Confusion Regarding Long Author Lists

I really don't understand the confusion that some people (like Rose) have with this. An "author" on a scientific paper doesn't have to write any text, and this is as true for a paper with 5 authors as it it is for 50. It is just generally accepted that anyone who contributed data or analysis to a paper ought to be an author. That's just being fair. Otherwise you are dishonestly claiming that you did their work. And I don't recall anyone "battling for authorships on as many papers as will accept them". Sure, there is some squabbling over being *first* author, but not just being *an author*.

And it is hardly surprising that genome papers have a lot of authors; once upon a time sequencing papers dealt with the sequence of just one gene; if anything we should be surprised that genome papers don't have the thousands of authors that would be expected on a linear extrapolation from that.

And, finally, if Rose is honestly wondering how such papers are written, my experience at being first author on a genome paper taught me that the first author writes most of the actual text (with perhaps a few paragraphs sent in by other authors scattered around the globe), but bases the bulk of the writing on the data and reports provided by the other authors, all of whom read the final draft to make sure that their results have been accurately interpreted.

You know, this causes confusion for those outside of science regularly. Given that scientists are the only ones doing it this way, it really seems rather absurd for them to be the ones whinging when no one "gets it."

Good lord, why not just respect the authorial convention by using it for the 1-3 who were actually authors, and add a contributor field for all the other work? Everyone gets proper, thorough recognition that is clear, concise, logical, properly classified, and easy to understand.

Two other situations that result in long author lists are papers involving patient data (often collected at great expense and effort from numerous clinics who all deserve some degree of acknowledgement generally in the form of authorship), and the case where a group that doesnt have a huge amount of money gets some research done by getting other groups to collaborate on the expensive experiments (for example microarray analysis or massive sequencing analysis) with the resultant accumulation of authors from these collaborators.
As you said, the papers are generally still almost entirely written by a single author - with minor revision based on comments of the other collaborators.

You know, this causes confusion for those outside of science regularly. Given that scientists are the only ones doing it this way, it really seems rather absurd for them to be the ones whinging when no one "gets it."

Rose is a scientist, so I'm complaining about a scientist not getting it.

Good lord, why not just respect the authorial convention by using it for the 1-3 who were actually authors, and add a contributor field for all the other work? Everyone gets proper, thorough recognition that is clear, concise, logical, properly classified, and easy to understand.

The problem is that in order to get credit, one must be listed as an author. Being included on an acknowledgment list means absolutely nothing. In science, the writing is only one step in the process -- compared to other fields where the writing is the entire process. The authorial convention is to award credit to those people who contributed to the intellectual developments described in the paper -- that includes people who may have done little to no writing, but played an important role in generating or analyzing the data.

In Japan, it only takes one person to sequence a genome. :-)

Yeah, many prokaryotic and microbial eukaryotic genomes can be sequenced and analyzed by a small research group (or a single person, I guess). But the larger eukaryotic projects (and, I presume, the multi-species projects from all taxa) still require large consortia.

Well said in response. Much like Dr. Rose my field (ecology) often does not have multiauthored papers but I originally did physiology and biochemistry where the papers had many authors for exactly the reason you suggest. Even in ecology though there are cases where I am one of a dozen authors (reviews by a group of researchers, 10 year long experiments where 5 grad and 5 undergrad students participated) - and in those cases, some ecologists still tend to question lists of more than a half-dozen. Admittedly, I have seen cases where an author has been put on for dubious reasons but these tend to be the exception and journals don't manage for exceptions.

My admittedly uninformed and cynical take on long author lists: If the 3rd through 20th authors had actually contributed substantial ideas and data they would have published their results in a separate paper.

Most of the authors I assume were functionally no more than technicians. An interchangable set of hands operating a pipetter shouldn't be included in the author list even if those hands were pipetting 12 hours a day. Nor should it matter if those hands belong to a grad student or post-doc. They get listed in the acknowledgements if they're listed at all.

A paper ought to be a work of scholarship. Therefore, a person shouldn't get authorship unless they contribute intellectually to the project. If 20 people had actually contributed original ideas they would have produced seven or eight papers between them rather than just one.

Well, at least in genomics it's a bit more than just listing technicians, Herb. For example, many people at TIGR/JCVI have specialties that they are known for. One person is an expert on transporters. Another is an expert on transposons, another is an expert on sugar metabolism, or phylogeny and so on. While these people have their own projects where they often write papers as lead author, it is generally expected of them that they also contribute their expertise to see if there is anything of interest in each new genome that the institute does. Each individual analysis may not be worthy of a paper in itself, but together the analyses from the various experts may make an interesting story.

