Like I’ve said before, I think the fact that Google is a commercial venture shouldn’t blind the courts so that they don’t see what a huge boon Google Print will be to (non-commercial) scholarship, teaching, news reporting, etc. I think the way it’s structured now, Google Print should definitely be considered fair use and allowed. But, I’m not sure this will be crystal clear to courts.

First, you’re right; a parody is technically a derivative work. But the larger point I was making is that a parody will never substitute for an original or a conventional derivative, such as a translation or a film adaptation. Thus it will never have an effect on the market for the original or conventional derivatives. Parodies, therefore, will never have a tough time with the fourth factor. It is clear that fair use was developed for things like parody that cannot substitute for, or compete with, the original and at the same time create lots of new social value. In such cases, we don’t see any societal gain in allowing the copyright holder to prevent such a use.

Second, I’m afraid that a positive market impact for the original is in fact irrelevant. UMG v. MP3 was cited approvingly on this point in the Napster case. Also, the Supremes said in Harper & Row and again in Cambel v. Acuff-Rose that the enquiry “must take account not only of harm to the original, but also of harm to the market for derivative works.”

Copyright law grants creators certain exclusive rights. If you write a book and several studios want to make a movie out of it, you have the right (during the copyright term) to not allow anyone to make the movie. Whether to make the movie or not is your exclusive right. Now, if I go ahead and make the movie anyway and it is a huge hit and as a result you sell three times as many books as you had before the movie, is it a fair use? No. I still violated your exclusive right. Even if Google Print has a wonderful effect on the market for books, this doesn’t get them fair use. There is still a negative effect on the market for licensing books to be used in searchable databases.

The question then is–and you put your finger on it–whether there is a market for the derivative work or whether that market is merely hypothetical. In my example above, a market for movies is well known. A court would not focus merely on the original product as you contend–they would look at the market for the derivative work. So, is a market for licensing books for inclusion in searchable databases hypothetical? I would argue that it is not. Amazon.com and Google Print Publisher are two examples of such a market. Lexis and Westlaw are other examples.

All this said, my point, again, is this: In my mind the fourth factor will be the most important. I don’t think it’s clearly in Google’s favor. Like you, I think the courts should follow the lead of Kelly v. Arriba Soft and find that the market for search engine licensing is negligible and outweighed by the great new social value that is being created by Google Print. But to me it’s a 50-50 shot at best whether a court will be that enlightened and not simply find that there is a market for licensing that Google is harming.