Do scientists have a miraculous hotline to truth, or are they just as prone to having their assertions skewed by their prior beliefs as everyone else?
Every now and then I stumble upon a paper that's truly revealing about some aspect of the zeitgeist. Joshua Greene's "The Secret Joke of Kant's Soul" is just such an artefact: it alleges to disprove deontology in general (and thus Kantian ethics in general) on the basis of functional magnetic imaging, evolutionary metaphysics, and a handful of psychology experiments. Yet everything about the paper is spectacularly wrong: the neurobiology, the moral philosophy, the metaphysics – every brick in Greene's wall is ill-shaped, and a gentle push is enough to cause it to crumble. Yet the paper has generated considerable secondary literature – Jonathan Haidt cites Greene's results with approbation, as if his argument held water. I begin to suspect an ideological slant motivated this research, that Greene and team set out to 'disprove' Kant based on a prior commitment to Consequentialism and positivism. If this accusation holds water, how did such a shockingly biased paper survive peer review?
Before answering that question, let me review Greene's case, and the presented evidence. In essence, this paper reports fMRI results obtained from subjects in response to various Trolley Problem thought experiments. The two main examples used are: should you flip a switch to reroute a runaway tram to kill one person tied to the tracks instead of five (Phillipa Foot's original Trolley Problem), and the Footbridge variant whereby instead of a switch you have to physically push a fat man from a bridge to stop the tram, killing him in the process. The nub of the new data is that in thought experiments involving direct harm, like Footbridge, Greene claims 'emotion' centres light up, while in impersonal cases he claims 'cognitive' centres light up. Greene alleges that the former involve quintessentially deontological (duty-focussed) decisions, and the latter involve basically consequentialist (outcome-focussed) decisions. Then, on this basis, invents a stunningly spurious evolutionary How-Why Game, of the kind I criticise in The Mythology of Evolution, to explain why our emotion reactions, having allegedly evolved in different contexts to those we face now, can't be trusted and therefore Kantian ethics is false and consequentialism is the One True Ethics.
Trouble is, no part of that argument makes sense or is based on viable research. Picking just a handful of the papers that have offered rebuttal to Greene demonstrates the extent of the problem. Colin Klein demonstrates that the fMRI inferences are badly reasoned, since the weight of research does not favour a dual track model of moral cognition involving 'emotional' and 'cognitive' modules, but a single network involved in imaginative self-projection. Richard Dean is one of several people who have demonstrated that even if the core experiment held up, it wouldn't be evidence against any salient form of deontology. Guy Kahane makes the point that if the evolutionary debunking were plausible, it would undermine the cognitive judgements just as much as the emotional ones! More worryingly, John Mikhail points out that all the thought experiments used, if treated as legal cases, would fail to absolve the actor of blame for the murder they commit.
Indeed, if we were to take Greene's case seriously, we could easily conclude precisely the opposite: the fact that most people would consider flipping the switch but not murdering the fat man shows how easy it is to act immorally when the circumstances are de-personalised. Rather than disproving Kant, Greene's paper perhaps sheds light on why Milgram's famous electroshock experiment was accepted by so many subjects as reasonable. But on this reading, Greene's case does not disprove Kantian ethics, but rather the Consequentialist theories he favours! I do not want to draw that conclusion too strongly – outcome-focussed ideals such as 'the greatest good' are just as valid as their duty-focussed rivals – but it does underline just how flimsy Greene's case is that the same evidence can be used to derive a contrary conclusion.
Ultimately, the sole reason Greene believes he has constructed a case against Kantian ethics is that he has a prior commitment to Consequentialism (which is admitted later in the paper) which allows him to conclude that people are morally right both to flip the switch and to murder the fat man, thus something has gone wrong (Greene believes) when we are ready to murder an innocent by flipping a switch but not by pushing someone off a footbridge. I am far more inclined to use Allen Wood's savage critique of these kinds of thought experiments to expose just how far from morality they take us:
The deceptiveness in trolley problems... consists at least partly in the fact that we are usually deprived of morally relevant facts that we would often have in real life, and often just as significantly, that we are required to stipulate that we are certain about some matters which in real life could never be certain. The result is that we are subtly encouraged to ignore some moral principles (as irrelevant or inoperative, since their applicability has been stipulated away). And in their place, we are incited to invoke (or even invent) quite other principles, and even to regard these principles as morally fundamental, when in real life such principles could seldom come into play, or even if they did, they would never seem to us as compelling as they do in the situation described in the trolley problem... In the process, an important range of considerations that are, should be, and in real life would be absolutely decisive in our moral thinking about these cases in the real world is systematically abstracted out. The philosophical consequences of doing this seem to me utterly disastrous, and to render trolley problems far worse than useless for moral philosophy.
