How do people learn to be scientists? We’re very good at teaching our students how to titrate a solution, take a derivative, label a dissected earthworm, or calculate the p-value from a one-way ANOVA. One might get the impression that learning these skills is an important part of training to be a scientist. Well, arguably they’re not unimportant; but they’re more skills used by scientists that they are skills that make us scientists. In Being a Scientist: Tools for Science Students, Michael Schmidt tackles the much more interesting question of that latter set.
Being a Scientist covers the softer skills that let scientists do what they do: philosophy, creativity, reading and writing, and so on. For some of these things (like philosophy of science) there’s plenty of literature, but few students (or working scientists) read much of it. For others, there’s little written down* but a huge body of unwritten lore, passed on from mentor to student or hashed out over coffee or beer or in lab meetings. There’s enormous value in making it easier for students and early-career folk (and also us grizzled old fogeys) to learn these things. I found plenty to think about in the book. Sometimes I learned new things; sometimes I found myself nodding vigorously in agreement; other times I found myself disagreeing. The most useful books are those that provoke all three reactions.
Being a Scientist is a succinct book, with 15 chapters organized into three major parts. I’ll summarize, and comment on some bits that interested me in particular.
Part I, Thinking and Behaving Like a Good Scientist, covers the philosophy, ethics, and ethos of science. Like many working scientists, I have little formal training in these areas, so I found these chapters very interesting. They’re brief – philosophy from Aristotle to Latour via Bacon, Hume, and Kuhn in just 15 pages, for instance; but the discussion is accessible and engaging. I was particularly impressed with Schmidt’s treatment of social construction, as he avoids the nihilism of all-claims-are-subjective-and-unreliable but also shows the real value in the idea of knowledge as constructed by human societies.
Ethics receives a similarly whirlwind treatment, with a little more attention to what Schmidt calls the “ethos” of science: the ways of thinking that characterize scientists. Borrowing heavily from Merton, Schmidt discusses communalism (knowledge belongs to everyone), universalism (value of an assertion doesn’t depend on who asserts it), disinterestedness (what we think is true shouldn’t depend on what we want to be true), originality (new insights are valued more than confirmation of old ones) and skepticism (constant interrogation claims, even those we’re making ourselves). An especially interesting discussion considers whether communalism is compatible with capitalism and the system of profiting from patents; Schmidt concludes that it is (oddly, he doesn’t mention open-access vs. subscription publishing in this context).
Part II, Standing on the Shoulders of Giants, deals with the scientific literature in its various forms (from Wikipedia through technical monographs, conference proceedings, and published paeprs). There’s a nice discussion of the history and evolution of scientific papers and journals; advice about finding literature; a history and assessment of the function of peer review; and most valuably, a chapter on how to read a scientific paper.
That key “reading” chapter begins by explaining that reading papers is hard, and argues that that isn’t the fault of readers or writers but simply of the fact that we need to communicate highly technical material in very brief form. I’d object a bit here: that reading papers is hard actually is often the fault of writers, including writers who follow some of the advice that Schmidt himself has to offer (more about this below). But that aside, the tips for reading are valuable – notably, that one shouldn’t plan to read papers as you read novels (from page 1 to the end, to see what unfolds). Instead, Schmidt points out that IMRAD and other structural device exist so that a reader can decide what they need from a paper, and access that without reading the whole thing with equal attentiveness. A key reading task, he argues further, is to figure out the paper’s arguments: its local argument (how it interprets its own data) and its larger argument (how it applies its findings to a broader question or field). Reading done well is active learning, he’s pointing out (as is, I would add while ducking to avoid the flung rotten vegetables, listening to a lecture).
Part III, Planning, Documenting, and Presenting Science, takes a piece of research all the way from coming up with an idea through to presenting or publishing the results. This is a big ask for 115 pages of text, so it’s no surprise that it has to be thin in places. Schmidt begins with an interesting treatment of divergent and convergent thinking in the generation of research ideas, along with some suggestions about how to spark creativity. Even after 30 years in science, I’m still mystified by where research ideas come from – perhaps this is the most unwritten lore of all! There’s a short chapter on refining ideas and writing proposals to fund them. Schmidt pithily summarizes the four questions all proposals must answer: “Who are you? What do you propose doing? Why should we believe you can do this? Why should we give you money to do this?”. He doesn’t delve much into the rhetoric by which one answers these questions; but fortunately, there are entire books that do.
