By Mike Dorf (with thanks to my colleague Tom McSweeney for pointing me to some historical sources)
A couple of days ago at breakfast, I was playing the podcast of my latest Verdict column to check the sound quality, when my seven-year-old daughter overheard the opening line: "Can Congress use its commerce power to require people to eat broccoli . . . .?" She asked me what I was talking about. I opted not to try to explain to her all about Wickard v. Filburn, Gibbons v. Ogden, or the Bank debate in the Washington Administration. Our conversation went more or less like this:
Me: Well, suppose there was a law that said people have to eat broccoli.
Her (incredulous): Is there such a law?
Me: No, but what if there was?
Her: That wouldn't be so bad.
Me (encouraged): Why's that? Is it because you like broccoli anyway?
Her: It's okay but you and mommy make me eat it.
Me: But what if there were a law that said everyone had to eat it?
Her: That's weird because laws tell people things they shouldn't do, not what they should do.
Me (inside my head): Crap, the Supreme Court is going to strike down the PPACA.
Me (out loud): That's very interesting, but some laws tell you what you have to do, like the law that says you have to go to school now.
What to make of the foregoing exchange? Okay, so I was proud of my daughter for her precociousness, but also struck by how she seemed to intuit the act/omission distinction as a basis for sorting permissible from impermissible laws. Of course, my daughter was not distinguishing between federal and state laws. Her thought was simply that laws don't tell people what to do, just what not to do. And she meant this mostly as a descriptive generalization rather than as a normative proposition. Was she right?
I think it likely that in most civilizations ruled by law, prohibitions outnumber affirmative obligations, but there are still plenty of the latter. Laws are not usually categorized as positive and negative, so one would have to go through the statute books carefully to get a clear sense of the relative prevalence of positive and negative laws. But I do know of one legal tradition that expressly utilizes the positive/negative distinction. Jewish law reckons 613 commandments in the Torah: 365 are negative and 248 are positive. So the negative outnumber the positive, but only by about a 3:2 ratio. Of course many of the positive (and negative) obligations involve rituals. Do civil governments impose many positive obligations?
Apparently yes. Bracton, writing in the 13th century, put it this way: "[L]eges Anglicanæ et consuetudines regum auctoritate iubent quandoque, quandoque vetant, quandoque vindicant et puniunt transgressores." That translates as "English laws and customs, by the authority of kings, sometimes command, sometimes forbid, sometimes castigate and punish offenders." That's at Page 21 in both Latin and English here. The note indicates that Bracton borrowed his description from the Roman law scholar Azo of Bologna's Summa on the Institutes, a commentary on Justinian's Institutes. Azo in turn was characterizing Justinian's Digest at D.1.3.7, which said: "the virtue of law is this, to command (imperare), to forbid, to permit, and to punish."
It's clear from the context that "command" is being used in contrast to "forbid," so that a "command" is an affirmative obligation. Neither Bracton nor Azo says whether affirmative commandments are less common than negative ones, and if so, how much less common, but the casual inclusion of both categories in a list of only four categories of laws (commandments, prohibitions, permissions, and punishments) pretty clearly shows that affirmative commandments are a deep feature of our legal tradition.
Thus, the Bible, Roman law, and medieval English law all accept affirmative obligations as a standard form of legal obligation. So I guess my daughter's descriptive generalization was basically wrong. And oh yeah: So are the libertarians who think (on libertarian rather than federalism grounds) that the PPACA "mandate" is somehow unprecedented. But they're only wrong by a few thousand years.