This morning an e-mail came in from the “Science Publishing Group”, a call for “Editorial Board Members, Reviewers and Paper” for their open access journals, advertised as
Full peer review: All manuscripts submitted to our journals undergo double blind peer review.
Fast publication: Fast peer review process of papers within approximately one month of submission.
This included a special deal on the “Article Processing Charge”: \$70 or \$120 before May 15. I’ve been highly suspicious of all “author pays” open access schemes in math or physics, so I decided to check into what this one was. When I went to their web-site and looked at their list of journals, the first on the list that looked like it would have material in it I would know something about was the American Journal of Modern Physics. The first paper that showed up on the journal web-page was MSSM Neutral Higgs Production Cross Section Via Gluon Fusion and Bottom Quark Fusion at NNLO in QCD by Tetiana Obikhod, so I took a quick look at it.
It looked perfectly competent, but oddly it wasn’t on the arXiv, and the only papers by that author on the arXiv appeared to be some papers on F-theory and D-branes from 1997-98. A little bit of investigation quickly showed that much of the paper was plagiarized from elsewhere, including at least a 2003 paper by Harlander and Kilgore, Higgs boson production in bottom quark fusion at next-to-next-to-leading order and a 2011 paper by Bagnaschi et al. Higgs production via gluon fusion in the POWHEG approach in the SM and in the MSSM (neither of which are listed in the references).
For instance, the AJMP paper introduction has
In the Standard Model the gluon fusion process [12] is the dominant Higgs production mechanism at the LHC. The total cross section receives very large next-to-leading order (NLO) QCD corrections, which were first computed in [13]. Later calculations [14, 15] retained the exact dependence on the masses of the top and bottom quarks running in the loops. The next-to-next-to-leading order (NNLO) QCD corrections are also large, and have been computed in [16]. The role of electroweak (EW) corrections has been discussed in [17]. The impact of mixed QCD-EW corrections has been discussed in [18]. The residual uncertainty on the total cross section depends on the uncomputed higher-order QCD effects and on the uncertainties that affect the parton distribution functions (PDF) of the proton [19].
while Bagnaschi et al. has
In the Standard Model (SM) the gluon fusion process [4] is the dominant Higgs production mechanism both at the Tevatron and at the LHC. The total cross section receives very large next-to-leading order (NLO) QCD corrections, which were first computed in ref. [5] in the so-called heavy-quark effective theory (HQET), i.e. including only the top-quark contributions in the limit mt → ∞. Later calculations [6, 7, 8, 9, 10] retained the exact dependence on the masses of the top and bottom quarks running in the loops. The next-to-next-to-leading order (NNLO) QCD corrections are also large, and have been computed in the HQET in ref. [11]. The finite-top-mass effects at NNLO QCD have been studied in ref. [12] and found to be small. The resummation to all orders of soft gluon radiation has been studied in refs. [13, 14]. Leading third-order (NNNLO) QCD terms have been discussed in ref. [15]. The role of electroweak (EW) corrections has been discussed in refs. [16, 17, 18, 19]. The impact of mixed QCD-EW corrections has been discussed in ref. [20]. The residual uncertainty on the total cross section depends mainly on the uncomputed higher-order QCD effects and on the uncertainties that affect the parton distribution functions (PDF) of the proton [21, 22, 3].
In the body of the AJMP paper, for example starting at the bottom of page 3 with
The subprocesses to be evaluated at the partonic level are given as following…
the following material in the paper including the equations is an edited version of Harlander and Kilgore, starting at their page 4 with
The subprocesses to be evaluated at the partonic level are given as following…
As far as I can tell without spending more time on it, the author did run some kind of package to calculate something (the plots in the paper aren’t in the older papers), and then wrote the surrounding paper largely by plagiarizing the other two papers. There’s a good reason this one isn’t on the arXiv: they now run an automated system which would have immediately identified the plagiarism problem.
It’s possible that I just got unlucky, that there was a problem only with the first of the papers I looked at, but this seems unlikely. I realize that this is a very obvious case of a journal with extremely low standards, run to make money off of the increasingly popular “author pays” model of financing journals, but I’m hoping that those that are trying to move high-quality journals to this model are seriously thinking through the issues involved. Just this month in the AMS Notices, there is discussion of a proposal to move two of the AMS journals in that direction. Yes, this is very different than AJMP, but there’s an argument to be made about the “author pays” model that it is best avoided, since it’s a good idea to keep academic and vanity publishing strictly separate endeavors.