As for single author genome papers, I seriously doubt that even a bacterial genome can be adequately analyzed by one person. While it's possible that Nakagawa is some sort of super genius familiar with all areas of genomics, the fact that he or she has decided to publish in an obscure non-web accessible journal (and in Japanese - unreadable by the vast majority of scientists even if they could find a hardcopy of the paper!) suggests that communication of interesting analyses was not the goal of this publication.

The problem is that the definition of "author" is: someone who writes.

As a writer myself, I have to fall in with the argument that a "large consortia" cannot all be called authors, and claim credit for authorship, if none of the words in the paper in question were written by them. "Collaborators" would seem to be the more appropriate term. You really have to do some serious redefining of the concept of authorship to do it the way it's currently done. 30 authors for a journal article? It's just nonsensical.

Just because there are no projects in your field that require manuscripts with large author lists doesn't mean that you should look at papers with more than 20 authors with disdain.

Amen.

It seems to be a truism, though, that many scientists think that the way things work in fields other than theirs (or even *sub*fields other than theirs) are all wrong, and that their own subfield's way of operating is the One True Way.

In response to Dirkh -- "authorship" on a scientific paper does not mean the same thing as it does on a newspaper article or a novel. Indeed, I've heard some who don't speak English as their native language refer people "signing" a paper rather than authoring it; it may be that other languages have terms that offer less confusion. The simple fact is that in most scientific fields, anybody who contributed meaningfully and crucially to the scientific work of the paper is a co-author on a paper. And, in some fields, that will lead to 10, 20, 100, or even 1000 authors on a paper.

-Rob

"And, in some fields, that will lead to 10, 20, 100, or even 1000 authors on a paper."
-----
1000 co-signers, anyway. As we all know, sometimes people appear as "authors" who had nothing to do with the paper at all, but are listed there for reasons of lab politics or general clout, or a dozen other reasons. (I'm no scientist but I'm married to one who publishes frequently).

When doing research, I ignore all the co-authors and just look at "to whom correspondence should be addressed," as those individuals are likely to have actually had something to do with writing the paper, rather than strictly with doing the research. It just seems to me that the distinction matters.

When doing research, I ignore all the co-authors and just look at "to whom correspondence should be addressed," as those individuals are likely to have actually had something to do with writing the paper, rather than strictly with doing the research. It just seems to me that the distinction matters.

Very often this is the case. But, often, the person to whom correspondence should be address didn't actually write the paper -- correspondence is often directed at the senior author, while the 1st author actually wrote the paper.

More importantly, doing science is more than writing up the results. To get credit for doing research, one must publish the results. Sometimes a research project involves many people and gets written up in a single paper. Maybe only a few people actually write the manuscript, but all the other people either analyzed data (possibly making a figure/table for the paper), designed the actual experiment, or collected the data -- each one contributing a certain level of expertise. Those people are given credit for the work -- beyond just the writing aspect.

I seriously doubt that even a bacterial genome can be adequately analyzed by one person.

Some genomes are sequenced for a particular type of analysis -- especially if the genome is small. For example, check out this paper with 6 authors.

As a physicist, having 50 authors can seem small. Some papers, particularly those using particle accelerators, can go into the hundreds!

@dirkh

You said "The problem is that the definition of "author" is: someone who writes.

As a writer myself, I have to fall in with the argument that a "large consortia" cannot all be called authors, and claim credit for authorship, if none of the words in the paper in question were written by them."

If you look at dictionary.com's entry for author from the american heritage dictionary you'll see a note at the end.
"The verb author, which had been out of use for a long period, has been rejuvenated in recent years with the sense "to assume responsibility for the content of a published text."

Sort of like having your name on a paper implies responsibility for the content of that paper.

I second my former colleague Jonathan Badger's comments. I've been an author on many genome papers, virtually all of them with >20 authors, and most of the names listed contributed in real, substantive ways. The people complaining here - including the original poster, Robin Rose - either don't understand how genomic science works (likely), don't understand scientific authorship in general, or have some kind of axe to grind. The quaint historical view of a lone scientist laboring away, perhaps with one or two colleagues, before writing up their discoveries, just doesn't apply to much of modern science (though it still happens).

