What an Argument Does
The moves of a dispute, with an example from Hume
A skill everyone praises
We spend a good part of our days giving reasons and receiving them. We ask a colleague to move a deadline, a child to put the phone away, ourselves to put off a purchase; in each case someone offers reasons and someone else weighs them. Not everything we say is argument, of course. Greeting a neighbor is not, and neither is asking the time. Still, a fair share of what we end up believing, doing, wanting and hoping for passes through exchanges of this kind.
One would conclude that the ability to argue well holds a high place in education. In curriculum guidelines it does; in classroom time it does much less. It is one of those skills that everyone calls important and that few think they need to teach, or to learn properly, perhaps because we assume that anyone who can talk can also reason. The assumption is convenient. Any condo board meeting is enough to cast doubt on it.
In recent years debate has brought argument back to the center of attention, and that is to its credit. But it tests a mastery that is formed elsewhere, the way a match tests a player who has trained beforehand. During a debate there is time to answer, but rarely time to understand why one answer worked and another did not. Someone who learns to play only by playing matches learns something; usually he also learns to repeat the same mistakes with greater confidence. At most, debate pushes us to take care of our ability to argue. It remains to say what that care consists of.
What an argument does
We usually look at an argument by asking whether it is good: whether the premises are true and whether the conclusion follows from them. The question is the right one and should always be asked. In a dispute, however, every argument is also an action. It serves a limited purpose: to prove a claim, to bring down the other side’s, to narrow what one had asserted, to shift the burden of proof onto the opponent. These actions combine toward an overall purpose, which is usually to persuade someone, or at least not to be persuaded by him. I will call these actions moves.
The moves are not infinite. This is not obvious, since the sentences we argue with are countless. The things we do with those sentences, on the other hand, are few, and they keep coming back. We can build our own position; we can attack someone else’s; or we can reposition ourselves, conceding a point, drawing a distinction, clarifying what we meant. Each of these families admits many variants, but the variants can be counted and classified.
Chaïm Perelman, who studied argumentation in the 1950s together with Lucie Olbrechts-Tyteca, observed that his schemes are not mere logical forms. An a fortiori argument, an analogy, an appeal to consistency each have a structure and, along with it, an intended effect on an audience. That effect depends on who uses the scheme and on the point in the dispute at which it is used. It follows that the same scheme can serve different moves: an analogy can support a claim or make the opponent’s claim look ridiculous. To understand an argument, then, it is not enough to recognize its form. We have to ask what the person using it is trying to achieve, and at that precise moment of the dispute.
The spectator and the player
Someone who knows chess and watches a game sees things a beginner does not. He recognizes an opening, understands why a pawn was sacrificed, anticipates the reply before it comes. At the end he can say who played better, even when the better player lost. The same holds for disputes. The expert reader recognizes the moves, senses the ones to come and, once the exchange is over, can judge it. This is the skill we call critical reading.
Playing is another matter. The player has no time to contemplate the game. He has to grasp what the other side is trying to do and answer fittingly, preferably at the point where the opposing position is weakest. The spectator can afford to be right afterward; the player has to find the answer now. The two skills are connected, because no one answers a move well without first recognizing it. They remain distinct all the same. A school that only teaches students to read other people’s disputes produces spectators, and it would do well not to believe it is producing players too.
The simplest way to see the difference is to watch a game played well. I take one from a book that few people would open to learn how to argue.
An eighteenth-century match
In David Hume’s Dialogues concerning Natural Religion, published posthumously in 1779, three characters argue. Cleanthes believes that the order of the world proves the existence of a designing god; Demea is an orthodox believer; Philo is a skeptic. In Part I they are talking, on the surface, about how to educate the young. Demea holds that the weakness of human reason should be taught early, so that young people come to religion already distrustful of their own reasoning. Philo seizes the opportunity.
Get Pietro Alotto’s stories in your inbox
Join Medium for free to get updates from this writer.
First move. Philo builds an a fortiori argument. If reason stumbles over everyday questions, all the more will it stumble when it claims to settle the origin of the universe. The move also has a side effect: the skeptic finds himself allied with the orthodox believer, and Cleanthes is left alone to defend the powers of reason.
The reader can stop here and ask what he would do in Cleanthes’ place. He could challenge the premise and show that reason is not so weak; that would be long and uncertain work. He could challenge the step from small to great. Cleanthes takes a third road.
