<<

. 38
( 62 .)



>>

during saccadic motion creates stunning lapses in change detection. On
the sensorimotor contingency theory, the ¬rst of these involves a subject
learning a new sensorimotor contingency within one sense modality that
comes to be integrated with the rest of the adapting subject™s behavior.
The second not only reveals the limitations to the information that is en-
coded in visual representation, but also shows the importance of active
and continual sensory interaction of the surrounding environment for
complete perception.13
The account also predicts that there should be cases of sensory substi-
tution just when alternative sensory channels preserve the same structure
of sensorimotor contingencies. This arguably is precisely what is found
in cases of remote tactile sensing and the visual experiences of users of
tactile visual sensory systems (TVSSs). The best known of the latter of
these use a camera to feed visual information into cutaneous stimulators
attached to the subject™s back, stomach, ¬nger, or forehead, where tac-
tilely stimulated, blind subjects report visual experiences. On the senso-
rimotor contingency theory, what subjects learn through their practice
in actively controlling the camera is a new series of tactile sensorimotor
contingencies, each of which is isomorphic to those that the visual system
would normally develop.14
Critical to the success of these TVSSs in producing reported visual
experiences in the blind is the ability of the subject to control the
movement of the camera. This makes sense on the sensorimotor the-
ory, since what is critical to visual experience, on that view, is establishing
or maintaining sensorimotor contingencies, not simply receiving or pre-
serving information that codes for visual features of the environment.
(For readers with CAVE “ Complete Audio-Visual Environment “ virtual
reality experience, note the experiential difference when you walk with
the virtual reality control, as opposed to when you follow someone else
holding the control.) The reported visual experience of TVSSs are, to be
sure, incomplete vis-` -vis those using their eyes “ for example, subjects do
a
not report seeing objects as colored. But this can be explained in terms of
Expanding Consciousness 235

the limitations to the isomorphism between the visual information that
the subject controls through the camera and the tactile stimulation she
receives on her skin. The relevant (and surprising) ¬nding is that there
is any visual experience reported at all.
Likewise with remote tactile sensing more generally. Tactile sensations
are usually felt at a particular location in the body, but they can also be
felt in objects that extend beyond the body, typically in (and through)
objects to which the body is reliably connected. On the sensorimotor
theory, and TESEE views more generally, this is because sensory experi-
ence is a matter of embodied and embedded agents doing things in the
world. While the body is a critical mediator for that action, it forms only a
contingent boundary for sensory experience that can be, and sometimes
is, transcended. Blind people experienced in the use of a cane feel ob-
jects in the world, not the cane itself; bicyclists can feel the ¬‚atness of a
tire while they are riding, and without touching the tire itself; and experi-
enced hockey players feel the puck on the end of their sticks. Just as in the
case of visual perception using TVSSs, here the phenomenology requires
practice and is more restricted than that gained through regular tactile
stimulation. In particular, the sensations are chie¬‚y spatial in character,
concerning the position, size, texture, and shape of what is felt, and do
not correspond to the full range of bodily sensations. On this view, tactile
sensations are felt at a bodily location, despite the fact that it is in the
brain where they are registered, because the body is systematically linked
to the brain as a matter of physiology. When the body becomes similarly
causally linked to other features of the world, at least some of those felt
sensations can be transferred to those features.
A second way in which a TESEE view has been developed is through
Susan Hurley™s attack on the equation of perception with input (and ac-
tion with output). Hurley™s alternative idea is that perception and action
are better thought of as forming dynamic cycles. In the previous chapter,
I brie¬‚y mentioned Hurley™s caricature of the traditional view of the mind
as a sort of sandwich in which cognition is the ¬lling. Her conception of
perception and action as forming dynamic cycles is one of her positive
alternatives to the traditional characterization of perception and action.
On the traditional view, perception is input and the cognitive archi-
tecture of the organism is conceptualized in terms of what Hurley calls
vertical modules:
Each vertical module performs a broad function, then passes the resulting rep-
resentations on to the next. Within the perceptual module, information about
locations, color, motion, and so on, is extracted from inputs by various parallel
Thinking Through and Beyond the Body
236

streams of domain-speci¬c perceptual processing. The representations produced
by the various streams of input processing converge and are combined by percep-
tion. The uni¬ed result is sent on to cognition, the central module that interfaces
between perception and action.15

