How did Darwin decide which book to read next?

A new study published in Cognition blends information theory, cognitive science and personal history

By Christian Jarrett

Between 1837 and 1860 Charles Darwin kept a diary of every book he read, including An Essay on the Principle of Population, Principles of Geology and Vestiges of the Natural History of Creation. There were many others: 687 English non-fiction titles alone, meaning that he averaged one book every ten days. After Darwin finished each one, how did he decide what to read next? In this decision, a scientist like Darwin was confronted with a problem similar to that afflicting the squirrel in search of nuts. Is it better to thoroughly search one area (or topic), or to continually jump to new areas (topics)? Foraging, whether for nuts or information, comes down to a choice between exploitation and exploration. In a new paper in Cognition, a team led by Jaimie Murdock has analysed the contents of the English non-fiction books Darwin read, and the order he read them in, to find out his favoured information-gathering approach and how it changed over time.

The researchers analysed digital versions of 665 of the books Darwin read, counting and categorising the topics mentioned within. From this, they calculated whether each successive book reflected a foray into surprising new territory or a deeper dig into a similar subject area. They found that Darwin began his note-keeping period with a greater emphasis on exploitation, largely tending to master one area at a time, rather than jumping sporadically from topic to topic. Over time, and as we get nearer to his writing of his masterpiece On The Origin of Species, he shifts to a greater emphasis on exploration: more sporadic and surprising jumps to new topics.

This finding is unexpected in the sense that it is usually considered optimal for a forager – be that a squirrel or academic – to show the opposite pattern to Darwin, to scope out many different areas before later zooming in on a particular area for deeper scrutiny. Murdock and his team propose two explanations. First, perhaps Darwin had already shown a stronger emphasis on exploration earlier in his career, before he began keeping a reading diary and so missed by the current analysis. Or second, perhaps Darwin required an earlier phase of exploitation and mastery of narrower areas as a way to build his confidence and hone his abilities.

Another way the researchers approached Darwin’s reading choices was to see whether the order he went from book to book was more surprising, in terms of topic changes, compared with the natural progression of topics as reflected in the order in which books were published. This showed that his chosen journey through the reading material was more exploratory and surprising than the chronological progression of ideas based on publication history. That’s to say, over the course of his career, he deliberately mixed and matched what he read, bringing disparate ideas together, rather than passively encountering new ideas as they emerged.

Lending validity to the new findings are the fact the identified phases of relative exploitation and exploration in Darwin’s reading map well onto the biographical phases in his career. First, as he assembled his research journals from the Beagle voyage from 1836 onward, he was in a clear exploitation phase of reading. Then from 1846, he began a period of intense research into barnacles (having realised their taxonomic study had been neglected in the literature), and this coincides with a shift in his reading style from exploitation to exploration, especially in the sense of reading books on topics that were new to him compared with his past reading habits. Finally, from 1854, as he began preparations for writing Origin, Darwin’s reading enters its most exploratory phase, in terms not only of reading topics that were fresh relative to his past reading material, but also literally fresh from one book to the next.

The researchers said their study represented the first time that information theory has been applied in this way to study the reading and research habits of a single individual. In principle, the same kind of analysis could now be applied to any other individual living or dead, for whom their reading history is available. In time it will be fascinating to see what similarities and differences there are between Darwin’s intellectual journey and those embarked upon by others.

Exploration and exploitation of Victorian science in Darwin’s reading notebooks

Christian Jarrett (@Psych_Writer) is Editor of BPS Research Digest

One thought on “How did Darwin decide which book to read next?”

Comments are closed.