The Ditherington Mill

Book Reviews

From an 18th-century revolution in building design to a 21st-century revolution in political activism, via a frank look at how scientists learn from accidents and mistakes

Oxford University Press

Failure: Why Science is so Successful

By Stuart Firestein, £14.99, ISBN 9780199390106

Any parent who’s made the mistake recently of trying to help a child with homework by suggesting that what’s needed is a ‘trial and error’ approach may have learned that it’s now known as ‘trial and improvement’.

Whatever you think of the semantic change, the new description is definitely in the spirit of Stuart Firestein’s fascinating new book. A professor of biological sciences, Firestein has become notorious for a TED Talk in which he describes his experience of how scientists work as being akin to “farting around”. For him, the history of science isn’t an incremental tale of hopping from breakthrough to breakthrough. Instead, he maintains, it’s more like a continuum of varying degrees of success - and equally failure - that from time to time culminates in a major advance in knowledge.

Does this matter? Well, when so much influence is wielded by politicians and other decision- makers who are likely to judge how funding is being spent based on whether or not researchers have found something new, then yes, it matters.

Another bugbear of Firestein is the perennial issue of public understanding of science - or lack of it - which begins at school when we’re taught the scientific method of coming up with an idea based on previous observations, then carrying out an experiment to see whether we’re right. Most of the time experiments don’t work, so most of science is negative results. It’s just that no one bothers recording them. Firestein has little time for this approach, believing that it often stifles creativity and puts youngsters off when they start to discover how incompatible it is with the real world of innovation.

The fate of Francis Bacon, who was largely responsible for codifying the approach of observing, hypothesising and experimenting, illustrates the paradox well. One account of his death relates that while travelling by coach he was struck by the idea that freezing might be a good way of preserving meat. He immediately stopped and, having acquired a chicken, stuffed it with snow. Shortly after, he contracted a fatal case of pneumonia.

The chilly conditions of the experiment and his death may have been a coincidence. What’s important is that there was nothing in Bacon’s previous experience or ‘observation’ that would have led him to believe that low temperatures could help to preserve meat but it seemed like a good idea at the time. No one’s quite sure where the idea came from, and this random approach is more common than we teach young people.

‘Failure’ is a short, polemical book which benefits from not being weighed down with hundreds of examples, and there are many that Firestein could have chosen from all branches of science to prove his point. More than a history, it’s a well-argued case for changing the way we teach science to make it more appealing than simply acquiring a body of knowledge, and at the same time reconsidering our own view of the breakthroughs that have created the modern world.

As novelist, poet and playwright Samuel Beckett put it in an aphorism adopted by Firestein to introduce one of his chapters: “Ever tried. Ever failed. No matter. Try again. Fail again. Fail better.” A delightfully accurate description of how science works in the real world.

Dominic Lenton

Princeton University Press

Political Turbulence: How Social Media Shape Collective Action

By Helen Margetts, Peter John, Scott Hale and Taha Yasseri, £19.95, ISBN 9780691159225

The idea that the Internet has encouraged people to get involved with politics and pressure for social change won’t come as news to most of us. But whereas a single retweet or approval of a Facebook post carries little weight on its own, pervasive social media makes it easier for shared opinions to rapidly gain momentum until the many small voices have become a single noisy one.

In this environment, movements that once might have taken months to acquire critical mass can rise and fall within hours. The resulting turbulence is dubbed by the authors of this book ‘chaotic pluralism’. The model of democracy that is emerging, they argue, has evolved from the ordered, organised form of discourse described by the early pluralists into a whole new form of politics.

At the same time, the notion of political engagement being associated with a higher income and better education is changing. As people from all backgrounds spend increasing proportions of their lives online, they’re being invited and encouraged as never before to support myriad political causes, whether by sharing, liking, endorsing, or downloading. The resulting chain-reactions form a growing part of collective action.

As the authors of ‘Political Turbulence’ acknowledge, social science doesn’t yet understand the ecology of mass mobilisations. Some undoubtedly work, on scales from local campaigns to global political movements and even regime-changing revolution, but many more fail. Why do some succeed? And how genuinely resilient are the networks behind them? This timely scientific analysis is an attempt to combine a psychological perspective with the quantitative analysis of social media that is made possible by ‘big data’ techniques.

Their foundation is that users interact with social networks in two fundamental ways: as a source of information and opinions; and as a way of demonstrating to others what they think and what they’re doing about it. Carefully structured experiments test how social media influence citizens deciding whether or not to participate. The results illustrate how different personality types react to social influences, and identify which types of people are willing to participate at an early stage in a mobilisation when there are few supporters or signals of viability.

The conclusion is that data science and experimentation with social data is capable of providing a toolkit for understanding, shaping, and perhaps even predicting the outcomes of a state of democratic turbulence which shows no sign of diminishing. Perhaps as important as learning how and why some movements gain traction online is keeping up to speed with the never-ending new environments where this takes place. Facebook and Twitter have been effective, but they’re being superseded by other environments that users find better suited.

As ‘Political Turbulence’ makes only too clear, governments and security services will have to come to terms not just with the technology driving this new age of activism, but also with how it changes the whole way in which a previously indifferent public is going to become increasingly involved in trying to influence policy.

Dominic Lenton

Recent articles

Info Message

Our sites use cookies to support some functionality, and to collect anonymous user data.

Learn more about IET cookies and how to control them