Self-plagiarism, or building on previous ideas?

In our April "how can we help you?" thread, a reader asks:

I am writing a short article for an anthology. Because of the length limit, I cannot fully develop some ideas and arguments. However, I still want to include many or even most of them. Will it be appropriate if I write a separate and longer paper later that basically includes the same ideas but develop them in detail? Or will this be a case of self-plagiarism? I think see some "famous" philosophers doing this sometimes, but I wonder what would be a good rule of thumb here.

At least offhand, this doesn't sound inappropriate, but maybe it depends on how much one develops the relevant ideas or arguments in the anthology chapter? I do know that many journals require authors to attest that papers are original and haven't been previously published "in whole or in part" (or some such). So, perhaps if the ideas are fleshed out in some detail in the anthology paper, then defending them again at greater length might run afoul of this standard?  

Another reader submitted the following reply:

Why wouldn't you just write another paper on the topic, without including "the same basic ideas"? Of course you can build on what you argue in this shorter paper, but why not write a new paper on the topic?

What do you all think?

Comments

[Laurence B. McCullough]

Plagiarism occurs when an individual represents as his or her own the ideas or words of another, without attribution following accepted conventions for quoted passages and citation of their sources. On this account, self-plagiarism is a meaningless word and should not be used.

The proper classification is duplication, which occurs when one publishes what one has published before without acknowledging that one has done so. This includes using one's previous work in a longer piece or in a book. The way to prevent duplication is to include an acknowledgement that has complete information about the source, following citation guidelines from the publisher. If one does not own the copyright, one should check the publication agreement from the journal for the scope of already-granted permissions. If what one proposes to do is not within that scope, then one should obtain permission from the copyright holder. Then add, 'used with permission', to the acknowledgement.

Duplication without such acknowledgement sends the message that one has nothing new to say and is simply padding one's CV, which is reputation-destroying. In the sciences this is not acceptable, which can create very serious problems when a promotion or tenure package reaches the college-wide or university-wide committee.

Beware that publishers now routinely run submitted work through review systems to determine if the submitted work had been plagiarized or (pertinent to duplication) has been previously published by the author and may reject the submitted work even if it includes acknowledgements. Best to check with editor(s) of the volume about this matter. If duplication even in part is not allowed, then one must prepare an original piece of work, citing and building on one's previous work as appropriate.

[Caligula's Goat]

An important distinction here to add to Laurence McCullough's helpful post is between:

1. Literal copy/paste duplication

2. Writing a brand new paper that covers similar ideas

Both are bad form but only the first, in my view, is going to get you in serious trouble. What "famous" (?) folks often do, especially because they often get invited to submit to so many different things, is to take a theory and apply it to a novel topic or in a novel way.

For example, suppose I had a theory of non-ideal justice. I might write six different papers that applied that theory to specific political situations (elections, immigration, abortion, police funding, etc). These papers would all ready very similarly in the run up to the application (since they'd all need to explain and defend the same theory of justice) but, unless I'm literally committing what would usually be called plagiarism (literaly copy/paste or very light paraphrasing), then there's no problem here at all.

If, on the other hand, I write six articles that all essentially defend the same theory of non-ideal justice using the same sorts of arguments in the same sorts of ways and using similar structures then even if each article is written from scratch this looks like I'm trying to get credit for the same idea six times. Bad form. It wouldn't tank a tenure case at my university unless these were your only papers (because they would, essentially, be only one paper).

[Mickey Mouse]

Sounds fine to me. Remember to rewrite, don't copy paste. Cite your own work and mention that you build on the previous work and you should be fine.

This does not mean you will get the second work published though. Referees my say not original enough.