7' Moral Facts and Scientific Inquiry

1
7. Moral Facts and Scientific Inquiry

• Antti Kauppinen

2
I Non-Reductivist Moral Naturalism
3
Responding to Error Theory I The Argument from
Relativity

• The mere fact of disagreement shows little (some
  people disagree about the shape of the Earth)
  what poses a challenge is disagreement among
  equally rational people
• First line of response much disagreement is only
  apparent
• Same basic moral principles lead to different
  derivative principles in different circumstances
• Eskimos may think it right to kill off the old
  and sick, unlike us, but what if the alternative
  was that both the old and the young perish? Dont
  we accept sacrificing some to save everyone?
• Same basic principles lead to different
  derivative conclusions given differences in
  factual beliefs
• Some people think high taxation and
  redistribution best helps the poor, others think
  low taxes and faster economic growth best help
  the poor
• Some of these belief disputes may be very hard to
  resolve (how malleable is human nature? How much
  does poverty influence crime?)

4
Responding to Error Theory II The Argument from
Relativity 2

• Second line of response moral disagreement is
  more common than factual disagreement, because
  there are more distorting factors
• Moral principles concerns largely the regulation
  of our interaction and the distribution of
  benefits, which means that peoples
  self-interest, prejudice, and ideology are more
  likely to distort their views than in, say,
  physics
• Religion and religious institutions have long
  restricted moral inquiry more than scientific
  secular moral theory is still a new enterprise
• Third response some moral disagreement reflects
  the fact that moral ties and incommensurability
  are possible
• A realist can grant that there may be no fact of
  the matter about the comparative value of opera
  and pop music
• Debates can be underspecified perhaps abortion
  is wrong in some cases and right in other cases
• Fourth response if persistent disagreement among
  rational inquirers means that there is no fact of
  the matter or correct answer, most philosophical
  disputes go out the window (partners in crime)
• After millennia, philosophers cant agree whether
  we have free will does that mean there is no
  fact of the matter about it?

5
Responding to Error Theory II Queerness

• Recall that Mackies argument from queerness had
  two elements
• Moral facts would be queer
• We would need a queer faculty to know about them
• Contemporary moral naturalists deny both of these
• Moral facts are just natural facts
• We can come to know moral facts much like other
  natural facts

6
Varieties of Moral Naturalism

• Synthetic moral naturalists believe that moral
  terms cannot be defined in terms of natural
  terms, but argue that moral properties can still
  be natural
• Non-reductive naturalists believe that moral
  properties are natural properties of their own
  kind that are not reducible to other natural
  properties
• Reductive naturalists believe that moral
  properties are reducible to other natural
  properties
• Analytic moral naturalists believe that moral
  terms can, after all, be defined in natural terms

7
The Open Question Argument

• Moore and other non-naturalists employ the Open
  Question Argument to show that
• moral terms are not definable in natural terms
  (semantic claim)
• consequently, moral properties are not identical
  with any natural properties, but are instead
  non-natural (ontological claim)
• Error theorists like Mackie accept both a and b,
  but reject non-natural properties as queer
• Contemporary synthetic naturalists accept a (so
  there are no analytic truths of the form If x is
  N, then x is M), but reject b

8
The Open Question Argument, Synonymy Version

• Two properties are identical only if the
  predicates that ascribe them are synonymous (the
  semantic test of property identity)
• For example, the property of being a vixen is the
  same as the property of being a female fox,
  because vixen and female fox are synonymous
  (or it is analytically true that vixens are
  female foxes)
• So, goodness is identical with a natural property
  N only if good and N are synonymous
• If two terms are synonymous, substituting one for
  the other does not change the meaning of a
  sentence (the substitutional test for synonymy)
• Vixens run fast means the same as Female foxes
  run fast
• Substituting any N for good changes the
  meaning of any sentence
• For example, Pleasure is good says more than
  Pleasure is pleasure asking Is pleasure
  good? is not just asking Is pleasure pleasure?
• So, good is not synonymous with any
  naturalistic term N (3, 4)
• So, being good is not identical with any natural
  property N (1, 5)

