What is science? I think about this a lot. Before I broke my notebooks apart into fine-grained projects, I used to keep an entirely separate series of notebooks just for science, as opposed to my writing and sketching, and in each science notebook I’d attempt to redefine exactly what I thought science was.
So it might surprise you that I stopped listening to a psychology audiobook when the professor said (paraphrased): “Wilhelm Wundt defined psychology as the science of mind, thus dooming it to failure, because science is based on the study of phenomena that are public, repeatable and measurable.”
I literally hit eject a few seconds after hearing that, and said aloud (paraphrased), “Now, this is what people mean when they say people in the soft sciences are trying too hard to emulate the hard sciences without understanding how the hard sciences work.” (There may have been a few curse words in there as well.)
Now, I said I stopped listening a few seconds later – right around the point where the professor said “psychology is defined as the study of behavior” because if the professor had that poor a grasp of what science is and of how his own science is defined, how could I trust anything else that he would say?
My friends in science, psychology is defined as the study of mind and behavior. There are plenty of well-formed psychology experiments that can be conducted that have no overt behavior at all – for example, whether a visual image causes signs of recognition in the brain as detected by a brain scanner.
Furthermore, whether something is public or not has nothing to do with with whether it can be the subject of science. Quarks are not public – they only exist in bound states – and if our theories of quarks are correct, their properties cannot be directly measured, only inferred from the behavior of particle aggregates.
Furthermore, whether something is repeatable or not is not a measure of whether it is scientific. That’s an empirical question. We have reason to believe that the phenomena of genetic inheritance, thermal noise in materials, and radioactive decay are fundamentally random, and can only be predicted in the aggregate.
So what is science? Richard Feynman was fond of saying that science meant “the sole test of any idea is experiment” and I love modifying that to say “the sole test of any idea open to observation is experiment” because observation crystallizes the key distinction between mathematics, science, and speculation.
This brings us to the part I like most about the “public, repeatable, and measurable” part of the professor’s definition, because while the subject of a science may not be measurable, there certainly has to be something that’s actually observable, or you’re dealing with mathematics or metaphysics.
And that brings us, in turn, to FOOM.
My definition of science is that it is the formal observational / operational method (FOOM) of bringing phenomena in the world in contact with modeling and experiment so those phenomena can be explained and our understanding of the phenomena in the world can be expanded. What does this mean?
Well, first off, the scientific method is formal. You may become like Thoreau and retire to a cabin to suck the marrow out of life, and you may learn a lot by doing that, and you might even write it down in a famous book that transforms many people’s lives – but that’s recounting your experience, and is not science.
The first step in science is observation of the world – actually, direct observation of phenomena in the world like Thoreau did – but what starts to transform it into science is formality. Formal means we bring to observation some kind of structure which enables us to collect a raw body of facts about a phenomenon.
For example, sleep researcher J. Allan Hobson recorded a dream journal over decades to start collecting a body of data that occur in dreams. Without a comparable body of data, it’s impossible to say what “typically” happens in a dream and we’re reduced to sharing anecdotes.
But of course, the dream journal of one particular person who happened to be a sleep researcher might not be typical – and that’s where the operational part of the method. If the dream journal seemed to produce useful data, then researchers can start to develop protocols that analyze them more rigorously.
Formal observations that have been operationalized have been embedded in a set of procedures which enable the observations to be collected reliably – at which point we start calling them measurements (or more broadly evidence, if the observations are not very number-like, for example, narratives).
Science doesn’t start with physicists building a two-mile-long particle accelerator to measure the charge of the electron. It starts(ish) with Ben Franklin doing some damfool thing with a kite, string, key and bottle, progresses through Michael Faraday having to instruct people on how to interpret his magnetic induction experiments, and ends with the discovery of the electron by J. J. Thompson a century and a half later.
By the time that happened, the operational methods were so good that Thompson got a read on the mass / charge ratio of the electron when he isolated it. But here’s the thing: the electron itself, that is, the particle, was a discovery, not at all apparent in Franklin’s time – when electricity was thought of as a fluid.
Now we have principled reasons to believe, based on public, repeatable measurements, that electrons are fundamental in the same sense that private, quantum mechanical, unmeasurable quarks are fundamental, and in the same sense that composite objects like protons, atoms, and psychology professors are not.
But we didn’t get there by starting off with phenomena that were public, repeatable, and measurable. We started off wonding about a phenomenon as unpredictable as lightning, and by a centuries-long process of creating formal methods to turn observations into operationalizable measurements – which we used to craft experiments to explore the phenomena we discovered along the way, no matter how weird they were.
That’s why I say that the formal observational / operational method, which I call “FOOM” in my conceptual lexicon, is the foundation we need to lay in order to subject the ideas we have to experiment. And that difficult process of reducing chaos to order while being open to surprises, to me, is the essence of science.
-the Centaur
Pictured: the Physics section of Moe’s Books in Berkeley
