Latour undertakes this investigation by not listening to what philosophers say about Truth, not listening to what sociologists say about Society, and not listening to what scientists say about Nature, but instead to observe the scientists at work. Find out what they do, and not what they say. And some of his interpretations of their actions are quite surprising.
Latour starts with analyzing technical papers, as those consume a great portion of most scientists' time. He decomposes the dense tangle of references, citations, and figures and explains how this tangle is necessary as a defense against those who would attack the paper. By referencing and citing others, the scientist mobilizes an army against those who would attack him. "Aha!" he says. "To attack my claim, you would first have to disprove all of these other claims!" Figures serve a similar purpose, except that they connect him to the laboratory. To disprove the figures requires having a similarly equipped lab to run a similar experiment.
It was really interesting to me to see this take on papers. I had never really thought about why technical papers were so hard to read, but it makes sense if one thinks of the scientist as being hunkered down in a bunker ready to defend their claim against all attackers. And this was just the first chapter, so I looked forward to the rest of the book.
Latour goes on to make several more striking observations about how science is actually done. He spends some time studying how science gets funded, tracing out the vast web of connections that must hold for science to actually take place. And he describes this web by noting how the scientist must convince others that their interests are aligned with his. He must convince the company that funding his research will lead to tangible benefits in their product. He must convince the government that his research can make better weapons, or save lives. Ironically enough, the more esoteric the subject, the more networking has to occur to allow it to happen at all. Thus, the "purest" of sciences such as particle physics (to pick a random example) that would seem to be the most remote from everyday interests, has to do the most work to convince people that it is necessary.
Another interesting concept he follows is the idea of black boxes. He
labels scientific theories as black boxes that are used as the bases
for other theories. For instance, we treat the idea of DNA as a
double helix as a fact now that all biology experiments now must take
into account. Latour dives back into history, using
In fact, the entire idea of black boxes is a useful one, I think. It plays into my thoughts of how the one axiom of science is that everything can be questioned. For our convenience, we tend to treat most well-checked theories as facts, but, at any point, we can open up the black boxes, see how they were constructed, and possibly attack them if more evidences has come to light. Richard Feynman describes such an incident in his autobiography: he had been wrestling with a theory that wouldn't fall into place because he "knew" that beta decay of the neutron is S and T. One day, some experimenters told him that it might be V and A - they opened up a black box for him - and his theory fell into place immediately.
Of course, we can't go around and question things all the time, or we would never get anything done. So when theories reach a certain point of acceptance, we begin to treat them as true, as being a black box rather than a theory. Eventually, these black boxes become so accepted that they become invisible, and only pop into focus again when an outsider questions them. This idea was built upon by Bowker and Star in Sorting Things Out.
Latour also brings up some interesting questions about who actually does science. When we think of scientists, we think of the lonely researcher, alone at his workbench, separate from the rest of society. However, Latour demonstrates that this picture is incomplete. He traces the itinerary of a laboratory director as he flies around the world, talking to government officials to drum up more funding, talking to journal editors to convince them to open up a new section for submittals, talking to companies to refine their instruments to make the research that his lab does more effective. Is this lab director doing science in the colloquial sense? Certainly not! But when the researcher in the lab uses the extra funding to buy the new improved instruments to get results for a paper to be published in the new section of the journal, it becomes clear that the director is indispensable to science.
So who is actually doing the science? Latour answers the question by noting that because science must enlist many social actors in order to happen. So, in a sense, everybody is contributing to science. Governments through their funding, companies through their instruments, etc. He mentions the more typical theory that a genius scientist comes up with a brilliant theory which steamrolls its way across society, forcing millions of people to follow in its wake. He contrasts this with his theory that the millions of people are enlisted before the theory ever comes into existence, through various re-alignment of interests performed by the scientist. Since the people already have an interest in the research succeeding, once it does produce a theory, it is no wonder that it sweeps across society so fast.
I've only begun to scratch the surface of Latour's ideas in this book. I'm definitely interested in picking up other books by him to see what he has to say.
Updated with links to my reviews of other Latour books: