Physicists undertake stamp-collecting

Ernst Rutherford, the "father" of nuclear physics, once airily declared "In science there is only physics. All the rest is stamp collecting". By this he meant that the theory of physics is the only significant thing in science. Such mundane activities as taxonomy in biology were just sampling contingent examples of physics.

So it is with some amusement that I note that in order to make sense of string theory, a group of physicists have been trying to do taxonomy over string theories. Why this is more than a "gotcha!" is that since the late nineteenth century, philosophers of science have ignored classification, although some of the more important advances in physics relied on it, such as Mendeleev's Table, which drove theoretical advances in both chemistry and physics (and led even more ironically to the understanding Rutherford had of radioactivity).

The last significant English speaking philosopher of science to discuss classification in science at length was W. Stanley Jevons in his 1873 book, second edition 1878. John Dewey treated classification as an important, but largely operational, activity, akin to (and leading to) classification in libraries. That classification had a crucial role to play in the scientific epistemology was not argued, so far as I can tell, until fairly recently, with the taxonomy wars of the 1970s and 1980s, and even then the biologists appealed to Karl Popper, who has three dismissive comments I can find in his entire corpus, on classification. Popper clearly agreed with Rutherford on this.

Classification dropped out of philosophy of science for a century. Arguably it is still out, since those arguing for it rely on a philosophy of science in which classification is uninteresting. I am going to propose how classification is not only interesting but one half of science. Suppose we take science to be on a continuum of ordering activities, ranging from theory-building to classification. Suppose we also take science to have an orthogonal empirical axis ranging from experimental manipulation to passive observation. Science is thus played out on a field like this:


Now, traditionally, and by that I mean since around the 1930s, philosophy of science has held increasingly to two views about science. One is that science is all about the construction and testing of theories. The other is that all observation is theory-bound. This effectively meant that under the standard view, science was properly restricted to one quarter of the field of epistemic possibilities:


Passive observation went out the window with the fact-value distinction, the myth of the given, and observation sentences, while the construction of classifications was relegated to the activity of museum curators, librarians and possibly teachers.

Needless to say, biologists, and in particular field biologists, refused to go along with this for a very long time, and there remain a number of holdouts. However, the introduction of a physics (and chemistry) based discipline in biology - molecular biology - meant that even in that domain, the standard view was at least palatable, if not entirely helpful.

Hence my amusement. While these physicists are trying to order the vast field of possibilities into a logical structure, which is only one kind of classification, they are not, as yet, happy about passive observation, and indeed in the context of string theory, or physics in general, it is hard to see what might count as passive observation. But there is a kind of passivity to observation even when it relies on technical apparatus like the Large Hadron Collider - you see the readouts, displays or cloud chamber photos without much in the way of theory - the interpretation is what requires theory.

It seems to me that the assumption of classical (not 19th century, which is too classical, but 20th century) philosophy of science is that all useful and actually scientific activity happens in that quartile, and the debates over what the scientific method is, are about what the trajectories should be. Do we start with experiment and move to theory? No - problem of induction. Start with theory and move to testing. Problem of Quine-Duhem, and so on...

But what if there is no standard trajectory? And what if the field of scientific activity includes the other three quartiles?


Now we have a slew of scientific activities that can include the field biologist, the museum curator, the naive experimentalist, and so on, all of which was excluded from "real" science by the analytic post-logical empiricist tradition. And if the entire field is open to occupation by a researcher, a research group, a tradition, research program or even an entire discipline, and there is no "next move" one has to follow like a computer following programmed instructions, now you get a much richer notion of what it is to do science.

Jevons, William Stanley. 1878. The principles of science: a treatise on logic and scientific method. 2nd ed. London: Macmillan. Original edition, 1873.

More like this

Over at Evolving Thoughts, John Wilkins pokes string theorists: Ernst Rutherford, the "father" of nuclear physics, once airily declared "In science there is only physics. All the rest is stamp collecting". By this he meant that the theory of physics is the only significant thing in science. Such…
Theory: A word that gets used a lot in discussing science, or attacking it. Theories are only verified hypotheses, verified by more or less numerous facts. Those verified by the most facts are the best, but even then they are never final, never to be absolutely believed. [Claude Bernard, 1865,…
Carl Zimmer has one of his usually clear and precise articles on recent work on the nature of life, focussing on the work of Carol Cleland, who is at the National Astrobiology Institute, despite reduced funding for actual science by the present administration. I met Carol last year when we both…
In this post, I want to propose my own view, or rather the views I have come to accept, about the nature of science. [Part 1; Part 2] There are three major phases in the philosophical view of science. The first was around in the nineteenth century - science is the use of inductive logic based…

John, I suggest that it is far to early to define 'string theory' as a part of physics or any other science. It is an intriguing mathematical speculation but as physics it is 'not even wrong': i.e. it has not, at least yet, made any falsifiable predictions!