9
The Semantic Test for Property Identity

• For many contemporary naturalists, the weakness
  of the Open Question Argument lies in the
  assumption that two properties are identical only
  if the predicates that ascribe them are
  synonymous (premise 1, the semantic test for
  property identity)
• The semantic test is motivated by a central
  thesis of Freges philosophy of language, sense
  determines reference
• Once critics of Frege like Kripke and Putnam
  rejected this thesis, new metaphysical
  possibilities opened up
• The basic point is that analyticity is not the
  same as necessity
• There are necessary truths that are not analytic,
  like water H2O
• A consequence is that even if two expressions
  have a different sense (are not synonymous), they
  can designate the same property in all possible
  worlds

10
Non-Reductive Naturalism

• So, even if good does not mean the same as
  conducive to pleasure, it can still be the case
  that being good is being conducive to pleasure,
  where the is is not that of predication but of
  identity
• Another way for moral properties to be natural is
  for them to be constituted by natural properties
  (the is of constitution)
• Constitution is not the same as identity
• Even if a table is nothing over and above a
  certain arrangement of particles, it could remain
  the same while there was some change in that
  arrangement
• Similarly, even if injustice is nothing over and
  above a certain distribution of rewards, a social
  arrangement can remain unjust while there is some
  change in that distribution (say, Tom rather than
  Harry gets a lot while doing nothing) injustice
  is multiply realizable

11
Which Natural Facts Are/Constitute Moral Facts?

• We identify the neurological facts that are or
  constitute psychological facts on the basis of
  what the true psychological theory says
• Similarly, we identify the psychological/sociologi
  cal/etc. facts that are or constitute moral facts
  on the basis of what the true moral theory says
• So, if hedonistic utilitarianism is true, moral
  facts are facts about what produces most pleasure
  to most people

12
II Coherentist Moral Epistemology
13
Mackies Epistemological Challenge

• If there were objective values, the only way we
  could be aware of them would have to be by some
  special faculty of moral perception or intuition,
  utterly different from our ways of knowing
  anything else.
• But if moral facts are just higher-order natural
  facts, it should be possible for us to come to
  know them just like other natural facts, by way
  of scientific investigation and observation
• To be sure, observing moral facts requires
  background theory, but on a coherentist view of
  science, any observation is theory-laden

14
Foundationalism and Coherentism

• Foundationalism there are some beliefs that are
  not justified by other beliefs
• Instead, they are justified by virtue of being
  produced by a reliable mechanism or are
  self-evident
• The basic argument for foundationalism is the
  regress argument if every belief required
  justification by another belief, this would lead
  either to infinite regress or to vicious
  circularity
• Coherentism not all circularity is vicious a
  belief is justified because it forms a part of a
  coherent system of beliefs
• Negative thesis there are no foundational
  beliefs
• Positive thesis justification is a matter of
  mutual support between beliefs

15
The Negative Thesis of Coherentism

• You read on a poster that someone is giving a
  talk on Iraqi refugees at 5.30pm, and form the
  belief that someone is giving a talk on Iraqi
  refugees at 5.30pm (call this p). What does it
  take for your belief that p to be justified?
• Anyone should grant that justification of p
  depends on your being justified in believing that
  the poster is a reliable source of information
  about events taking place at the university (call
  this q), instead of being a prank etc.
• How about your perceptual belief that it says on
  the poster that p?
• Foundationalist a perceptual belief is justified
  without support from further beliefs, for example
  because it is produced by a reliable mechanism,
  such as normal vision
• Coherentist to be justified in believing
  anything, you must have a reason for it. Your
  reason for believing that things are as they seem
  to you is that you believe that vision is a
  reliable mechanism for producing beliefs (call
  this r)
• For the coherentist, just like you dont have
  justification for believing what a poster tells
  you unless you believe it is reliable, you dont
  have justification for believing what your eyes
  tell you unless you believe they are reliable

16
Coherentist Justification

• But what justifies your second-order belief r
  that vision is a reliable mechanism?
• It coheres with your other second-order beliefs,
  such as the belief that evolution has equipped us
  with senses that are generally reliable under
  suitable conditions
• The beliefs that vision gives rise to match
  beliefs that other mechanisms you believe to be
  reliable produce you see a helicopter, you hear
  a helicopter
• And finally, r coheres with an ever-expanding set
  of first-order observational beliefs, such as
  your belief that the poster says that p, your
  belief that youre looking at a slide, your
  belief that these words are written in black...
• Thus, observational beliefs and second-order
  beliefs are mutually supporting
• In the light of new experiences, both may have to
  be adjusted

