, O’Connor places his protagonist-antihero, Frank Skeffington, in a scene that I always thought showed and was meant to show Skeffington’s deep contempt for his voters and supporters:

To Adam [Skeffington] said, “A good few minutes’ work. Could you hear anything at all”.

“Very little,” Adam said. “The wind took most of it away. But did I hear you mention Portugal a couple of times?”

Skeffington nodded. “Foreign policy,” he said gravely.

“Foreign policy?”

“Very important. A man can’t run for mayor on the domestic issues alone. Not in this day and age. We all have to cultivate the wider vision.”

“You mean that in a local election you have to talk about, say, Russia?”

“No. That’s one of the great handicaps for the local politician: he can’t call his opponent a Communist. It’s a shame, but there you are. Of course you can call your opponent a Communist if you really want to, but it won’t do you any good; nobody’ll believe you. They all know he goes to Mass on Sunday, so he can’t be a Communist; you might just as well say that the Cardinal and the Kremlin exchange pen-pal letters. No, Russia and Communism never have been much of an issue around here. We’re under the disadvantage of having to evolve a foreign policy that meets local requirements.”

“Which includes what?” Adam asked. “Portugal?”

“You’d be surprised how important Portugal becomes,” Skeffington said, “when you’re speaking to the Portuguese. These fishermen, almost all, came originally from the Portuguese mainland or the Azores. I find they appreciate an occasional reference to the glorious country of Henry the Navigator. I’ve been trying to find a more contemporary figure than Henry, but with Portugal that’s not so easy. However, it isn’t a major point. There aren’t enough Portuguese. When you come right down to it, there are only two points that really count.”

“Such as … ?”

Skeffington held up two fingers.

“One,” he said, ticking the first, “All Ireland must be free.

“Two,” he said, ticking the second, “Trieste belongs to Italy.”

“They count. At the moment the first counts more than the second, but that’s only because the Italians were a little slow in getting to the boats. They’re coming along fast now, though; in twenty yyears the Irish issue will be about as burning as that of Unhappy Ethiopia. Fortunately, I don’t expect to be among those present at the time.”

Leave a comment

If you wanted to be more friendly to Skeffington than I think O’Connor is to his character, you could say that even though the passage is satirical, it is not simply a Skeffington sneer at the voters’ intelligence, but merely a commentary on the transactional, performative, and sometimes absurd nature of mid-1900s urban machine politics, where ethnic identity and symbolic gestures often mattered more than policy substance, with Skeffington being not so much contemptuous as he is pragmatic and a little cynical about the rituals of politics. He knows the game, and he plays it with a wink.

But that is belied by the end of the story. Because the crux of the story told in the book is that Skeffington does not in fact know the new game, and does not play the new game successfully;

What happened next was not recognized by everyone. Characteristically, it was Skeffington himself who caught the first sign. The Ward Seven tabulation had stopped; the vote from Gorman’s ward continued to yield its comforting harvest; one by one, the other wards had begun to declare themselves. Now the blank spaces… received their first chalked markings; Skeffington examined them swiftly with a sharp, professional eye. He had hardly begun the examination when he stopped short. Like an engineer, powering at full speed down a long-familiar stretch of track, he was jolted from routine…. It was nothing more than a small return from a single precinct…. In this small, reliable precinct, the race was proving to be surprisingly dose; he was barely squeezing by…. [Was] the fluctuation was part of a pattern, or… in fact a freak….

“Sam,” he said in a low voice, “go into the office and get on the phone. Talk to Teddy, Charlie Ferrino, and Mike Gallaher. Don’t talk to anybody else; stay there till you get them but get them fast. Find out what’s going on. If anything.” Weinberg went off, wordlessly. In the same low voice Skeffington said, “What’s your guess?” “I don’t know, Frank. I don’t like it much,” Gorman said candidly….

He looked towards the door of the private office: the reappearance of Weinberg was what mattered now. Even as he looked the door opened: Weinberg emerged and came towards him…. “Well, Sam?”

“I dunno,” Weinberg said. “Something’s up, and it ain’t good. I don’t figure it yet, but I tell you this: I think we got trouble.”

“All right,” Skeffington said impatiently. “We’ve got trouble. Where? What kind of trouble? Come on, Sam, out with it. Fast.” In swift, blunt terms, Weinberg told his story; it was not a comforting one…. McCluskey was revealing unexpected strength. He was piling up respectable totals in a number of areas considered inviolable; in several marginal precincts he was actually leading. What it all meant, whether this was or was not the beginning of a turning tide, was not quite clear; the next half-hour would tell….

In the office, he snatched up the telephone and made a half-dozen calls to key points; in each call he snapped a series of abrupt questions and quickly cut short any sustained reply. He was after the bare bones of information, and by the sixth call he had them. He put down the phone and said, “All right. Now at least we’ve got some sort of general picture. It’s not a pretty one. We’ve got trouble: you were right, Sam. I should say reasonably serious trouble. The boy’s begun to pull in about thirty precincts, and widely scattered ones at that. In a couple of them he’s started to snowball.”…

Gorman agreed; he said thoughtfully, “I’d say it’d swell before it shrinks, that’s sure. It damn well could hinge on how the East End holds. You’ve heard nothing from Cavanaugh or Paddy Montgomery?”

Skeffington shook his head. “No. It’s too soon.”… He had wanted his position defined; it had been defined with a brutal clarity. He knew now exactly how desperate it was. And, faced with the fact, he could arrive at no reasonable explanation for it; he had been hit especially hard because he had been completely unprepared for the blow. Too shrewd and too experienced to be guilty of overconfidence, he had come to his headquarters tonight secure in the knowledge that he had kept his guard up, that he had taken every necessary precaution, that no stray sign of danger was to be seen. And yet the cyclone had struck, rushing out of nowhere, preceded by no warning; he knew now, better than anyone else, that in the face of this development his chances of winning comfortably had vanished….

He said steadily, “Very well. We may be all right. If the East End holds, we surely will. In any case there’s just one thing we can do now and that’s to sit tight and see what happens. Not a very aggressive course, but at the moment we haven’t a great many alternatives.”…

Skeffington’s right hand moved in a slight gesture of impatience. “No. Does it matter? At this point, that is? Obviously there were slip-ups all along the line. All right. We can attend to them later. What matters now - and the only thing that matters now - is what’s going to happen in spite of the slip-ups. We can play parlor games and fix responsibilities when this is all over.”

Weinberg had risen. Anger over probable betrayals had driven him from slouching immobility; he walked restlessly across the limited floor-space of the small office, his dull, lackluster eyes now bright with fury. “Those bastards!” he said. “O.K. So we fix responsibilities later. So what do we do now? Sit around on our keesters and wait for the moon to come up? Or do we go out and try to fix this mess?”