The sequencing and analysis of a genome requires a very broad range of expertise - in sequencing technology, bioinformatics, biochemistry, evolution, and usually specialized expertise on the species being sequenced. And the *only* credit that scientists get is authorship - we don't get paid to write, and as one commenter mentioned already, a note in the Acknowledgements is worthless. So if you want a high-quality genome paper, you have to recruit experts to help, and you have to include them.

The bottom line is that someone takes the lead on every genome paper (usually the first or the senior author), and that person decides who else deserves co-authorship. It's a good model and it is working well.

"either don't understand how genomic science works (likely), don't understand scientific authorship in general, or have some kind of axe to grind."

-----------
None of the above. Scientists either don't understand how journalism works (likely), don't understand that "taking responsibility" for a paper is not the same as authoring it, or have some peculiar scientific ax to grind.

It's really quite simple: There's no way to know who actually wrote the paper--and everybody seems to think that's just fine. "Authorship", under such terms, really doesn't mean anything at all.

Scientists either don't understand how journalism works (likely), don't understand that "taking responsibility" for a paper is not the same as authoring it, or have some peculiar scientific ax to grind.

Science is not journalism. Authoring a journalistic article differs from authoring a scientific paper. This isn't the first time I've had a journalist misunderstand how scientific publications get produced (see here).

I take your point--although one could argue that science becomes journalism when science is written up (evidently by authors who wish to remain anonymous) and published in a journal.

All I've been trying to say is that there is an obvious difference between the researchers who write the words that are published, and the researchers who fax some results or figures to the researchers who write the words that are published. Why conflate the two? But enough.

And you make it seem like it's trivial to do anything other than write the paper. The majority of the work (based on time) is not in writing the paper, but generating and analyzing the data.

Hehehe! Bloated lists of authors are obviously wrong. I don't remember EVER seeing single-author paper in molecular biology. EVER.

If science is really THAT cooperative, maybe we should award professorships to groups of 2-100 people and split salary accordingly? ;)

Seriously:

Bloated lists of authors are just end result.

Big wrong is reliance on publications and citation index as almost sole measure of scientist's quality. Most scientists perhaps consider it sensible, progressive and mimicking result-oriented approach in bussiness. THIS IS NOT. Real company evaluating workers in such oversimplistic way would quickly suffer big losses. Science does, too.

Worse than run for co-authorships is death of innovative research. Einsteins and Edisons exist no more. They run for small contributions on well-trodden paths, to pimp their publication lists.

Bloated lists of authors are obviously wrong.

If bloated author lists (a loaded description if I ever read one) are wrong, then what is the remedy? How else do we award credit for the work performed on the research?

I don't remember EVER seeing single-author paper in molecular biology. EVER.

Does this count as a single author paper in molecular biology?

If science is really THAT cooperative, maybe we should award professorships to groups of 2-100 people and split salary accordingly?

Red herring. Giving people authorship as credit for work done is not equivalent to awarding tenure. As I mentioned above, you're contribution is recorded by your position in the author list. Being author #38 out of 152 won't mean much for a tenure track faculty member, but it could be important for an undergrad applying to grad school. Authorship for someone going after tenure only matters if they're 1st or last author.

My comment got waaaay too long so I placed it here instead.

This is ridiculous. There's no point whatsoever in arguing about what "author" should mean based on what it means in other fields. The situation is absolutely unambiguous: in science, an author is someone whose name appears at the top of a paper, because they played a sufficiently important role in the research being reported (exactly what sort of role can vary, and it should be clarified in a "contributions" section if anyone cares). This is simply what the word "author" means in science. Complaining about it is about as absurd as complaining that journalists often publish so-called "articles" that don't even cite their sources in a bibliography at the end.

"Good lord, why not just respect the authorial convention by using it for the 1-3 who were actually authors, and add a contributor field for all the other work?"
"The problem is that the definition of "author" is: someone who writes."
Check the dictionary. Actual writing is only one sense of the word. It also means "one that originates or creates" (Webster). The paper itself is only the final step in creation. Maybe more accurately, it's the organization of what has been created.

"When doing research, I ignore all the co-authors and just look at "to whom correspondence should be addressed," as those individuals are likely to have actually had something to do with writing the paper, rather than strictly with doing the research."
1) The corresponding author is often (perhaps even typically) more like an editor than an author in the sense Dirkh wants to use. The first author is more often who you want if you're interested in the person who actually penned the words.
2) The point of the author list is to show who contributed to the final product. If you don't want to call them authors, fine (although, it matches just fine with the dictionary definition).

