and
the Immortality of Illusions:
The Subjective Perceptions of “Self” and “Continuity
of Consciousness”
and the Costs and Benefits of Preserving
These Perceptions Unto Eternity
Ben Goertzel
Novamente, LLC /
Applied Research Laboratory, Virginia Tech
National Capital Region
Many of us have the feeling that we would like to live forever – or at least to live a very long
time, without the fear of involuntary death, until such (hypothetical, perhaps
never-arriving) point as we decide freely that continued life is no longer of
interest.
My goal in this essay is to probe a little deeper into this feeling than is
usually done. What does it really mean
to want to live forever? Delving into
the meaning of “live” and “forever” isn’t that fascinating to me, though it does give rise to
some significant issues (to what extent is our sense of self tied to our
physical body? what are the odds that the physical universe containing us will
end in a Big Crunch or some similar calamity, making literal immortality
impossible?). What intrigues me more is
digging into the ”I” who wants to live forever. What exactly is it that’s being perpetuated, in the hypothesis of
eternal life? How can we tell whether,
in a particular scenario, this “I” is
really being perpetuated or not?
I take a lot for granted before I even begin this conceptual
investigation. I assume that human
immortality is physically possible, sociologically, economically and
psychologically practical, and morally acceptable. I realize that all of these points have been and still are
disputed by a variety of intelligent and thoughtful individuals, but the
arguments on both sides have been repeated many times and there would be no
point in my recounting them. It seems
very clear to me that pharmacology or nanomedicine will eventually enable human
physiological immortality, and that uploading will eventually allow humans to
copy their minds into computers of some sort.
It seems very clear to me that socioeconomic problems ensuing from
widespread human life extension would be solvable via deployment of appropriate
technologies –
e.g. overcrowding of the planet could
be solved at first via digging underground and building huge skyscrapers, then
later by colonizing other planets, building huge spaceships and/or uploading
into virtual realities. And although I
agree that much of the meaning of current human life ensues indirectly from the
inevitability of death, it seems to me that humans freed from the fear of death
would find new sources of meaning. All
these points, though currently controversial from a mainstream point of view,
seem obvious and not worth debating.
But the debates over these trivial points tend to obscure the deeper
issues that make the issue of immortality really interesting.
To illustrate the deeper issues I want to raise, I’ll first outline a few thought-experiments
regarding hypothetical scenarios that may be enabled by future technology. The general points I want to make are not
tied to these particular scenarios; the scenarios are just illustrative.
The first scenario involves a future version of myself – FutureBen, let’s call him – who is allowed to grow, learn and change
freely as he wishes. Suppose FutureBen
lives ten billion years and increases his intelligence by a factor of
ninety-seven quintillion. His human
body was shed after a few thousand years of life – and he’s placed the episodic memories of his first
century of life (the part that took place in humanly-embodied form) in a
very-rarely-accessed portion of his memory, since it’s really not very interesting compared to
some of the things that have happened to him since.
Now let’s
suppose President George W. Bush has also spawned a future version of himself – FutureBush. And suppose that FutureBen and
FutureBush made friends in their eight
billionth year and decided to link their lives together in a kind of posthuman
mind-partnership. I submit that, after
a billion years of collaborative, mutually-coupled growth and change, it may be
rather difficult to distinguish FutureBen from FutureBush. Perhaps they have even exchanged ancient
episodic memories, so that each of them has complete first-person memories of
the other one’s
life. They may be using the same black
holes in the same galactic cores as their wormhole-coupled quantum-gravity
cognitive processors. FutureBen and
FutureBush may be an awful lot like each other -- and very, very little like
Ben Goertzel or George W. Bush from 1985, 2005 or 2035.
The question is: What difference does it make that FutureBen happened to
evolve out of Ben Goertzel instead of George W. Bush, PeeWee Herman, or for
that matter, one of 2005-Ben’s guinea pigs Coffee or Tea? It
surely doesn’t
matter much to FutureBen. If I, 2005
Ben Goertzel, am simply going to serve as the initial state for a completely
different sort of being, then in what sense am “I” really becoming immortal? Why not just let myself die, and then let other sorts of beings
comparable to FutureBen and FutureBush take my place? Or why not, for example, die and let my children take my place,
and let their children take their place, eventually after N generations
resulting in completely different sorts of beings?
