The Social Brain: culture, change and evolution | Bret Weinstein (Full Video)


We’re heading into a very dangerous phase
of history; human beings being addicted to growth are constantly looking for sources,
so when we feel austerity coming on we tend to become more tribal. Unfortunately a perfectly free market will
not allow benevolent firms to survive in the long run. My argument is not an argument for centrism. I regard utopianism as probably the worst
idea that human beings have ever had. We find ourselves unfortunately stuck in an
archaic argument about policy; frankly the left and right are both out of answers and
they should team up on the basis that they agree at a values level about what a functional
society should ideally look like. — Human beings, like all creatures, are the
product of adaptive evolution, but they are highly unusual amongst evolved creatures. In order to understand them it is very important
to recognize certain things that make us different from even the most similar creatures, like
chimpanzees. The most important difference is something
I call the omega principal. The omega principal specifies the relationship
between human culture and the human genome. The most important thing to realize about
human beings is that a tremendous amount of what we are is not housed in our genomes;
it’s housed in a cultural layer that is passed on outside of genes. Culture is vastly more flexible, more plastic,
and more quickly evolving in an adaptive sense than genes, which is why in fact cultural
evolution came about in human beings. It allows human beings to switch what they
are doing and how they are doing it much more quickly than they could if all information
that was adapting was stored in DNA. One of the very important benefits of understanding
this relationship between the genome and the cultural attributes of human beings is that
it frees us to engage in an analysis of the evolutionary meaning of behaviors without
having to know where exactly the information is stored. This is especially important with complex
phenomenon, which may be partially housed in the genome and partially housed in the
cultural layer—something like human language, for example. Human language as a capacity is obviously
genetically encoded, but individual human languages are not. And so if we are to talk about the adaptive
utility of human language, being obligated to specify what is housed where could put
off that discussion for generations, whereas if we recognize that the cultural aspects
of language—as well as the genomic aspects of language—are all serving a united interest
then we can begin to understand the meaning of something like language in rigorous, adaptive
terms. The hypothesis of cultural evolution, which
has now has been sufficiently tested to be regarded as a theory—of human cultural evolution,
is the invention of Richard Dawkins, who in 1976 in The Selfish Gene coined the term ‘meme’
as an analog for gene; it’s a unit of cultural evolution. The genome creates a brain that is capable
of being infused with culture after an individual person is born. If culture was evolving to do things that
were not in the genome’s interest they would effectively be wasting the time and resources
that the genetic individual has access to on frivolous things at best. So the genome would shut down frivolous culture
were it a very common commodity. So the theory of memes tells us that there
is a process, very much like the one that shapes our genomes, at work in the cultural
layer. That does not mean, however, that the cultural
evolving layer is free of obligation to the genome. In fact, the cultural layer is downstream,
and one of the things that we have repeatedly gotten wrong is we have attempted to just
simply extend the rules of adaptive evolution as we have learned them from other creatures
and apply them to human beings, and it leads to some unfortunate misunderstandings. The fact that we are primarily culturally
informed tells us that culture serves the genetic interests almost all of the time. Which is to say, if you look at a long-standing
cultural trait, it doesn’t matter what it is—whether it’s music or religion or humor—all
of those things must be paying for themselves in terms of genetic fitness. Once we’ve recognized that, we can skip
to the much more interesting question of: “in what way do some of the remarkable cultural
structures that we see serve genetic interests?” Some of them seem absolutely paradoxical if
we try to imagine that they are serving our genomes, and yet that is the conclusion that
we have to reach when we realize that the genome is not only tolerating the existence
of that culture, but it is facilitating its acquisition. This suggests a very odd state of affairs
for human beings, in which we have minds that are programmed by culture and that can be
completely at odds with our genomes. And it leads to misunderstandings of evolution,
like the idea that religious belief is a mind virus—that effectively these belief structures
are parasitizing human beings, and they are wasting the time and effort that those human
beings are spending on that endeavor, rather than the more reasonable interpretation, which
