Rethinking “Science” Communication
Old Sam Hamilton saw this coming. He said there couldn’t be any more universal philosophers. The weight of knowledge is too great for one mind to absorb. He saw a time when one man would know only one little fragment, but he would know it well.
Last year, 8,411 science and engineering journals published just over 1.1 million peer-reviewed articles. Another 190,000 papers were published in 3,016 social science journals. This works out to more than two peer-reviewed articles being published every minute of every hour of every day for the entire year. These staggering numbers should change how we think—and talk—about both science and scientists.
Consider that the smallest field, marine engineering, published 564 articles across 14 journals in 2013 alone. In this single field it would be difficult to keep abreast of the research. It is all but impossible for materials science, which had over 61,000 papers. After the Philosophical Transactions of the Royal Society became the world’s first scientific journal in 1665, it took about 200 years to reach 3,000 journals. We’ve added almost that many in just the past 10 years.
So what does it now mean to be a scientist? When it involved a few European men studying a handful of topics, it may have made sense to treat science as a coherent idea. Not anymore. Is it fair to think of theoretical physicists and ecologists in the same way when scientists in the same discipline often don’t even overlap?
At meetings of the American Geophysical Union, it always struck me that soil scientists are completely separated, physically and intellectually, from space physicists. Even within space physics, those of us studying near-Earth objects did not interact much with the folks studying Mars.
Sure we were all geophysicists. But I can’t tell you the first thing about soil science, and I suspect soil scientists would say the same about my research. So how much do different groups of scientists really have in common? How much do we understand beyond our own specialty?
The dilemma only grows when we look outside academia. We tend to think of science as the abstract pursuit of knowledge conducted at a university. Again, this picture may have been true at some point. Today it’s a lot more complicated.
We now have environmental science, science for international development, disability research, pedagogical research, science for policy, and science of policy. We use science to improve air quality in the United States and to produce cleaner cookstoves in Africa. Universities, government labs, nongovernmental organizations, foundations, and private companies all play a role.
Looking at how diverse scientific research has become, it’s hard to understand what common phrases like “the scientific method” and the “scientific way of thinking” mean. Consider what should be a straightforward question: Does science involve experiments? At some point we all learned that scientists use experiments. But is that always true?
Consider space physics. My colleagues and I often argued that you cannot experiment with the Sun because space conditions change continuously. We instead focused on constructing models, making observations, and collecting and analyzing data. To this day some of my friends insist that we never ran experiments. To them, space physics is an observational science only. That’s a pretty powerful claim worth repeating: Practicing space physicists believe they do not conduct experiments.
So saying that science or the scientific method involves experiments doesn’t make any sense. Some fields of science run careful, repeatable experiments. Areas such as atomic physics come to mind. This type of research, where every parameter can be diligently controlled, is how we usually think of science. But some fields cannot do this.
The phrase “science proceeds by testing falsifiable hypotheses” is similarly imprecise. Sometimes hypothesis testing requires double-blind controlled trials. Other times it requires computer simulations or observations. Is a double-blind trial including real people anything like a computer model of a simulated future? Sometimes we precisely monitor all variables and make repeated measurements. At others we control just a few variables and do not repeat. Some hypotheses are clear and are clearly falsifiable: I hypothesize that this solution is acidic rather than basic. Others depend on so many different branches of science, bodies of data, and physical models that the relevant scientists cannot agree on a test, and falsifiability is clearly not possible: Did global warming intensify the last hurricane?
In this vein, consider the phrase “how scientists think,” as if in some way we all think alike. As described in journalist Chris Mooney’s book Storm World, the data-driven climatologist Chris Landsea, at the National Hurricane Center, weighs observations more than theoretical calculation and is skeptical that humans can detect a link between global warming and stronger hurricanes. MIT’s Kerry Emmanual, attacking the same problem, uses the opposite approach and concludes that there is a strong link. This divide between emphasizing observation versus theory goes back to the founding of meteorology. Even scientists in the same field wrestling with identical questions do not think the same way.
And then there’s “the purpose of science,” perhaps the most perniciously misleading phrase of them all. There is no one purpose to science. Scientists have purposes. They are as complicated and varied as all human purposes. Some scientists want to explore nature, some want to prevent disease, some want to create technology, some want to educate children, some just want to do their job, get paid for it, and go home. None of these goals are more worthy than the others.
It appears that thinking like a scientist, testing a hypothesis, and conducting an experiment can have a different meaning depending on the research and the scientist in question. By themselves, these phrases are vague and uninformative. Yet they are used in public all the time as if they communicate something specific and essential to science, as if they add up to a single “purpose.” But they don’t.
The world is a very complicated place, and we can’t expect scientists to apply the same rules and procedures to the millions of questions they ask. Scientists need intellectual and methodological flexibility. Certain tools and methods are appropriate for atomic physics, others for space physics.
The problem is that these differences are rarely discussed. Scientists instead typically give the opposite impression, presenting science as uniform and monolithic. We promote the scientific method instead of scientific methods. We discuss “how science works” rather than emphasizing that science is too diverse to work in just one way.
