When I read Dominic Cummings’…err…article/thesis/rant/manifesto/short story (delete as appropriate) this weekend, I was struck by two reactions. One was to yell “Nazi” and head to the hills as images of eugenics programmes flooded my mind. The other was to think “Oooh Odysseus curriculum led by big questions – I like the sound of that.” I really, really want to dislike Cummings, but I suspect that he, like most human beings, is a complex person with some likeable traits. Perhaps he has inherited a 70% unlikeable personality which has been tempered by 30% worth of warm, human nurture. Or the other way round. Either way, shouting “Nazi” and running for the hills will not help this debate one iota – we should always be careful not to confuse genetic science with the horror of eugenics. So let’s look instead at the implications of his statement that intelligence is genetically predetermined.
Cummings’ claim is based on the work of behavioural geneticist Robert Plomin. Now with the kind of respect I argued for in my blog yesterday, let’s explore what exactly it is that is being said here. Plomin’s earlier work in attempting to explore the heritability of intelligence focused very much on IQ as an intelligence measure. There have, over the years, been enough critiques of IQ as a measure of intelligence so I needn’t go into the flawed nature of this interpretation, but at the time, Plomin found through a series of studies on twins separated in different adopted families that there was a roughly 50% heritability factor of this form of ‘intelligence’. In the last decade, his work with other scientists has led to a clarification of this and an engagement with the complexity of measuring intelligence. In a series of papers beginning with the title ‘More than just IQ’, Plomin and others (Chamoro-Premuzic et al 2010) found that self-perceived abilities – i.e. belief in one’s own efficacy and ability to improve was a more significant factor on achievement than cognitive ability was. That is to say that while intelligence may or may not be heritable, a sense or feeling that one was able to succeed was more important. This chimes strongly with Carol Dweck’s work on growth mind sets.
More recently, Plomin and others have published research conducted on 7 and 12 year olds that he claims suggest genetic ‘architecture’ which support intelligence across a slightly wider range of verbal and non verbal tasks (Trzaskowski et al 2013). The tests for the 7 year olds were conducted by telephone and for the 12 year olds on-line. Let’s leave the obvious flaws with that methodology aside for a moment. The summary of the report suggests that there is evidence that genes impact on intelligence across a range of tasks.
The conclusion to be drawn it would seem, is that even if genes affect the architecture of what we have traditionally considered to be intelligence – i.e. verbal reasoning and non verbal problem solving – there is considerable evidence that self-perception is a factor; evidence supported by the study from the London School of Economics this year, that showed that competition and rank ordering in classes damaged the performance of all but those at the top.
Let’s assume also, that Plomin’s educated guess that genes account for 70% of this architecture is true (and it is certainly disputed), so what? Even if 30% of subsequent achievement was accounted for by nurture, that is a significant factor – put crudely in examination terms, the difference between the Russell Group and the dole. Let’s imagine for a moment, that we only have 30% to play with, is this a problem or limitation or an opportunity? Even within this deterministic model, a genetically enabled child would have plenty of room to flunk and a genetically disadvantaged child plenty of opportunity to thrive based on the development of self perception, confidence and crucially, expert teaching.
Moreover, research by Cheety et al (2013) published in the Economist suggests that good teachers make significant value added impact on the achievement of all children throwing into doubt the assertion that teaching has little or no impact on outcomes. To be fair, Cummings’ writing does recognise this complexity, but this caveat becomes lost in a text in which the relationship between the poverty gap, genetics and teacher quality become confused.
Cummings reminds me of a man standing with head in hands, crying in despair that he will never fly because he has no wings while behind him some enterprising brothers build a plane. Let’s not be drawn into this limiting and damaging mode of thought. Architectures are structures from which a dwelling might be built, but they are not the dwelling itself. It takes more than a steel frame to make a home.
Chamorro-Premuzic, T, Harlaar, N, Greven, CU & Plomin, R (2010) , ‘More than just IQ: A longitudinal examination of self-perceived abilities as predictors of academic performance in a large sample of UK twins’ Intelligence, vol 38, no. 4, pp. 385 – 392.
Cheety, R. Friedman, J and Rockoff, J “Measuring the impacts of teachers II: teacher value-added and student outcomes in adulthood“, NBER working paper 19424, September 2013.