Gallagher 3
Content
Phenomenology and embodied
cognitive science. University of
Jyväskylä, Finland (September
2006).
Outline
1. Conditions for moral personhood (Dennett)
2. In an attempt to deepen our understanding of
these conditions: An Aristotelian argument:
moral personhood depends on both embodied
and social aspects of human existence which
give us the capacity for phronesis.
3. The Dreyfus-Collins debate on expertise
4. Expertise and phronesis
5. Revisiting the conditions for moral personhood
Moral personhood and phronesis
Shaun Gallagher
Philosophy and Cognitive Sciences
University of Central Florida
[email protected]
Moral personhood (moral agent)
The capacity for moral action and responsibility
• Dennett (1978): for an entity to have the status of
moral personhood it must meet six conditions
1. Rationality
2. We must be able to take the intentional stance
toward it -- we treat it as if it had a mental life
3. Recognition by others -- we take a certain attitude
toward it
4. Capacity for reciprocity
5. Communicative capacity
6. Capacity for self-consciousness, defined as a
sophisticated reflective (second-order) mental
state.
• An Aristotelian argument: moral personhood depends
on both embodied and intersubjective aspects of
human existence
– Someone is a moral agent, that is, they have the status of
moral personhood and have the capacity to do the right or
wrong action, only if they are capable of having (practicing)
phronesis (practical wisdom).
– The person who is capable of moral action/ responsibility
(the moral agent) is fully embodied (is more than a brain),
and this embodiment shapes the very nature of moral
possibilities.
– To be capable of having (practicing) phronesis depends on
an intersubjective existence -- and specifially a form of
“endogenous intersubjectivity”
– In light of these considerations, we will come back to
Dennett's six conditions as conditions for the possibility of
having phronesis.
1.
2.
3.
4.
5.
6.
Rationality
Intentional stance
Recognition
Reciprocity
Communication
Self-consciousness
• 1-5 are necessary conditions for selfconsciousness.
• Young children are not capable of higher-order
self-consciousness (Dennett, Frankfort, Wilkes).
• Conditions 2-5 explicitly and importantly involve
dimensions of intersubjectivity, although, for
Dennett, the precise nature of these social
dimensions is still an open question.
• Dennett (e.g., 1982, 1991) suggests that a brain in a vat
or a computer might be able to attain this kind of selfconsciousness. This implies that these conditions do not
depend on embodiment in any strong sense,
and it raises questions about the notion of
intersubjectivity involved.
Starting point: a debate on expertise
• I set aside the question of whether phronesis
is a kind of ethical expertise until later.
• But getting a good understanding of what
constitutes expertise will throw some light
on the nature of phronesis.
• A debate between Dreyfus and Harry
Collins (sociologist, Cardiff) on expertise.
1
Dreyfus: phenomenological
account of the embodied
basis for expertise
• Expertise is an instance of embodied human
performance -- on a continuum with basic life-world
practices.
“We are all experts at many tasks and our everyday
coping skills function smoothly and transparently so as
to free us to be aware of other aspects of our lives
where we are not so skillful” (Dreyfus and Dreyfus,
1990).
Phenomenological background
• Merleau Ponty’s notion of the lived body and the
concepts “intentional arc” and “maximal grip” -practiced activities controlled by body schematic
adjustments that operate as the basis for expert
practices
• Heidegger’s emphasis on pragmatic contexts as our
primary way of engaging in the world -- just those
contexts in which expertise is practiced.
• Higher order reflection on how one does what one
does only occurs when things or procedures fail to
work effectively (Heidegger’s hammer)
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
Monday, May 8, 2006
• Dubner and Levitt: The Birth-Month Soccer
Anomaly: elite soccer players are more likely to
have been born in the earlier months of the year than
in the later months.
• Could astrology explain expertise in soccer?
• Answer: No. Research by Anders Ericsson at Florida
State University over the past 30 years has
demonstrated that “deliberate practice” is what
creates expertise: setting specific goals, obtaining
immediate feedback and concentrating as much on
technique as on outcome.
• Dreyfus’s target: any account of expertise given
in terms of purely cognitive, propositional, or
representationalist terms -- expert knowledge as a
set of explicable rules or as a form of theoretical
knowledge.
• Expert skills -- a matter of of “knowing how”
rather than “knowing that”
• Knowing how (vs propositional knowledge)
involves embodied practice rather than cognitive
deliberation -- the exercise of skills of which one
cannot fully give an account or full articulation
(e.g., chicken sexing)
• Expertise is a matter of intuition, not
intellectualization: “Action becomes easier and less
stressful [as the expert] simply sees what needs to be
done rather than using a calculative procedure to
select one of several possible alternatives” (Dreyfus
2001).
• The expert not only sees what needs to be done, but
also how to achieve it without deliberation,
immediately -- non-reflectively recognizing new
situations as similar to previously encountered ones,
and intuiting “what to do without recourse to rules” -he or she recognizes important features as contextually
sensitive (Dreyfus and Dreyfus 1986).
• Expertise is in the practice, and the expert is a
practitioner.
• Expert knowledge is not third-person propositional
knowledge; it is not theoretical knowledge. To think of
it in this way distorts the first-person embodied
experience -- the form of life -- that constitutes
expertise.
• Expertise is not captured in a set of rules; it is not a
technical kind of knowledge in this sense. Dreyfus
thus objects to the artificial, computational creation of
expertise -- vs e.g., Doug Lenat
2
Doug Lenat (the Cyc project)
• Expertise is reducible to rational
(computational) rules and propositional
knowledge -- it can be rationally reconstructed as
primarily a mental intellectual phenomenon
• Lenat argues for a kind of expertise that Dreyfus
rules out -- an expert who is rich in propositional
facts about X, but has never done or experienced
X.
• The expert need not be a practitioner. This kind
of expertise can be disembodied and instanciated
in a computer.
Harry Collins:
The social explanation of expertise
• Vs both Dreyfus’s emphasis on
embodiment and Lenat’s emphasis on
propositional knowledge
• Dreyfus’s explanation remains asocial.
In an account that takes social dimensions seriously,
expertise is thought of as “distributed” -- embedded
in social practices, particular settings (laboratories
and social networks), standardized in technologies,
and promoted in specific rhetorical means of
recruiting professional experts (Mialet 1999)
For Collins, expertise constituted in language or languagegames, and does not depend on embodied practices.
– He cites Wittgenstein’s example of a talking lion -- an animal
that divides up the world in a different way.
– For example, for us, a chair is a different kind of affordance
(Gibson) than for the lion -- we have a body that can use a chair
to sit -- so ‘chair’ if in the lion’s vocabulary, would mean
something different.
– Collins claims that if the lion hung around and communicated
with humans long enough, the lion would come to know the
meaning of chair, despite differences in embodiment.
“the language of a community embodied in one way can
be acquired by individuals with different shaped
bodies, and who, therefore, cannot participate in the
activities of that community.” -- therefore, bodies
contribute little to the acquisition of expertise.
Lenat points to the case of Madeleine as an example of
purely intellectual and disembodied expertise.
• Oliver Sacks (1985): Madeleine was born
blind and disabled. She became so dependent
on others that she was even unable to use
her hands to read brail or for any other
purpose -- she was utterly inactive
throughout her early life
• Nevertheless, Madeleine learned about the
world from books read to her by others. She
was able to speak “freely indeed eloquently
... revealing herself to be a high-spirited
woman of exceptional intelligence and
literacy” She was “of exceptional
intelligence and literacy, with an
imagination filled and sustained, so to
speak, by the images of others, images
conveyed by language, the word.”
