Niels Bohr and the argument about what complementarity claims: 7 October

Niels Henrik David Bohr was born in Copenhagen on 7 October 1885, the son of Christian Bohr, professor of physiology at the university, and Ellen Adler, daughter of a Jewish banking family. He died in the same city, at the Carlsberg honorary residence, on 18 November 1962. In between he gave the atom its first quantum model, in three papers in the Philosophical Magazine in 1913; won the Nobel Prize for it in 1922; opened the Institute of Theoretical Physics on 3 March 1921 and ran it as the capital of the new physics; escaped the German occupation by fishing boat to Sweden on 29 September 1943; and spent his last years arguing, in an open letter to the United Nations of June 1950, for a world in which no state kept scientific secrets.

Any of that would secure the date. What earns Bohr a place in a philosophy calendar is something narrower and stranger: a single idea, announced at a conference on Lake Como in September 1927, about which the philosophy of science has been unable to agree for ninety-nine years. The disagreement is not about whether complementarity is true. It is about what kind of claim it is.

A physicist with a philosopher's upbringing#

Bohr's philosophical education was not an accident of later reading. Harald Høffding, Denmark's leading philosopher, was his father's friend and taught the young Bohr; he was also the first occupant of the Carlsberg residence that Bohr would later inhabit. Bohr read Kierkegaard, sent his brother Harald one of Kierkegaard's books in 1909, and disagreed with a good deal of what he found there; Richard Rhodes and others have argued that the Kierkegaardian either/or reached him filtered through Høffding all the same. The habit of thinking in terms of standpoints that exclude one another, and of the observer who cannot step outside the scene, was in his head before there was a quantum theory for it to apply to.

That matters for how one reads what came next. The physics of 1925 to 1927 had produced two formalisms, Heisenberg's matrices and Schrödinger's waves, which agreed in their predictions and disagreed about everything else, and Heisenberg's uncertainty relations of early 1927 had shown that position and momentum could not both be sharply assigned to the same particle at the same time. Bohr's response at Como was not a new equation. It was a reading of what the equations meant.

What was said at Como#

The Como lecture argued, first, that Heisenberg's relations could be derived from classical considerations about the resolving power of instruments, and then drew the conclusion Bohr would spend the rest of his life refining. A quantum object admits two kinds of description. One is spatio-temporal: where it is, when. The other is causal, in Bohr's sense: what energy and momentum it carries, and so what it will do. Each description requires an experimental arrangement that rules out the other. Light looks like a wave in one apparatus and a particle in another, and the two pictures exclude each other; yet both are needed, since neither alone exhausts what can be known. That is complementarity. It was presented again at the Solvay conference in Brussels a few months later, where Einstein began the long series of objections that the Niels Bohr Institute's own history now calls the Bohr–Einstein debates of 1927 and 1930.

Two features of the doctrine are easy to lose. The first is that Bohr insisted the language of classical physics, position, time, momentum and energy, was not optional. We need it to say what an experiment is and to tell one another what happened. Those concepts are refinements of everyday ones, and the quantum of action limits where they can be applied, but there is no other vocabulary in which a result can be stated. The second is that Bohr did not think the wave–particle pairing was the deepest form of the idea. The Stanford Encyclopedia's account notes that he came to treat the kinematic–dynamic pairing as the lasting case and quietly dropped the wave–particle version, because what matters is that the two attributions rest on mutually exclusive experiments, not that the pictures look different.

1935 and the turn#

The hardest test came in 1935, when Einstein, Boris Podolsky and Nathan Rosen published a thought experiment designed to show that quantum mechanics was incomplete: two particles, once they have interacted, can be so correlated that measuring one tells you the position or momentum of the other without touching it, so the other must have had those properties all along. Bohr's reply changed his vocabulary. Before 1935 he had spoken as if measurement disturbed the electron, which his critics could read as conceding that the electron had definite properties to disturb. After it he denied that atomic objects possess such attributes at all, apart from the whole experimental arrangement in which they show up. 'The whole situation in atomic physics', he wrote, 'deprives of all meaning such inherent attributes as the idealization of classical physics would ascribe to such objects.' A phenomenon, in his late usage, is not the object but the object together with the apparatus and the account of how the apparatus was set.

