The Quantum Pontiff

Slow Science

The “slow movement” is a vast beast: there’s Slow Food, Slow Travel, Slow Money, and even, I kid you not, Slow Reading. These movements all begin with the premise that modern culture emphasizes ever increasing speed and convenience (cue the Eagle’s: “Listen, baby. You can hear the engine ring. We’ve been up and down this highway; haven’t seen a goddam thing.”) The prescribed medicine is a moderance in life. More smelling of the roses (but watch out for Ringo), more taking the long road, and most definitely more chewing your food slowly. While the movement suffers from large doses of overly nostalgic pastoralism, I find myself resonant with the slow movements search for a good pace and balance in how I try to live my life.

Thinking about this the other day (while chewing slowly, of course) I wondered, well, what about “Slow Science?” And like most thoughts you think might not have ever been thought, it turns out that this phrase has come up before: “Taking time to savour the rewards of slow science” Lisa Alleva, Nature 443, 271 (2006). To quote from the letter:

In shedding the ambition of my peers, I have discovered a secret: science, slow science, is perhaps the most rewarding and pleasurable pastime one could ever hope for. My supervisor’s lab is small — two postdocs only, with no teaching responsibilities. We are free to read the literature, formulate ideas and carefully plan our experiments so as to execute thoughtful strategies. We do not plough through genomes hoping to discover something interesting; we formulate a theory, and then we go in and test it.

Perhaps we are old-fashioned, but I feel my education as a scientist has benefited far more from my five years of slow science than the preceding five years of fast science. What’s more, we are on the brink of something big, exciting and wonderful, that spurs my slow science forever onwards.

So what about it? Who’s in for a slow science movement?

i-9b7c738fe82e831351878057c0c77ba9-Traffic_Slow_Chinese.png

Okay, okay, first let’s get one thing out of the way. Yes, if you look at my publication record, you might say, “Dave you’ve been performing slow science already!” And then, like all good anonymous commentors, you would follow that up with a smack in the face, “And look where that’s gotten you in your career! You just want slow science because you’re lazy!” My own situation is one reason I rarely post career advice on this blog (that and my strong belief that all advice is highly contextual and generalizations are what lead us into the stifling social norms which cause much of the problem in the first place.) But put that unfortunate background fact aside, and ignore the shouts of “lazy” from the slave driving biologists in the back of the room, and humor my daydreaming a little: what would the life and times of a “slow scientist” be like? What changes are needed to bring about a “slower science?” And would there really be benefits for this slowness?

How could one practice slow science? Well the first thing would be, of course, to “stop and smell the roses.” When is the last time that you didn’t have a deadline to meet and had time to actually reflect on your work, work at a pace not set by a deadline, and spend time deeply engaging in a subject, especially a subject not directly in you’re field of research? Slow science means, then, taking a day, a week, a month, a year, not just continuing on with the project you need to get finished because of the conference deadline coming up and your peers beating down the door to finish that journal article, but saying “it is more important to me that my work is not rushed, and is of high quality than it is to meet the deadlines which will get me more points on the great tenure meter in the sky.” Slow science means ignoring the rushing sound of papers flying off your colleagues desks and thinking: a leap is better than the random walk of a thousand small steps.

Does slow science mean working less? Well, I think, it could. But it could also mean changing your habits to avoid the detrimental rat race that exists in academia while not reducing how much you actually work. It’s not the hours that count but how your hours count. In a similar vein I would argue that the idea that one must be maniacally hard working to do good science or the idea that one must not be maniacally hard working to do good science are both two sides of a style of guiding our lives which is detrimental to good science. There are times when completely and totally absorbing yourself in your work is the only way you will make a step forward. There are times when you need to go outside and canoe around a lake to loosen up your neurons. These depend completely on circumstance but the most dangerous pill is the guilt that either is the right and only way to do good science. Slow science does not mean not working hard, it means working hard for the right reasons.

Slow science would, I think, also mean that you spend time savoring results. Yes, you could easily take that very cool result that your pal produced and tweak it a little such that you have another paper, but why take another bite when your next bite will be a sour patch in a field of sweet? Slow science would mean that you don’t kick yourself for missing the next step in a line of research. Savor the good papers, the good experiments, and, I think, you just might find that you understand more clearly what makes these papers and experiments so alive. This probably also means that a good part of slow science would be searching out the best of the best and sharing these results with all who are interested. I cannot tell you how many times my day has been made by a simple conversation about a new piece of science that a colleague or friend has shared with me.

