4. How
reliable is moral cognition? Does evolution debunk morality?
Evolution does not “debunk morality” (a stupid and annoying phrase). Hume
gave the main arguments debunking strong (non-teleology-based) moral
objectivism long before Darwin. Evolution put the final nail in the coffin of
teleological metaphysics, which further reinforced the importance of Hume’s
arguments, and 20th Century work in evolutionary biology, ethology,
evolutionary psychology and (evolutionary) neuroscience has helped bring out
the important implications of the fact that our ‘moral faculty’ is at least grounded in a limbic system and
norm-generator that have their specific form only because this form has been
selected for. Nonetheless, it would be wrong to claim that these discoveries
have had earth-shattering conclusions for meta-ethics: they have not shown that
a Mackean Error Theory or some form of expressivism is the correct account of
human moral language, because they have not shown that it is impossible to
reason about ethics. I argue that, in roughly
the same way that we accept that knowledge of evolution does not “debunk”
mathematics, we ought to accept that knowledge of evolution does not “debunk”
the grand philosophical tradition of reasoning about ethics, from Plato to
Parfit (or show that we should give up debating ethical dilemmas).[1]
Long before Darwin, David Hume in A
Treatise of Human Nature famously argued: a) that one can’t deduce a
prescription from any number of descriptive statements, b) that beliefs are not
in themselves motivating and c) that it is a category error to think that
desires or preferences are rational or irrational in themselves. The first is a
straightforward logical truth,[2]
with (of course) extremely important consequences for very naïve views about
ethics. Nevertheless, the observation of this logical truth was not fatal to
full-blooded moral objectivism for scientific rationalists before Darwin,
because a pre-Darwinian scientific rationalist could still rationally believe
in some kind of teleological metaphysics.
In fact, most pre-Darwinian Enlightenment figures did believe in some kind of
teleological metaphysics (either taking the form of Christianity (like Newton,
Locke or Kant) or some kind of Deism (like Spinoza or Voltaire)). Galileo and
Newton had obviously shown that Aristotle’s teleological physics was hopelessly
flawed, but their theories hadn’t shown that there could not be purpose in
creation. Newton certainly didn’t think he had shown that!
However, as philosophers have been pointing out for
more than a century, with Darwin’s theory of evolution came the deathblow to
the ancient Aristotelian worldview that design was inherent in nature (that,
e.g., the undeniable telos of Man,
“the rational animal”, is to be a zoon
politikon who fully exercises his powers of reason and attains eudaimonia). This is because Darwin
showed that all appearance of design in nature was just that: teleonomy, not teleology, because all life evolved from completely unconscious
processes over very long stretches of time.
This conclusion destroys any claim for the objectivity of teleological
ethics, thereby rendering unjustifiable arguments from “the natural” for or
against actions or practices.[3]
Without any contribution from an understanding of
evolution, b) and c) do further violence, if true, to naïve ethical views,
because they jointly imply the conclusions i) that even if there were absolutely ‘true’ moral judgments which all
sufficiently intelligent species would eventually converge on, an omniscient
(and logically omniscient) organism could recognise these ‘true’ judgments and
act monstrously nevertheless, because of a lack of motivation to do good, and
ii) that it would be a category error to think that this lack of motivation
would be in itself rational or irrational. Although the most extreme Humean
view on rationality of preferences certainly runs into trouble (because having
an intransitive preference set seems clearly irrational and because, as Derek
Parfit showed [1984, 2011], one can at least come up with niche examples of
arbitrary, self-defeating preferences that seem clearly irrational by any
standard), a slightly moderated version seems basically impossible to deny.
Pertinently, Darwin’s theory gave this line of thought a further power, by bringing
out the massive historical contingency of
our feelings towards con-specifics. In The
Descent of Man, Darwin himself
illustrates this point:
“In the same manner as various animals have some sense
of beauty, though they admire widely different objects, so they might have a
sense of right and wrong, though led by it to follow widely different lines of
conduct. If […] men were reared under precisely the same conditions as
hive-bees, there can hardly be a doubt that our unmarried females would, like
the worker bees, think it a sacred duty to kill their brothers, and mothers
would strive to kill their fertile daughters.” [Darwin, 1981/1871: 73]
The basic thought here is that different species have
different limbic systems and different forms of social cognition (mainly(?))
because of different historical trajectories, and that there’s no basis for the
intuition that one kind of limbic system or form of social cognition produces
the true judgments (whereas there is
an obvious basis for thinking that having a bigger neocortex may allow for
better reasoning, etc).
