The Gray Market
(Read the title in a spooky voice. Maybe imagine lightning booming in the background.)
Now let us bring up the specter of the gray market. A print of a film is in the public domain or it isn’t. There are no shades of gray there. We’re not involving donor restrictions if you’re sourcing from an archive print — that’s another matter entirely — but if you possess a print of a film unencumbered by copyright, that print is yours to do with as you see fit. The film belongs to everyone and that includes you — that’s the very concept of the public domain.
The “gray market” does not exist. It’s a term invented by publishers to discredit resellers, especially those legitimately buying items cheaply in one market and reselling them in another market that the publisher is trying to price-gouge. Think buying a textbook in East Asia and reselling it in the US, where the same book retails for a hundred times more. This is perfectly legal as the courts have ruled numerous times. It is the right of first sale, as expressly defined by section 109 of title 17 of the US Code. All the publisher can do to legitimize their price inflation is to try and tarnish the resellers by linking them to black market bootleggers and counterfeiters in the minds of potential buyers. That’s what the “gray market” is.
Now there are orphan works where the trail of ownership has been lost and nobody is quite sure who the film might belong to. Releasing these films is sometimes referred to as “gray market” but it isn’t. It’s simply piracy — black market — but it’s banking on the fairly safe bet that whoever does own the film has no idea that they do and will not pursue the infringement. Harpodeon does not deal in orphan works, but I personally have no issue with pirating films in this case — there’s zero chance you’ll see them otherwise.
There are those who won’t watch releases from Grapevine, Alpha, etc. — including Harpodeon, of course — because they’re “gray market” distributors, and frankly, that’s idiotic. Their videos are no less legitimate than those from Kino and a great deal more legitimate those from Edition Filmmuseum. Now, Alpha’s image quality ranges from passable to you’re not even sure if there’s a picture beneath the murk and avoiding them in that case is understandable. Granted, several of our videos are sourced from very poor prints — that’s why we try to be upfront about the quality and why we provide screenshots and video clips to give an unvarnished idea what the print looks like. There are those that want perfection and would rather have nothing than fall short of it. They usually get nothing. Far be it from me to call them wrong, but when it comes to rare films, I’ll take what I can get and be happy it’s available at all.
You’ll see this idea frequently in old Usenet discussions but there are those that cling desperately to it still that one who restores a film then has some copyright claim on it. To begin with, there’s a question of what “restoration” means. In some cases, it means a great deal of work assembling disparate elements and working backwards to find what the original release must have looked like. It can mean nothing more than cleaning and transferring the film, with the cleaning either done chemically/mechanically to the film before the transfer or digitally after. Some “restorations” don’t even bother with the cleaning bit: restoration can simply refer to finding a definitive print and copying it. And let’s not pretend, “definitive” in this case often means “the one we have”.
For all our post-HD releases and the later SD releases as well, we chemically clean the film itself, transfer it, disassemble it into its individual shots, stabilize the image of each shot, diminish or remove the flickering, adjust the levels, diminish the grain where it’s become excessive from multiple generations of copying, sharpen it, tint it where the tints are known or can be confidently guessed, and then reassemble the film. Particularly distracting damage (e.g., a torn sprocket hole on 9.5mm, which leaves a heavy scratch up the center of the film) is painted out frame by frame. If the titles in the master are replacements (all Excel prints from the 1930s, for example — they didn’t use the original Selig titles) or if they’re badly cropped (example of what I’m talking about from The Juggernaut) or illegible, then the titles are re-created to more resemble the originals. When we have several prints of a film, all of them are transferred and combined in the chance that one print has more or longer scenes than another. When you’re dealing with old film, absolutely no two prints are identical.
Of course, the more slavish your restoration work is to bring the film back to its original release form, the less eligible for copyright protection it becomes. That’s where the old Usenet argument falls flat on its face. There may be a great deal of labor involved in restoring a film but no creativity at all. Copyright protection is about promoting creativity. If your goal is a copyrightable work, then you don’t want to restore anything, you want to transform it into something new. For silent films in the past, that was commonly done by replacing the titles with your own or by Raymond Rohauer’s old trick of rearranging the scenes to tell a slightly different story, but that’s mostly fallen out of favor nowadays. The “special content” copyrights you see claimed on public domain titles refer only to original content, i.e. the score.
The definitive case on photographic copyright law in the US is Bridgeman v. Corel (1999), which found that “a photograph which is no more than a copy of a work of another as exact as science and technology permits lacks originality. That is not to say that such a feat is trivial, simply not original.” The last line is a reference to Feist v. Rural (1991), which established that labor and skill are immaterial to copyright and a collection of uncopyrightable material remains uncopyrightable no matter how much work went into it. Simply put, “the sine qua non of copyright is originality.” This was expanded upon in Assessment Technologies v. Wiredata (2003), which found that a work based on public domain material that nevertheless meets the threshold of originality for copyright protection still has no claim on the underlying public domain material itself. Further, any attempt to assert such a claim is considered an abuse of process, forfeiting whatever intellectual property case they might have had and simultaneously opening themselves to countersuit. They cannot prevent anyone from extracting the underlying public domain material for their own use, or for undoing whatever transformative process made the work original in order to access the public domain material. Cases like Bridgeman v. Corel have been brought several times since with the exact same outcome, as in Meshwerks v. Toyota (2008), which found again that “reproductions do nothing more than accurately convey the underlying image” and are ineligible for copyright.
This is for the US copyright law, elsewhere may vary. Canada is more or less the same with regard to creativity (see Tele-Direct v. American Business Information (1997)). From my little familiarity with EU law, I know that whatever their objective for copyright is, it doesn’t make a great deal of sense and isn’t at all consistent. I wouldn’t be surprised if labor is considered there, even if it isn’t transformative, particularly if such a ruling would be protectionist to a member country.
As for what is transformative, the bar is rather high. Bugs alone may deter copying but are not transformative enough to qualify for protection. That’s doubly true if the bug is blurred or blacked-out in the copy. But really, you stop caring so much about your films being “stolen” when you accept that they were never yours to steal. Rohauer didn’t give a damn about film beyond his ability to monetize and litigate it. Don’t be Rohauer. If you want to get into releasing silent films, get into it because you want to release silent films, and don’t expect to make a dime.