"WHAT HAVE YOU CHANGED YOUR MIND ABOUT?"
Computer Scientist, UC Berkeley, School of Information
Analysis of Language Requires Understanding Language
me, having my worldview entirely altered is among the most
fun parts of science. One mind-altering event occurred during
graduate school. I was studying the field of Artificial Intelligence
with a focus on Natural Language Processing. At that time there
were intense arguments amongst computer scientists, psychologists,
and philosophers about how to represent concepts and knowledge
in computers, and if those representations reflected in any
realistic way how people represented knowledge. Most researchers
thought that language and concepts should be represented in
a diffuse manner, distributed across myriad brain cells in
a complex network. But some researchers talked about the existence
of a "grandmother cell,"
meaning that one neuron in the brain (or perhaps a concentrated
group of neurons) was entirely responsible for representing the
concept of, say, your grandmother. I thought this latter view
one day in the early 90's I heard a story on National Public
Radio about children who had Wernicke's aphasia, meaning that
a particular region in their brains were damaged. This damage
left the children with the ability to form complicated sentences
with correct grammatical structure and natural sounding rhythms,
but with content that was entirely meaningless. This story
was a revelation to me -- it seemed like irrefutable proof
that different aspects of language were located in distinct
regions of the brain, and that therefore perhaps the grandmother
cell could exist. (Steven Pinker subsequently wrote his masterpiece, "The
Language Instinct," on this topic.)
after this, the field of Natural Language Processing became
radically changed by an entirely new approach. As I mentioned
above, in the early 90's most researchers were introspecting
about language use and were trying to hand-code knowledge into
computers. So people would enter in data like "when you
go to a restaurant, someone shows you to a table. You and your
dining partners sit on chairs at your selected table. A waiter
or waitress walks up to you and hands you a menu. You read
the menu and eventually the waiter comes back and asks for
your order. The waiter takes this information back to the kitchen." And
so on, in painstaking detail.
as large volumes of text started to become available online,
people started developing algorithms to solve seemingly difficult
natural language processing problems using very simple techniques.
For example, how hard is it to write a program that can tell
which language a stretch of text is written in? Sibun and Reynar
found that all you need to do is record how often pairs of
characters tend to co-occur in each language, and you only
need to extract about a sentence from a piece of text to classify
it with 99% accuracy into one of 18 languages! Another wild
example is that of author identification. Back in the early
60's, Mosteller and Wallace showed that they could identify
which of the disputed Federalist Papers were written by Hamilton
vs. those written by Madison, simply by looking at counts of
the function words (small structural words like "by", "from",
and "to") that each author used.
field as a whole is chipping away at the hard problems of natural
language processing by using statistics derived from that mother-of-all-text-corpora,
the Web. For example, how do you write a program to figure
out the difference between a "student protest" and
"war protest"? The former is a demonstration against
something, done by students, but the latter is not a demonstration
done by a war.
the old days, we would try to code all the information we could
about the words in the noun compounds and try to anticipate
how they interact. But today we used statistics drawn from
counts of simple patterns on the web. Recently my PhD student
Preslav Nakov has shown that we can often determine what the
intended relationship between two nouns is by simply counting
the verbs that fall between the two nouns, if we first reverse
their order. So if we search the web for patterns like:
that are * by students"
find out the important verbs are "draw, involve, galvanize,
affect, carried out by" and so on, whereas for "war
protests" we find verbs such as "spread by, catalyzed
by, precede", and so on.
lesson we see over and over again is that simple statistics
computed over very large text collections can do better at
difficult language processing tasks than more complex, elaborate
Scientist; Personal Computer Visionary, Senior
Fellow, HP Labs
Big Mind Change At Age 10: Vacuums Don't Suck!
age 10 in 1950, one of the department stores had a pneumatic
tube system for moving receipts and money from counters to
the cashier's office. I loved this and tried to figure out
how it worked The clerks in the store knew all about it. "Vacuum",
"Vacuum sucks the canisters, just like your mom's vacuum
cleaner". But how does it work, I asked? "Vacuum",
they said, "Vacuum, does it all". This was what adults
called "an explanation"!
I took apart my mom's Hoover vacuum cleaner to find out how
it worked. There was an electric motor in there, which I
had expected, but the only other thing in there was a fan!
How could a fan produce a vacuum, and how could it suck?
had a room fan and I looked at it more closely. I knew that
it worked like the propeller of an airplane, but I'd never
thought about how those worked. I picked up a board and moved
it. This moved air just fine. So the blades of the propeller
and the fan were just boards that the motor kept on moving
to push air.
what about the vacuum? I found that a sheet of paper would
stick to the back of the fan. But why? I "knew" that
air was supposed to be made up of particles too small to
be seen. So it was clear why you got a gust of breeze by
moving a board
— you were knocking little particles one way and not
another. But where did the sucking of the paper on the fan
and in the vacuum cleaner come from?
it occurred to me that the air particles must be already
moving very quickly and bumping into each other. When the
board or fan blades moved air particles away from the fan
there were less near the fan and the already moving particles
would have less to bump into and would thus move towards
the fan. They didn't know about the fan, but they appeared
The "suck" of
the vacuum cleaner was not a suck at all. What was happening
is that things went into the vacuum cleaner because they
were being "blown in" by the air particles' normal
movement, which were not being opposed by the usual pressure
of air particles inside the fan!
my physiologist father came home that evening I exclaimed
"Dad, the air particles must be moving at least a hundred
miles an hour!". I told him what I'd found out and he
looked in his physics book. In there was a formula to compute
the speed of various air molecules at various temperatures.
It turned out that at room temperature ordinary air molecules
were moving much faster than I had guessed: more like 1500
miles an hour! This completely blew my mind!
I got worried because even small things were clearly not
moving that fast going into the vacuum cleaner (nor in the
pneumatic tubes). By putting my hand out the window of the
car I could feel that the air was probably going into the
vacuum cleaner closer to 50 or 60 miles an hour. Another
conversation with my Dad led to two ideas (a) the fan was
probably not very efficient at moving particles away, and
(b) the particles themselves were going in every direction
and bumping into each other (this is why it takes a while
for perfume from an open bottle to be smelled across a room.
experience was a big deal for me because I had thought one
way using a metaphor and a story about "sucking",
and then I suddenly thought just the opposite because of
an experiment and non-story thinking. The world was not as
it seemed! Or as most adults thought and claimed! I never
trusted "just a story" again.
Claremont McKenna College; Past-president,
American Psychological Association;
Differences in Cognitive Abilities
A Simple Truth To "It All Depends"
are men underrepresented in teaching, child care, and related
fields and women underrepresented in engineering, physics,
and related fields? I used to know the answer, but that was
before I spent several decades reviewing almost everything
written about this question. Like most enduring questions,
the responses have grown more contentious and even less is
"settled" now that we have mountains of research designed
to answer them. At some point, my own answer changed from what
I believed to be the simple truth to a convoluted statement complete
with qualifiers, hedge terms, and caveats. I guess this shift
in my own thinking represents progress, but it doesn't feel or
look that way.
am a feminist, a product of the 60s, who believed that group
differences in intelligence or most any other trait are mostly
traceable to the lifetime of experiences that mold us into
the people we are and will be. Of course, I never doubted the
basic premises of evolution, but the lessons that I learned
from evolution favor the idea that the brain and behavior are
adaptable. Hunter-gatherers never solved calculus problems
or traveled to the moon, so I find little in our ancient past
to explain these modern-day achievements.
is also the disturbing fact that evolutionary theories can
easily explain almost any outcome, so I never found them to
be a useful framework for understanding behavior. Even when
I knew the simple truth about sex differences in cognitive
abilities, I never doubted that heritability plays a role in
cognitive development, but like many others, I believed that
once the potential to develop an ability exceeded some threshold
value, heritability was of little importance. Now I am less
sure about any single answer, and nothing is simple any more.
literature on sex differences in cognitive abilities is filled
with inconsistent findings, contradictory theories, and emotional
claims that are unsupported by the research. Yet, despite all
of the noise in the data, clear and consistent messages can
be heard. There are real, and in some cases sizable, sex differences
with respect to some cognitive abilities.
practices are undoubtedly important, but there is also good
evidence that biological sex differences play a role in establishing
and maintaining cognitive sex differences, a conclusion that
I wasn't prepared to make when I began reviewing the relevant
literature. I could not ignore or explain away repeated findings
about (small) variations over the menstrual cycle, the effects
of exogenously administered sex hormones on cognition, a variety
of anomalies that allow us to separate prenatal hormone effects
on later development, failed attempts to alter the sex roles
of a biological male after an accident that destroyed his penis,
differences in preferred modes of thought, international data
on the achievement of females and males, to name just a few
types of evidence that demand the conclusion that there is
some biological basis for sex-typed cognitive development.
