Moving beyond dichotomies: Liao, S. Matthew (ed.), Moral Brains: The Neuroscience of Morality, Oxford University Press, 2016

Abstract

Matthew Liao's edited collection Moral Brains: The Neuroscience of Morality covers a wide range of issues in moral psychology. The collection should be of interest to philosophers, psychologist, and neuroscientists alike, particularly those interested in the relation between these disciplines. I give an overview of the content and major themes of the volume and draw some important lessons about the connection between moral neuroscience and normative ethics. In particular, I argue that moving beyond some of the dichotomies implicit in some of the debates advanced in the book makes the neuroscience of moral judgments much more useful in advancing normative ethics. Recent theoretical and empirical research in the psychological sciences has significantly advanced our understanding of moral thinking. In this changing landscape, Moral Brains: the neuroscience of morality does a great job at featuring leading researchers in moral cognition from a wide range of disciplines and summarizing the last two decades or so of scientific and philosophical discussion in moral cognition. This is an especially appealing book for researchers working on moral judgments, emotions and reasoning, moral decision-making and epistemology, personality disorders associated with impaired moral judgment, and the neuromodulation of moral thinking. Researchers with a broad interest in moral psychology, philosophy of 2 psychology, normative ethics, and metaethics will also find the book valuable. The volume supplies a collection of readings in moral psychology and neuroscience that works well as an introduction for advanced undergraduates and graduate students, as well as a stimulating reading

Full text

REVIEW ESSAY
Moving beyond dichotomies: Liao, S. Matthew (ed.),
Moral Brains: The Neuroscience of Morality, Oxford
University Press, 2016
Ivan Gonzalez-Cabrera
1,2
Received: 8 May 2017 / Accepted: 4 September 2017
Springer Science+Business Media B.V. 2017
Abstract Matthew Liao’s edited collection Moral Brains: The Neuroscience of
Morality covers a wide range of issues in moral psychology. The collection should
be of interest to philosophers, psychologist, and neuroscientists alike, particularly
those interested in the relation between these disciplines. I give an overview of the
content and major themes of the volume and draw some important lessons about the
connection between moral neuroscience and normative ethics. In particular, I argue
that moving beyond some of the dichotomies implicit in some of the debates
advanced in the book makes the neuroscience of moral judgment much more useful
in advancing normative ethics.
Keywords Dual-process theory Moral judgments Moral neuroscience Moral
psychology Moral reasoning Normative ethics
Recent theoretical and empirical research in the psychological sciences has
significantly advanced our understanding of moral thinking. In this changing
landscape, Moral Brains: The Neuroscience of Morality does a great job at featuring
leading researchers in moral cognition from a wide range of disciplines and
summarizing the last two decades or so of scientific and philosophical discussion in
moral cognition. This is an especially appealing book for researchers working on
moral judgments, emotions and reasoning, moral decision-making and epistemol-
ogy, personality disorders associated with impaired moral judgment, and the
neuromodulation of moral thinking. Researchers with a broad interest in moral
psychology, philosophy of psychology, normative ethics, and metaethics will also
&Ivan Gonzalez-Cabrera
ivan.gonzalez-cabrera@anu.edu.au
1
School of Philosophy, RSSS, Australian National University, Fellows Road, Room 3229, ANU,
Canberra, ACT 2601, Australia
2
Konrad Lorenz Institute for Evolution and Cognition Research, Klosterneuburg, Austria
123
Biol Philos
DOI 10.1007/s10539-017-9590-2

