Being No One with Thomas Metzinger

Being No One with Thomas Metzinger

October 9, 2019 100 By Stanley Isaacs


(gentle music) – This year’s first
lecturer, Thomas Metzinger, is a leading philosopher of mind working on consciousness and subjectivity, whose central thesis is
that there is no subject. I will let him tell you
more about his ideas. I was thinking over his accomplishments and thinking what might be
good for me to tell you now by way of introduction, and I
thought that I would tell you something that you might
not know, which is that Thomas Metzinger is a community activist in a distinctively
philosophical sort of way. Philosophy, unlike science,
is not a team sport, but it is a collective
undertaking nonetheless. Philosophers need teachers,
they need interlocutors, they need readers, they need an audience, they need a community. Thomas Metzinger has done
much work establishing philosophical community
where there was none before, and I wanna just give two
examples, I can think of many, and I had to pick and
choose among the examples to give you just two. If you’ve done work in
philosophy at the frontier of philosophy and neuroscience,
then you will know that that frontier can
sometimes be a fairly cold and inhospitable place. Well, that certainly was true. It is much less true today,
and I think in good measures as a result of the works
of Thomas Metzinger. And in particular, I’m
thinking of his work establishing the international society, the Association for the
Scientific Study of Consciousness, which seeks to further
international collaboration between philosophers and scientists. A little bit closer to home, closer to Thomas Metzinger’s home, Thomas has, under the
auspices of his office as professor of philosophy,
has brought together young philosophers and neuroscientists working in the different
corners of Germany in relative isolation on
topics at the intersection of philosophy and neuroscience. He’s brought them together,
provided opportunities for interaction,
discussion, collaboration, and thus established community. I think you can safely say
that today philosophical, there’s a real critical
mass of work going on in philosophical
neuroscience, especially among young scholars in Germany. And there wasn’t before,
and that is thanks to Thomas Metzinger. Now Immanuel Kant, in the
Critique of Pure Reason, wrote that if you measure
the length of a book not by how many pages it has, but by how difficult it is to read, by how much time it takes to read, then many a book wouldn’t
have been so long if it hadn’t been so short. I’m sure we can all think of examples. Thomas Metzinger’s recent
book, Being No One, published by the MIT Press,
is not an example of this. It is not a short book. Weighing in at something like 700 pages, it is a very long book. However, it is a book of
which one can say, I think, that this is a book that
wouldn’t have been so short if it hadn’t been so long. It wouldn’t have been so
accessible, so available to writers working and
thinkers coming from different traditions, it wouldn’t
have been so influential, it wouldn’t have been so
excellent if it didn’t exhibit the kind of muscular scope which is in evidence in its length. So on that note, and
without further delay, I introduce to you
professor Thomas Metzinger. (audience applauding) – Thank you very much, Tony. Thank you very much, Alva, for
your beautiful introductions. You must have all noticed
how full of exaggerations Alva’s introduction was. I’m very grateful for this invitation. Let me say this. And I feel deeply honored
for you to permit me to speak to you today. And it’s always a pleasure
to come back to Berkeley. When I got this invitation,
I thought, Thomas, when actually were you the
first time in Berkeley? Then it appears it was more
than a quarter century ago. It must have been September 1977. I had just finished high
school, grabbed my backpack, and flown into Montreal, Canada. And I hitchhiked down to
New York, came back up, hitchhiked all the way through
the Trans-Canada Highway and came down here from Vancouver, and somebody dropped me,
pitch-black dark night, 10 o’clock at night on the freeway. And the next thing that
happened is a car went over my backpack and tore everything apart. And as I was there on
the freeway at 10 o’clock at night gathering my belongings, my toothpaste here, my sleeping bag there, my camera there, the next car
that came was a policeman. So I ran to the man and
said, hey, hey, listen, somebody just destroyed everything I have, and it’s actually easy, it’s
the next car down the highway. Can you help me find these people? And he said no, you get $100 ticket for pedestrian on freeway. And to be quite honest
with you, I spent almost all of last Sunday trying
to find this notice to appear in court for you,
but I actually didn’t find it. And the next day, for the first time, I worked the Berkeley campus, and I saw this policeman
walking the campus in a really relaxed way,
swinging his, how do you say, riot stick or something like that in a really relaxed way, and
he had this nice clean uniform and a beautiful cap and hair down to the belt of his trousers. And I thought that is really convincing. That would not have been
possible in Germany in 1977, and I first realized there was something about Berkeley that I like. And so I’m very happy to be back here. How many of you, do you
know the story of Mary? Which was actually invented
by an Australian colleague of mine, Frank Jackson, in 1982. It’s the second half
of a thought experiment that actually goes like this. Mary is the best neuroscientist
mankind has ever had. She knows everything you
can know about the physical, neuro-computational,
what-have-you underpinnings of conscious color vision. The first premise of
the thought experiment is as strong as you may want to have it. Neuroscience has come
to its historical end. Mary knows all the
physical facts underlying conscious color experience. The second premise of a thought experiment is she’s never had a
color experience herself. Through some allergy or
something, she had to live in an achromatic prison under the earth. Everything was black and white
and she did all her studies through the internet with
a black and white monitor. In short, she has never had a conscious color experience herself. She doesn’t have any
first-person knowledge. Now the question is what happens if Mary leaves her achromatic dungeon and for the first time sees
the blueness of the sky and a red apple on a tree? Does she learn anything new about reality? For most people, it seems
obvious that the answer is yes. She learns new facts
she didn’t know before. For instance, facts about the
minds of other human beings. She knows suddenly what
her subjects she conducted experiments with via her
black-and-white monitor have been talking about all the time when they talked about
their red experiences, blue experiences and so forth. But if this is true, says Frank Jackson, there is a hole in a scientific worldview which cannot be repaired, because there are phenomenal
facts, non-physical facts. Here is how the argument goes. Before leaving her achromatic
prison, Mary knows everything that can be known physically
or neuroscientifically about the conscious color
vision of human beings. When first viewing a
colored object, she acquires new knowledge, and this
knowledge is factual knowledge. Here’s our man. Therefore, Mary, before
having her first conscious color experience did
not know all the facts one can know with regard
to color experience. Therefore, there are
non-physical facts, for instance, about conscious human color
vision that can only be grasped by phenomenal knowledge,
by first-person knowledge. And therefore, physicalism is false. I don’t buy this. And I don’t want to
speak about the knowledge argument today, but what
I want to point out is, is that there’s a very
important central term, the notion of a first-person perspective, and the notion of a self from which this first-person perspective originates. And in philosophy, it’s
very fashionable these days to say things like third-person facts are not reducible to first-person facts. But nobody ever asks what a
first-person perspective is and what a self actually is,
and this is what I will try to do together with you
today, and I also give you two promise, at the end, I
will come back to the question if there’s any deeper meaning
in all this popular talk you may have heard, that
the self is an illusion, or that no such thing as a self exists. And of course also as
this is our first lecture, I will come back to the question of the immortality of the soul. If we want to understand what
a consciously experienced first-person perspective
is, there are three phenomenal target properties
we have to understand. The first one is the property of mineness. A phenomenal property is something that exists in consciousness. In the Anglo-Saxon
world, you also call this the sense of ownership. Mineness is a higher-order
property of particular forms of phenomenal content. And here are some examples
how we refer to this private, subjective property from public space by using linguistic representations. We say things like I
experience my legs subjectively as always having belonged to me. I always experience my
thoughts and my emotions as part of my own consciousness. Voluntary acts are initiated by myself. Let me give you first brief example. And forgive me as I take off my jacket. This is something from a very
recent August issue of Nature. Have you ever, you can do this at home, heard of the rubber hand illusion? Here is how it goes. Activation of the premotor cortex during the rubber hand illusion. In the illusion, normal
individuals experience an artificial limb, a rubber hand, as if it were part of their own body. The subject observes a
facsimile of a human hand, the rubber hand, while
one of his own hands is hidden from view. Both the artificial hand and
the subject’s hands are stroked repeatedly and synchronously with a probe. The green and yellow
areas indicate the tactile and visual receptive fields
respectively for neurons in the premotor cortex. You see, something shifts. The subject experiences an
illusion in which the felt touch is brought into alignment
with the seen touch. This brings the visual
receptive field into alignment with the rubber hand,
resulting in activation of premotor cortex neurons. This is just an illustration. I will not go into any technical details, but that’s what I’m interested in. The moment where you
suddenly have the feeling this is my hand, although you
cognitively know it’s not. The property of ownership,
and as you see today, we have good nuts and
bolts stories are appearing about what actually happens in the brain. This is why we are living
in such an interesting time. You see how these people
