Speaking of physics

‘Comps’ season has started in the Physics and Astronomy Department — that’s the nickname for the ‘senior integrative exercise’ required of all graduating seniors at Carleton. In our Department, it works like this: The students come up with a topic that they would like to explore and on which they would become the local experts, and spend the equivalent of one very intense course digging into the physics – it doesn’t have to include original research, but it could if they wanted. The students have complete freedom, with the only constraint that their project be ‘integrative’, that is, contain ideas from more than one of their courses, and be significantly deeper than introductory course-work. The topics have ranged from ‘The physics of catfish’ through ‘Deriving the BCS theory of superconductivity’. At the end of all that, they make a 70 minute oral presentation open to the community, and write a 20 page paper that goes through 2 revisions – the first with a single faculty reader, the second with 3.

I remember spending my first comps season at Carleton with my jaw hanging open. During most of the presentations the kids are in control of the material – they know what they know and what they don’t know, and do a very good job of convincing you that they basically understand even the stuff they don’t know cold, and we all learn something as a result. The best ones have this, and more: The speaker ‘owns’ the material, the talk is carefully thought out, pedagogically soundly structured, and entertaining to boot. My reaction during these outstanding talks – about a couple a year or so — is that if I didn’t know for a fact that this kid hadn’t graduated college yet, I would give him/her a faculty job if (s)he wanted it. One particular joy is watching students who have not necessarily shone in the coursework get the bit between their teeth and blow every one out of the water. It serves to remind me yet again that raw technical talent does not necessarily indicate a good research student, and a good research student isn’t necessarily good at creating and directing a research program … and a good talk may be completely unrelated to all of the above.

Having come to Carleton and having seen undergraduates capable of giving good talks has served to further my annoyance with/lower my tolerance for bad seminar and colloquia speakers (see Chad, Biocurious, and YoungFemaleScientist’s recent posts). I have sat through talks at multiple places, and in particular, I was at Rice when they were in serious hiring mode, and I had the privilege of seeing some great talks from the candidates as they went through. In general, I am going to bet that job-talks are some of the most engaging talks you’ve seen. It really is no mystery why job talks are so good: there is a lot at stake during the talk, and at least at Rice, the colloquium talk is supposed to show off both research and teaching skills.

The bottom line for the quality – and even the level — of a talk seems reasonably obvious + fairly simple ‘economics’ analysis of the speaker’s self-interest, coupled to his/her evaluation of the audience. For example, if the seminar speaker thinks that there are only a couple of experts in the audience who (s)he wants to impress, the talk is aimed at their level or higher, and the rest of the audience suffers terribly as a result. I don’t mean that everyone has to be entertaining and put on a great show – but in planning your talk, at least have some respect for the bulk of the audience, think about what they do and don’t know, and so on.

You can’t get very far in science without persuading people of the value of your work. Those who direct that persuasion only at a few people at a talk are focused on those with power over them – the power to affect career, publications, and grants. It’s a short-sighted perspective, reinforcing a hierarchical system where many people simply ignore junior colleagues/students and postdocs since they have no power over them. (I’m exaggerating somewhat for effect, I doubt that there are many people out there who are so short-sighted.)

On the other hand, when visitors from graduate programs come to Carleton, the power structure shifts remarkably – we are in a seller’s market with many sharp and well-trained undergraduates, and the speakers work hard to engage these young minds.

Most talks are given in situations with a mixture of those senior to you and those junior to you, particularly at conferences. There are usually many people in the audience who may never have direct power over you, but to whom you are trying to convey some physics. And it is crucial for the long-term health of physics, your area of specialty, and your research program to get everyone excited about what you do.

A mathematical case for diversity

This is going to be merely a pointer to a recent column/blog-post by one of my favorite science writers Philip Ball. Among other things, he says:

A company sets out to hire a 20-person team to solve a tricky problem, and has a thousand applicants to choose from. So they set them all a test related to the problem in question. Should they then pick the 20 people who do best? That sounds like a no-brainer, but there situations in which it would be better to hire 20 of the applicants at random.

… and then goes onto discuss some other interesting cases in agent-based situations, where many agents are competing for some fixed quantity – energy in a power grid, for example. As he points out, the results he is reviewing show that ‘diversity in decision-making may fundamentally alter the collective outcome. … [These results are] another example of how difference and diversity can improve the outcome of group decisions. Encouraging diversity is not then about being liberal or tolerant (although it tends to require both) but about being rational.

Undergraduate research

Carleton has a Learning and Teaching Center that facilitates conversations about teaching issues on campus (though as one of my now emeritus colleagues once said, “What on earth do we do on campus except learn and teach — why do we need a Center for that?” But let it pass, let it pass). This Tuesday we had a visitor, Sandra Laursen, who spoke about undergraduate research and its benefits. It was mostly preaching to the choir — almost all the science faculty at Carleton work with undergraduate students and will be happy to tell you about how wonderful it is to see students flourish during, and as a result of, their research experiences. Irrespective, it was good to see a reasonably controlled study (interviews with those undergraduates who did research — both immediately after their research experience, and a follow-up sometime later, compared with those who did not do research, and also interviews with faculty) that tried to flesh out our anecdotal ideas.

It turns out, not surprisingly, that research experiences help students clarify their ideas about whether they want to go to grad school or not, it brings them immense amounts of confidence about their scientific abilities, they feel like they learn to think and work ‘like a scientist’, they understand the process of science a lot better as a result, and they love owning a project, however small a slice of the big picture it might be. All of these experiences are significantly better — on average — at small liberal arts colleges rather than at research Universities (not surprisingly again) even if the actual project may be more exciting at the latter.

The punch line was about what the benefits and costs were to the faculty who supervised it. The personal benefits are great — it is always an immense pleasure to see raw talent transform into seasoned thinking and to send these kids off to adventures in graduate school and beyond. It is less clear what the professional benefits are, on average, and the costs can be quite high.

It’s something the Anacapa Society has hashed over a few times, and let me summarize the main points. What it comes down to is that by the time you’ve trained an undergraduate to do something, you’ve probably put in about 5 times the amount of effort you would have had to if you wanted to do it yourself — particularly true for theorists, I think. Now this might be true of a graduate student as well, of course, but with a grad student, you get a pay-off: They are around for a lot longer after their training period to deliver on the training. There is far less opportunity for that kind of pay-off from undergraduates. So working with undergraduates on research should be understood as a lot of teaching, and a little bit of research.

One of the ways I’ve dealt with this issue is by having multiple balls in the air at all time — some are those I can only conceivably work on with colleagues, and some slower-moving or smaller ones where undergraduates can participate. Sometimes the projects with colleagues will deliver an opportunity for undergraduates to be able to deliver, and we jump on it, but that’s rare. Another way I deal with it is that I choose projects (particularly those with undergraduates) that don’t focus on detailed technicalities. Instead, I take a broader perspective, attempting to uncover the deeper ideas underlying the technicalities. Overall, I have to say that I’ve done all right so far –  but I’m not going to be one of those prolific theorists that Doug was talking about a few months ago!

Ways of knowing

Last week’s ECC session was devoted to reports by the team-leaders of Carleton’s curricular redesign teams. Carleton, in the middle of the transition to a 5-course (over 3 terms) teaching load for faculty, and an ambitious capital campaign, is — sensibly enough in terms of timing, but insanely in terms of pressure on the faculty — also reconsidering the general graduation requirements for the first time in well over a generation. That is, the requirement for various majors are being left untouched, and most of the broad structure as well, but the ‘distribution’ requirements are being reformulated (or not, if the faculty votes the changes down).

There are 3 teams, and each came back with a distinctive plan (the details of which I cannot discuss here, because they are not complete, they are confidential, and it would be totally unfair to the full faculty, which gets to see them in a few weeks). Despite the differences, there was a common theme to them all: (1) They took the recently formulated Carleton College Mission Statement seriously. (2) They moved away from the current content-based distro: ‘3 from the sciences, 3 from the social sciences, 2 from humanities, 2 from arts and literature + language + sundry other’ system to a ‘ways of knowing’ system (more below). (3) The overall impact in all cases was to slightly ease the requirements (though not necessarily, depending on the student’s choices).

The shift to ways of knowing is in keeping with the way the College (and its faculty) have been talking about education for a while now. That is, there is are far more similarities between a suitably mathematical Economics class and a Physics class than between that Econ class and an Anthropology class in terms of how students are challenged or how they learn — even though both the latter would count as Social Science. And if we claim to teach our students how to think rather than content (content-based education is looking pretty rusty in the information age/post-google world), then our requirements should reflect that. So on campus we’ve been talking about quantitative reasoning courses, criticism and interpretation courses, practice of aesthetics courses, empirical reasoning courses, etc.

Anyway, it was very interesting to listen to the current state of the redesigns, and I made myself a pain as usual, in this case by insisting, for example, that my Revolutions in Physics class seemed to fit in the category of a course on the ‘Foundations of Western Civilization’ because I didn’t really see how anyone could claim that modern Western culture did not rest on physics, specifically via technology. Most of my heckling was done — and received by the team-leaders — in the spirit of forcing people to define their terms more carefully. (It’s one of my favorite things about being at Carleton, actually, it’s small enough and collegial enough that this sort of argument can be conducted — usually — in a pretty spirited but friendly fashion). Similar arguments were thrown at them from my colleagues from across the campus, and the leaders promised to take our comments under advisement before the big event: Presentation to the full faculty.

That’s the part I am looking forward to. The faculty will not be allowed to comment during the first presentation to give the teams a fair chance of actually being able to present their plans. As Bill, the co-chair of the ECC, said: We’re going to need duct tape.

DJ for a day:music stories

A couple of months ago, the Carleton Alumni magazine did an article about the recently-tenured faculty, with the organizing theme being their question: ‘What inspires you?’ Mine was pretty easy — music. I listen to it incessantly, and it really does inspire me to think/focus/be. A student who has a show on the Carleton student-run radio station picked up on that and asked me to come in for an hour for her ‘DJ for a day’ show, where she invites/cajoles people into sharing their music.

Well, okay, then. The biggest problem as far as I was concerned was picking an hour’s worth of music out of my library. Being a good statistical mechanic, I decided to go with randomness — I allowed the party shuffle to play, and created a playlist by picking whatever showed up (within limits of size — some of my favorite Hindustani classical pieces can run 45 minutes without a break of any sort, for example. Ok, to be completely honest, I also used the criterion of familiarity — there’re a few songs on my ipod I actually know diddly-squat about). I walked in with the playlist which was quite a bit longer than the hour — and again used the randomizer to choose what it wanted and I told whatever stories I associated with the music. It was fun, and now my mumbled stories and chosen songs are permanently archived — the first few tunes are someone else’s and then there’s an hour of my music.

There is some intellectual/physics content to this post, if you’ve borne with me so far. When I taught my first-year seminar on complexity a few years ago, one of the books we read was Per Bak’s how nature works: The Science of Self-Organized Criticality. One point he makes in there has stayed with me, even as I acquire more evidence for this from my readings into psychology, and even as I clarify my own thoughts on the difference between material determinism (+ randomness) and psychological freedom.

The point is roughly as follows: Humanists ascribe a lot of meaning to contingent phenomena, and spend their intellectual energy creating (or discovering) stories to explain the specific sequence of events that have occurred. A good statistical mechanic realizes, however, that it is almost impossible to explain complicated nonlinear phenomena given the incredible amounts of randomness and deterministic unpredictability involved, and instead recommend focusing on the probabilities of various events happening.

Human beings live in this tension: We are a lot happier when we understand why something happened, and from the ‘inside’, but honestly, the only hope is to understand it statistically, that is, from the outside.

Anyway, the ipod randomizer did a pretty good job of choosing songs for my playlist, and even finished with an amazing segue: The last song it chose from my playlist was Stevie Ray Vaughan covering Jimi Hendrix’s ‘Voodo Chile (Slight Return)’. The next person in the booth, Ross Currier, a Northfield resident, had burned his CD with his playlist, and the first song on that was Jimi himself doing ‘Red House’. Given that I’d never met Ross before, and had no idea that he liked similar music, I was pretty impressed. Randomness rocks!

Volunteer recognition, finally!

Gene Sprouse wrote me yesterday (as well as to many many many other physicists):

The Editors thank you for your help during the past year as a referee for the journals
of the American Physical Society. Your thoughtful and well-informed reviews are crucial
for our decisions, and your comments help authors improve their manuscripts. We sincerely
appreciate your assistance and look forward to your ongoing contribution to the physics
community.

To express our appreciation for the essential work that anonymous peer reviewers do for
our journals, we are starting an annual program to recognize approximately 130 of our
42,000 active referees each year as “Outstanding Referees”. Each awardee will receive a
certificate and pin, and we hope to thank each in person at one of our annual APS meetings.
The list of Outstanding Referees will be published online and in the journals although each
can choose not to be recognized publicly. We will select the referees based on many factors
that may include diligently returning reports in a timely manner, reviewing many papers
over many years, or providing especially insightful advice. The award will be one-time and
the referee must be living. Neither nationality, APS membership, nor field of scientific
expertise will be a factor. We hope that all referees will be pleased, whether or not they
are chosen, that we are recognizing some who have done an especially outstanding service to
the community.

Well, finally! Not because I’m expecting to be one of the Outstanding Referees, but some sort of recognition for this is a very good idea. It’s a little strange that the nominal ‘make or break’ activity for reward in academic physics — publications — is handled by unpaid anonymous refereeing. I don’t mean to say that the referee’s name should be published along with the authors (although in certain moments I have wondered if that might change things for the better considerably; transparency and acceptance of responsibility is almost always a good thing. But it would make it hard within the current power system for junior referees to do a good job of reviewing senior authors’ papers).

But some sort of lifting of the veil might help. Here’s why:

Physics has a relatively high publication rate (compared to Math, say), which I’ve heard attributed to a tendency to publish a lot more papers than are actually useful (or read, perhaps). There is a sense in which we believe, I think, that the good stuff out of all that will survive/rise to the fore, and so some sloppy papers getting through, or some inappropriate rejections don’t mean a lot. Great attitude in general, and probably responsible for the rise of arXiv (the free online unrefereed repository so many of us use).

I try my hardest to do a conscientious job when called to referee: Respond promptly, be honest about reading the paper carefully and understanding it, be clear and respectful about what I don’t understand, or about things that look wrong to me, and to be as constructive as possible in the comments/reports. In general, I try to act like a colleague down the hall from the authors — and I love it when I get referees who treat me the same way.

But the acceptance of slop in the system combined with the anonymity means that no one beats up on the referee if a particular paper that (s)he let through turns out to be flawed or incomplete and there is no payoff for being careful, compassionate, intelligent in your refereeing (except for serving the cause of physics itself, don’t get me wrong). What’s the external incentive then for doing a good job?

There was the usual reception hosted by the editors at Physical Review the last time I was at a DAMOP conference and I ducked into the tail end of it because I was feeling hungry and remembered that they always had good nibbles at the Phys Rev reception. I landed up chatting briefly with one of the Editors and in passing suggested to her that some sort of name recognition might be very valuable. It would give us an incentive — however minor it might seem, academics thrive on minor incentives — to do a good job and it would give us some sort of responsibility for our product. All good things. The American Journal of Physics publishes the name of all their referees every year, and I think it’s worked really well for them.

I very seriously doubt that mine was the only such feedback they’ve gotten over the years, or that this was what was led to this new initiative, but about time!

Good classes

Students in almost all my physics classes read the textbook, send in questions which I use to guide me in constructing my lecture, and then we work on problems together in some of the classroom sessions. I asked the students in the quantum class how it was going yesterday, and one of them said: ‘Well, I don’t really get it at all when I read the book, and the lecture — I get it a little bit more, and it’s only when we do the problems, I sort of start feeling good about what we learned. ‘ And I couldn’t help getting a big smile on my face: ‘Yes! That’s not a bug, it’s a feature.’

The ‘Revolutions’ class had a visit from Bob Russell, Director of the Center for Theology and the Natural Sciences in Berkeley, as promised. I enjoyed myself thoroughly — I got to provide a provocative intellectual experience for the students, and the speaker got some wonderful interaction with the students — and I didn’t have to do a thing except introduce them to each other. He talked about how the ‘equation’ of science with atheism was not valid, and that it was possible to be either an atheist and a scientist, or religious and a scientist, with the argument presented mostly from the perspective of philosophical underpinnings of inquiry — that is, epistemology.

