“We build tools to assess controversies”

RESEARCH: Why are you interested in scientific and technical controversies?

BRUNO LATOUR: Scientific and technical questions have become so pervasive in everyday life that everyone is obliged to take an interest in them! Should we or should we not eat bluefin tuna, as some claim the species is endangered? Should we, or should we not, use energy-saving light bulbs, as some claim they generate a potentially dangerous electromagnetic field? These controversies are never framed in terms of pros and cons, as on television, or guilty/not guilty, as in court. They arise as a tree of questions to which different actors give different answers. Mapping a controversy therefore means listing the positions present, for each one describing who they are supported by (scientists, industrialists, non-governmental organizations, etc.) and with what arguments. The ultimate goal is to help citizens make up their own minds about these controversial issues.

Can’t we trust the experts to decide them?

BL No, public opinion increasingly doubts the experts, and moreover the authorities in general. I started this work fifteen years ago with students from École des mines de Paris (today Mines ParisTech). Like all engineers, they had a slight tendency to think that there is a best solution to every problem and therefore to every controversy, even if their teachers did not know the answer to the questions asked. The mapping of controversies was therefore born as an educational tool to prepare future engineers for a world where their expertise will no longer be automatically respected. Since then, the huge increase in digital techniques has changed the game. The generalization of digital tools makes it possible both to be flooded with information that is not always very reliable, but also to invent tools that make it possible to navigate, to find around, to make visible the diversity of positions.

What types of positions? Must e.g. expert positions and “lay people” are treated separately?

BL No, there is no need to impose the distinction between the two, as we can now put them together without confusing them. If digital techniques have changed the game, it is because the tools for following scientific journals and private blogs have become the same. Before, it was very complicated to follow both scientific literature and opinions. Now the search tools on the Internet are the same for following producers of fact and producers of rumor or opinion. I am quite sure of the possibility of simultaneously measuring the galaxy of scientific positions and the galaxy of social positions. For example, what is said about vaccination against the H1N1 influenza virus both scientifically (in specialized journals) and rumor (on the blogosphere), or even political decisions.

Why do you choose to put all these opinions on the same level?

BL Because controversies that once remained confined within the scientific community are now brought to the public and concern thousands of people who have their own ideas on the matter. Fishermen have an opinion about the decline in fish stocks; beekeepers have an opinion about the scarcity of bees. Before, a scientist could say: “Your ideas are not scientific, here is the correct explanation.” Now, whether we like it or not, such an intervention will no longer stop the discussion. That being said, there are only two positions left. Either despair, lament the rise of irrationality, etc., or invent instruments to correct this diversity. It is not a question of placing all positions on the same level, but of registering them in the same space, which will be the public discussion. This is all the more important since in a democracy it will be necessary to convince all the parties present to end the discussion. We may not necessarily have a consensus, but at least an agreement to disagree. To chart a controversy is not to look at it from Sirius, in an ideal of perfect objectivity, but to adopt a second-rate objectivity.

What do you mean by “second rate objectivity”?

BL The first rank, sweet sin for the students at the School of Mines, is to say: I look at a controversy, I examine the positions, I choose the one that seems best to me, and I support it with my hierarchy. It is not the role of an expert or an engineer to do so. Its real role is to tell its bosses: there are about ten opposing positions, which depend on interests, technical devices, investigative methods. To be objective is not to take the position of an expert but of a cartographer. This implies taking marginal elements seriously, just as when you draw a map of France, there is not only Paris but also Saint-Martin in the Allier. Recently, a scientist attending a paper presentation on the bee depletion controversy asked: Why are you presenting this position that has been invalidated by research? This is a typical reaction. It is invalid, to be sure, but nevertheless exists in the landscape of opposing attitudes, such as those of the conspiratorial zozos and those of the pesticide industry. The second layer of objectivity is to present all the positions.

But does this not lead to putting all scientific positions on the same level when they do not all have the same influence?

