[My withering-away but not-dead-yet blog seems to be turning into all-Swartz all-the-time when I manage to post. But sigh, I feel an emotional connection here.]
In a legal debate over the Aaron Swartz prosecution (does any really need an explanatory link now?), Lessig has posted a response paper which has some interesting details for those who follow the case closely. It's worth reading for its treatment of various claims, via correcting errors in the paper to which he's replying. I'll extract some important parts below.
The first plea deal (everything afterwards was worse) is given as:
The initial offer was (1) 90 days in prison, (2) then time in a half-way house or home detention, (3) two years banned from using a computer, and (4) pleading guilty to a felony. As was discussed with the prosecutor, because of a medical condition, Aaron could not qualify for a "minimum security facility."
Note that computer use ban for two years. That's essentially two years of professional unemployment and isolation these days.
Further, the part of being a convicted Federal felon deserves emphasis. It's not just about being unable to vote or own firearms. It's a huge burden for one's entire life, affecting employment, housing, travel, and more (yes, it would have hurt someone of Aaron's privilege much less than it would, say, a poor black man, no debate there - but it's still immensely destructive).
I suspect this is one of those things which is outside common experience, so simply doesn't enter consciousness when people think about the case. I've seen many comments that assume the downside of a guilty plea would be a few months at most of prison, then he could just put it all behind him (and then some people go on to presume since that wouldn't be so bad, it's obviously irrational not to take the deal, so his suicide must be irrational, end of thought-process). But that calculation is wrong. The downside of a plea includes a certainty of lifelong "Federal felon" damage.
And pleas are not even bounded to the downside in themselves:
I know many advised Aaron to accept the plea. His lawyers warned him that there was some chance the prosecutor would not stick to her offer (in US v. Gonczy, 357 F3d 50 (CA1 2004), she was admonished for promising to support one sentence and then arguing for a more significant penalty), but even so, many viewed the costs of fighting the charge (both the estimated $1 million in legal fees and the threatened penalties if he lost) as greater than admitting his own guilt.
Note, regarding penalties, I've seen several objections to quoting the maximum years of prison of the various felony charges, in terms of this being misleading. Yes, the maximum isn't the average, and almost nobody gets the maximum. But that's sort of the definition of "maximum", i.e. it's the worst possible outcome, not the typical result. HOWEVER, when people then seem to implicitly argue that the value has no relevance, I think there's a severe reasoning error there. To wit: Almost nobody will get the worst case result, but everybody has to worry that it'll be them.
It's also struck me that "maximum" != "average" would be a particularly unhappy argument to someone in Aaron's position, as he'd just had the situation go extremely, horribly, incredibly badly. Putting aside finger-wagging moralizing about accepting consequences, I don't think he expected to wind up facing multiple Federal criminal charges. The key question I suspect he'd be thinking about isn't "What do Federal Sentencing Guidelines recommend?" (which is complicated in itself). It's "What if the end result is as bad as everything else has been up to now?".
Lessig also has an interesting comment on motive, similar to what he's said before:
Neither you nor the prosecutor knows Aaron's motives here. I do. But without revealing what I know, let's be clear about what could reasonably be believed: one could reasonably believe only one of two possible motives: either he was (1) building a database to analyze bias in scholarship (as he had done while at Stanford using the Westlaw database, resulting in a published article), or he was (2) intending to make available (consistent with the Manifesto, and likely in the Third World) in disobedient protest this scholarly material.
I've always thought that Aaron intended to make available only public-domain material, and was later going to sort out what could be legally redistributed from what was still restricted by copyright (I should hasten to disclaim I have *zero* actual knowledge of his motives, and we never even discussed this particular topic). That would be consistent with what he'd done before, e.g. making available court filings. But as Lessig has never mentioned this mild variant, I guess I have to accept I was wrong on that (which saddens me greatly).