John, thanks - I fixed it.

Time - I am so not getting into that issue. From a philosopher of biology perspective, it looks like physics from here ...

By John S. Wilkins (not verified) on 28 Mar 2008 #permalink

I thought you were exaggerating, but this really does look like stamp collecting. But is it physics? And what did the author mean by "nearly infinite?" That might be a lot of theories, but I can't tell for sure as I have no idea how much that really is.

By John Vreeland (not verified) on 28 Mar 2008 #permalink

I don't get your point. Much of astronomy has been done by building new instruments (like radio/IR/gamma telescopes) Then observing as much as possible, classifying the results in to categories and then waiting for the theoreticians to catch up. That lead to entirely unexpected discoveries like black holes, pulsars and dark matter.

I agree that classification hasn't been much of an issue in the history of philosophy of science, but that doesn't mean it isn't picking up steam. The best examples I can think of are investigations of folkbiology. Scott Atran has done some great work here, as well as non-anthropologists like David Hull. It's one of those interdisciplinary explore how the human mind works through examination of a universal strategy towards the environment kind of things. There is such a thing as folk physics, as well, although it usually isn't as culturally prominent.

The "String Vacuum Project" is a quite different classificational project than the ones you're concerned with. What they would like to classify is not things coming from experimental observation, but an exponentially large number of mathematical models, none of which have been shown to have anything to do with the real world. It's kind of like stamp collecting, as done by someone who doesn't have any stamps, just a lot of ideas about what stamps might look like...

Peter: Oh I like that, I really do. I realise they are doing that sort of taxonomy - and I recognise that it's rather different to what biologists typically mean by it; rather more like using theory to form the classes.

Don, as my work has been on Hull and Atran (it's all right; you weren't to know) for the past ten years or so, among others, it is because of them and those others I am now concerned with classification.

And no, Bob, the Willi Hennig Society wouldn't have me.

By John S. Wilkins (not verified) on 29 Mar 2008 #permalink

Bad taxonomy is the root of all evil.

By Jim Thomerson (not verified) on 29 Mar 2008 #permalink

The war between the stamp collectors and the "scientists" actually has its origins in modern science at about the middle of the 17th Century and was originally between the Baconians and the Mathematicians (later to become the Newtonians) and played a significant role in the social history of British Institutional Science up till about the middle of the 19th Century (about the time Jevons was active!). Bacon was the stamp collector supreme and rejected both theories and mathematics as starting points for science, resulting in his famous rejection of the work of both Copernicus and Gilbert, thus evoking the wonderful remark from Harvey, his one time secretary, that Bacon writes science like a Lord Chancellor.

When the Royal Society was founded in 1660 it was predominately Baconian and several mathematicians refused to become involved for that reason. When Halley, a mathematician, became secretary in the 1680s a group of prominent Baconians under the leadership of Flamsteed left the society and set up a short lived rival organisation in Oxford. When Newton, the mathematician supreme, became President in 1703 the real power was still in the hands of the secretary Hans Sloane, a true Baconian, (whose natural history and antiquities collections formed the basis of the British Museum after his death) and there ensued a period of guerrilla war fare within the Society until Sloane retired in 1712 and Newton finally gained full control. From then until Newton's death in 1727 the mathematicians ruled supreme. After Newton died the Baconians staged a coup and Sloane was elected president and the Baconians reigned throughout the rest of 18th Century and well into the 19th. This dominance of stamp collecting was particularly strong during the presidency of Joseph Banks. The result of all this was that the Baconian sciences, biology, geology, palaeontology etc, flourished in Britain in this period and the mathematical sciences withered.

Early in the 19th Century a group of mathematicians at Cambridge started to campaign to bring the mathematical sciences in Britain up to the level of Continental Europe which had forged ahead in the 18th Century under the influence of Euler, the Bernoullis and others, this group was called the Analytical Society and included Peacock, Babbage, Herschel and De Morgan among others. Members of this group were instrumental in the founding of The British Association for the Advancement of Science at the beginning of the 1830s to counterbalance the influence of the still strongly Baconian Royal Society.

