The Public Domain, Digitization, and Copyright

Dr. Klaus Graf, the author of the useful Archivalia blog and a the maintainer of the extensive German site on museum law and cultural property, has supplied an interesting comment on my post on Copyright Term and the Public Domain in the United States. In it, Dr. Graf wonders about when digitization projects can exert copyright ownership over scanned versions of public domain works. In particular, he cites several statements from Cornell University sites, and asks if the University is not overstepping what the law allows. The comment raises a number of interesting issues that are worth an extended response.

First, a disclaimer. While I work at Cornell University, I am not responsible for any of the statements or policies that Dr. Graf cites. Furthermore, I am a big believer in Bridgeman v. Corel, which argued that faithful copies of public domain works did not warrant their own copyright protection. I am also a believer that archives should limit what rights they exert over the public domain portions of their holdings. My presidential address to the Society of American Archivists entitled "Archives or Assets?" and found in American Archivist as well as at <http://hdl.handle.net/1813/52> should make this clear.

Nevertheless, I recognize that many libraries and archives wish to also claim some rights over the use of materials, and this may be legal as well. There are a couple of different approaches that are followed.

First, the policy of many institutions is to claim a compilation copyright in the digital collections they put online. This would in theory prevent someone from copying the entire collection. Furthermore, given the amount of effort (and expense) that goes into making digital files, many institutions have argued that scans of digital files should be copyrightable (even after Bridgeman). Digital photographers with whom I have spoken assure me that their work is both skilled and creative, and hence worthy of copyright protection - though I am not so sure. We may be in a gray area here: scans done on a flatbed scanner are not original, but scans made by a digital photographer with a camera may add some originality.

But even if we conclude that scans of public domain works cannot themselves be copyrighted, I think the copyright statements Dr. Graf cites are still accurate. For example, one statement notes that "Copyrights on all images presented on this site are either held by Cornell University or the artists (and/or their representatives)." The statement does not say that all images on the site are protected by copyright - merely that if an image is protected by copyright, then that copyright is held either by the artist or possibly by Cornell (if it was done by one of Cornell photographers, for example, or if there is a copyright in the digital scan).

Would it be a service to scholarship if digital collections identified which of the images on their sites were in the public domain? Perhaps, but most digitization projects cannot take either the time or the responsibility for this, preferring to leave it up to the researcher to determine if an image is in the public domain. Furthermore, it might also be deceptive. Dr. Graf wants to identify public domain images so that they can be freely used in other projects. Many repositories, however, still wish to use their physical ownership of the original public domain item as a basis for controlling the use of that item. Explicitly stating that an image is in the public domain might lead researchers to believe that they could use the image freely, when in reality their use is limited by other factors.

Dr. Graf cites a good example of this, the copyright statement for the Andrew D. White architectural photograph collection. First, it starts with the standard disclaimer that all American libraries must use to protect themselves against possible liability for copyright infringement. Since most of the images in the AD White collection were published in the 19th century, they should be in the public domain. There may be some unpublished or later images in the collection, however, so including the Federally-required disclaimer language does no harm and may offer some protection.

The statement then discusses the library's rights as physical owners of the original items. Nowhere does the page indicate that the library thinks that it owns any copyright in the images - rather use of the images is predicated upon the library's ownership of the physical original and the contractual arrangement it makes with the user.

This approach is best summarized in the library's general Guidelines for Using Text and Images from Cornell Digital Library Collections. A really excellent summary of the issues regarding physical ownership versus copyright ownership is found at <http://www.library.yale.edu/special_collections/copyright.html>.

Lastly, Dr. Graf wondered whether libraries could claim that they have published unpublished photographs by making them available on the web, thus giving them the extended term of copyright that accrues with publication. There is no case law on this issue of which I am aware, though some have argued that this would indeed be the case. I would argue that even if a library published an unpublished photograph on the web, for the purposes of copyright it would remain unpublished - unless it was done with the approval of the copyright owner. Furthermore, at that point the copyright owner, and not the library, would own the extended copyright in the image.

In short, while many libraries may have in the pre-Bridgeman past claimed copyright in digital scans, and while many might still wish to be able to do so, I believe that most libraries and archives are moving away from copyright claims and instead asserting contractual claims based on their physical ownership of an image. Whether this is a good thing or not is something that I tried to address in the essay I cited at the beginning of this too-long message.

Comments

[Mary]

Update: a historical society that licenses public domain images has been sued. Discussion continues at LibraryLaw Blog August 25, 2005

[Dave M]

I don't think that's even possible under current laws..? i may be wrong tho..

---
Public Domain Books: Online - http://www.PDBooksOnline.com

[James Agenbroad]

Re NGA(nee NIMA, nee DMA) copyright claims. These claims to protection for works of the U.S. Government are, I believe, based on House Report 94-1476 http://www.title17.com/contentLegMat/houseReport/houseReport.html on the 1976 copyright act. The House Report is a sort of a gloss by the House of Representatives (one of the chambers of the U.S. Congress) on the legislation before them. It says in part:
"The prohibition on copyright protection for United States

Government works is not intended to have any effect on protection

of these works abroad. Works of the governments of most other

countries are copyrighted. There are no valid policy reasons for

denying such protection to United States Government works in

foreign countries, or for precluding the Government from making

licenses for the use of its works abroad."

Since the law as passed makes no mention of this, it's a somewhat teuous basis but there it is. I guess that the idea is that while the U.S. doesn't grant works of the U.S. Government copyright, they may be granted copyright in other countries. Why then have any statement, since notice is no longer required? Possibly bureaucratic intertia, but perhaps it is because the Dutch government publications are uncopyrighted UNLESS there is a copyright notice on them. Therefore under the national treatment principle, this claim wouldn't be possible in the Netherlands without notice.

