There are pretty clear tradeoffs in your example (if we stipulate that there's no way to get the compulsive, habitual shoplifter to reliably stop shoplifting).

There are pretty clear tradeoffs in your example (if we stipulate that there's no way to get the compulsive, habitual shoplifter to reliably stop shoplifting). > Does the repeat shoplifter, which can't be trusted to stop, deserve life in prison then? Two choices: 1) Accept the cost to society of letting a person run around stealing from various different businesses, probably over the course of years. Not only do those businesses lose money, but it makes their employees' work experience more unpleasant, and adds to a general low-trust atmosphere among strangers. 2) Accept the cost to the shoplifter of putting them in a miserable environment where they won't be able to thrive as a person. Either dispersed costs for lots of people who aren't responsible for the wrongdoing, or concentrated costs to the one person who is doing the objectionable actions. I loosely favor the former... but this is a constrained example, and the available options kinda suck. (Not dissimilar to real life in that way...) If you can actually get the shoplifter to stop their bad behavior, then the options and attendant tradeoffs change.