One possible answer is that this is not really immortality for Ben Goertzel
of 2005. The difficult of this answer
however is: Where do you draw the line?
2005-Ben is very different in many ways from 1985-Ben, and more so from
1970-Ben (who was 3 years old). Is it
the continuity of the physical body that’s critical? It’s easy to make counterarguments to
that, as my next hypothetical scenario will illustrate. Is there some critical threshold of change,
beyond which “selfness” is not preserved? Is there some special emergent “self-pattern” which exists in all the human Bens
mentioned above (emerging in different ways from the differently mature Bens,
but still maintaining its own integrity), but is lost in SuperBen? I think the latter is fairly close to the
truth – there
is an emergent self-pattern spanning all these human Bens which is not there in
the hypothesized far-future SuperBen.
The question then is: What is this emergent self-pattern? How real is it? Why is it important?
These are critical questions that I will return to a little later.
Another possible answer to the FutureBen question is that this scenario is
real immortality for 2005-Ben, and that there is some sort of value in the
continuity of consciousness between 2005-Ben and FutureBen. In this view the immortality lies in the
process of continuous awareness, even if this includes continuous radical
growth and change. Continuity of
awareness is posited as a primary value, right along with awareness and life
itself. This is a philosophically
respectable view, but it gives rise to further subtle questions. What is this continuity-of-consciousness and
why it is important?
Following up on the prior
thought-experiment, I will now introduce some additional hypothetical
scenarios, intended to highlight the philosophical subtleties associated with
various uploading techniques. These
scenarios get directly at the issue of continuity of consciousness raised at
the end of the prior section. I’ll define a number of different
NearFutureBens (NFBens), and the question is which ones have a greater claim to
Ben-ness.
NFBen1 is produced via recording
complete information about the physical parameters of all particles inside Ben
Goertzel at a particular point in time, storing this information in a database,
then completely annihilating Ben Goertzel – and then, 666 hours later, re-creating an
exact copy of the previously annihilated Ben Goertzel from the database
records.
NFBen2 is produced via making the
same kind of recording and re-creation, but doing the re-creation 10
microseconds after the annihilation, rather than 666 years.
NFBen3 is produced the same way but
with a delay of 3 femtoseconds.
NFBen4 thru NFBen7 are produced like
NFBen1 thru NFBen3, but the reconstruction takes place inside a computer rather
than in physical reality. These Bens
are software, though they may feel like they have physical bodies, due to
having the option to cruise around in Ben-like bodies in an Earth-like virtual
reality. When embedded in the virtual
reality, they feel like physical Bens, but they know they’re “just software.”
NFBen 8 thru NFBen10 are produced
like NFBen1 thru NFBen3, but the reconstruction takes place differently: Ben’s physical body is re-created
correctly particle by particle, but Ben’s brain is replaced by an Asimov-style positronic brain
or a digital computer that contains all Ben’s memories and realizes Ben’s thoughts and feelings exactly in
spite of having a different physical substrate.
Next, imagine variant NFBen11, who
is an exact particle by particle replica of regular old Ben, but with one
difference: the particles from the original Ben are gradually transplanted into
him. First NFBen11 is created by the
same method as NFBen3, then one by one the actual cells from the original Ben
are exchanged with the cells in NFBen11, until eventually all the original Ben’s cells are in NFBen11, and all the
original NFBen11’s
cells are in what used to be the original Ben.
All of these NearFutureBens will
feel like they are Ben – they will have Ben’s memories and Ben’s self and will feel like their consciousness is continuous with Ben’s.
They will have as much right to feel like Ben as I, writing these words,
have to feel that I am the same Ben who went to sleep last night. I went to sleep, disappeared, and then woke
up in the morning – a new
person, with a feeling of having the same self as the guy who went to sleep in
my bed last night, and a feeling of having a sort of temporarily interrupted
continuity of consciousness with that guy.
One can argue that all these NFBens
really are Ben. Or one can argue that
some of them are just new people who feel like they’re Ben.
My sympathy is strongly with the former position. The distinction between the Bens who look
and feel and think and act exactly like Ben, and the Bens who actually are Ben,
feels to me like a mystical distinction-without-a-difference. If this is accepted, then what it means is
obvious: Ben is uniquely identifiable via a certain a pattern of arrangement. Ben is not uniquely identifiable via a
particular collection of particles, nor even necessarily by a pattern of
arrangement of particles – he is uniquely identifiable via a pattern of arrangement that may emerge
from particles or bits or potentially anything else.
