If any researchers ever seemed well-equipped to write a popular book, they would be neuroscientists Stephen Macknik and Susana Martinez-Conde. Ten years ago, the husband–wife team launched the 'Best Illusion of the Year Contest', which honours magic tricks and other perceptual illusions. In 2007–08, they wrote two feature articles for Scientific American and started a monthly online column there about the neuroscience of illusions. A few publishing companies took note and contacted the pair about writing a book.

Credit: Adapted from Hong Li/Getty

But Macknik and Martinez-Conde, now at the State University of New York Downstate Medical Center in Brooklyn, were hesitant. They felt comfortable producing 2,000-word articles, but the idea of writing 8,000-word chapters, braiding multiple narratives and developing an overarching structure was intimidating. “We didn't know if we could handle a book,” says Macknik.

Then, at a conference, the pair met Dan Ariely, a best-selling author and a behavioural economist at Duke University in Durham, North Carolina. He introduced them to his literary agent — who, in turn, suggested that they work with freelance science writer Sandra Blakeslee, who had co-authored several books, including one with a noted neuroscientist.

The three hashed out a plan for a book about the neuroscience of magic, and the proposal garnered interest from more than a dozen publishers. The team eventually went with Henry Holt and Company (part of Macmillan Publishers, which owns Nature). Over the next 9 months, the scientists wrote 500- to 2,000-word chunks and sent them to Blakeslee, who pushed for clearer explanations, revised the text, wrote additional sections and assembled the vignettes.

The result — Sleights of Mind: What the Neuroscience of Magic Reveals About Our Everyday Deceptions — hit the shelves in 2010. Although not a huge commercial success in the United States, it sold well in the United Kingdom and Spain, led to more speaking engagements for the scientists and won positive reviews and an award. Their confidence boosted, they are now on book number two.

Many scientists eager to write a book similarly wonder if they can do it on their own. Partnering with a professional writer, often a science journalist, can ease the pressure and vastly improve the manuscript. Writers can help to translate complex concepts for the public, construct compelling narratives and organize tens of thousands of words into a clear structure. Many also do a good deal of the required research. And the use of a co-author can also motivate the scientist to meet deadlines.

Good with the bad

But co-authorship comes with challenges. The scientist will not necessarily save time: working with writers requires extensive discussions and review of many drafts. The co-authors may disagree on what phrasing or details to include. And a skilled co-author is not cheap: the writer usually expects a flat fee (which can run as high as a couple of hundred thousand dollars), a share of payments from the publisher, or both. Still, it can be a rewarding — if not highly remunerative — experience. To best protect their interests and to streamline and smooth the process, scientists who aim to pursue this route should seek an experienced writer with a compatible personality, explicitly plan the schedule and division of labour before the project begins, consider the pros and cons of sharing a literary agent and line up a lawyer to review the contract.

Strong starts

If the scientist brings a co-author on board early, the writer can help to prepare a marketable book proposal and craft an enticing sample chapter. Usually, a writer charges about US$5,000–10,000 for this work, which can take a few months. Although some researchers may hesitate to invest that amount up front, the ultimate deal — and amount of the advance — hinge on a strong proposal. “It's what will sell the book,” says Blakeslee.

You've just got to get on with it. Teamwork is very effective at driving it along.”

To find a co-author, scientists can contact journalists who have interviewed them, written about similar topics or been recommended by colleagues. Whether candidates must have experience in writing books is debatable — Blakeslee believes that it is not essential, but Macknik argues that since the project will be time-consuming and risky, scientists should seek writers with a strong track record in book authorship.

If the scientist's first choices are not available, those writers can often recommend others, says Thomas Hayden, a science communication lecturer at Stanford University in California who has co-authored two books. Researchers can also post ads with writers' organizations, such as the US National Association of Science Writers (www.nasw.org) or the Association of British Science Writers (www.absw.org.uk). Blakeslee recommends that a would-be scientist-author interview four or five candidates face to face, look for a good personality fit and evaluate samples of their work for accuracy and flair. It will not be clear how heavily the writer's work was revised by an editor, but an interview will help the scientist to gauge the candidate's intelligence and capabilities.

The writer with the most recognizable name is not always the best choice, says Hayden. Instead, the scientist should consider whether the writer asked thought-provoking questions during their meeting or previous interviews. “You want somebody who can go as deep as you can into the material,” he says.