What I took away from it was that if you were, like him, deeply religious and trained as a physicist, you could manage to keep the two ideas together in your mind without seriously compromising either. I would also add: for the most part. Not all of the arguments convinced me, or felt substantial enough to hang something this big upon.

He tried to lay this out mostly in the context of cosmology and the issue of a t=0 moment (at the origin of the Big Bang) and how that was handled by various people from their atheistic or religious perspective. The students were wonderful — they are mainly humanists, and they drew from their training and their basic intelligence to keep Bob on his toes, forcing him to reach deep to clarify and re-frame his points, etc so that his arguments steadily became more precise. A classic liberal arts moment.

And physics, too

Some of my research kicked back to life.

The first is my stymied project on parameter scaling in the difference between classical and quantum mechanics. A friend had asked why I didn’t ask a ‘competitor’ to comment, and thinking about it, I inverted the way in which I have recently gotten into collaborations. That is, recently, people have come to me with an idea to see what I thought of it, and my questions and comments have proved valuable enough that I got involved with the research. This time, I wrote to Nathan, a bright young grad student who I had run into at a conference and who has published some results related to mine, and asked if he would care to look at my paper and see if he could figure out a way past the road-block of neat results, but no clear physical intuition. He has some spare time at the moment, and no constraints on him to not start new collaborations, so he gave it a whirl.

He had a great question almost immediately. The focus of this paper is to show that a particular distance function to measure quantum-classical difference works marvelously in demonstrating parameter scaling. But there are factors of hbar (ok: hbar is Planck’s constant, and critical, crucial, in setting the scale for quantum mechanics) that make no sense, and if you can’t explain your results, how on earth can you trust it, let alone publish it? He thought that the issue was that we were looking at a quantum-classical distance function that was not appropriately dimensionless. Which immediately led us into the issue that everything in phase-space has weird dimensions (I know I’ve lost all the non-physicists but pretending you are reading along, phase-space is where you look at both the position and the momentum of a particle. Since you are in this hybrid space with distances in different directions having different meaning, it is intrinsically weird).

Anyway, it was good to be able to talk about a physics puzzle that was bothering me, and now Nathan’s going to take a longer and more detailed look at this issue of dimensions. As always, it is discussion that gets me going when I am stymied, and nowadays even more so with a lack of time to sit down and focus. I’m happy to have a new collaborator.

I also got back in touch with Arik — we’ve both been distracted, and unable to proceed on the response to our paper from Physical Review Letters. But there was a useful conversation this afternoon: We are still going back and forth on whether we have enough data already to answer the referee. Today’s discussion landed up on the issue of whether chaos induced by measurement counted as chaos due to quantum effects. I say, sure, measurement’s quantal, darn it. But I can see the referee’s point somewhat.

And another call, not really physics, but sort of — an alum, now in a newly cooked-up interdisciplinary bio-and-physics-and-chemistry program, wanted to talk about issues of feeling mathematically rusty, feeling not quite trained enough mathematically for what he wanted to do, etc. I tried to remind him that the problem always outranked the solution, and that elegant mathematical solutions were in general not available in biology — it is far too nonlinear, complex, and downright messy (from the theoretical physicist’s perspective) so feeling like he didn’t have enough math to do theoretical biology was, to my mind, a question of attitude rather than fact. Particularly for him — this kid took a ton of math here at Carleton. We’ll find out.

Money, Science and Atheism, the FCC

Separately those are the things on my mind.

Money: Today the CISMI advisory board had the pleasure of sitting down together to figure out the process for spending some money. This is for student research/internships and comes to us thanks to the generosity of a Carleton Physics alum, David Ignat. He has endowed a fund that students can use on or off-campus during summers and winter breaks. The formal announcement of procedures etc should be going out soon, but it’s always nice to be able to make it easier for those interested in science to find out if they want to pursue it further. And what better way to find out, beyond classes, than to be plunged into the ‘doing’ of science. I so wish I could’ve spent my summers in college doing science instead of futzing around in the doldrums of an Orissa summer. Ah the pleasures of being at a liberal arts college with a decent endowment.