find the book valuable. The volume supplies a collection of readings in moral
psychology and neuroscience that works well as an introduction for advanced
undergraduates and graduate students, as well as a stimulating reading for
experienced researchers interested in the connection between moral psychology
and the life sciences broadly construed.
Like most collections, Moral Brains explores a wide range of topics without a
clear unifying theme. However, a recurrent thread running through the book
concerns the relation between the neuroscience of moral judgment and normative
ethics, especially the implications of the former for the latter. In what follows, I will
give an overview of the content and major themes of the volume and draw some
important lessons from it. I will argue that moving beyond some of the implicit
dichotomies that permeate some of the chapters of the book may help research in
moral neuroscience to advance normative ethics.
Overview of the chapters
The book begins with a helpful overview by Matthew Liao (editor and contributor to
this volume) of the main issues discussed in the fourteen articles featured in the
book. The chapters are divided into four parts, which are quite variable in focus,
content, and methodology. The first part focuses on the role of emotions and
reasoning in moral judgments, and how these different aspects of cognition can be
eventually integrated into human moral thought. The second part discusses the
reliability of deontological versus consequentialist judgments, focusing on the most
recent version of Joshua Greene’s dual-process theory of moral judgment. The third
part presents new findings and methods on the neuroscience of moral judgment,
emphasizing the importance of clinical, pharmacological, and model organisms in
the study of moral cognition. The fourth part deals with fundamental theoretical
issues that overlap with many of the debates addressed in the book.
In the first part of the volume, on the role of emotions and reasoning in moral
judgments, Jesse Prinz claims that there are philosophical reasons and behavioral
evidence to support a version of moral sentimentalism according to which emotions
are a constitutive part of sincere moral judgments. Since the model is considered
compatible with most of the empirical evidence in moral neuroscience, Prinz argues
that this version of moral sentimentalism can help us to understand how different
brain structures contribute to moral cognition. Jeanette Kennett and Philip Gerrans
do not necessarily disagree with Prinz’s constitutive model of moral judgments but
they advocate in their contribution for a much broader role of reasoning in moral
deliberation. Their model contrasts with that of moral intuitionist in which moral
judgments are the result of tacit affective processes that are partially encapsulated
from explicit reasoning (Haidt 2012; Haidt and Bjorklund 2008). In Kennett and
Gerrans’ model, reasoning plays a major role in moral decision-making once the
diachronic aspects of human agency are taken into consideration—e.g., the fact that
an agent has to resolve long-term conflicts between opposing intuitive moral
responses or deal with conflicting moral responses from different agents.
I. Gonzalez-Cabrera
123

Perhaps the most compelling contribution in this section is that by James
Woodward who explicitly rejects a sharp distinction between human cognitive and
affective pathways to moral judgment. According to Woodward, areas commonly
identified as involved in emotional processing contribute causally to the construc-
tion of moral judgments in neurotypical subjects. Brain areas such as the
ventromedial prefrontal cortex (vmPFC), orbitofrontal cortex (OFC), anterior
cingulate cortex, insula, amygdala, and the ventral striatum are involved in
emotional processing. What these brain regions do is computing values associated
with (positive and negative) reinforcers and the actions undertaken to provide those
reinforcers. The computation of these values is essential for all kinds of decision-
making because, otherwise, agents’ choices would not be motivating. Among these
structures, the vmPFC and the OFC integrate reward signals from different stimuli
and representations from cognitive systems (Rolls 2005). Since empirical evidence
shows that moral judgments in neurotypical subjects are often causally influenced
by value signals in the vmPFC and the OFC (Greene et al. 2004; Shenhav and
Greene 2010), areas commonly identified as involved in emotional processing
would play a central role in moral judgment in neurotypical subjects regardless of
whether moral judgments and choices are in fact supported by reasoning or effortful
thinking. Moreover, this would mean that we cannot make a sharp distinction
between cognitive and affective pathways to moral judgments in neurotypical
subjects and that the moral judgments of this population would be ‘sincere’ in the
sense of being intrinsically motivating.
The second part is arguably the heart of the book. A great deal of it is dedicated
to discussing different aspects of Greene’s dual-process theory of moral judgment
and the second part specifically focuses on his most recent formulation (Greene
2014), which is reprinted in this volume. In this contribution, Greene proposes two
routes through which neuroscience research could have implications for normative
ethics. In the direct route, independent normative assumptions are combined with
neuroscientific research about the factors that our moral judgments are sensitive to.
In the indirect route, neuroscientific research identifies the conditions under which
automatic and effortful moral judgments are more cognitively efficient.
Greene’s central argument focuses on the indirect route. He argues that current
neuroscientific research favors a certain form of consequentialism. Drawing upon an
analogy with digital SLR cameras, Greene maintains that human moral cognition
operates in two complementary modes: a set of automatic settings and an effortful,
general-purpose reasoning mode. According to him, we should not rely on our
automatic moral settings when attempting to resolve moral problems with which we
have inadequate evolutionary, cultural, or personal experience (or as Greene calls
them ‘unfamiliar’ moral problems) since it would be a cognitive miracle if we turn
out to have reliable good moral instincts under these conditions (p. 131). So effortful
thinking is best suited for dealing with this class of moral problems. This has
important consequences for normative ethics, for Greene maintains that automatic
emotional responses typically support characteristically deontological judgments,
while processes of effortful thinking typically support characteristically conse-
quentialist ones. As defined in this chapter, the former are judgments that are
naturally justified by appeals to rights, duties, and so on, whereas the latter are those
Moving beyond dichotomies: Liao, S. Matthew (ed.), Moral…
123