can clearly delineate per certain areas of premotor
cortex where you have a shift of functional properties in the
moment where the rubber hand becomes a part of your experiential self. And you can do this at home, try it. The thing is, you must
stroke synchronously and you must stroke at
least 30 to 60 seconds, and you must make your
subject not look in your face, but must make your subject
really look on the probe or the brush or whatever you’re using. So much about mineness. Here’s the second target
property we have to understand. It’s the property of selfhood. Some crazy German philosophers call this (speaking in foreign language). (audience laughing) Pre-reflexive self-intimacy,
the way of being infinitely close to yourself
before you even start any thought or cognitive activity. Here are examples. How we speak about this, we
say things like I am someone, I experience myself as being
identical through time. The contents of my
phenomenal self-consciousness form a coherent whole. Before initiating and independently of any intellectual operations,
I’m already directly, whatever that may mean,
acquainted with the contents of my self-consciousness. And here’s the third
and last target property we must understand,
and that’s the property of perspectivalness. That is a structural property of your experiential space as a whole. It possesses an immovable center. Now here is the mystery, a mystery Thomas Nagel wrote a lot about. For each one of you, it is true that you are the center yourself. To be phenomenally aware means to possess an inward perspective and
to take on this perspective and the subjective experience of the world and of your own mental states. But if you say I am this center myself, you don’t really understand
what you’re saying. This is where the puzzle
occurs, when you flip from a third-person
description of a property of conscious space into a
first-person description by using a concept like myself. What I want to do now is I want to analyze these phenomenal properties
on the representational list and functionalist level on
lower levels of description. I want to ask what would
it mean for a given information processing system to instantiate these properties. Let us start by analyzing
these target properties. My first step will be to introduce
a new theoretical entity, that’s the phenomenal self-model or the conscious self-model or the PSM. I say something like that
exists, and it will be found on all levels of explanation. It’s a distinct theoretical entity. And at the very end of
this talk, I will introduce a second theoretical entity. It forms the representational
instantiation basis of the phenomenal, the
conscious properties that we want to explain. Let me give you some
ideas about what I mean by the notion of a self-model. It’s only episodically active, it’s a representational entity, and the content of that
representation is the very system in which it appears. From a logical point of
view, you can distinguish three classes of information
processing systems. Some can do simulations. Think of the meteorology
department, a computer that simulates weather
movements, cloud movements. Then there are more complex machines, machines that can emulate
non-observable properties of another computer. For instance, if on your Windows desktop you make the pocket calculator come up, or if a very clever Turing machine pretends to be a stupid Turing machine, then you have the situation where one information processing
system emulates another one. It emulates the internal information flow. What I am saying is that we are a system that do both of these things. We simulate and we emulate. And if you have the special
case that target system and processing system are identical, then you have the case of self-modeling. So what I’m saying is that
you all as you’re sitting here are systems that simulate
and emulate themselves for themselves as they
are listening to me. A background assumption
is that this self-model possesses a true
neurobiological description, some activation vector. It is some complex activation
pattern in the human brain, but I will not talk about this
today being a philosopher. The phenomenal, the conscious
self-model is that part of the mental self-model
which is currently embedded into the highest-order
integrated structure, the global model of the world. This will be very easy to
see for all those of you are interested in psychotherapy. Human beings have an integrated
self-model in their brain. Not much of it is conscious,
but of course it is clear that parts of your unconscious self-model can have causal properties and influence, say, endocrinal output,
psychosomatic interactions as we call them. So there is a conscious
and an unconscious layer, and what is conscious is variable. As philosophers say,
the phenomenal content of the self-model supervenes locally. That is, for the experiential
content it is true that if all of your brain
properties are fixed, all of your properties of
your phenomenal self-model are fixed as well, which doesn’t yet imply that there’s a reductive explanation. The phenomenal self-model in
our own case is a plastic, structure changes over lifetime. It’s a multi-modal structure. Many different sensory organs,
from your blood vessels, from your vestibular organ,
and so forth feed into it. And possibly it evolved
from a partially innate and hardwired model of the
spatial properties of the system. Some British philosophers
