In conversation with Rafe Champion

 

A group of unrelated and unknown people meet in a room. They don’t ask about credentials and they have nothing in common beyond their desire to be in that room. Outside on the street, in their cars, in restaurants, with their families and friends, these people are as insufferable and flawed as the rest of us: gossiping, threatening, sweet-talking, fighting – dogmatic, tribal, loyal, and full of prejudice. But inside that room everything changes: they argue and criticise and interrogate, but all their other human baggage is left at the door. If you were to accidentally walk in and watch them for a few minutes, you would find it impossible to guess who the senior members were, and who were the juniors, who had had more professional success and who had less, or what their lives and personalities might look like when the meeting ends and everyone goes home.

These hypothetical people are scientists, and for most of us this is enough to explain the unusual behaviour of that room. By extreme good fortune we have become accustomed to the sudden shift in attitude and rigour. But what actually happens beyond that threshold takes some clearing up. And asking the scientists themselves likely won’t get you there – for the most part, they don’t know either! So how is it that these deeply individualistic people manage to leave their egos at home and work towards something impersonal and collective, without being conscious of the process themselves?

It is what has come to be known as the “social turn” by Ian Jarvie, the “institutional turn” by Rafe Champion, or “the organization of inquiry” by Gordon Tullock, in the philosophy of science; and it belonged first to Karl Popper:

what we call ‘scientific objectivity’ is not a product of the individual scientist’s impartiality, but a product of the social or public character of scientific method; and the individual scientist’s impartiality is, so far as it exists, not the source but rather the result of this socially or institutionally organized objectivity of science.

Now this seems odd… counterintuitive. Most Popperian readers will have a clear-enough image in their minds of what science ought to look like. It involves as many competing theories as possible, as much criticism and as many tests as possible, and as much advocacy and argument. We will never reach an upper limit on this, a place where we stop and decide that science has become drunk on its own health, and so the question of organisation doesn’t seem like much of a question at all: if you want science to succeed, simply get out of its way; let the free choice of free individuals rule, hugging as close as we can to laissez faire principles.

Popper would agree with those words, and disagree with the implication. How those individuals fit together also plays an important role in this – without which, ‘scientific objectivity’ would be impossible. At the end of the day, science actually achieves things, it solves problems, moves forward, improves the world, and makes progress. It is more than just a collection of unconnected scientists building their own hypotheses and running their own tests. If it were only this, it would be a miracle if even a single problem was solved.

Perhaps a clearer way of looking at the problem is in reverse. What would you do if you wanted to cripple the progress of science and the growth of knowledge? Here is Popper:

By closing down or controlling laboratories for research, by suppressing or controlling scientific periodicals and other means of discussion, by suppressing scientific congresses and conferences, by suppressing Universities and other schools, by suppressing books, the printing press, writing, and, in the end, speaking. All these things which indeed might be suppressed (or controlled) are social institutions…Scientific method itself has social aspects.

There are two interrelated motives to everything scientific: understanding the natural world and controlling the natural world. It is the difference between “pure” and “applied”, between curiosity and a practical purpose. Recognised in nearly every lab, in every country, the line is as appropriate as it is fuzzy. Like it or not, once the step of understanding the natural world has been successful, the controlling of the natural world has already – largely – come to life, with the many immediate uses, tests, experiments, applications, and implications, being present and explained by the pure research discovery. There is never a hard and obvious demarcation in the scientist’s mind – unavoidably slipping between the pure and applied titles as he works.

The objective truth of things is unknown and out there to be discovered, and we only ever get there by guessing (conjectures) at what that unknown world looks like. This is the way knowledge creation happens, whether it is in the mind of a great theoretical scientist or in the dusty corner of an undergraduate laboratory (still in the mind, of course). These people might have different pressures, different funding incentives, or different motivations (creating something vs. becoming rich), but as they are both trying to create new knowledge (the knowledge of how something works or the knowledge of what to do with it), they are both scientists of the same kind; the same kind of scientist that we all are… every single one of us!

At all stages, the same game is being played and the same question asked: do my theories about the world actually correspond to reality? And there is no such thing as a theory which doesn’t try to connect with the world beyond our minds. Without a phenomenon of some kind that needs explaining, a scientific theory is never born. This was Galileo’s great crime, not discovering a new motion of the planets, but claiming that this represented a new law of nature and a better way to explain what was actually out there in the universe. The charge against him was presented by the Jesuit cardinal in this way: “act prudently” and “content yourself with speaking hypothetically and not absolutely.”