Second move. Cleanthes leaves the argument alone and looks at the man who holds it. Total skepticism, he observes, cannot be lived: “we shall learn by and by, when the company breaks up: we shall then see, whether you go out at the door or the window.” This is an attack that turns the tables, what rhetoricians call retorsion: it sets the opponent against his own conduct. The joke adds ridicule, which speaks to the onlookers more than to Philo. Cleanthes has shown only that the skeptic does not practice his doubt all the way; about whether reason can reach God he has said nothing.
Here too it pays to stop. Philo has three possible answers. He can defend total skepticism and accept the role of the philosopher who pretends to doubt the door. He can give up everything. Or he can give up something to save the rest.
Third move. Philo concedes and draws a distinction. He readily admits that in common life the skeptic acts and reasons like everyone else, because he cannot do otherwise and because experience keeps correcting his mistakes. In theology that correction is missing. There, he says, “we are like foreigners in a strange country, to whom every thing must seem suspicious.” Philo’s claim has narrowed, and for that very reason Cleanthes’ retorsion no longer finds its target. This is a repositioning: Philo changes the boundaries of what he holds, without adding arguments. An inexperienced spectator may not notice that the game has changed.
Fourth move. Cleanthes attacks the distinction. Skeptics, he says, accept without difficulty Newton’s explanation of the rainbow and the Copernican system, both far removed from common experience. If they accept these, their rejection of theology reveals a doubt applied at their own discretion. Perelman would call this an appeal to the rule of justice: similar cases must be treated alike. If Newton yes and God no, it is up to Philo to say what tells them apart.
Fifth move. Cleanthes then shifts the dispute onto historical ground. Religion and skepticism, he notes, have joined forces and parted ways more than once, and Locke was one of the first Christians to assert openly that faith is a form of reason. Philo accepts the story and changes its meaning. Those alliances, he replies, are explained by the convenience of the moment more than by principle. The historical account, meant to put the skeptic in difficulty, becomes one more example of reason bent to interest.
In five moves, then, all three families appear. Philo builds; Cleanthes attacks twice; Philo repositions himself and finally turns the historical attack against the one who launched it.
Who won
An attentive spectator can now try to reach a verdict. Cleanthes scored a real point: total skepticism cannot be practiced, and whoever professes it drops it as soon as he gets up from his chair. Philo, however, by drawing his distinction, kept the only ground that mattered, namely whether reason can reach God. Cleanthes therefore won the exchange on a point that, by the end, no one was contesting anymore. Whoever leaves the room remembering only the joke about the window thinks Cleanthes won; whoever followed the moves knows the question is still open, and that it stayed open thanks to Philo.
The example shows something that lovers of truth tend to forget. Winning a dispute settles who got the better of the opponent; it does not settle who was right. Schopenhauer, in The Art of Being Right, separated precisely the objective truth of a claim from the art of making it prevail. One can draw a cynical conclusion from this separation, namely that only winning counts. One can also draw a more modest and more useful one. We can be certain we stand on the side of truth and still find ourselves needing to win, because a truth badly defended loses just as a falsehood does, and the audience goes home convinced of the opposite.
That is what knowing the moves is for. It lets the spectator avoid mistaking a clever line for a demonstration. It lets the player avoid forfeiting a game in which he was right. Measured against the ambitions of philosophy, these are small advantages; measured against the disputes we have every day, they are considerable.
Moves are learned the way one learns to read a game: by watching many of them, by putting a name to what one sees, and then by playing against someone who concedes nothing. How to organize that training, and what AI can do to make it repeatable, is something I will come back to in future issues.
To anyone who wants to try, I suggest rereading Part I of the Dialogues with a single question in mind, repeated at every turn: what is this man trying to do, right now? It is a simple question. Few people ask it, and almost no one teaches it.
References
Hume, D. (1779). Dialogues concerning Natural Religion. London.
Perelman, Ch., and Olbrechts-Tyteca, L. (1969). The New Rhetoric: A Treatise on Argumentation. Trans. J. Wilkinson and P. Weaver. Notre Dame, IN: University of Notre Dame Press. Original: Traité de l’argumentation. La nouvelle rhétorique, Paris, PUF, 1958.
Schopenhauer, A. The Art of Being Right. Trans. T. Bailey Saunders. London, 1896. Original: Eristische Dialektik, published posthumously in 1864.
