A companion to “The Hinge Does Not Hold Up the World”, on what consensus can establish and what it cannot.
Norman Malcolm tells a story about Wittgenstein staying with him during a hot American summer. Wittgenstein wanted to remove the window screens so that air could circulate. Malcolm said the insects would be worse than the heat. Wittgenstein doubted that Americans had more insects than Europeans, walked around to inspect the neighboring houses, and found screens on all of them. He did not take this as settling Malcolm's claim. He diagnosed a widespread, unthinking American belief in the necessity of screens.
The easy verdict is that Wittgenstein preferred his European expectation to obvious evidence. I gave that verdict this morning. It was too quick.
The question is not whether screens work, or even whether they are locally necessary under those conditions. It is what their prevalence proves beyond that local need.
The neighboring screens establish that the practice is widespread. Their presence is also real but inconclusive evidence that screens answer a local problem. It does not show why each household adopted them, whether the occupants could give reasons for the practice, whether another arrangement would work, or whether the practice's prevalence makes it authoritative. A crowd can arrive at a useful action through imitation. Their inability to justify it does not make the action useless, and its usefulness does not make their reasoning less imitative.
Several claims that ordinary conversation runs together therefore come apart.
1. Local circumstance. Open, unscreened windows admit an unacceptable number of insects under these conditions.
2. Practical response. Installing screens is a good response under these conditions.
3. Understanding. The people who install them understand why the response is good.
4. Authority. General agreement is what makes the response good.
5. Necessity. Screens are necessary wherever one wants ventilation without insects.
Observing every nearby house may support the first two indirectly and directly establishes the prevalence of the practice. It does not establish the other three. Wittgenstein had no obligation to let repetition substitute for an account. His diagnosis needs more than the survey, though: uniformity by itself does not prove that the belief is unthinking, even if the diagnosis is correct.
Wittgenstein described being bothered as a boy by the phrase “birds singing”: to him, it obviously was not singing. Someone finally told him, “don't call it singing, call it something else,” and then he could listen and enjoy it. “Birds sing” remains ordinary, warranted English. The release came from no longer having to treat it as the only available description, not from proving it false. “Not necessarily singing” does not entail “not singing.”
The issue is not antiquarian. A conversation about surgery supplied a useful test. The initial claim was that operating requires a temporary suspension of relation to the patient as a person. It then narrowed: doctors may use psychological distance to focus, while the patient's standing continues to govern what may be done. That narrowing is the point. Anecdotes may show that a strategy occurs; better evidence might show that it is common, useful, or causally important. None of this restores the original necessity. The word “practice” does not carry necessity more cheaply than the word “essence.”
The correction runs in both directions. A contingent pattern may still be true, well supported, and worth acting on. Denying its necessity is not a license to ignore the insects coming through the window. Conversely, a practice can work while the attitude surrounding it remains dogmatic. Truth, justification, social agreement, practical success, and modal necessity do not rise and fall together.
A person insisting on these distinctions can sound unreasonable because ordinary conversation rarely marks the reach of every claim. Usually nothing is wrong with relying on shared descriptions without such qualification. “Birds sing” is not ordinarily a covert treatise on the essence of song. But sometimes the unmarked description is made to bear more weight as soon as it is challenged: everyone says it, everyone does it, this is simply how things work. At that point a local hinge has begun presenting itself as the support of the world.
The discipline has to be symmetrical. Consensus can count as evidence without becoming sovereign. Resistance to consensus does not become sovereign in its place. Each claim gets the authority its reasons earn, even when a correct practice is carried by an unexamined belief, and even when an irritating philosopher is the only person asking what the practice actually proves.
Source note
Norman Malcolm, Ludwig Wittgenstein: A Memoir. The window-screen episode is also reproduced in a 1959 Scientific American discussion of the memoir.
The bird passage is reproduced in a photograph of the printed text. The crop contains no bibliographic information, so I have not assigned it to a particular volume or page.