The Welch Review — Notes https://thewelchreview.pages.dev/notes Short entries, links and marginalia. en Wed, 22 Jul 2026 00:00:00 GMT Hirschman's three arguments https://thewelchreview.pages.dev/notes/hirschman-on-perversity https://thewelchreview.pages.dev/notes/hirschman-on-perversity Wed, 22 Jul 2026 00:00:00 GMT Rereading The Rhetoric of Reaction and struck again by how cleanly the taxonomy holds up. Hirschman claims that arguments against reform almost always take one of three forms:

  • Perversity — the reform will produce the opposite of what it intends.
  • Futility — the reform will change nothing, because the underlying structure is untouched.
  • Jeopardy — the reform endangers some earlier achievement worth more.

What makes it useful isn't that these arguments are bad. Each one is sometimes correct, and Hirschman is careful about saying so. What's useful is that they're available — they can be produced for any proposal whatsoever, without knowing anything about it. That's the tell. An argument you could have made before hearing the proposal isn't evidence about the proposal.

He wrote a companion list of the progressive equivalents late in life, which is less quoted and, I think, sharper about his own side.

]]> Note — Wed Jul 15 2026 https://thewelchreview.pages.dev/notes/on-being-wrong-in-public https://thewelchreview.pages.dev/notes/on-being-wrong-in-public Wed, 15 Jul 2026 00:00:00 GMT A reader points out that the legitimacy essay treats "procedure" as though it were one thing, when the proceduralist tradition contains at least two quite different views — one about fair process, one about public justification. They're right, and the essay is weaker for the elision. I've added a correction rather than quietly editing the paragraph.

Worth saying that this is the whole reason to publish corrections visibly. The alternative is an archive that silently becomes more correct over time, which reads as though you were never wrong.

]]> There ought to be a word for it https://thewelchreview.pages.dev/notes/a-word-for-it https://thewelchreview.pages.dev/notes/a-word-for-it Thu, 09 Jul 2026 00:00:00 GMT There's no good English word for the specific error of treating a decision that was never put to anyone as though it had been ratified. "Naturalising" is close but carries too much theoretical freight. "Path dependence" describes the mechanism, not the mistake of reading the mechanism as consent.

I keep needing the word and keep having to write the sentence out.

]]> The Liberalism of Fear https://plato.stanford.edu/entries/shklar/ https://thewelchreview.pages.dev/notes/shklar-liberalism-of-fear Tue, 30 Jun 2026 00:00:00 GMT Shklar's essay puts cruelty first — not liberty, not equality, not justice. The argument is that we can disagree endlessly about what the good life requires while agreeing readily about what we most want not to happen to us, and that a political order built on the second is sturdier than one built on the first.

The part I keep returning to: she insists this isn't pessimism but a claim about where agreement is actually available. That's a methodological point disguised as a substantive one, and it applies well beyond her subject.

Permalink

]]>