Since the middle of the 20th Century, a
huge amount of progress has been made in the fields of theoretical evolutionary
biology, ethology (particularly primatology), social psychology and
neuroscience towards understanding the evolutionary ‘roots’ of our moral
faculty and the way people commonly make ethical judgments (the role of
reason). Some have claimed that some of this research has dramatic meta-ethical
conclusions. I disagree. In the short space I have left, I will do my best to
show that this is the wrong inference.
In the introduction, my words suggested that I was
mostly concerned with those who have used Darwinian research to argue either
for an Error Theory about ethics or some form of expressivism, but there’s
actually another, more interesting group who have used this research for a
strong meta-ethical conclusion: utilitarians like Peter Singer, Joshua Greene
and (possibly) Steven Pinker who think that theoretical evolutionary biology
and psychology show a) that we rely on gut intuitions which we rationalise constantly, but that this is
specifically a failure to think in consequentialist terms, b) that a rational
person ought to become a utilitarian, and c) that ‘reason’ is the main engine of
moral progress, where moral progress is basically identified with “expanding
the moral circle”. This argument is expounded in greatest detail by Greene in
his 2013 book Moral Tribes, although
Singer’s well-cited 2005 paper “Ethics and Intuitions” instantiates this
argument as well. Quite apart from the simplistic science that is often
marshalled in service of this argument (see my first assignment on “Ethics and
Intuitions”)[4],
it is a category mistake, as Massimo Pigliucci [2012] and Thomas Nagel [2013] argue,
to think that empirical research could in itself strongly favour one moral
theory over another. Of course, Singer and Greene realise that they also need a
philosophical argument to support their interpretation of this evidence –
however, like Pigliucci and Nagel, I believe that their arguments fall short of
establishing their very strong conclusion that utilitarianism is the ‘true
theory’, or the one sensible choice for a rationalist.
As Nagel points out in his 2013 review of Moral Tribes, “You can’t learn about
morality from brain science”, there appears to be a contradiction in Greene’s
position. On the one hand, Greene does not want to say that utilitarianism is the true theory, i.e. taking a strong
realist stance, (as Singer has in recent times been strongly tempted towards
[2016]) and yet his advocacy of a sub-Millian utilitarianism with principles
that are deeply counterintuitive and unfollowable – which even Greene admits he
doesn’t come close to following – strongly suggests the view, as Nagel puts it,
that morality is something “more […] than human”. Why a non-realist would hold
such a view is inexplicable.
Moreover, as Nagel suggests, it is strongly contrary to available evidence to
hold that deontic principles can’t form the basis for a universal human ethic,
as we can see in the phenomenal international success of ‘human rights’ since
World War II [Pinker, 2011 (funnily enough)]. Like Singer, therefore, Greene
does not have the grounds he thinks he has to repudiate a more Rawlsian
approach to ethical reasoning, without the extreme ambition of de-humanising
ethics.
More generally, I think that one thing that unites
this utilitarian group of Greene et al, and
the as-yet-undiscussed group who use Darwinian research to defend some form of
strong anti-rationalism (represented by Haidt) is that they infer from evidence
that a massive amount of “rationalisation” goes on in the production of moral
judgments by non-philosophers [Haidt, 2012] that major philosophical debates
must boil down merely to a clash between very elaborate competing
rationalisations. I myself believe that Haidt’s research casts serious doubt on
some of the basic assumptions that motivated many great philosophers (e.g.
Kant), but I also think there is much of importance that it doesn’t cast doubt
on. It doesn’t cast doubt on the fact that human beings can’t change their mind
when exposed to moral reasons (in the Rawlsian sense), nor that an astute
philosopher can’t tell when a person’s commitment to abstract principle x is in conflict with their position on
specific issue y.
