4' The Rise of Expressivism

1
4. The Rise of Expressivism

• Antti Kauppinen
• PY 3702
• Martinmas Semester 2007

2
I What is Expressivism?
3
I.1 Expressivism, Non-Cognitivism, Emotivism...

• I will use expressivism as the general term for
  the sort of theories on which the primary
  constituents of moral judgments are attitudes
  rather than beliefs
• This sort of views used to be called
  non-cognitivist, but as well see, recent
  expressivists are willing to talk about moral
  knowledge (and even, in a sense, moral beliefs),
  so this can be a slightly misleading
  characterization
• Varieties of expressivism include emotivism,
  prescriptivism, quasi-realism, explanatory
  expressivism, neo-expressivism, ecumenical
  expressivism, and so on

4
Defining Characteristics

• Expressivism is the conjunction of the following
  three claims
• Moral psychology moral judgments consist in
  non-cognitive (world-to-mind) attitudes toward
  natural facts
• Moral semantics the deep structure of moral
  discourse must be understood in terms of its
  function of expressing moral attitudes its
  primary function is non-descriptive
• Moral metaphysics the only facts there are are
  those certified by natural and social science
  talk of moral facts is a matter of projecting our
  attitudes onto the world
• So, expressivism is characterized by
  non-cognitivism, non-descriptivism, and
  projectivism
• Note some recent expressivists reject some of
  these

5
I.2 Expressivism and Subjectivism

• It is essential not to confuse expressivism with
  subjectivism
• According to subjectivism, when I say Cheating
  on your spouse is wrong, I am describing or
  reporting my attitude toward cheating
• Crudely, this is saying I dont like cheating
• This is straightforwardly true iff I dont like
  cheating
• According to expressivism, when I say Cheating
  on your spouse is wrong, I am expressing or
  manifesting my attitude toward cheating
• Crudely, this is saying Boo cheating!
• This is neither true nor false

6
The Basic Problem for Subjectivism

• (Simple) subjectivism cant explain moral
  disagreement
• If you say Collecting an estate tax is wrong,
  and I say, Hang on, collecting an estate tax is
  not wrong, we certainly appear to disagree about
  the wrongness of estate tax
• However, on the subjectivist account, you just
  said I disapprove of estate tax and I said I
  dont disapprove of estate tax these two
  sentences may well be true at the same time, so
  where is our disagreement?
• On the subjectivist account, to contradict you, I
  should say No, you do not disapprove of estate
  tax!
• There are better ways to capture whats worth
  capturing in the idea that something can be good
  for x but not good for y

7
Disagreement in Attitude

• Stevenson
• Disagreement in belief two people have beliefs
  that cant both be true
• Disagreement in attitude two people have
  attitudes that cant both be satisfied
• Two men are planning to have dinner together.
  One wants to eat at a restaurant that the other
  doesnt like. Temporarily, then, the men cannot
  agree on where to dine.
• Typically, ethical disagreement involves both,
  but only disagreement in attitude is essential to
  it
• You say Boo dancing!, expressing an attitude
  with a world-to-mind direction of fit that will
  be satisfied if the world is such that there is
  no dancing and I say Hurrah dancing!, which is
  otherwise the same except that it will be
  satisfied if there is dancing so both cant be
  satisfied at the same time
• D-in-a determines which beliefs are relevant, and
  ethical argument will terminate with A-in-A
  regardless of D-in-B

8
I.3 A Problem for Expressivism?

• Philip Pettit and Frank Jackson have recently
  argued that if we accept a Lockean picture of
  language, expressivism collapses into
  subjectivism
• On the Lockean picture, words are voluntary
  signs that come to stand for something when we
  agree to use them in the presence of certain
  beliefs
• Any explanation of how we English speakers came
  to use the voluntary sign good' for the attitude
  we do use it to express, according to
  expressivists, must allow that we recognise the
  attitude in question in us. For, to follow Locke,
  we could hardly have agreed to use the word for
  an attitude we did not recognise and failed to
  believe we had, since that would be to use the
  word for we know not what'. But that is to say
  that expressivists must allow that
• we use the word sincerely only when we believe
  that we have a certain kind of attitude. And then
  it is hard to see how they can avoid
• conceding truth conditions to That is good',
  namely, those of that belief.
• So, my utterance of That is good is true iff I
  approve of the demonstrated object

