Greg Detre
17 October, 2001
Wed, 2nd week
John Campbell, Wilde Prof. of Mental Philosophy
what�s the importance of the notion of consciousness?
it�s experience of the world that makes it possible for us to think about it
without it, we wouldn�t know what it�s like
creatures with very different experiences of the world, will think about it in very different ways
e.g. someone who�s not had experience of colours (e.g. Jackson�s Mary, or Locke�s blind colleague � scarlet �is like the sound of a trumpet�)
most direct way to understand someone�s remarks is to experience the object being discussed
consider the most elementary case of experience making thought possible
�who�s that man there?� -� you know which person�s being talked about in a direct way because the other person directs your conscious attention towards them (like putting a yellow highlighter on them)
considers a robot zombie � no conscious experience, but can act and operate and sense like we can
such a creature wouldn�t have any glimmering of what�s going on, wouldn�t be able to think about its environment if it wasn�t conscious
in addition to being able to think, we�re conscious � what�s the evolutionary value of consciousness on top?
the reason this seems a difficult question is because it�s a bad question � it�s not that there are 2 separate things � it�s conscious experience of the world that makes thinking possible
intends to propose a test for a theory of consciousness � illustrate with 2 particular analyses
technical notion � a demonstrative = a term like �that chair� or �this clock�, referring to an object that you�re currently perceiving
they�re interesting because it�s your experience of the object that enables you to understand what a demonstrative is
How must we think of experience of objects, if it is the experience of objects that provides you with your knowledge of the reference of a demonstrative (i.e. the actual chair/clock)?
experience of the world has an explanatory role to play
if someone says, �to be conscious of an object is to be in state X, and here�s my analysis� � we can assess that proposal by seeing whether being in state X is what puts you in the position to think about the object
the test for whether a theory of consciousness is correct: could a given theory (e.g. particular brain activity) explain your knowledge of the reference of the demonstrative?
the problem with reductionists is usually that consciousness is given no explanatory role � hence the impasse
Rosenthal � a mental state is conscious iff it�s accompanied by a thought to the effect that you�re in that state
what�s the difference between being in a perceptual state that is vs is not conscious?
being in a state that represents redness or whatever, and your having a thought to the effect that you�re in that state
�consciousness of the object� =
(a) perceiving the object
(b) having a thought to the effect that one is perceiving that very object
according to Rosenthal, you can be in state (a) without being in state (b), e.g. blindsight
blindsight patients have no awareness of objects in one half of their visual field � nevertheless, patients will guess reliably about the contents of the field they�re unconscious of � initially, they�re very surprised at their own success
blindsight = perception minus consciousness
it�s very difficult to find a counterexample � any part of your conscious life is necessarily accompanied by a thought about the perception
the complexity of our conscious experience changes and grows as we get older as our conceptual repertoire expands
criticism: could X be what explains my perceiving the demonstrative, e.g. the glass?
no, because X is just the accompanying thought that I experience the glass
experience has to be something that is more primitive than thought, because it needs to be able to explain the thought
this analysis gets it round the wrong way, presupposing (and hence not explaining) my understanding of what it is to experience the glass
can�t falsify it easily just through introspection � hence this argument
consciousness of an object has to explain your understanding of a demonstrative referring to an object
why say it�s consciousness rather than perception that�s doing the work?
maybe it�s the informational part that makes thought possible, rather than some ectoplasmal consciousness?
if I ask you to point to and describe the person I�m thinking of in a crowd, and you get it consistently correct, to my surprise � it�s only when the crowd ceases to be a sea of faces and resolves into my conscious attention being directed to an individual that you can say that I �know� who you mean
connection between the notion of reference and the notion of truth. you can understand a proposition without being able to verify it � connection between the meaning and verification of a proposition. 3 parts to meaning:
knowledge of what it is to be true (truth condition)
inputs that will lead you to accept it
outputs from the proposition
knowing the reference of a demonstrative = knowing what it takes for that proposition to be true, and so you can verify and act upon the proposition
a blindsighted person can verify propositions about the contents of the blindsighted visual field � suppose there�s some water in your visual field, try reaching for it � they can do that
they can verify propositions
but they don�t have knowledge of reference, knowledge of what it is we�re talking about � the difference is that normally I have conscious experience of the glass, and it�s that conscious experience that causes neural information machinery to swing into play to let me act on the glass
our cognitive machinery is at the disposal of your conscious experience � can formulate conscious objectives, that you can act on
can a blindsight person verify the proposition? do they really know if they�re guessing, if they don�t know what it is that you�re talking about? they can answer the question, �is it upright or tilted�, but Campbell says they can select information from the right object to control verbal response, without experience of the object, and so without verification (he says)
classical view: knowledge of the reference of a term is what causes, and justifies, your use of particular ways of verifying, and finding the implications of, propositions containing the term
knowledge of reference is what causes and justifies the � (functional account)
you can do a hell of a lot even with blindsight � there are limitations, but they are continually being overturned by more ingenious experiments
it could be that a zombie is someone without conscious experience, but can select (like blindsight), but cannot think about it � not understood what control mechanisms it is that are enabling the right selection of information, but they aren�t the same as for you and I (we rely on the mechanism of conscious experience)
sleepwalking = like zombie/blindsight? yes, being able to switch on the TV and even tell you about it the next day, but not actually think about it at the time
perception, control, short-term memory, speech, problem-solving
this is just cognitive processes, ignores consciousness
he says that�s just what consciousness is � to have a brain that implements that kind of cognitive system is consciousness
common sense objection (Chalmers) � tells us nothing about why it should be like something to be a system to be undergoing these processes � you can�t empathise with it - it feels as though the model is leaving something out
Dennett�s reply � there is something it is like to be a system engaging in this kind of processing � that is what it is to be like something
Campbell has been saying that an account of conscious experience has to explain our pattern of use of demonstratives � does Dennett�s model play the explanatory role of consciousness?
