Recently, a friend of mine who works in finance told me about a graduate his firm had employed. The new hire was highly motivated and gifted in the quantitative space – the models they created were, evidently, the products of exceptional academic training. And yet they couldn’t effectively apply those models. They didn’t understand that the referents required contextual application. They were not a good employee.
While the recent focus – even obsession – with quantification in society is a helpful one, it’s important to ask what costs it comes with. Numerical literacy will be essential for the next generation of researchers, clinicians and corporate executives, but understanding how to apply mathematical understanding is, as the graduate’s story suggests, no less important.
In the evolutionary sciences (my own field, broadly speaking), the dichotomous approach to research has manifested itself in several, sometimes unpleasant ways. The explosion of work in the cultural evolution space has, for example, created a confrontation between those who believe we can view culture in terms of quantifiable units – “variants” – and those, often social anthropologists, who do not. Fiery letters and exchanges between prominent academics on both sides highlight the underlying issue: many people reject the possibility that we may understand culture as a set of variants that, like genes, can be modelled changing through time.
Analogous issues appeared during the Covid-19 pandemic. Researchers tracking SARS-CoV-2’s genetic evolution developed complex models that allowed us to determine when new variants of concern arose, and through these techniques we were able, for the first time, to adjust policy as the disease changed.
These models were, however, largely just descriptions of change in the language of genetics. They were not evolutionary models in the sense that they didn’t use Darwinian logic to make and test predictions about how the virus would behave as it changed. In this way, these descriptive genetic trackers implicitly divorced themselves from evolutionary thinking, undoing the century-old marriage between Mendelian genetics and Darwinian understanding known as the modern synthesis.
I have written about the implicit de-synthesis of recent work in the evolutionary sciences elsewhere, and my aim here is merely to suggest that over-emphasis on either quantitative or conceptual thinking is likely to lead, in any field, to substantive shortcomings. As major thinkers found a way to combine quantitative Mendelian thinking with qualitative Darwinian thinking in the 1920s, leading to our modern understanding of biology, we should focus today not only on quantification, but also on critical analysis of the questions we ask.
Social scientists seem aware of the problem on both sides. In the past few years alone, we have seen calls for social scientists to embrace quantification – a laudable and much-needed suggestion as we start to rely, more and more, on large datasets to answer our research questions. But others have said that greater emphasis should be placed on the “philosophy” element of the PhD – the ability, among other things, to critically analyse the questions themselves and to better understand the logic underlying the scientific method.
Even in the corporate space, philosophical thinking is valuable to some: when giving $75 million (£59 million) to the Johns Hopkins philosophy department in 2018, investor William H. Miller III cited the analytical training he received as a graduate student in the department.
Both approaches to research undoubtedly have merit, and the obsession with quantitative analysis, algorithms and machine learning might eventually fade in favour of close attention to language and analytical thinking. But it’s important for us all, as philosophers and scientists, to recognise that we can disagree about meaning just as easily as we can disagree about facts – and that the problems that conceptual and evidential objections might pose to our thinking are equally important.
Many funders and university administrators would agree with all this but assert that the appropriate response is to promote interdisciplinarity. This approach plays to each person’s particular skills while also capitalising on both qualitative and quantitative techniques. Yet despite interdisciplinary thinking being high on the policy agenda in education, there’s little evidence that it’s taken seriously in academic departments.
It would be better, in training future generations of thinkers and researchers, to consciously inculcate a methodological synthesis within individual minds. Failing that – or in addition to it – we should encourage genuine inter- and intra-disciplinary dialogue between those who predict, those who quantify and those who contextualise, in frameworks where all understand each other.
In these ways, we might start to see the tangible manifestation of interdisciplinarity in academic work, rather than mere rhetorical gestures towards it that are belied by the reality.
Jonathan Goodman recently completed a PhD at the Leverhulme Centre for Human Evolutionary Studies, University of Cambridge. He is currently writing a book on cultural evolution in modern society for Yale University Press.