is that these belief systems have flourished because they have facilitated the interests
of the creatures involved. Our belief systems are built around evolutionary
success and they certainly contain human benevolence—which is appropriate to phases of history when there
is abundance and people can afford to be good to each other. The problem is, if you have grown up in a
period in which abundance has been the standard state you don’t anticipate the way people
change in the face of austerity. And so what we are currently seeing is messages—that
we have all agreed are unacceptable—reemerging, because the signals that we have reached the
end of the boom times, those signals are everywhere and so people are triggered to move into a
phase that they don’t even know that they have. Despite the fact that human beings think that
they have escaped the evolutionary paradigm, they’ve done nothing of the kind; And so,
we should expect the belief systems that people hold to mirror the evolutionary interests
that people have rather than to match our best instincts. When we are capable of being good to each
other because there’s abundance, we have those instincts; and so it’s not incorrect
to say that human beings are capable of being marvelous creatures and being quite ethical. Now I would argue there’s a simple way of
reconciling the correct understanding—that religious belief often describes “truths”
that in many cases fly in the face of what we can understand scientifically—with the
idea that these beliefs are adaptive. I call it the state of being literally false
and metaphorically true. A belief is literally false and metaphorically
true if it is not factual, but if behaving as if it were factual results in an enhancement
of one’s fitness. To take an example: if one behaves in, let’s
say, the Christian tradition, in such a way as to gain access to heaven, one will not
actually find themselves at the pearly gates being welcomed in, but one does tend to place
their descendants in a good position with respect to the community that those of descendants
continue to live in. So if we were to think evolutionarily, the
person who is behaving so as to get into heaven has genetic interests. Those genetic interests are represented in
the narrow sense by their immediate descendants and close relatives; in the larger sense they
may be represented by the entire population of people from whom that individual came,
and by acting so as to get into heaven the fitness of that person, the number of copies
of those genes that continue to flourish in the aftermath of that person’s death will
go up. So the believe in heaven is literally false—there
is no such place—but it is metaphorically true in the sense that it results in an increase
in fitness. If you think about all of the things that
you know human beings to have done over the course of human history you’ll realize that
humans must have shifted from one niche to the other again and again. Effectively humans are a niche-switching creature. That is the human niche—to discover new
things to do when the ancestral ways have petered out and are no longer useful. Innovating new ways to be is very much the
human toolkit. When human beings adopt an opportunity, their
population grows in proportion to the size of that opportunity, and that opportunity
essentially should be thought of as a frontier. Now, there are many kinds of frontiers that
humans have discovered. The most obvious kind of frontier is a geographic
frontier: when a population discoverers an uninhabited island—or in an extreme case
they discover a continent that has no people on it—that is a tremendous opportunity,
and a tiny population can grow to gigantic size given such a bit of good fortune. But there are less-obvious kinds of frontier
as well. A technological frontier occurs when people
discover a mechanism for doing more with the territory that they have. So for example, if you think about a piece
of territory that has been inhabited by hunter-gatherers, at the point that farming is either invented
or brought in (discovered by some other population that same piece of territory), if it is hospitable
to farming, can support a much larger population. So it functions just like having discovered
a new landmass, because the size of the population that exists on the current landmass goes up
as a result of the fact that the land is made more productive. There’s a third kind of a frontier, which
I call a “transfer” frontier, which is not really the same in the sense that it is
zero sum: somebody has to lose in order for somebody else to win. But from the point of view of an individual
population, another population that cannot defend the resources that it has is an opportunity,
and so many of the worst chapters in human history involve one population targeting another
population that can’t defend the resources that it owns. And so for the population doing the targeting,
capturing those resources functions like having discovered a new landmass or a new technology
that allows productivity to go up. All of these types of frontiers eventually
run out. There is simply a limit to the number of geographical
locations that can be inhabited. There may always be a next technology, but