9
No Problem for Expressivism

• Smith and Stoljar
• We have agreed to use square for square things
  (to refer to them), and so utter That is square
  only when we believe something is square, insofar
  as were sincere. In virtue of the referential
  convention, claims like That is square can be
  true or false depending on whether the object is
  square or not.
• By contrast, while we have indeed agreed to utter
  That is good only when we approve of something,
  insofar as were sincere, we havent agreed to
  use it for the state of approval it does not,
  by our linguistic conventions, refer to our own
  psychological states any more than square does
• In short, while saying Thats good may indeed
  conventionally convey the speakers belief that
  she has the attitude of approval, but those are
  not its truth conditions
• Compare That is square even though I only
  sincerely utter this when I believe that I
  believe the object is square, its truth doesnt
  consist in my so believing, but precisely in the
  object being square!

10
Varieties of Expressivism

• By and large, expressivists agree on metaphysics
  all there is are natural facts
• Differences between expressivists come out in
  moral psychology, semantics, and epistemology
• What exactly does thinking that something is
  wrong consist in?
• What is the non-descriptive function of moral
  discourse?
• Can there be something worth calling moral
  knowledge if there are no moral facts that would
  make our moral representations true?

11
II First Wave Expressivism
12
Basic Arguments for Expressivism

• Depending on their philosophical background
  commitments, expressivists articulate different
  varieties of the following four basic arguments
• The Argument from Moral Internalism Only
  expressivists can make sense of the essential
  action-guiding character of moral judgments
• The Argument from Metaphysical Parsimony
  Expressivism can make sense of the phenomena of
  morality without problematic metaphysical
  assumptions
• The Argument from Epistemic Inaccessibility
  Expressivism can make sense of our talk of moral
  knowledge without assuming peculiar and
  implausible cognitive capacities
• The Argument from Semantic Openness
  Expressivists have the best explanation for the
  open feel that Moore noted

13
II.1 The Argument from Moral Internalism

• This is the familiar Humean argument
• Moral judgments essentially motivate
• Stevenson goodness is magnetic, so that a
  person who recognizes X to be good must ipso
  facto acquire a stronger tendency to act in its
  favour than he otherwise would have had (1938,
  16)
• Only non-cognitive states (those with
  world-to-mind direction of fit) essentially
  motivate
• So, moral judgements are non-cognitive states
• So, cognitivist theories like those of Moore,
  Ross, and naturalists must be false

14
Emotivism and Internalism

• The emotivism of Ayer and Stevenson is in the
  first instance a thesis about the function of
  moral language
• For emotivists, moral judgments consist in
  feelings of approval and disapproval, and moral
  utterances, in spite of surface grammar, are
  vehicles for expressing these feelings
• Thus if I say to someone, You acted wrongly in
  stealing that money, I am not stating anything
  more than if I had simply said, You stole that
  money. ... It is as if I had said You stole
  that money, in a peculiar tone of horror, or
  written it with the addition of some special
  exclamation marks. The tone, or the exclamation
  marks, adds nothing to the literal meaning of the
  sentence. It merely serves to show that the
  expression of it is attended by certain feelings
  in the speaker. (Ayer)
• Ayer and Stevenson are not particularly clear on
  exactly which feelings and attitudes are moral
  (as opposed to aesthetic or gustatory), but
  plainly they are meant to give rise to action, or
  at least a tendency to act
• Ayer simply speaks of ethical feeling and
  special sort of moral disapproval

15
Emotivism and Moral Language

• Stevenson emphasizes that the key purpose of
  moral language isnt merely expressing attitudes
  but exhorting others to share them
• When you tell a man he oughtnt to steal, your
  object isnt merely to let him know that people
  disapprove of stealing. You are attempting,
  rather, to get him to disapprove of it. ...
  Ethical terms are instruments used in the complex
  interplay and readjustment of human interests.
  (19)
• Some words have an emotive meaning, a tendency
  arising through the history of its usage, to
  produce (result from) affective responses in
  people
• The word good has only an emotive meaning
• Words like democracy and slut have both
  emotive and factual meanings

16
Emotivism and Moral Argument

• For Stevenson, giving reasons in a moral
  argument is a matter of citing facts that are
  likely to influence the emotional reactions of
  others
• The alternative is non-rational persuasion, as in
  presenting the suffering of the poor with such
  appeal that the opponent changes her attitudes
  toward the facts
• Similarly, Ayer argues that moral argumentation
  presupposes shared attitudes toward facts and
  consists in trying to show that the opponent has
  been mistaken about the empirical facts of the
  case or made a logical error
• If our opponent concurs with us in expressing
  moral disapproval of all actions of a given type
  t, then we may get him to condemn a particular
  action A, by bringing forward arguments to show
  that A is of type t.
• If, on the other hand, the opponent, as a result
  of different conditioning, doesnt share our
  reactions, we finally resort to mere abuse

17
II.2 Semantics and Metaphysics

• For the early expressivists, the main rivals were
  intuitionist non-naturalism and varieties of
  analytic naturalism
• Both of these rivals drew metaphysical
  conclusions from semantic premises
• In effect, both shared the assumption that good
  refers to a property, and analysis of its meaning
  can tell us which property it is an irreducible
  non-natural one or a natural one
• Thus, the anti-realist expressivists defended
  their view by attacking the semantic premises of
  the realists

18
The Argument from Semantic Openness

• As Moore had argued, whatever natural property N
  x has, it always makes sense to ask whether x is
  good
• For expressivists, this amounts to asking whether
  to approve of x this is always an open question
  because
• to think that something is good is to approve of
  it, but
• to think that something has a natural property,
  whatever it may be, is not, as such to approve of
  it
• So, expressivism can explain the openness of the
  open question without assuming that goodness is a
  simple, indefinable non-natural property
• Early expressivists join Moore in inferring from
  the lack of natural definability that goodness
  cant be a natural property

19
Hare Against Descriptivism

• An alternative expressivist argument against
  analytic naturalism is based on the possibility
  of moral disagreement and reform
• Grant that we use terms like good on the basis
  of some natural properties that the object has
• Still, it is possible that A thinks things that
  are x are good and B thinks different things that
  are y are good
• If good means x for A and y for B, there is no
  disagreement between A and B things that are x
  are good-A and things that are y are good-B
• But, intuitively, A and B disagree, so good
  cannot mean either x or y
• Similarly, moral reform is possible someone may
  intelligibly deny that something we all have
  thought is good isnt such

20
The Argument from Unverifiability

• Ayers argument was based on the verificationist
  theory of meaning, according to which only
  statements whose truth or falsity can (in
  principle) be verified are cognitively
  significant
• More precisely, meaningful synthetic statements
  are those that some possible experience can
  support
• No possible observations, however, support moral
  conclusions, any more than they do metaphysical
  or religious ones
• Thus, there is no way, in principle, to settle
  the argument between someone who thinks that
  abortion is wrong and someone who thinks its not
  wrong by appeal to facts
• So, ethical statements lack cognitive meaning
• They are unverifiable for the same reason as a
  cry of pain or a word of command is unverifiable
  because they do not express genuine
  propositions. (Ayer, 43)

21
Problems with First Wave Expressivism

• No explanation for why we make the error we
  apparently do when we think and talk as if there
  were moral properties and facts
• No explanation for the inferential role that
  moral statements, unlike expressions of emotions,
  can play
• Verificationist theories of meaning have few
  supports, and the open question argument makes
  questionable assumptions, so the case for
  expressivism cant be build on them

22
III Second Wave Expressivism
23
Saving the Apperances

• First wave expressivism could be attacked by way
  of rejecting the four basic arguments
• Even if one were to accept their basic thrust,
  most philosophers now agree that there is a
  further desideratum that early expressivists
  missed, namely that of saving the apperances
• We seem to argue about ethics, assert
  propositions, make inferences, think that some
  moral conclusions can be true independently of
  how we feel in short, our ethical thought and
  talk has a realist surface
• To save the appearances, the expressivist must
  how this realist surface comes about if ethics is
  about attitudes

24
Quasi-Realism and Explanatory Expressivism

• The two most sophisticated contemporary
  expressivist theories, those of Simon Blackburn
  and Allan Gibbard, are attempts to carry out this
  project of saving the appearances
• In addition, they develop further the classic
  expressivist arguments from internalism and
  metaphysical parsimony

