It’s about the time of year when prospective graduate students are making one of the most important decisions of their lives: where to go to grad school. So we really should give some advice about that, but happily we already have! And it still seems pretty relevant. Meanwhile, today I’m at the KITP in Santa Barbara, speaking on a panel on The Perils and Pitfalls of Speaking to the Press. (One in a series organized by the KITP’s Journalist in Residence.) So I have to give a short talk about that, and thought I could take advantage of the opportunity by turning it into a blog post.
Sadly, I eventually realized that I do not have a Grand Unified Theory of interactions between scientists and journalists. It is a complicated relationship, in which there is much overlap in objectives on both sides, but also undeniably some tensions here and there. Consider the following two anecdotes:
- My first direct interaction with the science press was as a grad student, when I was working with Edward Farhi and Alan Guth on whether it was possible to build a time machine out of cosmic strings (as proposed by Richard Gott). Our work was written up in Science News, and they did an extremely careful job — Ron Cowen interviewed us in depth, asked good questions, and the magazine even sent us a draft copy of the article to check for accuracy before it was printed. (That almost never happens, don’t expect it.) But when we saw it in print, an editor had helpfully inserted just one new sentence to make things more clear — explaining that open universes were ones that would expand forever. Except that we were working in the slightly unusual context of 3 spacetime dimensions, not the usual 4, and in that case open universes don’t really “expand” at all. Good intentions gone awry.
- I was once in the audience for a panel featuring David Kestenbaum, a science reporter for NPR. He played us a tape of a radio journalist talking to a scientist about the fear of avian flu spreading from the Bronx Zoo. The scientist babbled on at length about open systems and complex environment and disease vectors in a rapid-fire stream of utter incomprehensibility. The journalist stopped him for a second, and basically said “Look, cutting to the chase, does the zoo pose a danger?” The scientist said “No, absolutely not.” “Okay, could you say that directly?” “Sure, no problem.” And then the journalist asks the question again, to which the scientist — well, you can guess. A rapid-fire stream of dense jargon, in which the word “No” never appeared. Completely useless for the radio.
As far as the Very Big Picture is concerned, scientists and journalists are on the same side. We all want to tell interesting and true stories to a wide audience. But when it comes to specifics, aims and competencies often diverge. Understanding what each others’ goals and constraints are can definitely help to make for a better final product.
So here are some things that I, as a scientist, have figured out about what journalists want. At least I think I have figured them out; actual journalists are welcome to jump in and explain what they really want in their own words.
- Journalists want stories that are interesting. This goes without saying, but for better or for worse the judgment about what is “interesting” may be different from person to person, and from scientist to journalist. I was once interviewed about a paper I had written, in which I mentioned that we had calculated a number that was possibly the smallest positive number ever to appear in a physics paper. To me it was just a joke, but the journalist seized on it and wanted to make it the centerpiece of the story. I thought that our speculations about the origin of the universe were really more important than that, and I didn’t even know for sure whether any other smaller numbers had appeared, but to no avail.
- Journalists want stories that are understandable. Again, pretty obvious, but again a danger lurks, and here is a case where a conscientious scientist can do some good. It is often possible, as we all know, to string together a series of words that conveys the illusion of understanding, without actually conveying any real information. To a journalist who is not an expert in the field, it can be hard to tell the difference. Scientists should be careful to ensure that the explanations they give are actually increasing the amount of understanding in the recipient’s brain, not simply providing a warm feeling of being smart.
- Journalists want statements to be tangible. Abstract thought is a necessary component of being a physicist, but we are used to taking leaps in ways that non-experts are not. (Back when you were learning various bits of abstract math and physics for the first time, was it all clear to you immediately, or did it take some practice?) So journalists feel extremely grateful when a lofty-sounding idea can be brought down to earth with a useful analogy or illustration. No professional cosmologist should ever be surprised when they are asked “What is the universe expanding into?“, and they should have an answer ready.
