Thursday, 25 July 2013

Rosalind Franklin, Social Science

Today's google doodle honours Rosalind Franklin, who was born 93 years ago. To me it's one of the better doodle subjects, as it draws attention to the way women's contributions often get overlooked. Franklin is a famous case; there are many others we don't know about.

So, a quick summary: Franklin was working as a research assistant at King's College London. While there she applied x-ray crystallography to DNA in an effort to deduce its structure. X-ray crystallography isn't like taking a picture; the crystals have to be prepared carefully and the pattern that comes out requires a fair bit of interpretation and deduction to figure out the actual structure of the molecule, particularly in an era without easy access to computers. Franklin produced the best x-ray data on DNA in the world, and was using it to build models of DNA. Watson and Crick were also building their own models at Cambridge, using Franklin's data. But the interaction was one-way; Maurice Wilkins, another research at King's College, was showing Franklin's data to Watson and Crick (without her permission), who used it to build their famous model, but Franklin was excluded from the conversations that her male colleagues were having and so was building her DNA model largely on her own. Watson and Crick published their famous paper proposing the double helix, which minimized Franklin's contributions, and later Watson wrote a book that further minimized Franklin's contributions (a particularly low blow since, by that point, Franklin had died).

It's an important story to tell, not least because it explodes the myth of an abstract, impersonal something overseeing a meritocratic process in which the best work always rises to the top; I'm going to call that something SCIENCE. It's a myth that underlies the "scientific method" so often taught in primary schools, which presents science as an abstract cycle that can be done anywhere, and notably leaves out such steps as "placing your work in the context of the field" and "convincing other people it's worth their time to read what you've done." It's a myth that underlies the movies and books and comics in which the (usually mad) scientist works in seclusion for years before unveiling their creation to the world, which looks on in awe--while conveniently sweeping away any details about how one gets "the world" to pay attention long enough to look on in awe. It's a myth that many scientists have helped to foster, by extolling the supposed ideal of pure research, unhindered by such mundane realities as "politics", in which the invisible hand of the "marketplace of ideas" selects the most worthy contributions.

This is, of course, not the way science has ever worked. Not least because science isn't an impersonal force, it's a collection of people. Papers get reviewed by people, data and ideas get shared by people, hiring and tenure decisions get made by people. And those people have biases, likes and dislikes, and ideas about what a good scientists looks like. Science, in short, is not SCIENCE.

And while we might like to promote SCIENCE as an ideal to strive for, the reality is science is simply too big to work like that. Here's an example of what I mean: on, which is a repository for physics research articles, there are 63 articles listed under "condensed matter" for July 24, 2013. So for one subfield in physics, on one day, scientists produced about 300,000 words of research articles. That's about the length of three typical novels (or one George R. R. Martin novel). Someone in the field, then, who wants to keep up with current research, has three novels a day to read. Three novels of physics, which, in my own experience, generally takes more time to get through than actual novels. Add to that the articles published in the literature of chemistry or other fields that could be relevant, and older articles the new articles refer to that are necessary to fully understand them, and our hypothetical condensed-matter-physics researcher has a rough estimate of half a million words to read every day to keep up with the research.

Of course no-one can read that much. So scientists, like professionals in every other field, use a collection of heuristics and skimming techniques to sift through the mountain of potentially relevant research and pull out things that actually interest them. And these tools are very dependent on social networks and name recognition. See a big name in your field as the author? Probably worth going through. Oh, that person gave a talk at that conference that was quite good; maybe her paper is worth reading. If you're new to the field, you likely have a supervisor who sends you articles to read; your supervisor's choice of articles is influenced by their professional network.

I'm using journal articles here as a proxy for overall research. Less formal avenues are even more prone to be dependent on social networks; a lot of science happens over beer in settings that blur the lines between friend and colleague. The point is that a) any research you do is only valuable if other people see it and use it in their own research, and b) there isn't a good way of navigating the enormous amount of scientific research out there without relying on professional and social networks. So talking to people, making contacts, and participating in "politics" (a term that scientists seem to use solely to describe social interactions they dislike) is and always will be important. It is also, unfortunately, a major mechanism by which implicit and explicit biases at the individual level are magnified into entrenched systemic bias.

This brings us back to why Rosalind Franklin's story is important to tell. Because if we keep insisting that SCIENCE is the ideal, we don't address the biases that are running wild in science, since they're simply a bi-product of the non-ideal aspects that shouldn't be there anyway. If we acknowledge that all science is social, we can look at how to address systemic bias by addressing the individual biases that shape the social network of science. And then maybe we can keep future Rosalind Franklins from being marginalized and ignored in making world-changing discoveries.

No comments:

Post a Comment