Sheepishly

Our nearest neighbour in the Lake District, photographed shortly before we left for home. Herdwicks (which are native to the Lakes) are such beautiful animals.
Quote of the Day
”It is discouraging how many people are shocked by honesty and how few by deceit.”
Noël Coward
Musical alternative to the morning’s radio news
Lord Inchquin
One of my favourite Irish airs.
Long Read of the Day
When Decentralization Fails
This essay by Alex Chalmers is a long, long read, but in a way that’s appropriate because it addresses one of the biggest problems that societies will have to face (assuming we survive climate catastrophe), namely how to govern two kinds of technologies that we have recently invented — AI and (although Chalmers doesn’t address it directly) synthetic biology.
In a way the only comparable problem the world has had to address — at least in my lifetime — was the challenge of controlling nuclear weapons, and to date we have just-about managed that (though there have been a few close calls).
These newer technologies are different in one important respect, though. Whereas nukes were the sole prerogative of nation states, AI and synthetic biology are the property of giant corporations. And the bi-polar world in which the nuclear threat evolved (and was to some extent resolved) is fragmenting into a potentially chaotic, multipolar one.
Chalmers ends on an interesting note. Most governance questions are deployment problems where domain-specific knowledge is generated through practice — oncologists, not central regulators, hold the relevant knowledge for AI in cancer medicine, say. Existing institutions (professional bodies, universities, malpractice insurers, procurement officers, open-source communities) can govern within their domains, with state or interstate coordination reserved for the genuinely cross-cutting issues like frontier model weights.
This won’t fix production-side concentration of corporate power, but it can build countervailing power on the deployment side — so that AI giants have to satisfy many authorities rather than capturing one regulator. What happens at the deployment level of these technologies is where humans — and democracies — still have critical leverage and power, if they choose to exercise it.
I hope you can find time for it.
My commonplace booklet
Stockholm syndrome in Mar-a-Lago: The belief that “something must be done” and the sanewashing of economic policy in the age of Trump
Historian Adam Tooze, after reprising arguments that were largely incomprehensible to me, provides this helpful summary at the end:
None of us really knows where this clown car is headed and what drives it on its crazy course. It seems like a mystery even to many on board. Quite reasonably we look for elements of rationality. We ask: who inside MAGA 2.0 is thinking and what are their thoughts? We then relate that to our own efforts to diagnose America’s history and the history of the world economy. At the very least we need to explain how Trump 2.0 happened. Sometimes we will find a match between a strand of policy from inside MAGA and our own analysis and it is tempting to label that as “MAGA for thinking people” and to look for continuities with the Biden team etc. That mode of analysis is reasonable. To historically minded people it is appealing for obvious reasons. But it puts us at risk of is underestimating the radicalism of the break marked by the Trump administration. In search of historical context we miss what is most historically significant. We avoid facing the conclusion that the vision of a Mar-a-Lago Accord may have more in common with grift, a protection racket or a facelift pandering to the ignorant vanity of an old man than with economic policy as we have hitherto known it. Faced with Trump, the risk is that conventional realism is a form of escapism.
Which, oddly enough, is what I’ve felt since Trump took office.
Linkblog
To see or not to see? Every single Shakespeare play – ranked!
This is truly wonderful: the great theatre critic Benedict Nightingale’s analysis of all Shakespeare’s plays, in order of his preferences.
Many thanks to Christine Happel for alerting me to it.
Feedback
Lots of interesting emails arrived about the problems of being famous. Reading them made me remember Nuala Ó’Faolain, who was a great — and very well-known — Irish feminist when I was a student. A publisher asked her to write an autobiographical introduction to a collection of her newspaper columns, and she was stuck for a title. Then one day, when queueing at a supermarket checkout, a lady came up to her, looked at her closely, and asked “Are you somebody?” And Bingo, she had her title! And a world bestseller.
This Blog is also available as an email three days a week. If you think that might suit you better, why not subscribe? One email on Mondays, Wednesdays and Fridays delivered to your inbox at 5am UK time. It’s free, and you can always unsubscribe if you conclude your inbox is full enough already!