- Journalists are not in the business of science education, generally speaking. This seems puzzling to scientists, who like to see a greater understanding of all of science, both the brand-new parts and the more established parts. But more than one journalist has tried to explain to me that they’re job is reporting news, not providing a general education — the same would hold for reporting on economics or politics. Of course, as scientists, if we can figure out a way to educate people about a general principle by relating it to a specific news story, all the better for everyone.
- Journalists are not in the business of allocating credit. Another sticking point for scientists, who (as academics more generally) live and die by an allocation of credit. Journalists don’t want give credit to the wrong people, of course, but doling it out to every possible person in precisely appropriate measure is not their primary concern. Scientists will generally want to see all of the collaborators on a paper being mentioned in a story, and funding agencies certainly want to carefully distinguish who made what telescope and so on. And that’s not even counting what happens when more than one group is responsible for something, and a story only mentions one. It makes sense to try to strike some sort of middle ground here, giving credit as fairly as possible, but scientists need to understand that a detailed list of all the people involved in some piece of work is generally not what a journalist is looking for. (Maybe in a magazine article, unlikely for a newspaper, and essentially never for broadcast media.)
But there is one goal that is worth separating out from the others, and should be shared by scientists and journalists in equal measure.
- Everyone is interested in saying only things that are true, not things that are false. Well, sure. But it’s worth emphasizing, as there are so many other pressures — on both sides! — that the truth occasionally gets compromised, and in my view that should never be acceptable. The WMAP headlines we wrote about some time back are a great example. Both scientists and journalists worked hard to turn a scientific result that was undoubtedly interesting and important into a story that was punchy and palatable, and a lot of truth got sacrificed in the process. When that happens, we can only blame ourselves when the public gets confused about what is going on.
So, absent a detailed underlying theory, I can suggest just a few helpful hints to keep in mind when you get called on by a journalist. Again, only idiosyncratic impressions by someone who hasn’t really done this all that often — feel free to chime in with your own.
- Think of what to say ahead of time. This is probably the single most important thing I have learned, and is especially important when you are being interviewed on the air. You might think “It’s an interview, I will just be asked questions and answer them.” Even if that were true, you can very often anticipate many questions, and your answers will make much more sense if you’ve thought about them ahead of time. But a lot of the time the interviewer won’t know the best questions to ask — you are more likely than they are to understand what are the interesting bits, and there’s no reason why you shouldn’t gently guide them in that direction.
- More generally, take an active role. Talk to the journalist about what they have already learned and plan to say. You might be able to make useful suggestions about other people to talk to, or directions to take the story; or you might be able to correct some lurking misimpression.
- Think of analogies and metaphors. Bring it down to earth, and they will love you for it. Tell them that visible matter is “the olive in the martini of dark matter,” and they will want to have your children.
- Use language carefully. One of the hardest things for professionals in any field to remember is how to disentangle their terms of art from ordinary language. When we use words like “energy” or “dimension” or “vacuum,” we have specific things in mind, but so does everyone else — they’re just not the same things. Try to anticipate the connotations your words will have in people’s minds. Even when we use our own technical jargon, we don’t always do so consistently, and it’s worth the effort to sort out the precise meanings of each word we use. (Does “Big Bang” mean a framework in which the universe expands from a hot, dense early state, or a specific moment in time of infinite density and curvature?)
- Put things in context. What is important, less important, known, still speculative? I’m a big believer that it’s good to let the wider world in on the messy process that science really is, showing them our work in progress rather than waiting until we have rock-solid findings to reveal to the unwashed masses. But if you do that, make sure you draw very specific distinctions between what we know (“the universe is expanding”), what we feel has a good chance of being true (“there was a period of early inflation”), and what is a simple speculation, as well-motivated as it may be (“the quantum state of the universe is described by the Hartle-Hawking wave function”).