For just an instant Skeffington’s temper flashed through. “Fix it with what?” he said savagely. “What would you suggest? Sticking plaster? Staples? How in hell can you do any fixing at this stage of the game? Everybody’s voted. It’s all over. Do you think we ought to bomb the polls? Assassinate McCluskey? For the love of Christ, wake up to realities!”…

He heard the office door open; he sensed, rather than saw, the swift approach of Weinberg, bringing with him the news that, at worst, would be additionally unpleasant - at best, utterly useless. For the East End, which just a few minutes before was of such vast significance, now no longer mattered at all. With or without it, Skeffington was beaten, for the signs he had read did not point merely to defeat; they pointed to cataclysm. Everywhere, in all precincts, the McCluskey totals were on the move; of all the wards in the city, only John Gorman’s held firm in the Skeffington camp.

Throughout the room the loyal supporters watched these soaring figures but, mindful of the words of their leader, they were waiting eagerly for the turn to come; only Skeffington - and now Gorman - knew that there would be no turn. For he was not simply to be beaten; he was to be buried, buried under a great, roaring, sliding avalanche of votes which at any moment now would come sliding down for everyone to see. It was the thing of which he had not even dreamed: it was the McCluskey landslide. It was unthinkable, unimaginable, and yet - it was happening….

But most of all, the defeat was a mystery to Skeffington. Less limited than his colleagues, he could not bring himself to ascribe the catastrophe to the simple, single cause. He had thought immediately of the possibility of betrayal, of a failure in his organization; he had dismissed the thought. He knew that neither singly nor in combination could they have occasioned his defeat. Undoubtedly they had been there, but they had always been there, and what had beaten him now was not something old, but something altogether new. What it might be, he simply did not know. He tried desperately to think it through, but it was no use. His mind would not work properly. It churned away, but its edge had been dulled by the disaster; it bit into nothing, it merely raced by images and impressions, bouncing off them, assimilating nothing and arriving nowhere. In a kind of narcosis, he could grasp solidly only the one fact: he had ridden into the election a confident man, backed by four years of his most effective administration; in the twinkling of an eye, he had been knocked out of office, repudiated overwhelmingly, subjected to humiliation which he had never thought possible. And always the question returned: Why? Why the debacle?… All through these stunned minutes he came no closer to the solution; he stood staring straight ahead at the board, instinct and habit controlling him, warning him to be still…

Give a gift subscription

In the book, it is the New Deal, immigrant assimilation, and the rise of impersonal, bureaucratic government with its social-democratic social-insurance programs that have made the old boss system based on the idea that he can get your cousin Timmy a job with the fire department obsolete. Urban political machines like Skeffington’s had flourished when American cities were teeming with new immigrants—Irish, Italian, Azorean, Jewish, other Eastern European—who had arrived with little money, often, limited English, few social connections, and no social-network power to maneuver in the New World society. The bosses—archetypes like Tammany’s “Big Tim” Sullivan or Boston’s James Michael Curley—offered tangible help: jobs, coal in winter, legal assistance, and a sense of community. In exchange, the machine expected political loyalty at the ballot box. This system, for all its corruption, was rooted in a direct, personal quid pro quo that alltowed immigrants to cash in the societal power that was their ability to vote for concrete things that made their lives easier.

But the coming of the social-democratic New Deal Order shifted the provision of social welfare from the hands of local bosses to the impersonal state. No longer did a desperate widow need to petition the ward heeler for a city job or a bag of groceries; she could now rely on a federal check. The “iron rice bowl” was now filled by Washington, not the ward office. At the same time, assimilation and upward mobility eroded the ethnic solidarity that was the machine’s foundation. The children and grandchildren of immigrants moved to the suburbs, entered the professions, and shed the old tribal loyalties. They became less interested in the politics of the parish hall and more in the politics of the school board and the zoning commission. The rise of bureaucratic, professionalized government—civil service exams, regulatory agencies, and the like—further diminished the scope for personal patronage. In this new world, the old boss was not just obsolete; he was an anachronism, a figure of nostalgia rather than power.

And thus by the time of Edwin O’Connor’s novel in 1956 there were no longer elections to be won by, among other things, flogging over and over again the only two things you needed to know about foreign policy—that all Ireland must be free and that Trieste must belong to Italy. These gestures—and everyone knew they were gestures—mattered because they spoke to voters’ lived experience of exclusion and their yearning for respect and recognition. But the grandchildren of immigrants, educated in American schools and working in American firms, were less interested in the politics of Lisbon or Limerick and more in the politics of Levittown.

This is not to say that the kowtows to symbolic identity politics ever disappeared—they merely mutated. The politics of the old neighborhood, with its saints’ days and parades, gave way to the politics of television, polling, and mass marketing. Skeffington’s mastery of the old rituals became, in this new environment, a kind of fossil skill—impressive, but no longer decisive.

And we can hope that we see an analogous change now, and a change for the better.

What once seemed daring and effective early in the social-media age —public outrage, ethnic signaling, and performative politics—may have become, in the age of social media saturation, pure background noise. Kamil Kazani’s analysis of Zohran Mamdani’s victory in New York hopes so: Kamil claims that Mamdani’s calm, adult comportment stands out precisely because it is so rare in a marketplace glutted with outrage merchants. So Skeffington’s old dance moves flopped in the mid-1950s.

However, a warning: It may be that Mamdani’s victory is a sign of a similar shift. It may be that Mamdani’s position here was a plus among the turning-out Democratic primary electorate, and so a good sign for an electorate capable of pursuing a politics and ethics of responsibility, in New York City. But even if so, it is only in New York City. And it is only in the Democratic primary. It may not be so in the general election. It may not be so elsewhere in America.

The primary electorate, often more engaged and informed, may reward appeals to reason, policy, and civic virtue. Or the primary electorate may be composed of True Believers who demand ideological purity above all. It depends. On many things.

But what is true is that the general electorate is different. It is broader, more heterogeneous, less informed, and also less committed. It may or may not be more susceptible to spectacle, charisma, and demagoguery.

We have seen, time and again, that the politics of spectacle and unreason are never far from the surface—whether in the populist surges of the late 19th century, the radio demagogues of the 1930s, or our current social media hell The decline of the old boss system did not guarantee the triumph of reason; it merely changed the terrain on which politics was fought.

As then, so now.

Refer a friend

References:

Get 50% off a group subscription

Leave a comment

Subscribe now

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…

#public-reason
#mamdani-vs-skeffington
#the-last-hurrah
#zohran-mamdani
#edwin-oconnor

Politics as Making Hardwood Furniture with Dull Tools: Mamdani's NYC Primary Win Is Not a Win, But Only an Uncertain High-Stakes Bet

The Democratic primary is just the preseason, in that it does not count at all. A clear-eyed view of this is essential for successfully building coalitions and avoiding catastrophes: what Max Weber and John Maynard Keynes have to say to overenthusiastic supporters of Zohran Mamdani…

Share

Share Brad DeLong’s Grasping Reality

It is reasonably well known as these things go that since the first Trump administration, I have been a firm believer that those of us write social Democrats or left neoliberals who wish good things for America in the world have no live political options, except to pass the baton to those on our left. We should then give them good advice. We should hope they do not drop the baton. We should hope that they run in the right rather than the wrong direction, but until the center right possesses any ac executiontors with political force who are neither grifter, morons, fools, nor cowards, that is our only option if we seek to accomplish good things for America and for the world.

