3

I'm attempting to demonstrate that a film registered in 1947 did not have its copyright renewed by the registrant in its 28th year, and thus would have fallen out of copyright and become public domain (because it was published between 1922 and 1964 before the law with respect to renewal changed). The film itself is already present in several public domain archives, so there is good reason to believe this is the case; however, one needs to be certain about these things before publishing a work that incorporates the original.

I can pay a firm to do a search of registrations and renewals based on the data present at the Library of Congress (the data is not present in the LOC's online search due to its age), or I can pay them a lot more (4x) to do a more exhaustive search for assignment of the copyright to other entities as well.

My question is this: if the production company had assigned or transferred their copyright to another entity prior to the 28th year, and that other entity had renewed the copyright themselves, would we still expect that renewal to be reflected in a search of registrations and renewals for the copyrighted work?

Alternatively, is there a better approach to demonstrate that a work is now public domain?

(If so, then presumably the less-exhaustive but much-less-expensive search should be adequate to show that the film did not have its copyright renewed in 1975 and therefore became public domain.)

2
  • If it was published in 1992, that puts it before the 1923 cutoff, doesn't it? Apr 10, 2018 at 4:15
  • 1
    My understanding is that if it was published before 1923 it's already public domain... works published from 1923 to 1964 had to be renewed in their 28th year or they became public domain. If they were renewed, then they'd be good for 95 years from the registration date. The work in question was published and registered with the copyright office in 1947, so it would need to have been renewed in 1975 or fall into the public domain. Apr 10, 2018 at 4:53

2 Answers 2

3

The renewal registrations, published by Project Gutenberg (PG) are available online, at https://www.gutenberg.org/cache/epub/11800/pg11800.html The registrations are also available in smaller chunks, 6 months at a time, for example at https://www.gutenberg.org/cache/epub/11801/pg11801-images.html

The intro says:

Produced by Michael Dyck, Charles Franks, pourlean, and the Online Distributed Proofreading team, using page images supplied by the Universal Library Project at Carnegie Mellon University.

RENEWAL REGISTRATIONS—LITERATURE, ART, FILM

An alphabetical list under title of all works (with the exception of musical compositions) in which the renewal copyright was registered during the period covered by this catalog. Included in the list are cross-references from all essential names associated with the work and from variant forms of these names.

Project Gutenberg itself publishes much of its content based on the "lack of renewal" situation. I myself have done searches using these online documents which were accepted (when confirmed by a different volunteer) as sufficient due diligence for PG to publish works not found to have been renewed.

The registrations list the original title and author, as well as known (to the US Copyright Office) alternate names and titles. For esample, here is a sample record for a film:

ADAM'S RIB, a photoplay in ten reels, by Famous Players-Lasky Corp. © 7Feb23, L18658. R58624, 17Feb50, Paramount Pictures Corp. (PWH)

Doing such a search, using these online records, is relatively quick and simple. One thing to note, the law specifing "renewal in the 28th year" actually allows for plus or minus 1 year, so one must check the 27th, 28th, and 29th years after publication.

One possible source of false negatives: if there is an alternate author name (pseudonym) and alternate title not known to the copyright office (so that no cross-ref is included) and also not known to the searcher, and he work was registered nd renewed under the alternates, then it would be missed. But the renewals all link back to the original registration, so the alternates would have to have been used in the original registrations.

Later assignments will show up under the name of the original author. (For a film that will often be the production company.) The original title will also be listed, as will the original registration number. If you can find the original registration in the copyright office records, that will give the registration number (R-number) which can also be searched on.

I should mention that an error could occur because of a transcription error on the part of Distributed Proofreaders (DP). But this is unlikely. I have worked with DP, and the text is first created by an OCR program from the scanned images, and then checked against the original images by six separate volunteers. The result is quite accurate, errors are rare.

One could do a self-search first, the cost would be zero, and much easier than using the old card catalogues. The odds are high that if the renewal is present, it will be found. If it is found, the work is not PD, and there is no need to pay anyone.

If it is not found, one could then consider paying a private firm, or the LC searchers, to try to find any odd cases. But a search on these files is quite likely sufficient due diligence to establish "innocent infringement" if a false negative occurs.

1

Did some additional research on this, and the answer is probably "it depends on who is doing the search." I contacted CompuMark, one commercial entity which runs its own copyright search service and asked them if their "Registration and Renewal" search product would catch this corner case. They assert that it would. I can't vouch for them, and I am sure there are multiple companies in this market from which to choose.

The Library of Congress itself also offers a search service; by their description this search is pretty thorough, and while I haven't gotten a definitive answer from them yet, I strongly suspect based on their description that they would also discover such a renewal.

Less clear is what would happen if you visited the Library of Congress and attempted to perform the search yourself. According to circulars published by the LOC the card catalog contains "idiosyncrasies" and someone unfamiliar with its quirks might miss something.

So I think the bottom line is that you're probably best off paying someone (be it a private searcher or the LOC itself) who understands the LOC filing system to do the search on your behalf, and ask them ahead of time what their search will and won't discover before you order your search, rather than trying to execute such a search in-person. Make sure that whatever you're paying for will cover the cases that matter to you.

1
  • 1
    This is an accurate assessment. It isn't that different in principle from a search of real property records, in which it is possible to find all properly indexed records, but many people lack the technical expertise to conduct the search properly.
    – ohwilleke
    Apr 11, 2018 at 5:23

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .