Would a publisher accept a completely nonsensical manuscript if the authors were willing to pay Open Access publication charges?  After being spammed with invitations to publish in Bentham Science journals earlier this year, I decided to find out.

Using SCIgen, a software that generates grammatically correct, “context-free” (i.e. nonsensical) papers in computer science, I quickly created an article, complete with figures, tables, and references.  It looks pretty professional until you read it. For example:

In this section, we discuss existing research into red-black trees, vacuum tubes, and courseware [10]. On a similar note, recent work by Takahashi suggests a methodology for providing robust modalities, but does not offer an implementation [9].

The manuscript, entitled “Deconstructing Access Points” was submitted on January 29th, 2009, to The Open Information Science Journal (TOISCIJ), a journal that claims to enforce peer-review.

The manuscript was given two co-authors, David Phillips and Andrew Kent.  Any similarity to real or fictitious, living or dead academics is purely coincidental, as was their institutional affiliation: The Center for Research in Applied Phrenology based in Ithaca, New York.  If the acronym didn’t reveal the farce right away, phrenology is the pseudoscience of reading personality traits from the lumps on one’s head.

Bentham confirmed receipt of my submission the very next day (January 30, 2009).  Nearly four months later, I received a response — the article was accepted.  The acceptance letter read:

This is to inform you that your submitted article has been accepted for publication after peer-reviewing process in TOISCIJ. I would be highly grateful to you if you please fill and sign the attached fee form and covering letter and send them back via email as soon as possible to avoid further delay in publication.

The letter was written by a Ms. Sana Mokarram, the Assistant Manager of Publication.  She included a fee schedule and confirmation that I would pay US$800, to be sent to a post office box  in the SAIF Zone, a tax-free complex in the  United Arab Emirates.  The manuscript was subsequently retracted:

Dear Ms. Mokarram,
I’m afraid that we have to retract this article.  We have discovered several errors in the manuscript which question both the validity of the study and the results.

I have yet to receive a response.  What is surprising is that the assistant manager claimed that the article went through peer-review although there is no evidence that it actually did.  Anyone with English proficiency — with or without a degree in computer science — would recognize that this manuscript makes absolutely no sense.  Had it gone through peer review, I should have received reviewer comments.  If you are skeptical that I might be misreading the response of someone whose first language is not English, I clarified the decision in a previous email with the simple question, “Does this mean that our manuscript was accepted for publication?”  Her answer was the above quote.

From this one case, we cannot conclude that Bentham Science journals practice no peer review, only that it is inconsistently applied.  Earlier this year, I reported on a case in which a nonsensical article submitted to another Bentham Science journal was rejected after going through peer review [1].

While one should be careful not to generalize these results to other Open Access journals using similar business models, it does raise the question of whether, at least in some cases, the producer-pays-to-publish model may unduly influence editorial decision-making.  One may also question whether publishers like Bentham see a lucrative opportunity from the OA movement, considering that academic libraries are establishing author publication funds to pay Open Access charges.

Notes:

[1] Adventure in Open Access Publishing (March 12, 2009). The first manuscript, “A Study of Wide-Area Networks” was submitted to the Open Software Engineering Journal.  No other submissions to Bentham journals were made.

Reblog this post [with Zemanta]
Phil Davis

Phil Davis

Phil Davis is a publishing consultant specializing in the statistical analysis of citation, readership, publication and survey data. He has a Ph.D. in science communication from Cornell University (2010), extensive experience as a science librarian (1995-2006) and was trained as a life scientist. https://phil-davis.com/

Discussion

72 Thoughts on "Open Access Publisher Accepts Nonsense Manuscript for Dollars"

Well done! I have had my doubts about that journal for quite some time. Actually, I have long ago withdrawn from the so-called editorial board because I felt something was wrong. I am actually a bit shocked to find out that my name still figures on the list!!

Thanks!!!

