Why we need non-scientific solutions to moral problems

Or what genetics can't teach you about personal responsibility

By Joachim Skanderby Johansen

Last year, Kathryn Paige Harden published The Genetic Lottery where she argues that we should pay attention to genetics’ powerful influence upon our lives when trying to build a just society. As part of this argument, Harden proposes that we are only personally responsible for the behaviour that cannot be explained by genetics or social factors.

While Harden states that her take on personal responsibility attempts to avoid murky metaphysical discussions, her argument leads straight into the jaws of a classical will-free paradox – showing that those who do not engage with the history of philosophy are bound to repeat its follies.

The genetic lottery and absolution by luck

Harden's arguments in “The Genetic Lottery” can be summarised in two central claims; in the first half of the book, Harden argues that genes are causes of important life outcomes such as level of education and financial success. In the second half of the book, which is the subject of this post, Harden asks what political implications it ought to have that genes are important drivers for how our lives turn out. Drawing on the framework of all-star philosopher John Rawls, she argues that since genes are random causes of inequality among people, we should seek to remedy genetic inequalities for the same reasons we should remedy inequalities stemming from social class. Harden argues that this has implications for our notion of personal responsibility – she ponders:

“If you take genetics seriously as a source of difference between people, then what does that mean (if anything) about the responsibility that we bear for how our life turns out?”

Before seeking to answer this question, Harden states that she wishes to avoid “falling down a black hole of never-ending metaphysical debate” and starts her line of argument by “putting aside metaphysical questions…” This will return to haunt her later, as we will see.

Harden argues that we are only accountable for our choices. As an example, we hold people responsible for murder and not eye-colour because murder is seen as the outcome of choice, whereas eye-colour is not, writes Harden. According to her, the question of whether we could have done things differently is essential to whether we have a choice or not and, therefore, whether we are accountable.

Harden then proposes that we measure the degree to which we have choices in our lives with the so-called free will coefficient, assuming that the question of choice must be equal to the question of whether we have free will. The free will coefficient is defined as differences in life outcomes (e.g., educational attainment) that are not driven by differences in genes or social circumstance. This is measured by studying differences in the life outcomes of identical twins who are raised together because these are exposed to similar genetic and social causes in their lives. These differences that Harden propose as a measure of free will, are also called the non-shared environment and abbreviated e2, Harden explains. The e2 value can range from 0 to 1, where 0 means that identical twins raised together are perfectly alike in their outcomes and where 1 means that they are no more alike than other people picked at random. Measured by e2, free will therefore becomes the behaviour that remains unexplained after taking social and genetic factors into account.

Harden then proceeds to describe that studies have found low e2 values for outcomes like IQ, general cognitive abilities and educational attainment. From this, Harden concludes that:

”Once we account for the powerful effects of luck – both environment and genetic, together – there is remarkably little territory left for “personal responsibility.”

Bottom-line, Harden argues that our social circumstances and genetic make-up are a product of luck, and that the degree to which we are personally responsible must depend on whether our behaviour can be explained by these factors. Finding that genetics and social factors explain the lion’s share of differences in life outcomes, Harden concludes that we are near-absolved from, or deprived of, personal responsibility.

The problems of Harden’s account of personal responsibility

Harden’s claim raises two questions: first, whether e2 is a good measure of free will and personal responsibility, given Harden’s own examples. Harden describes that e2 for height is scarcely greater than e2 for educational attainment in Scandinavian cohorts. Harden uses this as an argument against personal responsibility for educational attainment, but by her logic the fact that e2 is above 0 indicates that we are, at least to some small degree, personably responsible for our height. In the same vein, Harden mentions that identical twins may differ in outcomes such as whether or not they suffer from schizophrenia. Supposedly, this difference would lead to a higher e2 value, but it hardly seems fair to hold people responsible for suffering from schizophrenia, even to a small degree.

Secondly, and more fundamentally, Harden’s account raises the question of whether we can simply measure free will and personal responsibility on a scale, while bravely ignoring all philosophical question about free will. I will argue that we cannot.

To see this, let’s start by exploring the question: What could be the causes behind the behaviour that is unexplained by social and genetic circumstance and quantified by e2? In the first instance, it might be tempting to simply answer that personal choice or free will is the cause behind that behaviour. But what is then the cause of this personal choice or that particular will? Here, it seems we end up with three equally dissatisfying answers.

First, the cause behind the behaviour might be the environmental factors that are unique for an individual, even for the identical twins in Harden’s studies that share genetic and social backgrounds. These could be differences caused by the twins being in different classes in their schools, doing different sports etc. This would seem like an apparent explanation as e2 is exactly an abbreviation of the non-shared environment. The problem is that under this interpretation, the e2 appears unfitting as a measure of personal responsibility or free will. It is hard to see why the twins in the studies should be held more accountable for their actions if they are caused by the non-shared environment than by the shared environment; what classroom one ends up in seems just as governed by luck as social class.

Secondly, e2 might measure the influence of true randomness in our lives – that is, a part of our lives that is fundamentally up to chance and therefore not explainable by social and genetic factors. However, once again this makes e2 an unfitting measure of personal responsibility. It is difficult to argue that the twins in the studies should be held responsible for the part of their behaviour that is caused by pure chance. If it is pure chance that one twin gets into a prestigious university and the other does not, it does not seem fair for there to be pride in the latter or shame in the former. Actually, Harden’s very argument for not holding people responsible for behaviour caused by genes or social factors is that these are products of luck!