- Boil it down to the essence. You’re not giving a lecture or teaching a course; you can’t rely on the audience’s attention over the long term. What is the point you would like them to remember a year later? (And if there isn’t one, why are you bothering?) Take that point and express it in a sentence of no more than twelve words. You’l genuinely be doing some good.
Sean wrote:
>
> So journalists feel extremely grateful when a lofty-sounding idea
> can be brought down to earth with a useful analogy or illustration.
Well bless my soul! I thought scientists loathed and despised analogies, considering them unreliable and misleading. But if analogies are OK after all then the sky’s the limit. For example, Veneziano and Gasperini’s pre-Big Bang scenarios could be likened to an Amish drive by shooting (clip clop clip clop clip clop clip clop BANG!)
Ellipsis, New Scientist reports on ideas and results that are out there, that scientists are publishing in journals, presenting at conferences, discussing on blogs or posting on the arxiv. We’re a weekly magazine and I’d love it if physicists were making 52 ground breaking discoveries a year that stood the test of time and were interesting to 850,000 readers every week. But it just isn’t so. We check our stories carefully and if a story stands up, we’ll run it. By “standing up”, I mean if research is published in a reputable peer reviewed journal, if research has been awarded a major prize, a researcher has received a large grant or if other well-respected scientists are taking a piece of research seriously.
Believe me, there’s heaps of research out there that we reject. Of course, we’ve reported results in physics that has later been overturned but, hey, isn’t that what the the scientific process is all about? It’s true that New Scientist hasn’t always gone back and made it clear to readers that the research has been overturned. And we haven’t always made clear if the work in question is peer reviewed or not – we’re making a big effort to address this.
You’ve got to realise that New Scientist, Scientific American and Discover are “leisure reads”. Their main goal is to entertain, not educate or disseminate scientific results. We’ll leave that to educators and journals, such as Science, PRL and Nature.
I have seen a number of times how the different interests of journalists and scientists can result in an article that misrepresents the scientist rather badly, and that hurt the scientist of their project in a very real way. A few experiences like that and no sane person will want to commit themselves to anything quoteworthy or definite.
If a journalist wants to make a piece about a new product or development in a company, they are not going to get access directly to the people responsible for that development; they’ll get to talk to a well-briefed communications specialist – a PR person. If they want to talk about an earlier fire with the fire brigade, they will likely talk to the press spokesman. Even if someone wants to write about something involving a newspaper reporter, the paper emplying them will send forth their PR department, not the journalist.
Scientists should realize it might be a good idea to have PR people of their own, and do most press contacts through them rather than trying to do it themselves. There’s good reasons why almost all other organizations use press spokespeople; it’s perhaps time scientists and their employers did that too.
Valerie, if I still got your attention, maybe you can enlighten me on something. Why is it so urgent to write articles based on fresh preprints, not even yet read by the majority of the community, let alone peer-reviewed? what is going to change in the period of the few weeks the community is going to decide whether or not the article is worth publication? I can see that some of those stories are going to go away, if they are based on preprints that are never, ever going to be publishable (examples are readily available), but isn’t that a good thing?
I wish I could know the criterion USNews uses to rank universities
There are many organizations which rank school differently, which one I should follow ?
Off the topic question:
Is there any lower bound from theory on the confinement scale? or from cosmology?
for example is it possible for a QCD-like theory to exist with a confinement scale as low as a few MeV?
Linda
If you have a blog then you can write a blogposting on the topic and make sure that the journalist will refer to that blogposting for more details.
Moshe,
Let me jump in and give one answer to the question you asked Valerie about writing stories based on preprints. But let me start by asking you a question, cribbing your language: Why is it so urgent to write scientific papers based on fresh preprints, not even yet read by the majority of the community, let alone peer-reviewed?
The path from preprint to peer review to publication, especially in physics with the arXiv, is a long process. With 24/7 news reporting, and a competitiveness among news organizations, we simply can’t afford to wait until all research passes through peer review. Waiting weeks (you’ve got an optimistic view of peer review if you think it will be through in weeks!) is too long to wait, or else we end up writing stories that everybody else published weeks/months/years ago.