This is one example of why the “pay to publish” model has to end. Even top journals like PlOS can justify their fee schedule all they want, but at the end, pay to publish journals simply shift the revenue stream from institutions and readers(libraries and organizations) to the actual researcher, creating an incredible conflict of interest that compromises quality. Nestor.

As Peter Suber reports (http://www.earlham.edu/~peters/fos/2009/06/hoax-exposes-incompetence-or-worse-at.html), Bentham’s behavior has been questioned in the past. See, for example, this thread from liblicense in April 2008.

http://www.library.yale.edu/~llicense/ListArchives/0804/msg00027.html

I think it is possible to set up safeguards to avoid this conflict of interest. Let’s not paint all OA publishers with the same brush.

On a lighter note, $800 seems like a bargain compared to other journals’ author fees!

My last paragraph clearly illustrates that I’m not painting all OA publishers with the same brush.

This case study suggests that libraries need to be careful when establishing OA author funds and establish some guidelines to prevent abuse of the institution’s goodwill.

Is $800 a bargain? I guess it depends on what you get. Caveat emptor!

Nice job exposing this. In the end, will people read or subscribe to content which is pure jibberish (as in this case) or even just content that is not good quality? Will they cite it? Of course not. And who is going to submit to journals/publishers who publish such stuff? “Bogus” journals will find themselves without quality submissions and without readers. Users will gravitate toward quality and IMO the system will (hopefully!) “police” itself.

Phil Davis Replies:
Thank you. There are several self-correcting mechanisms in science (this case study is but one of them). Remember that journals serve author interest as well as reader interests. There is a long tail of subscription journals that exist because they fulfill the need for second-tier faculty to publish their articles and to sit on editorial boards. The vast majority of the articles published in these journals are neither read nor cited.

Let’s rework this a little based on the Sokal affair and see if it flies:

While one should be careful not to generalize these results to other subscription-based journals using similar business models, it does raise the question of whether, at least in some cases, the subscription model (where publishers charge higher subscriptions the more papers they publish) may unduly influence editorial decision-making. One may also question whether publishers like [chose a subscription publisher at random] see a lucrative opportunity from the subscription movement, considering that academic libraries have established massive acquisition funds to pay for subscriptions.

Good question! What moderates the subscription model are economically conscientious individuals we call “librarians”, who are willing to cancel titles when their cost greatly exceeds the value to their institution, and who refuse to subscribe to titles they deem as poor quality.

Can you imagine a librarian (responsible for managing an OA author payment fund) writing to a faculty member:

“Dear Professor, we have deemed that the journal in which your article has been accepted is complete rubbish and are therefore denying you reimbursement for your open access author processing charges”

Librarians have no say where a faculty submits his/her manuscripts (this is contrary to academic freedom), but does have a say in whether the library is willing to purchase a subscription. In that sense, the subscription model empowers librarians to become responsible stewards of the institution’s funds. While they do not exert complete control, it is far better than an alternative model where librarians merely foot author’s publishing bills.

I’m still a student but from the literature I’ve been reading it seems that the power you describe for librarians disappeared some years ago when academic libraries were pressured into all-or-nothing big bundle deals. Costs that exceeded value just meant consortium deals and cancelling subscriptions to smaller journals. In that sense the subscription model no longer empowers librarians it binds them.

I can indeed imagine a perhaps more diplomatically expressed letter with the gist you propose. If as you say a librarian might have responsibility for managing an OA author payment fund (I haven’t ready about this yet) I cannot imagine it being without criteria in place for how that would be done; not least of which would include evidence of the appropriateness of potential intended journals.

Academic freedom involves freedom to express ideas, not to be fiscally irresponsible. If a faculty member cannot, even with a librarian’s assistance find an appropriate open access journal willing to publish their paper, then an institutional repository or self-archive would still allow the freedom of expression without wasting money.

Sorry, Phil, I should have been clear. My “broad brush” comment was a reply to Nestor’s, comment.

