Philosophy & Critical Thinking: Definition forcing

What you'll be able to do: Force precise definitions of the terms doing the work in a philosophical dispute, and tell whether a disagreement is about the world, substantive, or about the word, verbal.

The squirrel that ended the argument

William James opens the second lecture of Pragmatism with a story from a camping trip. A squirrel clings to one side of a tree trunk; a man walks around the tree trying to see it, and the squirrel moves around the trunk at the same rate, always keeping the trunk between them, so the man never sees it, though he has gone all the way round the tree. Does the man go round the squirrel?

James's campers argued for an hour. James's answer: it depends what you mean by "going round." If you mean passing from north to south to east to west of the squirrel, the man does go round it. If you mean being successively on the squirrel's north, south, east and west side, he does not, because the squirrel's compensating movements keep the same side facing him throughout. Once "going round" is defined, the dispute evaporates. No fact about the squirrel was ever in question. That's a verbal dispute, and the test for finding one is exactly what James did: taboo the contested word, restate each side's claim without it, and see whether the disagreement survives.

A harder case: is it always verbal?

Free will is where the test gets interesting, because philosophers disagree about whether applying the test settles anything. Compatibilists (Hume, Harry Frankfurt) and incompatibilists disagree about whether "free" requires the strong, contra-causal ability to have done otherwise given exactly the same past and laws, or whether it requires acting from your own reasons and desires without external compulsion. Restate each side without the word "free" and a lot of the apparent disagreement looks like two groups arguing over which of two real things deserves a label.

But hard incompatibilists such as Derk Pereboom resist that diagnosis. Their claim is that the dispute is substantive, not semantic bookkeeping, because real practices, punishment, praise, blame, desert, genuinely depend on which notion of freedom is correct, not merely on which word wins. Bring this case into a session as the harder test: the taboo-the-word move doesn't always settle the question of whether it was verbal. Sometimes philosophers disagree about that too, and noticing that second-order disagreement is itself progress.

Worked example: the open question

G. E. Moore's open-question argument (Principia Ethica, 1903) is definition forcing built into metaethics itself. Take any proposed natural-property definition of "good" , say, "good = pleasant." If the definition were correct by meaning, the question "is the pleasant good?" would be as closed and trivial as "is a bachelor unmarried?" It isn't, the question stays meaningfully open for a competent speaker of the language. Moore's diagnosis: no naturalistic definition of "good" survives this test, which is why he calls the attempt the naturalistic fallacy. Run the same test on any proposed definition before you accept it: does the question "is X really Y?" stay open once you substitute the definition for Y? If yes, the definition hasn't captured the concept.

This connects directly to Hume's is-ought passage (Treatise, 3.1.1): moral writers slide from "is" premises to an "ought" conclusion without explaining "how this new relation can be a deduction from others, which are entirely different from it." Forcing a definition of "good" before evaluating any argument that moves from facts to values is the practical version of taking Hume's observation seriously.

How to run it

1. Identify the term carrying the disagreement's weight.
2. Ask each side's claim to be restated without that term at all.
3. Check whether the disagreement survives translation.
4. If it dissolves, name the two different things each side meant, and stop arguing about the word.
5. If it survives, you've found the real, substantive crux, now argue that.
6. Run Moore's open-question test on any proposed definition before accepting it.

Here's a dispute: [position A] versus [position B], both using the term "[term]". Restate each side's actual claim without using that word at all. If the disagreement survives being stated that way, tell me what it's actually about. If it disappears, tell me that too, and name the two different things each side meant by the word.

The tell

You can state your position fluently using the contested word, but can't restate it without the word at all. If you can't, you haven't located the claim you're making, you've located the label you're attached to.

Try this today

Take an argument you're currently having, in an essay or in life, where the same word keeps doing different work for each side. Restate both positions without that word before you write another sentence defending either one.