Pulling back the curtain on conferences

Before I made the good decision to give up tenure I went to academic conferences in history, journalism, and sociology that were both fluffy and petty. I thought conferences of biologists, chemists, and physicists—folks dealing with hard realities—must be much better. Not so, according to University of Cambridge scientist Jeremy J. Baumberg. In The Secret Life of Science, he pulls back the curtain on scientific pecking orders, groupthink tendencies, journal competition, the “fluffy marketing” of theories, and—in the excerpt below, courtesy of Princeton University Press—conferences. The Secret Life of Science made WORLD’s short list for 2018 Science Book of the Year. —Marvin Olasky

Competition of conferences

Most unhelpfully, the pool of conferences is not subject to much competitive pressure. Scientists gain by being organizers of meetings, and promotion or prizes may depend on being seen to be heavily involved in such community leadership. As a result, it becomes important to make such conferences successful and make sure good science is represented. Twisting arms of highly reputed colleagues into attending is one of the real jobs of organizers. Killing a conference would remove esteem from a field and from the current generation of its motivators.

While there is clear sense of prestige attached to publishing science (as we saw in the last chapter), that attached to giving conference talks is fuzzy. Gaining the opportunity to talk at a massive-scale conference has high worth, but scientists know opportunities to discuss new science here are lower. Learned societies are happy when their conferences increase in size, reporting this as their enhanced weight in science, and bestowing them greater presence in national lobbying. Topical meetings are instead where the peers you really want to impress are, but they proliferate continually.

Killing a conference also “disinvests” science funds from businesses of a host city. All meetings are now encouraged to evaluate their local “benefits.” Typically they use an “event impact calculator,” in which the money brought into a city by the descent of scientists into airports, taxis, hotels, venues, restaurants, and bars is added up. Such calculators (encouraged by city halls and trade organizations) make no attempt to balance these against the other side of the equation: lost time, poorer management, environmental impacts, or the true benefits to society. More to the point, there is no sense of truly evaluating the impact and benefit of conferences—it is yet to be built into the science ecosystem at all.

Underpinning all this is the support of funders to underwrite the activity. Overall increases in science funding are not connected to any careful scrutiny of the percentage spent on conferences. Funders assume that being invited to talk at conferences is a good outcome, boosting the need for researchers to focus more on them. Nowhere in the science ecosystem is there any explicit attention to how many conferences it is worthwhile to attend. While it is possible to argue that even one rewarding encounter is beneficial, this has to be set against the opportunity costs. Funds can be spent elsewhere, and most importantly, time can be spent on other activities.

Nowhere in the science ecosystem is there any explicit attention to how many conferences it is worthwhile to attend.

This lack of self-scrutiny means that no one even collects statistics on who goes to how many conferences (though see the website [that goes along with this book]). Current measures of a successful conference are: did it break even, did enough people come, and will they come again? Circumstantial data on the best size for banquets, the length of poster sessions, and how much food or drink they should be accompanied by, as well as much else, is left as implicit convention. Meetings without long-term leadership bounce from style to style, while the yearly megameetings find it hard to avoid ossification. Myriad opinions rather than any consensus suggest the “right way.” And none explicitly focus on engendering those magic moments that set researchers alive.

Let us imagine that researchers instead spend one of their conference weeks each year differently—collectively discussing recent research directions. Assigned collaborations, committing a week of time but virtual rather than requiring travel, could produce a joint review reflecting on advances in a subfield. This would be accessible by others, use electronic media and informality, exploit group views rather than individual polemic, and help guide researchers through the web of emerging new published material. Key is to find a way to reward individuals who develop such activities.

In the end scientists sustain the current system because alternatives cannot yet be envisioned and funded. Conferences can still deliver the “zing” of new insight, creativity, and problem solving, but only sporadically. They also form the lifeline of connections that keep the whole ecosystem bound together, albeit with increasingly shorter lengths of tighter-lashed ropes. As a result the only choices scientists consciously ponder as consumers is the selection of which meetings to go to.