• Lenat takes this to mean that Madeleine had
propositional but not embodied knowledge of
• Collins (2004) also argues for a kind of expertise that
Dreyfus rules out -- a kind of expert knowledge that
someone may attain via social learning procedures -specifically, linguistic and communicative processes:
interactive expertise vs contributory expertise
• vs Lenat, this expertise is not based on propositional
knowledge -- it is the result of a socialization process that
requires conversational interaction with another
• Collins himself is example.
– A sociologist of science learns gravitational wave physics-- not
by learning the propositional science itself, and not by doing the
science, but by hanging out with the scientists, conversing with
them, seeing what they do, etc. -- to the point that he becomes
proficient in the language of gravitational wave physics and not
only can hold up his side of a conversation, but make intelligent
suggestions -- even though he is not a practitioner.
• Collins offers a different
interpretation of the case of
Madeleine to show that through
communication and a passive
reception of information (she
attained much of her knowledge by
having someone read to her) -thus, through the medium of the
written [heard] and spoken word -Madeleine could easily be an expert
on X.
“Madeleine had a minimal `body' with
almost no ability to take part in
the normal activities of the
members of the surrounding
3
The criticism of Dreyfus for ignoring the social aspects
of expertise has been extended in a slightly different
way:
• Both Iris Young (1998) and Maxine SheetsJohnstone (2000) criticize Dreyfus for assuming
that the body which acquires skill has no relevant
biography, gender, race, or age. They emphasize
the cultural embeddedness of the body.
• This approach nicely explicates the external
limitations on expertise that cultural factors impose,
and has much to say about social-economic-political
factors that limit embodied processes, but it does not
provide any positive account of how expertise
develops except by adding these external limitations
to Dreyfus’s account.
• Selinger and Crease (2002) point out that Dreyfus
does have a place for the idea that “cultural styles”
affect how skills are learned (Dreyfus, 2000), but that
this notion is simply not developed.
– “From Dreyfus’s perspective, one develops the
affective comportment and intuitive capacity of
an expert solely by immersion into a practice;
the skill-acquiring body is assumed to be able,
in principle at least, to become the locus of
intuition without influence by forces external to
the practice in which one is apprenticed.”
In summary:
• Lenat rules out any important role for embodiment
and emphasizes a cognitive - computational model
consistent with traditional views of expertise as a
mentalistic or intellectual phenomenon.
• Collins also rules out any important role for
embodiment and emphasizes a socially contextualized
model of expertise. This model is primarily a
linguistic-communicative one.
• Dreyfus critiques traditional and computational models,
but ignores social/intersubjective dimensions and
emphasizes pre-reflective embodied skills as the
basis for expertise.
Neuroscience of Resonance systems:
Brain areas associated with my own action are also
active during imitation and observation of others
(Gallese, Jeannerod)
Shared neural representations (mirror system)
– action
– observation of another’s action
– imaginative enactment of action Observation
– preparing to imitation
Imagination
– Neuroscience: shared representations and mirror neural
systems
– Developmental psychology: primary and secondary
intersubjectivity (Trevarthan)
– Phenomenology: embodied intersubjective experience that
involves both perceptual processes and pragmatic contexts.
Developmental Psychology
Trevarthan’s notions of primary and secondary
intersubjectivity
Primary intersubjectivity: the infant’s ability to
perceive meaning in the other’s gestures, movements,
facial expressions, intentional actions.
Action
Preparing to
imitate
Are these not the embodied apprenticeship
experiences that build expertise -- and
doesn’t this make them already social?
The interaction approach
Expertise (including our everyday expertise and coping
skills) requires embodied and intersubjective practices,
where embodied practices are always intersubjective
and intersubjective practices are always embodied.
This connection between embodiment and
intersubjectivity can be found in infancy and continues
as the basis for intersubjective understanding even in
adults -- and it directly ties in to questions about the
acquisition of expertise.
Infants engage and interact with
others -- from neonate imitation
to emotional mirroring, to eyetracking, to infant’s ability to
parse intentions of others -perceptual access to others (vs
theory or simulation)
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
4
Phenomenology of primary intersubjectivity
“The very first of all cultural objects, and the one
by which all the rest exist, is the body of the
other person as the vehicle of a form of
behavior” (Merleau-Ponty PhP 348). “I live in
the facial expressions of the other, as I feel him
living in mine … (Merleau-Ponty, N 218).
“… the perception of others is anterior to, and the
condition of, such observations [simulations,
theoretical stances], the observations do not
constitute the perception. … between this
phenomenal body of mine and that of another as I
see it from the outside, there exists an internal
relation which causes the other to appear as the
completion of the system” (352)
Secondary intersubjectivity: Around the age of 1 year,
the infant goes beyond person-to-person immediacy
and enters contexts of shared attention -- shared
situations -- learning what things mean and what they
are for.
“The defining feature of secondary intersubjectivity
is that an object or event can become a focus
between people. Objects and events can be
communicated about. … the infant’s interactions
with another person begin to have reference to the
things that surround them” (Hobson, 2002, p. 62).
– Studies of shared attention in infancy.
– Infants understand the failed intentions of others
(Meltzoff 1995).
Infants at 18 months imitate to completion the
unfulfilled intentions of humans; they do not
imitate the movements of the mechanical
device.
Phenomenology of secondary intersubjectivity
• Trevarthan’s developmental concept of
secondary intersubjectivity is already
foreshadowed by phenomenological analyses of
Heidegger and Gurwitsch,
• Understanding the meaning of an object or an
instrument is dependent on our involvement in
pragmatic contexts
• In the pragmatic we also find intersubjectivity -other people appear in meaningful roles within
pragmatic contexts.
• Aron Gurwitsch (following Heidegger’s
analysis of equipment and circumspective
engagement with the surrounding environment,
and the larger action contexts of human
existence)
– our understanding of the other’s expressive
movements depends on meaningful
instrumental/pragmatic contexts
– Things and situations provide scaffolds for
understanding the actions of others -- and in
those pragmatic contexts we see and come to
learn and imitate what they do
5
• Dreyfus takes up Heidegger’s analysis of
equipment, and his account of expertise follows
this line of thought -- expertise is embodied and
pragmatic -- but what about intersubjectivity?
• Clearly, the embodied-pragmatic expertise (of the
sort described by Dreyfus) is dependent on
intersubjective interaction.
• To add a corrective to Heidegger-GurwitschDreyfus: It is not that our primary access to
others is through pragmatic contexts; it's that
our access to pragmatic contexts is primarily
through others.
Conclusions:
• In opposition to Collins and Lenat: It is not feasible or
necessary to eliminate embodiment in order to make room for
social and/or even what appear to be “purely” cognitive
dimensions of expertise (cf. embodied cognition; Mark Johnson)
• As a supplement to Dreyfus: Nor is it necessary or feasible to
eliminate intersubjective dimensions in order to make room for the
role played by embodiment
• To the extent that expertise depends on the grasping of the
intentions and meanings of others -- on acting, observing others
act, imagining ourselves act, imitating -- it also depends on
embodied resonance processes that are simultaneously
intersubjective.
• To suggest that expertise depends only on embodiment without the
intersubjective dimension (Dreyfus), or only on the social without
embodiment (Collins) is to give an incomplete account of
expertise.
Aristotle : phronesis
• The good person, the person with
phronesis, like the person with expertise,
sees what to do in an immediate way, and
does the good thing in a close to automatic
way, as if it were second nature.
• But, in regard to acquisition, Aristotle
takes a position that is similar to Collins on
expertise: one acquires phronesis through a
good upbringing, and this means hanging
around with the right people – good people
who provide good examples of good
actions.
• You don’t get phronesis by taking an
ethics course (vs Lenat).
Merleau-Ponty
“Insofar as I have sensory functions … I am
already in communication with others ….