If the work of science was to only state what we see, and then revise what we see, and never draw a connection to underlying laws of nature – making our theories “simply abbreviated statements of observations” – then we have some hard lifting before us. When the Einsteinian system replaced the Newtonian one, it wasn’t because it explained more observations; after all there are very few observations that Einstein’s theory can handle more simply, or effectively, than Newton’s. If the whole game were about data collection and accuracy, then general relativity would never have been regarded as anything more than a “minor step forward”.

Albert Einstein was supposed to have said, “if you want to know what a scientist really believes, don’t listen to what he says, but observe what he is working on.” And very few will ever be caught working on theories or experiments in a way that would indicate a scepticism about objective reality, or as if their research were only “devices for conveniently summarising experimental results.” Pick a scientist at random, controlling only for their being committed to their job, as well as being honest. Working day and night, sweating and suffering for their theories and experiments, what they probably don’t do is study their own interests and motivations; asking themselves why certain problems were chosen, why particular methods used, or how truth relates to what they are doing.

Because they don’t spend much time on these issues, and because so much of what they do understand is inexplicit, they are liable to make some strange mistakes when pushed – grabbing hold of “various ill-conceived theories”. It is, after all, a hard thing to say publicly – as well as to oneself – that I don’t really know what I am doing! Left alone to scurry around their labs, these scientists take plenty for granted, and stumble onto unspoken answers: the possibility of truth is the only thing that brings meaning to their work, and progress to science. Without it, all their achievements would be miracles!

We (scientists and non-scientists alike) don’t begin by collecting data, we form theories, and then test those theories through experiments, against other theories, and finally against the collection of data. The point being, the accumulation of information gets us absolutely nowhere, until a human mind creates a theory to make sense of it; as well as to make the accumulation possible. Without a theory how does anyone know what to collect in the first place, or what constitutes a data point?

Straighten this out as well as you can, find your methodological groove, and what is left – according to Popper – is only still ever conditionally accepted as true. It is always open to revision and scepticism, and this also leads many scientists to make elementary mistakes about the nature of their work, thinking that the critical attitude which has been drilled into them has unhelpful implications about the scientific enterprise. It goes like this: if I am required to be sceptical about every possible theory, this surely means that they are all untrue. Or: if I am required to be sceptical about every possible theory, then I ought to also be sceptical about the very existence of truth.

And all sorts of horrible little ideas fill that opening space: science is about consensus or science is about workability or science is about the collection of data. The shame and the errors are magnified by one important – and true – implication, being the one which is most commonly missed: our “theories seem not to last”; that the history of science has been a history of radical change and of disproof. Nothing that we do ever seems to stand the test of time – all it takes is a little bit of well applied criticism, and everything we once thought to be true crumbles inevitably to our feet. We are always likely to be wrong (in one way or another), and yet never likely to see this for ourselves.

Its origins are as long and deeply carved as you could hope for, stretching back to The Logic of Scientific Discovery and to The Open Society and Its Enemies, but this social turn in Popper’s philosophy always happens at the intersection of human fallibility. As a small proof of this, Popper himself seemed to not fully understand the significance of his turn; snow-blinded to his own enormous discovery. “Popper’s consistent ability to think socially also does much to account for his originality” writes Jarvie in The Republic of Science, “since it is hard to do and its difficulty is attested by how often readers and critics of Popper do not grasp that this is what he is doing.”

So wrapped-up in defending science against the charge of being a “mere social construction”, Popper’s snow blindness was largely self-inflicted. Perhaps the word turn is not helpful here, indicating a slight or casual change, a subtle mention or glance in an otherwise ignored direction. This underplays just how essential “thinking socially” is to Popper’s work, much more so than “logically or psychologically”. It is so central to everything Popper thought, and to the scientific method, that Jarvie believes it amounts to a “proto-constitution of science.”

It is what keeps science afloat and functioning, even when the best methodology is not used. Alexander Fleming’s discovery of Penicillium rubens was not the accident that most people think it to be. Thousands of other researchers had seen the same contamination in their slides – and millions more had seen it on the non-microscopic level – but only Fleming realised its importance. An accident may have led to the contamination, but not to the creation of the theory; that was all Fleming! Still, it was an example of bad methodology, and the reason why the theory holds today is the same reason why most others fail: community!

Not a community of like-minded colleagues, nor a community hugged together by a governing body or a set of laws, but a group of otherwise disconnected people committed to doing the one thing for each other that we cannot do (particularly well) for ourselves: expose our mistakes. No one gives commands, there are no hard organising principles, and none of it is consciously designed. It all hinges on the key truth that, if you desire new knowledge, then you must want to find and eliminate error. And it is this which requires a community… of a kind.