Which conferences do scientists want to go to?

Homes give us safety, secure our assets, and are where we let our hair down. But how many conference homes can a scientist have? This is an ever more difficult question, as the number of subfields and conferences that might sustain each scientist grows. At the start of their careers they get little guidance, and the practice within any subfield varies widely.

Young researchers drool over announcements for conferences in exotic locations, talking up the esteem of this meeting to their boss. One of the joys and perks of being a scientist has always been the international interconnections. But since the cost of travel, hotels, and conference fees is typically several thousand dollars, any research head will be cautious about committing funds unless the reason to attend is compelling.

For young researchers who just completed their first project, this is their opportunity to discuss it with other scientists, make connections, and feel for the first time part of a field. Rather than a panic-stricken outsider wondering why everyone else is so much smarter, they can attain the feeling of belonging and acquire a reason for all those dazed late nights and enthusiastic weekends in the lab. Research supervisors like students to go to conferences as soon as they have results, to motivate them (to work harder). Partly this can show students just how competitive science is, and how people all over the world may be working on the same problems. Students hopefully return with a much wider view of research and start to think in their own way what good directions to pursue.

It is at the next stage of career, where a researcher is in charge of their own funds, that things become more complex. Repeated contact with colleagues and competitors in a topical subfield forms a stable tribe for them, and they can see where they stand within it. Once they find a group of scientists friendly and inclusive, they will be happier to go to their meeting again, as long as the others also keep attending. The most important asset for meetings then is their aura, the atmosphere they manage to engender, which becomes associated with the meeting brand. Like a microclimate formed by a biocommunity, its assets are its sanctuary and stability.

Repeated contact with colleagues and competitors in a topical subfield forms a stable tribe for them, and they can see where they stand within it.

As a scientist’s work becomes significant to others, they may find themselves “invited” to present at a meeting, rather than submitting the short summary abstracts that are ranked to bid for time slots. Invited talks are markers on a CV, indicators of research esteem. Compression of scientific presentations into the scarce available time means high competition for such longer talks. It is hard to turn down the offer of giving an invited talk, and it becomes still harder for the next levels of esteem, “plenary” and “keynote” talks, normally given to the massed attendees. Researchers thus compete to access the most prestigious conferences. But as we shall see, conferences compete to access the best researchers.

Location, location, location

One way scientists choose between conferences is where they are held. European nobility have donated many astonishing palaces and castles to universities and foundations for use as venues, motivated by their realization that investing in the future is the only powerful legacy. Scientists always find some rationale to organize new meetings in such stunning locations, relentlessly adding to the number of conferences.

Carefully constructed venues combine the lure of novelty with subtle transportation problems that prevent attendees from disappearing to the brilliant beach instead of darkened chambers. Spectacular Italian hilltop castles miles from anywhere, or remote mountain lodges, are spectacularly effective. All scientists accumulate stories of travel quests progressing from airplane to goat cart, of baggage that arrives just as they leave, or being stranded in unexpected places. During the 2010 Icelandic volcano lockdown of the upper atmosphere, my students called from Mexico to ask what to do since they couldn’t get home from their conference. I had to convince them to find something to explore rather than stew in their rooms waiting to get back to their labs.

How to choose a conference

Scientists also choose between conferences on the basis of interesting invited speakers. If I know all their names it is not such a good sign—perhaps I only hear the same people time and time again, so learn little beyond reinforcing group-think. If I know none of the speakers, then perhaps they really aren’t doing anything terribly exciting. It is very hard to guess if I will learn something important from a conference before I go, or experience stimulation and creativity from an open exchange of ideas. There is no openness index for meetings that I can check.

Sometimes, the pressure to attend comes from the dangers of being missed. Will I no longer be seen as a significant actor in the discipline? Will I not be party to conversations that build a mutual support club? Will I not be “important”? Competition in science intensifies this endless participation, ratcheting up as competitors speak ever more.

Sometimes, the pressure to attend comes from the dangers of being missed.