No sooner has my gaze fallen upon a living
body in the process of acting than the
objects surrounding it immediately take
on a fresh layer of significance”
Intersubjective contexts cannot be reduced to or
encompassed by purely instrumental ones -- the social
aspects of experience transcend the purely pragmatic.
• Evidence from recovery after brain injury (Gallagher and
Marcel 1999)
• Evidence from studies of gesture in cases of
deafferentation (Cole, Gallagher, McNeill 2001)
Return to the question of moral agency/moral
personhood and phronesis
• Is phronesis equivalent to expertise? Is the acquisition
of phronesis best explained on a skills-acquisition
model?
• Phronesis is like expertise in certain ways. Following
along the same lines of the debate about expertise,
phronesis is not equivalent to theoria – that is, a
theoretical knowledge that is propositional and learnable
in a purely intellectual way. Phronesis cannot be
programmed into a computer.
• It is, rather, as Dreyfus says of expertise, a kind of
"know-how," but is nonetheless not reducible to a set of
rules, and should be distinguished from techne.
• Yet this is not sufficient.
– On the Collins model of expertise, there is no guarantee
that one's social interaction with experts will necessarily
make one an expert.
– Plato's complaint in the Meno – a son who is raised by
good parents and given the best education amongst the
best of society still may turn out bad.
• To attain phronesis, one must also act and interact in
a good way. It would not be enough simply to
watch, or to converse with good people. One needs
to imitate them, to act as they do and to do the kinds
of things that they do: phronesis, like expertise, is in
the practice -- Aristotle = Dreyfus + Collins
6
More on acquisition
• In every case (for expertise and for phronesis) Dreyfus
outlines a multi-step acquisition process: novice to
advanced beginner to competence to proficiency to
expert.
• In each case the novice stage starts like this: “Normally,
the instruction process begins with the instructor
decomposing the task environment into context-free
features that the beginner can recognize without benefit
of experience. The beginner is then given rules …”
(Dreyfus and Dreyfus 2004).
• Seemingly we start with rules and/or theory and work our
way out of dependency on these mentalistic beginnings to
non-mentalistic expertise through practice. This may
work for learning to drive or play chess. But what about
learning first language or learning to walk?
• How do we gain phronesis? Aristotle: by hanging out
with the right people. Learning from example.
Imitating. Picking up practices.
Aristotle is Trevarthan, not Allison Gopnik
• Dreyfus seems to be a friend of theory in the
acquisition/developmental stages (although he also talks
about the importance of practice)
“On analogy with chess and driving, it would seem that
the budding ethical expert would learn at least some of
the ethics of his or her community by following strict
rules, would then go on to apply contextualized maxims,
and, in the highest stage, would leave rules and
principles behind and develop more and more refined
spontaneous ethical responses” (Dreyfus 2004).
The difference between phronesis and expertise
• Is phronesis "ethical expertise”?
• Just as Aristotle makes a distinction between phronesis
and cleverness, as well as a distinction between virtuous
action and techné (1105a22), a similar distinction should
be clearly made between phronesis and expertise.
“There is a faculty which is called cleverness; and this is
such as to be able to do the things that tend towards the
mark we have set before ourselves, and to hit it. Now if
the mark be noble, the cleverness is laudable, but if the
mark be bad, the cleverness is mere smartness; hence we
call even men of practical wisdom clever or smart.
Practical wisdom is not this faculty, but it does not exist
without this faculty. … practical wisdom is to cleverness
-- not the same, but like it” …. “it is impossible to be
practically wise without being good.” (N. Ethics, VI)
More on acquisition
• Language: we learn the rules (grammar) after we learn to
speak; Walking -- are there any rules? -- these are things
we learn by pure practice.
• How do we learn our everyday coping skills? Not by
working our way through theory or a set of rules.
• How do we come to understand others and gain our
“people skills”-- not by theory -- we are not given rules.
We are given people and we start to interact with them and
imitate them (primary and secondary intersubjectivity). In
the longer term we learn by narratives. Not theories, not
rules, but pure doings.
• Driving? Rules or practices?
• But Dreyfus is not a friend of theory in
regard to the actual practice of expertise or
in regard to phronesis.
• Vs Habermas and Benhabib’s implicit
moral theory, ethical cognitivists, relying on
rules and principles.
• Pro Gilligan in emphasizing care over the
over-rational justice.
Phronesis according to Dreyfus
• Dreyfus is right and properly Aristotelian in
characterizing phronesis as non-mentalistic and as
not relying on rules or maxims; the phronimos copes
case-by-case, attending to differences in situations.
• But this is not the complete picture of phronesis –
something more is required. This goes beyond the
question of the educational backdrop.
• Consider the following characterizations made by
Dreyfus (2004), following Heidegger.
7
• The phronimos is the “master of his or her culture’s
practices”
• Ethical experts: “experts capable of responding
appropriately to a wide range of interpersonal
situations in their culture. Such social experts could be
called virtuosi in living …. This is obviously
Aristotle’s phronimos.”
• Following Heidegger: “people have skills for coping
with equipment, other people, and themselves”
(Dreyfus 2004)
• Is phronesis reducible to people skills or virtuosity in
interpersonal dealings? Is virtuosity equivalent to
virtue? Just as I can be a clever criminal, I can be a
virtuoso in selling used cars, managing an organization,
convincing people to vote for me, managing a
classroom.
• But none of this requires that I do the good or right
thing.
• A person could know and have the know how for
exactly the right thing to do, to act morally (= ethical
expertise), and be inclined to act that way, but decide
not to act ethically, and use her knowledge to act in a
way that is not ethical.
• Such a person might have ethical expertise, but would
not have phronesis.
• What exactly is it that makes phronesis so different
from ethical expertise -- why not say, for example, that
phronesis is expertise in what constitutes the moral life?
• On the Dreyfus model of expertise, one would have to
practice one's skill. But what precisely is the skill that one
practices when one has phronesis?
Phronesis is about the self
• What makes phronesis different from expertise, and
even expertise in how to live the good life (if there is
such a thing), is, I suggest, the particular object or
target involved.
• The particular target of phronesis is one's self -oneself in various but very particular situations.
"Practical wisdom also is identified especially with
that form of it which is concerned with a man
himself-with the individual; and this is known by
the general name 'practical wisdom'" (1141b28).
– One might claim that to have phronesis is to have expert practical
knowledge and skill in how to live the good life in the company
of others.
– One might claim that the expertise of an expert in human affairs,
for example, a marriage councilor, is really a kind of phronesis.
– The problem that doesn't go away: one might remain an expert
marriage councilor and for whatever perverse reasons,
intentionally deliver advice that will destroy the marriage of your
clients.
– The expertise used to improve lives, which may be the same as
that used to destroy lives, simply cannot be equated with
phronesis. Rather, phronesis is precisely the thing that would
prevent you from using your expertise for bad purposes.
• The target is not one’s self as an object; but oneself
as situated agent, moral practitioner.
• While there are textbooks on different areas
of expertise there is no textbook on one's
own self or on the unique situations in which
one finds oneself.
• Regardless of who you are, or the kind of person
you are, you can read the text book on X, and then
practice, practice, practice to the point that you
become the intuitive expert on X. This gives you a
skill and makes you an expert, but it doesn’t
necessarily change the kind of person you are.
• In contrast, for Aristotle, it is not the character of
the actions that make them virtuous, but the
character of the agent:
“The agent also must be in a certain condition
when he does them; in the first place he must have
knowledge, secondly he must choose the acts, and
choose them for their own sakes, and thirdly his
action must proceed from a firm and unchangeable
character. (1105a31).”
• Phronesis, in contrast to expertise, involves
making decisions about my own actions, and what
is genuinely best for the situation defined as
including myself.
