Typesets of public domain autographs that have 1st Eds

[rkram53]

I've posted a number of typeset editions of public domain autograph scores on IMSLP, primarily C. Graupner. I was told that there may be copyright issues in doing this if a First Edition of the work exists published within the last 25 years? I can't imagine there are copyright issues with typesetting an autograph accessible to the entire world on IMSLP, especially if I live in the US. Can you please tell me if there are any issues to consider here? I assume the IMSLP copyright staff vetts this kind of thing, if it is even an issue these days.

[Carolus]

This is pretty complicated. The problem arises from the fact that works of composers which have never been formally published are treated differently in different places and at different times, so here's a short summary:

1. Canada. Canada's posthumous publication provision (50 years from publication) only applies to works which have never been "performed, recorded, or delivered". Thus, the only thing that can be protected with a composer like Graupner is a new edition or adaptation. If the item was never performed but the composer is long dead (like Graupner), the situation is murky, but at least one provision makes it likely that all older works like this are free and only "new adaptations" can be protected.

2. USA. As always, it's own circle of copyright hell. If a given work was first published (which can be a fairly complicated thing to define) between 1923 and 1977, the work itself could be protected for 95 years from the date of first publication. However, if published between 1923 and 1963, it had to be published with a proper and compliant copyright notice and duly renewed 28 years later. Items first published between 1964 and 1977 with a proper notice are protected for 95 years. Items first published between 1978 and 1988 with a proper notice are protected until 2048, while those first published between 1989 and 2002 are protected until 2048 regardless of the copyright notice or lack thereof. On 1/1/2003, all unpublished works of authors dead over 70 years automatically entered the US public domain. So, if unpublished on 1/1/2003, no problem. If published before 1923, no problem. If published 1923-2002, maybe a problem.

3. EU. Unpublished works are covered by a doctrine called editio princeps, which stipulates a term of 25 years from first publication. Note, this effects things in the USA published between 1923 and 1971 without a notice or an improper notice - they are ineligible for restoration under GATT because they were public domain in the country of origin on 1/1/1996. Recently, a EU court ruled in the application of editio princeps to an opera by Vivaldi. It held that first publication took place in the 1700s when the parts were sent from one opera library to another (=distribution) and that the work was therefore ineligible for editio princeps protection.

[rkram53]

Thanks. This is a minefield. How about derivative works. If I make an arrangement of a public domain autograph for other instruments and even change things, add parts, lines, etc and there happens to be a valid 1st edition of the original. Are there issues here?

[jossuk]

A note of thanks to Carolus for a lucid presentation of a thorny subject!

[Notenschreiber]

@ Carolus.

A general question, but I try to make it clear by an example. You write under 2.USA:

"those first published between 1989 and 2002 are protected until 2048 regardless of the copyright notice or lack thereof."

Does that mean: "first published in the USA and protected in the USA until 2048", or "first published somewhere and protected
in the USA until 2048" ?

[Carolus]

The USA law is somewhat confusing and contradictory, but the most conservative view is that a work first published anywhere between 1989 and 2002 is protected in the USA until 2048 - even if the composer has been dead for 500 years. By this I mean the work itself, not just the edition. However, the issue actually gets quite complicated in at least two respects:

1. Under US law, only authors and their legal heirs are qualified to make the initial copyright claim. A publisher can be the legal claimant only if there is a written agreement with the author's legal heirs to publish the work. Ownership of a manuscript does not convey the ownership of the work written upon it, so an agreement with a library would only be valid if the author or an heir willed the copyrights in donated unpublished works to the library along with the manuscripts. Therefore, if a EU publisher issued a Graupner work in 1997 naming themselves as the copyright claimant without an agreement from Graupner's heirs, it could be argued in court that it was an "unauthorized publication" - which means that the work remained legally unpublished and therefore entered the US public domain on 1/1/2003 (although the edition might still be protected as a derivative work). Under the EU's editio princeps doctrine, which does not exist in the USA, there is no problem with a publisher being the claimant of a work like this.

2. The definition of what exactly constitutes "publication" can be very tricky. It is entirely possible that the listing of a work in a library catalog which was distributed to other libraries around the world actually fulfilled the definition of "first publication" under US law. This is a similar line of reasoning as that used by the German court in the Vivaldi opera case to deny the editio princeps claim being made by a publisher.

[Carolus]

Thanks. This is a minefield. How about derivative works. If I make an arrangement of a public domain autograph for other instruments and even change things, add parts, lines, etc and there happens to be a valid 1st edition of the original. Are there issues here?

If the work itself was first published between 1923 and 2002, there is some possibility that you would have to obtain permission from the copyright owner of the first edition to make an arrangement or other derivative work in the USA. As you can see from the discussion above, there are a fair number of complicating factors in all of this, so the actual probability of you running into an issue is still reasonably low - even in the USA. Since IMSLP's principal server is in Canada, the issues are really much, much less in terms of making things available for download. The chances of a work of a long-dead composer like Graupner being eligible for Canada's 50-year term for posthumous publication are very slim. Canada's copyright law is quite advantageous in many respects and was even well-designed in places to weed out the kind of sheer insanity one encounters in the USA. (The whole snake-pit of copyright nonsense surrounding the well-known piece Happy Birthday being a superb example.)