rather though, Dennett�s model is a consequence of consciousness � it�s more like the blindsighted patient�s
yes, it can verify and act upon propositions containing the demonstrative �you�ve got all the inputs and outputs, but you�re missing the knowledge of the reference of the term
introspection + reportability
but if consciousness is an epiphenomenon, then isn�t consciousness more like a self-directed inaudible report, rather than a mechanism per se???
it�s conscious experience of the world that makes thinking possible � isn�t it equally possible that it�s thinking about (i.e. processing neurally) the world that makes us conscious???
by implication, can you have (b) without (a), e.g. hallucinations???
so Rosenthal�s analysis/definition is necessary but not sufficient, right???
his main point is giving consciousness an explanatory role �
consciousness is what enables us to think about the object � why should it
necessarily have an explanatory role??? don�t you have
to first show that experience of the object does play an explanatory role???
he�s using �thought� to mean something more linguistic and complex and higher-order than just the perception/sensation???
what sort of explanation could satisfy state X???
is your thesis empirically falsifiable � if I do build a fully-functional robot, does that mean it�s conscious, or does it prove you wrong???
why can�t you define state X in
causal/functional terms???
how does Campbell explain blindsight???
does it pose a problem for Campbell that we can seemingly do many things better unconsciously??? is it just that all of these are �thoughtless� acts, i.e. do not require us to think about the object/action???
campbell thinks that it�s �[our] experience of the object that enables [one] to understand what a demonstrative is� � I think it�s the natural pattern-matching/categorising/recognition systems/modules in the brain that do that, and that naturally give rise to the notion of the demonstrative, in conjunction with emergent(???) processes like attention�
in his criticism of Rosenthal, why
is it necessarily the case that �experience has to be something that is more
primitive than thought�???
being conscious of the object must be logically prior to having a (conscious) thought about it, i.e. if we weren�t conscious of the object in the first place, how could we consciously choose to have a thought about it??? � good objection
you�ve explained why there need to be internal representations in our minds of the external world � presumably instantiated in our brains � and you can define representations as being �things you can play around with� � but why do we need to be conscious of these internal representations???
�normally I have conscious experience of the glass, and it�s that conscious experience that causes neural information machinery to swing into play to let me act on the glass� � are you not then waiting for science to show how conscious experience plays a causal role in the physical world???
perhaps it counts as verifying (e.g. for a blindsighted person) if it�s a reliable method of accessing the truth???
if you could teach a blindsighted patient to just trust their intuitions/guesses, they�d function almost normally, without conscious experience, wouldn�t they??? why isn�t that knowledge of the reference of the term???
surely all that�s missing is a particular pathway that somehow casts a thought process into consciousness, but otherwise they�re functioning more or less identically, aren�t they???
is this whole issue (e.g. about
blindsight) one of free will??? or is it about the complexity and type of
thoughts that we are capable of having???
can�t the knowledge of reference be
captured in connectionism???
how recent is the Dennett map???
what�s the difference between Dennett�s attention and conscious attention???
why can�t you take Dennett and add panpsychism or epiphenomenalism??? � any account along your lines has to rest on top of a cognitive model � what do you have to say that is positive that keeps most of the progress that cognitive science has made plus whatever you think is missing???
surely Dennett�s model can play the role of state X???
Minsky and multi-modal representations
what more is needed of the
explanatory role played by knowledge of the reference than functioning??? do
you think that I should be able to tell in the 3rd person whether someone has
conscious experience???
what role does conscious experience actually play??? what couldn�t someone without conscious experience do???
you�re just talking about
representations, aren�t you???
no, because the representations need to be caused and justified by �knowledge of the reference�
what does �causing and justifying� the functional account mean/require???
a functionalist would have to say that there is nothing to cause and justify, only the functional use � but that leads you to lots of trouble with your theory of meaning
Witt, Dummett and Quine all thought that all you have is the pattern of use � once you remove knowledge of reference, then it�s very difficult to know what constitutes making a mistake � since nothing causes or justifies, there�s nothing for you to keep faith with � that�s Witt�s problem with following a rule
no, the info is contained in the environment, in our shared perceptual mechanisms, and our genes � altogether, we more or less opt for the same optimum solution from within a very restricted space � plus, we have a shared means of agreeing and discussing our chosen causal/justifying convention/system
why can�t you add a super-box on
another level to analyse causes and justifications and direct the overall
system???
could it be hidden in M (working memory)???
no, it�s at the wrong level � that�s just another cognitive process
are the boxes defined by what they�re in between???
yes � this is functionalism (inter-defining all the boxes functionally in terms of their inputs and outputs)
why can�t you just add a �consciousness� box to Dennett�s map???
that would be just adding complexity � you need something that would control which patterns of use you�re engaging in
Russell�s notion of acquaintance with objects is the same � that acquaintance is more primitive than thought, but allows you to think about the thing
�it�s conscious experience of the world that makes thinking possible� � it seems more plausible that it�s thinking about (i.e. brain-processing) the world that makes us conscious???
what does �causing and justifying� the functional account mean/require???
doesn�t your �classical view of knowledge of the reference
of a term� just amount to a definition of a representation??? you�re just talking about
representations, aren�t you??? what further role does consciousness play???
you�ve explained why there need to be internal representations in our minds of
the external world � presumably instantiated in our brains � and you can define
representations as being �things you can play around with� � but why do we need
to be conscious of our internal representations??? it would make much more
sense to see the explanatory role that you think Dennett�s account is missing
as a high-level emergent mechanism.