However, this morning I confess I wince, very strongly wince, at the headline and the vibe here. I pick on John Ganz because he is (a) hard-working and knowledgeable, (b) intelligent, and © good-hearted:

John Ganz: What It Took To Win <https://www.unpopularfront.news/p/what-it-took-to-win>: ‘Thoughts on Zohran Mamdani’s Popular Front…. We were told…. We were told repeatedly…. And we were told…. The lesson Mamdani and his strategists evidently took… was that… [there was] an opening for another kind of protest politics entirely, one that was constructive and positive…. There was an entire universe of motivated voters (and, importantly, volunteers) out there just waiting to be reactivated: Veterans of Bernie 2016, 2020, Warrenistas, and all the civic movements of the 2010s…. Mamdani does well in the middle, which in New York, with its high cost of living, stretches well into the six figures…. Unionized wage laborers, junior white-collar professionals, and small business owners. Say what you like about their feasibility, the major policy portions of Mamdani’s campaign were about cost-of-living issues, and he targeted a coalition that goes across cultural and racial backgrounds but were all struggling to build decent lives in New York….

The old civic associations’ power to control and mobilize the electorate has been steadily weakening…. The “machine”… pre-dates even the labor union, the radio, and television!… Mamdani mastered the new and dominant form of civic association: the Internet…. Politics is ultimately about speaking in public. Find someone with powers of self-expression and you’re in business….

The electorate doesn’t really exist until election day, and the politician and his or her campaign are actively creating that electorate. All political errors, from the level of action to analysis, are based on reifying the situation, believing in a static, factual situation that cannot be changed. And all great political successes are based on the opposite: the art of the impossible; believing in a chance for something new…

Unpopular Front
What It Took To Win
The old saw goes that politics is the art of the possible, but it often seems like the art of the impossible: putting together winning coalitions that “objectively” seem to defy logic—or at least, the logic of the pundits.) My favorite example of this is Trump managing to spike the white working class vote and fulfilling the lifelong Republican dream of…
Read more
5 hours ago · 71 likes · 6 comments · John Ganz

Leave a comment

But: “Winning” a Democratic-Party primary wins you nothing.

The primary election is a choice process in which people who regard themselves as Democrats together—by whether they turn out or not and who they then vote for—select a candidate to put forward to the general electorate for its consideration on the Tuesday after the first Monday in November. John Ganz should thus not be enthusing about “winning”, “something new”, “great political success”, “creating the [primary] electorate”, a “a coalition [of]… all struggling to build decent lives”, “protest politics… constructive and positive”, “universe of motivated voters… and volunteers”, and so forth.

John should be looking backward: Did this process select a candidate who is in the sweet spot with respect to properly balancing (a) the likelihood that the general-election voters will in fact approve the Democratic Party’s candidate to be Mayor of New York City, and (b) the likelihood that that candidate as Mayor will build governmental-bureaucratic coalitions to implement policies that will successfully advance the good parts of Democratic-Party values and preferences? And John should be looking forward: © What needs to be done, starting today, to maximize the long-run policy and governance wins should we be lucky enough to find Zohran Mamdami mayor of New York City come next year?

Read more

Fourteen-Lion Parade

Parking this here so I can find it easily. In 2007 earthquakes reopened the Savuti Channel from the Okavango Delta—where the Cuando/Okavango River flows into the Kalahari Desert and dries up—to the Mababe Depression. What had been a November-March estival wetland then became, once again, a year-round wetland…

Share

Share Brad DeLong’s Grasping Reality

Dawn…

The ten lions from the previous evening led by their matriarch—on the road, to move fast and silently, without rustle, looking for breakfast. They are trailed by the three three year-old cat-boys from the previous morning, and by the five year-old zebra kill-hogging Fred from the previous evening.

And so we have the fourteen-lion parade:

They will turn right and start hunting upwind…

Now heading upwind and about to go belly-to-dirt. In a minute, even with our monkey eyes, it would be very difficult to tell by sight if there were any lions—let alone fourteen—within fifty yards, crawling slowly upwind through the grass.

We were extraordinarily, extraordinarily privileged to have the world-class Jonah Seboko as our guide…

Share Brad DeLong’s Grasping Reality

References:

Give a gift subscription

Leave a comment

Subscribe now

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…

#mababe-depression

#fourteen-lion-parade

HOISTED FROM COMMENTS: RAFAEL KAUFMANN: Carving Nature at the Joints: Faithful Representation, the Platonic Dream, & the Unreasonable Near-Success of GPT LLM MAMLMs

Perhaps why reality’s low-dimensional secrets let machines mimic us so well—and where the limits remain to the surprising power—and peril—of compressing the world into latent spaces. Or, when brute-force statistics become near-“intelligence,” and what’s still missing from the algorithmic mind…

Share

Share Brad DeLong’s Grasping Reality

Why do large language models seem so uncannily adept at predicting human discourse? And given that they are so successful at language games, why are they so inept at reality-description. Except, of course, where they amaze us by not being inept at reality-description. But unless you are a subject-matter expert, you can be easily fooled by their hallucinations?

Rafael Kaufmann’s incisive commentary, spotlighted here, contends that the surprising efficacy of LLMs is rooted in the principle of “faithful representation”—the idea that the world, at some deep level, is compressible into a handful of latent-variable dimensions, just as Plato once imagined. The real reality is not the buzzing chaotic confusion to which we try to give names, but the underlying logic and order of the Forms. This property, he argues, is why both scientific models and machine learning algorithms can “work” at all: reality’s underlying simplicity allows for effective compression, prediction, and imitation.

But as we automate the carving of nature at its joints, the question of true understanding looms ever larger. In the process of being stochastic parrots that emulate internet s***posters—LLMs must construct internal representations that map the hidden structures of language. The hope was that such internal representations would also map onto the underlying structures of reality—for since language is a tool to describe the world and how it works, the two should be, if not one-and-the-same, very close. Their statistical training will first find the same compressed mappings that undergird human language-games and then make the leap to the same compressed mappings that undergird human understanding. That was the hope. That is still the bet of those who see AGI arriving from GPT LLM MAMLMs any day now.

Yet, so far, they largely fail. But might they succeed? Someday? Soon?

Kaufmann’s commentary, hoisted for your consideration, explores how statistical models find the Platonic forms beneath our words. Yet, the leap from correlation to causation and from plausible verbiage to actual intelligent insight remains a chasm even the cleverest algorithm struggles to cross. Tracing the lineage from ancient philosophy to modern machine learning, we try to grasp the promise—and the peril—of machines that claim that their stochastic parrotage is on the edge of true understanding.