All of the usual disclaimers apply, I am not a lawyer etc...

[Dr. Klaus Graf]

Sorry I thought public domain is public domain and not non-commercial domain. I do not see any difference to
http://blog.librarylaw.com/librarylaw/2004/07/declaration_of_.html

[Peter]

The copyright notice to which Klaus Graf refers says the following:

"© This work is the property of the University of North Carolina at Chapel Hill. It may be used freely by individuals for research, teaching and personal use as long as this statement of availability is included in the text."

My guess is that Klaus may be troubled by UNC claiming copyright in a work published in 1863. My take would be that UNC is claiming copyright in the mark-up they have done (including encoding the text as per TEI level 4), and possibly in OCR and maybe even the scanning.

UNC allows free reign for personal use. They have invested a lot in making the scanned and marked-up text, and would seem not to want a commercial service to come along, grab their work, and then sell it. That position does not seem to me to be unreasonable. While the copyright on the mark-up may be "thin," I suspect there is protectible content there. Of course, I don't think this has been tested in court yet.

[Dr. Klaus Graf]

What do you think about this (c)-Note:
http://docsouth.unc.edu/imls/bartlett/bartlett.html ?

[Peter]

Dr. Graf asks about the legality of contracts for visiting web sites. This is a very complex question, with few definite answers. My best guess is that in states that have passed UCITA laws (such as Maryland and Virginia), there is a good chance that a click-through license is binding. Whether a "terms of agreement" license is also binding is less clear - and it becomes even murkier when one moves outside of those states (or overseas).

Part of the issue here is preemption. The copyright law says that copyright rights are protected by federal law, not state contract law. My reading of the preemption clause and some court cases citing it is that if one is only interfering with the right of reproduction, then preemption might apply and there would be no cause for state action. On the other hand, I think most preemption cases have found that non-copyright rights are usually implicated in these cases, and the courts have allowed these cases to proceed in state courts.

Most of all, I think the issue is moot. Johns Hopkins University, for example, has a web site devoted to reproductions of manuscripts of the Roman de la Rose. Since JHU is in Maryland, the click-through license on the web site would create a binding contract in Maryland. I just don't see JHU or its museum partners ever suing someone for using one of the images - unless for some reason it became a huge commercial product. The question that each individual must answer is whether he or she wishes to ignore the stated desire of the museums that actually own the manuscripts to control the use of reproductions. If one does ignore the museums' desires, it seems less likely to me that they would digitize new items.

[Mary]

I fully agree that a valid contract would be necessary to control images of public domain work, even if a lot of time and expense goes into making them. Digital Restrictions Management (DRM) is the only other way I know of to protect the works.

Copyright law is not likely to protect the images. The Bridgeman case is only a district court case, but its reasoning is sound, and even counsel for museums agrees (below).

Peter writes: "Digital photographers with whom I have spoken assure me that their work is both skilled and creative, and hence worthy of copyright protection - though I am not so sure."

I'd say that the more skilled the photographer is, the less creative it is. The Bridgeman court put it rather starkly, saying that 'slavish copying,' "although doubtless requiring technical skill and effort" to make transparencies of the two-dimensional works lacked any original authorship as required by U.S. copyright law.

To readers not familiar with the case: Corel was selling a CD-ROM with images of well-known paintings by European Masters, including some that Bridgeman claimed to have sole authorization to reproduce. Bridgeman lost, since the Court said there were no valid copyrights to the individual images, making any copyright infringement impossible.
Bridgeman Art Library v Corel ,36 F. Supp. 2d 191 (S.D.N.Y. 1999) http://www.constitution.org/1ll/court/fed/bridgman.html

Barry G. Szczesny, Esq., Association of American Museums Government Affairs Counsel gives further insight into the case at http://www.panix.com/~squigle/rarin/corel2.html

He said that museums tried to get Bridgeman to drop the suit since it was expected to get a decision harmful for museums that want to "copyright" their digital images. He noted that "Just about every museum attorney looking at the case objectively thinks it came out the correct way according to U.S. copyright law ­ that's why no museum had ever brought such a suit. The U.S. Copyright Office informally agrees. Thus, no one saw any real chance for reversal on appeal. In addition, an affirmation by the Second Circuit, arguably the most influential court on copyright issues, would be even more damaging."

-----------
Side note: If you click on the Bridgeman link, you'll notice it's a LEXIS printing of the case. This seems to be a good illustration of contracts having inadequate control over public domain works. LEXIS took a public domain work, a court decision. It then licensed that work to its customers, holding them to a contract that governed its use. Someone who had LEXIS access apparently went ahead and posted it. Not me, by the way. I happened to find it when I googled the case. I'm not bound by that LEXIS license, and could have reprinted it here. Actually, all I did was put in a link and links are FACTS, not subject to copyright. I'm not linking to a site that I know is infringing copyright - though it could be violating is a private contract agreement that LibraryLaw blog readers are not party to.

[Dr. Klaus Graf]

http://www.library.yale.edu/special_collections/copyright.html

This source doesn't mention Bridgeman v. Corel!

Physical access to public domain works is one thing, visiting a website is another. I would appreciate to hear a clear lawyer's statement that there is a valid contract between the institution and the visitor which urges the visitor to seek permission to publish public domain works.

I wonder on which legal ground http://www.nga.gov forbidds any use of all images on their website without permission. There is no own copyright it can claim. NGA is not a private institution like Cornell.