Are word lists sufficiently original to be copyrighted?

I recently pushed to GitHub a set of word lists intended to be used to generate secure passphrases called the Orchard Street Wordlists. I’m pretty proud of them – you can read more about the lists here.

The words on the list come from two sources: Google Ngram data and Wikipedia word frequency data, via this project. I essentially made a large word list by “blending” these two lists together, alternatively taking one word from each, starting with the most frequently appearing. I then cut words to make lists for various purposes.

Google Ngram data is licensed under Creative Commons BY 3.0 Unported. Wikipedia text, as of this writing, is licensed under CC BY-SA 3.0. That “SA” stands from ShareAlike, which basically means you can use it, but you have to share your work under the same or a similar license. Thus, since all of my lists include words I got from Wikipedia word frequency data, I figured that I would have to use that same license (CC BY-SA 3.0) for my word lists.

However, I’m obviously not doing what most people do with Wikipedia text, like quote a sentence or few paragraphs from an individual article. By using word frequency data, I’m doing something a bit more broad, and almost completely removing all context of each word (though it does “touch” more content). Do I really need to maintain the ShareAlike license?

Is an alphabetized word list sufficiently original?

But setting even that question aside: We should also ask if any alphabetized word list, which is what makes up my project, is original enough to even be copyrighted in the first place. I think this is a necessary condition to use CC BY (3.0 or 4.0). That all of my lists are in alphabetical order likely helps the argument that they are not original enough to copyright.

Clearly everything is not copyrightable, even under Creative Commons. I can’t write a very simple programming function that, say, lower-cases any string given to it and then license it under GPL or write a simple a sentence and offer it under CC 4.0 BY.

While most word lists I know of are not copyrighted or licensed, the EFF fandom word lists are licensed under Creative Commons Attribution 3.0 United States license (proof). However that seems to the exception. Most word lists included in code projects seem to just sit under the project’s overall code license, like MIT or GPL, etc.

While my pride might be a little hurt if I couldn’t copyright my word lists, it would make sharing the lists easier.

Next steps?

In practical terms, I think replacing the current CC BY-SA 3.0 Unported license with the CC0 license would be a proper way to show/assert/embrace this lack of originality and thus lack of copyright.

Unfortunately I just don’t have enough information to know exactly what to do. Leaving the CC BY-SA 3.0 license in place feels safe, but part of me wants to slap the CC0 on there and not think about this anymore!

Feel free to comment on this GitHub issue or find me on the Fediverse if you have thoughts!