cc
>> I'm Jessica Flack and I'm the
co-director of the Center for
Complexity and Collective
Computation in the WID, and
before I introduce tonight's
topic and speakers, specials
guests, I would like to first
thank some of the folks in WID
who made tonight happen.
Maria Peot, Marianne English,
Jenny Eagleton, Sarah Shapiro,
and Beth Misco, WID's
outstanding administrative
outreach team.
I'd also like to thank Wisconsin
Public Television, who's in the
back filming the lectures
tonight.
Everybody at University
Communications who's done a
great job with help with
advertising.
Folks at the Isthmus for
accommodating changes the last
minute.
And, of course, the IT team,
Alan Ruby in particular here at
WID for their support tonight.
And, finally, the Center for
Complexity and Collective
Computation is indebted to
John Wiley and to the Templeton
Foundation for very much
appreciated critical support,
financial support of tonight's
lecture series.
Without this support, we could
not have these kinds of events.
So, tonight's lecture on the
roles of energy and information
in the 21st century biology is
part of a new public lecture
series sponsored by C4, that's
the Center for Complexity and
Collective Computation, and the
series is called John
von Neumann Public Lectures
in Complexity and Computation.
And the lectures are going to be
each month at about 7:00 PM the
second or third Wednesday
over the course of the fall
and spring semesters in 2012
and 2013.
We have a great lineup this
fall.
The next speaker, the next
lecture is on October 17th, and
it's by Graham Spencer who you
might know as one of the
original founders of Excite, and
he's currently a partner at
Google Ventures.
And he's going to be talking
about programming the Internet.
He's a super smart, broadminded
computer scientist.
He gives fun, insightful
lectures, so I hope to see you
guys all there.
Energy and information are two
of the most fundamental concepts
in science, effecting phenomena
over a vase range of scales.
From quantum mechanics to the
origin of life to the evolution
of cooperation in human and
animals society to, also as
you'll hear about tonight, the
size and pace of cities.
In tonight's lecture,
Dr. Geoffrey West and Dr. David
Krakauer, who I will introduce
momentarily, will explore how
information and energy have
shaped nature, and they'll
discuss the implications and the
principles of the future of
science, the future of biology,
and for society.
Some of the questions that we
will tackle over the course of
the evening tonight include,
what really are energy and
information?
The laws of physics, and
specifically the first law of
thermodynamics, tells us that
energy cannot be created or
destroyed.
As such, it's said to be a
conserved quantity.
Meaning that the total energy in
the universe is a constant.
It is this property of energy
coupled to the fact that it can
be hard to transform into
something useful that makes it
seem like a finite expendable
resource.
Now, is the same true for
information?
Or can information be creative
and destroyed?
And if so, what does this mean?
So, as you'll hear tonight,
existing theories of information
stress that information
increases our understanding of
the world around us.
In formal terms, we say that
information reduces the
uncertainty about the
environment.
But there's nothing in this
definition that captures, I
think, the fact that most of us,
for most of us in this audience,
information is really about
meaning.
It's the knowledge itself, not
about reducing uncertainty about
the knowledge that we're
interested in.
So existing formal theories of
information have nothing to say
about this, what you might call,
semantics.
Do we need a new theory of
information that includes
meaning or semantics?
That's one of the questions
we'll ask tonight.
Can we use existing formal
theories of information?
Or the new ones that we might
build to overcome energetic
constraints.
Or does information only help us
find solutions to harness energy
more efficiently?
These are deep and unanswered
questions in physics and
biology.
So, finally, we are going to
ask, what would a science of
energy and information mean for
physics, for computer science,
for biology, and for society?
So our first speaker is
Dr. Geoffrey West.
Geoffrey is a theoretical
physicist.
His primary interests have been
in fundamental questions in
physics, especially those
concerning the elementary
particles, their interactions in
cosmological implications.
Geoffrey received his BA from
Cambridge in 1961, his doctorate
from Stanford in 1966 where he
returned in 1970 as a member of
the faculty.
He was the leader and founder of
the Higher Energy Physics Group
at Los Alamos National Lab.
And then he moved to the
Santa Fe Institute in 2003 where
he became a distinguished member
of the faculty and eventually
became president of SFI for 2005
to 2009.
His long-term fascination has
been in the origin of universal
scaling laws that pervade
biology from the molecular
genomic scale up through
mitochondrian cells to whole
organisms and ecosystems.
He's currently working to extend
the theory of scaling as it was
developed in biology to social
systems, with the goal of
understanding the structure and
dynamic of social organizations
such as cities and corporations.
Tonight, he will talk on energy
scaling the future of life on
Earth for about 45 minutes.
Our second lecturer is David
Krakauer.
David is the director of the
Wisconsin Institute for
Discovery, co-director of the
Center for Complexity and
Collective Computation,
professor of genetics at UW
Madison and external professor
at the Santa Fe Institute.
A graduate of University of
London and Oxford University,
he's authored more than a
hundred scientific papers,
served on several journal
editorial boards, and works to
bridge the gap between academia
and business and also between
the physical and biological
sciences on the one hand and the
social sciences and humanities
on the other.