The chief puzzles about visual experience on this view are where, when,
and how consciousness of particular features of visual experience come
about, and how they are uni¬ed to form the seamless experience that,
for the most part, we typically have.
In challenging this division between perception and action, Hurley
introduces the idea of horizontal modules, content-speci¬c layers that loop
“dynamically through internal sensory and motor processes [as] well as
through the environment.” There are several features of horizontal mod-
ules relevant to our rethinking of visual experience. The ¬rst is that they
are dynamic in that they involve continuous feedback between percep-
tion and action. For this reason, any process that they govern must be
conceptualized as temporally extended, rather than simply as a series
of temporal atoms, and as embodied in an active agent. The second
is that there is no barrier to such modules being either entirely inter-
nal to an organism, or extending partly into that organism™s environ-
ment and so locationally externalist. Perception does not occur ¬rst,
with perceptual inputs being processed “centrally” and then instructions
passed off to action. Rather, the many systems engaged in perceptual
processing, including in the production of the “what it™s like” of experi-
ence, cut across the input-output distinction. As with the sensorimotor
contingency theory, on Hurley™s view there is a “there” to sensory ex-
perience, but sensory experience is not properly conceived as a type
of internal event, or a property of some speci¬c internal state of the
organism.16
There are several af¬nities between the views of O™Regan and No¨ e
and Hurley. Like the sensorimotor contingency theory, Hurley™s views
of perception challenge the traditional way of thinking of visual experi-
ence as some sort of internal event that is then monitored or processed
further in some other way to give rise to consciousness. And like that
theory again, Hurley™s view makes sense of a range of actual and hypo-
thetical experiments concerning perceptual experience, intention, and
action. These include some of those that O™Regan and No¨ also discuss “
e
including the left-right inversion prism experiments of James Taylor
and Paul Bach-y-Rita™s studies of TVSSs “ but also Ivo Kohler™s classic
experiments with blue-left and yellow-right inverting goggles and cases
of asymmetrical bodily and perceptual neglect.17
Expanding Consciousness 237

Particularly relevant is Hurley™s discussion of the spatial inversion
experiments that Taylor performed using the mathematician Seymour
Papert as a subject. In these experiments, the subject was ¬tted with left-
right inverting goggles, and then trained on a range of tasks that required
moving his body through the actual (uninverted) world. Taylor reports
that Papert adapted to the wearing of goggles ¬tted with these lenses to
the extent of being able to ride a bicycle with them on, and also readily
adapted to the removal of the goggles, even while riding the bicycle. As
Hurley recounts, to perform successfully on such tasks requires calibrat-
ing a range of cognitive systems, including those for motor control and
for proprioceptive feedback and control, for now we have a subject act-
ing in the world, not simply recording information about it visually and
reporting that information. To the adapted subject, the world appears
the right way up, and his performance on a range of spatial tasks suggests
that his adaptation is not simply visual but proprioceptive, intentional
(in the sense of concerning the agent™s intentions), and motoric as well.
What has adapted is the whole subject, or at least cognitive systems that
cut across the perception-action divide of traditional views of perception.
One conclusion that Dretske draws in his argument for global exter-
nalism can be expressed by saying that visual experience is metaphysically
determined in part by the thoughts and language that we use to describe
it. We can draw a similar moral from left-right inversion cases: that visual
experience itself is metaphysically determined in part by what one does.
More speci¬cally, visual experience lies in the coordination and integra-
tion of visual “input” and behavioral “output,” in the exploitation of visual
information for the direction of behavior.
Hurley also considers a variation on the actual experiment that places
the subject in “Mirror Earth,” an environment in which there are left-
right reversals that are, in turn, reversed by the left-right inverting lenses.
This scenario parallels Ned Block™s “Inverted Earth” thought experiment,
which Block uses to challenge externalist, particularly representationalist,
views of the nature of consciousness. What is distinctive about the Mirror
Earth thought experiment that Hurley offers is that, as with the left-right
inversion experiments, it involves a “where” rather than a “what” system.
As such, this Mirror Earth case allows “ indeed, requires “ the subject™s
exploration and manipulation of things in his environment. Since the
inverting lenses correct only for the worldly inversion as it is perceived
visually, further inversions are required if the subjects are to remain in
the same internal states through their respective explorations of their
environments. As Hurley points out, unlike the case of Inverted Earth,
Thinking Through and Beyond the Body
238

which involves color inversion, this case of Mirror Earth requires that
motor output and proprioceptive feedback be systematically inverted in
order to have a case in which we have two individuals identical in their
intrinsic, physical properties.18
Part of Hurley™s point in considering the dynamic Mirror Earth variant
on the actual left-right inversion experiments is that it is in fact very
dif¬cult to satisfy the assumption that an individual on Mirror Earth can
be a duplicate of an individual on Earth, since the requisite inversions in
the dynamic case seem to require changes in the subject himself. It is with
this Mirror Earth case in mind that I want to turn to some broader issues
about consciousness and philosophical methodology in the next section.


