I wrote ‘Squarespace’ in a bit of a flurry along with some other songs that have ended up in various places, all inspired by reading Susanna Clarke’s ‘Jonathan Strange and Mr Norrell’ at the end of 2015. I was on holidays in New Zealand with my partner Adele, recovering from doing my honours year in psychology. Lying in our room above a pub in Auckland, I was absolutely hooked by Clarke’s novel, so much so that I finished it in a day and then read it again the next day. Ostensibly about a competitive rivalry between two great magicians, I was most invested in the backdrop of magic returning to England and the love story between Jonathan Strange and his wife Arabella. There was a part in the novel where Jonathan Strange is working obsessively on his book, not noticing that Arabella is essentially being possessed by magic. Having just spent an intense year working on my honours thesis (not always to the benefit of my personal life) I identified with this, and wrote Squarespace from the perspective of someone half way between myself and Jonathan Strange. It takes the form of a half-apology (more of an explanation) from someone who is obsessed with finding truth to their significant other. The feeling it tries to evoke is one of nervous intensity, but with a touch of tragedy - many relationships have been lost because of single-mindedness.
I wrote the song in 2016, and it sat on the shelf for a while. It almost ended up attached to a few different projects, but never quite landed. I recorded the vocals and guitar that can be heard on the final track around this time. From 2017 until 2020 I was mostly occupied with my PhD thesis, which made Squarespace feel even more relevant to my life. The narrator of the song is desperately trying to find truth, but finds themselves ‘drowning circles in a square-shaped space’, which is a pretty good way to describe the frustration and satisfaction of doing a research PhD. Indeed, I was becoming very interested in finding truth. Psychology was beginning to go through the ‘replication crisis’, where many experimental studies were found to fail to replicate because it was generally acceptable to try and make the results fit your hypotheses by selective reporting, changing the hypotheses, and/or opportunistic statistical analysis. I was interested in what it would mean to do ‘good' science in psychology, beyond the steps that were already being taken (such as preregistration of hypotheses before data collection). How would I know whether or not I had found truth in my research? Being a PhD student meant I had the freedom to spend a bit of time on a detour like this, and I feel like it was one of the most valuable parts of the whole experience. So, I dived headfirst into philosophy of science. I wanted to know what makes some science more ‘truthy’ than other science.
There is a ‘big 4’ in the history of 20th century philosophy of science, and I went through the works of each of them in turn. I started with Karl Popper, who (among many other things) came up with the idea of ‘falsification’. The idea behind this is that accumulating evidence to support a scientific theory doesn’t prove anything when the theory is so vague that almost anything could be used to support it. In particular, a theory is unfalsifiable where one outcome supports the theory but so does the exact opposite outcome - it doesn’t allow itself to be proven wrong. According to Popper, good scientific theories are specific enough that they could be wrong, but have so far survived testing. These theories are not not necessarily ‘truthful’, but they are ‘not yet false’ - the possibility of being false is what gives them credibility.
Falsification is a brilliant way of getting around the problem of induction (that is, no amount of supporting evidence can prove that something is true, because an exception might still be found). On the other hand though, trying to prove your own theories wrong is a bit of a depressing enterprise, particularly in a field like psychology where the tools and methods (not to mention the theories) are still quite primitive. Research is a creative process, and it’s demotivating to be searching for falsehood rather than truth, particularly when falsehood is so easy to find. The next author I went to was Thomas Kuhn, whose book ‘The Structure of Scientific Revolutions’ introduced the concept of the ‘paradigm shift’, where theories are overthrown because they eventually fall apart in the face of reality, being replaced with a new ‘paradigm’ that provides a better explanation. Kuhn’s analysis was not so much about formal logic as about how science is actually done. In this sense, it is quite accurate, if depressing for a truth seeker. It certainly doesn’t offer much help if you want to do good science yourself.
I next turned to Imre Lakatos, who combined the insights of Popper and Kuhn in proposing that theories should be evaluated in terms of whether they are progressive or degenerative. A progressive research program is one where the theory creates new knowledge about the world, while a degenerative program creates ever-more tangled additional clauses to prevent it from being totally overthrown. When a research program becomes too degenerative, it should be replaced by a new theory. What Lakatos was trying to address was something I had also struggled with in taking a Popperian approach to doing science - what should be done when a theory is partly proven false? I thought his solution to this was quite elegant.
However, then I read the work of Paul Feyerabend, who gleefully demolished his friend Lakatos’ ideas in the great work ‘Against Method’. He pointed out that there is no way to know whether a currently degenerative research program might not suddenly start to become generative, or vice versa. Feyerabend thought that we should keep all ideas and ways of thinking around just in case they become useful for a given context. Moreover, he argued that there is no consistent method of doing science that fully separates good science from bad science. Any attempt to nail down a specific way of doing science can only serve to prevent great science being done. In his words, ‘anything goes’. This could be seen as a cop out, but I found it freeing - just do science as best you can, either in accordance with how things are usually done or not (depending on how much you care about other scientists liking you). The plurality of ideas this approach generates makes more sense when you think about science as a collective enterprise - there has to be room for different ideas, methods, and perspective because it is this that allows science to adapt to an endlessly complex universe.
So that’s where I ended up with my search for truth (in science at least). You can never logically prove that anything is true, and in trying to find truth ‘anything goes’. Believe it or not, this was actually incredibly helpful for me - it took the pressure off that I was putting on myself to do science ‘right’ according to some universal set of guidelines. Instead, I just did (and still do) my research as best I could. Rather than trying to find truth, I started to aim for ‘usefulness’. It’s much easier to operationalise usefulness, ironically because it is subjective! Psychologists are in their element measuring subjective thoughts and feelings. This didn’t stop me from working on more theoretical work either - in fact, most of my work is theoretical. But I treat that work less as a search for truth than as an attempt to improve on what exists, or to address a specific problem. Science as problem-solving is more tangible and ultimately satisfying than science as truth-seeking.
So that’s how I escaped Squarespace - putting circles into circle-shaped spaces, and looking for squares to put into the square-shaped spaces. There’s enough madness in the world without self-inflicting it.
Comments