"All I've been trying to say is that there is an obvious difference between the researchers who write the words that are published, and the researchers who fax some results or figures to the researchers who write the words that are published."
Perhaps not intentional, but that sounds rather dismissive of the researchers who "fax some results". The results are the important part that takes the majority of the time, effort, and money. Writing it up is just the final step.

All I've been trying to say is that there is an obvious difference between the researchers who write the words that are published, and the researchers who fax some results or figures to the researchers who write the words that are published.

Yes, the latter have obviously done the bulk of the scientific work, if they have also designed the experiment and analysed the results. (Unless the writer is the only one who knows what the heck all those results and figures mean, but in that case, the experiment design seems fucked from the start.)

In non-Anglo research groups, the person writing up the paper often is often simply the one with the best command of English. Are they automatically the most important scientists? Should scientists with poor English skills get no authorship credit?

These days some big lab groups hire professional writers to pen the papers. Thus, the only person who is NOT the author of the paper is the person who wrote the words. Which just goes to show that the process of writing (using language) is the most peripheral role in the authorship of scientific knowledge, a final touch, an afterthought, something that an outsider can do and has nothing to do with the creative process and hard labor that go into authorship.

"...the process of writing (using language) is the most peripheral role in the authorship of scientific knowledge, a final touch, an afterthought...
------

Well, that certainly explains why so many scientific papers are so poorly written.

I was under the impression that the idea of publishing a scientific paper was to communicate as clearly as possible the work that was done. But in order to do that, writing, or as it's quaintly known, "using the language," would have to be more than an afterthought.

Don't be obtuse - read what I linked above. Nobody is talking about the quality of writing which is in some cases bad in others brilliant (as it is in fiction, poetry and journalism). We are talking about what it means to be an "author" and it has nothing to do with writing or using language, it is about creating something (art, knowledge, music...) - the creaitvity and the hard work that go into making something. This has nothing to do with language and writing. An author is a creator. Whoever writes about it at the end is a scribe.

And, the idea of publishing a scientific paper is not to do creative writing, but to prevent the piece of knowledge that the authors of science have discovered from being lost forever - it is scribbling it down so others can enjoy it and disseminate it further. Hopefully the scribbling will be well done, but it is a process outside of - and something done after - the process of scientific authorship itself.

Outside the scientific community, and contrary to Coturnix' claims, authorship has a different and specific meaning. An author is someone who wrote the material in question. Leonardo da Vinci did not author, he painted the Mona Lisa. Mozart did not author, he composed A Little Night Music.

Those scientists who participated in the work that led to the paper but did none of the writing contributed. Those scientists who put words to paper are the authors. Way back when somebody made an error in terminology, and now it's time to correct that error.

Arrrrgh! This has nothing to do with scientific community. Authorship has nothing to do with language and writing in any area of life.

Authoring = creating

One can be an author of an idea, a slogan, a piece of art or a piece of music, and yes, a story. And one can be an author of a piece of knowledge. Creating or discovering something that no person has done before is authoring. Describing it post facto, by using written language, is not authoring. It is equivalent to clicking on "Save".

Tolstoy is the author of War And Piece because the plot and characters are his creations, not because he used ink and paper to write it down. He could have dictated it and the scribe would not have been the author, Tolstoy would.

Just divorce authorship from language in every sphere of human acitvity and you will start understanding what authorship is about. Then, you will understand why a 100-author scientific paper is a perfectly normal thing - some things can be created only by teams.

Way back when somebody made an error in terminology, and now it's time to correct that error.

Nonsense. First, language changes with time, and you can't retroactively declare that widespread, well-established usages are based on an initial error and must therefore be revoked. Whether you like it or not, in science the word "author" doesn't mean what you think it should.

Second, I'm not convinced that it has ever been strictly limited to your meaning. I looked up the etymology on dictionary.com (not the best source, but free and quick) and the word derives from "auctor", from the Latin verb "augere", meaning to increase. The meaning of someone who writes texts dates back only to 1380. Considering that the first scientific journals date to 1665, the scientific notion of authorship is now older than the entire notion was when science started using the term. (OK, I admit I don't know that the authors of the first scientific journal publications were called "authors", but I assume they were. Does anyone know for sure?)

"Don't be obtuse - read what I linked above."

I read it, and it didn't help. Leonardo Da Vinci, the "author" of Mona Lisa? Please. Alan is on the right track: No one I know of has ever routinely referred to Da Vinci or Mozart as "authors."

