Don't put a weak argument before a strong one (usually).

Lawyers are often told to put their best arguments first in a brief, their weakest arguments in the middle, and the second best argument last (called "sandwiching"). I think this conventional wisdom is wrong.

In a recent LinkedIn post, Chad Colton discusses the concepts of "primacy" and "recency" with respect to juries. Put simply, jurors are more likely to remember the first and last thing that they hear. For this reason, the sandwiching approach makes sense for juries.

But why do we think it makes sense for judges? Jurors are in an environment they've never been before, facing a torrential data dump. Because of this, primacy and recency are incredibly powerful psychological tools. But judges and clerks are sitting in chambers, cup of coffee in hand, unpacking and analyzing each argument in a brief. There's no reason to assume that the same strategies will work with them.

When I was reading briefs as a clerk, a weak argument was an instant hit to a lawyer's credibility. Upon encountering a weak argument, I viewed the rest of the arguments more skeptically. So, if a brief has three arguments, putting the weak argument second made me more skeptical when I encountered the stronger third argument.

In contrast, when a lawyer put their weak argument last, I implicitly understood that this argument was more of a Hail Mary or fallback position. It didn't make me go back and treat their initial argument with skepticism.

This isn't to say that the sandwich method never has its uses. One situation where it can be helpful is when you’re responding to an argument, rather than making affirmative arguments. For example, if the opposing party has moved to dismiss on three bases, and you lose on any of them, it can be useful not to end on your weakest point. This is where sandwiching can come in handy. But I would caution against reflexively sandwiching simply because it’s the conventional wisdom. In many cases, you’re better off putting arguments in order of strength.

(This is a longer version of a post I recently made on LinkedIn.)