I still think that you can not avoid such conflicts of interest under a pay to published model. And I feel strongly that such a model needs to end – even among top OA journals such as PLoS.

Here is the problem, under the traditional model… the revenues of the journal were directly linked to the journal’s ability to publish quality research. As quality increases so individual and institutional subscribers.

Instead, under the pay-to-publish model, the revenues of the journal are linked directly to manuscript acceptances and largely (although I admit not entirely) independent of the quality of the accepted manuscripts. All pay-to-publish journals get paid each time a manuscript is published regardless of the quality of the manuscript. And we have not even began to talk about how Pay-to-publish models discourages and discriminates against smaller researchers at liberal arts colleges (despite their efforts to waive or reduce fees for certain researchers).

But this is not an issue of OA. It is an issue of Pay to publish. There is a reason this model was abandoned by most fields many decades ago: it simply is not the best model to distribute the best science. OA simply needs to find a different model (pure ad revenue, charging for institutional access, grants, etc etc). Nestor.

Excellent stuff!

One criticism though. Context-free does NOT mean nonsensical — it refers to the kind of formal generative grammar model SCIGen uses to produce syntactically-correct text.

Context-free grammars (CFGs) are also used in, amongst other things, parsers for computer languages and human languages (although real human language isn’t always context-free).

http://en.wikipedia.org/wiki/Context-free_grammar

The type of grammar used to parse or generate text only tells you about the syntactic structure (grammaticality) of that text and nothing about its meaning.

Although you might refer to SCIgen’s output as ‘content-free’ 😉

I think the phrase should have been “content-free”, which is the document engineering term for “rubbish”.

I warned about this particular publisher before (http://gunther-eysenbach.blogspot.com/2008/03/black-sheep-among-open-access-journals.html).
In the meantime, we have created the Open Access Scholarly Publishing Association (OASPA). Members commit to peer-review and other quality standards. Bentham is not a member of OASPA.

Could you repeat this experiment by sending fake papers to small subscription-based journals which are desperate for submissions? Those have “pre-sold” journals to subscribers, and are now in a contractual obligation to fill pages – wouldn’t there also be just the same incentive to just accept papers, bypassing thorough peer-review?

You frame this as if “pay for publication” is more prone to abuse than the traditional model, but this isn’t what you have shown, because you didn’t have any controls.

Good point. Publishers who promised subscription agents and libraries, say 4 issues, would merely combine issues.

But this was a dangerous practice because it sent out a very bad signal to librarians who would consider canceling the title for the next subscription year.

Similarly, filling issues with conference abstracts, republished articles and feschrifts would also ring an alarm for anyone who has control and responsibility for their collection budget.

Where do you get the notion that this is an open access publisher? Just on their own say-so?

This seems to be just another shady operation – the tip-off is the spam emails encouraging you to submit.

Check out the “Current Psychiatry Review” at http://www.bentham.org/cpsr/index.htm. It says: “Indexed in:
Chemical Abstracts, EMBASE, Scopus, PsycINFO, EMNursing, Google, Google Scholar, Genamics JournalSeek”. Indexed in Google? Who isn’t?!

Now go to Chemical Abstracts and search for the journal. Find anything? Check out Google – 24 hits. Who quotes this, what is its impact factor?

There are a lot of fake conferences and fake journals out there, making money off the pressure to publish.

To be fair, you need to submit the same paper to an ACM journal and other computer science journals. I am willing to bet that the paper will be accepted at one or two journals that don’t profess open access. You should check Mahoney’s work on peer review biases: Mahoney, M. J. (1990). Bias, controversy, and abuse in the study of the scientific publication system. Science, Technology, & Human Values, 15, 50-55.

“… the assistant manager claimed that the article went through peer-review although there is no evidence that it actually did … Had it gone through peer review, I should have received reviewer comments.”