8
• Phronesis involves a practical knowledge about
oneself from the inside out, and from within the
particular situation in which one exists.
• E.g., an expert in oil drilling shares knowledge
of (and a know-how about) oil drilling
with all other experts – it is in some sense
the same knowledge, and in this regard, it
has an objective character.
• Phronesis, however, does not involve an objective
perspective that would be just as appropriate for
anyone – and indeed, because we need to conceive of
the moral situation as including the moral agent, no
two situations are the same.
• But, the objection continues, one can say that expert
practice may depend on my ability to empathize
with my clients, or to put myself in their place. In
this regard the self is not left out of expertise.
• Okay, but still, as empathy is normally
characterized, it does not involve confusing
ourselves with the other. We know that it is not
ourselves in that particular situation.
• Our ability to empathize with the other's
perspective, of course, depends on precisely the
capacities discussed in the previous section, which
condition both expertise and phronesis, and are
doubly relevant to the kinds of expertise that
involve human affairs
• One way to express the difference between expertise and
phronesis is to recall Gabriel Marcel's (1935, 1962)
distinction, between a problem and a mystery.
– A problem is something that can be defined in
objective terms, in a complete way and as having a
definite answer (whether or not we know the answer
yet) -- it can be treated as objective and complete
because it is entirely external to the person who has to
deal with it.
• For example, a machine breaks down and a
mechanic has the expertise to fix it in such a way
that she does not become existentially involved in
the machine. She may have the right intuition of
what is wrong simply by listening to it.
– A mystery is not something that one can gain
objective distance from, because it involves one's own
existence.
A possible objection
• Phronesis is much more like the kind of expertise
had by a marriage councilor than by an oil drilling
expert (techne); my expertise as a marriage
councilor can not be easily summarized in a set of
rules or propositional statements,it requires
intuition, etc.
• Okay, but an expert marriage councilor will attempt
to see the entire situation in objective terms, and at
the same time not get personally involved -- it’s not
her marriage after all -- and the situation is really
someone else’s situation.
• This means that even if phronesis is about the
self, we are not entirely alone in our phronesis.
• The basis for the practical knowledge of oneself
required for phronesis is found precisely in the
embodied and intersubjective capacities that we
discussed before.
• Although this is a know-how gained from the
inside out, it is not a purely subjective knowledge,
since from the inside (endogenously), and from
birth, we are intersubjectively involved with
others, and our self is shaped by these encounters.
• Being in love or caring for another is a mystery rather
than a problem. When you attempt to analyze it in an
objective way, it ceases to be what it is. To love
another person is not a form of expertise.
• One's own body is a mystery rather than a problem
(even if from your physician's perspective it can be a
problem), since one is unable to step outside of
it in order to gain a completely objective
(scientific or technical) perspective on it.
– I can be an expert on the human body, and I
can even take up an objective perspective on
my body and treat it as a problem. But on Marcel's
definition, I can never treat my own body as a mere
objective body, or completely as a problem, insofar
as I am my body and there is always something in
excess of that which I can objectify.
9
• Using this distinction, I suggest that one way to
capture the difference between expertise and
phronesis is this:
– expertise is directed at problems
– phronesis is directed at a mystery
• The mystery at which phronesis is directed is one's
own situated self.
• Moreover, it is impossible to have expertise about
one's own self, since the self, or who I am, is
always more than I can objectify, and is constantly
involved in new situations that potentially change
it.
Moral personhood (Moral agency) revisited
Let's revisit the conditions for moral personhood as they
were outlined by Dennett.
(1) The kind of rationality involved in moral
personhood is not the sort that can be captured in
computational models, but the kind of practical
rationality that is involved in phronesis. Even if it
were possible (and Dryfus is right that it is not) to
reduce expertise to a set of rules and a disembodied
body of propositional knowledge, as Lenat suggests,
this is certainly not possible for phronesis. The kind
of rationality required for phronesis is at once
embodied and intersubjective, and we begin to pick
it up from our earliest encounters with others.
(2) We must be able to take the intentional stance toward
the person who would be a moral agent.
• Dennett cites Strawson on this. He “identifies the
concept of a person as ‘the concept of a type of entity
such that both predicates ascribing states of
consciousness and predicates ascribing corporeal
characteristics’ are applicable” (Dennett 1978).
• This says something about all of us, both “ascribers”
This capacity is clearly a condition for the development
and “ascribees.” Our ability to do this, which is the
of phronesis to the extent that it is the beginning point
to recognizeofan
entitythat
in the
environment
for ability
an understanding
others
involves
our ownas
another
person,
and thus alsoand
to be
a personprovides
to whom
motoric
(action)
reverberations,
therefore
are made (because
interaction
goes two
the ascriptions
basis for understanding
and forming
our own
ways – see condition 4), exists from infancy.
intentions.
(3) In the intersubjective, and richly affective
interaction that characterizes primary
intersubjectivity, it is clearly possible even for
infants to be the target of a certain
emotional attitudes -- and, in a certain
manner, demand our attention, if not our
respect.
(4) It is also clear that normal infants (and some nonhuman animals) emotionally reciprocate --emotional
interaction shapes our sense of self in an
intersubjective mirroring that forms a necessary basis
of social life and the possibility of moral practice
(5) Dennett's fifth condition involves the ability to
communicate with others -- not just the verbal or
signed transference of propositional knowledge, but
the capacity to implicitly pick up and understand the
non-verbal expression of others.
– The communication of intentions and feelings is
accomplished not simply in verbal (propositional)
discourse, but through embodied and perceptually
informed interactions -- something that reverberates
in an intuitive way in one's own action system, and
as such forms the basis of the intuitive sense of
what the other expects or approves.
– We pick up on emotions and this makes empathy
possible.
– Gallese (2001): the neural mirror system includes emotion.
– Scheler (1970): we see the joy in the face of the other.
– Hobson (2002): infants look to their mothers' gestures for
reassurance when they encounter a new object or situation
These emotional interactions are clearly part of what
Aristotle identified as the source of phronesis; not
only being with and observing others, but acting with
their emotional confirmation or caution, and coming
to know what actions are good and what ones are bad.
10
(6) Self-consciousness. If, as seems reasonable, a
higher-order reflective self-consciousness is
necessary for making explicit moral decisions, it is
not clear that on Aristotle's conception of phronesis
responsible moral action always involves this kind of
self-consciousness.
• Phronesis, to the extent that it involves something of
a second nature, often leads to action that is
intentional, but also close to automatic. The good
person intuitively knows what to do and does it
without much deliberation. Such intentional action,
however, is not done unknowingly.
• Intentional action is always accompanied by a prereflective self-consciousness – a self-awareness that is
implicit to experience itself. Thus, the person with
phronesis knows what they are doing on an implicit level -and this is best made explicit not by a higher-order,
detached, reflective self-consciousness, but by a situated
reflection that considers action on the highest pragmatic
level of discourse.
• To explain what makes my action appropriate may require
further self-conscious deliberation, and may be difficult to
express or justify, but prior to that the person with
phronesis has a pre-reflective self-surety about the
rightness of the action.
• In principle, this is not the kind of thing that could be
instanciated in a disembodied non-situated machine or
brain in a vat.
Conclusions
• It is possible, then, to define moral personhood by
the capacity for phronesis.
• On this definition, moral personhood depends on an
embodied and intersubjective existence in which the
rationality at stake is practical rather than theoretical,
and is characterized by a pre-reflective selfconsciousness emotionally informed by shared
reverberations endogenous to our own action
systems.
• One's capacity to act as a moral agent (that is, to act
morally or immorally, responsibly or irresponsibly)
is just this capacity to act on an intuitive insight into
one's own self in a way that is not divorced from but
rather fully implicated in our relations with others.
