Arsenic-in-the-water paper with “interesting data” first corrected, now retracted

Note (4/9/13): John McArthur contacted us with a few corrections, which we have made below.

The Journal of Contaminant Hydrology has retracted a 2008 paper by a group of Indian scientists for plagiarism and the failure to adequately reference their sources.

What makes this case somewhat unusual is that the journal allowed the authors to issue a correction (of the mega variety) attempting to acknowledge the problems, but then evidently decided that the patient was too sick to live — and that part of the disease was iatrogenic.

Here’s the retraction notice for the article, titled “Hydrogeochemical behavior of arsenic-enriched groundwater in the deltaic environment: Comparison between two study sites in West Bengal, India”:

Despite the inclusion of interesting original data in this paper, serious concerns have been raised about the manner of their interpretation in the Discussion section. Parts of papers Appl Geochem, 19 (2004), 1255–1293http://dx.doi.org/10.1016/j.apgeochem.2004.02.001 and Water Res. Research, 37(2001), 109–117http://dx.doi.org/10.1029/2000WR900270 have been plagiarized. An initial attempt to address this issue with the authors resulted in a Corrigendum but the explanations offered in this were not subjected to sufficient scrutiny, due to a combination of errors by the Publisher and the Editors-in-Chief, and were later judged to be incomplete.

About that correction. It’s a doozie:

i) Although we cite many publications from the London Arsenic Group, we did not cite Ravenscroft et al. (2001), which is an influential work in this area, at the beginning of the discussion section — we do so here.

(ii) On page 27, second paragraph (above Fig. 9), a reference was inadvertently omitted. The correct and final version should read as:
The axial intercepts reflect mineral weathering and microbial respiration processes that do not involve As mobilization (cf. Nickson et al., 2000).

(iii) On page 29, the first sentence of the second paragraph,the nature of the Bagla and Kaiser (1996) “News & Comment” publication was not made clear. The correct and final version should read as:

Based on their research findings, Bhattacharya et al. (1997), Bhattacharya et al. (1996) and previously Bhattacharya and Jacks cited by Bagla and Kaiser (1996) put forward a model that “arsenic may be released … from iron oxide[s] … through reduction [in the Bengal Delta Plain]”

(iv) Several sentences in the second paragraph, page 29, section 4, unintentionally closely follow other publications without full attribution. The correct and final version should read as:

In addition to that, there are several other geochemical processes that may result in poor As–Fe correlations:  (i) differential sequestration of arsenic and iron by the diagenetic pyrite (McArthur et al., 2001 after Moore et al., 1998 and Rittle et al., 1995) or secondary iron oxide phases (cf. Coker et al., 2006); (ii) weathering of biotite may result in increased dissolved iron in the groundwater (McArthur et al., 2001) and/or (iii) the release of arsenic during Mn-oxides reduction may incorporate into FeOOH (McArthur et al., 2004). Lastly, the residual Mn-oxides may act as a sink for dissolved arsenic (McArthur et al., 2004 after Stüben et al., 2003).

Additionally, the name of the principal author Aishwarya Mukherjee-Goswami is wrongly spelt as Aishwaya Mukherjee-Goswami.

We thank Prof. John McArthur for his assiduousness in reading the manuscript after its publication and highlighting 4 omissions of references to his papers as well providing other useful comments. We express our sincere regret to him and to the Journal editors for these unfortunate errors, which we correct here.

The original study has been cited 10 times, according to Thomson Scientific’s Web of Knowledge.

John McArthur, of University College-London, is leading expert in arsenic contamination of groundwater, and it was he who in 2011 discovered  the article while looking for data and noticed right away that it was a “cut-and-paste” job from his own studies.

McArthur — who stressed that the most of the authors listed on the article were innocent bystanders in the event — contacted the corresponding author, Debashis Chatterjee, but didn’t get anywhere.

If they had said, terribly sorry, that shouldn’t have happened … I would have dropped everything. But they didn’t.

He then attempted to contact Chatterjee’s vice chancellor at the University of Kalyani, but again hit a wall.

But not long after, McArthur says someone sent him the correction notice.

Basically it was the corrigendum that really got up my nose. It was personally and professionally offensive to me.

McArthur said that one of the authors, Abhijit Mukherjee, suggested “literally in passing” at a meeting that his group would correct the paper.

I thought that odd, but said nothing.
Later, he e-mailed me that a Corrigendum would be published.  So I waited for a pre-publication site of it, never dreaming that the authors would by-pass me and go straight to publication.
Hence my shock when the Corrigendum was published – not just because of the poor editorial control but also because the Corrigendum was not a correction at all.

Basically it was the corrigendum that really got up my nose. It was personally and professionally offensive to all of the authors of the plagiarized papers.

Eventually McArthur went to the corresponding author’s institution, and then the journal,

It was at that point that McArthur went to the journal, which moved, slowly but surely, he says, to retract.

So, why the correction? We have a guess. It turns out that the original article had appeared in a special edition of the journal, one with several guest editors. Among them were Abhijit Mukherjee and Prosun Bhattacharya, who also happen to be authors on the offending article. Did they manage to slip in a correction without the sanction of the main editors? That’s something we’ve been trying to ask the journal, but we haven’t been able to raise anyone over there yet.

McArthur, for his part, is convinced that that’s the case:

They attempted in my opinion to represent plagiarism as error, which it clearly wasn’t, and use their special guest editorship of that journal to slip something into print that shouldn’t have been there.

The region that includes West Bengal and Bangladesh, by the way, is one of the world’s best places for the study of arsenic pollution. Part of the Bengal basin, it hosts the confluence of three rivers, including the Ganges, and the region’s 240 million or so inhabitants get roughly 90% of their water from the ground. How? By digging wells. As McArthur puts it:

It’s very easy to install a well: Sink a hole 50 feet deep and install a screen and that’s your well, and that’s where your arsenic is. And it’s a huge problem.