o . input
TRANSCRIPT
1
24.09 spring 061
24.09 Minds and Machinesspring 2006
• handout onfunctionalism
• problem set, writingassignment this week
• final exam 5/24 9-12• current week
highlighted
24.09 spring 062
24.09 Minds and Machinesspring 2006
• functionalism,contd.
24.09 spring 063
“troubles with functionalism”
Imagine a body externally like a human body, sayyours, but internally quite different. The neuronsfrom sensory organs are connected to a bank oflights in a hollow cavity in the head. A set ofbuttons connects to the motor-output neurons.Inside the cavity resides a group of little men.Each has a very simple task: to implement a“square” of an adequate machine table thatdescribes you. (96)
24.09 spring 064
Block’s homunculus head
G
M
I16 I17 I18
stateG-man
input
O5CO77ZO8H H
O45PO191MO25K G
I18 117 I16. .O190 O191
input
state
output
24.09 spring 065
there is prima facie doubt whether [thehomunculus head] has any mental states atall—especially whether it has what philosophershave variously called “qualitative states”, “rawfeels”, or “immediate phenomenologicalqualities”…there is prima facie doubt whetherthere is anything it is like to be the homunculi-headed system. (97)
24.09 spring 066
from the philosophical toolkit:
a priori and a posteriori• (knowable) proposition p is knowable a priori iff p
can be known independently of experience• otherwise, proposition p is knowable (only) a
posteriori
a posteriori andcontingent:
it’s sunny
a posteriori andnecessary:
water=H2O
a priori andcontingent:
??
a priori andnecessary:
2+3=5
2
24.09 spring 067
Functionalism vs.psychofunctionalism (see Block)• suppose functionalism is true• could you (at least “in principle”) write down the
functional characterization of mental states just byreflecting on the meanings of mental vocabulary?
• the (“commonsense”, “analytic”) Functionalistsays ‘yes’
• the (“scientific”, “empirical”) psychofunctionalistsays ‘no’—science will tell us the functional story,not conceptual analysis
24.09 spring 068
Functionalists andpsychofunctionalists
• Lewis and Armstrong are Functionalists (“analyticfunctionalists”)
• Putnam is a psychofunctionalist• psychofunctionalism is probably the more popular
version• we will return to this issue when we discuss
Chalmers’ paper “Consciousness and its Place inNature”
24.09 spring 069
“mad pain and martian pain”
• defends Functionalism—the functionalspecifications of mental states can benoodled out from the armchair
• defends “realizer state” as opposed to“role state” functionalism
• replies to the “knowledge argument” (to bediscussed later)
24.09 spring 0610
his pain is caused bymoderate exercise
the madman
intense pain causeshim to snap hisfingers
24.09 spring 0611
the martian
• his hydraulic mindcontains nothing likeour neurons
• the causes and effects ofhis pain are like thecauses and effects ofour pain
24.09 spring 0612
• a simple identity theorysolves the problem of madpain, but not martian pain
• a simple functionalism goesthe other way: right aboutthe martian, wrong about themadman
• Armstrong’s and mytheory…wriggles betweenScylla and Charybdis
3
24.09 spring 0613
• the concept of pain is the concept of astate that occupies a certain causal role
• whatever state (e.g. c-fibers firing) doesoccupy that role is pain
• but something else might have occupied therole (just as someone other than SusanHockfield might have occupied the MIT-president role)
• so, pain might not have been pain• ‘pain’, as Armstrong and I understand it, is a
nonrigid designator
24.09 spring 0614
the martian
• the thing to say about martianpain is that the martian is inpain because he is in a statethat occupies the causal role ofpain for martians (hispopulation)
• ditto (mutatis mutandis) for us
24.09 spring 0615
the madman is in painbecause he is in thestate that occupies thecausal role of pain formankind
he is anexceptionalmember of thatpopulation
the state thatoccupies the rolefor the populationdoes not occupy itfor him
24.09 spring 0616
Minds and Machinesspring 2006
read Putnam, Burge