When have you ever heard a scientist publicly acknowledge that “the scientific method” is just a generic phrase that doesn’t tell you how to approach a specific question, and may mean something entirely different for a subatomic particle physicist and for an evolutionary biologist? Or that not all scientists conduct experiments? Or, to go back to my previous example, that observational and theoretical meteorologists think differently about the same problem?
Addressing the public
Given this diversity in scientific methods, how should scientists approach public communication? To begin to answer this question it might be helpful to first consider how we speak and think about any large category. There are two important principles to keep in mind. First, broad overviews do not always help you understand particular situations. Even more strongly, a specific claim can contradict a general one without undermining its truth. The fact that Orange County is very conservative does not undermine the generalization that California is a liberal state. These ideas can simultaneously exist without conflict because they speak at two different scales. The truth of one does not negate the other; it only complicates it.
Second, the members of any category share differences as well as similarities. All 50 U.S. states share common features. But there are also differences between Alabama and California, between northern and southern California, and even between Palo Alto and East Palo Alto.
Whether we emphasize similarities over differences, or whether we speak in terms of generalities rather than specifics, depends on the issue at hand. A comparison between U.S and British politics will use national generalizations and ignore the local distinctions between San Francisco and San Jose. A housing search will take the opposite approach.
SCIENCE IS NOW TOO BIG AND TOO DIVERSE, AND ABOVE ALL TOO DEEPLY WOVEN INTO AN ARRAY OF SOCIETAL ACTIVITIES EVERY BIT AS DIVERSE AS SCIENCE ITSELF, TO BE CONSIDERED ONLY “SCIENCE.” THERE IS NO SCIENCE ANYMORE, THERE ARE ONLY SCIENCES.
I suspect we all instinctively grasp these principles. We have all discussed categories at different scales. We all know that statements about red and blue states do not pertain to everyone within the state.
In my experience, though, we scientists tend to neglect these principles in our public communication. We overemphasize the broad generalizations and do not spend enough time explaining the local distinctions. Science is now too big and too diverse, and above all too deeply woven into an array of societal activities every bit as diverse as science itself, to be considered only “science.” There is no science anymore, there are only sciences.
Nevertheless, there will always be occasion for academic scientists—even those, like me, who left academia after getting a Ph.D.—to explain their research to nonscientists. And these occasions may require discussing “science.” When we find ourselves in this situation, our use of language must be meaningful and precise. Our words should transmit coherent thoughts, and there should be no doubt about what we are trying to say. In short, what we as scientists say in public should at least be consistent with our own ideals, if for no other reason than that we cannot expect the public to do so if we do not. Though we exhort the public to use evidence, to value truth, and to not distort, exaggerate, or cherry-pick examples, we often violate these principles when we discuss “science.”
Consider again that standard nostrum: “Science is about testing hypotheses.”
How should someone interpret this phrase? Is it that all scientific discoveries occur because a scientist is testing some hypothesis? Well, we know that isn’t true. Many discoveries have happened by pure luck. Scientists can’t test a hypothesis they don’t even know exists. We also know that some research can involve nothing more than observations and data collection without any conscious hypothesis testing.
Perhaps we mean only scientists test hypotheses? But we know that isn’t true either. Plumbers also test hypotheses: They hypothesize that Draino will fix the problem and modify their belief based on the result. You could argue that we all do so sometimes: I hypothesize that restarting my computer will get rid of the glitch in the behavior of my mouse.
So if everyone tests hypotheses at some point, and scientists do so only part of the time, what’s the big deal? Why bother emphasizing it at all? And since a “test” in geophysics is so different from one in atomic physics, it makes even less sense to use a single phrase to explain all scientific explorations.
And what’s with “is about”? All the professors I know spend most of their time applying for grants, managing them, and sitting on committees. In academic science at least, it seems much more accurate to say “science is about grant-writing and committee-sitting.”
Finally, how can “science” do anything on its own, much less test a hypothesis? Science must be done by scientists— by actual human beings.
A more precise version of the above might be: “Among their dozens of activities, some scientists—along with many other people—sometimes test hypotheses of various forms in many different ways to accomplish their goals.” This version is both accurate and understandable. Unfortunately, it is not very interesting.
How can scientists paint a more accurate and interesting picture of science? The answer is by not communicating “science.” Imagine that sportscasters spent most of their time discussing “sports” and nothing else. We would be left with nothing but clichés: Sports are about hard work and teamwork! The keys to the game are hustle and good defense! Fans rightly tune out such commentary. We are enlightened, however, by deep, insightful analysis of a specific game. We want to know how an injury will change team strategy, and when they should run the ball instead of passing it.
Intelligent science communication should emulate intelligent sports communication: low on generic phrases, high on detail and particulars. In this aspect, our public communication should emulate our research communication: precise, clearly defined, and modest. When we have to make generalizations, we should do so carefully and reluctantly, with many caveats and as few clichés as possible. We should acknowledge, privately and publicly, that any statements about science as a whole will often be overly simplified and not very useful.