that are naturally justified in terms of cost–benefit reasoning (p. 122). As a result,
Greene concludes that characteristically consequentialist judgments are best suited
for dealing with moral problems with which we have inadequate evolutionary,
cultural, or personal experience.
This part of the book includes two comments by Julia Driver and Stephen
Darwall as well as a reply to these comments by Greene. Driver is a long-standing
advocate of consequentialist views in moral philosophy. She argues that
consequentialist and deontological moral theories are in general immune to
Greene’s argument because the debate about whether moral deliberation is more
reliable in consequentialist than deontological terms requires assuming a back-
ground moral theory that allows us to determine whether subjects’ responses are
morally correct or not. Darwall, in contrast, is a well-known expert on deontological
approaches to ethics. He points out that arguing that consequentialist moral theories
are more reliable than deontological ones implies claiming that consequentialism is
a better theory of moral right.
Darwall’s argument requires some attention. He begins emphasizing that Greene
aims to support a particular form of consequentialism, namely act-consequentialism.
According to act-consequentialists, an action is morally right if and only if that
action yields the best available consequences, regardless of whether it would be best
for us to be disposed to act upon non-consequentialist moral intuitions in order to
bring about those consequences. This makes act-consequentialism an ‘esoteric’
moral theory (Williams 1995, p. 165). For example, people could not be reasonably
held accountable for acting upon those moral intuitions which best dispose them to
bring about these consequences, even in situations in which those actions actually
do not meet the act-consequentialist standards of a morally right action. This makes
the notions of moral right and moral accountability conceptually independent of
each other. But Darwall argues that the notions of moral right and wrong are tied
conceptually to the idea of moral responsibility or accountability in the sense that if
an action is wrong, then it is of a kind that is blameworthy unless the agent has an
excuse (p. 167). Thus, he claims that on conceptual grounds, there are superior
theories of moral right, including some versions of rule-consequentialism. A better
account, for instance, would be one in which an agent is obligated to perform
actions of which it is true that the general acceptance of a rule requiring those
actions would have better consequences than would the general acceptance of any
other rule in similar circumstances. Darwall himself does not endorse this form of
consequentialism. Yet, he exemplifies with it his key conceptual claim without
making non-consequentialist assumptions. For unlike act-consequentialism, this
form of rule-consequentialism would make the notion of moral right conceptually
tied to that of accountability. Since such a version of rule-consequentialism does
lead to characteristically deontological judgments, we should not conclude that
characteristically consequentialist judgments are more reliable than characteristi-
cally deontological ones.
In response to Driver, Greene’s reply proceeds in terms of both the direct and the
indirect route. I will focus only on the latter since Driver main argument focuses on
the role of background moral theorizing in Greene’s argument but the direct route
relies on independent normative assumptions to reach substantive moral
I. Gonzalez-Cabrera
123