That doesn’t mean that we as science journalists don’t consider carefully whether a preprint might be likely to withstand the process of peer review. We ask other scientists what they think of the paper (essentially overseeing an independent peer review process), we look at it through the eyes of trained scientists (you might be surprised how many science journalists have been trained to PhD level in the science they are covering), and we use the judgment we have developed through years of covering the field.
But to answer your question directly – it is about competitive pressure, for good or for bad.
Imagine that you couldn’t write a scientific paper based on a preprint and any time you did work that cited the work of a preprint, you had to hold off on publishing the work until the cited paper had passed peer review. What kind of competitive disadvantage would you be at in the world of science research?
David
Pingback: A bit more unsolicited advice « The Inverse Square Blog
Thanks David, this is very useful. I am afraid that in my mind the situations are not analogous at all . My training allows me to recognize which preprints are reliable prior to them being judged more formally by the community. You on the other hand, at least to a certain extent, have to choose between being prompt and being reliable. I guess it is up to me, and all other consumers of science journalism, to decide which kind of journalism they find more valuable, there is certainly a whole spectrum out there.
Just to second and third one of Valerie’s comments. Science writing is a public good, but not quite in the way most of us engaged in it like to think. Most science writing is, as Valerie writes, entertainment. That’s no bad thing; people learn a lot, and in particular, shape their emotional reality through the ways they entertain themselves. Better good science writing than any number of things I could think of. But it does mean that there is a gap between what a scientist knows to be important about a given body of work, and what a writer can make us of for the very different function to which she or he intends to put it.
I have to say that I bridle a little at Moshe’s last comment as well; from where I sit, it assumes more reflection on the part of the anonymous scientist than is always warranted (i.e., I can recall lots of unjustified gullibility in response to seemingly interesting preprints by pros), and a greater willingness to throw caution to the winds than most good science journalists I know display. A high batting average for significance and accuracy is basically one’s stock in trade — and it is a matter of ordinary practice to keep a few Moshes (e.g.) in one’s hip pocket against that time you need a quick reality check on a story an editor may clamor for.
That there are crappy journalists (and crappy scientists) is a given….which leads to the obvious: the best way to protect yourself as a scientist is to pay attention to whom you speak. Get to know the journalists whose bylines you come to respect, and try to work through them. When someone new comes along, google them. Do your due diligence; these are the people, after all, who form a virtual extension of your lab/research into the wide world. Sometimes someone will come along with such urgency that you can’t do your due diligence, but if you ‘ve paid attention to the difference between people who know what they are doing and those who don’t, you will be better armed to further your own goals.
I’m a specialist journalist, although not in science, and I’d like to say that Sean’s points cover most of the bases. The important thing is to realise why journalists speak to scientists, and what they hope to get out of it. If you approach an interview with those points in mind, you’re much more likely to get what you want out of it, especially when it comes to quotes – you have basically no control over the content of an article, although you can certainly influence it.
From my perspective, and I imagine it’s broadly similar in science journalism, the main things I’m looking for in a quote is one or more of colour, explanation and authority. Colour is (basically) the snappy quote that provides an interesting perspective. Explanation is pretty self-evident, clarification of a technical point by an expert. Authority allows the journalist to put a contentious or subjective view in an article, by virtue of the earned authority of the speaker. So be aware that anything you say that falls into those categories is liable to be used, even if you consider the comment tangential at best.
Another key issue interviewees have is talking to a journalist for an hour, say, and only getting a short quote. This is inevitable, especially in news journalism. We talk to lots of people about a story, and most of the conversation will be used to inform the content of the article, rather than quoted directly. In a limited space, it can be better to keep the quotes pithy and communicate the facts directly.
Incidentally, the Guardian’s Ben Goldacre over at Bad Science has a lot of things to say about the state of science journalism (mostly in the UK, though).