11
cognitive science. University of
Jyväskylä, Finland (September
2006).
Outline
1. Conditions for moral personhood (Dennett)
2. In an attempt to deepen our understanding of
these conditions: An Aristotelian argument:
moral personhood depends on both embodied
and social aspects of human existence which
give us the capacity for phronesis.
3. The Dreyfus-Collins debate on expertise
4. Expertise and phronesis
5. Revisiting the conditions for moral personhood
Moral personhood and phronesis
Shaun Gallagher
Philosophy and Cognitive Sciences
University of Central Florida
[email protected]
Moral personhood (moral agent)
The capacity for moral action and responsibility
• Dennett (1978): for an entity to have the status of
moral personhood it must meet six conditions
1. Rationality
2. We must be able to take the intentional stance
toward it -- we treat it as if it had a mental life
3. Recognition by others -- we take a certain attitude
toward it
4. Capacity for reciprocity
5. Communicative capacity
6. Capacity for self-consciousness, defined as a
sophisticated reflective (second-order) mental
state.
• An Aristotelian argument: moral personhood depends
on both embodied and intersubjective aspects of
human existence
– Someone is a moral agent, that is, they have the status of
moral personhood and have the capacity to do the right or
wrong action, only if they are capable of having (practicing)
phronesis (practical wisdom).
– The person who is capable of moral action/ responsibility
(the moral agent) is fully embodied (is more than a brain),
and this embodiment shapes the very nature of moral
possibilities.
– To be capable of having (practicing) phronesis depends on
an intersubjective existence -- and specifially a form of
“endogenous intersubjectivity”
– In light of these considerations, we will come back to
Dennett's six conditions as conditions for the possibility of
having phronesis.
1.
2.
3.
4.
5.
6.
Rationality
Intentional stance
Recognition
Reciprocity
Communication
Self-consciousness
• 1-5 are necessary conditions for selfconsciousness.
• Young children are not capable of higher-order
self-consciousness (Dennett, Frankfort, Wilkes).
• Conditions 2-5 explicitly and importantly involve
dimensions of intersubjectivity, although, for
Dennett, the precise nature of these social
dimensions is still an open question.
• Dennett (e.g., 1982, 1991) suggests that a brain in a vat
or a computer might be able to attain this kind of selfconsciousness. This implies that these conditions do not
depend on embodiment in any strong sense,
and it raises questions about the notion of
intersubjectivity involved.
Starting point: a debate on expertise
• I set aside the question of whether phronesis
is a kind of ethical expertise until later.
• But getting a good understanding of what
constitutes expertise will throw some light
on the nature of phronesis.
• A debate between Dreyfus and Harry
Collins (sociologist, Cardiff) on expertise.
1
Dreyfus: phenomenological
account of the embodied
basis for expertise
• Expertise is an instance of embodied human
performance -- on a continuum with basic life-world
practices.
“We are all experts at many tasks and our everyday
coping skills function smoothly and transparently so as
to free us to be aware of other aspects of our lives
where we are not so skillful” (Dreyfus and Dreyfus,
1990).
Phenomenological background
• Merleau Ponty’s notion of the lived body and the
concepts “intentional arc” and “maximal grip” -practiced activities controlled by body schematic
adjustments that operate as the basis for expert
practices
• Heidegger’s emphasis on pragmatic contexts as our
primary way of engaging in the world -- just those
contexts in which expertise is practiced.
• Higher order reflection on how one does what one
does only occurs when things or procedures fail to
work effectively (Heidegger’s hammer)
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
Monday, May 8, 2006
• Dubner and Levitt: The Birth-Month Soccer
Anomaly: elite soccer players are more likely to
have been born in the earlier months of the year than
in the later months.
• Could astrology explain expertise in soccer?
• Answer: No. Research by Anders Ericsson at Florida
State University over the past 30 years has
demonstrated that “deliberate practice” is what
creates expertise: setting specific goals, obtaining
immediate feedback and concentrating as much on
technique as on outcome.
• Dreyfus’s target: any account of expertise given
in terms of purely cognitive, propositional, or
representationalist terms -- expert knowledge as a
set of explicable rules or as a form of theoretical
knowledge.
• Expert skills -- a matter of of “knowing how”
rather than “knowing that”
• Knowing how (vs propositional knowledge)
involves embodied practice rather than cognitive
deliberation -- the exercise of skills of which one
cannot fully give an account or full articulation
(e.g., chicken sexing)
• Expertise is a matter of intuition, not
intellectualization: “Action becomes easier and less
stressful [as the expert] simply sees what needs to be
done rather than using a calculative procedure to
select one of several possible alternatives” (Dreyfus
2001).
• The expert not only sees what needs to be done, but
also how to achieve it without deliberation,
immediately -- non-reflectively recognizing new
situations as similar to previously encountered ones,
and intuiting “what to do without recourse to rules” -he or she recognizes important features as contextually
sensitive (Dreyfus and Dreyfus 1986).
• Expertise is in the practice, and the expert is a
practitioner.
• Expert knowledge is not third-person propositional
knowledge; it is not theoretical knowledge. To think of
it in this way distorts the first-person embodied
experience -- the form of life -- that constitutes
expertise.
• Expertise is not captured in a set of rules; it is not a
technical kind of knowledge in this sense. Dreyfus
thus objects to the artificial, computational creation of
expertise -- vs e.g., Doug Lenat
2
Doug Lenat (the Cyc project)
• Expertise is reducible to rational
(computational) rules and propositional
knowledge -- it can be rationally reconstructed as
primarily a mental intellectual phenomenon
• Lenat argues for a kind of expertise that Dreyfus
rules out -- an expert who is rich in propositional
facts about X, but has never done or experienced
X.
• The expert need not be a practitioner. This kind
of expertise can be disembodied and instanciated
in a computer.
Harry Collins:
The social explanation of expertise
• Vs both Dreyfus’s emphasis on
embodiment and Lenat’s emphasis on
propositional knowledge
• Dreyfus’s explanation remains asocial.
In an account that takes social dimensions seriously,
expertise is thought of as “distributed” -- embedded
in social practices, particular settings (laboratories
and social networks), standardized in technologies,
and promoted in specific rhetorical means of
recruiting professional experts (Mialet 1999)
For Collins, expertise constituted in language or languagegames, and does not depend on embodied practices.
– He cites Wittgenstein’s example of a talking lion -- an animal
that divides up the world in a different way.
– For example, for us, a chair is a different kind of affordance
(Gibson) than for the lion -- we have a body that can use a chair
to sit -- so ‘chair’ if in the lion’s vocabulary, would mean
something different.
– Collins claims that if the lion hung around and communicated
with humans long enough, the lion would come to know the
meaning of chair, despite differences in embodiment.
“the language of a community embodied in one way can
be acquired by individuals with different shaped
bodies, and who, therefore, cannot participate in the
activities of that community.” -- therefore, bodies
contribute little to the acquisition of expertise.
Lenat points to the case of Madeleine as an example of
purely intellectual and disembodied expertise.
• Oliver Sacks (1985): Madeleine was born
blind and disabled. She became so dependent
on others that she was even unable to use
her hands to read brail or for any other
purpose -- she was utterly inactive
throughout her early life
• Nevertheless, Madeleine learned about the
world from books read to her by others. She
was able to speak “freely indeed eloquently
... revealing herself to be a high-spirited
woman of exceptional intelligence and
literacy” She was “of exceptional
intelligence and literacy, with an
imagination filled and sustained, so to
speak, by the images of others, images
conveyed by language, the word.”