thinking about this controversial topic has changed. I have
come to understand that nature needs nurture and the dichotomization
of these two influences on development is the wrong way to
conceptualize their mutual influences on each other. Our brain
structures and functions reflect and direct our life experiences,
which create feed back loops that alter the hormones we secrete
and how we select environments. Learning is a biological and
so, what had been a simple truth morphed into a complicated
answer for the deceptively simple question about why there
are sex differences in cognitive abilities. There is nothing
in my new understanding that justifies discrimination or predicts
the continuation of the status quo. There is plenty of room
for motivation, self-regulation, and persistence to make the
question about the underrepresentation of women and men in
different academic areas moot in coming years.
all complex questions, the question about why men and women
achieve in different academic areas depends on a laundry list
of influences that do not fall neatly into categories labeled
biology or environment. It is time to give up this tired way
of thinking about nature and nurture as two independent variables
and their interaction and recognize how they exert mutual influences
on each other. No single number can capture the extent to which
one type of variable is important because they do not operate
independently. Nature and nurture do not just interact; they
fundamentally change each other. The answer that I give today
is far more complicated than the simple truth that I used to
believe, but we have no reason to expect that complex phenomena
like cognitive development have simple answers.
Climatologist, Stanford University; Author, Laboratory
Change: Warming Up To The Evidence
public appearances about global warming, even these days, I
often hear: "I don't believe in global warming" and
I then typically get asked why I do "when all the evidence
is not in". "Global warming is not about beliefs",
I typically retort, "but an accumulation of evidence over
decades so that we can now say the vast preponderance of evidence — and
its consistency with basic climate theory — supports
global warming as well established, not that all aspects are
fully known, an impossibility in any complex systems science".
it hasn't always been that way, especially for me at the outset
of my career in 1971, when I co-authored a controversial paper
calculating that cooling effects from a shroud of atmospheric
dust and smoke — aerosols — from human emissions
at a global scale appeared to dominate the opposing warming
effect of the growing atmospheric concentrations of the greenhouse
gas carbon dioxide. Measurements at the time showed both warming
and cooling emissions were on the rise, so a calculation of
the net balance was essential — though controlling the
aerosols made sense with or without climate side effects since
they posed — and still pose — serious health effects
on vulnerable populations. In fact for the latter reason
laws to clean up the air in most rich countries were just getting
negotiated about that time.
I traveled the globe in the early 1970s to explain our calculations,
what I slowly learned from those out there making measurements
was that two facts had only recently come to light, and together
they appeared to make me consider flipping sign from cooling
to warming as the most likely climatic change direction from
humans using the atmosphere as a free sewer to dump some of
our volatile industrial and agricultural wastes. These
facts were that human-injected aerosols, which we assumed were
global in scale in our cooling calculation — were in
fact concentrated primarily in industrial regions and bio-mass
burning areas of the globe — about 20% of the Earth's
surface, whereas we already knew that CO2 emissions are global
in extent and about half of the emitted CO2 lasts for a century
or more in the air.
there were new facts that were even more convincing: not only
is CO2 an important human-emitted greenhouse gas, but so too
were methane, nitrous oxide and chlorofluorocarbons (many of
the latter gases now banned because they also deplete stratospheric
ozone) , and that together with CO2, these other greenhouse
gasses were an enhanced global set of warming factors. On the
other hand, aerosols were primarily regional in extent and
could not thus overcome the warming effects of the combined
global scale greenhouse gases.
was very proud to have published in the mid-1970s what was
wrong with my early calculations well before the so-called "contrarians" — climate
change deniers still all too prevalent even today — understood
the issues, let alone incorporated these new facts into updated
models to make more credible projections. Of course, today
the dominance of warming over cooling agents is now well established
in the climatology community, but our remaining inability to
be very precise over how much warming the planet can expect
to have to deal with is in large part still an uncertainty
over the partially counteracting cooling effects of aerosols — enough
to offset a significant, even if largely unknown, amount of
the warming. So although we are very confident in the existence
of human-caused warming in the past several decades from greenhouse
gases, we are still are working hard to pin down much more
precisely how much aerosols offset this warming. Facts on that
offset still lag the critical need to estimate better our impacts
on climate before they become potentially irreversible.
sad part of this story is not about science, but the misinterpretation
of it in the political world. I still have to endure polemical
blogs from contrarian columnists and others about how, as one
put it in a grand polemic: "Schneider is an environmentalist
for all temperatures"
— citing my early calculations. This famous columnist somehow
forgot to bring up the later-corrected (by me) faulty assumptions,
nor mention that the 1971 calculation was based on not-yet-gathered
facts. Simply getting the sign wrong was cited, ipso facto in
this blog, as somehow damning of my current credibility.