He was equally firm that none of this made the observer a magician. 'It is certainly not possible for the observer to influence the events which may appear under the conditions he has arranged', he insisted; the arrangement fixes what can be said, not what will be found. The legend that Bohr put consciousness into physics is one he spent his later essays trying to kill.

Why philosophers cannot leave it alone#

Here the disagreement begins, and it has several parties. Karl Popper and Mario Bunge read Bohr as a subjectivist or a positivist, a man who had given up on describing nature and settled for describing measurements; Popper's hostility to the Copenhagen school was lifelong, and the Stanford entry argues he had the wrong man. A second camp reads Bohr as a Kantian of a naturalised sort. The classical concepts are the conditions under which any experience of the physical world is possible, and the quantum of action marks the limit of their legitimate use, much as Kant's categories had conditions and limits; Jan Faye, the entry's author, calls Bohr a 'naturalized' or 'pragmatized' Kantian. A third reads him as a pragmatist for whom the meaning of a concept is fixed by its use in experimental practice. A fourth insists he was a realist about entities, since he never doubted that atoms were real things rather than useful fictions, and an instrumentalist only about theories, the formalism being a symbolic tool for prediction rather than a picture of what is there.

These are not four ways of saying the same thing. They disagree about whether complementarity is a claim about nature (the world really is such that no single description fits), about knowledge (we are so constituted that we cannot have one), or about language (the concepts we have are the ones that fail to combine). Bohr wrote in a manner that permits all three readings, and he was aware of it; his admirers call this depth and his critics call it fog. Don Howard's study in Philosophy of Science (2004), under the title 'Who invented the Copenhagen interpretation?', argues that the thing most textbooks call the Copenhagen interpretation was assembled by others, Heisenberg above all, after the fact, and that Bohr's own view was never quite it. The Institute that bears his name says, carefully, that the interpretation is still disputed and still probably the one most physicists hold.

What makes the quarrel more than scholastic is that the alternatives have not gone away either. David Bohm's hidden variables, Hugh Everett's many worlds, the various collapse and information-theoretic readings, each proposes to do without complementarity. Every one of them, though, has to answer the question Bohr put. Is there a fact about a quantum system independent of any arrangement for finding it out? Bohr said no, and said it in a way that was neither idealism nor instrumentalism, and the attempt to say what it was instead is a large part of the philosophy of physics since. It is also an old question in a new coat. Ludwig Boltzmann, a generation earlier, had fought the same fight over whether atoms were real or merely convenient, and the physicists who took Bohr's side in 1927 had been Boltzmann's heirs.

Bohr himself extended the idea well beyond physics, finding complementary pairs in biology (mechanism and purpose), in psychology (the content of a thought and the act of thinking it), and in the comparison of cultures. Few philosophers have followed him there, and the extensions are the part of his work that reads worst now. The core, however, is still the argument. His ashes lie in the family plot at Assistens Cemetery in Copenhagen, a short walk from Kierkegaard's grave, which is the kind of coincidence a Dane would call complementary and an Englishman would call a coincidence.

Also on the day#

Edgar Allan Poe died at Washington College Hospital in Baltimore on 7 October 1849, four days after being found delirious in a tavern. The year before, he had published Eureka, a 'prose poem' and, in his own description, an essay on the material and spiritual universe, which describes the cosmos as unfolding from a single primordial particle and which has been credited, with much qualification, with an early answer to Olbers' paradox: the night sky might be dark because light from the furthest stars has not yet had time to reach us. Poe himself offered the idea sceptically. Marilynne Robinson has called the book a startling anticipation of modern cosmology; John Horgan, who agrees that it is startling, doubts that it is sound. Either way it is a document of the same impulse as Bohr's: a writer deciding that the universe would not be understood until someone said what kind of thing a description of it was.

Desmond Mpilo Tutu was born on 7 October 1931 at Klerksdorp in the Transvaal. He won the Nobel Peace Prize in 1984 and, at Nelson Mandela's invitation, chaired South Africa's Truth and Reconciliation Commission. His theology rested on the Southern African idea of ubuntu, which he liked to gloss as 'I am human because I belong, I participate, I share', and the Commission was its institutional form: a wager that a nation could be reconstituted by telling the truth about itself rather than by punishing those who had lied. He died in Cape Town on 26 December 2021.

The series' previous entry, on Matteo Ricci and the problem of translating one tradition into another, raised a question Bohr would have recognised; and the month's gatherings are in the October conference diary.