And if the meaning of author as "someone who writes texts" only dates back to 1380, that's good enough, I think.

Nobody can stop scientists from assigning a new meaning to the word, if they insist, but it's a perfectly good term, and everybody but scientists seems to know exactly what it means.

But here's my favorite: "Authorship has nothing to do with language and writing in any area of life."

Hmmm. I think I'll frame that and set it beside the books I've written.

"c.1300, autor "father," from O.Fr. auctor, from L. auctorem (nom. auctor) "enlarger, founder," lit. "one who causes to grow," agent noun from augere "to increase" (see augment). Meaning "one who sets forth written statements" is from c.1380."
~etymonline.com

In other words, the more general sense came first, the application to writing came later, not that it really matters. The use of author to mean creator is widespread outside of science, and I don't understand why scientists should not be allowed to use it in this sense. Heck, "author of creation" is a somewhat common phrase to describe the Christian God.

Authoring is not writing. Writing is preserving what was authored.

I am unaware of anyone who does NOT use the term 'author' when referring to the creator of Mona Lisa, the creator of Little Night Music, or the Creator of the Universe. I just discovered my first one on this thread and I am amazed.

Nobody can stop professional writers from assigning a new meaning to the word (in 1380 and after), if they insist, but it's a perfectly good term, and everybody but writers seems to know exactly what it means. Why try to hijack it from its usual meaning and apply it only to one's own profession? What's the emotional motivation for such a need?

Okay, fair's fair, I've been roaming the dictionaries, and I must admit that the definition of the word "author" in the way coturnix and colst want to use it is more common than I had supposed (although it is not the primary usage). For example, American Heritage, 3rd definition:

AUTHOR (NOUN):1a. The writer of a book, article, or other text. b. One who practices writing as a profession. 2. One who writes or constructs an electronic document or system, such as a website. 3. An originator or creator, as of a theory or plan.

So you guys are free to use it your way, and I'm freed from the charge of hijacking the word for my line of work.

However, I still think a division of names into "authors" and "contributors" on a paper would go a long way toward clarifying the matter of who did the writing, for those of us who would like to know. And if it's ghostwritten by a professional, I see no reason not to disclose that.

Dirkh:
"However, I still think a division of names into "authors" and "contributors" on a paper would go a long way toward clarifying the matter of who did the writing, for those of us who would like to know. And if it's ghostwritten by a professional, I see no reason not to disclose that."

You are still not getting it. When we look at the author list of a paper, we are mostly interested in who made the scientific contributions that resulted in the paper and not really in who did the actual task of writing. It's not that writing is a trivial work, but actually doing the research is what scientists spend most of their time and efforts. Who ends up writing it is relatively unimportant and we know anyway that usually the first author and/or the corresponding author do most of the writing.

To me it's all a matter of giving fair credits to those who did the work. So, it is justified that papers on experimental high energy physics or genome projects have long author lists. On the other hand, there certainly are cases when someone who did little work is added as an author for a political reason and I don't like it. Anyone remember Gerald Schatten in the Hwang Woo-suk cloning scandal?
http://doctorfreeride.blogspot.com/2005/12/authorship-matters.html
Specify constribution

Whoops. I meant add that many journals nowadays are encouraging the authors to specify their contributions to the paper (who did the experiment, who did the writing and so on) and I think it's a good trend.

What HI says. We want to know who is the author of the work, not who wrote it up, and we especially like to know what is the exact scientific contribution (idea, money, expertise, lab space, technical skills, physical labor, data analysis, graphing...) of each author of a multi-author project, which more and more journals are asking the authors to disclose.

"clarifying the matter of who did the writing, for those of us who would like to know."

Of all the things to wonder about the origins of a scientific paper, I can't think of anything of less relevance than which of the authors wrote it.

Not that it hasn't been said before: in science, the actual writing of the publication is almost irrelevant. The only thing you expect from the actual writer is that i) he/she makes the effort look interesting/important, ii) describes what has been done in a logical order, iii) makes no factual mistakes, and iv) behaves according to scientific ethics, meaning that previous work is properly acknowledged. Of all participants in a scientific effort, it is typically the person who does the writing who can be most easily replaced by somebody else.
As a matter of fact, in all of the places I have worked so far (several labs in Germany and Switzerland), the actual writing of a paper is normally done by the person who speaks the best English. It's that simple. Often, this is the last author, as he/she typically has the most experience.