How Ethical is Geoscience?
If you are a regular visitor to RetractionWatch, you see on a daily basis papers that are shown to be falsified or duplicates of earlier work or plagiarized from others’ work. And as a geoscientist, GG is often relieved to see that geoscience papers are a rare commodity over at RetractionWatch. Is this because we are as a community more pure of heart and mind, or is it that there are aspects of our science that are more capable of hiding misbehavior?
On other points, it isn’t clear that geoscientists are saints. Reading Hope Jahren’s blog, for instance, would cure many of any delusion that sexual harassment and discrimination doesn’t exist in the geosciences. GG looked to see if he could find any summary of campus sexual harassment/discrimination cases by academic discipline and failed (if you know of one, please post a link in the comments). But the nature of geologic fieldwork–long periods out in the field with small numbers of people–has long been a potential minefield for female scientists, and plenty of geoscience departments stayed “boys’ clubs” far longer than could be justified. Some geologists have been banned from supervising female students, for instance, because of inappropriate behavior in the field.
So if geoscientists are not particularly saintly on other scores, is there something about earth science that either hides or precludes academic misconduct?
Let’s consider what might tend to preclude misconduct. A lot of geoscience is based on, not surprisingly, observing the earth. And so the earth is still sitting there for somebody else to come and look at it. If you wanted to make a fake geologic map, there would be no appeal to special lab conditions, a unique strain of bacteria or cultural differences or any of a hundred other excuses possible in other disciplines. Disagreement over interpretation is quite likely (for instance, go to the south side of the Alexander Hills near Death Valley and decide if the base of the Noonday Dolomite is a fault or an unconformity), but map a granite where there is sandstone and prepare to be outed. Similarly, things like seismological records are now archived centrally; if GG makes a tomographic image of California and you want to double check things, you can access the raw data pretty easily, should you care. Falsifying terabytes of seismic records might just be way too much work to make a fake tomography.
And what of the incentives to cheat? Biology and medicine are very high profile fields at the moment. Fame and fortune can literally come to researchers who make breakthroughs in these fields. NIH is very well funded compared to the earth sciences, and sponsorships from corporations are both very common and very lucrative. In contrast, those who have made big money in earth science have generally done it by finding resources and collaborating with companies that can exploit those resources. No big oil or gold find, no big paycheck. [Probably there are some exceptions–those who have claimed some special technique will lead to a resource find only to have a failure be excused as due to some complication while they pocket a consulting paycheck. But again, you won’t get a lot for long]. Certainly there can be some who might bend the truth for some pay as an expert witness, but this isn’t going to get you too far financially (though that might change if lawsuits against oil and gas companies start to get to be more common). Neither of these latter two cases is apt to produce anything in the scientific literature and so is outside our purview here.
What might allow misconduct to go unchallenged in geoscience? Arguably the analyses of materials taken from the field provides a real opportunity. Say you found the oldest zircon and in measuring its age, you destroyed it. Could you have cooked the numbers to get a publication from a younger zircon? Sure. And if the samples are from some unique and difficult to reach locality, it would take quite a bit of interest to want to go and resample such a spot. Although geology in the field is fairly easily verified, reproducing many kinds of geologic analysis is harder.
Even so, there is a real limit to what you can pull off. For instance, there were questions about the interpretations of a signal of life in some very old rocks in a pretty obscure part of Greenland. This was a significant enough point that field trips were made out to the sampling locality to verify what was sampled and to allow others to make their own collections. If you are going to cheat, you can’t cheat on too grand a conclusion or, as in other fields, you will get found out.
From rumors GG hears from time to time, this is probably where most geoscience cheating is occurring: in more out of the way places where the results aren’t so unexpected, so publication of any results that seem to confirm existing ideas gets less scrutiny. Why do people cheat at this level? It isn’t terribly clear to GG, but perhaps it is to clear a bar on getting some number of papers published to get promoted or gain a raise, and publishing non-results can be difficult. Given the increasing pressures of publishing in order to both retain an academic job and get the next grant, it is possible that this mode of cheating is increasing but remaining under the community’s radar.
So maybe there is a low-level of background cheating. The last aspect of this is that the geoscience community may be much less interested than other scientists in retractions, instead preferring to simply move on past worthless papers. There are a lot of specific details on areas that only occasionally get studied. If you come along and are the first in 30 years to look at some rocks in detail and you find the previous work was way off base, do you worry about complaining to the journal or do you simply write a new paper saying the old guys were wrong? It might well be that the slower pace of modern geologic inquiry limits the incentive to get papers retracted.
So in the end, what’s the judgement? GG kind of thinks that geoscience literature might be somewhat cleaner than other fields, but probably not to the degree indicated by relative numbers of retraction notices. Frankly, the sad part is that any thought needs to be devoted to such behavior.