Scientist and writer also should agree on the approach to the subject. That was important to David Spiegelhalter, a statistician at the University of Cambridge, UK, who co-wrote The Norm Chronicles (Profile, 2013), a book about risk assessment in everyday life. He and his co-author, UK journalist Michael Blastland, had previously given joint lectures about how to communicate statistics and shared the view that experts should not belittle the public's understanding of risk. So when the two decided to write a book together, they knew that they agreed on an approach: both wanted to respect readers' gut feelings about risk but to help them to balance instinct and analytical thinking. The collaboration was smooth, and the book won acclaim for its avoidance of a patronizing tone.

The scientist will also need to find an agent (who typically takes a 15% commission on payments from the publisher). To find one, they can search the acknowledgements pages of similar books or ask scientist-authors for recommendations. The website Publishers Marketplace sells reports about which agents have recently sold books, including those on science, around the world.

Sharing an agent with the co-author can simplify matters by reducing negotiation time and the number of people involved in decisions — but it also carries risks. If the relationship later falls apart, the agent is likely to take the side of the most important client, which might be the writer, says Madeleine Morel, a literary agent at 2M Communications in New York City. The scientist should also retain a lawyer who specializes in publishing to review the collaboration contract, which should spell out issues such as payment (see 'Dividing the spoils'), cover credits, division of labour and deadlines.

Once the contract and book deal are in place, the real work begins. Sometimes, the scientist provides most of the ideas and resources for content — such as the subtopics each chapter should cover, lists of studies, articles or videos of lectures. And the writer may spend many hours interviewing the scientist to learn key concepts and gather anecdotes.

But the division of labour varies widely, and writers, too, can bring scientific expertise to the collaboration. Blakeslee had more than two decades of experience covering brain science when she began working with Macknik and Martinez-Conde, and she was more familiar with some neuroscience concepts — such as peripersonal space, the space around our bodies that can be reached by our arms — than the scientists were. Because following the news is part of a journalist's job, the writer may also be better at keeping up with the latest research in the field. This skill impressed Douglas Emlen, an evolutionary biologist at the University of Montana in Missoula, during his collaboration on the textbook Evolution: Making Sense of Life (Roberts, 2012) with Carl Zimmer, a science writer in Guilford, Connecticut. “He brings examples to the table that never in a million years would I even be aware of,” says Emlen.

Yet even if the journalist is responsible for all the writing, a scientist cannot expect him or her to produce a book alone. “It doesn't happen by magic,” says Catherine Dold, a freelance science and health writer in Boulder, Colorado. In addition to interviewing the scientist for concepts and anecdotes, the writer will need frequent feedback on chapters. Blakeslee recommends that the co-authors meet in person regularly to work on the book, maintain rapport and ensure clear communication. If the scientist weighs in only at the end, the writer may focus on points that the scientist would not have chosen to highlight, or might neglect areas that the scientist considers important.

It is also crucial to stick to the schedule because book-publishing schedules tend to be less flexible than those many scientists are used to. If one partner falls behind, the other team member or members may have trouble adjusting their schedule. And the publisher may cancel the book if the manuscript is late. Spiegelhalter found that having a co-author made him less likely to procrastinate.

“You've just got to get on with it,” he says. “Teamwork is very effective at driving it along.”

Researchers should expect disagreements to arise, especially over wording or the appropriate level of technical detail. They must ensure that the text is accurate, but should also recognize the writer's expertise in communicating to a lay audience. Writing engaging chapters that would maintain a reader's attention “required shattering a lot of the conventions that I'm comfortable with”, says Emlen. Zimmer used more active verbs and shorter sentences than Emlen was used to, for example.

But if the scientist begins to feel that the relationship is foundering, it is better to terminate sooner than later, says Morel. The collaboration agreement should contain a termination clause that specifies what will happen if the partnership dissolves. The scientist might keep the copyright to the text, for example, while the writer keeps any payment received so far.

Well-matched co-authors can avoid such pitfalls. Emlen says that it was “a dream” to work with Zimmer on the evolution textbook: Emlen provided deep background knowledge of the field, and Zimmer conveyed the material with compelling stories and clean, accessible language. “I spent a lot of the past few years realizing just how hard it is to write like that,” says Emlen. At the end of the day, whatever the bumps along the way, both parties want the same thing: to write a great book. “You're in this together,” says Blakeslee.