The Calgary Sun reports that a judge has found a man not guilty of a brutal assault (actually, "not criminally responsible by reason of a mental disorder"), accepting the defense argument that the man suffered from "parasomnia." That is, the judge found him not responsible for beating someone up with a baseball bat and raping her because he had been sleepwalking at the time.
The defendant had hired an "escort" to spend four days with him in Calgary (at $1,000 a day, which seems like a lot although I've never been to Calgary). Their simulated relationship was fine for three days, but on the fourth, she testified, he hit her with a baseball bat and then choked her until she passed out. The report says that when she came to "he was having sex with her," although normally the more accurate term for that situation would be "rape." She was later able to escape though the defendant pursued her, apparently still snoozing.
He claimed he had no memory of the attack. The defense argued that he had been suffering from "parasomnia," a disorder in which a person "commits involuntary acts while sleeping." Sleepwalking is a subtype of this disorder, which also includes "restless leg syndrome," "night terrors," and, perhaps, sleepassaulting. The prosecution, of course, argued that he was just making this up, but the judge found it credible. "I do not accept the Crown's assertion that [defendant] was fabricating a story," he wrote. The finding is not an outright acquittal, and the defendant will have to undergo a psychiatric assessment and treatment. But the judge basically ruled he was not capable of forming the necessary intent to commit a crime.
So, is this total crap? It's actually hard to say. Parasomnia is a real disorder, but the extent to which it can cause violent behavior isn't clear. The sources linked above refer to another subtype called "sexsomnia," which is just what it sounds like, and can allegedly produce "complex sexual behaviors" during sleep. Whether they can be complex enough to involve baseball bats and chasing after victims is probably quite debatable, though.
Nor is this the first time the sleepwalking defense has been tried in court, or the first time it has succeeded. This article in U.S. News and World Report lists seven cases in which the defense has been tried, dating back to 1846. It worked in three. (In Fain v. Commonwealth, in the 1870s, the defendant shot someone and then "repeatedly yelled 'Hoo-wee!'" Conviction reversed on appeal.) More recently, another Canadian was acquitted of murder in 1987 for this reason (parasomnia, not yelling "Hoo-wee!"). In that case there was a strong family history of parasomnia, and police also noted that the man - who had driven 14 miles to get there and drove himself to the station afterward - seemed to be unaware he had severed tendons in both hands during the attack. That probably went a long way in terms of credibility.
In most cases, of course, this is likely to be just as bogus as its polar opposite, the too-much-caffeine defense. But here, it worked.
Please note the complete absence from this post of any joke along the lines of "the defense rests."