Three notes, a simple decision and a complicated issue. Let's see if I can dissect this properly.

The Sixth Circuit has issued an opinion in two related cases we'll call Bridgeport Music v. Dimension Films. The item at issue is the sampling of three notes from an old George Clinton tune. The notes (re)appear in a number of new rap recordings and initially - back in 2001 - there was a big action to sue about 800 defendants over this unauthorized sampling. Those cases got severed and now a couple of them have been treated by the 6th.

What the court has said is that there is no sample too small to be infringing. The court stated that if performers want to sample they need to get permission, period. This applies to the sampling of performed material. Nothing in the court's opinion prevents someone from 'sampling' the original score and performing those three notes themselves.

In legalese what the court is trying to say is that there is no de minimis exception for music. And, as Lessig points out, that is true on a strict reading of the statute (Copyright Act of 1971). You can copy performed music; you just can't sample, says the Sixth.

The court is trying to draw a so-called "bright line" in this area. A bright line is a simple and strict test that can be applied in an objective fashion. Sort of the opposite of the infamous "I'll know it when I see it" definition of what is pornography. In this case the Sixth specifically noted that it was going beyond the arguments of the plaintiffs in an attempt to create this line.

The problem is that this is a very odd line to draw. It seems to turn on the presence of a particular kind of technology (digital) that permits cut-and-paste. It also seems to discriminate against a particular set of creative endeavors (digitally produced rap, hip-hop, scratch, etc.) in a way that doesn't apply to other art forms, either musical or non (e.g. bricollage).

It's worth noting here that the three notes weren't just directly pasted into the new material. The original sample is four seconds long; the edited version uses two of those four seconds, pitch-shifted, looped, and extended to seven seconds. Susan Butler has a nice analysis of this process in Entertainment Law Weekly noting that the notion of "originality" has gotten somewhat short shrift in this discussion. It's unclear whether the Sixth meant to state that any digital recording of any analog sound is original or not.

The case is also a little odd in that the samplers did not deny the origin of the sample. Presumably if they had we would have watched the parties engage in some kind of shell game in which the original artist or suing copyright holder would have had to demonstrate in some fashion that those three notes were "his." There's more than a little controversy in sampling over whether, when, and to what degree you acknowlege your sources. Some popular artists (e.g. Moby) delight in talking about their influences and their search of old used-LP stores for rare and novel sample sources. Others are much more closed-mouth. Perhaps the latter are wiser?

The court also ignored what is called "legislative history" in this case. Sampling didn't exist when the law was written in 1971. What that means is that judges usually look to decided cases and recent legislation to help interpret the language of older statutes. Since 1971 several interpretations of the Copyright Act have ruled that recordings have less protection than compositions, not more. This ruling reverses that history by ignoring it in favor of reading the literal language.

Finally, as the techno-geek commenters in Lessig's blog have noted, the theory of Fourier transforms pretty much guarantees that you can recreate a digital waveform by first breaking down the original and then producing two or more original waveforms that are digitally combined to be equivalent to the original. I imagine that a patch or plug-in for digital sampling software that did this would be a simple matter of programming. What the Sixth would make of that is anyone's guess.

So yes, the court has drawn a bright line, but it's a pretty poor place to draw one. If three performed notes are protected expression then we're in pretty sad shape. A sarcastic observer might say that the court opted for a simple (even if wrong) test because it is keeping in mind that it has something north of 800 more sampling copyright cases that may come before it for appeal.

If courts keep issuing idiot opinions like this I may have to revise my SCOTUSwatch opinion. Someone has to sort this mess out and the Supremes will find a bunch of these cases looking for docket space. They'll have to take at least one, even if it's just to remand it back down for clarity.

P.S. For the curious: the title of this entry refers to a legendary sign that appears in the MIT Laser Lab. On the door to the lab from the outside is a sign warning "Do not stare directly at laser with your eye." On the housing of the laser itself is a second sign that says "Do not stare directly at laser with your OTHER eye."