David and Tom,
I realize I should make myself more clear. My question (comment 29) was directly specifically to Valerie who is working for New Scientist, and was concerned specifically with using preprints that are completely new, a week or two old at best when the story is published, which seems to be their niche. This is thankfully not a wide spread practice, I did not mean to imply that it is, and as I mentioned I do enjoy quite a bit of science journalism myself.
I’m also glad to hear that “high batting average for significance and accuracy is basically one’s stock in trade”. That is the part I was missing in all the above “realpolitik” comments about being entertaining and competing in the market place etc. etc.
Moshe,
I regularly discuss on my blog the many examples of misleading, overhyped stories about theoretical physics that appear in New Scientist and elsewhere, and not waiting for peer review is rarely the problem. A large number of bad stories actually come from press releases issued by physicists or their institutions at the time of publication of a paper in PRL or another major letters journal. Such papers have gone through a rigorous peer review, but this doesn’t stop people from putting out a misleading press release about what is in the paper and its significance. Whether or not a paper has been peer-reviewed, journalists have to do exactly the same thing: talk to experts in the subject and try to judge for themselves based on this. For most papers, experts in the subject can quickly judge for themselves what the significance is, and in the few cases where this is not possible, they can tell the journalist this.
Some journalists do a good job of gathering information from experts and putting it together into something sensible, some don’t. But, very often, problematic claims being made in these articles are not the fault of the journalists, but of the scientists themselves making claims they shouldn’t be making.
Hi again, Moshe,
I still stand by what I said earlier as relevant here but want to reinforce the timeframe issue that others have also mentioned. You say you have concern “with using preprints that are completely new, a week or two old at best when the story is published.”
A week or two is a lifetime in journalism. If you wait that long, you might as well forget about covering the story at all in many cases. And if you work for a weekly like New Scientist, then you have to have the story early enough to get it through the editorial and layout processes, and then printed and distributed so that when it appears on newsstands and in subscribers’ mailboxes, it isn’t too old and it hasn’t appeared in every other publication first.
I personally think it is great that people are writing stories based on the preprint server and not just waiting for the weekly Science and Nature press packets to come out to dictate what will be written. And this way, the journalists are going to primary sources rather than writing stories based on a secondhand source like a news release that might have dubious accuracy. (As Peter pointed out.)
Now, I could be biased here because I started writing news articles based on the arXiv before it was even called that (about 10 years ago), and it gave me a great advantage in being able to find fresh interesting news that others hadn’t done to death. Of course, I would always get expert advice beyond my own physics knowledge, and it didn’t seem to lead to lots of dubious stories.
Keeping tabs on the arXiv is absolutely essential now if you want to be covering physics for a news organization.
I’m an ex-science journalist who used to work for Popular Science and Discover magazines (and for Tom on a TV project before that). I recently gave it all up for science librarianship (and health insurance) but I still keep an interest. I have thoughts on some of Sean’s points.
Journalists want stories that are interesting: we want stories that are interesting not necessarily to us or you, but to our editors and readers. We’re insiders and what’s most interesting to us is often not what will get a casual browser to read the article. Part of the journalist side of the science journalist job is to find that hook that will catch the casual browser. I rarely found a scientist’s suggestions on interesting-bits, angles or the like to be helpful.
Journalists want statements to be tangible: if you can’t make a concrete straightforward statement about the implications of your research, we will make a best guess at what you were trying to weasel your way around being quoted as saying and paraphrase it for you. The best way to control your quotes is to actually say them.
Think of analogies and metaphors: please don’t. The art of the metaphor is the at the heart of science journalism; It’s what makes good science journalists shine. Let them do their job. Now, if you’re talking to a plain-old journalist, they’ll probably thank you for the help, but in my experience most scientists are too close to their work to make the abstracting mental leap to create a metaphor that truly illuminates it for the outsider.