Note that I’m not committing to a completely
reductionist view of the mind here. I
tend to think there are certain aspects of subjective experience that aren’t well-captured by the scientific,
reductionist perspective. I wouldn’t want to quite say that “Ben is a pattern of arrangement.”
But I would say that from the perspective of looking at various physical
systems and deciding which ones are Ben or not, the pattern of arrangement is
all that matters. The subjective view,
from inside Ben, is at least in part another story – but we’ve already posited that, in these
scenarios, all these NearFutureBens subjectively feel just like Ben.
So what does immortality mean, really?
One case is absolutely clear: If a person maintains their human body
forever, and doesn’t
alter their ways of thinking and feeling too much, then they will “live forever” in the same sense that a person now lives
60 or 75 or 90 years or whatever. This
doesn’t
solve any of the philosophical puzzles, however, it just defers them to
commonsense.
I think it is worthwhile to question the meaning of the commonsensical
sense in which a person now lives 60 or 75 or 90 years or whatever. We change immensely over our lifetime, and
we disappear nearly every night and reappear in the morning – with what justification do we say
that it’s the
same person?
I do believe there is some continuity of structure there – I have many traits in common with
my 1970 self, and there are in particular many aspects of my self-model
that are still there after all these years.
At a high level of organization, Ben is still Ben, even though the
particular memories and ideas out of which this high-level organization emerges
have changed a lot. And this high-level
organization is linked in with a bunch of physical and emotional peculiarities
that haven’t
changed over the years. There are many
continuously existing patterns there, unlike the Ben vs. FutureBen contrast
made above.
There is a lot of neuropsychological research showing that the “self” is in a strong sense an illusion – much like its sister illusion, “free will.” Thomas Metzinger’s recent book Being No One (2003)
makes this point in an excellently detailed way. The human mind’s image of itself – what Metzinger calls the “phenomenal self” – is in fact a construct that the human mind
creates in order to better understand and control itself, it’s not a “real thing.” Various neuropsychological disorders may
lead to bizarre dysfunctions in self-image and self-understanding. And there
are valid reasons to speculate that a superhuman mind – be it an AI or a human with tremendously
augmented intelligence – might not possess this same illusion. Rather than needing to construct for
itself a story of a unified “self entity”
controlling it, a more intelligent and introspective mind might simply perceive
itself as the largely heterogenous collection of patterns and subsystems that
it is. In this sense, individuality might not survive the transcendence of
minds beyond the human condition.
But, illusory or not, the patterns of my human self-model have largely
persisted in me since early childhood – and in that sense “I” do have some persistent existence, even if this “I” is really an illusory phenomenal self a la Metzinger.
Next, in addition to the continuity of structure, there is a perceived
continuity of consciousness. I feel
like I’m
participating in a largely continuous stream of thought and feeling from one
minute to the next, and when I wake up in the morning I often (not always) feel
like I’m
resuming the stream of thought and feeling from the night before. Sometime this continuity is quite vivid – as when I fell asleep thinking
about some problem, and I’m still thinking about it when I wake up.
Other times the perceived continuity is hardly there at all, as when I’m traveling and I wake up in a
hotel room and have a hard time remembering where I am or how I got there,
until I fully re-emerge into consciousness.
My contention is that, just like the phenomenal self, the continuity of
awareness is also a psychologically-constructed illusion. I have a memory of my prior thoughts and
feelings, and so I construct within myself a story of continuous flowing from
these prior thoughts and feelings to my current ones. But this process is not unlike how I construct within myself a
story of a “whole
self” rather than a disparate and
multi-faceted population of mental and emotional processes; and, to digress
slightly, it’s not
that different from how I construct within myself a story about “free will”, telling myself that I consciously and
rationally control my actions when in fact it’s usually the case that my conscious
rationalizations follow the unconsciously-determined decisions the rest of my
brain has made (Goertzel, 2004).
If a person wants to preserve their continuity of consciousness, and their
internal self-model, that’s fine – it’s a valid value judgment. But they should make this value judgment
based on the understanding that these things are not “real” – they’re
psychologically constructed illusions, that our brains have come to make
through some combination of evolutionary utility-seeking and
self-organizational pattern formation.