conclusions. So Greene’s argument seems to be stronger when framed in terms of
the indirect route than when framed in terms of direct route. According to the
former, we should not rely on our automatic moral settings when attempting to
resolve moral problems with which we have inadequate evolutionary, cultural, or
personal experience because this would amount to expecting a cognitive miracle.
Greene argues that no additional normative premise is required to support this claim
since such a claim is true regardless of the standard we apply for determining
reliability. To illustrate this, he considers the case of novice drivers who lack
personal experience behind the wheel since it would be a cognitive miracle if they
succeed in their first attempt at driving a car. Finally, Greene clarifies that he
understands consequentialism not only as a decision procedure for unfamiliar moral
problems but as a higher-order ‘metamoral’ standard, i.e., a normative standard that
adjudicates among competing tribal values and interests (p. 175). Thus, he thinks
that there is a standard for everyday cases and a standard for hard cases that is the
same, even though the decision procedure changes depending on the nature of the
decision problem.
In response to Darwall, Greene agrees with Darwall that act-consequentialism is
unfit for directly guiding everyday moral behavior, but he denies that this entails
that act-consequentialism is ‘interpersonally’ esoteric since people do have access to
the foundational moral standards upheld by act-consequentialists. Moreover, Greene
argues that since consequentialism is only a good normative guide for dealing with
difficult moral problems, his argument does not entail the complete rejection of
characteristically deontological judgments. Therefore, it is not a problem if the
dictates of rule-consequentialism are characteristically deontological since the kind
of metamoral theory he defends would accommodate both characteristically
consequentialist judgments and characteristically deontological judgments.
Overall, Greene’s responses do not fully address the objections raised by Driver
and Darwall. Assuming that we have inadequate evolutionary, cultural, or personal
experience to solve a moral problem is in itself a moral assumption, which means
that his main argument proceeds through Greene’s direct, rather than indirect, route,
and therefore it requires some background moral theorizing. Take the case of
driving a car. According to Greene, this example only works within the range of
plausible conceptions of good driving: ‘‘Of course, if by driving ‘‘well’’ you mean
crashing immediately into a tree, then all bets are off. But within the range of
plausible conceptions of good driving, we can say with confidence that new drivers
cannot drive well based on automatic responses (intuition) and must instead rely on
explicit, controlled decision-making’’ (p. 173). By parity of reasoning, this would
hold true in the moral case only within the range of plausible conceptions of what
making good moral judgments is. The problem would then be that determining this
set of plausible conceptions seems to require moral theorizing. Even assuming that
automatic moral responses are unreliable in situations in which we have inadequate
evolutionary, cultural, or personal experience, we still require background moral
theorizing. For we should not rely on our automatic moral settings only if (or to the
extent that) we have inadequate evolutionary, cultural, or personal experience to
solve a moral problem. However, assuming that we have inadequate evolutionary,
cultural, or personal experience to solve a moral problem is a moral assumption.
Moving beyond dichotomies: Liao, S. Matthew (ed.), Moral…
123

Since we need background moral theorizing to determine when (or to what extent)
we have inadequate evolutionary, cultural, or personal moral experience, then the
argument would still require background moral theorizing to support the
assumption.
Greene also seems to overlook Darwall’s key conceptual claim about the relation
between the notions of moral right and moral accountability. If act-consequential-
ism is unfit for directly guiding everyday moral behavior as he agrees, then people
cannot be held accountable on an everyday basis for following a different policy or
acting in ways that do not meet act-consequentialist standards. Therefore, even if
the kind of metamoral theory Greene defends encompasses characteristically
consequentialist judgments and characteristically deontological judgments, his
argument still would not fully support consequentialism construed as a theory of
moral right to the extent that it does not address Darwall’s key conceptual concern
about the connection between the notions of moral right and moral accountability.
The third part of the book is perhaps the most attractive for those readers engaged
in methodological issues around moral neuroscience. In their contribution, James
Blair, Soonjo Hwang, Stuart White, and Harma Meffert defend an integrated
emotion systems model of psychopathy, which aims to understand the functional
properties of the neural systems involved in psychopathic traits and the compu-
tational implications of their dysfunction (Blair 2007). They argue that emotional
systems allow norms to acquire their prohibitive power by guiding our attitudes
toward these norms and their violation. In the next chapter, Ricardo de Oliveira-
Souza, Roland Zahn, and Jorge Moll focus on developmental psychopathy and
acquired sociopathy. Their goal is reviewing and extending previous attempts to
infer the neural underpinnings of moral cognition through research on normal and
abnormal moral behavior. From a methodological point of view, they integrate
information from functional neuroimaging on normal subjects as well as lesion
studies on psychopaths and subjects with antisocial personality and conduct
disorders either in vivo or through postmortem exam.
My highlights of this part of the book are Molly Crockett’s and Jana Schaich
Borg’s contributions. On the one hand, Crockett’s chapter focuses on the influences
of the neuromodulator serotonin on moral judgment and behavior. The evidence
reviewed in this chapter reveals, for instance, that pharmacological enhancement of
serotonin function increases people’s aversion to harmful actions, and thus makes
people less likely to judge harmful actions as morally permissible in hypothetical
scenarios. Similarly, increased levels of serotonin have been shown to reduce
people’s willingness to inflict financial harm on others in retaliation for unfair
treatment in ultimatum games. Since there seem to be no healthy levels of serotonin,
and it is currently impossible to determine a baseline physiological state from which
we can generate reliable moral judgments, Crocket argues that the influence of
serotonin could have important normative implications, as moral judgments would
be sensitive to non-normative factors that are significantly variable. In other words,
these results warn us about potential noise introduced by serotonin function at the
implementation level of moral judgment and decision-making.
On the other hand, Schaich Borg’s chapter discusses the relevance of rodent
models of negative intersubjectivity in the study of moral behavior and cognition.
I. Gonzalez-Cabrera
123