What if journals worked with open peer review – all peers need to make comments open for all, and signed. Would be harder to pretend you journal is peer reviewed, if it is not.

The really disheartening thing about this process is that some peer reviewers might not be capable of recognizing gibberish, because that is what they themselves produce.

Back in grad school I had a professor (chair of the department) who gave a required course called something like “Cognitive Behavioral Approaches.” Sounded normal enough.

The first day in class I realized with a start that this individual’s lectures sounded exactly like a guy I’d heard (perhaps on the Ed Sulliven show?) decades before, who could speak gibberish fluently and with marvelous inflection.

How to take notes?!

Well, our entire grade was to depend on a single essay at the end of the course, with the question to be provided one week before the deadline.

So I began jotting down, verbatim, every golden phrase that issued from the professor’s, uh, mouth. Sure enough, the essay question made no more sense than the lectures.

I cobbled together a 5 page report that consisted entirely of the professor’s own meaningless phrases, with elements of the essay question sprinkled here and there… and earned an A+.

I bet if this professor were one of the reviewers for that distinguished journal, he could have written peer review comments indistinguishable from the gibberish article itself.

Come to think of it, so could I!

Excellent job! Expect to get a fair amount of ranting responses from conspiracy theorists at some point, as one of the 9-11 “Truthers” published a paper in one of Bentham’s journals. You’ve popped their balloon.

Actually 9/11 Truthers have published 2 papers in different Bentham Open Journals (BOJ’s) . In both cases the editor-in-chief said he or she had not seen the paper before publication. This seems to SOP at the company, 2 different scientists posted the e-mails inviting them to be E-in-C’s of BOJ’s and other than their names and areas of specialty the text was identical including, “You will not be expected to process any submitted manuscripts to the journal nor referee them”. In one of those cases they offered the top job at their Demography journal to Dr. Lisa Wynn, an anthropologist. And not to sound like an academic snob (I only have a BA) but Dr. Wynn was (and still is) an associate lecturer who had only received her doctorate 5 years earlier. There have been numerous reports of them offering positions on editorial boards to people outside their areas of specialty. They even spammed at least one PhD candidate to be a board member.

The review process is run by people in the publisher’s office. The same pairs of people are responsible for 8 – 9 journals in unrelated fields. One (pair) for example is responsible for running the “peer review” of journals covering Addiction, Chemical Physics, Numerical Methods and Urban Studies among others. How can they possibly be able to choose competent reviewers in such divergent feilds? We they the same people who invited a fish Endocrinologist to an editor of an Anatomy journal, or an economist a position with an Education journal? I imagine they use the referees indicated by the authors. Although they nominally use the anonymous peer review the nominal lead author of the 2nd “truther” paper acknowledged he knew both reviewers.

Most (all?) of the above is documented in posts I made on the following forum page
http://forums.randi.org/showthread.php?t=110489&page=14

Lenbrazil

You could also have published the same paper for free, in a regular journal.
A similar case occurred two years ago to a traditional scientific journal of a traditional publisher, but that attracted much less attention than this.
In 2007, Elsevier’ journal Applied Mathematics and Computation (Impact factor: 0.821, rank 61 out of 165 in its ISI category, publishing since 1975) accepted a paper entitled “Cooperative, compact algorithms for randomized algorithms”, by Rohollah Mosallahnezhad (DOI link: http://dx.doi.org/10.1016/j.amc.2007.03.011 ).
The paper was as well generated with SCIgen;
after some online discussion, the article has been retracted, as you can see now on the journal web site.
However, if you would like to look at the original, is still available here: http://ce.sharif.edu/~ghodsi/soft-group/misc/AMC-paper.pdf .
For some reason, this case had much less follow-up, and did not drive to immediate step down of editors, which is a scientifically honest reaction to such kind of accident. As this is not open access, not even online publishing, and so, perhaps attention about quality should be put elsewhere.