But the leap from successful representation of the shadows on the walls of the cave to true understanding of Nature’s Forms remains to be made. Learning causal models—discerning which states of the world cause which others—remains a far harder technical nut to crack, requiring intervention and experimentation rather than mere observation. Kaufmann notes that while LLMs can interpolate between the “little stories” humanity tells to explain the world, their ability to extrapolate reliably is tightly bound to the quality of their training data and the absence of direct grounding in reality.

The automation of representation is a monumental achievement, explaining the leap from “just autocomplete on steroids” to the breathless claims of AGI and ASI superenthusiasts. Yet, the absence of grounding in direct experimentation and the sensitivity to training data quality mean that LLMs, for all their prowess, remain fundamentally limited. The Platonic ideal is within reach, but the Promethean fire of true understanding has yet to be stolen.

Leave a comment

Apropos of: J. Bradford DeLong: Spreadsheet, Not Skynet: Microdoses, Not Microprocessors <https://braddelong.substack.com/p/spreadsheet-not-skynet-microdoses>

Give a gift subscription

Rafael Kaufmann of Gaia Network <https://engineeringideas.substack.com/p/gaia-network-an-illustrated-primer> <https://engineeringideas.substack.com/p/gaia-network-a-practical-incremental> writes:

Rafael Kaufmann: <https://braddelong.substack.com/p/spreadsheet-not-skynet-microdoses/comment/129169221>: ‘What Brad/Cosma’s explanation of LLM misses (along with most others) – the reason why Brad’s intuition “strongly leads [him] to think that they should not be able to do half as well as they do”—is the concept of faithful representation. Put simply: Reality allows itself to be mapped in a low-dimensional latent space, and some compression schemes are just good mappings. [Or at least (the instrumentalist view), of humanity’s currently-known shared ways to describe and predict our observations of reality.]

This is why our scientific and statistical models work in the first place. Brute-force ML works because (or to the extent that) it finds these mappings. LLMs work because their training process finds the same latent representations as (or equivalent to) the representations that internet sh*tposters have in their heads as they’re writing [1]. Therefore, they can quite accurately predict what a sh*tposter would have said, in pretty much any context.

Yes, this is just Plato’s concept of “carving nature at its joints” from ~2400 years ago. Quite surprisingly, it turns out that we’re all living at the unique moment in humanity’s history where we are discovering that such a carving is an objective possibility, and that we can, for the first time, automate this carving [2, 3]. And this does not just apply to language, BTW: if the success of multimodal models weren’t enough, tabular foundation models [4] demonstrate that good old statistics has the same property.

There is still the crucial question of whether/when, beyond “just” the latent representations, ML can learn the true causal world model: which latent states at t cause which latent states at t’ > t. This is, in a technical sense, much harder than learning temporal correlation, and at least in some cases, it’s impossible to learn just from the data, requiring the capability to intervene/experiment [5]. However, it has recently been proven that learning a good causal model is a strong requirement for robust behavior—ie, reliable extrapolation beyond the data [6].

We’ve been lucky so far that the Internet already has a lot of natural-language descriptions of causal models of just about everything. These can be compressed into “meta-representations” that let LLMs “interpolate extrapolations”. This is similar at some level to how humans learn much of their own extrapolation capability—not by experimenting themselves, but by learning theory from other people and representing it in their heads as “little stories” that they can interpolate.

However, because these meta-representations can’t be directly grounded in observational data, the way LLMs learn them is very sensitive to the quality of the training corpus. That is why stuff like embodied learning and synthetic data are hot topics in AI: This is stuff you want to get exactly right.

Regardless, “just” getting the representation right is a huge step in the direction, and it goes a long way to explain/justify the conceptual leap from “just statistics on steroids” to “AGI” made by LLM stans.

[1] https://arxiv.org/abs/2405.07987

[2] https://arxiv.org/abs/2505.12540v2

[3] https://aiprospects.substack.com/p/llms-and-beyond-all-roads-lead-to

[4] https://www.nature.com/articles/s41586-024-08328-6

[5] https://smithamilli.com/blog/causal-ladder/

[6] https://arxiv.org/abs/2402.10877

Get 50% off a group subscription

Refer a friend

Leave a comment

Subscribe now

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…

#carving-nature-at-the-joints

In the Country of the Tswana

In 2007 earthquakes reopened the Savuti Channel from the Okavango Delta—where the Cuando/Okavango River flows into the Kalahari Desert and dries up—to the Mababe Depression. What had been a November-March estival wetland then became, once again, a year-round wetland…

Share

Share Brad DeLong's Grasping Reality

Leave a comment

Yes: that’s Puumba up there, because hyenas and jackals cannot climb trees, but leopards can, even carrying a full warthog in their jaws…

Give a gift subscription

Dawn breaks…

Get 50% off a group subscription

Cat tracks—moving down the road: move faster with no rustle…

Cat pee—recent…

Cape buffalo...

The cats we had been following—three 2-3 year old males…

The cat-boys close in…

By now, one of the three should have diverted off and gone belly-down in the grass—waiting for the other two to show themselves, rush, and hope to drive a young/old/weak/cape buffalo into the ambush spot. None of them have…

Herd bull says: “You are too close, cat-boys!” But what are the other bulls doing? Grading him on form?

Warning given…

Cat-boys move off…

Cat-boys take a short rest…

Coming our way…

Cat-boy #3 gives us a wider berth…

Snoozing cheetah at mid-day…

Our first vulture sighting…

Five year-old Fred with a zebra kill—a kill he did not make…

Ten other lions—the ones who made the kill, led by a matriarch—around in the grass; Fred is hogging the kill, taking occasional bites, and growling at others who get too close…

Three of the ten licking up some spilled entrails…

The next day: dawn…

The ten from the previous day—again: on the road, to move fast and silently…

The Fourteen-Lion parade: the ten juveniles and females from the previous evening led by the matriarch, joined by the three cat-boys from the previous morning, plus the five-year-old Fred bringing up the rear. Again: moving swiftly and silently along the road with no rustle; and then they will turn right and start hunting upwind…

Now heading upwind and about to go belly-to-dirt. In a minute, even with our monkey eyes, it would be very difficult to tell by sight if there were any lions—let alone fourteen—within fifty yards, crawling slowly upwind through the grass. We were extraordinarily, extraordinarily privileged to have the world-class Jonah Seboko as our guide…

Refer a friend

References:

Leave a comment

Subscribe now

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…

#mababe-depression

Spreadsheet, Not Skynet: Microdoses, Not Microprocessors

The modest power relative to the size of the economy of GPT LLM MAMLMs as linguistic artifacts. Of course, since the economy is super huge, a relatively modest effect on it is still huge one. But more an Excel-class innovation than an existential threat to humanity. Or am I wrong?