25
The Argument from Metaphysical Parsimony

• Unlike the logical positivists, contemporary
  expressivists do not reject metaphysics as such
• Rather, they embrace a particular metaphysical
  outlook, naturalism or physicalism at the end of
  the day, all that there is are the sort of
  properties that play an explanatory role in
  science
• The project is fitting ethics into this picture
  telling a story that explains the features of our
  moral thought and practice using only
  naturalistic building blocks natural facts and
  our (natural) reactions to them

26
III.1 The Argument from Internalism Refined

• Internalism is an essential part of a
  naturalistic, evolutionary explanation for why we
  have a distinct capacity to make moral judgments
  in the first place
• As social animals, we need to cooperate and
  coordinate our actions and reactions
• The key function of morality is to make this
  possible
• But morality can serve this function only if it
  is motivationally effective
• Evolutionary success may attend the animal that
  helps those that have helped it, but it would not
  attend any allegedly possible animal that thinks
  it ought to help but does not. In the competition
  for survival, it is what the animal does that
  matters. This is important, for it shows that
  only if values are intrinsically motivating is a
  natural story of their emergence possible.
  (Blackburn, FE 48)

27
Blackburn on Moral Attitudes

• As weve seen, Blackburn wants to leave room for
  failure to be motivated
• He also wants to avoid defining moral attitudes
  in phenomenal terms, since it is implausible that
  you must always experience a particular feeling
  when you sincerely call something wrong, say
• Instead, Blackburn appeals to emotional ascent
  and stability to explain what makes an attitude
  moral
• Start with preference, disgust, anger. Say I am
  angry with you for treating her like that. This
  is not yet moralizing, but suppose Im further
  disposed to accept others being angry with you,
  too, and even encourage them. Even stronger, Im
  angry with those who are not angry with you. Im
  even angry with those who are themselves angry
  but are not angry with those who arent.
• Someone who moralizes in this way would
  disapprove of herself changing her attitudes, so
  her valuing attitudes tend to be stable
• Being moralistic is treating everything from
  dress codes to food preferences high up on the
  ladder

28
The Common Point of View

• We cant very well expect others to share in our
  anger (and anger at those who fail to be angry
  etc.) if we moralize things that merely stand in
  the way of our own self-interest
• There is thus an inherent pressure toward
  adopting what Hume called the common point of
  view when moralizing
• When a man denominates another his enemy, his
  rival, his antagonist, his adversary, he is
  understood to speak the language of self-love,
  and to express sentiments peculiar to himself and
  arising from his particular sentiments and
  situation. But when he bestows on any man the
  epithets of vicious or odious or depraved, he
  then speaks another language, and expresses
  sentiments in which he expects all his audience
  are to concur with him. He must here, therefore,
  depart from his private and particular situation
  and must choose a point of view common to him
  with others he must move some universal
  principle of the human frame and tough a string
  to which all mankind have an accord and
  symphony. (Hume, Enquiry IX)

29
Moralizing, Guilt, and Coordination

• I feel guilt when I think that the anger of
  others would be justified in expressivist
  terms, I myself approve of disapproval of someone
  like me
• Guilt is thus a matter of internalizing the anger
  of others. It characteristically derives from
  ones own actions and motivates one to make
  amends.
• Shame, in contrast, is about internalizing the
  disdain of others. It characteristically
  motivates to hide, and may derive from anything
  associated with oneself.
• Without feelings like guilt and shame, people
  would be much less likely to curb the pursuit of
  their short-term self-interest and pleasure,
  which in turn would lead to conflict and
  instability

30
Gibbard on Moral Judgment

• Gibbards account of moral judgment is cast in
  terms of rationality of emotions
• A judges that x is morally wrong iff A judges
  that (in the absence of excusing conditions) it
  is rational for her to feel guilt for having done
  x and for others to be angry with her because of
  it
• A judges that x is morally obligatory iff A
  judges that it is rational for her to feel guilt
  for not having done x and for others to be angry
  with her because of it
• Thus, it is possible to feel guilt for x without
  judging that x is morally wrong, since one might
  think ones guilt is not rational
• So, the inquiry into understanding moral judgment
  splits into two what is it to judge that
  something is rational and what is it to feel
  guilt?

