Scientific writing + Plagiarism in different disciplines

So I read through the manuscript that I was asked to review, and I've decided to go for it. It's not directly in an area with which I have tons of experience, but I know enough about it in a general way that I've been able to do some background reading and browsing through the references to get up to speed. If this were a longer paper, one with more technical difficulty, or for a higher level journal, I'd ask for a rain check. But I've decided that I can do this, and it will be a learning experience, and I have to start somewhere. I'm up to the challenge. And hopefully I'll be able to get a reality check/advice from either my advisor or a good friend and post-doc in our lab.

I'm only about halfway through an intensive reading of the manuscript, and it's already raised some interesting challenges and presented me with issues I've never thought about before. I'm feeling more and more like I'll actually have some useful advice and critiques to share. I have suggestions already for improving the intro, extending and clarifying the analysis, and even on a technical aspect of a certain statistical test/related assumptions.

One aspect though is particularly tricky; to summarize (while maintaining confidentiality)...

My big question for this post is: What constitutes plagiarism in scientific literature? (I've heard more than one person refer to the Supreme Court quote "I know it when I see it" during discussions of this topic). I have some previous experience working as a peer tutor for college level english courses. As you might expect, as a discipline, English/Literature seems to me to place much more emphasis on particular sequences of words or phrasings as being the intellectual property of an author. To avoid plagiarism we advised students to substantially paraphrase and reword similar sentences, or to use a cited, direct quote employing lovely punctuation: the ever popular " and ". In this paper I'm reviewing, after looking up references, it has become apparent to me that entire sentences can be traced to previous articles by different authors, almost verbatim (usually with the omissions or substitutions of only a couple of words).

Now I'm left to wonder how big of a deal this is. It makes me uncomfortable because I know it would be a problem for the english students I used to advise, but this is a different context. Scientific writing is necessarily more precise, and often there are fewer alternative ways of saying the same thing. The emphasis in science is on the creativity of research ideas and methods, the ideological content that is conveyed by language, and less so on the creativity of language used for communication.

Further muddying the waters, after looking at several of the citations, its clear that existing papers in the literature also use very very similar language when describing this particular kind of model, although usually these are multiple papers by the same author (not the same group as the one behind the paper I'm currently reviewing, and in better journals by and large). Is it ok to plagiarize yourself? If other papers in the literature use such identifiably similar language, is this standard for the discipline and I just haven't paid close enough attention before?

I think that a really important aspect of plagiarism is intent. Intentional and willful plagiarism without appropriate citations is a much more clear cut situation in my opinion - it's flat out wrong and demands serious consequences. Unintentional plagiarism is more grey, and I think requires more subtlety in its resolution. In this particular case, it's likely that the authors don't speak English as a first language, and there's no obvious intent to disguise the source of the models and ideas as citations are given. If direct quotes and quotation marks were used, I wouldn't blink (except for how rarely direct quotes seem to show up in journal articles). The wording is just soooo similar. Some journals have (relatively) clear policies on this, but not the one I'm reviewing for.

I'd love to hear what you all think... Are standards for plagiarism different in different disciplines? Should they be? And, (this is the part I'm stuck on right now), what is the appropriate role of a reviewer in this kind of situation? What kind of accommodations, if any, should be made due to linguistic and cultural differences in the peer review and publication process, especially as research becomes more and more international?

In upcoming posts I'll say more about this process as I work through it and figure out more about the role of a reviewer.

Theo out.