Okay, so suppose that we want to move to a world with more slow science (big assumption!) How could we get there? One can, of course, approach how to change this from a variety of different viewpoints: scientists at major research universities, scientists at small liberal arts universities, scientists in industry or scientists living in their van on the beach in maui. I’ll take the perspective that I most closely know, scientists at major research universities, but I think some of what I say may carry across the divide. A (direct, logical) approach to thinking about this is to ask what is so hectic about the life of a modern scientist today? If you ask that to most professors, I’d wager the first response you get would be “grants,” the second response would be “teaching,” the third would be “publications”, the fourth response would be “committees,” and the fourth would be “managing a group.” Everyone, I’m sure, can come up with ways in which the above could be modified in order to promote slow science. I’ll just venture into one relm, grants.

One thing I’ve never understood about the grant making process is, at least from my perspective, the inflexibility of the system to deal with the timescales involved in doing good science. For example, most of the programs I apply to at the NSF work on a three year time period. Now there is definitely something to be said for deadlines, but is it really true that every single research project should be set on a three year timeline? I certainly think that doing really groundbreaking work usually takes much longer, in part because a requirement for such work is often exploring more dead ends that more routine science. It would go a long way towards promoting slow science to have longer grants for the more speculative (but well motivated) research programs. Such a lengthening would, I’m sure, make these grants highly competitive, but at the same time this would give huge opportunities to promote projects that think beyond the three year time horizon.

Certainly there are other ways in which one could attempt to move to a world in which slow science is more the norm. But the big question lurking in the background is: why in the world would you want to do this? I mean, look, I went to Caltech. I know something about working long hours. The point here isn’t about the hours. The point is about how you approach your overall goal of doing science. The rat race causes a lot of harm in the actual progress of science. Indeed, when I go over my list of people I admire for doing great science, I would argue that a vast majority of these people are already practicing slow science of some form. I don’t want to name names, but I do ask you to consider your scientific heros and to think about how they approach their science. Certainly when I think about these people they often show a lack of concern with the metrics so heavily touted for achieving academic success.

All right so you made it to the end of this particularly badly motivated diatribe. Sometimes its just fun to throw out an idea and let the vultures eat. Soon, I’m guessing, this post will be ripped to shreds. Which is fine by me, as the ripping and the tearing and what with the feathers flying everywhere is exactly the kind of example I’d like to see of why slow science isn’t such a crazy idea.

Comments

  1. #1 Boris
    June 23, 2009

    Well written, Dave. I am not going to say anything insightful, I just like it. Very much.

  2. #2 uber snotling
    June 23, 2009

    Interesting Dave.

    From a consulting perspective, I can’t imaging having the luxury of a multi-year research project. But the questions I’m being asked are not of the same nature that yours are.

    Perhaps the funding model for basic research is wrong. Science as a paid profession is a relatively modern concept. Perhaps slow science is best suited to being the pasttime of men of leisure. Especially since the slow science work is almost philosophical rather than results-oriented.

  3. #3 matt
    June 23, 2009

    We don’t necessarily need multi-year research projects, but we do need a lot fewer “correct but boring” research papers. I’m sure all of us know the kind I’m talking about. The community can help by harshly reviewing such papers and counting such papers on the CV negatively when considering people for jobs. I blame the NSF’s emphasis on education for producing these papers. There is a big incentive to have grad students, as it helps the broader impacts on NSF grants, which leads to a desire to find make-work problems which lead to these uninteresting papers.

  4. #4 Janne
    June 23, 2009

    I don’t think it’s the grant process that’s the main problem. With today’s culture you could be entirely self-financed and still running yourself ragged.

    The problem is tight deadlines and multiple tasks in general. As you say, there just isn’t time to just stop your current work for two or three months to go on a detour learning something else (that may be very beneficial to your current work in the end). Even with no grant deadline, you have coauthors and other people waiting on you for every single thing you do. And there’s no large free blocks of time around where you can just spend a month or so on one single thing without lots of other incidental tasks breaking it up. Frankly, as long as research is a group effort you’re just not going to be able to change this.