INTELLIGENT SCIENCE COMMUNICATION SHOULD EMULATE INTELLIGENT SPORTS COMMUNICATION: LOW ON GENERIC PHRASES, HIGH ON DETAIL AND PARTICULARS. IN THIS ASPECT, OUR PUBLIC COMMUNICATION SHOULD EMULATE OUR RESEARCH COMMUNICATION: PRECISE, CLEARLY DEFINED, AND MODEST.
Communicating about anything as vast and heterogeneous as either science or sports can be done only by glossing over distinctions and muting differences. But at least in sports we know not to take these clichés too seriously. We also have phrases that present a more complex and nuanced image. We hear that professional sports are about money and greed as well as teamwork.
No one feels the need to choose between these two views of what sports is “about” because we understand that complex systems must be analyzed from different vantage points and that there is no single way to encapsulate all of sports in a few sentences. Professional sports are both a business and a game. They involve both greed and hard work. Our appreciation would diminish if we forgot that these traits exist simultaneously.
Scientists, on the other hand, do not provide such competing images. We don’t hear that science is about writing grants as much as it is about hypothesis testing, that in some ways science is objective and value-free, while in others it is not. We don’t recognize sentences with “science is,” “science is about,” “science involves,” “the scientific method,” “the scientific way of thinking,” or “how science works” for the vague, broad generalizations that they are. We don’t see that any attempt to explain “science,” as opposed to one specific area of science, will often resemble bad sports commentary.
To a certain degree, it’s understandable why scientists speak like this. Categories exist for a reason, and at times it can be very important to stress the differences between science and nonscience. For all their differences, atomic physics, space physics, and ecology have more in common with one another than with acting and literary criticism. At times it may be appropriate to speak of these disparate activities as a whole. Because the authority of science is often abused and misused, it can be important to stress a few general, even overly general, attributes. But it can be dangerous to take this approach too far. Almost by definition, a big-picture view of science will elide details and hide more than it reveals. Perhaps most important, these generalizations have contributed to making scientific authority more vulnerable to attack than would a more honest portrayal of the richness and diversity of the scientific world.
Consider climate change, where a childishly simplistic model of falsification has been so powerfully deployed to attack the scientific consensus. Imagine a world where scientists did not continually intone phrases like “science proceeds by falsification” and “scientists change their mind when the evidence changes.”
Perhaps in that world the public would know that sometimes falsification is straightforward and sometimes it is not. And that in any case the very term “falsification” is a broad concept that must be applied to a specific research field. Perhaps in that world the public would know that there are distinctions between falsification in analytical chemistry and climate science. And that scientists don’t always change their minds in response to every paper but instead look at the full body of research. Perhaps in that world, the latest paper on global warming would be viewed in context rather than as one shot “proving” or “disproving” global warming.
Why doesn’t that world exist? In no small part it is because we in the scientific community have fought its creation. And so we see the same dangerously simplistic model of science invoked in endless debates over uncertainty about the risks of nuclear power or the effectiveness of various educational or social policies. Politicians and the public seem to expect that epidemiology or sociology can create the same sorts of knowledge and levels of certainty that we see in experimental physics. We have told them that science is science, and they believe it. And since we in the scientific community do nothing to elucidate the very real and significant differences among the sciences, we are often powerless to combat misrepresentations.
It would help to acknowledge our limitations as scientists. I am not qualified to discuss condensed-matter physics even though I’ve taken a few classes on the subject and have a Ph.D. in applied physics. For any field of science outside my own, I always hesitate, speak carefully, and qualify my expertise. So how did I ever claim to know the process of “science”? How does anyone?
I know how strange this idea sounds. I understand how hard it will be to swallow. Somewhere along the way we all cultivated a false confidence in our knowledge and understanding. Something about our training makes us believe we can speak for and explain “science.” No one taught us that in the year 2013, individual scientists are a minuscule part of a $1 trillion, 5-million-person global enterprise. No one reminds us that we contributed to only a few of the over one million papers published this year.
To get a sense of how we might approach this problem constructively, it might help to take the sports analogy even farther. Perhaps we should consider ourselves scientists in the same way basketball players are considered athletes. Michael Jordan rarely spoke for all of basketball, much less for soccer, cricket, or all of sports. Whatever his accomplishments, his opinion was just that—his opinion. For all his greatness, Michael Jordan was a small part of something much bigger. We all would have a cramped and deeply flawed image of sports if we forgot that there are hundreds of sports, all with different rules and conventions, requiring athletes with different skills and abilities.
Each scientist too is just a small part of something much bigger, offering only one individual’s opinion among many. And as with sports, we should not forget that there are hundreds—thousands—of sciences, with different rules and conventions, requiring scientists with different skills and abilities, using different methods, standards of proof, and types of evidence. Any one of us can provide only a single perspective.
So when speaking to nonscientists, rather than grandiose proclamations about the scientific enterprise or the process of science, try to make simpler, more specific, and more human ones about your own research. Your research, the particular scientific methods you use, the nooks and crannies of your work, your personal journey to your own small corner of science are wonderful and awe-inspiring by themselves. They are the corner of science that, if you’re fortunate, you love, and that you know well enough to explain in a way that is compelling because it is your own.
If science is “about” anything, it is about scientists. It is a profoundly human story. It is your story. If we remember that, then we can all help the public gain a deeper, richer understanding of science.