Thanks to Vincenzo (VDM) for drawing attention to the previous experiment with a previous case of a SCIgen generated paper, accepted by an Elsevier toll-access journal, which is widely ignored in media reports of the Davis/Anderson experiment.

It is bothersome that Phil Davis and Kent Anderson – although they know better – frame this as an issue of open access journals, not citing previous studies which used SCIgen to uncover inadequate or non-existing peer-review in a number of other venues, INCLUDING subscription journals, conferences, etc. (see SCIgen blog).

Phil Davis’ statement that “producer-pays-to-publish model may unduly influence editorial decision-making” needs to be balanced with the statement that preselling subscription-journals to subscribers promising a given number of issues per year and then creating subsequent pressure on the editor to fill those presold pages, and to pay him on a per-article/per-issue basis (which is the practice at many toll-access publishers) poses exactly the same risks to editorial integrity – if not more, because at most OA journals the editor is actually not paid at all, and there is no pressure to publish a given number of articles per time-period.

Perhaps somebody really needs to start sending the Davis/Anderson/SCIgen manuscript to a number of subscription-based low-impact journals to make this point.

I had to review a paper for a fairly high ranked economics journal. The author(s) (mistakenly) submitted two papers, which were virtually identical. I notified the journal, who at that time were more interested in obtaining payment from the author than checking the submission package. The paper was withdrawn.

Much to my suprise, one version of this paper appeared at a Berkeley OJ outlet. I have brought this issue to the attention of the Editor but do not expect much.

So the bottom line is: if you want to publish crap, you can. There is nothing holding you back. The open journal concept just made that whole process a lot easier.

This is for Phillip Davis. Pacakage it as anything you want, but what you did is scientific dishonesty. I hope the scientific community will remember you as the “Author of a bogus paper”, and will treat your future manuscripts accordingly: with a lot of suspicion.

This was hardly a bogus paper – he retracted it before publication, and has widely reported his findings. How else could you check whether journals reviewed?

http://paleopolis.rediris.es/cg/CG2009_A04/
Starting in 1996 and for almost a decade, M.M. Imam contributed to twelve papers published in international geological journals [Note: none of these peer-reviewed journals are open access!]. These papers dealt with the micropaleontology and biostratigraphy of Cretaceous to Miocene series from Egypt and Libya. They were abundantly illustrated in order to support the author’s findings and interpretations. However most photographic illustrations (189 at least) were fabricated with material lifted from the publications of other authors, commonly from localities or stratigraphic intervals other than those indicated by M.M. Imam.

I disagree “In this section, we discuss existing research into red-black trees, vacuum tubes, and courseware [10]. On a similar note, recent work by Takahashi suggests a methodology for providing robust modalities, but does not offer an implementation [9].”

It’s far more intelligent than a lot of gibberish that’s published !

No computer has ever been designed that is ever aware of what it’s doing; but most of the time, we aren’t either. Attributed to Marvin Minsky

Nice investigative work, Phil. Does anyone have evidence as to whether or not TJPLI is bogus: Tiberian Journal of Pharm and Lifescience Innovatives (TJPLI)?

No…. seems its not a bogus one. They have mentioned about ISSN, INDEX, IMPACT and everything within the website. Seems its upcoming journal in the field of Pharm Science.
Also they have quoted in the Journal website that ” TJPLI is an open access non profit journal which means that all content is freely available without charge to the user or his/her institution. Users are allowed to read, download, copy, distribute, print, search, or link to the full texts of the articles in this journal without asking prior permission from the publisher or the author. The Journal never charges any handling fees, publication fees or any similar from author or authors institution”. I don’t think its a bogus one..

While Browsing I found there two journals in computer science and also don’t reveal their identity but publishes papers only for money and I submitted one scigen and got approved.
International Journal of Computer Science and Information Technologies http://www.ijcsit.com
International Journal of Computer Technology and Applications http://www.ijcta.com
Beware of these two journals.

Comments are closed.