17
Coherentism In Science

• Logical positivists believed that the mark of
  scientific theories was that they could be
  verified by observation, at least in principle
• Popper argued that science makes many claims it
  is impossible to verify, but at least scientific
  theories are falsifiable by observation
• Quine rejected this picture of science in favour
  of a coherentist view
• Observational data always depend on auxiliary
  hypotheses that form a part of a background
  theory that is held fixed for the purposes of an
  experiment, but can itself be put in question
• Neuraths boat metaphor just like you can fix a
  boat at sea one plank at a time, relying on other
  planks to keep it floating, we can fix our belief
  system one belief at a time, provisionally
  relying on other beliefs that may themselves come
  into question

18
Reflective Equilibrium in Ethics

• Rawls gave the classic formulation of the
  coherentist method of reflective equilibrium in
  ethical theory
• Start with a lot of moral judgments about
  particular cases
• Throw out those that result from obvious sources
  of distortion, are unstable over time etc. what
  is left are considered judgments
• Formulate a general principle that explains as
  many considered judgments as possible
• Reconsider and modify the considered judgments to
  cohere better with the principle
• Reformulate the principle to fit the new set of
  considered judgments
• Repeat steps 4 and 5 until you arrive at a
  reflective equilibrium between particular
  judgments and principles
• Narrow reflective equilibrium involves only
  ethical beliefs wide reflective equilibrium
  takes into account all our beliefs

19
III The Challenge of Explanatory Impotence
20
Testing Moral Theories

• When joined together with suitable auxiliary
  hypotheses, moral theories have testable
  observational implications
• Case 1 Justice and death penalty
• Theory In a just state, there is no death
  penalty
• Assumption Britain is a just state
• Observable implication There is no death penalty
  in Britain
• Case 2 Good people and promises
• Theory Good people keep their promises
• Assumption Jane is a good person
• Observable implication Jane will keep her
  promise to Joe
• But can the theories and assumptions themselves
  be directly justified by observation?

21
Harmans Challenge

• Can moral principles be tested like scientific
  hypotheses, on the basis of observation rather
  than mere thought experiments?
• Harmans challenge
• If you round a corner and see a group of
  hoodlums pour gasoline on a cat and ignite it,
  you do not need to conclude that what they are
  doing is wrong ... you can see that it is wrong.
  But is your reaction due to the actual wrongness
  of what you see or is it simply a reflection of
  your moral sense, a sense that you have
  acquired perhaps as a result of your moral
  upbringing?
• Harman will argue for the latter option

22
Theory-Laden Observations

• Harman grants that all observation is
  theory-laden
• To see something as an activity of burning cats
  you need a bunch of concepts and background
  beliefs about animals, agents, smoke, flames, and
  so on with a different background theory, the
  same pattern of light on your retinas would give
  rise to different perceptual beliefs
• No difference yet to the moral case if you hold
  a moral view, whether consciously or
  unconsciously, you will be able to perceive
  rightness or wrongness
• In both science and ethics, observational beliefs
  may conflict with your explicit theory
• You must choose whether to modify the theory or
  reject the observation as an error

23
Inference to the Best Explanation

• Harman endorses a principle known as inference to
  the best explanation, according to which we are
  entitled to ontological commitment to entities or
  properties that are indispensable to the
  explanation of our experiences
• Compare House M.D.

24
Best Explanation in Science

• According to Harman, the best explanation of
  scientific observational beliefs appeals to
  scientific facts
• Theory If a tungsten atom is bombarded with
  neutrons, it releases a proton in a cloud chamber
• Background belief Protons leave a vapor trail
• Suppose a physicist sees a vapor trail in a cloud
  chamber and thinks to herself There goes a
  proton. This is just what she expected on the
  basis of her theory.
• The observation supports the theory only if it
  could not be equally well explained without
  assuming the existence of a proton in the cloud
  chamber

25
Best Explanation in Ethics

• Harman argues that in ethics, the best
  explanation of observations never appeals to
  moral facts or properties
• The best explanation of your thought That is
  wrong! involves only your perception of
  cat-burning and the fact that youve been brought
  up (or have an innate disposition) to disapprove
  of cruelty to animals
• Thus, moral observations dont give evidence of
  moral facts, just your moral sensibility youre
  the sort of person who disapproves of tormenting
  cats
• Moral facts are not causally effective
• Protons can cause a vapor trail, which reflects
  light to the physicists eyes in such a way that,
  given her training and theory, she can
  immediately judge that there goes a proton
• In contrast, there does not seem to be any way
  in which the actual rightness or wrongness of a
  given situation can have any effect on your
  perceptual apparatus.