I don't get your point. Much of astronomy has been done by building new instruments (like radio/IR/gamma telescopes) Then observing as much as possible, classifying the results in to categories and then waiting for the theoreticians to catch up. That lead to entirely unexpected discoveries like black holes, pulsars and dark matter.

Observational astronomy is not physics and is very definitely stamp collecting and there is a very big difference between observational astronomy and cosmology. One of Newton's most famous disputes was with Flamsteed the Astronomer Royal and this dispute is described in almost all Newton biographies however very few of the biographers give the real reasons for this dispute. Newton is/was the founder of the modern scientific method and used observed facts to test his theories and wished to use Flamsteed's observations for this purpose. Flamsteed was however a true Baconian, theories should only be formed when enough observational facts have been collected, collated, catalogued and categorized then and only then. He regarded Newton�s use of his data to test, in his opinion, premature, and therefore dubious, theories as a misuse and an insult. What we actually have here is not an argument about data but a war between two conflicting philosophies of scientific method, Rutherford's "scientists verses stamp collectors".

In the twentieth century almost all philosophers of science, before about 1980, took classical physics as their role model in order to develop their philosophies resulting in them actually producing philosophies of classical physics that are at best difficult to apply to other areas such as the "Baconian" life sciences. Mr Wilkins' plea for a wider understanding of scientific method and a much broader based philosophy of science is a much-needed correction to this tendency.

The brief comments above should of course bristle with foot notes and references to back up the statement there in. Unfortunately it is several years since I read up on the topics mentioned here and it would take me at least a month to provide the necessary references (and I can't be bothered!). If however anybody is particularly interested in any of my claims and wishes to follow them up I can provide some road signs to help them on their way.

Thorny C. "Observational astronomy is not physics and is very definitely stamp collecting and there is a very big difference between observational astronomy and cosmology."
As a graduate student back in the '60's astronomy was under the umbrella of the physics department - at least in my institution. Astrophysics was still relatively new and accepted, by most of the astronomers that I came into contact with, as a means of explaining the phenomena they observed: Cosmology, however, was still considered to be speculative fantasy by many!

um... ironic. Rutherford develops a taxonomy of science then uses his taxonomy to argue that taxonomy is useless....

on a more serious note: i don't believe that any observation is 'passive'; i postulate that any organism capable of observing analyzes its observations, and in so develops 'theory' (or at least, hypotheses).

Hmm. Yes, but there's an important litmus test for all philosophy of science: if you put it to a practicing scientist, does he find it enlightening, or even interesting? Rutherford's philosophy should be just as subject to this as anyone else's.

And if you posed it to a serious ecologist up to their knees in mud, they're likely to say, "Alright buddy, YOU tell me what the snapping turtles in this swamp eat!"

Thanks for this. It is exactly the method I've used when teaching science to young children. I tell them that science is all about looking at the world very closely, then asking questions about what they see.

All things flow from observation; how else could the pursuit of predictive knowledge, and its subsequent categorization, have begun?


Thony, I'm going to have to get you to write a guest blog soon.

But despite the rise of rationalism over Baconian induction and empiricism, the topic of classification and naive observation was live as late as Jevons, and in some sense later. [And Jevons has some strident comments about laws of anture: "The laws of nature... are simply general propositions concerning the correlation of properties which have been observed to hold true of bodies hitherto observed"! - page 738 of the second edition]

So we do not need, in this instance to go that far back. But I think I will write this up as "a plea for Baconian stamp collecting (induction by enumeration)".

By John S. Wilkins (not verified) on 31 Mar 2008 #permalink

Snowflake wrote:

Thony, I'm going to have to get you to write a guest blog soon.

If you do I shall do a blog on Australia's ultimate style icon ONJ! ;)

If you do I shall do a blog on Australia's ultimate style icon ONJ! ;)
Posted by: Thony C

With lots of pictures of ONJ? Pretty please :o)

By Chris' Wills (not verified) on 04 Apr 2008 #permalink

John, your post and Thony C's comment (well, the first one) gave me a very good education of a Sunday evening. Thank you.

Jim, I missed your comment before:

Bad taxonomy is the root of all evil.

to which I can only say that I thought evil was unrooted, by consensus.

By John S. Wilkins (not verified) on 06 Apr 2008 #permalink