Over the last two years I’ve been attempting to get two papers retracted, both published in separate journals run by the Ecological Society of America (ESA), the most prominent professional society of ecologists in the world. The ESA has five journals, all published by Wiley, one of the very largest science publishers along with Springer, Elsevier and a couple of others.
Around 2010, I became aware of the work of William Baker, professor at the University of Wyoming, via an email interaction with one of his (then) grad students, who was aware of a paper I had published in 2008 and contacted me with some technical questions. The topic area involves the estimation of historic American forest or woodland conditions using the tree data collected by land surveyors in the 1800s. Most of these data were collected by the US General Land Office (GLO), and thus the survey is usually referred to as the GLO, or rectangular, survey. This survey was cadastral (land parcel delineation) and massive in extent, stretching from Ohio and Florida in the east, to the Pacific states and Alaska in the west, excepting Kentucky, Tennessee, Georgia and Texas.
Whenever GLO surveyors were working in areas of woodland or forest vegetation, they were required to record the taxon and diameter of two to four trees close to each defined survey point, and the angle and distance from those points to the trees. The survey points form the corners of square land parcels of 160 acres which were the basis of all subsequent land ownership transactions. The trees are known as “bearing trees”, and were marked as such on blazes on the trees. The survey point locations, called “corners”, were marked with either posts, rock piles, a set of four trenches, or some combination thereof.
Beginning in 1921, ecologists began to use the bearing tree data and other recorded vegetation data to estimate vegetation conditions at the time of the survey, which in most cases preceded settlement and thus the major land cover changes that came with it. Just how to make these estimates, however, has been a recurrent methodological issue in the many dozens of papers that have now appeared, the most difficult problem of all being the estimation of tree density from the measured distances. Nevertheless, most of the principle issues and potential biases in so doing had been identified by the 1970s, especially the 1950s, and most authors since have avoided dubious or wrong density claims, or just avoided density estimation altogether, concentrating instead on species composition or diameter distributions, at various spatial scales.
Jump to 2011. Into the scene step Baker and the former grad student referred to, Mark Williams, with a paper published in Ecological Monographs in which they introduce a new and entirely different type of method that they claim solves the longstanding problem of accurate density estimation from bearing tree data. Ecological Monographs is the most prestigious of the ESA journals, dating to the early 1930s and ranking near the top of all ecology journal rankings, with a very strong reputation. It is designed for long and comprehensive papers that either authoritatively review a topic, present a large mass of new data and analysis from a long term effort, or some combination thereof.
Occupied by other things at the time, I was unaware of Williams and Baker’s 2011 paper for several years, as well as two or three others that Baker was sole or co-author on. I didn’t look closely at the 2011 Monographs paper until maybe five years ago. When I did, I was amazed–but not in the way that monographic work is supposed to impress you.
To begin with, it wasn’t monographic in nature at all–it was instead just a presentation of a new density estimator that the authors invented, followed by a comparison against various other existing estimators. Assuming valid methods, this would comprise a standard type of paper, not a monograph. However, the methods are not valid, not even close. They are in fact horrendous–mistaken on conceptual, biological, and mathematical grounds, showing repeated misunderstandings of the existing literature, and of the relevant mathematics, and even fundamental principles of analysis, requiring impractical field methods, and leading to completely unreliable conclusions. To explain all this would require far too long and technical of an essay. The point of this post is my efforts to get that, and a later paper by Baker, retracted from ESA journals. Which goes like this:
About two years ago I contacted the Editor in Chief of Ecosphere, Debra Peters, regarding my concerns with a 2014 paper of Baker’s. At the same time, I was carefully reading the 2011 Monographs paper, because the former paper used the methods advocated in the latter, among other serious problems. The two are among the worst articles I’ve ever read, on any ecological topic, and far and away the worst on this topic–and I’ve read a lot of them. But this would not be apparent to those without a strong understanding of both the GLO survey and the math behind distance-based density estimators. I and soon realized that both articles merited retraction. The leading global scientific integrity organization is COPE, the Council on Publication Ethics, and their clearly stated criterion for article retraction, is unreliability of conclusions. Evidence of fraud or deceit are not required: unreliable conclusions are sufficient.
I received an encouraging response from Debra Peters, who offered to work with me to get retractions, if merited. She was aware that Baker had already been the target of other published pieces criticizing various of his papers, but which did not call for retractions. She wanted to know if I had additional arguments that had not been presented by these others, and I assured her that I did, as well as much additional data that added strength to certain points made by others. I then sent her a brief synopsis of the major problems in an attempt to demonstrate this and also mentioned the problems with the 2011 paper.
However, I soon found that there was a major impediment to the retraction process. Of course, you don’t just write a letter to an Editor-in-Chief and demand a retraction; you have to formally lay out in detail all the issues, detailing the exact reasons, which is then peer reviewed, and constitutes a paper in itself. That part goes without saying. I also knew that ESA had high per-article charges for their papers, but I simply assumed that these would be waived for any article successfully demonstrating the need for retraction. These charges are not minor–roughly $2000 depending on situation–and it became clear from the correspondence that ESA intended to charge me roughly $4000 for the two calls for retraction.