the discovery of new technologies comes in fits and starts, and there can be long dry
periods where you have reached or exceeded the limits of a technological opportunity
and the next one is nowhere on the horizon. So human beings, being addicted to growth,
are constantly looking for sources. And when geographic frontiers and technological
frontiers don’t provide those opportunities, human beings will sometimes look within their
own population and figure out who can’t defend the resources that they hold and they
manufacture reasons that they are not entitled to keep them. And so when we feel austerity coming on we
tend to become more tribal. And this is a very dangerous pattern of history,
for example, what took place during the Holocaust when the German population decided to target
European Jews, and it made up reasons that those Jews were not entitled to continue. So, what we are effectively seeing in the
present is a circumstance in which we have reached the end of a boom, and human beings
are becoming tribal because that is the natural transition at the end of a growth period,
and we are naturally inspired to look for something to replace the growth that has run
out. This is why many of these abhorrent messages
have become resonant in the present to many people. They are waiting to hear somebody explain
what population isn’t capable of defending its resources and to explain what justification
will be used to pursue those resources—and to transfer them. Many people are optimistic that technological
breakthroughs will continue to provide access to growth. And this is an unfortunate perspective, because
it leads us into a false sense of security, not realizing that—being evolutionary creatures—we
are not programmed to preserve that state of growth and make it last a long time. What we are wired to do is to capture the
benefits of them and bring them into use. What that means for most creatures is: when
a non-zero sum opportunity has been discovered, creatures create many more like themselves;
basically more mouths to feed. For modern people, sometimes creating more
mouths is not the natural reaction, but creating greater consumption is. And so as much as we are wired in a way that
is beneficial—where we discover new ways of doing more with less that provides abundance—we
are also wired to use up that abundance in consumption. And in fact we have a dynamic in which our
economic theories—the ones that we run society on the basis of—actually define economic
health. Growth is the conversion of useful energy
into useless heat and the conversion of useful materials into useless waste. So we have what I call a throughput society
where we view ourselves as doing something right as we are taking resources that might
be made to last a long time and we use them up. So, for example, if I were to invent a microwave
oven that is just as useful as the one you have but would last ten times as long as the
one you have, intuitively it seems that that should be a very good thing; but from the
point of view of the economy it will result in a reduction in growth because fewer microwave
ovens will be sold. So by defining our economic terms such that
they lead us to “correct” for improvements in efficiency and cause us to capture resources
and use them up in one kind of consumption or another, we set ourselves up for a situation
in which no matter how good an opportunity we discover it is inherently temporary. Ethics evolved to limit the self-destructive
behavior internal to a population. Unfortunately in our present circumstance,
where we have handed over so many functions to an anonymous marketplace, people that learn
where the ethical landscape (that we describe to ourselves), where on that “landscape”
there are opportunities that are unpoliced—actually come out ahead. If you discover things that are unethical
that therefore many people will not engage in, but you’re willing to engage in those
things and there’s no penalty—either informal or legal—built in the system, then you will
come out ahead as a result of your increased freedom because you’re not ethically limited
to the narrower set of acceptable behaviors. And so unfortunately society has begun rewarding
people who are good at figuring out where we are not policing our ethical standards
and exploiting those opportunities as a competitive advantage. The fate of benevolent firms in the market
is a very important topic. Unfortunately a perfectly free market will
not allow benevolent firms to survive in the long run. Now that may seem like an overly declarative
statement, but the problem is this: if you imagine two firms, one of which is perfectly
amoral and will do absolutely anything that generates a profit and the other one, which
is constrained by ethical beliefs that prevent it from availing itself of certain economic
opportunities, then it doesn’t matter what the economic circumstance is: the amoral corporation
or firm always has an advantage. The best that can be true is that the ethical
choice is also the strategically best choice and the two will be dead even, but in any
case where there’s any distinction whatsoever between the ethical choice and the perfectly
amoral choice, the amoral firm has the advantage of being perfectly free to avail itself of