31
Expressivism about Rationality

• To think that f-ing is rational (warranted) is to
• accept a set of
• norms that
• permit f-ing
• Norms can be thought of as rules that pair
  actions with deontic statuses (permitted,
  obligatory, forbidden)
• So, the Ten Commandments pair killing and
  fornicating with your neighbours wife with the
  status of forbidden and honouring your parents
  with obligatory
• For any set of norms, it is a matter of fact
  which deontic statuses it assigns to actions
• Ideally, a set of norms covers every situation
  that is, you can derive a deontic status for
  every possible action

32
Accepting Norms

• Hypothesis we have a special motivational
  control system that
• responds to linguistically encoded precepts
  (norms) because
• coordination and planning through language
  conferred an evolutionary advantage
• Being guided by language could enhance a
  proto-humans biological fitness, both by
  enabling him to develop complex plans for action,
  and by leading him to coordinate his emotions and
  actions with those of his fellows. (68)
• Those who can work out together reactions to an
  absent situation what to do and what to feel
  are ready for like situations. 72
• Coordination requires aiming for consensus and
  being motivated by one when it is found

33
Norms and Rationality

• According to Gibbard, all norms are basically
  norms of rationality of various responses
• The various different kinds of norms governing a
  thing moral norms, aesthetic norms, norms of
  propriety are each norms for the rationality of
  some one kind of attitude one can have toward it.
  Just as moral norms are norms for the rationality
  of guilt and resentment, so aesthetic norms are
  norms for the rationality of kinds of aesthetic
  appreciation. Norms of propriety are norms for
  the rationality of shock, so that something is
  improper if it makes sense to be shocked by it.
  (51-52)
• Conversely, questions about rationality are
  questions about which norms to accept
• In trying to decide what is rational, we are
  engaging our normative capacities to try to
  decide which norms to accept. We do this in
  normative discussion, actual and imaginative, as
  we take positions, subject ourselves to demands
  for consistency, and undergo mutual influence.
  (81)

34
Emotions as Adaptations

• As Gibbard acknowledges, his account of moral
  judgment is viciously circular if guilt and anger
  themselves involve moral judgments
• So, he argues that they are plausibly adaptive
  syndromes
• An adaptive syndrome is a combination of a cause
  (like intrusion), expressive behaviour (barking),
  and other behavioural tendencies (chasing the
  intruder away)
• Its adaptive, because ancestors innately
  disposed to such behaviour in such circumstances
  were more likely to survive and reproduce,
  passing the relevant genes on

35
Guilt and Shame as Adaptations

• Gibbard suggests that negative human emotions
  result from threats to ones place in cooperative
  schemes
• Lack of resources to contribute shame
• Lack of motivation to contribute guilt
• Having such emotions is adaptive, because they
  help smooth out conflicts that would otherwise
  arise
• These emotions are induced by things that
  typically portend bad treatment by others they
  are first-person counterparts to anger and
  disdain
• Basically, early humans who had a tendency to
  modify their behaviour in these specific ways to
  mollify the anger and disdain of others were more
  likely to succeed in cooperative schemes and thus
  survive and leave offspring

36
III.2 The Argument from Moral Supervenience

• Briefly, Blackburns argument from moral
  supervenience against moral realism goes as
  follows
• It is a priori that moral predicates supervene on
  natural predicates
• A metaethical theory must explain why a priori
  truths about moral language hold
• Moral realists cant explain why moral
  supervenience is a priori
• Expressivists can explain why moral supervenience
  is a priori
• So, we should reject moral realism and embrace
  expressivism

37
Supervenience

• Supervenience can be either an ontological
  relation between sets of properties or a logical
  relation between sets of predicates
• In ontological terms, supervenience holds between
  base properties and supervening properties when
  it is necessarily the case that
• there is no change in the supervening properties
  without a change in the base properties (for
  example, your beliefs dont change without a
  change in your brain) and
• two objects (or worlds) that have the same base
  properties have the same supervening properties
  (for example, if two worlds have the same
  distribution of neurological patterns, people in
  them have the same thoughts)
• Thus, as it is sometimes put, if mental
  properties ultimately supervene on physical
  properties, then if God wants to create a world
  in which people think exactly the same thoughts
  as they now do, all She has to do is to create a
  world which has identical physical properties

