This Day in Philosophy — 18 August: helium was discovered in the sun

On 18 August 1868, while a total solar eclipse was being observed in India, the French astronomer Pierre Jules César Janssen saw in the sun's spectrum a yellow line that no one had recorded before. The English astronomer Norman Lockyer detected the same line independently, proposed that it belonged to a new element, and named it helium after the Greek word for the sun.

The element itself was not found on earth until years later.

How do we know something we have not seen?#

The discovery reduces a central question in the philosophy of science to a concrete case.

Janssen and Lockyer did not see helium. What they saw was a bright line at a particular place in light passed through a prism. Getting from there to the conclusion that a new element exists requires a whole set of theoretical assumptions: that every element has a spectral signature of its own, that matter in the sun obeys the same laws as matter on earth, that the instrument is reliable.

Observation, in other words, is not independent of theory. That is precisely the claim Norwood Russell Hanson and Thomas Kuhn made in arguing that observation is theory-laden: what we see depends on the theory we bring to the looking.

The argument about realism#

The helium case is cited in the debate over scientific realism too.

For the realists it is a story of vindication. A theory predicted the existence of something no one had yet held in their hands; years later the thing was found. That shows the theory does not merely work — it genuinely describes the world.

The instrumentalists are more guarded. Theories are instruments for organising observations and generating predictions; their success does not prove that the entities they describe really exist. The history of science is full of concepts that produced successful predictions and were later abandoned, the ether being the best known of them.

The argument in its present form goes by the name of the pessimistic induction: if most of the best theories of the past turned out to be false, why should today's be any different?

When the sky became a laboratory#

The third consequence of the discovery is methodological. Spectral analysis opened a way of producing knowledge about an object that cannot be touched. Almost the whole of modern astrophysics, from the atmospheres of exoplanets to the neighbourhood of black holes, rests on that principle.

Early in the nineteenth century Auguste Comte had written that the chemical composition of the stars could never be known, offering it as an example of a limit of knowledge in principle. The discovery of helium showed how difficult it is to know where such a limit lies.