Fairly stunned, I responded, straight up, that not only did I not even have the required $4000, but that, even if I did have it, that I would not pay it, nor in fact would I pay any amount at all. I bluntly told them that any such fee was unjust, that it was their mistake to publish these two articles in the first place, resulting directly from a lax peer review process that would allow any such shoddy pieces through the gate. One can legitimately ask whether it’s justified to even charge authors of new papers such high fees, much less those attempting to point out a problem of ESA’s doing. I further said that the amount of unpaid work involved in compiling and presenting the necessary arguments (massively time consuming–can’t even begin to describe it) was far more than enough of a contribution. These things were also stated in a separate letter to the Executive Director of ESA, to whom I had finally taken the completely stalled matter.
Somewhere in this timeline, Peters stepped down from her Ecosphere EiC position, which meant I would now have to start over with somebody new, and possibly unfamiliar with the situation. Meanwhile, no comment or reply of any kind had been received from the Editor-in-Chief of Ecological Monographs, Jean-Philippe Lessard, nor has any been since, which I interpret to mean that he has no intention of dealing with the matter.
The only offer made in the various discussions was that I could join ESA at their standard membership rate and then make use of a policy which allows ESA members to publish one article per year, in Ecological Applications, free of charge. Although I like that journal, this proposal I also rejected on the same basis, i.e. principle. First of all, neither I nor anyone else should have to join ESA to criticize the articles they publish; secondly the two offending articles were published in other ESA journals, not in Ecological Applications, and therefore pointed criticisms of them should obviously appear in the originating journals; thirdly, if I were to join ESA I should have the right to use the free article benefit for a paper of my origination and interest, like any other member does, not one pointing out somebody else’s incompetence.
In my letters to the two EiC’s, the ESA publications office, and the ESA Executive Director, I then specifically asked for an exemption to the article fees, which I argued, they could and should easily make an exception for, and which should be ESA policy in the first place anyway. In doing so, it became clear that they in fact had no policy at all on this issue, and were apparently going to charge high article fees of anybody who wanted to call for any article retraction.
On top of all this, and after long delays (months) in waiting for various responses, the ESA Executive Director stated that I had not presented enough information in my correspondence with Debra Peters at Ecosphere to justify a retraction. What??!! I hadn’t submitted any manuscript detailing the reasons for either call for retraction. The only explanatory piece of any kind that I had sent was the brief summary of problems that I mentioned above, so that Ecosphere EiC Debra Peters would have some idea of what my arguments were. In no way was that piece ever intended to be, or presented as, a full description of the two papers’ problems, which are quite technical and involved, nor was it in any kind of manuscript form whatsoever. Either the Executive Director didn’t understand some basic facts on the issue, or else she was blowing smoke, grabbing for excuses to dismiss the matter, I don’t know.
I could not understand why the ESA wouldn’t just make a simple exemption of fees in this case, given what I had told them and the existing criticisms against multiple Baker papers. I was puzzled by this until I realized–it’s probably not their decision to make. Wiley is the actual publisher, not ESA, and Wiley like all big publishers is in the game for profit–they very likely set all article charges and everything else financial. This would include having no policy whatsoever in place for potential authors under financial hardship, a fact which is probably best understood as the result of entities living in a world in which, to them, everyone either has a salaried academic position, a large grant, or both. The idea that someone outside the standard academic system would raise strong arguments, and not have that kind of money to throw down, probably does not fall within their awareness zone. And very possibly, the idea of having to retract any paper, much less one exhibiting evidence of possible fraud, or intentional negligence, which is a definite possibility in Baker’s 2014 paper, given that he used fraudulent historical data–well that probably doesn’t sit well at all with the desired image.
Thus completely stymied, I then thought that I might write up my arguments and put them up on the EcoEvoRxiv preprint server in the mean time, so people could at least be aware of the issues. So I then went to the ESA publications website to see if there was a policy on preprints, where I read that preprints can later be submitted for ESA publication, but only if they retain their exact, original form, without any changes. Now why should that be the policy? What if you come up with better arguments, or even just a better presentation of the original arguments? You mean to tell me that you can’t change your preprint so as to present the arguments better…or better arguments?
Nevertheless, given ESA’s position, that option is one of the few that I have. Others are (1) to present the arguments here, or (2) to submit a manuscript to a non-ESA journal discussing these problems, but framed more generally somehow, and thus giving up on retraction altogether. There is a third strategy in mind, but in battle you never signal possible intentions to the opposition. And a battle is what this now is, without a doubt.
On top of all this, it was actually never made clear in the first place, just how the publications involved in a retraction would work. Would my pieces in fact be published as full articles? Would the original articles also remain visible so that people could read and compare them against my arguments? What about the responses of the authors to my arguments, would they appear? Or, were the journal Editors-in-Chief just going to make a decision, based on my evidence but without presenting it publicly–just suddenly placing a cryptic notice stating little more than that the papers had been retracted due to “editorial concerns”. None of these questions were made clear at the ESA website, nor the journal websites, nor in my interactions, and the entire appearance of the whole thing is that ESA just doesn’t really have any retraction policy at all. Worse still, high article fees put in place so as to actively discourage calls for retraction, under orders from Wiley, is also a possibility.
So, these are some of the fun and games you might expect when you try to get even egregiously wrong papers retracted from prominent scientific journals, and how I’ve been relaxing since posting last.
Ecologically Oriented 