Alan Sokal of the Social Text hoax visits Stockholm

May 28, 2009

Many of the readers of this blog will be familiar with the 1996 hoax perpetrated against the journal Social Text. The physicist Alan Sokal, in an attempt to expose the low standards of intellectual rigour in contemporary post-modernist scientific debate, wrote a parody of an article of scientific theory, and got it accepted.

The article begins by asserting that many scientists continue to “cling to the dogma […] that there exists an external world, whose properties are independent of any individual human being […].” It continues by weaving a strange cloth of disparate concepts from physics and philosophy, without any real justification, increasing in preposterousness to a magnificent climax, where Sokal claims that Lacan (the notorious psychoanalyst) has derived a mathematical justification for the phychoanalysis of AIDS from differential topology theory. It ends by saying that mathematics must be revised to be able to participate in the struggle against capitalism, patriarchy, and militarism.

Afterwards, he published another article, revealing that the first one was a hoax. This caused considerable debate, most of which is found on Sokal’s website. Richard Dawkins has summarized the arguments from the side of the stringent scientists with his usual humour and brilliant ire in a review of Sokal’s subsequent book.

Alan Sokal and your humble blogger outside the Royal Academy of Sciences
Alan Sokal and your humble blogger outside the Royal Academy of Sciences

In his speech yesterday at the Royal Academy of Sciences, Sokal set out in more general terms to discuss the importance of the scientific world view.

In essence, his main argument was that the scientific method is simply an effort to find out facts with rational methods, and furthermore, that most of us are completely able to do so in most aspects of our lives. Yet we often keep double standards and cease to question some aspects of our world view. For example, religious people are perfectly able to examine the factual basis of any religion besides their own – i.e. when were the sacred texts written, by whom, on what basis, and why should we believe them. But when it comes to their own religion they revert to the circular logic that the doctrine of the faith is true because the doctrine says it is.

In his talk, Sokal pointed out four main enemies of scientific reasoning:

  1. Potsmodernist theorists and extreme social constructivists. Sokal noted with pleasure that this group appears to have retreated somewhat over the past ten years.
  2. Advocates of pseudoscience. Here, Sokal went to some length to explain the impossibility of homeopathy, mainly by showing that it is entirely inconsistent with our current knowledge of chemistry.
  3. Advocates of religion.
  4. Lobbyists and spin doctors. And here, Sokal went into a rather long critique of the Iraq war and the loose factual premises on which the war was founded.

After the talk, I and a few others brought up a discussion on whether we should really be so hostile to useful placebos, even when they have absolutely no scientific merit (e.g. homeopathy and acupuncture). After all, these methods are very useful for patients with certain conditions such as chronic pain, where real medicine often has little to offer.

He responded that there is a need for an ethical debate on what sort of deception you can subject a patient to, and that the complementary therapists have to face up to that discussion. An excellent argument.

All in all, an enjoyable evening! My favourite quote was: “I am a scientist, not a politician. I have the luxury of saying what I think is true:”

In the audience, I spotted dr. Martin Rundkvist of the Aardvarchaeology blog. He has also written a post about the event.

Anatomy of the cell’s brain

May 22, 2009

For some time, I have been wanting to write a post on the ”hardware” of information processing in the cell. AK’s Rambling Thoughts has however beat me to it, in a series of four blog posts describing certain core principles. If I try to sum up the most important points, here they are:

  • Enzymatic reactions can function, in principle, as analog computer circuits.
  • The sheer number of interlinked reactions, both enzymatic and involving the regulation of protein transcription, allow us to view the cell as a computer with a huge processing power.
  • Many of the analog circuits are wired with positive and negative feed-back mechanisms that enable them to give a digital, all-or-none response. This entails a loss of processing power, which is in AK’s opinion offset by advantages of speed.
  • The computing is “modularised” in the cell, meaning that some reactions occur only in specialised compartments (a synaptic bouton, for example), and communication within the cell can take the form of chemical gradients occurring as a consequence of reactions taking place only in one location.

AK gives many specific and detailed examples of the mechanisms that are involved in each of these processes. But to me, the real power of the discussion lies in the presuppositions that are not very explicitly stated, but very strongly supported by AK’s examples.

The cell deals in processing information
It is easy enough to see that cells process matter. They take up glucose and other nutrients and transform them into energy and structural components. It is not intuitive for many to think of these processes as manipulation of information, but here’s how I see it:

If the cell is going to be successful at all, it needs very fine-tuned regulation of its diverse activities. Any regulation that takes place in response to external factors can be thought of as an internal representation in the cell of its exterior. A map, if you wish, in many dimensions – oxygen tension, concentrations of nutrients, activities of hormones, and so on. This map will then be the input into the cell’s “decision machinery”, which will output some sort of behaviour.

