Copyright

Your project will probably end up on the open web, and that makes copyright a real question for this course, not a technicality. This page sets out general principles and rules of thumb. It is a starting point, not an answer.

First, a Disclaimer

None of us are lawyers. Not me, not the TAs. Copyright is ultimately a legal question, and when you really get down to it, it is a question of risk: how likely is it that someone will object to what you have done, and that a court would side with them? Nobody on the instructional team can guarantee you anything.

What we can do is offer guidance, and so can the University Library System, whose librarians can help you understand your options and point you to resources (though, as their own guide notes, they do not give legal advice either). If a question really matters, ask early: the earlier you settle on a source, the cheaper it is to change your mind.

The Public Domain

A work in the public domain is no longer protected by copyright (or never was), and anyone may use it, in full, without permission. Most projects in this course work with public domain texts, and that is the safest place to be.

In the United States, as of January 1, 2027:

Published before 1932: anything first published in the United States in 1931 or earlier is in the public domain. Each January 1, another year joins it.
Published 1932 to 1963: many of these works are also in the public domain, because their copyright had to be renewed after 28 years and often was not. Finding out takes some research.
Published 1964 to 1977: most of these are still under copyright, for 95 years from publication.
Published 1978 or later: protected for the life of the author plus 70 years, so assume they are under copyright.
Never published (letters, diaries, manuscripts): generally protected for the life of the author plus 70 years, so those by authors who died in 1956 or earlier are in the public domain.
Works of the US federal government (but not necessarily state governments) are in the public domain no matter when they were made.

The standard reference for all of this, including the many special cases, is Peter Hirtle’s chart, Copyright Term and the Public Domain in the United States (Cornell University Library).

Works from other countries are more complicated. US rules govern what you do in the United States, but a foreign work’s status here can depend on its status in its home country. The good news for many humanities projects: a medieval manuscript, a classical text, or a nineteenth-century novel is in the public domain everywhere.

The Version Matters

This is the most common trap. An old text can be in the public domain while the version you are reading is not.

Translations are new works with their own copyright. Homer is in the public domain; a translation of Homer published in 1990 is not. A translation published before 1932 is. (Or make your own, which is a fine project in itself.)
Modern editions can carry new copyright in everything the editor added: introductions, notes, critical apparatus, and sometimes editorial choices about the text itself.
Reprints do not reset the clock on the original text, but check what else the reprint added.

So always record exactly which edition or translation you are using, and when it was published.

Images and Digitized Sources

In the United States, a faithful photograph or scan of a two-dimensional public domain work (a painting, a page of a manuscript) does not get a new copyright of its own: a federal court held that such exact reproductions lack the originality copyright requires (Bridgeman Art Library v. Corel, 1999). That decision is widely followed, but it does not bind every court, and other countries see things differently.

Separately, the library, museum, or website you get an image or text from may have terms of use that you agreed to by downloading it. Those are a contract, not copyright, but they still matter. Read them, and when in doubt, prefer sources that say plainly what you may do.

Analyzing Is Not the Same as Publishing

Studying a copyrighted text computationally is one thing; posting it on the web is another.

US courts have held that copying copyrighted books into a database so they can be searched and analyzed, rather than read, can be a fair use (Authors Guild v. HathiTrust, 2014, and Authors Guild v. Google, 2015). This kind of computational analysis is sometimes called non-consumptive research. So a project might be able to analyze copyrighted song lyrics and publish its findings: counts, patterns, charts, short quotations for discussion. What it should not do is republish the lyrics in full on its website.

Fair use is decided case by case, weighing four factors: the purpose of the use, the nature of the work, how much of it is used, and the effect on the market for the original. Pitt’s Copyright and Intellectual Property Toolkit explains each. Note too that some databases and websites forbid automated downloading in their terms of use, whatever fair use might allow.

AI Tools and Your Sources

Uploading a text into an AI tool makes a copy of it. For public domain texts this raises no copyright question. For copyrighted material, the law around AI is still being worked out, so be more cautious: use the course’s university-provided tools where you can, and do not upload copyrighted material you would not be comfortable sharing in any other way. (See AI Policies for an AI Course.)

Rules of Thumb

Prefer public domain or openly licensed sources (such as material under a Creative Commons license that permits reuse). Read the license: some forbid commercial use or modification.
Know your version. Record the edition or translation, its publisher, and its date.
Publish results, not copyrighted texts. Analysis, data derived from a text, and brief quotations are far safer than full reproduction.
Respect terms of use of the archives and websites you draw on.
Always attribute. Attribution is not permission, and permission is not attribution. You need to cite your sources either way.
Ask early. Bring copyright questions to your project mentor at the start, and to the library when you need more help: Pitt’s Digital Scholarship Services can be reached at ULS-DigitalScholarshipServices@groups.pitt.edu.