Physicist as librarian; Henri Poincaré’s intriguing metaphor

Among the new additions to Oxford University Press’s admirable series of Very Short Introductions is a revised edition of J.L. Heilbron’s VSI to the History of Physics. An interesting read in general, it raises one intriguing idea in particular; the metaphor of physicist as librarian.

Libraries and documents feature throughout the book. Heilbron notes that when the Mongol ruler Hulagu Khan created an astronomical observatory for the scholar Nasir al-Din al-Tusi in the city of Maragha in modern Azerbaijan in the thirteenth century, its facilities included a library and a librarian (p.38). This shows an interesting and different side to Hulagu, often demonised for the destruction of the great library of Abbasid Baghdad in 1258. And Heilbron reminds us that from the fifteenth century, printing, by replacing imprecise and error-prone manuscript copying, gave standardised versions of texts, giving scholars the same materials on which to develop exact and quantitative physical ideas (p.64).

However, the specific idea of physicist as librarian, which Heilbron uses as the title of a section in the his book, was first set out by Henri Poincaré, the French mathematician, in his keynote address to the first international conference of physicists, held in Paris in 1900: “He advised his audience against developing a fondness for a revised-Cartesian or any other world system. rather, they should collect the facts of experiment and arrange them for consultation in the most convenient manner, a good physicist was more librarian than philosopher …The metaphor of the library suited a large part of 19th-century physics, which boasted many new laws or effects easily entered in Poincaré ‘s imaginary catalogue” (p.100-101).

Helibron takes up enthusiastically the physicist as librarian metaphor throughout the rest of the book:

• “The library adds new shelves” with the discovery that mechanical forces can create electricity and magnetism (p.101)
• “The most exact entries in the library of physics” around 1900 were the precisely measured wavelengths of spectral lines emitted by elements when heated sufficiently (p.101)
• “The librarians of physics … already had a pace for [the blackbody radiation problem] on the shelf between electrodynamics and thermodynamics” (p.105)
• around 1900, physics got its first elementary particle [the electron] and half a dozen rays for which its library had no shelf mark” (p.108).

Poincaré’s metaphor was examined in detail, and criticised, in a book chapter by Antonio Videira (in Poincaré Philosopher of Science: problems and perspectives, de Paz and DeSalle (eds.), Springer, 2014), and, in an unpublished paper Physics and Philosophy meet: the strange case of Poincaré (2014), by Howard Stein, an American philosopher of science. Poincaré had argued that science could be compared to a library, whose collection should brow continually, but for which insufficient funds were available, so that purchases should be carefully chosen. It was the job of the experimental physicist to make the purchases and increase the size of the collection. The theoretician was responsible for creating and maintaining the catalogue; while this does not increase the collection, it allows the reader to best use of what is there, and by showing gaps in the collection it shows what purchases, i.e. experimental results, are needed.

The job of the theorist, in Stein’s restatement of Poincaré’s idea, is to find the simplest and most systematic arrangement for storing scientific information, in effect that of information retrieval, and transformation of the retrieved information into usable form; the information itself having come from experiment (p.16).Calling this a “lucid metaphor”, but concluding that it is erroneous, Stein suggests that it would have been better to regard theoretical physics as writing not the library catalogue but the books themselves, which experimental physics judges which to keep in the library and which to weed (p.17).

At all events, it is an intriguing metaphor, particularly when set against the more common view, usually ascribed to Ernest Rutherford, that all of science is divided into physics and stamp collecting, with the implication that physics, unlike the other sciences, does not simply collect observations and experimental results.

It is also interesting to note that Heilbron’s book talks much more about libraries than does the equivalent Very Short Introduction for chemistry, which has generally been considered a more information-intensive science than physics, though the latter book certainly includes a good deal about chemical information and communication.

Even if the metaphor may be stretched too far, that someone like Poincaré would use it reminds us of the very clear link between disciplines and their knowledge base, and indeed their documents.