The evils of Open Science

The ever-reliable Jon Tennant brings our attention this week to Philip Mirkowski’s recently published essay on the evils of Open Science (capitalized here for dramatic effect). At first, I found the paper baffling. The first baffling aspect is that it is a pay-walled paper about open science.

To summarize his complaint: Open science equals neo-liberalism. I have never been accused until now of being a neo-liberal, and I must admit that my initial reaction was sub-optimal. But as I calmed down toward the end of the rather long text, I figured out that I actually agree with much of his critique, if one key assumption is given: that open science is driven, and has already largely been captured by, big business. That is a valuable warning.

Last week, I declared that the business model of traditional academic publishers is dead, but Mirkowski is quite right in warning us that they are working on a new one, and that we are again their target. They successfully hijacked our labour and monetized research already paid for by the public in their traditional model by inserting themselves as intermediaries in the academic process. They seek to do the same in the new world of the internet in which, by their own admission, they have been disintermediated.  Their pattern is to capitalize on what we do, whatever it is. And there certainly are ways of capitalizing on what open science advocates want to do.

Setting the stage

Mirkowski starts by listing what I see as some of the good things about the open science movement, but which he clearly considers are useless fads at best, and existential threats to academia at worst. “For some, [open science] denotes mere open access to existing scientific publications; for others, it portends a different format for future scientific publication; for yet others, it signifies the open provision of scientific data; for others, it is primarily about something like open peer review; and for still others, the clamor for openness purports to welcome the participation of non-scientists into the research process, under the rubric of citizen science”. It does other things that Mirkowski considers very bad, such as promote radical collaboration (to which I will return later).

He then gives a moderately not-too-strawpersonish list of problems that allegedly plague the old scientific order, and that open science is out to fix: Distrust of science by the population, the slowdown in scientific productivity, increasing retraction rates and the reproducibility crisis.

I agree with him that opening up science is not enough to restore public trust. Only a change of attitude by the current keepers of knowledge will do that. As he shows, the existence of an actual slowdown is very doubtful, but at least in theory, getting more people involved in the scientific process could speed things up. There are currently many people whose participation is hindered or completely prevented, mainly scholars from the periphery, those with little or no institutional support, and interested citizens with limited access to university resources. Open science, however, is exactly the right tool to address any reproducibility crisis. Even if there is no crisis, open science, by definition, helps reproducibility.

The real problem

But now we get to the more substantial and interesting part of his critique: “The notions that any of these open science initiatives exist to render scientific knowledge more accessible to the general public and research more responsive to the wishes of the scientist turns out to be diversionary tactics and irrelevant conceits“. He means, of course, and lists, mostly commercially backed open science initiatives and tools. He does acknowledge things like arXiv and its derivatives, but seems to consider that they cannot work as intended when they are scaled up. I keep telling myself that he doesn’t have a point on this.

“There is a logic”, her argues, “to platform capitalism: Radical collaboration deskills the vanishing author, dissolving any coherent notion of ‘authorship’ (Huebner et al., 2017), and tends inevitably toward monopoly, in the name of profit”. And later, in the midst of all this radical collaboration enabled by open science, “the ‘author’ has entirely disappeared and there is no finality to ‘publication’. All you have is one big blob, like some 1950s sci-fi nightmare”. Again, I believe these are fundamentally positive developments, but then, I also think 50s sci-fi is one of the great contributions of western civilization.

Final publication is an artefact of the paper journal and book era. It was necessary back then for purely logistical reasons, and certainly not for scientific ones. Thankfully, we can now do without it. Academic authorship is indeed an illusion, created by the need for professional structures. One person’s contribution grades into another’s, and it should. The more contributions we have access to, the better. Some people move the conversation faster and further than others. That is their contribution. ‘Papers’ and ‘books’ are only waypoints and never destinations.

The ideal situation is collective ownership of academia’s overall publically funded contribution. Ownership by “the author” is a distant second, and not without its problems. Ownership by commercial entities is potentially disastrous, but as Mirkowski argues, an outcome favoured by current open science ideas. I agree that we have to very actively guard against it and set up mechanisms to prevent it, while enjoying the benefits of a more open science.

There is, Mirkowski warns us, the danger of an Academic Turk system. “Neoliberal science disparages scientists who remain in the rut of their own chosen disciplinary specialty or intellectual inspiration; what is required these days are flexible workers who can drop a research project at a moment’s notice and turn on an interdisciplinary dime, in response to signals from the Market“. Again, my own view is that open science allows us to do these very good things: to turn from one subject to another, or to stay in one, and to be as interdisciplinary, or not, as we want to be.

These are all fine ideas, until I consider them in the commercial, for-profit context that Mirkowski assumes is in the process of recapturing the academic enterprise. Once again, the danger does not come from the internal actors in science (I feel Mirkowski does not properly acknowledge this). The people generating the open science revolution, and those producing the material, have no intention of misleading us into an Amazon-driven Brave New neo-liberal research Uberotopia. That danger comes from outside. It comes from the same forces that took control of the academic publishing process under the traditional journal system.

Research can be profitable. It generates value. We have to ensure that the value generated by publically funded research goes to the public and not to neo-liberal interlopers. Again. Given academia’s track-record on this, perhaps Mirkowski is justified in his anxiety.

Because of its tone and presentation, it is too easy, perhaps, for open-science advocates to dismiss Mirkowski’s paper as the elucubrations of some grumpy old crank, left behind by the changing world around him, and disregard it entirely. I think that would be a mistake. I still believe that open science is better than the alternatives, but I have revised my threat assessment.

References

Huebner B, R Kukla, E  Winsberg 2017. Making an author in radically collaborative research. In: Boyer T, C Mayo-Wilson, M Weisberg  (eds) Scientific Collaboration and Collective Knowledge: New Essays. Oxford: Oxford University Press, 95–116.

Mirkowski P 2018. The future(s) of open science, Social Studies of Science 48:171-203.