In other words, by stipulating away almost all the real world information necessary to morally interpret the implied situation, Trolly Problems present false dichotomies as if they were the most important part of the situation. If we think rationally about the basic Trolley Problem we are forced to ask: how is it that I am faced with control of the lever, and have perfect knowledge of the outcomes associated? Indeed, the basic problem with Consequentialist ethics, as Nietzche observed, is precisely that we don't know how to accurately predict the outcomes of our actions. The Trolley Problem makes it look as if moral decisions are as simple as five is bigger than two – a logical truth. Precisely what makes real moral dilemmas agonising is that the truly difficult problems for morality do not have such easy answers.
Which brings us back to the question: how did this paper survive peer review? I can only speculate, but on the basis of what Greene was allowed to say, the most plausible answer is that his metaphysical biases – towards consequentialism in ethics, and towards positivism in general – align with a fair proportion of those who were likely to peer review him. The evidence for this claim is the following scholarly clanger Greene plops into his paper smugly:
From a secular humanist's point of view... what is distinctive about religion is its commitment to the existence of supernatural entities... And they are right.
Consider me gob smacked by this racist dogma masquerading as rational fact. You'd think the peer reviewers would at least have picked up on his use of 'they', which surely should read 'we'. Greene cites no-one in connection with this, which is unsurprising since no-one who studies religion could possibly endorse such a muddled view of world religion. Those who aren't sure why this charicature of the Abrahamic faiths fails to characterize religion as a whole are welcome to check Ninian Smart, William James, Joseph Campbell or any one of a host of scholars who, unlike Greene, actually bothered to do their research rather than writing down their beliefs as if they were necessarily factual.
We have in Greene a classic example of prior commitments skewing supposedly neutral research – and, let's not forget, a paper which gets nothing right. Yet there is no outcry over Greene, no hordes of screaming science nerds tearing their shirts in horror at this risible display of metaphysical prejudices. In fact, for all the flag waving about how Religion purportedly distorts Science, nonreligion is far more effective at distortion of the sciences because this wooden horse is already safely inside the gates of Troy. But let us not panic too quickly and completely, or will miss the lesson in Greene's gaffe.
What is revealed by this and other papers is that if we believe in an abstraction called Science that corresponds to truth, the publications of the sciences cannot be part of it. Which means that the work of the sciences isn't Science. This would be shocking if we had failed to recognize that Science (like its rival Religion invented to shadow it) is merely a metaphysical image, a fictional phantom as impossible to prove empirically as God or absolute Truth. It is even tempting to call it supernatural, but only because doing so would offer an amusing mirror image of Greene's misunderstanding of world religion.
Bruno Latour correctly links the mythology of 'Science' – and its equally preposterous villain, 'Religion' – back to Plato's allegory of the cave (which was the very first philosophical artefact I tilted my quixotic lance at, some seven years ago). Via the magic wand of Science, the scientist supposedly flits between the true world and the false social world of culture without any issues at all. Trouble is, scientists actually are human, and there is no perfect world of truth outside of our imaginations (although theists may certainly claim God has one in his back pocket, so to speak). Meanwhile, the sciences are perfectly capable of reading data about the world from a dizzying variety of instruments – as, for that matter, are we all. To extend an example Latour uses, the wine taster's palette can detect things gas chromatography cannot.
In Latour's terms, the problem with Greene is that he is not a reliable witness: he reads his own mind from tools meant to speak for brains (or at least he did in this paper – I am hopeful his later work, which I shall be reading shortly, has improved). The valuable lesson we can learn from Greene's blunderous paper is to be suspicious of research that purports to speak truth instead of reporting the findings of instruments. It is this uncovering of what the different parts of the world can be made to 'say' that constitutes the excellence of the sciences. 'Science', by contrast, offers a mirage of truth on the faith that those who read instrumentation acquire superpowers to discern absolute Truth, like the origin story of a comicbook superhero (ScienceMan!). Greene's paper demonstrates the absurdity of this belief. And that may be it's most valuable contribution to our knowledge.