Share

I am pretty confident that “AI” will, in the short- and medium-run, have a minimal impact on measured GDP and a positive but limited impact on human welfare (provided we master our attention uses of ChatGPT and company rather than finding others who wish us ill using them to enslave our attention. I just said so, earlier today: J. Bradford DeLong: MAMLM as a General Purpose Technology: The Ghost in the GDP Machine <https://braddelong.substack.com/p/mamlm-as-a-general-purpose-technology>:

Share Brad DeLong’s Grasping Reality

But in my meatspace circles here in Berkeley and in the cyberspace circles I frequent, I find myself distinctly in the minority. Here, for example, we have a very interesting but I must regard as weird piece from the smart and thoughtful Ben Thompson:

Ben Thompson: Checking In on AI and the Big Five <https://stratechery.com/2025/checking-in-on-ai-and-the-big-five/>: ‘There is a case to be made that Meta is simply wasting money on AI: the company doesn’t have a hyperscaler business, and benefits from AI all the same. Lots of ChatGPT-generated Studio Ghibli pictures, for example, were posted on Meta properties, to Meta’s benefit. The problem… is that the question of LLM-based AI’s ultimate capabilities is still subject to such fierce debate. Zuckerberg needs to hold out the promise of superIntelligence not only to attract talent, but because if such a goal is attainable then whoever can build it won’t want to share; if it turns out that LLM-based AIs are more along the lines of the microprocessor—essential empowering technology, but not a self-contained destroyer of worlds—then that would both be better for Meta’s business and also mean that they wouldn’t need to invest in building their own…

For the sharp Ben Thompson, he has inhaled so much of the bong fumes that for him the idea that the impacts of LLM-based AIs… along the lines of the microprocessor is the low-impact case. The high-impact case is that they will become superintelligent destroyers of worlds.

This seems to me crazy.

Most important, I have seen nothing from LLM-based models so far that would lead me to classify the portals to structured and unstructured datastores they provide as more impactful than the spreadsheet, but for natural-language interface rather than for structured calculation. And I see nothing to indicate that they are or will become complex enough to be more than that. Back in the day, you fed Google keywords, and it told you what the typical internet s***poster writing about the keywords had said. Poke and tweak Google, and it might show you what an informed analyst had said. That is what ChatGPT and its cousins do, but using natural language, and thus meshing with all your mental affordances in a way that is overwhelmingly useful (because it makes it so easy and frictionless) and overwhelmingly dangerous (because it has been tuned to be so persuasive).

(Parenthetically, Thompson’s conclusion that in the low-impact case Facebook “wouldn’t need to invest in building their own” models is simply wrong. Modern advanced machine-learning models—MAMLMs—are much more than just GPT LLM ChatBots. The general category is very big-data, very high-dimension, very flexible-function classification, estimation, and prediction analysis. 80% of the current anthology-intelligence hivemind share right now is about ChatBots, and only 20% about other stuff. But the other stuff is very important to FaceBook: Very big-data, very high-dimension, very flexible-function classification, estimation, and prediction analysis is key for ad-targeting. And FaceBook needs to keep Llama competitive and give it away for free to cap the profits a small-numbers cartel of foundation-model providers can transfer from FaceBook’s to their pockets.)

Now I see the GPT LLM category of MAMLMs as limited to roughly spreadsheet—far below microprocessor—levels of impact because I see them as what I was taught in third-grade math to call “function machines”: you give them something as input, and they burble away, and then some output emerges. start with the set of all word-sequences—well, token sequences—that can be said or written.

Leave a comment

Read more

MAMLM as a General Purpose Technology: The Ghost in the GDP Machine

Silicon Promises, estimated statistics, sunnier realities: the slow diffusion of “artificial Intelligence”. rom steam to silicon, history’s most transformative inventions rarely deliver on schedule—or as advertised—in the short- and the medium-run. The true transformation of work, wealth, and society will be slower, messier, and far more uneven than Silicon Valley’s PowerPoint prophets would have you believe. If history is any guide, the productivity revolution will come not with a bang, but with a series of small, cumulative changes—provided we have the patience, and the institutional imagination, to see it through…

Share

Torsten Slok believes in AI, macroeconomically:

Torsten Slok: Productivity Gains Are Coming <https://www.apolloacademy.com/productivity-gains-are-coming/>: ‘The Census conducts a biweekly survey of 1.2 million firms, and one question is whether a business has used AI tools such as machine learning, natural language processing, virtual agents, or voice recognition to help produce goods or services in the past two weeks, see chart below. Nine percent of firms reported using AI, and the rising trend in AI adoption increases the likelihood of a rise in productivity over the coming quarters:

Share Brad DeLong’s Grasping Reality

But by how much?

Read more

Cattle, Social Cohesion, Culture, Civic Cohesion, Collective Uplift, & Civilization in Sub-Saharan Africa; Causality; & the TseTse Fly

Marcella Alsan’s bioculturalsociogeographical take on long-run African economic development: why the story is radically incomplete without the tsetse fly: one insect and the fate of a continent, for it looks like the most important player in Africa’s economic history wasn’t a king or a conqueror, but a bloodsucking fly…

Share

Nick Decker reminds me of the great Marcella Alsan’s job market paper of a decade and a half ago. How the tsetse fly, by killing cattle and stymieing plow agriculture, set the stage for centuries of institutional divergence: a lack of plows and of centralized states leading to the peculiar vulnerabilities of its societies to modes of imperial-colonial exploitation.

Her thesis is fully persuasive. And yet it is part of a pattern of very good articles that together I find somewhat problematic, for history can only be truly causally explained once.

Share Brad DeLong’s Grasping Reality

Nick Decker <https://x.com/captgouda24/status/1937388725299781764>: ‘Why was Africa so empty, and for so long?… Marcella Alsan argues that it was the TseTse fly which did it, by rendering vast swathes of land uninhabitable. It is not the direct effect of sleeping sickness which matters, terrible though it is, but how it kills livestock. The wild game of Africa is immune, but it is absolutely murderous to cattle. Without cattle, you can have no plow. You have no dung to fertilize crops. You have no carts or wheels, or even packhorses—you must rely upon human portage. Without carts, there are no roads. In short, productivity tanks and centralization is impossible…. The Bantu migration was recent, and they abandoned crops and animals when they crossedh terrain unsuited for them…. She finds truly enormous effects….

In particular, it increases the incidence of slavery, an institution which is incredibly toxic to sustained economic growth. Without livestock, the value of humans goes up, and without cities, they can’t run. What might Africa have been like without the fly? We already know — there would have been states, just as there was in Ethiopia and Great Zimbabwe. It is little surprise that those ethnic groups which perform the best today are those which have had the longest exposure to centralization, states, and reading. It would not surprise me if their descendants are still afflicted by the environment thousands of years hence. An incredible paper…

Give a gift subscription

Not “uninhabitable” for people, but uninhabitable for oxen, and oxen being an absolutely key factor in the web of biotechnology throughout all of the non-rice lands of Eurasia-Africa.

