Orphan Works

Books are flying

Marlene Schiffman

Catalog Librarian, Metadata Services and Processing

Marlene Schiffman
Marlene Schiffman, Catalog Librarian, Metadata Services and Processing

An orphan work is one for which the owner of a copyrighted work cannot be identified and located by those who wish to make use of the work—typically publishers, libraries, museums, historical societies, universities or archives–in a manner that requires permission of the copyright owner. The long duration of copyrights, and the absence of mandatory registration, notice, renewal and recording of transfers have given rise to millions of orphan works.[1]

Even if the user makes a reasonably diligent effort to find the owner, there is a remote risk that a copyright owner could bring an infringement action after the work is reused. Secondary users would have to initiate another search since they cannot rely on the search of a previous user. It would stifle creativity to prevent authors from remixing, sampling and otherwise building upon the works of their predecessors as raw material for new expression.[2]

A search may be conducted by delving deeper into copyright records at the Copyright Office.[3]  These records may help determine who owns the work currently through records of copyright transfers. Copyright Office renewal records will reveal if the publisher has failed to renew the copyright in the work, which puts the work in the public domain if it was published between 1923 and 1964. The three most common methods of searching Copyright Office records are hiring a search firm, paying the Copyright Office to do the search or searching the Copyright Office records using the internet. Each search requires a different amount of effort, and in many cases, the copyright owner still cannot be identified.

With visual or digital works, software and related tools that facilitate image recognition, fingerprinting, watermarking, audio recognition and licensing for copyrighted works are technologies already available, and startups are hoping to fill this role in assisting those who search for copyright owners.[4]

The problem with these schemes is deciding upon whom the burden would fall for creating the database; for example, who would have to register the billions of analog art works in a digital format? The creators of such works do not have the resources to engage in such a project.[5] Mass digitization projects may serve important public benefit objectives including preservation and accessibility and providing authors and copyright holders of out-of-print works with potential new sources of revenue. How could any archives possibly locate millions of copyright holders associated with such a project? Congress has only exacerbated the problem by enacting longer and longer copyright terms, abolishing the requirement that copyright holders must renew the copyright twenty-eight years from publication and that they register their works with the Copyright Office.[6]

Potential users of orphan works contend that if users conduct a reasonably diligent search, they should be protected from excessive penalties should the owner appear. Payment should be limited to a fee as if it had been negotiated beforehand between the user and the copyright owner. Photographers have suggested that use of orphan works be limited to individuals for non-revenue producing personal or community purposes. Non-profit educational institutions might be given permission to use them in exhibits, documentaries or websites and even in souvenir sales in connection with an exhibit.[7]

Ultimately, a legislative solution needs to be addressed by Congressional action. Bills were proposed in 2006 and 2015, but they have not resulted in any concrete legislation. Comprehensive copyright revision is not expected soon. So how should libraries and archives handle orphan works in the meantime?[8]

An institution must consider the following risk factors when proceeding without permission:

  • The investment in the project using the copyrighted work. The more money spent on the project, the greater the risk if its publication must be suspended.
  • The diligence of the copyright search. The more diligently it has been researched in good faith, the less risk.
  • The nature of the work and the reduction of risk. How easy will it be to remove the unauthorized material?
  • An analysis of the risk. This will determine how hard it would be to replace the material if the owner insists that it be removed.

The likelihood of discovery depends on how widely distributed the infringement is. Liability might result only in payment to the rights holder of the standard fee within the trade for a similar use.[9] Ultimately, stakeholders would benefit from legal clarification from Congress that would allow orphan works to be available for public use in a way that is equitable for both users and copyright holders.


[1] Neil Weinstock Netanel, Copyright: What Everyone Needs to Know (New York: Oxford University Press, 2018), 52.

[2] Netanel, Copyright, 203.

[3] Richard Stim, Getting Permission, 6th ed. (Berkeley, CA: Nolo, 2016), 44-45.

[4] Allan Adler Testimony, in Copyright and the Orphan Works Issue, ed. Andrew D. Owens (New York: Nova Science Publishers, 2010), 8. For a description of PicScout, “a young technology company based in San Francisco CA and Israel,” specializing in image recognition technology, see Maya Gura Testimony in the same volume, 15-16.

[5] Victor S. Perlman Testimony, in Copyright and the Orphan Works Issue, 29.

[6] For legislation concerning lengthening of copyright terms and strengthening of copyright protection, see Marybeth Peters Testimony, in Copyright and the Orphan Works Issue, 34-35.

[7] The American Society of Media Photographers suggests the possibility of limiting use of orphan works to institutions qualified under §501(c)(3) of the Internal Revenue Code. Perlman, in Copyright and the Orphan Works Issue, 26.

[8] Stim, Getting Permission, 46.

[9] Fishman, Stephen, The Copyright Handbook: What Every Writer Needs to Know, 13th edition, ed. David Goguen (Berkeley, CA: Nolo, 2017), 376.