It may be that an individual decides these illusions, like the illusion
of free will, lie at the essence of humanity, and are worthy of preservation as
a fundamental core value.
On the other hand, it’s equally valid to judge that the fundamental value lies in overcoming
these illusions of self, will and continuity – and seeing that they don’t have any true reality. Perhaps the overcoming of these illusions is
the right path toward discovering a better way to exist – the human condition having many well-known
and well-documented flaws. This is the
direction in which Zen Buddhism and other
mystical traditions point (Austin, 1998). Science has discovered the illusory nature of free will, self and
continuity of consciousness only recently, but in the domain of “wisdom traditions,” all this is quite old news.
One can perceive the preservation unto eternity of the human illusions of
free will, self and continuity of consciousness as a good thing – or one can view it as a burden,
like the preservation unto eternity of stomachaches and bad tempers and
pimples. An equally valid, alternate
perspective holds that human-style individual minds, ridden with illusions as
they are, are merely an intermediary phase on the way to the development of
really interesting cognitive dynamics.
Among humans, illusions like will, self and consciousness-continuity are
just about inevitably tied in with intelligence. Highly rigorous long-term routines like Zen meditation practice
are able to whittle away the illusions, but they seem to have other costs – I don’t know of any Zen masters who make
interesting contributions to science or mathematics, for example. Among humans, the reduction of these
illusions on a practical day-to-day basis seems to require so much effort as to
absorb almost the entire organism to the exclusion of all else. Yet the same will not necessarily be the
case for superhuman AI’s, or enhanced human uploads, or posthuman humans with radical brain
improvements. These minds may be able
to carry out advanced intellectual activity without adopting the illusions that
are built into the human mind courtesy of our evolved brains.
A mind without the illusions of self, free will or continuity of
consciousness might not look much like a “mind” as we currently conceive it – it would be more of a “complex, creative, dynamical system of inter-creating patterns.”
FutureBen and FutureBush, as envisioned above, are actually fairly
unadventurous as prognostications of the future of mind – as described above, they’re still individuals, with
individual identities and histories; but it’s not at all clear that this is what the
future holds in store. If one’s value system favors general
values like freedom, growth and joy (Goertzel, 2004a), rather than primarily
valuing humanity as such, such a posthuman relatively-illusion-free mind may be
considered superior to human minds … and the prospect of immortality in human form may appear
like a kind of second-rate “booby prize.”
All these issues center around one key philosophical point: What is the
goal of immortality? What is the goal
of avoiding involuntary death? Is it to
keep human life as we know it around forever? That is a valid, respectable,
non-idiotic goal. Or is it to keep the process of growth alive and flourishing
beyond the scope painfully and arbitrarily imposed on it by the end of the
human life?
Human life as it exists now is not a constant, it's an ongoing growth
process; and for those who want it to be, human life beyond the current maximum
lifespan and beyond the traditional scope of humanity will still be a process
of growth, change and learning. Fear of death will largely be replaced by more
interesting issues like the merit of individuality and consciousness in its
various forms -- and other issues we can't come close to foreseeing yet.
It may be that, when some of us live long enough and become smart enough,
we decide that maintaining individuality and the other human illusions unto
eternity isn't interesting, and it's better to merge into a larger posthuman
intelligent dynamical-pattern-system.
And it may be that others of us find that individuality still seems
interesting forever. Resource wars
between superhuman post-individuals and human individuals can’t be ruled out, but nor can they be
confidently forecast -- since there will likely so many resources
available at the posthuman stage, and diversity may still seem like an
interesting value to superhuman post-individuals (so why not let the retro
human immortal individuals stick around and mind their own business?).
These issues are fairly hard to “feel out” right now, stuck as we are in this human form with its limited capacity
for experience, intelligence and communication. For me, the quest for
radical life extension is largely about staying around long enough, and growing
enough, to find out more about intriguing (philosophically, scientifically and
personally fundamental) issues like these.
Acknowledgements. This essay owes a great deal to conversations with various individuals, including most of all Bruce Klein, but also Izabela Goertzel, Martine Rothblatt, Moshe Looks, Jeff Medina, and the members of the DC Future Salon.