Historically, the decision system has often been conceptualised as a set of enzymes acting linearly and without a lot of interconnections. With more sophisticated models derived from computer theory, it is far more likely that we will be able to capture some of the complex goings-on and generate accurate predictions of how these systems will behave.

Important principles emerge even from only a very general knowledge of the signalling networks
We do not have to know exactly which enzymes do what in order to make some pretty powerful inferences about the signalling network. Based only on a few known classes of interaction, like positive and negative feedback, it’s possible to envision properties of the whole system. Prime among these is the strong nonlinearity that must be expected when there are so many parameters.

In a way I envy physicists. If I poke my teacup, they will be able to predict exactly where it will move to, and when, and in what position it will stop. But if I poke a member of one of my cell cultures (gently), or Zelda, my mother’s Chinese Crested, there is no telling what will happen.

Zelda. Yes, she is cute.

Zelda. Yes, she is cute.

Suddenly the system might go into one of these digital, all-or-none responses. I do not believe they are motivated by speed, by the way, because I do not see why an analog response could not be as fast. Sometimes they are probably motivated simply by the necessity to decide some things one way or another. In particular, this concerns “cell fate decisions”, when the cell decides whether to replicate, or to commit apoptosis. There can be no half-measures with some things.

The cell is not a unit
Some things appear to us at first to be one and indivisible. For example, it came as a small shock to me when I learned that my brain sleeps in regions – it’s not at all necessary for the whole brain to sleep at once. I have also discovered an interesting subdivision when I am daydreaming. Rather often, I start constructing arguments or writing a speech in my head. No visual imagery attached. That’s left hemisphere activity. And then I find myself having hummed a tune with no lyrics, almost without noticing, perhaps for several minutes. That’s the right hemisphere entertaining itself when I am not watching.

The “I” of the cell can be similarly elusive. Some cells are several decimetres long, making it impossible for diffusion of small molecules to carry information from one part of the cell to another. Besides the specialised long-distance information transfer systems (such as axonal transport mediated by the cytoskeleton), such a cell is a set of functionally semi-independent decision-making centres. It’s a bit like an 18th-century colonial empire with faraway patches of land, only intermittently connected by slow sailing vessels.

There are times when I feel a pang of jealousy over discoveries already made, that other people have lived to experience. But then I remember that the pace of discovery has only increased and keeps increasing, and then I feel the same sense of wonder as one might before the sky on a starry night, when I think of what we will know about these systems in one or two decades.

Scientia Pro Publica is up

May 18, 2009

Scientia Pro Publica
The blog carnival Scientia Pro Publica has just been published at E. M. Johnson’s blog The Primate Diaries. It’s a cornucopia of interesting posts, among which one of my own has been included – on behavioural conditioning of the immune system. Among my favorites in this edition are a post by [weird things] on how a warp drive might work and another by biotunes about the evidence against antioxidant supplements.

Check it out!

I appear in French!

May 12, 2009

The WordPress software gives me interesting statistics. For example, it tells me which Google search terms have landed people here. Today the top-ranking search was “hund cm kg diagramm”, followed by “robert browne the physist” and “cells growing on top of each other”. Kind of poetic.

But not as lyrical as the French translation of my blog (again, thanks to Google), that is apparently out there on the web.

Try this:

Certains organismes pluricellulaires émettent de la lumière dans un chemin. Fireflies procéder balises qui brillent dans la nuit, et certains poissons de haute mer phosphorescente utilisation appendices pour attirer ses proies. Cependant, les renseignements contenus dans la lumière de ces émissions est sans doute pas plus de «Je suis ici”, ou peut-être “Je suis aller dans une direction avec une certaine vitesse”.

It’s the first paragraph of an older post of mine. Some metazoans shed light on the road, do you suppose? The original read like this:

Some multicellular organisms emit light in a conspicuous way. Fireflies carry beacons that shine in the night, and some deep-sea fishes use phosphorescent appendages to attract prey. However, the information content in these light emissions is probably no greater than “I am here”, or possibly “I am moving in a certain direction with a certain velocity”.

Absurdity. I love it! And on a more serious note, I wonder how long it takes before I will actually be able to communicate straight to the majority of the world’s population, through automated translation?

UPDATE: And it gets even better when the circle completes one more rotation:

Quelques métazoaires de faire la lumière sur la route, faites-vous supposer?

