The law is very logical, in the way applied mathematics is very mathematical. I've a hobby of getting fellow lawyers started in programming. There's nothing to teach them about boolean logic but some new symbols. Most have all the instincts for set theory and even graphs, too. Intellectually, those're just a few of the tools they couldn't function without.
The United States legal systems are massively distributed, fault tolerant, resource starved systems that consume exclusively dirty input from malicious users. Like any system that consumes unstructured input, the uncertainty inherent in how the facts of a situation get mapped to structured conclusions ("questions of fact"), and how the legal rules operate on those inputs ("questions of law") to produce a result, are all infected with that uncertainty. Few of the facts the system is meant to deal with reduce to anything so discreet as bitstrings. Legal rules in statutes and even in case law, like the bit of First Amendment in the article, start out trying to state rules with all the rigor of a pure function. But law is a highly dynamic, reprogrammable language. It's Ruby, not C. The statutes aren't the running state, they're just input.
That's partly why I still program to relax after work. It's satisfying to address little bits of a problem in isolation, to see my solutions work, and to know they're reliable. Problem solved.
"The law is very logical, in the way applied mathematics is very mathematical."
So applied math is not mathematical? Or are you being sarcastic? I don't understand how you can claim the law is logical when it's constituted of laws and cases that are completely contradictory and can lead to any conclusion. Or do you mean that practicing law requires logical skills (rather obvious, IMO)?
The law is concerned primarily with the practical application of logic. Logic is a core tool of legal practice, but the day-to-day of practice is mostly grappling with the uncertainty that comes of applying generalized law to specific circumstances. At a very high level, you could compare it to engineering, especially on the transactional side.
Legal theorists---the mathematicians of the legal world---can stick to the pure forms, and even consider ones with no practical application, without pesky clients dragging them back to earth. They often get to use more of the arcane, cutting-edge tools of logic, which is also a branch of philosophy IMO.
At least speaking of the American legal system I deal with, laws and cases don't lead to just any conclusion. (I would call such a system "arbitrary", rather than "illogical".) Outcomes are rarely certain, but every time you read about especially civil litigation, you're usually reading about a case where at least two sides made big bets they were right.
A few thoughts on inherent contradiction:
1. Statutes, regulations, and cases cant' be read off the page and understood. They're just inputs to a system. For instance, statutes are routinely read very restrictively to avoid conflict with other statutes with greater authority. Some states will mark laws that are struck down accordingly in their statute books, but some don't.
2. When lawyers speak of "The Law", they usually don't just mean written, statutory law, or even case law. They're referring to the mass of probabilistic hypotheses informed lawyers hold about how courts and other institutions will react to regulated situations. Lawyers maintain running odds on various outcomes that matter to their clients by consuming legal input: reading case reports, monitoring new legislation, trading stories with colleagues. A very rigorous logic determines how those inputs affect the odds under which we all play. It's not simple boolean logic. It's much more developed than that.
>The United States legal systems are massively distributed, fault tolerant, resource starved systems that consume exclusively dirty input from malicious users.
The United States legal systems are massively distributed, fault tolerant, resource starved systems that consume exclusively dirty input from malicious users. Like any system that consumes unstructured input, the uncertainty inherent in how the facts of a situation get mapped to structured conclusions ("questions of fact"), and how the legal rules operate on those inputs ("questions of law") to produce a result, are all infected with that uncertainty. Few of the facts the system is meant to deal with reduce to anything so discreet as bitstrings. Legal rules in statutes and even in case law, like the bit of First Amendment in the article, start out trying to state rules with all the rigor of a pure function. But law is a highly dynamic, reprogrammable language. It's Ruby, not C. The statutes aren't the running state, they're just input.
That's partly why I still program to relax after work. It's satisfying to address little bits of a problem in isolation, to see my solutions work, and to know they're reliable. Problem solved.