26
Evidence for Theories

• The physicists perceptual belief is evidence for
  her theory, because it results from an
  explanatory chain
• The physical theory explains why there is a
  proton, the protons existence explains why there
  is a vapor trail, and the trail explains why
  there is the observation
• By contrast, The explanatory chain from
  principle to observation seems to be broken in
  morality
• An assumption about moral facts would seem to be
  totally irrelevant to the explanation of your
  making the judgment you make.

27
Sturgeons Response

• To begin with, if Harmans argument is to be a
  new and interesting one, it must be that even if
  there are moral facts (even if it is wrong to
  torture cats, for example), they play no
  explanatory role with respect to our moral
  beliefs
• So, Sturgeon will assume that there are moral
  facts, and argue that they do explain our
  beliefs, and other things besides
• Candidate moral explanations
• Hitler was ordered the death of millions, because
  he was morally depraved
• South Africans, black and white, rebelled against
  apartheid, because it was a deeply unjust social
  system

28
The Counterfactual Test of Explanatory Relevance

• An assumption is irrelevant to an explanation of
  a fact if
• the fact would have obtained, and we could have
  explained it just as well, even if the assumption
  had been false
• In positive terms, A is relevant to explaining B
  if B would not have occurred had A not occurred
• Lee Harvey Oswalds pulling the trigger is
  relevant to explaining the death of JFK, since
  JFKs wouldnt have died as he did had Oswald not
  pulled the trigger
• Well leave aside issues of causal
  overdetermination etc.
• A lot of people believed in witches in the middle
  ages do we need to assume the existence of
  witches to explain this fact?
• No, because even if there are were no witches,
  people would still have believed so, and we can
  explain this belief just as well (say, by
  reference to the social function of the belief)
  without assuming the existence of witches

29
The Counterfactual Relevance of Moral Facts

• Is Hitlers moral depravity irrelevant to
  explaining the fact that he ordered the death of
  millions?
• What if Hitler had not been morally depraved?
  Would he still have ordered the Holocaust?
• Sturgeons response
• According to the moral realist, Hitlers moral
  depravity is constituted by his non-moral
  character traits, such as his insensitivity to
  other peoples suffering, his willingness to
  sacrifice others to further his ambition, and his
  flagrant racism.
• Hence, to imagine a situation in which he would
  not have been morally depraved, we have to
  imagine him lacking these qualities we have to
  imagine a Hitler who wasnt a racist, cared about
  the pain of others, and respected the rights of
  others.
• So the question is would Hitler have ordered the
  Holocaust, had he been a caring,
  rights-respecting non-racist. And the answer is
  obvious surely he wouldnt have.
• So, the fact to be explained would not have
  obtained, if we assume that Hitler was not
  morally depraved. So, Hitlers moral depravity is
  not irrelevant to the explanation.

30
Choosing the Right Counterfactual

• Harman seems to think that the counterfactual to
  evaluate would be Had Hitlers character been
  just as it is in non-moral terms (racist etc.),
  but had he nevertheless not been morally
  depraved, would he still have ordered the
  Holocaust?
• But this seems impossible, given that Hitlers
  depravity supervenes on his non-moral qualities
  he couldnt have had those exact non-moral
  qualities, yet not be morally depraved

• Compare with the phenomenon of imaginary
  resistance we can easily imagine pigs with wings
  or steel that melts at room temperature (anything
  that involves change in natural laws), but we
  cant, it seems, imagine a world in which raping
  a child would not be morally wrong
• So, take anything you think is morally wrong (in
  virtue of having some non-moral properties and
  relations), and try to imagine something that
  would have the exact same non-moral properties
  and relations, but would be morally right! The
  odds are against you.