His research focuses on the
evolutionary history of
information processing
mechanisms in adaptive systems.
But his mind, and I know him
well, is perhaps best described
as sitting in an intellectual
spaces deeply influenced by, I
guess, I think, a somewhat
idiosyncratic group of some of
history's most interesting
contributors including John
Louis Borges, John von Neumann,
Turing, Bill Hamilton,
and Sir Thomas Browne.
Tonight, he will talk for
30 minutes on the power of
information to transfer in the
21st century.
Following the lectures of David
and Geoffrey, Steve Paulson from
Wisconsin Public Radio will lead
a panel discussion on the role
of energy and information on
21st century biology.
Steve is the executive producer
and interviewer with To the Best
of Our Knowledge, a well known
science radio program.
Paulson has written for Salon,
Slate, the Huffington Post, the
Chronicle of Higher Education,
The Independent, and many other
publications.
His radio reports have also been
broadcast on NPR's
Morning Edition
and All Things Considered.
His recent book, Atoms and Eden:
Conversations on Religion and
Science, was published from the
University Press.
So please join me for a bit of a
marathon session and welcome
Steven, David, and Geoffrey.
Thank you.
[APPLAUSE]
>> Okay, thank you.
Thank you, David, thank you,
Jessica, and thank you for
inviting me to give this talk
and be part of this interesting
discussion about energy and
information So, this is the
title that was given to me,
mostly because I was delinquent
in giving one myself, and it
does sort of cover what I'm
going to talk about, but this is
kind of a landscape that I'm
going to be talking about
tonight in terms, primarily, of
energy and the ideas surrounding
metabolism in biology.
And it's a very idiosyncratic
view, and as is clear from the
introduction, I come to it as a
theoretical physicists wondering
into areas of biology and the
social sciences.
And I'm going to begin it, I'm
going to take as a point of
departure, something that's
relevant to this, namely this is
supported by the Center for
Computation and Complexity, or
whatever, complexity was in it.
[LAUGHTER]
And I wanted to use as a point
of departure a statement
by Stephen Hawking,
who is not someone that works on
complexity and not someone that
works on biology, but is a
highly reductionistic
physicist.
And here's something he said
about 10 years ago: "Some say
that while the 20th century was
the century of physics, we are
now entering the century of
biology, what do you think of
this?"
And many people believe the 21st
century is the century of
biology and it certainly will
be, but more importantly, and
this I do believe, is what
Hawking says, "I think the next
century will be the century of
complexity" because many of the
problems we're going to face and
are facing to do with complex
adaptive systems, and I'll be
talking a little bit about them,
and I want to use this, as I
say, as a point of departure
because implicit in this is an
idea of what complexity is, and
that's something I'm not going
to discuss except to use it to
talk about what is not complex
first because it brings in the
whole role of energy.
And this is a system we're
familiar with, the solar system,
and this is a system that is
often thought of as not complex.
It is simple, and by simple, it
means that the whole
dynamic of that system, the
motion of the planets around the
sun and, therefore, of the
satellites that power our
communication systems, are
encoded in two very simple
looking equations.
This is a highly non-technical
talk.
You don't have to understand the
equations.
Everybody is familiar with the
idea F equal MA.
But there are two laws proposed
by Newton that kind of covers
everything to do with the motion
of the solar system, the planets
around the sun, but much, much
else.
Much of modern physics is based
on these equations or versions
of these equations, but critical
in this is the flow of energy.
The idea that energy is the kind
of fundamental thing that's
underlying this motion.
And one of the crucial aspects
of that has already been
mentioned by Jessica: the
conservation of energy.
But if you add to it another set
of equations which you don't
have to understand but important
to understand conceptually, the
equations of Maxwell for
unifying electricity and
magnetism and giving rise to
understanding radiation,
electromagnetic radiation.
That pretty much encompasses
much of the kind of world that
we live in, and just to take
these two equations with these
equations is how your cell phone
works and allows everybody to
communicate and everything else
that's pretty much around us.
And the word simple applies too
because they invoke the idea
that you can encode everything
in the symbolic fashion, and
from that, everything else
follows, even down to the
quantum level.
And one of the things I want to
emphasize is that, again, these
are to do with the flow of
energy and because they are
simple and they are written in a
mathematically precise form, you
can calculate things somehow in
principled to any degree of
accuracy.
And one of the great triumphs of
20th century science is this
calculation which is calculating
the strength of the magnet
associated with the electron,
and here's the theory to 12
decimal places and here's the
experiment agreeing with it to
12 decimal places.
It's kind of extraordinary, and
this is quite unusual, actually,
in the world that we live in
because the world that we live
in, oh, I forgot about that.
I put this slide in just to show
you some of the main characters
that you're familiar with that I
think played an important role
in developing ideas of energy.
One is Newton.
That's a kind of rock star image
of Newton.
[LAUGHTER]
This is Einstein
looking like Einstein.
And this is Max Planck,
the origin of photons.
And this is someone--
Oops.
This is someone that most people
are not familiar with.