Fun with a fluorescence microscope

May 12, 2009

I have a new friend. It is a Leica DM RXA fluorescence microscope. She is not a baby anymore, but her senses are sharper than a razor, quite literally!

We spent some time together this afternoon. Here are some of the results:



You will get a chance to guess what it is before I give it away!

Open Science on knol

May 11, 2009

In my last blog post, I discussed the need for a way to make scientific results publicly available more easily.

Damien Chaussable of the Science 2.0 blog has an interesting idea in this vein. A current paper of his is being drafted as a knol, completely open for everyone to see.

The advantages are obvious: rapid, free dissemination, and in the best case: a chance to get valuable feedback before it’s submitted to a journal. The drawbacks are also glaring; it’s not searchable in science-specific search engines, meaning that people in the field will find it only by word of mouth. Nobody takes the responsibility of maintaining the URL over time so I’d have to cite it as a “personal communication”, and it will not count as a merit when the authors are applying for jobs or grants.

You might be surprised that scientists have been so slow to simply post their results on the internet. Indeed, this is the first instance I have seen. The reason is that journal articles are the currency that determines everything, and the data should be new when it’s presented there. Because of the inevitable lag times, new in that context usually means secret since a long time, rather than just discovered. Physicists and mathematicians are exceptions: immediately when they had invented the internet, they started circulating preprints there instead of by paper. This practice has grown to a huge database called arXiv, which is now the primary source of literature for many in those fields, and often the only place that a physics or mathematics paper gets published. Nature has started a similar pre-print server for the biological siences, which has however attracted limited interest. Since 2007, it has only archived about 800 pre-print manuscripts.

I like Chaussable’s idea, and wish more people were trying to find new ways for science to open up. But I am pessimistic about the use of knol and similar unindexed sites for scholarly communication. My next manuscript will be posted on Nature Precedings, and I will beg the readers of this blog to read it and give me comments and feedback!

On publication bias in the laborative biosciences, and my feeble attempts to avoid it

May 7, 2009

Clinical researchers and epidemiologists are very much concerned about publication bias. That means all factors causing less than the totality of performed experiments to be reported. A well-known bias, for example, is that positive results are much more likely to get published than negative results. If someone then tries to assess all the evidence for and against the benefit of a certain treatment, for example, the picture will be distorted. Therefore, it is important to publish results even when they are not spectacular.

In the laborative biosciences, the problem is far more pervasive than in clinical research. Therefore, I find it a bit surprising that people talk so little about it.

Sins of omission
When we write papers, the objective is to tell a good story. Usually, multiple types of laboratory methods and experimental systems are used in support of each other. Far are we from the studies of yore that investigated a single hypothesis with jut one method. In general, this is a positive development, of course, driven by ever-better methods to generate data easily.

Stories require sacrifices of the narrative. Sidetracks must be removed, darlings killed, and details fitted into the grand plot. Inevitably, data that are not interesting will be thrown out the window.

This is not to say that people deliberately withhold contradictory data, at least not very often. But nearly every experiment starts with a small pilot run, and if it turns out contradictory or confusing, it is so very easy to simply prioritise something else and then never publish it on the grounds that it was never conclusive. The final paper may report just half of the hypotheses that were addressed in the course of the study.

And in this manner, most papers are navigated through a sea of uncertainty, leaving dead hypotheses as corpses under the surface, invisible and unknown to all but the scientists who left them there. And that can be pruriginously annoying. Because then I have to do the experiments myself to find out, even if it is obvious that someone ought to have tried it before. Some experiments, such as optimisation protocols, are never reported simply because it’s boring for everyone except the ten people (sometimes including me) that have to struggle for days or weeks to get the procedure to work in our own laboratories.

Who is to blame?
The current system incentivises everybody quite heavily to publish very selected subsets of their data. Scientific journals want papers with strong evidence that points directly in one direction. Scientists’ grants and reputations are pegged to their success in publishing in the same journals.

In my last project, I did a small experiment which, if positive, would have warranted a bigger experiment to validate the positive finding. But it was negative. I did it three times, and it came out the same way, to I decided to throw it in the paper anyway. Then the next person who might want to do it doesn’t have to, and I haven’t hidden any data. I stuck it in the supplementary section: files that won’t even be printed but are accessible at the publisher’s website.

When we got the review comments they were generally positive, but one of the reviewers lamented, inevitably, that that particular experiment was too weak to prove anything. We considered taking it out. But I decided to argue the point instead, this time, and keep it.

I can’t shake the feeling that there has to be a better way for me as a scientist to make my data available to people!