The actual day-to-day conduct of research is covered only by a narrowly focused chapter on the lab notebook. Here Schmidt betrays his background as an experimental lab chemist, because there’s not much suggestion that the notebook so revered in chemistry and molecular biology might not suit a mathematician, a particle physicist, a bioinformatician, or a field ecologist. (In general, there’s little recognition in the book of non-benchtop-experimental science.) To be fair, I’m not sure there’s any way to cover nuts and bolts of conducting research while generalizing across all of science – and that might itself be the key lesson to take from the “notebook” chapter, even if it’s not the one Schmidt intends.
Schmidt turns for his last three chapters to scientific writing (including oral and poster presentations). Here you might expect me to have opinions, and I’d hate to let you down. The book is very strong on strategies for assembling a scientific paper or a talk: for example, which sections to write first; the importance of including the data your story needs and omitting what it doesn’t; how to make effective graphics and how these differ between papers and talks. There’s a wealth of good advice here, explained and justified clearly. But on the matter of style, Schmidt offers instead what I think is interesting bad advice. He’s firmly in the 30-years-outdated camp preferring Methods written in the passive voice. He allows that Science requests that authors use the active, but argues that“a more traditional journal like the Journal of the American Chemical Society will almost always have Methods written in the passive voice”. This isn’t true of science journals in general; and although it might be true for JACS (I didn’t check**), its potential truth is irrelevant – because the fact that our past literature is full of bad writing isn’t a reason to keep that up. In a similar vein, Schmidt advises that a scientific writer shouldn’t try “to express…personality or be creative in [their] writing”; rather,“scientific writing should sound as if any scientist might have written it.” Together, these counsels against the active and against personality are an excellent recipe for the continued production of the flavourless, unengaging tedium that unfortunately typifies our literature. I don’t mean that we should all be writing papers in blank verse (it’s been done, and it was a bad idea) – but Darwin didn’t write without style, and it wouldn’t hurt our modern literature to be less dull and maybe even a little bit beautiful. Now: I said the bad advice here was interesting bad advice. “Interesting”, because it’s probably good advice when offered to undergraduates writing for grades. It’s the risk-averse strategy, to produce writing that models what’s already out there, because that’s what sounds like scientific writing to most of the folks who will assess it. Sadly, it’s this circularity of expectation, and the failure of those who teach and grade writing to reach beyond it, that keeps our literature from improving.
But here’s the thing. I’ve just been arguing with Being a Scientist – and that’s exactly what I suggest readers should do. It’s even what Schmidt suggests (implicitly) readers should do, by including “skepticism” in his treatment of scientific ethos. As Schmidt points out in his Introduction, few questions around how science is or should be done have clean and simple answers. So Being a Scientist should start conversations – mentor with mentee, or labmate with labmate.
Being a Scientist is aimed at undergraduates, but I think it would be excellent reading for beginning graduate students too. It would be especially valuable as the centrepiece for a seminar course or series of lab meetings in which students take turn presenting a chapter or two, supplementing the book with deeper dives into the literature. Students could, for example, read the original (or at least more detailed) sources for some of the ethics and philosophy; or bring their arguments for how the book does and doesn’t apply to scientists doing theory or meta-analysis or field work; or compare recommendations for style across several different writing guides. I would have learned a lot from discussing this book early in my career. I learned some things reading it even now.
© Stephen Heard April 14, 2020
*^Or more accurately, little written down in formal contexts like books and papers. There are guidebooks (like the British Ecological Society’s excellent guide to peer review); there are blog posts (like this one on choosing journals, this one on attending conferences as an introvert or this one on how to write the teaching-statement part of a job application), and there are opinion pieces in society Bulletins and so on. But they’re scattered and not easy to find.