Roughly speaking, negative intersubjectivity is the process of disliking or feeling
negative (for whatever reason, selfish or not) when another individual feels bad (p.
248). Schaich Borg argues that a central reason to pursue this avenue of research is
that negative intersubjectivity is an important regulatory mechanism of immoral
action as shown in studies on the affective components of empathy and research on
callous personality traits. Another reason is that neuroscience tools available in
humans such as functional magnetic resonance imaging (fMRI) have poor temporal
and spatial resolution to study the type of processes we believe are responsible for
moral cognition and behavior. The question is, of course, whether rodent models are
actual models of moral cognition, but Schaich Borg argues that rodent models
should be complemented by similar tests in humans for validation and comparison.
This emphasis on the role of comparative psychology in the study of moral behavior
and cognition is particularly welcome since the study of non-human animal
cognition connects research in the psychological sciences to the phylogenetic
history, adaptive significance, and ontogeny of behavior and cognition. By focusing
on moral action, Schaich Borg’s contribution also reminds us of the risk of over-
intellectualized views of moral cognition that have limited practical implications.
Although understanding moral judgment might be philosophically deep and
genuinely important, so is understanding why bad, overly aggressive behavior
happens. In this context, non-human models of empathy and aggression control can
be enlightening, even if they are cognitively impoverished under some reasonable
anthropocentric standard.
In the final part of the book, Guy Kahane argues that the most interesting
arguments that allow drawing interesting normative conclusions are epistemic in
nature, i.e., arguments in which the causal origins of our beliefs affect their
justification (pp. 290–291). Since the epistemic status of moral beliefs will
frequently depend on whether their distal, as opposed to proximal, causes are
reliable sources of moral evaluations, findings on the neural mechanisms of moral
cognition will play only a minor role in such arguments. In the following chapter,
Matthew Liao argues that heuristics involve a form of reasoning, regardless of
whether one understands heuristics as an attribute substitution process (Kahneman
and Frederick 2005; Sinnott-Armstrong et al. 2010) or as a fast-and-frugal algorithm
(Gigerenzer 2008). Given that intuitions entail forming conclusion-judgments not
based on premise-judgments, they are different from reasoning, and thus different
from heuristics understood either way. So, Liao argues, normative claims, such as
those of Greene, that deontological intuitions tend to be inaccurate and unreliable
like the automatic settings in a digital SLR camera would be unwarranted. In the
closing chapter of this volume, Walter Sinnott-Armstrong draws heavily on his
previous work (Parkinson et al. 2011; Sinnott-Armstrong 2008; Sinnott-Armstrong
and Wheatley 2012,2013) to argue that no single set of common and distinctive
features of moral judgments that enables interesting psychological generalizations
can unify them. Unification here means to be able to test which judgments are moral
in order to reveal what it is that makes these judgments to be moral (p. 335).
However, Sinnott-Armstrong argues that there are reasons to think that moral
judgments are not unified in terms of their content, neural basis, and function—see
also Sinnott-Armstrong (2008) and Sinnott-Armstrong and Wheatley (2012,2013)
Moving beyond dichotomies: Liao, S. Matthew (ed.), Moral…
123

