Scientist Hofstadter dreams of butterflies and other odd things

Scientist Douglas Hofstadter creates a spectrum of electron energy levels and then goes on to ponder artificial intelligence and the nature of self

PUBLISHED : Sunday, 04 August, 2013, 12:00am
UPDATED : Sunday, 04 August, 2013, 3:22am

This spring, there was excitement in the world of physics as a long-predicted butterfly was proved to exist. But, being a creature of physics, this butterfly wasn't an insect, nor anything that would even occur without human minds to construct it.

This was Hofstadter's butterfly, a remarkable spectrum of electron energy levels. It was first described in 1976 by Douglas Hofstadter, who was then in the physics department at the University of Oregon. He was looking at the allowed energy levels of electrons restricted to a two- dimensional plane, with a periodic potential energy and a changing magnetic field. As Hofstadter put it in a summary of his work: "The resultant Schrödinger equation becomes a finite-difference equation whose eigenvalues can be computed by a matrix method."

To which you might respond, "Aha, but of course it does!"

Or even, "Huh?" — in which case, you might simply appreciate that when he plotted a graph of the spectrum, Hofstadter made a remarkable pattern that looked somewhat like a butterfly. And this pattern was recursive, so if you look at a small part of the pattern you see the same butterfly shape, which is repeated at larger and larger scales. The paper was published just one year after the term "fractal" was coined, and Hofstadter had discovered one of the very few fractals known in physics.

Physicists have since searched for proof of the butterfly, yet until recently it proved elusive. This is largely as it results from quantum effects, and when atoms in the two-dimensional plane are close together, observing the butterfly would need unfeasibly strong magnetic fields, while, if they are widely spaced, disorder ruins the pattern.

Graphene, a quirky form of carbon, has been the key to finding the butterfly. It is a one-atom thick layer of carbon atoms arranged in hexagonal patterns - somewhat like chicken wire. A layer of this was placed on atomically flat boron nitride substrate, which likewise has a honeycomb atomic lattice structure, but with slightly longer bonds between atoms. This combination resulted in the electrons experiencing a periodic potential, akin to a marble rolling over a surface shaped like the tray of an egg carton.

Cory Dean, assistant professor of physics at City College of New York, developed the material. He was a member of an international group that published its findings in May.

Separate groups at the University of Manchester in Britain and the Massachusetts Institute of Technology simultaneously reported similar results.

According to a City College press release, the light and dark sections of the butterfly pattern correspond to "gaps" in energy levels that electrons cannot cross and dark areas where they can move freely. While efficient conductors like copper have no gaps, and there are very large gaps in insulators, Dean believes the very complicated structure of the Hofstadter spectrum suggests as yet unknown electrical properties.

"We are now standing at the edge of an entirely new frontier in terms of exploring properties of a system that have never before been realised," he said. "The ability to generate this effect could possibly be exploited to design new electronic and opto- electronic devices."

Graphene planes have already shown promise as a new wonder material. They were first isolated in 2004, and have a thickness almost a millionth of a human hair. Graphene is stronger than steel and more conductive than copper, and can help make ultrafast optical switches for applications including communications. It may even prove to be the ideal material for 3D printing.

Rather as graphene may have multiple uses, the man who described the butterfly spectrum has proven multitalented. Douglas was the son of Stanford University physicist Robert Hofstadter, who in 1961 was the joint winner of the Nobel Prize for Physics "for his pioneering studies of electron scattering in atomic nuclei and for his consequent discoveries concerning the structure of nucleons."

Like father, like son, you might think, as Douglas also became a physicist. Yet he did not remain so for long. The year after his paper on the spectrum was published, Hofstadter joined Indiana University's computer science department, and launched a research programme in computer modelling of mental processes, which he then called "artificial intelligence research", though he now prefers "cognitive science research".

Hofstadter pondered the question of what is a self, and how can one come out of stuff that is as selfless as a stone or a puddle? In an attempt to provide an answer, he wrote a book, Gödel, Escher, Bach: An Eternal Golden Braid. This interwove several narratives, and featured word play, puzzles as well as recursion and self-reference, with objects and ideas referring to themselves. The book was a success, winning the Pulitzer Prize for general non-fiction. Yet in an interview with Wired, Hofstadter later expressed disappointment that most people found its point was simply to have fun, albeit noting that hundreds of people had written to him, saying it launched them on a path of studying computer science or cognitive science or philosophy.

Some of these people might have been startled when Hofstadter, by then professor of cognitive science at Indiana University, later told the New York Times: "I have no interest in computers," and adding: "People who claim that computer programmes can understand short stories, or compose great pieces of music - I find that stuff ridiculously overblown."

The New York Times interview accompanied the publication of a straighter book on questions of consciousness and soul, I Am a Strange Loop. Within this, Hofstadter wrote: "In the end, we are self-perceiving, self-inventing, locked-in mirages that are little miracles of self-reference."

Here, Hofstadter seems to echo the butterfly pattern he discovered, with its multitude of versions of itself.

Over 2,300 years ago, a butterfly featured in an anecdote by another thinker. The Chinese philosopher Zhuangzi wrote of dreaming he was a butterfly, and awaking to wonder if he was a man who dreamt of being a butterfly, or a butterfly dreaming of being a man. After Hofstadter, you might wonder if either of these is but a dream within a dream. Which may remind you of a movie, which you may find within another column, which is currently a hazy looping form within the mirage that's this writer … but will not feature butterflies. Martin Williams is a Hong Kong-based writer specialising in conservation and the environment, with a PhD in physical chemistry from Cambridge University.