As I’ve been (vaguely) following the Kagan hearings, I’ve come to a somewhat surprising conclusion: I think the sort of person for whom “intelligence” is a matter of being cute with language and/or logic is not just unhelpful, but actively destructive.
Software is a collection of symbols that, when manipulated by mechanical processes, should direct those processes to predictable and beneficial outcomes. It’s really hard to get right, even though it benefits from unambiguous languages, constrained problem domains, and predictable environments.
Law (or any equivalent system of rules) is like software written in an ambiguous language for a vaguely defined problem running on unspecified and unreliable hardware. It’s amazing that it works at all.
In my view, “cute” arguments are nothing more than bugs; they represent the introduction of strange conditions unforeseen by the designer of whatever logical context you’re working in. They’re easy to come up with, and, in a sense, not to be taken seriously.
Unfortunately, when cute arguments become an important tool for getting your way in society, two problems arise. First of all, these sorts of language games erode the foundation of society. They make it impossible to tell quite what the laws are (since laws are large, complex, and subject to the double hazards of cute drafting and cute interpretation). They also make it very hard to strike a bargain, since any complex contract is subject to those same hazards. (See, for instance, Hollywood accounting.)
Secondly, a society that places too much emphasis on the playing of logic games as a way to achieve one’s ends leaves itself unprepared to deal with circumstances and adversaries unimpressed by logic games. Clever arguments only have leverage inside a system that gives them force; they are quite useless in the world of blood and iron.
Consider the Deepwater Horizon for an illustration of the uselessness of clever talk.