opportunities that its competitor cannot reach. And what this does is it causes evolution
of firms in the direction of ruthlessness. It is often times the case that people who
set things in motion in the marketplace with the best of intentions are surprised at what
ultimately becomes of their innovations. Google famously began with the prime directive
“don’t be evil” and many people have recognized that over time Google has become
more ruthless than it was at the start. This is actually a perfectly predictable phenomenon. The reason that Google was able to have noble
objectives at first was that it existed in an immature market where having ethical restraints
on what is possible did not put it at a competitive disadvantage to any viable competitor that
it faced. As a market matures, its tolerance for firms
that restrain themselves is much reduced, and a perfectly efficient market has effectively
no tolerance for self-restraint. And as a result of this, firms evolve to become
more ruthless or they parish. And either way what we find is an increased
tendency in a market—as it becomes more efficient—towards ruthlessness. This does not have to be the case, but it
is the result of the fact that we leave the market free for this kind of evolutionary
trajectory. If we were wise about this we would realize
that a free market is not the ideal state. We don’t want to tinker, we don’t want
to meddle in a way that is overly disruptive of innovation, but we do want to tinker enough
that the evolutionary tendency produces the kinds of firms that we wish to see rather
than ones that behave in a way that horrifies us. We find ourselves unfortunately stuck in an
archaic argument about policy where right and left disagree about the wisdom of tinkering
with society to make certain things better. In general the left is overly enthusiastic
about meddling and it doesn’t appreciate the full danger of unintended consequences,
and the right is overly skeptical of the advantages of tinkering and prone to focus on the unintended
consequences, and be underambitious with respect to making society better. But the entire argument is based on ideas
of the 18th and 19th century, and those ideas are simply not up to date enough to deal with
the problems of the 21st century. So my argument would be that those on the
left and right who are in favor of liberty as perhaps the highest human value should
put aside their policy differences—because frankly the left and right are both out of
answers—and they should team up on the basis that they agree at a values level about what
a functional society should ideally look like. And we should actually begin a new conversation
about policy in which we investigate what we can do that the founders of this nation
(and others that are modeled on it) couldn’t imagine because they didn’t have the tools
at their disposal. In particular, we should be very careful that
whatever solution-making we engage in is evolutionarily aware. The founders of the United States did not,
of course, know anything about evolution. Those who have constructed our markets did
not know anything about evolution. And what they have done repeatedly is accidentally
set up an evolutionary system in which adaptation begins to take place without anybody’s awareness. And what that tends to do is it tends to take
the best intentions of those who set up these systems and overrun them with things that
simply function. It results in dangerous patterns like regulatory
capture, where those entities that figure out how to tinker those parts of the governance
apparatus that are supposed to regulate them come out ahead of those that don’t attempt
to tinker with those elements. My argument is not an argument for centrism. I believe that the answers we are looking
for are not actually on the map of possibilities that we are familiar with. We are effectively living in Flatland, and
what we have to do is learn to detect the Z axis so that we can seek solutions of a
type that will at first, be unfamiliar to us. There’s a great danger in doing this, of
course, which is Utopianism. I regard Utopianism as probably the worst
idea that human beings have ever had, and if anyone doubts that that’s the case you
should look at the history of Utopian ideas across the 20th century. The untold number of bodies that stacked up
as a result of Utopian ideas run amok is absolutely staggering. Utopians make two errors: the first error
is they prioritize a single value. Now, because of the way mechanisms function,
anytime you optimize a single value you create incredibly large costs for every other value
in question. By prioritizing things like liberty or equality,
if you do so in a narrowly focused way you can’t help but generate a dystopian result,
because all of the other values that people might hold are effectively destroyed in the
process. The other mistake that utopians make is they
tend to imagine that they know what the future state should look like and they miss what
every inventor knows, which is that your grandest ideas are crude to begin with. You have to build a prototype in order to
figure out what you don’t understand. And so I would argue we cannot describe the
future that we should be seeking. We can say what direction it probably is in
and we can head in that direction intelligently, but the minute we start telling ourselves
that we know what the state that we are trying to construct looks like we will suffer the
same failure that an inventor that wanted to bypass the prototyping stage would suffer. There are two kinds of conflict that people
can find themselves in: they can be in conflict when they have fundamentally different interests
from each other, but very frequently for human beings we will find ourselves in conflict
with somebody with whom we are aligned but we have a difference of opinion about what
to pursue or in what manner to do so. And there are a couple of things that evolution
can tell us about how to address such a conflict to be productive. The first thing is it is very important to
figure out what it is that causes you to disagree. Sometimes you may be disagreeing over values. So for example, if the two people prioritize
things differently they may have a different sense about what should be done. On the other hand, people may be disagreeing
about how to accomplish something while being completely aligned with respect to what it
is that is desired. So establishing what it is that has you disagreeing
is very important. I’ll give you an example. I frequently find, as somebody who comes from
the political left, that I have very easy conversations with Libertarians on the right,
and those conversations remain easy until we get to the question of policy. At the point that we get to the question of
policy we diverge. And the reason that we diverge is actually
that we have different expectations about the danger of creating new policy, but we
do not disagree over the values. A right-of-center economic Libertarian will
agree that ideally the market works best when opportunity is as broadly distributed as possible,
so that everybody has an opportunity to innovate. They may disagree about how equal the opportunity
is currently, and they may also disagree about the wisdom of attempting to redistribute opportunities
so that people who don’t have it can gain access to it, but there’s no disagreement
over whether it is a desirable characteristic. On the other hand, there may be disagreements
over what objectives are worthy of pursuing; that some people would like to see liberty
prioritized over equality, for example, and some other group of people may want to see
equality promoted at a cost to liberty—those are both valid perspectives and it is worth
understanding that when values are at issue there may actually be no resolution. Two reasonable people can disagree over how
valuable various objectives are, and when that’s the case simply recognizing that
the difference cannot be resolved at the level of discovering that somebody is correct and
somebody is incorrect because in fact both positions are equally valid. I would say that there is a failure in the
way we view argument, that in general I think our politicized and polarized atmosphere has
caused us to look at arguments as always tactical. And one thing that you find when you interact
with people who are very adept intellectually is that they are often capable of putting
aside suspicion about the motives of the people with whom they are arguing, and they will
argue not to win but to discover what is true. And it’s a very different state of affairs,
because although nobody likes being shown to be wrong the great thing about being shown
to be wrong is that it gives you the opportunity to correct your understanding and to be wiser
the next time you encounter the question, rather than entrenching yourself in a wrong
position and suffering the costs of being wrong every time you encounter the question. So, putting aside a desire to win and substituting
a desire to discover what is true is the key to discovering truth through argument, which
benefits everybody who participates; those who have turned out to be correct, those who
have turned out to be incorrect and in general what typically happens is one will discover
that nobody was perfectly correct, and then all sides have increased their understanding
based on hashing out the details of what they disagree over. We’re heading into a very dangerous phase
of history where a large number of people, especially young people, have become convinced
that the free exchange of ideas is not only no longer necessary but is actually counterproductive;
and so they set out to silence those who have opinions at odds with theirs. Some of the people who have opinions at odds
with theirs truly believe abhorrent things, but the problem is: until you fully understand
a topic, you don’t know which opinions to shut out. One has to actually engage beliefs that are
at odds with your own beliefs in order to figure out whether what you believe is correct,
and to improve where it isn’t correct. So shutting down speech has become the mode
for a large number of individuals who believe they see very clearly what is wrong with civilization
and what must be done to improve it—and they are unfortunately shutting down people
who have vital things to tell them that they definitely need to know.

Leave a Reply

Your email address will not be published. Required fields are marked *