for arguments against other potential ways to unify moral judgments. This raises the
question about what feature (or set of features) could possibly unify moral
judgments in the sense specified above. As a result, he suggests a bottom–up
methodological approach aimed to investigate more carefully defined subclasses of
moral judgments that might or might not lead to the desired unification.
Discussion
As previously mentioned, one of the central themes of the book is the implications
of moral neuroscience for normative ethics. In this part of the review, I would like to
reflect further on this issue. For one central feature of the book is that many of the
contributions, especially from philosophers, often point out how little we can
actually learn from this data—Prinz, Woodward, and Kahane are particularly
explicit on this point. Most of the contributions indeed focus on traditional
psychological methods. This is understandable since traditional psychological
methods are semantic (in the sense of targeting mental states with content about the
world), which seems more informative than mere data about, say, the formal
computations of cognitive systems or how they are implemented in actual neural
systems. One important exception is Crockett’s contribution since she focuses on
how moral judgments respond to neuromodulators such as serotonin that are, in
principle, morally inane and not clearly linked to morally relevant distal factors (see
Kahane, pp. 294–295, in this volume for discussion).
The claim I want to defend now closely follows that of Woodward in this
volume, for I want to argue that moving away from certain dichotomies prevalent
in, but not exclusive to, Greene’s dual process theory of moral judgments makes
neuroscience much more useful in advancing normative ethics. More specifically, I
want to challenge the following assumptions: first, the idea that either we rely on
automatic moral settings or we rely on conscious reasoning, and second, the idea
that either we have adequate evolutionary, cultural, or personal experience, or we
have not.
Rejecting these dichotomies makes it easier to derive normative conclusions
from premises about neuroscientific facts by focusing on the interaction between
automatic settings and effortful thinking as well as on the coordination and
integration of relevant disciplines beyond neuroscience such as evolutionary
biology, cultural evolution, and developmental psychology. Even assuming that
moral facts are natural facts, neuroscience alone cannot bridge the gap between
premises about neuroscientific facts and the moral implications that we aim to
derive from those facts. Moral judgments (understood as mental states) are just not
reducible to facts about neural architecture, as Sinnott-Armstrong argues in this
volume and elsewhere (Parkinson et al. 2011), and the causal connection between
our neural organization and the relevant facts that our moral judgments are supposed
to track (whatever they are) does not seem reconstructable by neuroscientific
research alone.
Regarding the first dichotomy, it is not true that we rely on either one mode of
cognition or the other since automatic settings and effortful thinking interact to
I. Gonzalez-Cabrera
123

influence moral judgment—Woodward makes a similar point in this volume with
respect to the emotion/reason dichotomy. For example, effortful thinking can
influence the prediction error upon which our automatic mode of cognition operates
(Daw et al. 2011). Similarly, automatic settings may provide estimates which we
employ through effortful thinking when forced by computational complexity to
prune its online evaluation of options (Crockett 2013). Furthermore, even if effortful
thinking requires to override our default intuition and replace it by, say, conscious
reasoning, the capacity to overrule intuitive responses is also a function of factors
such as the metacognitive feeling of rightness in the initial response (Thompson
2009; Thompson et al. 2011). Hence, it is not true that we rely on either one mode of
cognition or the other since we can rely on both automatic settings and effortful
thinking.
Relying on both modes of cognition can reduce computational noise. Compu-
tational noise can be defined as the chance variability of judgments due to the
influence of irrelevant factors. The complex calculations associated with effortful
thinking are often accurate but they are not immune to computational noise—e.g.,
time and stress pressure, limitations on attention, speed, the ability to multitask, and
depletion of other cognitive resources. Similarly, there is also noise associated with
incomplete and inefficient learning associated with our automatic settings.
Information gathered through experience is always partial and learning from it
requires significant time. Yet the interaction between automatic settings and
effortful thinking can help to reduce the computational noise of each other. For
instance, effortful thinking can train our automatic settings through offline
simulation (Ji and Wilson 2007), which reduces the exploratory risk and cost
associated with prolonged reinforced learning in the latter. Since effortful thinking
can influence the prediction error upon which the automatic settings are learned
(Daw et al. 2011), it can also reduce computational noise by speeding up learning.
Moreover, automatic settings can help to reduce computational noise associated
with effortful thinking by providing estimates which are used to prune the options
that the latter evaluates (Crockett 2013).
Less computational noise increases the computational robustness of the overall
decision-making system. Computational robustness is the ability of a computational
system to maintain its functionality across a diverse array of operational conditions.
In the context of moral decision-making, it would mean something like making
good moral choices in a wide range of circumstances. Reducing the chance
variability of judgments due to the influence of irrelevant factors would
consequently increase the chances of making good moral choices across a number
of possible scenarios. Therefore, relying on both automatic settings and effortful
thinking can increase the computational robustness of the overall decision-making
system.
Because relying on both systems can increase the computational robustness of
our moral decision-making, we can investigate how to increase this form of
robustness significantly more by looking at the interaction between both systems
than by looking just at the relative robustness of each system independently. To put
it another way, looking at the interaction between automatic settings and effortful
thinking can help us to advance normative ethics significantly more than thinking of
Moving beyond dichotomies: Liao, S. Matthew (ed.), Moral…
123