Use language carefully: and please try to notice the unspoken assumptions behind your language usage. It took a number of very confusing conversations before I realized that some biologists use “gene” and “protein” interchangeably.
Boil it down to the essence: Ideally, that’s the journalist’s job. When I interviewed a scientist I had studied up and done my background reading so I could understand the full unboiled version or at least keep asking the right questions until I did and took the time to go through it all and then think about it and consider what the essence actually is. (Maybe that’s why I couldn’t make a living at freelancing?) If it’s a longer article, some aspect of your research speaks to some angle on the main topic, but that aspect and angle aren’t necessarily going to be in the boiled down essence of either.
Generally, as for the poor general level of science information presented to the public, I think there are several factors working together. Most importantly, most science journalism isn’t being done by science journalists. There just aren’t that many trained and experienced subject specialists around and often the topics get passed over to the political reporters just as they’re getting interesting. A good science journalist just needs to be told the facts and will take it from there, but others will need the hand-holding and gentle education Sean’s hints offer. Know who you’re talking to and work with them accordingly.
David, I find your perspective valuable, thanks. I’d like to be able to know the system well enough to make good decisions, mainly as a reader but sometimes as a contributer as well. As a reader, knowing the system will hopefully allow me to avoid plain falsehoods, but also those heavy-handed and predictable “infotainment” stories (cure to cancer? details on page 10), which I frankly find as entertaining as an evening in front of C-SPAN…So, if I understand correctly, the shortest the time frame for the story is, the less I should expect in terms of quality, makes sense.
Let me also say that short pieces based on a single source (preprint or a press release) is not the only conceivable form of an article. More enjoyable ones attempt some synthesis of many sources, adding some personal perspective of the writer as well. Those are more creative, take more effort, and I doubt one often get scooped on them. They are also much more fun to read (for me). I understand that it is not reasonable to expect such stories from a weekly publication.
Journalists should bear in mind that a lot of the complaints come from scientists who resent the attention paid to hot topics, to cosmology in particular. When you have been working on some dull technical issue for years, it must be galling to see that journalists can in fact distinguish interesting from boring. You see a lot of this “cosmology envy” on the blogs, with boring people trying to console themselves by sniffing at “speculative” work in that field or at “premature” work on interesting topics in it. In short, science journalists should beware of scientists criticizing work outside their own tiny worlds, or criticizing journalists who ignore boring things.
Pingback: (Stolen Tag Line alert): The Way We Live Now « The Inverse Square Blog
Pingback: Professionals are professionals at Freedom of Science
I would add the following: When writing for the general public, writers tend to hook readers with the personal or social implications of a topic, not the technical implications. Your story is more likely to get written (and read) if it has a human dimension. Ask yourself the question: why do people care about this? What does this have to do with the average person’s life? Does it challenge one of their accepted beliefs about the universe? Does it create controversy around an issue that they care about? Could it affect their environment, their pocketbook, their family life, their sex life, their religious life, or their career? Does it expand the knowledge base that they use to inform their lives? (Note: to expand their knowledge base, you’ve got to build on whatever knowledge they already have, not on the knowledge you’d like them to have. You’ve got to reach out to the edge of THEIR understanding, not expect them to meet you at the edge of yours).
Scientists are trying to tell a technical story; they are obligated to do this. All of their training tells them to keep human passions away from their work. The problem is: writers for the general public are trying to tell a human story. They are trying to make scientific work personally relevant and, in order to do this, they invoke human passions; they simplify complex issues; they use hyperbole and metaphor. They ignore important science for the sake of sexy science. This drives scientists nuts.
The thing is- scientists aren’t alone in this treatment from the media. Look at political coverage, sports coverage, business coverage, entertainment coverage, etc. Every other industry has the same issue. But other industries expect it. In fact, they work to use it to their advantage and manipulate it. Scientists don’t do this. PR spin is reprehensible to people who have been trained in the quest for universally validated truths. The very thing that makes someone an excellent scientist can also make him a bad press representative.