Religion: Bob Russell’s going to talk to my class tomorrow, and here’s the abstract of the paper that I sent around to the students for their edification, which is what he’s going to talk about.

This paper explores the thesis that science, philosophy and theology can, do, and should interact in positive, constructive and fruitful ways. The paper consists of two parts: I first offer an historical case study of scientific cosmology in the 20^th century. We will see explicitly that philosophy and theology have played a constructive role within actual scientific debates, thus demonstrating that one cannot neatly cleave apart science from philosophy and theology. In the second part I explore current constructive interactions between science, philosophy and Christian theology regarding such issues as the creation of the universe at “t=0″ and the “fine-tuning” of the universe by God. I will suggest that both of these issues are important in the interaction but ultimately of only transitory value, given the changing character of scientific cosmology and Christian theology. I then close by looking briefly at a new set of questions surrounding cosmology and theology. I want to acknowledge at the outset, however, that many intellectuals take for granted that science unilaterally supports atheism and discredits religion, or at least theism. Accordingly, before developing the constructive arguments I will list several (well known) reasons why the ‘equation’ between science and atheism is not necessarily valid. Without first attending to these claims even if briefly, the constructive portion of this paper would be, or could seem to be, seriously undercut.

In the evening, a nice dinner, and then his ‘Barbour lecture’ on Barbour’s contribution to the dialog between science and religion.

Given how much I like to get the students thinking about these ‘big-idea’ issues, I’m glad to have this visitor.

The FCC: A student asked me if I’d like to DJ a music show on the Carleton radio station this weekend, and I agreed. So she sent me back the FCC guidelines, including the 7 swear words that I can’t use on the air. That email made me burst out into laughter, given those words right in the middle of a very casual email, and with an ‘excuse, please’ appended. Damn, now I feel really constrained :-). But I do have to figure out what to play: so many songs, so little time, how do I choose only an hour’s worth? I need an organizing theme of some sort …

 

Broadening access to science

(Warning: I am speaking off the top of my head. I expect that some of what I say will rub people the wrong way, and I hope to learn from responses to this post, either on the blog or directly by email or by phone call. To find my contact information for the latter, just google me).

There’s a group of us on campus — faculty and staff — who have informally and formally met and struggled for a few years with the issue of diversity in the sciences (see the post below on Freeman Hrabowski’s visit to Carleton and associated events at the end of January). I’m more formally involved with ’steering’ this now as part of my CISMI responsibilities, so expect occasional thoughts about such issues on this blog.

Today: What motivates this group?

Not surprisingly, different things. From my end, the challenge is why Carleton’s student and faculty population in the sciences looks different from the general Carleton population, and why this looks different from the population of the United States, considering that Carleton draws its students from a national (and increasingly international) pool. Physics, for instance, is one of the extreme cases nationally, even when it comes to gender equity. Carleton has done remarkably well over the years in that roughly 25% of its graduating seniors in Physics and Astronomy are female, but it’s been stuck at that number for years, for example. And as for black or Latino/a students, the numbers are tiny. Computer Science, too, has terrible numbers in this regard. But not Mathematics. So what explains this difference?

I am from India, and have watched the international view of Indians transform in a blink of a generation from them being regarded as poor, technologically incapable, and associated with mysticism and exoticism, to being regarded as scientifically and technically enormously capable and as a scientific-technological-economic power, threatening to the West (the real issue with ‘outsourcing’ isn’t that pay-scales are lower in India, but the immense size of the scientific and technological human talent pool there). In a case of things turning full circle, various Americans now spend a summer or year interning in India in the Silicon Valley there (I remember the strange feeling of talking to an alum Michael R. as he graduated and headed off to Bangalore — it was the best job he could imagine having). And Japan (Japan!) has a case of ‘Indian educational system’ envy.

I claim that there are two lessons in this: One, the cultivation of talent in one population helps all of humanity. Two, any population is capable of doing science, if that is valued and cultivated appropriately. Yep, this isn’t a massive intellectual point, just my starting point.

Why cannot these lessons be applied to sub-populations in the United States?

That’s my motivation.

—————-
Now playing: Barry Louis Polisar - All I Want Is You (Juno soundtrack)
via FoxyTunes