Even if I am invited to a conference, rarely will my costs be fully covered by it. Proselytizing scientists will mostly be supported by their national governments and taxpayers. The better they are, the more they’ll travel to conferences, but the more difficult it becomes to find enough time for them to carry on their best research. Thus they pare down time at each conference, depriving younger attendees of debates and serendipitous encounters. Invitations come weeks or years in advance, and it’s hard for scientists to decide which to accept, harder still to decide for those who want to privilege their own personal relationships. Such selection pressures unbalance the ecosystem by elevating only certain personality types, working against its overall interests.

With more conferences, the need to spread a message without it drowning in the sea becomes more important. These are the factors that propel scientists around the globe, and despite widespread disquiet about carbon footprints and unsustainability, no alternative model has traction.

How do speakers get chosen?

Having looked at what drives scientists to conferences, we now look at the conference ecosystem. Each conference aims to bring prestige to the organizers. Location, celebrity speakers, established clientele, and a good brand are needed, as with any events. Organizers have to build an attractive program early enough to convince their potential audience to earmark the meeting dates and identify funds to come.

A scientific advisory committee defines the scope of the conference and the range of subfields represented, setting how many people would be interested. They then suggest names of potential speakers for the invited, plenary, and keynote talks. A single keynote talk for the conference, for instance from a Nobel prize winner, and a plenary talk every day from a highly successful established scientist, work well to keep delegates attending for the longest time, and rouse them early each morning.

Invited talks delineate the range of subfields, ensuring groups send other delegates too, since invited speakers have a vested interest in making the conference a success. Targeting of larger audiences leads to inventing trendy sessions to capture the imagination of potential attendees. Since new areas do not often have many results yet, such sessions can rapidly become clones of each other.

Cartel carousel

How does this committee of scientists choose speakers? Often they take inspiring talks that they recently heard at another conference. This is very prevalent but leads rapidly to overusing a cartel of famous speakers. The selected few hop from conference to conference, while their research itself progresses only at the normal rate. One has heard their talk before, and while a good show, it ceases to stimulate new creativity. It also has the effect of driving other researchers mad to break into the cartel.

The selected few hop from conference to conference, while their research itself progresses only at the normal rate.

Once in the cartel, if a scientist brings in the crowds and has the golden touch then, as in mass media, they become the figure of the moment. Their name on the conference advertising gilds any meeting, and the invitations build up thick and fast. At this point many a speaker regrets the position that they gained, feeling closer to entertainer than scientist. Good speakers start to wonder if they can say just anything and still be adulated. Questioning is mostly timid in the face of such reputed greatness. Of course, if the science really is dull or weak, the crowds will drift away, and in time subfields trail off into the devoted few acolytes. Harnessing the cartel carousel primes the success of a meeting. It can then be used to reassure potential meeting sponsors of a high profile, and allow a secure financial budget to be put forward.

Another tactic for selection is by word of mouth, discussing with colleagues who they would like to hear. Program committees are chosen to give some balance across different countries, genders, or experience. Members are often partisan, focusing on their nation’s figures who have been doing strong science. But in many fields it is no longer easy to know who has been getting interesting data. Researchers even in a single country are disconnected from each other since they rarely attend national meetings whose scope is necessarily smaller than international meetings. A frequent but bizarre experience is finding time to talk to a colleague working in the building next door to mine, at the conference on the other side of the world we are both attending.

As with innovation or in any ecosystem, it is hard to introduce novelty because customers and backers are risk averse. Most conferences are remarkably similar to each other, not because the model is perfect, but because the system crushes out change. A disappointing aspect of topical conferences is the way they can develop into a closed group of eminent scientists who persist on inviting each other. There is a tradition in many fields of preventing committee members from speaking, however their protégés are often substituted. Such insular communities persist for a long time, their stability producing a comfortable support system, but reducing the intensity of competition and sometimes the pace of the field.

Excerpted from The Secret Life of Science: How It Really Works and Why It Matters by Jeremy J. Baumberg. Copyright © 2018 by Princeton University Press. Reprinted by permission.