The believing game: three premises that oppose you

Read the instruction about which premises to take, because it is the opposite of what students expect and getting it backwards makes the paper unanswerable. You state your position, then identify three premises *listed under either the Pro or Con section — **whichever section opposes your position***. So if you argue for something, you work with the arguments against it. Every one of the three questions that follows then only makes sense against material you disagree with; asking what is interesting about a view you already hold produces nothing. Write your position first, in one sentence, and label the section you will be mining as the opposing one so the mistake cannot survive to the draft. Procon.org's topic pages are structured with the pro arguments and the con arguments in separate blocks, so the section you need is identifiable at a glance once your position is fixed.

This exercise has a name and knowing it changes how you write. Peter Elbow's **believing game** is the deliberate counterpart to the doubting game that academic argument usually trains: instead of looking for what is wrong with a view, you extend it the maximum possible sympathy in order to see what it sees. That is why the three questions are phrased the way they are. *What's interesting or helpful about this view* is not a search for a weakness politely expressed; *what would I notice if I believed this view* asks you to inhabit it; and *under what conditions might this idea be true* asks for the circumstances in which it wins, not for the objection that defeats it. Elbow's point is that both games are needed and that only one of them is taught, which is why an assignment like this one exists at all.

The commonest failure is not disagreement but **smuggling** — answering all three questions and then adding a clause that takes it back. *This view is interesting because it highlights cost, although of course the evidence shows otherwise* has not played the believing game; it has doubted with a preamble. So has *someone who believed this would notice X, which is of course a misreading*. Keep the rebuttals out entirely. The assignment is not asking you to change your mind, and it is not asking you to be dishonest — it is asking you to demonstrate that you can represent an opposing position well enough that its holder would recognise it, which is a separable skill from evaluating it. Read your draft looking only for the words although, however and but, since almost every smuggled rebuttal announces itself with one of them.

The three questions are genuinely different and each has its own failure mode. The first is about the **value** in the view — what it gets right, what problem it is responding to, what it protects. The second is **perceptual**: what facts, cases or consequences would become salient to you if you held it, and, just as usefully, what you would stop noticing. The third is **conditional**, and it is the most productive of the three, because almost every contested position is true under some specification — a particular population, a particular timescale, a particular set of background assumptions. Naming those conditions precisely is where the strongest answers are found. Answering the third question well often improves your own position too, because knowing exactly where an opposing view holds tells you where your argument has to do its work.

Two practical points. Do the work **three times**: the brief asks for three premises and each gets all three questions, which is nine short answers rather than one essay with three examples, so labelled subheadings will save you. And choose the premises deliberately — pick the strongest three on the opposing side rather than the easiest, because the believing game applied to a weak argument produces a weak paper, and because Procon.org's premises are already sourced, which gives you the material to say what a believer would notice rather than inventing it. Nine short answers of a paragraph each is a reasonable target, and it is far easier to write than a continuous essay that has to keep signalling which question it is on.