BL I have the weakness to believe that the sociology of science, based on bibliometrics and scientometrics, is today able to measure distances between positions. As part of the European project Macospol (English acronym for Cartography of scientific controversies for politics), which I coordinated, our partners at the University of Amsterdam, for example, have designed a bias detector that works very well. It automatically detects conflicts of interest that may influence scientific results. It is sufficient to list the researchers who attend conferences, testify before the courts or intervene in the press in connection with a dispute, and to see in which other controversies they get involved. When the same researcher intervenes on the harmfulness of cigarettes or asbestos, on global warming and on acid rain, one can only doubt his expertise. The same applies if we find that all those defending a thesis are from the same research group. Or that they all received support from the same source. It is elementary journalism, but what is new is that digital tools put this type of investigation within reach of a student.

Do you think mapping out a controversy can bring closure to it?

BL This is one of our research topics. In particular, we are trying to find out whether cartography can make it possible to close artificially maintained controversies, such as about the carcinogenic nature of tobacco. A current example would be the controversy over the anthropogenic origin of global warming. Bias detectors play a very important role here.

Suppose that mapping the controversy over the anthropogenic origin of global warming shows that climate skeptics’ articles are cited only by other climate skeptics. Would it make it possible to reduce the influence of these ideas in public opinion?

BL The reason these controversies can fester is that climate science cannot reach the level of certainty that would end the debate. They, like all the modeling sciences, have their own epistemology, about which the public has never been informed, because it remains in a positivist conception of science, according to which it is enough that a fact contradicts the theory (for example, such a cooling room) in order to it becomes invalid. We must change our vision and establish trust in science in a way other than positivist certainty. One of the means is to evoke the scientific institution, the society, the instruments, the means of research. Today, we can have full confidence in the scientific establishment on this question of the anthropogenic origin of global warming: we have never obtained such a varied body of evidence, of arguments, to establish it. It is therefore a matter of replacing certainty with trust in the scientific community, but the problem is that we lack the instruments for that. It is therefore necessary to create new instruments adapted to this web environment, where everyone will look for their own information. Faced with a map, everyone must be able to place themselves: for example, to see that if I am against the anthropogenic origin of global warming, I stand with people who are against abortion, for the use of coal, and ask myself , about I really want to be next to these people.

The term map implies a two-dimensional space, but the controversies you mention have much more to them.

BL They are not maps in the sense that they would have latitude and longitude. These are network analyses, where there are as many dimensions as there are pointers. The bibliometric analyzes of co-citations or co-authorships of articles today, it must be admitted, are very difficult to represent simply. It looks like spaghetti! We are still very far from having a satisfactory visualization, it is clearly our bottleneck. Consequently, the mapping of a dispute is still unable to help with the decision. These tools remain too complicated for people who don’t have time, who have to write a note to their minister in a few hours, and who want to understand the space of existing positions at a glance.

And for the public interested in controversy?

BL In the end, our goal is to create a search interface on the Internet to answer a question by saying whether it is controversial, for how long, by whom, belonging to which professions, having which authority, in which country, according to which arguments , etc. We should be able to see this automatically with a color scheme. If this tool tells you that it is a closed controversy, you can directly consult an institutional information page or an encyclopedic notice.

Does this mean that the reader abandons forming his own opinion in order to align himself with those he trusts with opinion makers. If he doesn’t like Monsanto, he will listen to what the anti-GMOs say without trying to determine the truth.

BL This notion of truth is problematic. The time allotted to form an opinion is limited, all the public involved in a controversy can do is to discover the bearers of an opinion that seem the most reliable. It is a question for scientific and technical subjects that have consequences for my life, and there are many of them, to be able to discover as quickly as possible who is biased and who is not. Not even to be informed – there are encyclopedias, specialized journals for that – but to discover partisans with simple criteria. Does this person e.g. published in the field? Because if she hasn’t published, what is she doing in this debate? The number also matters: how many people support this position? Once the partisans have been discovered, one can go to information sites, which already abound. Historically, newspapers made it possible in the 19th century to form a political opinion and thus contribute to the development of democracy. Today, it is a matter of inventing new media in the form of web tools that enable everyone to form a scientific and technical opinion.

Photo credit: JOEL SAGET / AFP

Leave a Comment