these systems separately, for the more we understand how to increase the
computational robustness of moral decision-making, the more we can advance
normative ethics. A deeper understanding of these interactions and their
consequences for computational performance can help us, for instance, to find
more robust moral principles and theories.
Regarding the second dichotomy, if it were true that either we have adequate
evolutionary, cultural, or personal experience to solve a moral problem, or we have
not, then it should not be the case that sometimes we have partially adequate
experience about morally relevant facts. But it is difficult to conceive a moral
problem in which all our evolutionary, cultural, or personal experience turns totally
inadequate. For example, we often have evolutionary, cultural, and personal
experience about intentional facts that is relevant for moral evaluation. So, it is not
true that either we have adequate evolutionary, cultural, or personal experience to
solve a moral problem, or we have not, and thus we frequently have both adequate
and inadequate experience to solve moral problems.
This shows how we need to rely on normative ethics to tell us what facts are
morally relevant and when we have gathered information about them through our
evolutionary, cultural, or personal experience. The relevance of some of these facts
could sometimes be controversial, but not always. Relying on uncontroversially
relevant moral facts puts us on the safe side, as Greene remarks. But as Humeans
repeatedly remind us, we cannot logically derive a conclusion with explicitly moral
content from premises without moral content—a claim that could be true even if
moral predicates were synonymous with non-moral predicates (Pigden 2010). This
means that Driver is right to emphasize the background role of moral theorizing,
contrary to Greene’s assumption (p. 171). Whatever the metaphysical status of those
facts is, we need moral theorizing to shed light on what facts are morally relevant in
a particular moral situation and whether they support our premises about the
adequacy of our evolutionary, cultural, or personal experience.
Moreover, we would need facts about our evolutionary, cultural, and develop-
mental history that connect facts about our neural wiring with morally relevant
facts. That is, we would need to rely not only on facts about our internal wiring on
the one hand and on normative ethics to pinpoint morally relevant facts on the other
but also on connection facts that link these two. Connection facts are facts about
how our internal organization registers and tracks external circumstances. Consid-
erations about these connection facts allow us to assess the reliability of our inner
neural wiring to track those morally relevant facts—this tracking cannot just be a
matter of luck as if we took moral decisions by throwing a dice (see Kahane, p. 294,
in this volume). The life sciences can contribute much to this project because
understanding the connection between cognitive machinery and relevant moral facts
requires explaining how such machinery evolved, how it develops, and how it
relates to our environment under ecological conditions that we often engineered
through our cultural practices and which affect cognitive performance. Since we
need facts about our evolutionary, cultural, and developmental history that connect
facts about our neural wiring with morally relevant facts, then we need to integrate
research on the life sciences more broadly (including the study of cultural evolution)
for moral neuroscience to advance normative ethics.
I. Gonzalez-Cabrera
123