The theories that survive also owe their lives to this social world of scientists, without whom (and their best efforts at criticism and refutation) they would be indistinguishable from the crowd of false theories. A perfectly true hypothesis can never be considered as such until it moves from its host’s mind into the minds of other scientists; with all the doubt and difficulty and explanation and testing and predictions and implications that goes along with that.

There are better and worse ways for this to happen though. Dissemination is always a challenge and a balancing act – trying to get new theories exposed to the scientific community as quickly and cleanly as possible, whilst also filtering-out the frivolous, the nonsensical, the fictional, the non-rigorous, the fraudulent, and the fabricated. We all have a limited amount of time and attention to spare, after all. When this is done well, scientific knowledge grows sharply, benefiting us all. Done badly, and the social fabric tears under a deluge of un-distilled information and phony publications. The problems of science are the problems of the human condition; Tullock puts it like this:

There is no reason to believe that scientists are much more thoughtful and honest than other men. The obvious high degree of truthfulness in scientific research comes not from the superior moral probity of the individual scientists, but from the social environment in which they operate.

A scientist caught faking his experiments or fudging his results is not an existential threat to science – nor to the scientific community – but he does represent a muddying risk to the smooth function of things, and the pace of progress (a non-trivial problem when your whole goal is to improve things, through knowledge creation, as quickly as possible). Matching the seriousness of the crime, the punishment is invariably excommunication, the end of a career, the thorough collapse of reputation, and the questioning of all previous work.

In the 1920’s, Paul Kammerer was one of Europe’s most prominent and well respected biologists. After years and years of success and rigour and achievement, he was then associated with a single faked experiment… and it “ruined him”. Kammerer’s downfall was sudden, dramatic, complete, and irreversible; ending, sadly, with his suicide in 1926. The fraud he took part in speaks of a tragic desperation that he must have been feeling. Fabricated theories are unpredictable, fabricated experiments are unrepeatable, and fabricated discoveries have no practical applications. Kammerer was always going to be exposed, with his deception sooner or later bubbling to the surface; which is why the scientific world is more truthful and honest than the non-scientific: lies are better and more seriously policed!

So if everything comes back to the question of effective dissemination – allowing scientists to check each other's work as quickly as possible, whilst also filtering-out the obvious errors – then how should this be done? These days everyone has the answer, infecting their vocabulary like a pathogen escaping a laboratory: peer review! And it means absolutely nothing – no content implied! How peer review works, and how it should work to be more effective, remains a mystery beyond the plain meaning of those two words. At its bones, it is as crude as the worst aspects of human life: “scientific advances are disseminated” writes Tullock, “through the same channels of advertising, salesmanship, and public relations as other commercial products… [and] this does have some effect on the development of science.”

So, much of what is good science misses out on publication, not because it lacks merit or quality, but because it doesn’t line-up with editorial guidelines, audience expectations, or the reviewers simply don’t understand it. All journals are specialised, but this specialisation can only go so far, and sadly the “most important” new research often falls on deaf, and confused, ears. Then the reputation of the author becomes a problem: notable scientists get too much of a free pass, while the unknown are rejected on the assumption that their unknownness is a symptom of their poor research. After that comes the problem of the sheer number of journals that now operate (all trying to meet publication deadlines), meaning that ten rejections of an article has no necessary implication for the future of the research. Short on content for the next quarter, the eleventh will take it. If not the eleventh, then the twelfth…

How to save science from its own institution? And from becoming a victim of diminishing returns? Choose your poison: improve the tenure process, offering more protection for researchers at all stages in their careers; do away with the résuméd-importance of having a flock of grad-students (encouraged to support and advance their supervisor’s work); expand access to funding and equipment; place more value on the receiving of awards (encouraging boldness) rather than the accumulation of publications; or perhaps something much smaller, much simpler, but which would have a disproportionate downstream effect…

For this we go back to Gordon Tullock and his hard look at the structure buttressing the scientific enterprise. One idea, one requirement upon the editorial boards of academic journals, would be revolutionary: make them publish a list of their rejections. From this, we could see how discerning they are, how much they reject vs. how much they publish; we would also begin to see any biases that they might have, if they were consistently rejecting papers from one view point for example; and the next time a ground-breaking paper is published, you could go back and see all the journals that rejected it before it was published, popular, and acclaimed.

Or, as Rafe Champion points out, you could simply give new scientists “a good introduction to the works of critical rationalism” and Karl Popper.

 

*** The Popperian Podcast #17 – Rafe Champion – ‘Karl Popper’s Social Turn’ The Popperian Podcast: The Popperian Podcast #17 – Rafe Champion – ‘Karl Popper’s Social Turn’ (libsyn.com)