• Lenat takes this to mean that Madeleine had
propositional but not embodied knowledge of
• Collins (2004) also argues for a kind of expertise that
Dreyfus rules out -- a kind of expert knowledge that
someone may attain via social learning procedures -specifically, linguistic and communicative processes:
interactive expertise vs contributory expertise
• vs Lenat, this expertise is not based on propositional
knowledge -- it is the result of a socialization process that
requires conversational interaction with another
• Collins himself is example.
– A sociologist of science learns gravitational wave physics-- not
by learning the propositional science itself, and not by doing the
science, but by hanging out with the scientists, conversing with
them, seeing what they do, etc. -- to the point that he becomes
proficient in the language of gravitational wave physics and not
only can hold up his side of a conversation, but make intelligent
suggestions -- even though he is not a practitioner.
• Collins offers a different
interpretation of the case of
Madeleine to show that through
communication and a passive
reception of information (she
attained much of her knowledge by
having someone read to her) -thus, through the medium of the
written [heard] and spoken word -Madeleine could easily be an expert
on X.
“Madeleine had a minimal `body' with
almost no ability to take part in
the normal activities of the
members of the surrounding
3
The criticism of Dreyfus for ignoring the social aspects
of expertise has been extended in a slightly different
way:
• Both Iris Young (1998) and Maxine SheetsJohnstone (2000) criticize Dreyfus for assuming
that the body which acquires skill has no relevant
biography, gender, race, or age. They emphasize
the cultural embeddedness of the body.
• This approach nicely explicates the external
limitations on expertise that cultural factors impose,
and has much to say about social-economic-political
factors that limit embodied processes, but it does not
provide any positive account of how expertise
develops except by adding these external limitations
to Dreyfus’s account.
• Selinger and Crease (2002) point out that Dreyfus
does have a place for the idea that “cultural styles”
affect how skills are learned (Dreyfus, 2000), but that
this notion is simply not developed.
– “From Dreyfus’s perspective, one develops the
affective comportment and intuitive capacity of
an expert solely by immersion into a practice;
the skill-acquiring body is assumed to be able,
in principle at least, to become the locus of
intuition without influence by forces external to
the practice in which one is apprenticed.”
In summary:
• Lenat rules out any important role for embodiment
and emphasizes a cognitive - computational model
consistent with traditional views of expertise as a
mentalistic or intellectual phenomenon.
• Collins also rules out any important role for
embodiment and emphasizes a socially contextualized
model of expertise. This model is primarily a
linguistic-communicative one.
• Dreyfus critiques traditional and computational models,
but ignores social/intersubjective dimensions and
emphasizes pre-reflective embodied skills as the
basis for expertise.
Neuroscience of Resonance systems:
Brain areas associated with my own action are also
active during imitation and observation of others
(Gallese, Jeannerod)
Shared neural representations (mirror system)
– action
– observation of another’s action
– imaginative enactment of action Observation
– preparing to imitation
Imagination
– Neuroscience: shared representations and mirror neural
systems
– Developmental psychology: primary and secondary
intersubjectivity (Trevarthan)
– Phenomenology: embodied intersubjective experience that
involves both perceptual processes and pragmatic contexts.
Developmental Psychology
Trevarthan’s notions of primary and secondary
intersubjectivity
Primary intersubjectivity: the infant’s ability to
perceive meaning in the other’s gestures, movements,
facial expressions, intentional actions.
Action
Preparing to
imitate
Are these not the embodied apprenticeship
experiences that build expertise -- and
doesn’t this make them already social?
The interaction approach
Expertise (including our everyday expertise and coping
skills) requires embodied and intersubjective practices,
where embodied practices are always intersubjective
and intersubjective practices are always embodied.
This connection between embodiment and
intersubjectivity can be found in infancy and continues
as the basis for intersubjective understanding even in
adults -- and it directly ties in to questions about the
acquisition of expertise.
Infants engage and interact with
others -- from neonate imitation
to emotional mirroring, to eyetracking, to infant’s ability to
parse intentions of others -perceptual access to others (vs
theory or simulation)
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
4
Phenomenology of primary intersubjectivity
“The very first of all cultural objects, and the one
by which all the rest exist, is the body of the
other person as the vehicle of a form of
behavior” (Merleau-Ponty PhP 348). “I live in
the facial expressions of the other, as I feel him
living in mine … (Merleau-Ponty, N 218).
“… the perception of others is anterior to, and the
condition of, such observations [simulations,
theoretical stances], the observations do not
constitute the perception. … between this
phenomenal body of mine and that of another as I
see it from the outside, there exists an internal
relation which causes the other to appear as the
completion of the system” (352)
Secondary intersubjectivity: Around the age of 1 year,
the infant goes beyond person-to-person immediacy
and enters contexts of shared attention -- shared
situations -- learning what things mean and what they
are for.
“The defining feature of secondary intersubjectivity
is that an object or event can become a focus
between people. Objects and events can be
communicated about. … the infant’s interactions
with another person begin to have reference to the
things that surround them” (Hobson, 2002, p. 62).
– Studies of shared attention in infancy.
– Infants understand the failed intentions of others
(Meltzoff 1995).
Infants at 18 months imitate to completion the
unfulfilled intentions of humans; they do not
imitate the movements of the mechanical
device.
Phenomenology of secondary intersubjectivity
• Trevarthan’s developmental concept of
secondary intersubjectivity is already
foreshadowed by phenomenological analyses of
Heidegger and Gurwitsch,
• Understanding the meaning of an object or an
instrument is dependent on our involvement in
pragmatic contexts
• In the pragmatic we also find intersubjectivity -other people appear in meaningful roles within
pragmatic contexts.
• Aron Gurwitsch (following Heidegger’s
analysis of equipment and circumspective
engagement with the surrounding environment,
and the larger action contexts of human
existence)
– our understanding of the other’s expressive
movements depends on meaningful
instrumental/pragmatic contexts
– Things and situations provide scaffolds for
understanding the actions of others -- and in
those pragmatic contexts we see and come to
learn and imitate what they do
5
• Dreyfus takes up Heidegger’s analysis of
equipment, and his account of expertise follows
this line of thought -- expertise is embodied and
pragmatic -- but what about intersubjectivity?
• Clearly, the embodied-pragmatic expertise (of the
sort described by Dreyfus) is dependent on
intersubjective interaction.
• To add a corrective to Heidegger-GurwitschDreyfus: It is not that our primary access to
others is through pragmatic contexts; it's that
our access to pragmatic contexts is primarily
through others.
Conclusions:
• In opposition to Collins and Lenat: It is not feasible or
necessary to eliminate embodiment in order to make room for
social and/or even what appear to be “purely” cognitive
dimensions of expertise (cf. embodied cognition; Mark Johnson)
• As a supplement to Dreyfus: Nor is it necessary or feasible to
eliminate intersubjective dimensions in order to make room for the
role played by embodiment
• To the extent that expertise depends on the grasping of the
intentions and meanings of others -- on acting, observing others
act, imagining ourselves act, imitating -- it also depends on
embodied resonance processes that are simultaneously
intersubjective.
• To suggest that expertise depends only on embodiment without the
intersubjective dimension (Dreyfus), or only on the social without
embodiment (Collins) is to give an incomplete account of
expertise.
Aristotle : phronesis
• The good person, the person with
phronesis, like the person with expertise,
sees what to do in an immediate way, and
does the good thing in a close to automatic
way, as if it were second nature.
• But, in regard to acquisition, Aristotle
takes a position that is similar to Collins on
expertise: one acquires phronesis through a
good upbringing, and this means hanging
around with the right people – good people
who provide good examples of good
actions.
• You don’t get phronesis by taking an
ethics course (vs Lenat).
Merleau-Ponty
“Insofar as I have sensory functions … I am
already in communication with others ….