Thirdly, the cause of the behaviour that Harden wishes to measure as the free will coefficient might be – well – its own cause and effect, not stemming from a social or environmental driver. While this option might seem attractive, it is worth noting just how strange a phenomena e2 is supposed to measure under this interpretation. Our free will is here supposed to have causal properties in the sense of being able to cause behaviour but must itself be uncaused by other factors (otherwise we just end back at the first interpretation, were e2 measures effects of the non-shared environment). The notion of the mind as its own cause, which can shape our behaviour but is not influenced by other factors, has the reminiscence of theories of the mind from philosophers like Rene Descartes (1596-1650). Under this interpretation, e2 would indeed measure free will and a high e2 coefficient does seem to imply personal responsibility. However, such an account of free will is rare today exactly because it turns the will and mind into metaphysical entities that seem at odds with the cause-and-effect lens we use to understand the rest of the world with. Quoting her old PhD advisor, Eric Turkheimer, Harden appears to want to avoid exactly such metaphysics by using the e2 measure:

“The non-shared environment is, in a phrase, free will. Not the sort of metaphysical free will that no one believes in anymore, according to which human souls float free above the mechanic constructs of the physical world, but an embodied free will”

Harden’s argument ends in a dilemma: We can either understand free will, her pre-requisite for personal responsibility, according to the first or second interpretation above, in which case free will seems compatible with a naturalist worldview, but does not provide grounds for personal responsibility, or we can understand free will according to the third interpretation where free will does provide grounds for personal responsibility, but becomes the kind of metaphysical concept Harden seems keen to avoid.

Why the black hole of metaphysical discussion matters

It is notable that this problem around the paradoxical nature of free will is not new in philosophy at all. The no-nonsense Scottish Enlightenment philosopher David Hume (1711-1776) laid it out in a more general version in A Treatise of Human Nature. But because Harden does not wish to be drawn into “... a black hole of never-ending metaphysical debate”, her account ignores, rather than faces, these philosophical discussions head-on. This is a shame because there actually are accounts of free will and personal responsibility that are compatible with determinism but are not in danger of re-introducing metaphysics 500 years past their expiration date. Drawing on the aforementioned Hume, one can for example argue that an act of free will is simply one that stems from our desires or preferences, and which is not hindered or forced by some external force. Under this interpretation, my act of going to a café and writing is an act of free will, even if it can be predicted by social and genetic factors.

These ponderings are not only relevant to the philosophically pedant. If we subscribe to the Hume-style version of free will, Harden’s argument for highly limited personal responsibility crumbles. If free will is the pre-requisite for personal responsibility but acts of free will can be determined by social or genetic factors (again, such as my writing at a café), then the argument for not holding people responsible for acts explainable by social or genetic factors disappears. This shows that taking a philosophical stand is necessary to think about moral questions such as personal responsibility.

While it is understandable that Harden wishes to avoid murky metaphysical problems, her substitution of philosophical discussion with a scientific solution also means that her account is remarkably free from discussions of the moral implications of highly limited personal responsibility. Beyond being problematic on a philosophical level as pointed out, this is also politically perilous and can be seen as part of a larger trend where questions of right and wrong are warped into technical discussions.

The dangers of smart solutions

A criterion often used in modern philosophy is that the implications of a moral theory – like when we should hold someone accountable – should not be in violent opposition to our basic moral intuitions; if a moral theory suggests that it is morally praiseworthy to torture baby seals or set fire to kindergartens, we should take that as an argument against that moral theory. While testing theories by using our “philosophical gut” might seem unappealing to people shaped by the hard sciences, the method has the advantage that it is fundamentally democratic; the basic tools for discussing ethics are distributed to all of us. The dangers of disregarding such “moral weighing” are similar to those put forward by Michael Sandel in his fantastic book “The Tyranny of Merit.” Here, Sandel argues against the politics of just doing “what’s smart.” Sandel explains how Obama-era politics were driven by the idea that one could avoid pesky just-vs.-unjust- and right-vs.-wrong-discussions (maybe Obama saw these as never-ending too) and simply do “what’s smart”. Doing “what’s smart” turned out not to be some universal common good but rather what fitted the value system of some highly educated Harvard types. This did not unite Republicans and Democrats in a “politics beyond politics”, but rather left the larger population alienated as politics became a technical discipline; another branch of engineering taught at the Ivy League, which the uninitiated did not have access to.

Harden’s account suffers from similar problems; by trying to side-step philosophical considerations with quick scientific solutions, she not only ends in the philosophically paradoxical, but with a fully technical definition of personal responsibility without discussing its moral implications – and there are plenty to discuss: Does the assertion that “there is remarkably little tension left for personal responsibility” mean that we should not hold Putin responsible for the invasion of Ukraine, for example? That big businesses should not be held accountable? Or, reflecting on the positive side of responsibility, that we are not really responsible for the achievements in life that we feel most proud of?

Such discussions are essential for a society to engage in – and Harden’s standpoint is undoubtable a valuable contribution to them - but they will suffocate if we turn them into discussions of e2 values, alien to our shared moral intuitions.

Joachim Skanderby Johansen is a regular writer at Unreasonable Doubt. He writes on the ethics and practicalities of responsibility and uncertainty. He occasionally defends dead liberal ideas. Joachim works in the financial sector. He has a master’s degree from the London School of Economics and Copenhagen Business School.