7 individualistic residues
Both the sensorimotor contingency theory and the dynamic cycles con-
ception of perception-action exemplify how a TESEE view of visual expe-
rience can be developed. Together with representationalist views of visual
experience of the type that we have seen Dretske articulate, they provide
a basis for challenging much that has been taken for granted about vi-
sual experience that has suggested an individualistic view of it. I think
that the root assumptions at play here are that (a) particular states are
phenomenal in and of themselves, and that (b) phenomenal properties
are intrinsic properties of the states that have them, or (at least) of the
subjects of those states. The systemic view of taxonomy and individuation
introduced in Part Two provides a general reason to be skeptical about
(a), and the TESEE view of visual experience provides a positive frame-
work that either does without (a) or tempers its signi¬cance. Insofar as
particular visual experiences are phenomenal, they are part of an em-
bodied system as it dynamically interacts with a particular environment
and with an agent™s motoric, intentional, and proprioceptive capacities.
The rejection of (b) amounts to a rejection of “qualia,” or at least one
signi¬cant feature of them, but it is important to see that the TESEE view
is not eliminativist about visual experience, as other critiques of qualia
have been. There are phenomenal properties that we have ¬rst-person
access to, just as there is mental content to which we have such access.
But in both cases, these properties are not what we might have thought
they were. In particular, they are externalist, rather than individualistic,
in nature.19
What of the TESEE view and the remaining phenomenal states, those
of pain and bodily sensations? In section 5, I argued that Dretske™s
Expanding Consciousness 239

argument for global externalism was weakest in the case of these sorts
of phenomenal states. This is because their representational character is
not necessarily subject to the social deference that drives taxonomic ex-
ternalism, and they are realized in entity-bounded systems, and so are
not subject to an appeal to locational externalism. These points make it
prima facie less plausible to apply the TESEE view to these kinds of phe-
nomenal states across the board, even if many of the representational
characterizations we naturally give to such states follow those of visual ex-
perience and are at least taxonomically externalist. This constitutes one
sort of individualistic residue that remains once we push externalism as
far as we can, from the intentional to the phenomenal.
But there is another such residue, a sort of legacy of a philosophi-
cal tradition heavily populated by thought experiments. Philosophical
discussions have made it seem easier than it in fact is to articulate an indi-
vidualistic conception of phenomenal states, and correspondingly made
externalist views appear more counterintuitive than they in fact are in
part because of their reliance on what can and can™t be conceived in cer-
tain kinds of thought experiments. For example, a familiar line of attack
on materialist views of phenomenal states is based on the claim that one
can conceive of the possibility of either inverted or absent qualia. The
paradigm case discussed in such arguments is that of color experience
and its inversion or absence. If all the physical facts about an individual
are ¬xed yet it is possible for that individual to have either inverted or no
color experience at all, then physical facts (about an individual) do not
determine all of that individual™s properties.
The classic response to such arguments, one that opens the door to
the sort of argument that we have seen Dretske make for global external-
ism, is to point to the causal (and so, for some, conceptual) connection
between qualitative states and the ¬rst-person beliefs that they generate.
As the philosopher Sydney Shoemaker says, to think that absent qualia
are possible is “to make qualitative character irrelevant both to what we
can take ourselves to know in knowing about the mental states of others
and also to what we can take ourselves to know in knowing about our own
mental states.” And qualia severed from ¬rst-person knowledge of them
under at least some circumstances constitute some sort of conceptual
confusion. It is for this reason that Shoemaker holds that while inverted
qualia are possible, absent qualia are not. There are two points to make
about the dialectic here.20
First, such thought experiments themselves presuppose both (a) and
(b), for they hold, ex hypothesi, all physical or functional facts true of an
Thinking Through and Beyond the Body
240

individual while varying the phenomenal state: Green is the phenomenal
state present rather than red (or nothing), or pain is present in one
individual while it is absent in her physical or functional duplicate. Insofar
as the response that Shoemaker offers shares these assumptions, it too is
problematic.
Second, it is easy to radically underdescribe inverted and absent qualia
thought experiments, and misleading to draw substantive conclusions
on the basis of such underdescription. This is just the sort of point that
Hurley has made against Block™s appeal to Inverted Earth. Such thought
experiments take us so far, but only so far, in telling us something about
the nature of the mind and how we can and should conceive of it.21


8 global externalism and the tesee view
While I have rejected Dretske™s argument for global externalism, one
might wonder about how great the distance is between global externalism
and the TESEE conception of consciousness. The appeal to the preceding
arguments of Parts Two and Three shares in Dretske™s representationalist
strategy of arguing for an externalist view of consciousness on the basis
of assuming externalism about intentionality.

<<

. 38
( 62 .)



>>