I’m teaching an Ethics and Technology class this semester. I came up with a thought experiment today that I’m going to have my students discuss on the course blog. I’ll refrain from articulating what I think the philosophical upshot is. For now, I’m curious if anyone has intuitions about the case. Here it is:
The Case of the Copyright Hoarder
Imagine someone developed a computer program to help them generate every possible combination of sentences of English. For each possible set, they set up a blog to post it to a website and then (of course) copyright it. The result is that anytime an author writes a short story, poem, novel, it will already be in their database of works that they have claimed copyright to. Do they have a moral claim to determine how all future written works of English are produced/distributed?
If you think moral claims to copyright are more stringent, when the content is generated by some cognitive effort, then change the case so that the person is near omniscient, can quickly think through all the possible combinations and have them uploaded to a database (or something like that).
I’m curious what your intuitions are about either case.
Hi Andy, I think the premise of the example is false. Poems etc are not individuated by their words, sentences, etc, for reasons given by Borges, Danto, Levinson etc. So unless someone copied the blog, they would not be producing instances of the works on the blog only works word-for-word indiscernible. Now whether the law would recognise this, I do not know. Lee.
I’d guess I’m on the side of more or less everyone in having the intuition that this person as no moral claim at all. Whether I can articulate a clear rationale for that intuition is another matter, and I won’t even try here. But there’s a point that, whether ultimately relevant or not, struck me immediately.
Make the absurdly inadequate assumption that there are only 50 distinct sentences in English. Works could consist of any number of instances of those sentences, in any order. The simplest works would consist of just one sentence, the next simplest two sentences, and so on.
If we allow repetition (and why not?) then the sum total of all works no longer than 50 sentences formed from these 50 base sentences is 9 followed by 84 zeros.
So suppose we rule out repetition. We still end up with a total of over 8 x 10^64 works. And of course, if we made the number of possible sentences even remotely realistic, and the length of possible works long enough to include even a long essay, let alone a novel, we end up with a number so staggeringly large that the hoarder’s project is within an iota of absolutely impossible.
If we consider realistic cases of auto-generated works we might have qualms about whether the “author” has a moral claim to copyright, but if the works really are generated by mere combination/permutation, it’s pretty unlikely that it will matter. Most of the output will be pure garbage that won’t put any serious restrictions on any serious author. The closer the case comes to threatening real creativity, however, the closer it comes to sheer fantasy.
I’m not sure of the moral of this bit of arithmetic. But there it is, for whatever it’s worth.
I was about to post similar math to what Allen did. Seeing that he beat me to the punch, I’ll post my intuitions:
This seems to be another example of the letter of legislation getting in the way of the spirit. But forget that. Furthermore, suppose we accept we have identical works and we deny the computational impossibility and the intuitions that the effort of writing are what we respect about art.
My intuition is that it is the selection process is being left out here. In writing, it’s not the quantity of output you produce, but the quality that is important. If it can be assumed that the hoarder WOULDN’T have selected to put any particular work forward without being prompted by someone else’s success, then HE is actually the one that is producing derivative work. Consider what he would have to do to search his database. He’d have to type in the work word-for-word into his search engine. The fact that the work – and infinite others – were already uploaded is no more relevant than if alphabet soup had ever been accidentally rearranged into the works of Shakespeare.
I hope that makes sense, I’m several cups of coffee in for the day.
I’m glad you are making your students think about this stuff! The quality of our discourse on copyright (etc.) is extremely poor, due in part to its having been shaped by the big content holders. Here are a couple of thoughts.
The scifi story “Lobsters” by Charles Stross (2001) features a hero who holds a business method patent on the idea of feeding a ‘problem space’ into a computer to automatically generate patent applications on every possible solution to the problem. In the future, the patent application process is automated, so he generates millions of patents per day, which are automatically assigned to the Free Software Foundation in an attempt to disrupt the global IP economy. Unfortunately, the end of the story may be too disturbing/offensive to give to students. (On the other hand, the disturbing/offensive ending raises issues about gender, sexual consent, and reproductive ethics that may be relevant to your class.) It appears to be part of a series of stories, so maybe the author has written another story about this character without the bad ending.
Last year I taught an ethics class in which we briefly discussed Locke and Hume on property. I asked the following question on an exam:
“What factors would a Humean need to consider in evaluating the justice of copyright and patent (‘intellectual property’) legislation? How would a Lockean reason differently?”
Despite my scare quotes, not one student questioned the assumption that copyright and patent are kinds of property, subject to the same moral norms as tangible property.