have spoken about a long-term body image. You know the stories of
Damasio and Melzack maybe. More about this later. An active self-model it’s
important to understand this is not a little man in the head. It’s a subpersonal functional state. That is again philosophically speaking. It is characterized by
a distinct causal role, and from a strictly
analytical perspective, it’s a set of causal relations. And if anybody here still
believes in classical cognitive science, we might
say something like that. It’s a transient computational
module which is episodically activated by the system
in order to regulate its interaction with the environment. That’s a very complicated
way of saying what happens when you wake up in the morning. When you come to yourself, then
the organism which you are, that’s what I’m saying at least, has to achieve complex
sensory motor integration. You have to go to the
refrigerator or to the toilet. And then it meets this
transient computational model, the conscious self-model,
and it just switches it on. And this is the moment when you wake up, when you come into existence
as a conscious being. For those of you are interested in logic, there’s a formal proof
by Conant and Ashby 1970 that every complex system
that has a regulator that regulates its own
behavior will automatically by necessity turn this
regulator into a model of the system as a whole. That’s pretty intuitive. If you have to regulate
the different parts, you have to map them somewhere. I also make a teleofunctionalist
background assumption. The development of the
activation of this module plays a role for the system. It’s good to have it
in pursuing your goals. The functional self-model processes a true evolutionary description
that is, it was a weapon which was invented and
optimized in the course of a cognitive arms race. And that rather unromantic
quote comes from Andy Clark from his 1989 book. So the idea is that what we have now as our conscious self-model is something that has a long history and emerged out of a basically competitive process. It consists in a very specific achievement the capacity not only to
open representational spaces but to open centered
representational spaces. Space is centered around
a model of the self. And before all this gets too
boring, I want to give you two low-level examples of
what I mean by a self-model. Astronauts in space frequently get the following problem. They cannot feel where up and
down is in their body anymore. It’s like a radicalized
version of motion sickness. It’s difficult, if you’re trying to eat, if you can’t feel top
down in your body anymore. Every astronaut knows
how to help his buddy if he has that problem,
you just hit very hard on the heel from below
on the sole of the shoe. And instantly the body
image locks in again and there’s this conscious experience. This is down, this is up. And every astronaut knows
how to annoy his buddy, hit them on the head right afterwards. So what that shows is
that the human self-model is just a virtual model. It depicts a possibility, philosophically very interestingly as a reality. It’s just the best
hypothesis the system has about its own current state. And if it is under constraint by input, which it is in a spaceship,
then it can become highly context-sensitive. You just have to knock
on the back of the foot and it locks in, and
down suddenly is there. So the self model is a
simulation, a virtual model. Now I guess many of you
have heard of San Diego’s Ramachandran and phantoms in the brain and the many patients he has. And I used to go to lunch
with him when I spent a year at UC San Diego in ’89, ’99. And how many of you have heard of these mirror experiments,
mirror synesthesia? You have these people
have an amputated arm and they have a phantom limb. And usually that phantom
limb will go away within a couple of weeks with a
so-called telescoping effect. The phantom limb could
become smaller, smaller. You can make a fist in the
stump and then it disappears. Some people, however, have
a hurting phantom limb that is paralyzed and
stays up to 12 years. I mean, how do you cure pain
in a non-existing limb, right? If you want to help
these patients, one knows the first thing you have
to do is you have to regain volitional control, you
have to make it mobile. But it doesn’t work. So the task Ramachandran
did with one of his patient who had this paralyzed
phantom limb for 12 years was now make butterfly-like
symmetrical movements like this, what is your
conscious experience? And a patient says, well, Doctor, what is my conscious experience? My good arm moves, the
phantom limb is paralyzed. Now what you do is you put
a mirror down in the middle and say, can you please
do the same thing again, and look at the mirror from the side. And the patient will
exclaim, Doctor, Doctor, my phantom limb moves. I can move my phantom limb for
the first time in 12 years. You pull the mirror up
or you tell the patient close your eyes, and
with great disappointment they will say, oh, it’s frozen again. What moves in that experiment is what I call the phenomenal self-model. Here you have the setup. Technically speaking, you
install a virtual source of visual feedback which
points into just the right region of state space
where the system sends its motor commands and