    What we could do, instead, is to make sure people have blocks of “single time” in between “group time” periods. So, at the end of a year working on your cooperative project, you get an enforced “mini-sabbatical” of two months or so where you’re expected to go off on a tangent by yourself and be forbidden to contact your collaborators about the main project. No teaching, no advising, no nothing. But no holiday either – if anything this may be the one period where the university should check for attendance to help you not slack off.

    At my former place, a couple of us did a small silly side project: a humanoid hitting a baseball. It was meant to be just a fun demo – a neat video to show visitors – with little to none original research. To my surprise, though, that side project is by now generating as many publications as the main project I worked on did. You never know what turns out to be useful.

  5. #6 Janne
    June 23, 2009

    BTW: Perhaps using the “MAN” character as a symbol isn’t the best choice as the most common word in which it occurs in Japanese is “gaman” – endurance – and it frequently occurs as a compound for words like “haughtiness” and “pride”. Not the associations I’d be looking for.

  6. #7 Jonathan Vos Post
    June 24, 2009

    Not completely silly, at all.

    “Was Feynman a Kantian Physicist?
    Einstein was!
    http://ysfine.com/feynman/feykant.html

    “According to Feynman, the adventure of our science of physics is a perpetual attempt to recognize that the different aspects of nature are really different aspects of the same thing.”

    “Why am I so crazy about what he said above? The reason is quite simple. I am a Kantianist, and Feynman talks like a Kantianist. I was able to talk to Wigner because he was a Kantianist. The most prominent Kantianist in physics was of course Albert Einstein.”

    “According to what he said above, Feynman is saying or at least trying to say the same thing in his numerous papers. Thus, his ultimate goal was to combine all those into one paper. Feynman published about 150 papers. I am not able to combine all those into one paper….”

  7. #8 Rocky Humbert
    June 24, 2009

    Definition of a paradox: “Slow science advocated by a researcher in quantum computing.”

    Moralistic paradigm: “Since we have a limited amount of time on this earth, the slower-we-move, the less we can accomplish.”

    Lofty platitude: “When traveling from NY to LA, a Boeing 777 is preferable to a DC-7, even though the seats are more comfortable on the DC-7.”

  8. #9 Tobias J. Osborne
    June 24, 2009

    I suspect some of my acquaintances in “safe” positions are in the habit of, paraphrasing what you said, kicking themselves for missing the next step in a line of research. So many times I’ve heard people sighing/cursing when a new paper hits the arxiv: “oh no! I could’ve/should’ve done that!”

    Of course, I also feel the pressure to be the one to be quickest to point out the next logical step in a line of research. I’ve spoken to various people about this, trying to find out where the pressure comes from and what surprises me is that most people I spoke to don’t actually feel this pressure coming from grants or students or whatever; they weren’t actually able to tell me why they feel this pressure. Maybe it’s a competitive thing…? It’s weird because the incentive structure eg. here in the UK, rewards good papers rather than many papers…

    Do you feel a pressure to write a certain number of papers? Eg. “it’s the end of the year and I’ve only written 4 papers: oh no!” Where on earth does that pressure *really* come from? I don’t know, perhaps it’s a hangover from one’s early postdoc days…? Surely it’s better to write one amazing paper every 3 years than 12 crappy papers!? Certainly my scientific heroes (mostly) work more like this.

    Speaking personally, overcoming this pressure to produce papers and be in the limelight pointing out all the next steps in research is the greatest impediment to slow science (which I think is a *great thing*, BTW).

    But I also think “slow science” shouldn’t mean “working slowly”. I feel that to make progress one should “make as many mistakes as fast as possible”! (Is that quote due to Gell-Mann? I forget…)

  9. #10 Matt Leifer
    June 24, 2009

    “I feel that to make progress one should “make as many mistakes as fast as possible”! (Is that quote due to Gell-Mann? I forget…)”

    No, it’s an unmistakable John Wheelerism.

  10. #11 krusty the kobold
    June 24, 2009

    on #3 and #9:

    If one removed “correct but boring” papers, how many journals would there be? How many pages in the surviving journals?

    If you pick your favorite paper and remove all the “correct but boring” articles that cite it, how many times would it be cited?

