Getting Science Wrong Why the Philosophy of Science Matters

So you think you know what science is? I thought I did. I mean, we notice patterns, formulate hypotheses, gather observations to see if our ideas are supported or not, and discard or accept our hypotheses. And this is what we do. Yet, as philosopher Paul Dicken shows in this lightly written introduction to the philosophy of science, there is no good definition of the scientific method, though there are plenty of misconceptions.

Dicken's narrative prominently features examples taken from the history of physics (Copernicus and Galileo's model of heliocentrism overthrowing geocentrism, Newton's theories on mechanics and gravitation, and Einstein's theories of general and special relativity), with some reflections on creationism and climate change denial thrown in.

In accessibly written prose, Dicken takes us through the problems with Popper's ideas on falsifiability, on which Popper wrote in his Conjectures and Refutations. Because scientists reject unsupported hypotheses, this sets them apart from conspiracy theorists who will not change their mind no matter the evidence marshalled. But every practising scientist knows that when interpreting data there are judgement calls to be made. In practice, data not in support of an idea rarely lead to its immediate dismissal. Faulty equipment, unaccounted for outside factors etc. could all influence your results. But that does not put scientists and conspiracy theorists on equal footing, as scientists have a different attitude and will eventually change their mind if refutations are repeated and convincing.

Dicken links Hume's "problem of induction" and the idea that you can never have enough data with the failures of the Big Data revolution, pointing out that more data is not always better. You can find all sorts of correlations in large datasets, but you still need educated guesswork (conjecture) and explanation to tell apart correlation and causation. And Thomas Kuhn's work establishing that groups of scientists can work within different, broadly accepted frameworks (so-called paradigms – Kuhn's The Structure of Scientific Revolutions is where the term "paradigm shift" originates) is shown to often be confused for relativism. This is the idea that we're all right from our point of view, something that has made a vicious comeback with Trump's declaration of "alternative facts".

In what is probably the most relevant chapter on the misconception of science, Dicken deals with the circus around creationism and its demands for "balanced treatment", and climate change science and "the argument from history". Creationists appeal to the open-minded spirit of scientific investigation to endorse a view that is itself dogmatic. Climate change deniers see the track record of failed environmental predictions (from Malthus's An Essay on the Principle of Population to Paul Ehrlich's The Population Bomb) and extrapolate from there that we therefore have no reason to trust future predictions. Both are examples of what Dicken's PhD supervisor calls "epistemological judo": trying to use principles of scientific methodology against itself (here these principles are open-minded assessment of alternative options and extrapolating from past instances to make predictions about the future). Especially climate change deniers conveniently forget that science is a process of gradual refinement.

Getting Science Wrong was branded as fun and accessible in one review, and I agree with that. For a book on the philosophy of science it is very readable, and Dicken does not shy away from lampooning venerated philosophers and scientists of the past, exposing all their foibles and human quirks. But the book does suffer somewhat in its structure. Writing this review I realised it was hard to summarise what misconception of science each chapter dealt with, and in my opinion the book would have benefited from chapter conclusions. Nevertheless, the book's message is an important one and should serve as a great springboard for further discussion and exploration of the literature, or perhaps even as an introductory read to coursework.