never gets feedback. And you see something is moving there. It’s the phenomenal self-model. Now let us start to continue our representationalist analysis of our three target properties. Remember the rubber hand illusion. What is mineness? All representational
states which are embedded into the currently active self-model gain the additional higher-order
property of phenomenal mineness. That is a nonconceptual
sense of ownership. Nonconceptual, something
that a nonlinguistic creature could have as well. An animal could have this as well. If this integration process is disturbed, different neuropsychological syndromes or altered states of consciousness result. Here are some examples. Consciously experienced thoughts are not my own thoughts anymore. You have that in florid schizophrenia. You probably have all heard about this. This is one of the main
problems schizophrenics have. If you cannot integrate your
own cognitive processing into your own self-model
anymore, you cannot experience your own thoughts as your own anymore. How many of you know what
unilateral hemi-neglect is? Happens every day in
hospitals around the world. Somebody wakes up after surgery, thinks the medical students have made a particularly distasteful joke, finds a dead leg in his bed, tries to throw the leg out,
falls out of bed with it, realize the dead leg is
grown to his own hip. The nurse comes, he
says, this is not my leg. The nurse will say, but then
you should have three legs. Can you count? Patient will say, I cannot explain this, but this certainly is not part of my body even if it is fixed to me. This is something that is very
well documented and studied, and it’s exactly the sense
of ownership that is missing. In Alien Hand Syndrome, my
arm performs goal-directed actions without my own control. The interesting thing about that syndrome is that you have goal-directed actions, typically with a sense
of intermanual conflict. The Alien Hand Syndrome
patient will pick up the phone and the other hand will
slam the thing down and interrupt it. One hand will button down the shirt, the other hand will button it up. The interesting thing is
that some patients even start to attribute
personality to that limb because it’s so obviously goal-directed. It’s clear you can have
goal-directed motor behavior which cannot be integrated into your conscious self-model. Every day people come into emergency wards and say things like this. Doctor, I am a robot. I am transformed into a mechanical puppet. My volitional acts are not my
own volitional acts anymore. You have that in depersonalization. You feel remote controlled and you lose what German philosopher and
psychiatrist Karl Jaspers called (speaking in foreign language), executive consciousness,
the phenomenal experience only of initiating an act,
but of carrying it through. But then you also have
states where people say I am the whole world. All events in the world are controlled by my own volitional acts. So to speak, the
self-model can also expand to the boundaries of the world. And a couple of years I gave a talk in a psychiatry institute in Germany, and they had two patients there. One of them was standing
by the window all day making the sun move, experiencing every little
movement as self-caused. And the other guy was
looking down into traffic making the cars drive, the puppets walk, turning the traffic lights on and off. So there are phenomenal
states in which every event that is consciously experienced is experienced as self-caused. And of course every
modern philosophy of mind has to account for all these phenomena or it is not an
interesting theory of mind. Let us move on to the
core property of selfhood. What is it? It’s the existence of a single coherent and temporally stable
self-representation which forms the center of the overall
representational space. And again, if this
representational module is damaged, if it disintegrates or
if multiple structures of this type alternate within the system, you get different
neuropsychological syndromes, different altered states of consciousness. Here are some examples for you. In Anosognosia and
Anosodiaphoria you have a loss of higher-order inside
into existing deficits. For instance, in blindness denial after a massive cortical
infarct, Anton’s syndrome. You will have people
who are totally blind. They will never see
anything in their life. Not because anything is
wrong with their eyes, but because something is
wrong with their brain. They show all functional
signs of being blind, they bump into walls,
they fall over furniture, but they claim to still be seeing people. If you say how many fingers are these, they say, oh, it’s kinda
dark in here and I don’t, I’m not interested in your experiments. And if you take your keys
and you say how many fingers, you know, what is this? And they say, well, it’s keys, you know? And they start to confabulate. I’ve had the most interesting discussions with German (mumbles) in my own country who claim that this doesn’t exist because it’s logically excluded that something like this exists. As a matter of fact, it
is documented since 1899 in many clinics around the world. It’s a well-understood phenomenon. But there still are philosophers who say this is logically impossible. The opposite is true,
and modern theory of mind has to explain it. You may have all heard of
multiple identity disorder. It now has been renamed into DID, Dissociative Identity Disorder. I do not want to go into
this in detail today, but in this framework, it