People began to herd domesticated cattle in Africa in large numbers starting around the year -5000. Remember: back then the Sahara was very different than it is today. It was in no sense a barrier. Crossing it was in no sense arduous. Then was the African Humid Period, when rainfall was much higher, monsoon patterns were stronger, and the region supported a rich array of wildlife, as well as human communities. Instead of the vast, arid desert we know today, it was a landscape of lakes, rivers, grasslands, and scattered woodlands—a true “Green Sahara.”

As of the year -1500, in Africa, we find substantial numbers of oxen and ox-cultures north of the tsetse fly belt, but not south. We find oxen no longer in the Sahara, which is now too dry, but along its northern and southern edges, in the Nile Valley, and in the Ethiopian highlands. Then the Iron Age, along with the yam and the oil palm. comes to the Bantu population around the Benue River Valley, in the borderlands of southeastern Nigeria and western Cameroon. And the Bantu expansion follows: not a conquest, or, rather, not just a conquest. We see a slow, multi-century process of migration, adaptation, and cultural diffusion, radiating out from this heartland. As Bantu-speaking peoples moved, they carried with them not just language, but a package of agricultural and technological innovations that allowed them to thrive in new environments—except, of course, where the tsetse fly or dense rainforest posed insurmountable barriers to their lifeways:

With newly-bread somewhat trypanotolerant cattle like the N’Dama in West Africa, and following the “green highway” that was the Lake Victoria-Tanganyika-Malawi corridor, they made their way through the highlands along the west edge of the rainforest into the Zambezi-Limpopo-high veldt corridor, bringing culture, technology, and cattle all he way down to the Cape of Good Hope

And so we get to Marcella Alsan’s results using bioecological climatic suitability of a region to the tsetse fly as her instrumental variable. Her TseTse Suitability Index (TS) predicts significant, detrimental effects on development only within Africa. Outside Africa, the same index has no explanatory power. And a one standard-deviation increase in the TSI is associated with:

  • a 22%-point decrease in the likelihood that an African ethnic group had large domesticated animals.

  • a 9 %-point decrease in intensive cultivation (i.e., less use of land-saving, labor-intensive farming).

  • a 7%-point decrease in plow use.

  • a 19.5%-point increase in the probability that females performed the majority of agricultural tasks.

  • a 12%-point increase in the likelihood an ethnic group used slaves.

  • an 8%-point decrease in the probability an ethnic group was politically centralized.

  • A one standard deviation increase in TSI is associated with

  • a 28% decrease in intensive farming.

  • a 46% reduction in population density in 1700.

Her simulations of Africa without the tsetse fly generate a continent with only half as much indigenous slavery, and nearly double the pre-colonial population. Plus the TSI has a negative correlation with current economic performance that is robust to including colonizer-legal origin effects.

What do I think of this?

Read more

Regency or Ruin: The 25th Amendment, Trump the Chaos-Monkey, & Zero Adult Supervision of the Use of American Power

What happens when a government built for grown-ups finds itself governed by the credulous and the credulous alone? And the Republican Party Grandees do not have the mojo and the ovaries to stage an adult intervention? We are finding out. It is not good…

Share

Share Brad DeLong's Grasping Reality

The American presidency was designed for the competent—yet we find ourselves ruled by a man who believes the last thing he hears, no matter how fantastical. And so Donald Trump’s “foreign policy” reflects only the sycophancy of those closest to him—and is the chaotic product of a president whose memory has deep-sixed all knowledge of the hot stove he touched yesterday. Impeachment, the 25th Amendment, a regency—there ought to be options. But with no George Shultz or Howard Baker waiting in the wings and with uniquely cowardly Republican Grandees, no improvement is possible.

This is not merely a Trump problem; it is a systemic failure, a sign that the institutions designed to constrain presidential incapacity are themselves in crisis. The addled president is repeatedly outwitted by his own courtiers—each new betrayal a surprise, each new lie a revelation. So here we are, very predictably.

Give a gift subscription

Chaos-monkey wages war:

Sanho Tree <https://bsky.app/profile/sanho.bsky.social/post/3lse2egwju226>: ‘Dear Donald: lithium isn’t just for batteries. Thank you for your attention to this matter!💊

No understanding of the situation that he stuck our B-2s into. Zero:

Patrick Chovanec

Patrick Chovanec: ‘He really is living in a fantasy world enabled by a myriad of eager sycophants:

Leave a comment

And so now we have Donald Trump, with a dawning realization that Binyamin Netanyahu lied to him when he told him “you do one, and then we all can be done”. He is not reacting well:

Andy Carvin: ‘We are truly through the looking glass now:

Tim O’Conner: ‘It hardly needs to be said, but if Biden had so visibly lost the plot and all self-control, CNN would already be melting its way through to the earth’s core: ((Tendar)))‬: “Trump is excusing Iran for the missile strike (“perhaps the missile landed by mistake”) this morning and blaming Israel for the latest escalation. He ends the statement in an emotional way:

Donald Trump: ‘They have violated it, but Israel violated it too…. Israel, as soon as we made the deal, they came out and dropped a load of bombs on the likes of which I had never seen before. The biggest load that we have seen. I’m not happy with Israel. You now, when I say “you have twelve hours”, you don’t go out and in the first hour drop everything you have. So I’m not happy with them. I’m not happy with Iran either. But I’m really not happy with it if Israel’s going out this morning. Because the rocket. That didn't land. That was shot by mistake. That didn’t land. I’m not happy with that…. You know what? We basically have two countries that have been fighting so long and so hard that they don’t know what the fuck they are doing. You understand that?…

And Max Kennerly (I still believe his uncle was innocent!) ha the appropriate reaction:

Max Kennerly: ‘Just one thing Mr. President: was that not apparent to you three days ago when you decided to embroil our country in their dispute?… The most pertinent question, because Trump's reaction here is at least 99% the product of whoever he talked to last, and somebody riled him up about Netanyahu. (Just as somebody else will probably talk him down later today. 🤷‍♂️):

‪Craig Ganzer‬: 'He probably genuinely thought that if he bombed their nuclear sites, Israel would have no reason to keep the war going and he could play peacemaker and win a Nobel for it…

Let me state the obvious: Iis, factually and constitutionally, long past time to invoke the 25th Amendment and remove Donald Trump from office.

The spectacle of a president who, with the regularity of a metronome, parrots the last person who whispered pleasing lies into his ear—only to be surprised, again and again, that these were indeed lies—ought to be enough to convince even the most hardened partisan that we are witnessing cognitive incapacity in real time. This is not a matter of ideological disagreement, nor of policy preference; it is a matter of the basic faculties required to discharge the duties of the presidency.

The American presidency, for all its performative trappings, is not a ceremonial post. He is not King Charles III Windsor.

The American presidency is a node of decision-making in a world where the margin for error is measured not in polling points, but in the lives and livelihoods of millions.

