sigh I swore I wouldn’t interject myself back into this thread, because arguing with the “it should all be free RIAA sucks may Sonny Bono rot in hell” crowd gives me fits. But the temptation is too strong.
Anyway.
One of the chief merits of the life + X term, over a fixed length term, is ease of administration. In the U.S., death is typically a recorded and verifiable event. This is in sharp contrast to date of publication, date of creation, or any number of other dates that could be picked. If you’ve ever been in litigation, trying to establish a factual issue like that can be a thorny issue.
Life plus X also helps the public to know whether a work is under copyright or not. Is the author alive? Yes/no. If no, when did they die? If that number is greater than X, the work is usable, otherwise not. Life plus X also allows the entire output of an author to enter the public domain at the same time, which makes it easier to administer estates, or for those interested in publishing to create more attractive packages. For example, X years after J.K. Rowling dies, anyone can publish any of the Harry Potter books, regardless of when they were originally published. A fixed term system would have “Sorcerer’s Stone” be available on one date, “Chamber of Secrets” on another, etc., etc. Such complete sets are often more marketable than a la carte works.
Now, those who advocate fixed terms often want to go back to a system where registration/ publication was king. The problem with registration is that it deals poorly with unpublished works. There are many creative works that, for whatever reason, an author may not choose to publish, and which society would generally agree should not be published. (Journals and correspondence come immediately to mind.) Requiring registration and publication as a prerequisite to protection strips these works of any protection, and deprives the author of the right to enjoin their further publication. So if someone breaks into J.K. Rowling’s house, steals her private correspondence, and publishes it on the Internet, she has no recourse, even though the loss of control was not her fault. Requiring registration therefore creates a perverse situation - to prevent publication of private, unpublished papers, the author has to register the copyright and disclose those papers.
There is something to be said for the idea that no one should be required to forfeit their property while they are alive, absent some compelling interest. In the case of patents, there is a compelling interest in requiring forfeiture, but the rights patentholders are granted for their limited term of protection are correspondingly greater. (There is no equivalent of “fair use” for patents as exists in copyright.) The existence of a patent more clearly inhibits future innovation than the existence of a copyright inhibits further creativity. A patent on a chemical process could prevent me from practicing or commercially exploiting an improvement on that process. But the copyright on the Harry Potter books does not prevent me from writing other works of children’s fiction, or works involving young wizards.
There is also the problem of deferred earnings. In most other professions, the practitioners are paid up front for their work. I’m a lawyer, and I get paid by my clients within two months of doing the work (most of the time). Some have said that “if an author can’t make good money off their work in ten or twenty years, they are a commercial failure and someone else should have a go at it.” But the nature of artistic creation is such that the innovation or creativity of a work may not actually be realized for some time after the work is published. Short copyright terms might stifle innovation and hurt the public in the long run by encouraging the produciton of crowd-pleasing mediocrities (more marketable in the short term) over innovative and creative works whose genius and impact might not be apparent until years afterwards.
So, a life term makes good sense. Life plus makes sense also, because authors who have families, like any other professionals who have families, would like to provide for their heirs after they die. Preventing copyright from passing into the author’s estate limits that ability, and may discourage people from pursuing careers in the creative arts. Furthermore, if the author dies just before the work is commercially successful, the author’s heirs lose out not only on the earnings that the author might have made had he/she pursued an alternate career, but also the earnings that the author would have earned had he/she lived. The case I’m thinking of here is the composer of the musical “Rent”, who died before the musical opened on Broadway and became a big hit. If he had heirs, they lose out not only on what he could have earned had he lived, but what he would have earned had he simply gone into insurance or accounting instead of writing musicals. (Obviously, there are limits to this, which is where the fixed number of years after death kicks in.)
Now, the proposal of having copyrights be shorter for works held by legal persons, as opposed to natural persons, makes some sense, since there is some logic to the idea that a major publisher has greater wherewithal to exploit the work than the author. I kind of like the idea of limiting rights granted to artificial persons (as opposed to those which arise under work for hire) to twenty year licenses, renewable at the option of the author or the author’s estate through the life of the natural copyright. This would prevent an unnatural forfeiture if, for example, the author forms a closely-held corporation to manage his/her intellectual property for liability or tax purposes.