is of course easy to imagine how a system uses different
and alternating self-models in order to deal with
extremely traumatizing and socially inconsistent situations. Most DID patients have been
sexually abused by a parent, and it’s very clear that if a love object becomes an aggressor
suddenly, you may have to do something like a
division of emotional labor. You have to divide the emotions you have to two different self-representations and create an amnesia in between, else you can’t go on living with that. But this shall be enough for today, and then you have what in Germany we call (speaking in foreign
language), hard to translate, identity disorders,
delusional misidentification. There’s a large class of
psychiatric disturbances which is accompanied by deviating forms of the conscious experience
of one’s identity. That is about how you experience it. That doesn’t as such touch
the philosophical issue of personal identity. Now what is it to have a
first-person perspective? The existence of a single, coherent, and temporally stable
model of reality, one world which is representationally
centered around or on a single, coherent,
and temporally stable phenomenal subject that
is not only a self, but a model of the system as experiencing. This is not circular. I will come back to this point in the end. That is if you have a model
of the system as being tied to objects in the world
through perceptual relations through grasping behavior
or something like that, then you get a perspective. The structural feature of the
global representational space then leads to what I call a nonconceptual first-person perspective. Again, please note, you
do not have to be able to think thoughts or speak
a language to have that. A dog that has visual
attention will have that. If this global representational
property is lost, again, the phenomenology
will change, and you will get different neuropsychological syndromes or altered states of consciousness. Again, I’ve brought examples for you. You can have complete depersonalization. You can lose the phenomenal
first-person perspective, and this can be accompanied
by dysphoric states, by fear reactions and functional deficits. There are some psychiatrists in Zurich who have developed a
big statistical battery, and they have tested very
different types of altered states of consciousness, drug-induced,
psychosis-induced, through rhythmic drumming,
many different stimuli, and have found what the
etiology independent functional clusters are
in these altered states. And there’s one very strong
cluster which they call (speaking in foreign language),
dreadful ego-dissolution. The interesting result,
however, is that there are also experiences which are described
as selfless and noncentered later by the people undergoing them, which are experienced and described as non-pathological and non-threatening. These Swiss researchers termed
this oceanic boundary loss. And it is of course true that any modern philosophical theory of
mind will also not be able to shut its eyes in that direction. I will very, very slowly
start to come to the end of my talk now by asking
the two major questions. What is the central theoretical problem on the functional level of description? And then I will ask what is
the central theoretical problem on the representationalist
level of description. And I will briefly give you my own answers and we can discuss this. Here’s the question, in which way does the phenomenal self-model
differ from all other phenomenal models currently active? Which functional property
characteristically marks it out? How does it precisely become the center of the phenomenal space of representation? I mean, there’s everything
here, bottles, laptops, flowers, and the
phenomenal model of myself, why is it the immovable center? Here’s my answer, the
self-model is the only representational structure
which is anchored in the brain by a persistent functional link. That is, by a continuous
source of internally generated input, of input you generate yourself. Whenever you have anything
like conscious experience at all, that is whenever a
stable and integrated model of reality is there, you also
have this continuous source of internal proprioceptive input. There is something in your
conscious experience now which is so invariant that it is almost unconscious. And it has something to do with the background
sensation of your own body. There’s a man at McGill
University in Canada, Ronald Melzack, who claims
that there’s a hardwired partition of the neural matrix underlying the spatial model of one’s own body which is independent of external input, and that becomes the center. That is, you have a
part of your body image which is autonomously active, and 50 times a second
tells you, this is me. Here are two empirical
hypotheses from a philosopher. But you see, they’re stolen. New results concerning pain
experienced in phantom limbs point to the existence of
a genetically determined neuromatrix, the activity
pattern of which could form the basis of the more invariant aspects of your bodily self-experience. I just talked about that. Melzack calls it the
phylomatrix of the body image. So it’s something we
share with our ancestors. Then there is the Damasio
story, which I’m sure many of you will know. There’s a second empirical
hypothesis which says that the homeodynamics as
regulated by upper brain stem in the hypothalamus generates