No sooner has my gaze fallen upon a living
body in the process of acting than the
objects surrounding it immediately take
on a fresh layer of significance”
Intersubjective contexts cannot be reduced to or
encompassed by purely instrumental ones -- the social
aspects of experience transcend the purely pragmatic.
• Evidence from recovery after brain injury (Gallagher and
Marcel 1999)
• Evidence from studies of gesture in cases of
deafferentation (Cole, Gallagher, McNeill 2001)
Return to the question of moral agency/moral
personhood and phronesis
• Is phronesis equivalent to expertise? Is the acquisition
of phronesis best explained on a skills-acquisition
model?
• Phronesis is like expertise in certain ways. Following
along the same lines of the debate about expertise,
phronesis is not equivalent to theoria – that is, a
theoretical knowledge that is propositional and learnable
in a purely intellectual way. Phronesis cannot be
programmed into a computer.
• It is, rather, as Dreyfus says of expertise, a kind of
"know-how," but is nonetheless not reducible to a set of
rules, and should be distinguished from techne.
• Yet this is not sufficient.
– On the Collins model of expertise, there is no guarantee
that one's social interaction with experts will necessarily
make one an expert.
– Plato's complaint in the Meno – a son who is raised by
good parents and given the best education amongst the
best of society still may turn out bad.
• To attain phronesis, one must also act and interact in
a good way. It would not be enough simply to
watch, or to converse with good people. One needs
to imitate them, to act as they do and to do the kinds
of things that they do: phronesis, like expertise, is in
the practice -- Aristotle = Dreyfus + Collins
6
More on acquisition
• In every case (for expertise and for phronesis) Dreyfus
outlines a multi-step acquisition process: novice to
advanced beginner to competence to proficiency to
expert.
• In each case the novice stage starts like this: “Normally,
the instruction process begins with the instructor
decomposing the task environment into context-free
features that the beginner can recognize without benefit
of experience. The beginner is then given rules …”
(Dreyfus and Dreyfus 2004).
• Seemingly we start with rules and/or theory and work our
way out of dependency on these mentalistic beginnings to
non-mentalistic expertise through practice. This may
work for learning to drive or play chess. But what about
learning first language or learning to walk?
• How do we gain phronesis? Aristotle: by hanging out
with the right people. Learning from example.
Imitating. Picking up practices.
Aristotle is Trevarthan, not Allison Gopnik
• Dreyfus seems to be a friend of theory in the
acquisition/developmental stages (although he also talks
about the importance of practice)
“On analogy with chess and driving, it would seem that
the budding ethical expert would learn at least some of
the ethics of his or her community by following strict
rules, would then go on to apply contextualized maxims,
and, in the highest stage, would leave rules and
principles behind and develop more and more refined
spontaneous ethical responses” (Dreyfus 2004).
The difference between phronesis and expertise
• Is phronesis "ethical expertise”?
• Just as Aristotle makes a distinction between phronesis
and cleverness, as well as a distinction between virtuous
action and techné (1105a22), a similar distinction should
be clearly made between phronesis and expertise.
“There is a faculty which is called cleverness; and this is
such as to be able to do the things that tend towards the
mark we have set before ourselves, and to hit it. Now if
the mark be noble, the cleverness is laudable, but if the
mark be bad, the cleverness is mere smartness; hence we
call even men of practical wisdom clever or smart.
Practical wisdom is not this faculty, but it does not exist
without this faculty. … practical wisdom is to cleverness
-- not the same, but like it” …. “it is impossible to be
practically wise without being good.” (N. Ethics, VI)
More on acquisition
• Language: we learn the rules (grammar) after we learn to
speak; Walking -- are there any rules? -- these are things
we learn by pure practice.
• How do we learn our everyday coping skills? Not by
working our way through theory or a set of rules.
• How do we come to understand others and gain our
“people skills”-- not by theory -- we are not given rules.
We are given people and we start to interact with them and
imitate them (primary and secondary intersubjectivity). In
the longer term we learn by narratives. Not theories, not
rules, but pure doings.
• Driving? Rules or practices?
• But Dreyfus is not a friend of theory in
regard to the actual practice of expertise or
in regard to phronesis.
• Vs Habermas and Benhabib’s implicit
moral theory, ethical cognitivists, relying on
rules and principles.
• Pro Gilligan in emphasizing care over the
over-rational justice.
Phronesis according to Dreyfus
• Dreyfus is right and properly Aristotelian in
characterizing phronesis as non-mentalistic and as
not relying on rules or maxims; the phronimos copes
case-by-case, attending to differences in situations.
• But this is not the complete picture of phronesis –
something more is required. This goes beyond the
question of the educational backdrop.
• Consider the following characterizations made by
Dreyfus (2004), following Heidegger.
7
• The phronimos is the “master of his or her culture’s
practices”
• Ethical experts: “experts capable of responding
appropriately to a wide range of interpersonal
situations in their culture. Such social experts could be
called virtuosi in living …. This is obviously
Aristotle’s phronimos.”
• Following Heidegger: “people have skills for coping
with equipment, other people, and themselves”
(Dreyfus 2004)
• Is phronesis reducible to people skills or virtuosity in
interpersonal dealings? Is virtuosity equivalent to
virtue? Just as I can be a clever criminal, I can be a
virtuoso in selling used cars, managing an organization,
convincing people to vote for me, managing a
classroom.
• But none of this requires that I do the good or right
thing.
• A person could know and have the know how for
exactly the right thing to do, to act morally (= ethical
expertise), and be inclined to act that way, but decide
not to act ethically, and use her knowledge to act in a
way that is not ethical.
• Such a person might have ethical expertise, but would
not have phronesis.
• What exactly is it that makes phronesis so different
from ethical expertise -- why not say, for example, that
phronesis is expertise in what constitutes the moral life?
• On the Dreyfus model of expertise, one would have to
practice one's skill. But what precisely is the skill that one
practices when one has phronesis?
Phronesis is about the self
• What makes phronesis different from expertise, and
even expertise in how to live the good life (if there is
such a thing), is, I suggest, the particular object or
target involved.
• The particular target of phronesis is one's self -oneself in various but very particular situations.
"Practical wisdom also is identified especially with
that form of it which is concerned with a man
himself-with the individual; and this is known by
the general name 'practical wisdom'" (1141b28).
– One might claim that to have phronesis is to have expert practical
knowledge and skill in how to live the good life in the company
of others.
– One might claim that the expertise of an expert in human affairs,
for example, a marriage councilor, is really a kind of phronesis.
– The problem that doesn't go away: one might remain an expert
marriage councilor and for whatever perverse reasons,
intentionally deliver advice that will destroy the marriage of your
clients.
– The expertise used to improve lives, which may be the same as
that used to destroy lives, simply cannot be equated with
phronesis. Rather, phronesis is precisely the thing that would
prevent you from using your expertise for bad purposes.
• The target is not one’s self as an object; but oneself
as situated agent, moral practitioner.
• While there are textbooks on different areas
of expertise there is no textbook on one's
own self or on the unique situations in which
one finds oneself.
• Regardless of who you are, or the kind of person
you are, you can read the text book on X, and then
practice, practice, practice to the point that you
become the intuitive expert on X. This gives you a
skill and makes you an expert, but it doesn’t
necessarily change the kind of person you are.
• In contrast, for Aristotle, it is not the character of
the actions that make them virtuous, but the
character of the agent:
“The agent also must be in a certain condition
when he does them; in the first place he must have
knowledge, secondly he must choose the acts, and
choose them for their own sakes, and thirdly his
action must proceed from a firm and unchangeable
character. (1105a31).”
• Phronesis, in contrast to expertise, involves
making decisions about my own actions, and what
is genuinely best for the situation defined as
including myself.
