Skip to content
Science, Maths & Technology
  • Video
  • 10 mins
  • Level 1: Introductory

Gazing into the future

Updated Wednesday 26th May 2010

Professor of Ecology, Jonathan Silvertown, discusses science in an uncertain world, and the difficulties of predicting the future when life is so complicated.


Copyright The Open University


The world is changing and as an ecologist I’m interested in how that’s happening and why. One of the tools that we use for looking at such questions are long-term experiments, experiments that have been set up outdoors in the wild twenty, thirty years ago. Now if we think forwards about what the world will be like in fifty or hundred years’ time, one thing we know for sure is that scientists will be asking the same kinds of questions: how did we get where we are now? And to answer that question they will need long-term experiments. But how are those going to be set up? Which government is going to provide money for a fifty or a hundred year long experiment? None, of course, and so what we need to do, and what we have done, is start a new charity called the Ecological Continuity Trust which is going to raise money for experiments that future generations of scientists can use to understand the predicament that we are leaving them in.

Love science? There's something for you in The Open University's range of science short courses

‘Prediction is very difficult, especially about the future’.

This witticism has been variously attributed: to the baseball player Yogi Berra, to the film producer Sam Goldwyn and to the physicist Neils Bohr. Bohr certainly made the comment, but he himself attributed it to a fellow Dane called Robert Storm Petersen, who may (or may not) have got it from Mark Twain. I find all this uncertainty about a quip pointing out how uncertain the world is deliciously apt, but as a scientist it is also a challenge to my belief that there ought to be something useful one can say about the future.

Science fiction entertains us with imagined futures; science itself is also predictive, but must be more circumspect and attaches probabilities to alternative scenarios. Martin Rees, this year's Reith Lecturer and as distinguished a scientist as you could hope to find, dares look into the future further than most scientists and the prognosis of his book, Our Final Century, is bleak. If you happen to live on the other side of the Atlantic, you could be forgiven for thinking that the book's prognosis is even worse because there the book has the title Our Final Hour.

There is a degree of uncertainty in every prediction, so how can science say anything useful about the future? There are three sources of scientific knowledge that can be used to place boundaries of likelihood around a prediction:

  1. Extrapolation – ‘the sun will almost certainly rise tomorrow because I saw it rise every day last week’
  2. Understanding of mechanism – ‘the rising of the sun is caused by the rotation of the Earth, so as long as Earth rotates the sun will rise’
  3. Experiment – ‘if I construct a model solar system and place a model of the moon in orbit around the model Earth, I can predict the location of an eclipse of the sun to within 100 metres and its timing to within 1 minute’

These examples are deliberately easy ones and do not illustrate why prediction is so difficult. In fact, I have heard Martin Rees say that the really difficult science is not astronomy but biological science, because living things are so complicated. These complications mean that biologists have to depend much more on extrapolation than any modern astronomer needs to do. Indeed, biologists and earth scientists use the fossil record to study long-past episodes of climate warming. This is a means of predicting what may happen in the future, as Earth's global climate warms as a consequence of human activity. The past is often used as a guide to the future.

A spiral seashell fossil Copyrighted image Icon Copyright: Thinkstock
A spiral seashell: Fossils and their occurrence within the Earth's rock strata are known as the fossil record
[Image copyright: Thinkstock]

Life's complicatedness also limits the degree to which we can use knowledge of mechanisms to predict the future. This is not just because the mechanisms are hard to work out in the first place, but also because the behaviour of complicated mechanisms, even when you know how they work, can be inherently unpredictable. Nonetheless we are making progress in understanding them.

A lot of the progress in understanding how ecological (and other biological) systems work comes from the kind of investigation that is not available to astronomers: field experiments. In a field experiment, you change one factor at a time in a small part of the real world in order to see what effect it has on the behaviour of the system. This is then compared to a 'control' that has not been changed. As an ecologist, the systems I am interested in are ecosystems – interdependent communities of plants, animals, microbes and their physical environment.

The world's longest-running ecological experiment is in a meadow at Rothamsted Experimental Station in Harpenden, Hertfordshire. It was started in 1856 by John Lawes and Joseph Henry Gilbert. Lawes was a pioneer of scientific agriculture and he employed Gilbert, an agricultural chemist, to help him investigate plant nutrition. The two scientists set up a whole series of field-scale trials that examined the effects of different fertilizers on a variety of crops, including hay – which was the red diesel of those days, providing the fuel for all the motive power on the farm... horses. Today, traditional hay meadows, including the control plots of the experiment in the Park Grass meadow at Rothamsted, are valuable remnants of the flower-rich grasslands that were once found on virtually every farm, but which are now very scarce.

In the time that the Park Grass Experiment has been running, it has taught us a great deal about how grassland ecosystems work and these results have been reported in some 400 scientific articles. One of the most important findings from the early days of this experiment is that adding plant nutrients, particularly nitrogen, encourages the growth of a few vigorous species at the expense of other plants and causes a dramatic fall in the number of species present. Fertilizers must be very sparingly applied, if the few flower-rich grasslands we have left are to be preserved. Because Park Grass has been running for so long, and because the original meadow and those parts of it that have not been heavily fertilized are so full of different plant species, it is a unique outdoor laboratory in which the process of extinction can be studied. This is particularly relevant in 2010, the International Year of Biodiversity, when species extinction is so much in the news.

I have been using the Park Grass Experiment in my research for many years. Five years ago, some friends and I decided that we should be thinking about the long-term future. Our own scientific research has benefitted so much from the foresight of our Victorian forebears over 150 years ago, we ought to be thinking 150 years ahead to what scientists of the next century will need in order to understand the environmental predicament that we know they will inherit from us. If scientists of the future asked: ‘Ecologists back in the 21st century knew the world was changing, they used long-term experiments in their research, but what resources did they leave us to understand how the world has changed since their day?’ The answer will be a shameful 'none' unless we start some new experiments now. What is most needed is a set of experimental platforms that can be used to study and predict the effects of climate change on ecosystems. We are not so naive as to think that any government agency is going to provide us with a grant for a 150-year project, so we have done what John Lawes did to protect his experiments: we have created a charitable trust for the purpose. The Ecological Continuity Trust will help with the hugely challenging, but important task of fulfilling our scientific debt to the future.


For further information, take a look at our frequently asked questions which may give you the support you need.

Have a question?