Moreover, as Pigliucci points out [2012], one needn’t
believe in the ‘truth’ of one moral theory to believe in the possibility of
making progress with ethical reasoning. One can, like Pigliucci and I, believe
that different frameworks can be applied to different problems. In short, it is
not inconsistent with Hume, Darwin or modern psychological and neuroscientific
research to advocate a weak form of moral rationalism, which would be committed
to the view that certain points of consensus among modern philosophers have
much to do with reason (that extreme speciesism is a mistake, that racism and
sexism are wrong), but not committed to the view that all issues or dilemmas
have a determinate, superassertible verdict
which all rational people must accept, or that there is one ultimate theory
which all rational people should ultimately converge towards, contra Derek
Parfit [2011].
Reference
List
Aitken, Thomas
(2017). “First Assignment for How Biology Matters to Philosophy: Report”.
Accessible (if need be) from:
<http://writingsoftclaitken.blogspot.com.au/2017/04/a-brief-philosophy-of-biology.html>
Darwin, Charles
(1981/1871). The Descent of Man,
Princeton University Press, New Jersey.
Greene, Joshua
(2013). Moral Tribes, Penguin Press.
Haidt, Jonathan
(2012). The Righteous Mind: Why Good
People are Divided by Politics and Religion, Pantheon Press.
Hume, David
(1737). A Treatise of Human Nature, 1992
edition, Prometheus Books, New York.
Nagel, Thomas
(2013). “You can’t learn about morality from brain scans”, New Republic.
Accessible from:
<https://newrepublic.com/article/115279/joshua-greenes-moral-tribes-reviewed-thomas-nagel>
Parfit, Derek
(2011). On What Matters, Oxford
University Press.
Pigliucci,
Massimo (2012). Answers for Aristotle, Basic
Books, New York.
Priest, Graham
(1997). “Sexual perversion”, Australasian
Journal of Philosophy, 75:3, 360-372.
Turchin, Peter
(2005). War and Peace and War, Penguin
US.
Turchin, Peter
(2015). Ultrasociety: How 10,000 Years of
War Made Humans the Best Co-operators on Earth, Beresta Books.
[1]
Knowledge of evolution can help us see that some of the great moral
philosophers were more seriously mistaken about various matters than others,
with the teleological philosophers (like Plato and his fellow ancients) the
most fundamentally mistaken about what can reasonably be said to justify moral
claims. The view I will defend is simply that ethical reasoning is not always
faux-reasoning or rationalising in
the sense that the best ethical reasoners in the philosophical tradition really
are deriving conclusions directly from the axioms (general ethical principles)
they set up (the constraint on possible axioms is just that any axiom has to
seem very appealing to (at least most of) us).
[2]
Easy to overlook, of course, because so much of our language is normatively
infused, as Hilary Putnam argued (I don’t have time to argue that it would be
wrong to think that the linguistic ‘entanglement’ shows that Hume’s
‘guillotine’ is unjustified).
[3] A
brief contemporary discussion of this point can be found in Graham Priest’s
1997 paper Sexual Perversion, which
points out that to a Darwinian and moral rationalist moral objections using
language like “perversion” and “unnatural” carry no weight: if by natural you
mean historically adaptive and selected for, then the objection doesn’t go
through because a behaviour’s historical adaptiveness has no logical bearing on
its rightness or wrongness (this is more or less Hume or G.E. Moore re-hashed).
[4] Pinker, perhaps Singer’s main influence and very much a
proponent of the gene-centric theory of evolution, seems unaware of the
evidence that Hobbes’ description of ‘human nature’ matches only about 25% of
the population in experiments of co-operation games (including outside the
West) [Turchin 2005], generally overlooks the massive historical and
sociological evidence that people can be highly self-sacrificing for what Atran
calls “sacred values” [2001] (though Pinker in The Better Angels of our Nature does cite Atran in relation to
Israel-Palestine), and seems unaware of the case for multilevel group selection
as an explanation of human ‘ultrasociality’, as articulated, for example, in
Turchin [2005, 2015].
