After reading a draft of Selfish Reasons to Have More Kids, Tyler Cowen bemusedly told me, “You have more enthusiasm for your own arguments than you do for the children themselves.”  A slight exaggeration, but I take it as a compliment.  I do indeed love a good argument – and every time I father a hearty new argument, I swell with pride.

What makes a good argument?

First, a good argument begins with premises that many people find plausible even if they disagree with your conclusion.  The less controversial and complex your premises, the better.  This doesn’t mean, of course, that a good argument has to convince everyone.  No argument does.  But every good argument reaches out to minds not-yet-convinced.

Second, a good argument carefully reasons from these initial premises to your conclusion.  The conclusion need not be certain; if your premises are merely probable, you can hardly expect to reach a conclusion free of doubt.  But a probable conclusion is still a conclusion.

Third, a good argument must have a conclusion more obvious than the denial of your initial premises.  You can’t argue that 1+1=3.  After all, any argument of the form “A, therefore B” is also an argument that “Not B, therefore not A.”  Corollary: If your conclusion is totally obvious, you shouldn’t argue on its behalf.  To do so wastes everyone’s time. 

Does anyone seriously dispute these principles?  Yes.  The staunchest opponents of argument, strangely, tend to be intellectuals.  Their main argument against argument is that “Argument doesn’t work.”  Since people are stubborn and dogmatic, even the best arguments fail to change people’s minds.

My reply is just, “Compared to what?”  Arguments clearly change some people’s minds.  The people most susceptible to argument are usually those we independently perceive as reasonable and fair-minded.  Why aren’t these facts sufficient to vindicate the practice of argument? 

The worst enemies of argument, though, are free-riders rather than critics – the people who invoke argument without applying it.  How often have you heard someone use the phrase, “I would argue,” then state their conclusion, then never actually argue for it?  For me, the answer is several times per week.

We all have time constraints.  Sometimes you just want to fast forward to your conclusion before you leave a party.  But if you habitually fail to argue for the claims you “would argue for,” you give the art of argument a bad name. 

Tomorrow, October 30, I’m debating “Let Anyone Take a Job Anywhere” for Intelligence Squared in New York City.  I will present the best arguments I know how, subject to a rather brutal time constraint.  The audience will judge how well I meet my own standards.  Hope to see you there.

Update: I initially wrote, “Third, a good argument must have a conclusion less obvious than the denial of your initial premises.”  I misspoke.  If your premises are only moderately likely to be true (say 80%), and your conclusion is extremely unlikely (say 1-in-a-billion), it makes more sense to deny the premises (20% chance of truth) than embrace the conclusion (1-in-a-billion chance of truth).  Thanks to David Gordon for pointing out my misstatement.