Ironically, inside the scientific world, this switch of sign
of projected effects is viewed as precisely what responsible
scientists must do when the facts change. Not only did I change
my mind, but published almost immediately what had changed and
how that played out over time. Scientists have no crystal ball,
but we do have modeling methods that are the closest approximation
available. They can't give us truth, but they can tell us the
logical consequences of explicit assumptions. Those who update
their conclusions explicitly as facts evolve are much more likely
to be a credible source than those who stick to old stories for
political consistency. Two cheers for the scientific method!
Tech Culture Journalist; Co-editor,
Boing Boing; Commentator, NPR; Host, Boing Boing tv
Communities Rot Without Daily Tending By Human Hands
changed my mind about online community this year.
co-edit a blog that attracts a large number of daily visitors,
many of whom have something to say back to us about whatever
we write or produce in video. When our audience was small in
the early days, interacting was simple: we tacked a little
href tag to an open comments thread at the end of each post:
Link, Discuss. No moderation, no complication, come as you
are, anonymity's fine. Every once in a while, a thread accumulated
more noise than signal, but the balance mostly worked.
then, the audience grew. Fast. And with that, grew the number
of antisocial actors, "drive-by trolls," people for
whom dialogue wasn't the point. It doesn't take many of them
to ruin the experience for much larger numbers of participants
acting in good faith.
of the more grotesque attacks were pointed at me, and the new
experience of being on the receiving end of that much personally-directed
nastiness was upsetting. I dreaded hitting the "publish" button
on posts, because I knew what would now follow.
noise on the blog grew, the interaction ceased to be fun for
anyone, and with much regret, we removed the comments feature
grew to believe that the easier it is to post a drive-by comment,
and the easier it is to remain faceless, reputation-less, and
real-world-less while doing so, the greater the volume of antisocial
behavior that follows. I decided that no online community could
remain civil after it grew too large, and gave up on that aspect
of internet life.
co-editors and I debated, we brainstormed, we observed other
big sites that included some kind of community forum or comments
feature. Some relied on voting systems to "score" whether
a comment is of value — this felt clinical, cold, like
grading what a friend says to you in conversation. Dialogue
shouldn't be a beauty contest. Other sites used other automated
systems to rank the relevance of a speech thread. None of this
felt natural to us, or an effective way to prevent the toxic
sludge buildup. So we stalled for years, and our blog remained
more monologue than dialogue. That felt unnatural, too.
this year, we resurrected comments on the blog, with the one
thing that did feel natural. Human hands.
hired a community manager, and equipped our comments system
with a secret weapon: the "disemvoweller." If someone's
misbehaving, she can remove all the vowels from their screed
with one click. The dialogue stays, but the misanthrope looks
ridiculous, and the emotional sting is neutralized.
once again, the balance mostly works. I still believe that
there is no fully automated system capable of managing the
complexities of online human interaction — no software
fix I know of. But I'd underestimated the power of dedicated
one early weed from a bed of germinating seeds changes everything.
Small actions by focused participants change the tone of the
whole. It is possible to maintain big healthy gardens online.
The solution isn't cheap, or easy, or hands-free. Few things
of value are.
Universite' de la Mediterrane' (Marseille, France); Author: What
is time? What is Space?
is nothing to add to the standard interpretation of quantum
have learned quantum mechanics as a young man, first from
the book by Dirac, and then form a multitude of other excellent
textbooks. The theory appeared bizarre and marvelous, but
it made perfectly sense to me. The world, as Shakespeare
put it, is "strange and admirable", but it is coherent.