31
Explaining Moral Beliefs

• How about Harmans cat-burning case? Would we
  believe that the youths action was wrong even if
  it wasnt?
• In this case, according to the moral realist, the
  wrongness of cat-burning is constituted by its
  being pointless, deliberate cruelty
• So, for it not to be wrong, it would have to lack
  these features it would have to be a tragic
  accident, say
• If the cat-burning was a tragic accident, then
  plausibly, we wouldnt believe that it was
  morally wrong
• Hence, by the counterfactual dependence
  criterion, the best explanation of our moral
  belief involves the moral wrongness of the action

32
Harmans Epiphenomenalist Response

• Harman argues that Sturgeons criterion of
  relevance is too permissive
• If mere counterfactual dependence were sufficient
  for best explanation, even an epiphenomenalist
  (someone who thinks that a supervenient property
  pulls no causal or explanatory weight) would have
  to accord an explanatory role to supervenient
  facts, which is not the case
• For example, in philosophy of mind, an
  epiphenomenalist is someone who thinks that
  mental states are causally impotent all the
  work is done by the underlying brain states
• Similarly, a moral epiphenomenalist thinks that
  Hitler had certain non-moral traits that both
  explained his actions and made it the case that
  he was evil. His being evil, by contrast, does
  not explain anything.
• What this shows is that mere counterfactual
  dependence is not sufficient to grant moral
  properties a role in best explanation

33
Strengthening the Counterfactual Dependence

• The epiphenomenalist challenge to the
  non-reductive naturalist seems to be an instance
  of a more general challenge to the explanatory
  potency of higher-order properties in cases where
  all the causal work seems to be done by
  underlying properties
• Thus, some realists borrow general response
  strategies to meet the challenge
• One such strategy is appeal to the notion of
  program explanation

34
The Idea of Program Explanation

• Pettit and Jackson higher-order properties can
  be causally relevant and necessary for best
  explanations without being directly causally
  efficacious
• The case of boiling water in a glass bottle
• Explanation 1 (higher-order, program
  explanation) the bottle broke because the
  temperature was high
• Explanation 2 (lower-order, process explanation)
  the bottle broke because a particular water
  molecule (call it x) hit the glass molecules,
  breaking their bond
• Assumption temperature is just a matter of mean
  kinetic energy of the molecules, so its causal
  powers are entirely derived from those of the
  molecules it is not directly causally
  efficacious
• So, we can explain the breaking of the bottle by
  the fact that x hit the glass with speed s, which
  was sufficient to break it (explanation 2)

35
Program Explanation

• In what sense is temperature of the water
  causally relevant, then?
• Pettit and Jackson the realization of the
  temperature property logically ensured that some
  directly causally efficacious molecule would
  produce the cracking if the water was hot
  enough, either x or y or z would hit the glass
  hard enough to break it
• It did not do any work in producing the cracking
  of the glass, but it had the relevance of
  ensuring that there would be some property there
  to exercise the efficacy required.
• Program explanation provides modal information
  about the event even if things had been
  slightly different (y would have hit the glass
  instead of x), the same result would have come
  about
• Thus, it enables better predictions if you heat
  water in a bottle, it will break

36
Moral Program Explanation

• The apartheid case
• Program explanation people rose against
  apartheid because of its injustice
• Process explanation people rose against
  apartheid because people in Soweto were beaten,
  blacks had no access to higher education or
  decent housing, and non-violent leaders like
  Mandela were imprisoned
• Since injustice consisted of the sort of things
  mentioned in the process explanation, it was not
  causally efficacious over and above them
• However, the system would still have been unjust
  had people been beaten in Transvaal instead of
  Soweto, had blacks had access to decent housing
  but been subject to random confiscation of
  property, and so on
• In these circumstances, people would plausibly
  still have risen against apartheid. This is
  predicted by the program explanation, but not by
  the process explanation. In this sense, the
  program explanation is the best explanation of
  what happened.

37
Moral Program Explanation II

• The cat-burning case
• Program explanation We believe Joes action is
  wrong because it is wrong
• Process explanation We believe Joes action is
  wrong because it involves causing pain to an
  animal for fun
• Since the wrongness of Joes action consists in
  causing pain to an animal for fun, it has no
  causal powers over and above that
• However, had Joe done something slightly
  different, like electrocute the cat, it would
  still have been wrong
• In these circumstances, we would plausibly still
  have believed his action was wrong. This is
  predicted by the program explanation but not the
  process one.