This point is nicely illustrated by Driver’s question on whether it may also be
considered a cognitive miracle that moral judgments track moral truth at all, to
which Greene replies that, in his understanding of cognitive evolution, it is generally
adaptive to have true beliefs. Yet this line of reply makes too many assumptions
about the evolutionary link between cognition and moral truth. Perhaps having
mental states that track facts about our environment is adaptive but it is less clear
why these mental states have to be belief-like. Perhaps having belief-like mental
states that track facts about our environment is adaptive but it is less clear why those
facts have to be moral. Perhaps having belief-like mental states that track moral
facts is adaptive but it is less clear why this was so in the hominin lineage—we still
need an evolutionary story about how having belief-like mental states that track
morally relevant facts (assuming that those facts exist) was indeed fitness-
enhancing. Moreover, even if having true beliefs were always adaptive, it does not
follow from that that all adaptations (cognitive or not) are traits for having true
beliefs. Systems involved in moral cognition can be adaptations, although not
necessarily adaptations for tracking moral facts—e.g., cognitive mechanisms for
language can be adaptations for solving coordination problems between multiple
agents rather than adaptations for tracking environmental facts.
To sum up, Moral Brains explores a wide range of issues in moral psychology,
even if still too attached to traditional debates such as the role of emotions and
reasoning in moral cognition or the reliability of deontological versus consequen-
tialist moral thinking. The title of the book might be indeed somewhat deceiving
since not all chapters engage with actual neuroscience and not all of them with the
same breadth and depth. Yet this is a rather enjoyable feature of the book and
certainly an essential part of its take-home message. For bridging the gaps between
neuroscientific facts and moral philosophy is an integrative enterprise, which
requires a more detailed understanding of how we relate as organisms to our
environments. Moral neuroscience is not, after all, reducible to mere neuroscience.
Acknowledgements For helpful comments and suggestions, I would like to thank Claudia Passos-
Ferreira and Zoe Drayson.
References
Blair RJR (2007) The amygdala and ventromedial prefrontal cortex in morality and psychopathy. Trends
Cogn Sci 11:387–392. doi:10.1016/j.tics.2007.07.003
Crockett MJ (2013) Models of morality. Trends Cogn Sci 17:363–366. doi:10.1016/j.tics.2013.06.005
Daw ND, Gershman SJ, Seymour B, Dayan P, Dolan RJ (2011) Model-based influences on humans’
choices and striatal prediction errors. Neuron 69:1204–1215. doi:10.1016/j.neuron.2011.02.027
Gigerenzer G (2008) Moral intuition =fast and frugal heuristics? In: Sinnott-Armstrong W (ed) Moral
psychology: the cognitive science of morality: intuition and diversity, vol 2. MIT Press, Cambridge,
pp 1–26
Greene JD (2014) Beyond point-and-shoot morality: why cognitive (neuro)science matters for ethics.
Ethics 124:695–726. doi:10.1086/675875
Greene JD, Nystrom LE, Engell AD, Darley JM, Cohen JD (2004) The neural bases of cognitive conflict
and control in moral judgment. Neuron 44:389–400. doi:10.1016/j.neuron.2004.09.027
Haidt J (2012) The righteous mind: why good people are divided by politics and religion. Pantheon
Books, New York
Moving beyond dichotomies: Liao, S. Matthew (ed.), Moral…
123

Haidt J, Bjorklund F (2008) Social intuitionists answer six questions about morality. In: Sinnott-
Armstrong W (ed) Moral psychology: the cognitive science of morality: intuition and diversity, vol
2. MIT Press, Cambridge, pp 181–217
Ji DY, Wilson MA (2007) Coordinated memory replay in the visual cortex and hippocampus during
sleep. Nat Neurosci 10:100–107. doi:10.1038/Nn1825
Kahneman D, Frederick S (2005) A model of heuristic judgments. In: Holyoak KJ, Morrison RG (eds)
The Cambridge handbook of thinking and reasoning. Cambridge Universty Press, Cambridge,
pp 267–293
Parkinson C, Sinnott-Armstrong W, Koralus PE, Mendelovici A, McGeer V, Wheatley T (2011) Is
morality unified? Evidence that distinct neural systems underlie moral judgments of harm,
dishonesty, and disgust. J Cogn Neurosci 23:3162–3180
Pigden CR (2010) Hume on is and ought. Palgrave Macmillan, Basingstoke
Rolls ET (2005) Emotion explained. Oxford University Press, Oxford
Shenhav A, Greene JD (2010) Moral judgments recruit domain-general valuation mechanisms to integrate
representations of probability and magnitude. Neuron 67:667–677. doi:10.1016/j.neuron.2010.07.
020
Sinnott-Armstrong W (2008) Is moral phenomenology unified? Phenomenol Cogn Sci 7:85–97. doi:10.
1007/s11097-007-9065-z
Sinnott-Armstrong W, Wheatley T (2012) The disunity of morality and why it matters to philosophy.
Monist 95:355–377
Sinnott-Armstrong W, Wheatley T (2013) Are moral judgments unified? Philos Psychol 27:451–474.
doi:10.1080/09515089.2012.736075
Sinnott-Armstrong W, Young L, Cushman F (2010) Moral intuitions. In: Doris J (ed) The moral
psychology handbook. Oxford University Press, Oxford, pp 246–272
Thompson VA (2009) Dual-process theories: a metacognitive perspective. In: Evans JSBT, Frankish K
(eds) In two minds: dual processes and beyond. Oxford University Press, Oxford, pp 171–196
Thompson VA, Prowse Turner JA, Pennycook G (2011) Intuition, reason, and metacognition. Cogn
Psychol 63:107–140. doi:10.1016/j.cogpsych.2011.06.001
Williams B (1995) The point of view of the universe: Sidgwick and the ambitions of ethics. In: Altham
JEJ, Harrison R (eds) Making sense of humanity and other philosophical papers, 1982–1993.
Cambridge University Press, Cambridge, pp 153–171
I. Gonzalez-Cabrera
123