8
• Phronesis involves a practical knowledge about
oneself from the inside out, and from within the
particular situation in which one exists.
• E.g., an expert in oil drilling shares knowledge
of (and a know-how about) oil drilling
with all other experts – it is in some sense
the same knowledge, and in this regard, it
has an objective character.
• Phronesis, however, does not involve an objective
perspective that would be just as appropriate for
anyone – and indeed, because we need to conceive of
the moral situation as including the moral agent, no
two situations are the same.
• But, the objection continues, one can say that expert
practice may depend on my ability to empathize
with my clients, or to put myself in their place. In
this regard the self is not left out of expertise.
• Okay, but still, as empathy is normally
characterized, it does not involve confusing
ourselves with the other. We know that it is not
ourselves in that particular situation.
• Our ability to empathize with the other's
perspective, of course, depends on precisely the
capacities discussed in the previous section, which
condition both expertise and phronesis, and are
doubly relevant to the kinds of expertise that
involve human affairs
• One way to express the difference between expertise and
phronesis is to recall Gabriel Marcel's (1935, 1962)
distinction, between a problem and a mystery.
– A problem is something that can be defined in
objective terms, in a complete way and as having a
definite answer (whether or not we know the answer
yet) -- it can be treated as objective and complete
because it is entirely external to the person who has to
deal with it.
• For example, a machine breaks down and a
mechanic has the expertise to fix it in such a way
that she does not become existentially involved in
the machine. She may have the right intuition of
what is wrong simply by listening to it.
– A mystery is not something that one can gain
objective distance from, because it involves one's own
existence.
A possible objection
• Phronesis is much more like the kind of expertise
had by a marriage councilor than by an oil drilling
expert (techne); my expertise as a marriage
councilor can not be easily summarized in a set of
rules or propositional statements,it requires
intuition, etc.
• Okay, but an expert marriage councilor will attempt
to see the entire situation in objective terms, and at
the same time not get personally involved -- it’s not
her marriage after all -- and the situation is really
someone else’s situation.
• This means that even if phronesis is about the
self, we are not entirely alone in our phronesis.
• The basis for the practical knowledge of oneself
required for phronesis is found precisely in the
embodied and intersubjective capacities that we
discussed before.
• Although this is a know-how gained from the
inside out, it is not a purely subjective knowledge,
since from the inside (endogenously), and from
birth, we are intersubjectively involved with
others, and our self is shaped by these encounters.
• Being in love or caring for another is a mystery rather
than a problem. When you attempt to analyze it in an
objective way, it ceases to be what it is. To love
another person is not a form of expertise.
• One's own body is a mystery rather than a problem
(even if from your physician's perspective it can be a
problem), since one is unable to step outside of
it in order to gain a completely objective
(scientific or technical) perspective on it.
– I can be an expert on the human body, and I
can even take up an objective perspective on
my body and treat it as a problem. But on Marcel's
definition, I can never treat my own body as a mere
objective body, or completely as a problem, insofar
as I am my body and there is always something in
excess of that which I can objectify.
9
• Using this distinction, I suggest that one way to
capture the difference between expertise and
phronesis is this:
– expertise is directed at problems
– phronesis is directed at a mystery
• The mystery at which phronesis is directed is one's
own situated self.
• Moreover, it is impossible to have expertise about
one's own self, since the self, or who I am, is
always more than I can objectify, and is constantly
involved in new situations that potentially change
it.
Moral personhood (Moral agency) revisited
Let's revisit the conditions for moral personhood as they
were outlined by Dennett.
(1) The kind of rationality involved in moral
personhood is not the sort that can be captured in
computational models, but the kind of practical
rationality that is involved in phronesis. Even if it
were possible (and Dryfus is right that it is not) to
reduce expertise to a set of rules and a disembodied
body of propositional knowledge, as Lenat suggests,
this is certainly not possible for phronesis. The kind
of rationality required for phronesis is at once
embodied and intersubjective, and we begin to pick
it up from our earliest encounters with others.
(2) We must be able to take the intentional stance toward
the person who would be a moral agent.
• Dennett cites Strawson on this. He “identifies the
concept of a person as ‘the concept of a type of entity
such that both predicates ascribing states of
consciousness and predicates ascribing corporeal
characteristics’ are applicable” (Dennett 1978).
• This says something about all of us, both “ascribers”
This capacity is clearly a condition for the development
and “ascribees.” Our ability to do this, which is the
of phronesis to the extent that it is the beginning point
to recognizeofan
entitythat
in the
environment
for ability
an understanding
others
involves
our ownas
another
person,
and thus alsoand
to be
a personprovides
to whom
motoric
(action)
reverberations,
therefore
are made (because
interaction
goes two
the ascriptions
basis for understanding
and forming
our own
ways – see condition 4), exists from infancy.
intentions.
(3) In the intersubjective, and richly affective
interaction that characterizes primary
intersubjectivity, it is clearly possible even for
infants to be the target of a certain
emotional attitudes -- and, in a certain
manner, demand our attention, if not our
respect.
(4) It is also clear that normal infants (and some nonhuman animals) emotionally reciprocate --emotional
interaction shapes our sense of self in an
intersubjective mirroring that forms a necessary basis
of social life and the possibility of moral practice
(5) Dennett's fifth condition involves the ability to
communicate with others -- not just the verbal or
signed transference of propositional knowledge, but
the capacity to implicitly pick up and understand the
non-verbal expression of others.
– The communication of intentions and feelings is
accomplished not simply in verbal (propositional)
discourse, but through embodied and perceptually
informed interactions -- something that reverberates
in an intuitive way in one's own action system, and
as such forms the basis of the intuitive sense of
what the other expects or approves.
– We pick up on emotions and this makes empathy
possible.
– Gallese (2001): the neural mirror system includes emotion.
– Scheler (1970): we see the joy in the face of the other.
– Hobson (2002): infants look to their mothers' gestures for
reassurance when they encounter a new object or situation
These emotional interactions are clearly part of what
Aristotle identified as the source of phronesis; not
only being with and observing others, but acting with
their emotional confirmation or caution, and coming
to know what actions are good and what ones are bad.
10
(6) Self-consciousness. If, as seems reasonable, a
higher-order reflective self-consciousness is
necessary for making explicit moral decisions, it is
not clear that on Aristotle's conception of phronesis
responsible moral action always involves this kind of
self-consciousness.
• Phronesis, to the extent that it involves something of
a second nature, often leads to action that is
intentional, but also close to automatic. The good
person intuitively knows what to do and does it
without much deliberation. Such intentional action,
however, is not done unknowingly.
• Intentional action is always accompanied by a prereflective self-consciousness – a self-awareness that is
implicit to experience itself. Thus, the person with
phronesis knows what they are doing on an implicit level -and this is best made explicit not by a higher-order,
detached, reflective self-consciousness, but by a situated
reflection that considers action on the highest pragmatic
level of discourse.
• To explain what makes my action appropriate may require
further self-conscious deliberation, and may be difficult to
express or justify, but prior to that the person with
phronesis has a pre-reflective self-surety about the
rightness of the action.
• In principle, this is not the kind of thing that could be
instanciated in a disembodied non-situated machine or
brain in a vat.
Conclusions
• It is possible, then, to define moral personhood by
the capacity for phronesis.
• On this definition, moral personhood depends on an
embodied and intersubjective existence in which the
rationality at stake is practical rather than theoretical,
and is characterized by a pre-reflective selfconsciousness emotionally informed by shared
reverberations endogenous to our own action
systems.
• One's capacity to act as a moral agent (that is, to act
morally or immorally, responsibly or irresponsibly)
is just this capacity to act on an intuitive insight into
one's own self in a way that is not divorced from but
rather fully implicated in our relations with others.
11