    If you look at your own CV and remove all the “correct but boring” papers, how many are left?

    Whenever I go to a conference, I feel that Europe seems more capable of slow science than the US. It may be the grant structure and ultimately the support of graduate students. Often faculty positions in Europe include a small number of research assistants. This is never the case in the US. All graduate students need to be supported by grants or fellowships or end up teaching.

    In the US, success is equivalent to grants. Grants require preliminary results which are often the “correct but boring” papers that ideally lead to something better.
    After the grant is received often the “great idea” is not achievable on a three year time span. “Correct but boring” papers describing your progress are essential for both grant-renewal and future grants on other topics from the same agency.

    The current system does reward “great ideas” but also generates a glut of “correct but boring”. I sometimes daydream about ways to fix this. I am often drawn to a 1 paper/person/year model. You can be a co-author on as many papers as you like, but someone has to use their 1 paper token. In the real world this would probably just lead to administrative assistants as co-authors, but in my daydream it leads to an end of “correct but boring”.

  11. #12 汉字 idiot
    June 24, 2009

    #6: There is a little country called China, where people also use 汉字 (in fact that’s where they get their name from, or so I’m told).

  12. #13 Janne
    June 24, 2009

    #12: Yes, there are. Which is why it’d be important to select a character with the right connotations for both languages (and perhaps even Korean should be taken into account, though kanji are no longer used widely there).

    Of course, an even better idea would be to create an actual icon instead of pressing a writing character into double duty. “white coat-and-night cap”? A beaker with “Zzz” coming out of it?

  13. #14 John Preskill
    June 24, 2009

    Kitaev ….

    I try to do science as slowly as he does, but didn’t Clint Eastwood say that “a man’s got to know his limitations”?

  14. #15 mick
    June 24, 2009

    Nice article Dave. Anyone who has looked at my CV will see that I’ve long been an advocate for “slow science” :-) There should be more of it and I’d like to see more people in semi-safe career positions slow down and go after more big questions.

    That isn’t to say that I haven’t been afforded that opportunity by my employers. I’ve been very grateful for the fact that my superiors (who, I suspect, would hate that I used that term) have given me lots of opportunity to pursue my flaky ideas and to learn new areas of science. I mostly wish that others could have (or take) similar opportunities.

    @9 The EU definitely rewards researchers based on the sheer volume of work produced. The folk in Brussels like numbers that “prove” your worth and a large number of publications always goes down well. However, it’s true that in the UK, as Tobias points out, a number of local research agencies will show some leniency to those who are clearly publishing high quality work.

  15. #16 Dana
    June 24, 2009

    I started thinking about “slow science” a couple years ago, but my thoughts led me a different direction. I started wondering, what are the longest-running experiments in history? And what are the rewards for keeping such an experiment going?

    Some examples:
    * The Egyptians built a Nilometer that tracked the levels of the Nile river for 1300 years (622 AD to 1922 AD). See http://www.waterhistory.org/histories/cairo.
    * A uniquely long-running physics experiment is the pitch drop experiment at the University of Queensland, running since 1927. But is this truly an experiment? I think it’s more of a demonstration. See http://en.wikipedia.org/wiki/Pitch_drop_experiment.
    * The longest-running experiments I could find in the U.S. were mostly agronomy and seed science, for example this 130-year-old experiment on the long-term viability of seeds: http://www.cpa.msu.edu/beal/research/research_frames.htm.
    The website calls this “the longest continuously monitored experiment in the world.”
    * The experiment that got me interested in “slow science” to start with was the NIST Stone Test Wall, which is monitoring the weathering of different kinds of building stone. See http://stonewall.nist.gov. That experiment has been running since 1948.
    * In public health, the Framingham Heart Study started in the same year, 1948. See http://en.wikipedia.org/wiki/Framingham_Heart_Study.

    Do these experiments have anything to say to modern scientists? Can today’s scientists even conceive of doing an experiment whose results would not be known for 50 or 100 years? In a world of 3-year funding cycles, could such an experiment ever be funded?

    If anyone has any answers to these questions, I would be interested in hearing them!

  16. #17 Jonathan Vos Post
    June 24, 2009

    Publish slowly or perish slowly.

  17. #18 Janne
    June 24, 2009

    “Publish slowly or perish slowly.”