To have at the apex of hte executive branch an individual who cannot distinguish between reality and sycophantic fantasy, who is repeatedly surprised by the consequences of his own credulity, is not merely a constitutional crisis. It is an abdication of the responsibilities of governance. The Founders provided a mechanism: impeachment. It was meant for precisely this scenario, among others. It is time to use it. But the Republican senators will not do that. However, the aftermath of Eisenhower’s heart attacks gave us a kinder, gentler version: the 25th Amendment. But the Republican senators will not do that either.

So here we are.

There is yet a third, stopgap option. The Republican senators—those self-styled guardians of institutional integrity—could walk down Pennsylvania Avenue and tell Donald Trump what is obvious to all: He must find his George Shultz. He must delegate the conduct of foreign policy to an adult, a person capable of making and keeping deals, of distinguishing between the real and the merely expedient, of understanding that the world is not a reality television set populated by easily manipulated extras.

But they have not yet mustered the collective will to do so. I do not expect that they will.

The Reagan administration, in its twilight, stumbled upon the solution: regency, in all but name. Howard Baker and George Shultz ran the show while the Gipper smiled and waved.

It was not ideal, but it was semi-functional.

One might hope for a similar arrangement now, both in foreign and domestic policy—a cadre of grown-ups to manage affairs while the president is wheeled out for ceremonial duties and the occasional photo op.

But—and here is the rub—there are no George Shultzes or Howard Bakers on offer. The Republican Party, having purged itself of seriousness in favor of performative loyalty, has no bench. The institutions that once produced statesmen now churn out cable news personalities, at best. Their modal product is a Twitter provocateurs.

The risks are not hypothetical. The costs are not abstract. We are living, I think, in the long shadow of institutional decay, and the bill is due.

Leave a comment

Subscribe now

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…

LECTURE NOTES: Lash, Cash, & Cotton in the Imperial-Commercial & Early SteamPower Age

& the racialization of slavery. A ten-mintue taste of what my lectures have to say about post-1500 slavery. This then becomes the lecture introduction when I give the longer lecture…

Share

Share Brad DeLong’s Grasping Reality

Why post-1500 did the world’s wealthiest societies outsource brutality? And how did they come to create the idea of race and use it as their rationalization. Let us move from ancient misfortune to modern ideology, tracing slavery’s transformation as merchant and then industrial capitalism rlse. And we will pause for a moment to note the unseen costs of cheap goods: how the middle class became more than complicit but rather the prime mover in Atlantic slavery’s global machinery.

Start here: Whenever and wherever something is valuable, two strategies for acquiring it inevitably emerge. The first is what I might call the “nice economist” approach: resources are used to purchase the desired good or asset, transferring ownership through mutually agreed exchange. This method avoids destruction, minimizes waste, and—at least in theory—leaves all parties better off. It is, as Adam Smith would have it, the invisible hand at its most benevolent.

But there is always a second, darker option: the use of coercion, violence, or outright theft. When a prosperous mercantile or industrial economy sits at the center of a wider system—think of Britain in the 19th century, or Athens in the 5th century BCE—its very demand for resources creates perverse incentives at the periphery. Those on the outside, unable to compete with the core’s purchasing power, may instead turn to force. Thugs with spears—or, later, guns and whips—find profit in seizing people and land, compelling labor to produce goods for export to the wealthy center.

This pattern is ancient. In the 5th century BCE, the Skythian and other horse-lord elites of the Black Sea region enslaved local populations to grow wheat, which was then shipped to feed the citizens of Hellas, especially the city of the Athenai. Fast forward to the 19th century: the industrial economies of Britain and the U.S. North depended on cotton produced in the American South. After the forced “removal’ of Native Americans, the land was abundant but labor was scarce and valuable—hence, the brutal expansion of slavery.

Slavery, in any form, is a moral abomination. Within the household, and on the small farm or in the small workshop, it is moderated by the fact that the slave master has to look the enslaved person in the eye. In some cases, it is not that much worse than the “normal” household domination lineage heads exercise over poor, third cousins who have no independent resources of their own, but even there the fact that the point of buying a slave is that they are an alien without social power, a social network, or the ability to form one, making it sharper and nastier as they are easier to exploit, and to exploit more completely.

As the scale of the slave enterprise grows, and as the household becomes embedded in a merchant-capitalist sector where the slaves work is not limited by what the household can utilize, but is rather unlimited because their work can be sold on the market for money, things get worse. And the human social practice of slavery reaches its most monstrous proportions on large plantations, where enslaved people are forced to produce cash crops for distant markets. Especially absentee plantations. Owners then never need to become within eyesight of what they have set in motion. They never face any of the brutal human cost of their profits. Instead, they simply write terse letters to overseers: “Why are my returns not higher? Why have you not squeezed more wealth from this enterprise?” And what kind of person can flourish in such an intermediate overseeing position, and so becomes an overseer?

The system is then designed for maximum extraction, minimum accountability, and a studied indifference to suffering—a pattern that, I regret to say, recurs over and over throughout human—or, rather, inhuman—economic history.

After 1500, merchant capitalism became not just a sector in an agrarian-age economy, but rather the predominant mode of organization in what rapidly became imperial-commercial society. And so the typical mode of slavery underwent a profound transformation. It became plantation, and thus truly, obscenely nasty. And it became racialized in a way it had not been before.

In earlier eras, slavery was largely a matter of misfortune—being captured in war, falling into debt, or simply being in the wrong place at the wrong time. The enslaved were, for the most part, still recognized as people, albeit people in an unfortunate legal and social position. Yes, Aristotle famously mused that some non-Hellenes were actually “slaves by nature”. This, however, was more a rhetorical flourish than a dominant ideology. Slavery fit into a broader hierarchical order, where your place—slave or free—was simply your lot, not your essence, and where there were many hierarchical gradations in some Great Chain of Being.

However, as agrarian-age societies were followed by less ascriptive, more mobile-contractual societies—ones where individuals were supposed to find their own place through markets and negotiation and the construction of social networks—this old logic of slavery as a low status into which you were fixed by birth or ill-luck became harder to sustain. The world increasingly committed, at least in rhetoric, to individuals’ both getting to choose and having to find their place. The result was the valorization of individual freedom and equality.

So what then could you do with the idea of slavery?

You could eliminate it. And eventually the British Empire led the way to doing so. But it was very, very profitable! It was increasingly profitable as the commercial transoceanic economy gathered strength! Its profitability was amped-up another order of magnitude by the coming of SteamPower society. And so societies doubled down on the idea that rights, freedom, and equality were only for the true, full humans. They began to focus on a belief that some groups were inherently suited to enslavement in order to justify the continued existence of slavery within an Enlightenment-era world of humans with natural rights.

And race became the marker: Africans and their descendants were cast as “slaves by nature,” their status justified not by circumstance but by supposedly immutable characteristics. This was the ideological sleight of hand that underpinned the Atlantic slave trade and the plantation economies of the New World.