this continuous source of input by which you feel yourself. I have termed this emotional embodiment, the way you are anchored in
your emotional background state. Here’s a quote from Ronald Melzack. He says, in essence, I postulate
that the brain contains a neuromatrix or network
of neurons that in addition to responding to sensory
stimulation, continuously generates a characteristic pattern
of impulses indicating that the body is intact and
unequivocally one’s own. He says, I call this
pattern a neurosignature. If such a matrix operated in
the absence of sensory inputs from the periphery of
the body, it would create the impression of having a limb even when that limb has been removed. Here’s a woman, Azet, a patient
from Zurich, Switzerland from a department,
people I cooperate with. And this person was born
without arms and legs. She’s an academic,
absolutely cognitively lucid, 41 years old. They told her please rate
how real your phantom limbs feel to you compared to the
torso, to your real body, on a scale from zero to seven. Say your real body feels like seven, and I feel nothing is zero. And you see an interesting variance. You see some toes are 5.0,
another toe will only be 3.2. So these phantom limbs which
are there are not as real phenomenologically as
is the body that sends continuous input through
normal proprioception. But they are also there all the time. Being a philosopher I will
not take a position there, but the interesting issue
is if something like that could actually evolve postnatally, or if there is an innate
functional core to the body image. Intuitively, to many of us
who are not neurologists, it seems like if something
like that can happen, you must have a body image
you’re born with, right? There must be something
that is pre-structured. Some of my expert friends doubt that. Here’s a bit of Damasio. Of course, he says, and
he tries to go deeper. The brain does represent
muscles and joints and bones, but before it ever gets
to the muscles and bones, there are other aspects
of the body to consider, namely the viscera and
the internal milieu. The internal milieu
corresponds to the chemistries of the fluids in which all of our living tissues are immersed. Not only must the body
model and the brain include the latter aspect of the organism, it is likely that the model is anchored on those aspects
because they are indispensable for the maintenance of life. The dynamic structure and
operation of internal milieu and viscera are the beginning
of the body-minded brain. And he points out that
it is the first division, the one concerned with
the organism’s interior, which is permanently
active, that’s my persistent functional link, permanently
signaling the state of the most internal aspects of the
body proper to the brain. The brain is truly the
body’s captive audience. For a philosopher’s point of
view, my friend Tony’s theory has many conceptual problems. It’s not entirely coherent. But I think he’s got a
very good point there, because he can explain how self-modeling, the process of becoming
self-aware, is actually anchored in molecular level dynamics, in autonomous self
regulation of body fluids around the brain stem. And you can see also how far that is from what any artificial
system can do today. The time by which we have
robots with body fluids and that ultra fine-grained
emotional background state we have is I guess is a
distant point in time. Now I want to ask the
second and last question. What is the central theoretical problem on the representational
list level of description? Here’s the problem. There seems to be no
necessary connection from the functional and
representational properties to our conscious target
properties of mineness, selfhood, and perspectivalness. Somebody like Dave
could come along and say all this could take conceivably place without the emergence of
a genuine phenomenal self or a subjectively experienced
first-person perspective. We can imagine biological
information processing systems which develop and use centered
representational spaces without the emergence of
true self-consciousness. Somebody could say,
Thomas, you’re cheating with the word self. A self model is not a
self, but only an internal representation of the
system itself as a whole. It is a system model. So the question becomes,
how does one get from the functional property of centeredness and the representational
property of self-modeling to the phenomenal property of selfhood? Here’s my answer. The transparency of the data structures used by the brain. What is phenomenal transparency? A standard philosophical
definition will say that only content properties of certain representational structures
used by the brain are available to introspection. The vehicles, as
philosophers say themselves, the physical states employed by the system are transparent, that is,
they do not represent the fact that they are representations on the level of their content. This is an old philosophical notion. Goes back to G.E. Moore, 1903. Just look at these flowers here. We would say as philosophers, this is a phenomenally
transparent representation because you introspect
your perceptual processing as hard as you want to, you
cannot recognize the fact that this is all a state
in your visual system. You cannot see the
representational state itself, and that is why you are a naive realist. You have the feeling you are in direct and immediate contact with reality. We are systems that so