I could not understand why people remained unhappy with such
a clear and rational theory. In particular, I could not understand
why some people lost their time on a non-problem called the "interpretation
of quantum mechanics".
have remained of this opinion for many years. Then I moved
to Pittsburgh, to work in the group of Ted Newman, great
relativist and one of the most brilliant minds in the generation
before mine. While there, the experiments made by the team
of Alain Aspect Aspect at Orsay, in France, which confirmed
spectacularly some of the strangest predictions of quantum
mechanics, prompted a long period of discussion in our group.
Basically, Ted claimed that quantum theory made no sense.
I claimed that it does perfectly, since it is able to predict
unambiguously the probability distribution of any conceivable
time has passed, and I have changed my mind. Ted's arguments
have finally convinced me: I was wrong, and he was right.
I have slowly came to realize that in its most common textbook
version, quantum mechanics makes sense as a theory of a small
portion of the universe, a "system", only under
the assumption that something else in the universe fails
to obey quantum mechanics. Hence it becomes self contradictory,
in its usual version, if we take it as a general description
of all physical systems of the universe. Or, at least, there
is still something key to understand, with respect to it.
change of opinion has motivated me to start of a novel line
of investigation, which I have called "relational quantum
mechanics". It has also affected substantially my work
in quantum gravity, taking me to consider a different sort
of observable quantities as natural probes of quantum spacetime.
am now sure that quantum theory has still much to tell us
about the deep structure of the world. Unless I'll change
my mind again, of course.
Psychologist & Computer
Scientist; Engines for Education Inc.; Author, Making Minds
Less Well Educated than Our Own
reporters interviewed me in the 70's and 80's about
the possibilities for Artificial Intelligence I would always
say that we would have machines that are as smart as we are
within my lifetime. It seemed a safe answer since no one could
ever tell me I was wrong. But I no longer believe that will
happen. One reason is that I am a lot older
and we are barely closer to creating smart machines.
have not soured on AI. I still believe that we can create very
intelligent machines. But I no longer believe that those machines
will be like us. Perhaps it was the movies that led us to believe
that we would have intelligent robots as companions. (I was
certainly influenced early on by 2001.) Certainly
most AI researchers believed that creating machines that were
our intellectual equals or better was a real possibility. Early
AI workers sought out intelligent behaviors to focus on, like
chess or problem solving, and tried to build machines that
could equal human beings in those same endeavors. While this
was an understandable approach it was, in retrospect, wrong-headed. Chess
playing is not really a typical intelligent human activity.
Only some of us are good at it, and it seems to entail a level
of cognitive processing that while impressive seems quite at
odds with what makes humans smart. Chess players are methodical
planners. Human beings are not.
are constantly learning. We spend years learning some
seemingly simple stuff. Every new experience changes what we
know and how we see the world. Getting reminded of our pervious
experiences helps us process new experiences better than we
did the time before. Doing that depends upon an unconscious
indexing method that all people learn to do without quite realizing
they are learning it. We spend twenty years (or more) learning
how to speak properly and learning how to make good decisions
and establish good relationships. But we tend to not know what
we know. We can speak properly without knowing how we do it.
We don't know how we comprehend. We just do.
this poses a problem for AI. How can we imitate what humans
are doing when humans don't know what they are doing
when they do it? This conundrum led to a major failure in AI,
expert systems, that relied upon rules that were supposed to
characterize expert knowledge. But, the major characteristic
of experts is that they get faster when they know more, while
more rules made systems slower. The idea that rules were not
at the center of intelligent systems meant that the flaw was
relying upon specific consciously stated knowledge instead
of trying to figure out what people meant when they said they
just knew it when they saw it, or they had a gut
give reasons for their behaviors but they are typically figuring
that stuff out after the fact. We reason non-consciously and
explain rationally later. Humans dream. There obviously is
some important utility in dreaming. Even if we don't
understand precisely what the consequences of dreaming are,
it is safe to assume that it is an important part of our unconscious
reasoning process that drives our decision making. So, an intelligent
machine would have to dream because it needed to, and would
have to have intuitions that proved to be good insights, and
it would have to have a set of driving goals that made it see
the world in a way that a different entity with different goals
would not. In other words it would need a personality, and
not one that was artificially installed but one that came with
the territory of what is was about as an intelligent entity.
AI can and should build are intelligent special purpose entities.
(We can call them Specialized Intelligences or SI's.)