It is crucial to remember, though, that the principal beneficiaries of this system were not the plantation owners. The principal beneficiaries were the middle-class consumers at the heart of the industrial and commercial economies. Their cheap sugar, cotton, and tobacco were made possible by the brutal labor of enslaved people on distant plantations. This is the uncomfortable arithmetic of global capitalism: prosperity in one place, purchased at the cost of suffering in another.

Yet, history is not without its ironies.

The American Civil War (1861–1865) saw the states of the United States that did not secede and formed the Union volunteer to pay a staggering price to end slavery: 400,000 young men killed, another 300,000 maimed.

The magnitude of the suffering he had set in motion drove Abe Lincoln mad. As he said in his Second Inaugural Address, it appeared possible—even likely—that the war would continue until:

all the wealth piled by the bondsman’s two hundred and fifty years of unrequited toil shall be sunk, and until every drop of blood drawn with the lash shall be paid by another drawn with the sword…

Give a gift subscription

Yet he had still done a good thing in the eyes of God to have refused to allow the Confederates to secede from the United States, and preserve for a while their institution of slavery. For:

as was said three thousand years ago, so still it must be said ‘the judgments of the Lord are true and righteous altogether’…

Leave a comment

The American Civil War. 400,000 Union young men dead. Another 300,000 maimed. To rescue 5,000,000 American slaves from complete bondage. But, then, because of the substantial failure of Reconstruction, return them to a form of caste serfdom in which they had little property, less voice, and no vote. For a century. And as Claudia Goldin has pointed out, even setting aside this immense human cost, the United States wasted so much treasure on the war that, by her calculations, it could have used it to more than purchase every enslaved person in the United States at peak 1860 market prices, and also provided each freed family with 40 acres of decent farmland, plus a mule or two. Plus 300,000 Confederate young men dead, and 250,000 maimed.

Yet, still, well done by good and faithful servants, in our eyes, and hopefully in the eyes or others wiser than me.

Get 50% off a group subscription

Time to sum up, provocatively:

Slavery was not always about race—until the coöccurence of modern capitalism with Englightenment ideas demanded a new justification for old cruelties. The middle-class comforts of the nineteenth century rested on a foundation of violence and forced labor, rationalized by a newly invented ideology of race. The economic engine of empire pressured the turning of something that had been seen as random misfortune into a natural and systemic racial hierarchy, all in service of cheap cotton and sugar. What linked the Skythian horselords north of the Black Sea in the -400s to the cotton fields of Mississippi in the 1800s? The relentless logic of the plantation and the market linked them. And in the American case, the bill, grimly, came due with bloody justice of a strange and peculiar sort in the Civil War.

Refer a friend

Leave a comment

Subscribe now

If reading this gets you Value Above Replacement, then become a free subscriber to this newsletter. And forward it! And if your VAR from this newsletter is in the three digits or more each year, please become a paid subscriber! I am trying to make you readers—and myself—smarter. Please tell me if I succeed, or how I fail…

Who Were the ‘We the People’ Back in 1787?: Race, Law, & Original Constitutional Sin

Plus the Reconstruction-Era XIV-Amendment redemption. In which I go down a rabbithole of “original public meaning” or whatever. A law student’s white nationalist screed wins a prize, echoing arguments once enshrined by the Supreme Court itself—but not since 1857, and definitely not at all since the XIV Amendment’s 1868 ratification. But were the Founders really such exclusionary white supremacists? And what do we do with Ben Franklin’s prejudices against the swarthy Germans immigrating into his beloved Pennsylvania?

Share

I have a section from my American Economic History lectures this spring on the post-1500 Atlantic-economy racialization of early modern slavery that I want to get into good enough shape to post here. It is not yet there.

And this morning a correspondent who does not especially wish me well this morning with respect to my mental harmony sends something across my screen. He sends me a piece from the New York Times that is relevant to that item on my to-do list, And so I go down a rabbit hole.

The piece:

Richard Fausset: A White Nationalist Wrote a Law School Paper Promoting Racist Views. It Won Him an Award <https://www.nytimes.com/2025/06/21/us/white-supremacist-university-of-florida-paper.html>: ‘The University of Florida student won an academic honor after he argued in a paper that the Constitution applies only to white people. From there, the situation spiraled: Preston Damsky… law student… white nationalist… antisemite. Last fall, he took a seminar taught by a federal judge on “originalism”… argued that the framers had intended for the phrase “We the People,” in the Constitution’s preamble, to refer exclusively to white people. From there, he argued for the removal of voting rights protections for nonwhites, and for the issuance of shoot-to-kill orders against “criminal infiltrators at the border.” Turning over the country to “a nonwhite majority,” Mr. Damsky wrote, would constitute a “terrible crime.”… Mr. Damsky, 29, was given the “book award,” which designated him as the best student in the class. According to the syllabus, the capstone counted the most toward final grades….

The Trump-nominated judge who taught the class, John L. Badalamenti, declined to comment…. The granting of the award set off months of turmoil on the law school campus…. Mr. Damsky’s paper includes… a call to “reconsider” birthright citizenship, and an assertion that “aliens remain second-class persons under the Constitution.” It also argues that courts should challenge the constitutionality of the 14th Amendment, which ensures birthright citizenship, due process and equal protection under the law, and the 15th Amendment, which protects the right to vote for nonwhite citizens.

Mr. Damsky concluded the paper by raising the specter of revolutionary action if the steps he recommended toward forging a white ethno-state were not taken. “The People cannot be expected to meekly swallow this demographic assault on their sovereignty,” he wrote, adding that if the courts did not act to ensure a white country, the matter would be decided “not by the careful balance of Justitia’s scales, but by the gruesome slashing of her sword”…

Share Brad DeLong’s Grasping Reality

The New York Times story does not tell us that Damsky has been banned from the campus for various of his other actions, “including a message calling for the elimination of Jews ‘by any means necessary’”. Thus the piece is, I think, not quite fair with respect to the law school administration, which I see as more likely than not doing a pretty good job in handling the situation in its full gestalt, both with respect to maintaining itself as a community and using this troll’s appearance as a teaching moment.

And I do confess that I do not understand how a paper that argues that courts “should challenge the constitutionality of the 14th Amendment, which ensures birthright citizenship, due process and equal protection under the law, and the 15th Amendment, which protects the right to vote for nonwhite citizens”. How can a duly-ratified constitutional amendment be argued unconstitutional? That claim that the XIV Amendment to the Constitution is un-Constitutional seems, to me at least, the leading candidate for MOST JUST WRONG. I cannot see how a paper that does that can be any good. This is especially the case since Damsky appears to draw very strong conclusions for the state of the law today that are totally consistent with the post-XIV Amendment Constitution.

That said, the argument that the original public meaning of “we the people” in the U.S. Constitution was “we the [white] people” seems to me to be intellectually stronger than much of what I see winning majorities in the Roberts court these days.

Leave a comment

Read more