    So the slower you publish, the slower you perish. The less actual papers in your CV, the stronger it becomes. It’s homeopathic science!

  18. #19 cm
    June 24, 2009

    I use a different phrase for slow science: good science. Good science is often slow, and painstaking.

    Slow science means ignoring the rushing sound of papers flying off your colleagues desks and thinking: a leap is better than the random walk of a thousand small steps.

    I’d rephrase that: “A thousand nonrandom small steps is better than a leap.” I disliked the “leap” guys in science–when you’re looking for the big thing, you tend to believe encouraging data too much. Get the craft down. Get the data hard. When the leap comes, it comes.

  19. #20 Dave Bacon
    June 25, 2009

    A thousand nonrandom small steps makes a leap, while a thousand random steps just makes you a drunkard.

  20. #21 Jon
    June 25, 2009

    Scientists work how they can, each finding their his own niche. I don’t know that there is anything wrong with fast science. Somebody has to do it.

    That said, it is not a race. If somebody else would derive my result in the next few months anyway, then why should I do it? My goal should be to advance science in the ways that nobody else can—as quickly as possible. :)

  21. #22 Ross McKenzie
    June 29, 2009

    Great post. It stimulated a lot of thoughts.
    My random response, including my answer to Tobias’ question is here

  22. #23 Jonathan Vos Post
    June 30, 2009

    The distance that you go with thousand random steps depends strongly on the dimensionality of the space in which you are stepping. In discrete 1-D or 2-D your probability of returning to the origin is 1. In the 3-D lattice, this probability of return is approximately 0.3405373296.
    See, for example:
    http://math.berkeley.edu/~chr/teach/Lec13_2006.pdf

    “The preceding simple analysis is not easily generalized to higher dimensions, where the situation can be quite different, e.g. if the drunk man wanders about in a two-dimensional field. As the dimension increases, it makes sense that the walker is less likely to ever find by chance the special point where he started.”

    Let alone the absorbing barrier that wins you tenure, or a major award.

    If you think that we live in an infinite dimensional Hilbert space, you may not even bother to take the first step. But you must. Slow Science, yes; stopped-dead-in-its-tracks Science, no.

    See also:
    “An Elementary and Superficial Introduction to Operator
    Algebra”, by Kenneth R. Davidson of the University of Waterloo [sort of the MIT of Canada], FieldsNotes, Sep 07, Volume 8:1, Fields Institute, reviewed by Juris Steprans, p.4:

    “As an illustration of the unexpected behaviour encountered in infinite dimensional spaces, Davidson posed the following question: Can a continuous curve be drawn in Hilbert space which is always moving perpendicular to earlier portions of the curve? Since the question refers to curves that are merely continuous, but not differentiable, a precise formulation asks for a curve such that the lines connecting distinct pairs of points along the curve are perpendicular so long as the two intervals along the curve between these two pair of points are disjoint. In finite dimensional Euclidean space this is impossible because there is not enough room in k dimensional space to have even k+1 mutually perpendicular lines. However, Hilbert space does allow for such exotic behaviour. There is
    a wrinkled curve which, loosely speaking, moves in a direction perpendicular to all previous directions at any point in time.”

    Or, alternatively:

    Jerry Garcia Interview
    November 12, 1987
    Part 2 of 4

    “Have you ever checked out any of the Grateful Dead clone bands?

    Yeah, but I don’t think that’s where it’s at, exactly. Really, it’s people who have to invent their own version of what the Grateful Dead is, starting now. Not doing what we’ve done, but digging the way we’ve done it, and doing what they’re going to do—continuing this notion.
    But somebody else has to see that—it isn’t going to work just by following our footsteps. It’s gonna work by taking off perpendicular to every direction we’ve gone off in.”

  23. #24 Michael Raymer
    July 14, 2009

    In this vein, I am particularly proud of the following reviewer’s remarks on a paper of mine recently rejected by a certain well-known letters journal, who wrote regarding some of the references we cited: “It should also be noted that Rytov’s original 1938 suggestion (Ref.[9]) waited about 50 years for experimental confirmation (Ref.[7]). Considering that Ref.[8] was published more than 20 years ago, one cannot conclude that this is a rapidly advancing field.” Exactly why I enjoyed working on it!