Elusive Causation

1
Elusive Causation

• Cei Maslen, Victoria University of Wellington,
  cei.maslen_at_vuw.ac.nz
• 19.7.2006

Some Details Contextualism about Singular Causal
Claims This is the view that statements of the
form event c is a cause of event e and event c
is not a cause of event e depend for their
meaning and truth conditions on the context in
which they occur. Event c is a cause of event
e is true in context C iff had c not happened
but c had happened instead, then e would not
have happened but e would have happened instead
(where c is a salient contrast to c in context
C, and e is a salient contrast to e in context
C). More generally (with sets of contrast
events) Event c is a cause of event e is true
in context C iff for every ci in ci there is
an ei in ei such that if ci had happened
then ei would have happened (where ci is a
salient set of contrasts to c in context C, and
ei is a salient set of contrasts to e in
context C). Putnams (less ambitious)
Contextualism about Causation Imagine that
Venusians land on Earth and observe a forest
fire. One of them says, I know what caused that
the atmosphere of the darned planet is
saturated with oxygen. What this vignette
illustrates is that one mans (or
extraterrestrials) background condition can
easily be another mans cause. Assume
Nonreductive Materialism (i.e., materialism is
true and mental properties are not reducible to
physical properties) Are there contexts in
which Epiphenomenalism is true? (i.e., are there
contexts in which all positive claims of mental
causation are false?) Impoverished
contexts? (A) A desire for chocolate was a cause
of her action. (B) The firing of neuron535 was a
cause of her action. p.t.o.
Outline A contextualist view of
causation Examples Potatoes are bad for
you Almond Crescents The Exploding gas tank
Field Impoverished contexts?
Carroll Mechanisms of Relevance Two Rules of
Attention A Rule of Probability A Default
Contrasts Rule Subject vs. Attributor
Factors Other Issues Do you have to be a
mind-reader? (No) Do you have to know what youre
talking about? (No) Some Conclusions
2
Some potential contrasts with my desire for a
chocolate flake (1) Replacement mental event
desire for a snickers bar. (2) Replacement
brain event neuron536 firing (instead of
neuron535) (3) Zombie event event
physically identical to c, but lacking all mental
properties (4) Ghost event event
mentally identical to c, but lacking all physical
properties (5) Twin-Earth replacement
event desire for a shmocolate flake. (6) Range
of Contrasts set of contrasts including
all of the above. (7) Zombie-world contrast
contrast-world which is physically identical
to the actual world but lacking any mental
properties. (8) Ghost-world contrast contrast-
world which is mentally identical to the actual
world but lacking any physical properties. (9)
Neuron-in-petrie-dish contrast event
physically identical to c but placed in a very
different extrinsic physical background. Some
mechanisms for establishing relevant contrasts
for contextualism about causation Two Rules of
Attention (1) Seriously supposing that c is a
relevant way for the cause to be absent ensures
that c is a relevant contrast. (2) Seriously
supposing that c is not a relevant way for the
cause to be absent (e.g. c is not of interest)
ensures that c is not a relevant contrast. A
Rule of Probability (3) A contrast that is very
improbable, or below a threshold that is salient
in the context, is not relevant. A Default
Contrasts Rule (4) The unspecific contrast, the
nonoccurrence of the cause, is relevant as a
default. A quote from Kim The very same
network of causal relations would obtain in
Davidsons world if you were to redistribute
mental properties over its events any way
you like you would not disturb a single causal
relation if you randomly and arbitrarily
reassigned mental properties to events, or
even removed mentality entirely from the
world. Some Conclusions - There are
(attributor-sensitive) salience rules in place
for fixing contrast events in causal contexts
that are similar to the Rule of Attention often
given for contextualism about knowledge. - There
are also (subject-sensitive) objective mechanisms
for fixing contrast events in causal contexts,
but these are trumped by the salience rules. -
Making salient certain possibilities (such as the
possibility of a zombie world) can establish a
context in which epiphenomenalism is true. -
Neither knowledge nor causation literally
disappear due to shifts in context but there are
contexts in which its harder to make true causal
claims about mental causation, just as there are
contexts in which its harder to make true
knowledge attributions. - You dont have to know
what youre talking about in order to make true
causal claims.