38
Supervenience, Reduction, and Identity

• Supervenience as such is neutral between
  reductivist and non-reductivist views about the
  relationship between base and supervening
  properties
• If A-properties and B-properties are really the
  same, there (trivially) can be no change in
  A-properties without change in B-properties
• The interest in supervenience stems largely from
  the fact that it is a relationship of dependence
  that is compatible with the two sets of
  properties being non-identical
• Supervenience is compatible with multiple
  realizability A-properties can be realized by
  many different bases
• So, pain could be realized by C-fiber activation
  in humans, D-fiber activation in baboons,
  activation of area 51 in Martians, and so on

39
The A Priori Supervenience of the Moral

• When it comes to the mind, it is an empirical
  discovery that thoughts supervene on brain events
• We can coherently conceive of a world in which
  peoples thoughts change without a change in
  their brain
• That is, Cartesian dualism isnt a priori false
• Arguably, this is not the case with ethics
• Everyone who grasps the point of moral language
  knows that if two people do exactly the same
  thing in the same context, it cannot be the case
  that one of them does something morally wrong and
  the other something morally acceptable
• Or, if you think that someone can go from being a
  good person to being a bad person without a
  change in her natural features (or
  circumstances), it only shows you dont properly
  grasp the concepts involved

40
Supervenience and Analytic Necessity

• If it was analytically necessary that euthanasia,
  for example, was always wrong, supervenience
  would be explained
• It could not be the case that something has the
  same natural properties and different moral
  properties, because having certain natural
  properties would entail have certain moral
  properties
• However, if non-naturalists are right, there is
  no necessary connection between any particular
  natural properties and being good
• So, there are no necessities of the form Vx (Nx
  -gt Mx)
• So, it is logically possible both that euthanasia
  is right and that euthanasia is wrong, for
  example
• To tell which moral quality results from a given
  natural state means using standards whose
  correctness cannot be shown by conceptual means
  alone. It means moralizing, and bad people
  moralize badly, but need not be confused.
  (Blackburn 1984, 184)

41
The Explanatory Challenge

• So, if it is possible that an act of euthanasia
  isnt wrong, how come is it a priori that if it
  is wrong, any other act just like it is
  necessarily wrong?
• If God could have created a world in which
  euthanasia has the property of being wrong or a
  world in which euthanasia doesnt have the
  property of being wrong, why couldnt he have
  created a world in which some acts of euthanasia
  have the property of being wrong and others
  dont? Why is there a ban on mixed worlds?
• Note that since we are talking about an a priori
  necessary truth, induction wont do the trick
• It isnt just that we observe act x of kind A to
  be wrong, act y of kind A to be wrong etc. and
  then conclude that all acts of kind A are wrong
  rather, we know in advance that if one action of
  kind A is wrong, all are

42
Expressivism and Supervenience

• Blackburn argues that expressivists, and only
  expressivists, can explain why moral
  supervenience is a priori in spite of lack of
  analytic necessities
• The purpose of moral language is to guide choices
  on the basis of natural features
• If we didnt evaluate like cases alike, we
  couldnt guide choices on the basis of natural
  features
• So, we must evaluate like cases alike to fulfil
  the purpose of moral language
• Competent users of concepts are in a position to
  know what is necessary to fulfil the purpose of a
  discourse
• So, we know a priori that that we must evaluate
  like cases alike

43
Blackburn on Supervenience

• When we announce our moral commitments we are
  projecting, we are neither reacting to a given
  distribution of moral properties, nor speculating
  about one. So the supervenience can be explained
  in terms of the constraints upon proper
  projection. Our purpose in projecting value
  predicates may demand that we respect
  supervenience. (Blackburn 1984, 186)
• If we allowed ourselves a system (schmoralizing)
  which was like ordinary evaluative practice, but
  subject to no such constraint, then it would
  allow us to treat naturally identical cases in
  morally different ways. ... That would unfit
  schmoralizing from being any kind of guide to
  practical decision making (a thing could be
  deemed schbetter than another although it shared
  with it all the features relevant to choice or
  desirability. (ibid.)