to speak look through their own representational
structures as if they were in direct and immediate
contact with their content. Now what you have to
do, now let me give you two empirical hypotheses,
why we are beings that are naive realists. The respective data
structures are being activated so fast and so reliable
that the system cannot recognize them as such anymore. And my idea on a small
time scale is simply because of a lower temporal resolution of metarepresentational functions. That is the earlier processing
stages, for instance, leading to this coherent
pattern here are not available for introspective attention. You cannot penetrate them in principle because they are just so much faster. That is, systems like
us cannot become aware of the construction process,
however hard they may try. I have a second empirical
hypothesis which operates on a much larger time scale. I think there has been no
evolutionary selection pressure on the relevant parts of
our functional architecture. For beings like us, naive realism has been a functionally adequate
background assumption, and it’s easy to understand that. We only needed to represent
the fact there’s a wolf there or there’s a bear there. We did not need to represent the fact, there’s an active wolf
representation in my brain now. That would have what the experts call a much higher metabolic price. You would increase the
computational load on the system, you would burn more sugar,
but that wouldn’t pay. It wouldn’t pay to burn
more sugar for that. At least that is my own hypothesis. For the functional properties
we needed to survive, we didn’t need to distance
ourselves from ourselves. Now the crucial point in
my talk is that you have to apply this point to the
notion of a self-model. I’m claiming that we are
systems which are not able to recognize their own
subsymbolic self-model as a self-model. Therefore, we operate
under the condition of what I would call a naive-realistic
self-misunderstanding. We necessarily experience
ourselves as being in direct and immediate epistemic
contact with ourselves. So to speak, as you are
listening to me right now, what I’m saying is that
metaphorically speaking, you are a system that
constantly confuses itself with the content of its own self-model. Now we can continue our
representationalist analysis and come to the very last question. We have an idea of what
consciousness could be maybe and an idea how a phenomenal
self-model emerges. But what makes conscious
experience subjective experience? Here is very briefly a
second theoretical entity. I call it the phenomenal model of the intentionality relation, the PMIR. And what I say is that we
are systems that co-represent the representational relationship
while they represent. That is, the PMIR is a
dynamical and transparent model of the self in the act of knowing. A PMIR is a continuously
changing inner representation of ongoing subject-object
relationships, that is, the relationship between
subject and perceptual object or subject and internally
represented action goal. We are systems of capably
of dynamically representing ourselves as selves
standing in relationship to object components in
the world or in our life. So if the phenomenal model
of one’s own perceptual state contains a transparent
representation of their causal history, then you will
have nested global states, the content of which the system itself can only truthfully describe
in the following way. It can say things like this. I myself, that is the content of
the currently active transparent self-model. I’m seeing this object, the content of a transparent
object representation, and I’m seeing it right now. Right now means as an element within a virtual window of presence, as an element of working memory. And I’m seeing it with my own eyes. And this with my own
eyes is the simple story the brain tells to itself
about direct sensory perception which sufficed for the
evolutionary purposes of the brain. Of course you don’t see with your eyes. I hope nobody of you believes that. You see with your visual system. But on the user surface so
to speak, on the top level, there is this abbreviated short story. The system tells itself
that it sees with its eyes. And it creates this little
avatar which you are, the phenomenal self-model. This avatar doesn’t know
it has a visual cortex. It just sees with its eyes. It doesn’t know it has a motor cortex. It just acts with its hands. So my final claim will be that to have a phenomenal
first-person perspective is to possess a transparent PMIR. And now I can come back
and keep the promise I gave to you at the beginning. Given all this, isn’t it true
that the self is an illusion? I think it is not true because it contains a logical mistake. On the level on which we are talking, there is no such thing
as truth or falsity yet. There is nobody who could have
an illusion in this system. So if you really wanted
to stay with the idea that the self is an illusion,
you would have to say that it is an illusion
which is no one’s illusion. And this brings me to the final point, to the question, as this is
this year’s first lecture, the Immortality of the Soul. If it is true that the self is not a thing but a process as I’ve described it, then it is also true that
the tragedy of the ego dissolves because strictly speaking, nobody is ever born and nobody ever dies. Thank you very much. (audience applauding) (gentle music)