Sep 062014

The Twenty-third Congress of the International Union of Crystallography met in downtown Montreal, one or two kilometers south of McGill University, whose medical school peeks over the stadium towards downtown:

Downtown Montreal sits right on the St. Lawrence River, and amidst the tall buildings there squats the Palais des Congres…

…whose southern entrance opens to a small park…

…a few blocks from Rue Wellington (sorry, as an Anglophile, I had to mention that).

Not being good at estimates, I don’t know how many people attended, but there were enough people to fill a very large auditorium and keep eight parallel sessions going through the last (seventh) day – the IUCr estimates over 2,200.

Like many attendees, I cherry-picked events, but I did have a revelation. Suppose you go to a conference on a sprawling, multidisciplinary field, where the plenary talks focus on The Latest Hot Topic (more about that below, and in the next post), and where there are approximately a zillion parallel microsymposia on various subfields. Where would you go to get a general impression of what is going on?

To the poster sessions, of course.

Posters and exhibits were relegated to a large multipurpose room tucked beneath the multi-story complex that makes up the conference space. To someone used to mathematics conferences, where the exhibits are dominated by publishers hawking monographs and textbooks, the IUCr exhibits seemed to consist largely of software and hardware. Almost like a (gasp) trade show. There was, so I heard (but I did not see it on exhibition), one book on display (the Little Dictionary of Crystallography, which quickly sold out). And surrounding the exhibits were hundreds of posters.

Poster presenters are an odd lot. My father preferred presenting posters over giving talks: he wasn’t convinced that people actually listened to the talks (!). And with a poster you could have a one-on-one conversation with passersby – which, as educational psychologists will tell you, is a more reliable way to impart information that talking at lots of people from a distance. Another thing about posters is that screening committees are less picky: you can take risks and talk about odd topics without getting hurled out the door. The result is that some posters talked about funky or fun stuff that would never have made it into a microsymposium. And considering how many graduate students present posters at a meeting this large, you can get an impression of what is going on at major research centers simply by walking along the aisles of posters.

Returning to the talks, especially the plenary and keynote addresses, Quasicrystals were what’s hot. Dan Schechtman spoke about Quasicrystals – A Paradigm Shift in Crystallography. “Paradigm shift” refers to Thomas Kuhn’s model of scientific progress. Normal science is conducted under a reigning paradigm, and every once and a while, a paradigm’s inadequacies become manifest, and there is a shift to a new and more adequate paradigm. Paradigm shifts can be unpleasant to live through: think of Ignaz Semmelweis, who proposed washing hands before surgery to remove contagions, and who was dismissed from his post, harassed by his colleagues, and committed to a mental asylum, where he was beaten to death by the guards. Schechtman’s opponents were more civilized and he weathered the storm (which appeared to have lasted about a decade) and ultimately won a Nobel prize.

Schechtman mentioned four things a scientist needs to effect a successful paradigm shift.

  • One has to be very good at transmission electron microscopy (“TEM”). Schechtman focused on one particular skill that made his discovery possible. It takes many years to become any good at TEM, and few succeed. This may be generally true: sociologists and psychologists are increasingly remarking on the need for a great skill to accomplish a great deed: a great idea, in of itself, is often not enough.
  • One needs tenacity. Schechtman told the story of a student who saw a quasiperiodic diffraction pattern before he did, recognized its significance, but (possibly reflecting on Semmelweis’ fate) said nothing. Schechtman said that an odd observation may be an artifact (it probably is), but you will never know if you don’t pursue it. Schechtman also mentioned resilience, which is not quite the same thing. (On the other hand, there is the tenacious and resilient example of Louis Agassiz – the fellow who discovered ice ages and later showed great tenacity and resilience as the last great holdout against evolution.)
  • Believe in yourself. Schechtman actually had an arch-enemy, Linus Pauling, who was convinced that quasicrystals were classical crystals with very large units. But Schechtman persevered. There are two ways of looking at this. One is that this is the basic advice for aspiring writers: you know you aren’t a real writer until you can paper your wall with rejection slips (Lord of the Flies, which won Golding a Nobel, got 20 rejections, as did Frank Herbert’s Dune, while Harry Potter and the Philosopher’s Stone garnered only sixteen – and A Wrinkle in Time got 26 rejections while Gone with the Wind got 38).

There are two things that Schechtman did not mention: recognition and communication.

  • When I was a graduate student at UCLA, Paul Erdös (one of the great eccentrics in mathematical history) visited us, and told us the following story (which I have not checked to see if it is true). Before Marie Curie’s work on radioactivity, workers in a lab observed that if you left pitchblende and undeveloped film together, the film would get fogged. In reaction, the lab workers put a rule in their books: pitchblende and undeveloped film should not be stored together. Henri Becquerel later observed the same phenomenon and, paying more attention, told his co-workers (the Curies) about it. From Alexander Fleming’s chance observation of penicillium’s chemical warfare against germs to Charles Goodyear dropping rubber on a stove, some of the greatest discoveries were made by people who were paying attention.
  • In the mid-Nineteenth century, the physicists knew that they needed an algebra of three-dimensional space. In fact, William Hamilton knew that they needed an algebra of umpteen dimensional space (and Hamilton’s quaternions did not fit the bill). Hermann Grassmann developed such an algebra, but being of a philosophical bent, communicated his discoveries in books that Hamilton and other physicists were unable to decipher. Decades later, Willard Gibbs independently developed a similar algebra – which we now call vector algebra – and Gibbs was much better at marketing: he delivered his discoveries in bite-sized articles aimed at the audience he had.

And as far as reception goes, it may be wise to follow the advice from the Yes, Prime Minister episode, The Ministerial Broadcast. When a politician is to announce something routine, the opening music should be Stravinsky, the politician should wear a modern suit, and the background should feature abstract art. When a politician is to announce something truly revolutionary, the opening music should be Bach, the politician should wear a dark suit, and the setting should be oak paneling, leather volumes and Eighteenth century portraits.

Schechtman was merely the most notable of the speakers on aperiodicity. Marjorie Senechal gave a keynote address on Mathematical Crystallography in the 21st Century, and started by saying that major areas for future research include folding and flexing, diffraction and imaging, superspaces, symmetry, self-assembly and self-organization, and mapping the aperiodic landscape. She then focused on aperiodic structures from two points of view. From the point of view of the final structure, she said that under certain conditions, a Delaunay set must periodic, and asked if there was a (nice) set of conditions for a Delaunay set to have a discrete diffraction pattern. Then from the point of view of the formation of the structure, she observed that icosahedral shell structures of less than two thousand or so atoms are more stable than cuboctahedral structures, and asked how the transition from icosahedral to cuboctahedral takes place. She concluded by mentioning the Defense Advanced Research Projects Agency (DARPA), which frequently posts “challenge” problems. In 2007, DARPA posted 23 challenge problems for mathematicians (which for some reason DARPA took down, but Professor Vasilios Alexiades of the University of Tennessee archived the list), and Challenge Number Eleven was:

  • Optimal Nanostructures. Develop new mathematics for constructing optimal globally symmetric structures by following simple local rules via the process of nanoscale self-assembly.

Senechal said that she asked someone at DARPA what all these terms meant, and was told that DARPA hadn’t assigned specific meanings to the terms. So the challenge is somewhat open-ended.

Officially, the big event was the Ewald Prize, which was awarded to Aloysio Janner and Ted Janssen for developing the “superspace model” of quasicrystals, and that leads into my next (curmudgeonly) posting…