Smart computers will indeed be created. But they will arrive
in the form of SI's, ones that make lousy companions
but know every shipping accident that ever happened and why
(the shipping industry's SI) or as an expert on sales
(a business world SI.) The sales SI, because sales
is all it ever thought about, would be able to recite every
interesting sales story that had ever happened and the lessons
to be learned from it. For some salesman about to call on a
customer for example, this SI would be quite fascinating. We
can expect a foreign policy SI that helps future presidents
learn about the past in a timely fashion and helps them make
decisions because it knows every decision the government has
ever made and has cleverly indexed them so as to be able to
apply what it knows to current situations.
AI in the traditional sense, will not happen in my lifetime
nor in my grandson's lifetime. Perhaps a new kind of
machine intelligence will one day evolve and be smarter than
us, but we are a really long way from that.
Director, the Center for Science Writings,
Stevens Institute of Technology; Author, Rational
My Mind About the Mind-Body Problem
decade ago, I thought the mind-body problem would never be
solved, but I've recently, tentatively, changed my mind.
and scientists have long puzzled over how matter — more
specifically, gray matter — makes mind, and some have
concluded that we'll never find the answer. In 1991 the philosopher
Owen Flanagan called these pessimists "mysterians, a
term he borrowed from the 1960s rock group "Question
Mark and the Mysterians."
of the earliest mysterians was the German genius Leibniz,
who wrote: "Suppose that there be a machine, the structure
of which produces thinking, feeling, and perceiving; imagine
this machine enlarged but preserving the same proportions,
so that you could enter it as if it were a mill… What
would you observe there? Nothing but parts which push and
move each other, and never anything that could explain perception."
decade ago I was a hard-core mysterian, because I couldn't
imagine what form a solution to the mind-body problem might
take. Now I can. If there is a solution, it will come in
the form of a neural code, an algorithm, set of rules or
syntax that transforms the electrochemical pulses emitted
by brain cells into perceptions, memories, decisions, thoughts.
recently, a complete decoding of the brain seemed impossibly
remote, because technologies for probing living brains were
so crude. But over the past decade the temporal and spatial
resolution of magnetic resonance imaging, electroencephalography
and other external scanning methods has leaped forward. Even
more importantly, researchers keep improving the design of
microelectrode arrays that can be embedded in the brain to
receive messages from — and transmit them to — thousands
of individual neurons simultaneously.
are gleaning information about neural coding not only from
non-human animals but also from patients who have had electrodes
implanted in their brains to treat epilepsy, paralysis, psychiatric
illnesses and other brain disorders. Given these advances,
I'm cautiously optimistic that scientists will crack the
neural code within the next few decades.
neural code may resemble relativity and quantum mechanics,
in the following sense. These fundamental theories have not
resolved all our questions about physical reality. Far from
it. Phenomena such as gravity and light still remain profoundly
puzzling. Physicists have nonetheless embraced relativity
and quantum mechanics because they allow us to predict and
manipulate physical reality with extraordinary precision.
Relativity and quantum mechanics work.
the same way, the neural code is unlikely to resolve the
mind-body problem to everyone's satisfaction. When it comes
to consciousness, many of us seek not an explanation but
a revelation, which dispels mystery like sun burning off
a morning fog. And yet we will embrace a neural-code theory
of mind if it works — that is, if it helps us predict,
heal and enhance ourselves. If we can control our minds,
who cares if we still cannot comprehend them?
Psychologist, MIT; Author, Evocative
Objects: Things We Think With
What I've Changed
My Mind About
Throughout my academic career – when I was studying the relationship between psychoanalysis and society and when I moved to the social and psychological studies of technology – I've seen myself as a cultural critic. I don't mention this to stress how lofty a job I put myself in, but rather that I saw the job as theoretical in its essence. Technologists designed things; I was able to offer insights about the nature of people's connections to them, the mix of feelings in the thoughts, how passions mixed with cognition. Trained in psychoanalysis, I didn't see my stance as therapeutic, but it did borrow from the reticence of that discipline. I was not there to meddle. I was there to listen and interpret. Over the past year, I've changed my mind: our current relationship with technology calls forth a more meddlesome me.
In the past, because I didn't criticize but tried to analyze, some of my colleagues found me complicit with the agenda of technology-builders. I didn't like that much, but understood that this was perhaps the price to pay for maintaining my distance, as Goldilock's wolf would say, "the better to hear them with." This year I realized that I had changed my stance. In studying reactions to advanced robots, robots that look you in the eye, remember your name, and track your motions, I found more people who were considering such robots as friends, confidants, and as they imagined technical improvements, even as lovers. I became less distanced. I began to think about technological promiscuity. Are we so lonely that we will really love whatever is put in front of us?
I kept listening for what stood behind the new promiscuity – my habit of listening didn't change – and I began to get evidence of a certain fatigue with the difficulties of dealing with people. A female graduate student came up to me after a lecture and told me that she would gladly trade in her boyfriend for a sophisticated humanoid robot as long as the robot could produce what she called "caring behavior." She told me that "she needed the feeling of civility in the house and I don't want to be alone." She said: "If the robot could provide a civil environment, I would be happy to help produce the illusion that there is somebody really with me." What she was looking for, she told me, was a "no-risk relationship" that would stave off loneliness; a responsive robot, even if it was just exhibiting scripted behavior, seemed better to her than an demanding boyfriend. I thought she was joking. She was not.
In a way, I should not have been surprised. For a decade I had studied the appeal of sociable robots. They push our Darwinian buttons. They are programmed to exhibit the kind of behavior we have come to associate with sentience and empathy, which leads us to think of them as creatures with intentions, emotions, and autonomy. Once people see robots as creatures, they feel a desire to nurture them. With this feeling comes the fantasy of reciprocation. As you begin to care for these creatures, you want them to care about you.
And yet, in the past, I had found that people approached computational intelligence with a certain "romantic reaction." Their basic position was that simulated thinking might be feeling but simulated feeling was never feeling and simulated love was never love. Now, I was hearing something new. People were more likely to tell me that human beings might be "simulating" their feelings, or as one woman put it: "How do I know that my lover is not just simulating everything he says he feels?" Everyone I spoke with was busier than ever on with their e-mail and virtual friendships. Everyone was busier than ever with their social networking and always-on/always-on-you PDAs. Someone once said that loneliness is failed solitude. Could no one stand to be alone anymore before they turned to a device? Were cyberconnections paving the way to think that a robotic one might be sufficient unto the day? I was not left contemplating the cleverness of engineering but the vulnerabilities of people.
Last spring I had a public exchange in which a colleague wrote about the "I-Thou" dyad of people and robots and I could only see Martin Buber spinning in his grave. The "I" was the person in the relationship, but how could the robot be the "Thou"? In the past, I would have approached such an interchange with discipline, interested only in the projection of feeling onto the robot. But I had taken that position when robots seemed only an evocative object for better understanding people's hopes and frustrations. Now, people were doing more than fantasizing. There was a new earnestness. They saw the robot in the wings and were excited to welcome it onstage.
It seemed no time at all that a book came out called Love and Sex with Robots and a reporter from Scientific American was interviewing me about the psychology of robot marriage. The conversation was memorable and I warned my interviewer that I would use it as data. He asked me if my opposition to people marrying robots put me in the same camp as those who oppose the marriage of lesbians or gay men. I tried to explain that just because I didn't think people could marry machines didn't mean that I didn't think that any mix of people with people was fair play. He accused me of species chauvinism. Wasn't this the kind of talk that homophobes once used, not considering gays as "real" people? Right there I changed my mind about my vocation. I changed my mind about where my energies were most needed. I was turning in my card as a cultural critic the way I had always envisaged that identity. Now I was a cultural critic. I wasn't neutral; I was very sad.
College Professor of Psychology at Harvard University;
Author, Stumbling on Happiness
Benefit of Being Able to Change My Mind
years ago, I changed my mind about the benefit of being
able to change my mind.
In 2002, Jane Ebert and I discovered that people are generally
happier with decisions when they can't undo them. When subjects
in our experiments were able to undo their decisions they
tended to consider both the positive and negative features
of the decisions they had made, but when they couldn't undo
their decisions they tended to concentrate on the good features
and ignore the bad. As such, they were more satisfied when
they made irrevocable than revocable decisions. Ironically,
subjects did not realize this would happen and strongly preferred
to have the opportunity to change their minds.
Now up until this point I had always believed that love causes
marriage. But these experiments suggested to me that marriage
could also cause love. If you take data seriously you act
on it, so when these results came in I went home and proposed
to the woman I was living with. She said yes, and it turned
out that the data were right: I love my wife more than I
loved my girlfriend.
The willingness to change one's mind is a sign of intelligence,
but the freedom to do so comes at a cost.