It's another, there was a famous
Bill Hamilton here, biologist,
very famous ecologist/biologist
that Jessica actually mentioned.
This is a really deep
Bill Hamilton.
This is Bill Hamilton.
He was an Irishman who sadly
drank himself to death, but he
encoded something extremely
important about energy, and the
physicists in the audience will
forgive me, but basically he
said that all physical systems,
in some sense, minimize the
amount of energy that needs to
be expended.
So this kind of a least minimum,
least energy principle
Hamilton's principle is actually
called, technically, the
principle of least action, but
roughly speaking, this is the
framework in which all of
physics that encompasses
everything from the macroscopic
down to string theory operates
from is using those principles.
And it's has to do based on
energy.
So, here's a picture of energy.
This is energy.
We're familiar with it.
This is highly complex use of
energy because these are highly
non-linear effects and you're
familiar with them.
You're also familiar with this.
This is the energy that we
create on the planet through
objects like this, and I'm going
to talk a little bit about that
shortly.
And these are driven, of course,
by us, which involve energy.
So they're these multiple levels
of energy, and they go all the
way down to the thing that
drives these, us, that drives us
to create the light and the
energy by which we live, our
brains and ourselves.
And these, of course, also
operate by energy themselves,
but they invoke something new
that has typically not been
considered in physics.
And they invoke information, the
idea that information
necessarily is needed to be put
into the system in order to
understand the way these systems
have evolved and the way they
organize and the way they
behave.
And these systems are typically
outside physics, and they form
much of the work of biology and
the social sciences, and they
are complex but, more
importantly, they evolve and are
therefore adaptive.
So these are very different in
their nature from the kind of
simplistic system that we see in
the motions of the planets or
even, might I say, in thinking
about string theory as the
origins of the universe because
we can write definitive, precise
equations in principle for
those.
Whereas for these, we have this
extraordinarily challenging
problem of both integrating
information in it and
understanding the role of
energy.
And what I'm going to explore
tonight is how far one can take
the traditional physics thinking
without invoking, explicitly,
information to understand some
of the organization and dynamics
of these systems.
And then David will talk about
information, and maybe in the
discussion we'll talk a little
bit about the integration.
So here's another kind of system
which you're familiar with, a
forest.
And all of these systems obey
the second law of
thermodynamics, the conservation
of energy, and, therefore, the
production of entropy which is
the kind of random stuff, the
dissipative kind of energy that
comes from doing work and
creating order.
It is the resulting disorder
that comes from ordering the
system.
And one of the challenges, and I
would even go as far as to say I
think the major challenge
conceptually in 21st century
science is understanding this
integration between energy and
resources on the one hand,
metabolism and biology with
information genomics, maybe, and
integrate and get a complete
understanding of these.
These are typically considered
separately, roughly speaking, in
the kind of academic way in
which we think about these
things.
So, what I'm going to talk about
in these systems is, as a
physicist, I want a
quantitative, predictive,
mathematizable in principle,
understanding of the dynamics of
organization of these systems.
And it is clear for the kinds of
systems I just referred to, we
cannot do what we did for
understanding the magnetic
moment of the electron.
We're not going to get equations
that produce precise
calculations, but maybe what we
can get is what we call
coarse-grained descriptions.
So, for example, the kinds of
questions that we ought to be
able to answer quantitatively
are questions like this: why is
it that I can look across this
room and say everybody will be
dead in a hundred years?
In fact, if I look more closely,
I would say that...
[LAUGHTER]
Including myself, might I say.
But where does a hundred years
come from for a typical
life span of a human being?
Why isn't it 10 years or a
thousand years?
And how is that connected to
underlying molecular structures
of the genome and the complex?
Where in the hell does that come
from?
And why is it that the same
stuff, if it were a mouse, only
lives two or three years?
So those are the kind of
coarse-grained questions that
one ought to be able to answer
in a predictive framework, and
I'm not going to talk much about
these, but in what role does
energy and metabolism play in
these?
So here's another set of similar
kinds of coarse-grained
questions.
Why do we need eight hours of
sleep?
Why do we need to sleep?
Why is it not two hours or 18
hours like a mouse or four hours
like an elephant?
Where in the hell do these
numbers come from, and why is it
that mice get many more tumors
during their life span per unit
volume than human beings?
And, therefore, why in the hell
are we doing all these
experiments on mice and trying
to draw conclusions about human
beings?
We need to understand those
kinds of questions, and the one
I really love is, why is it very
good to grow babies in your body
but not grow cancer which is
also part of you?
So these are kind of
coarse-grained questions, and
one that has occupied me in the
last few years is, are cities
and companies just large
organisms that are biological or
are cities organisms just, is
New York a great big whale and
Google a great big elephant, and
if that's true, how come
organisms die, companies die,
but cities don't?
And in fact, I would say,
roughly speaking, almost no
cities die, although you can
think of counter-examples.
We have done awful experiments
to test that.
We've dropped two atom bombs on
two cities, and 30 year later,
they're functioning.
Companies, all companies, die.
And so one of the questions is,
in a coarse-grained way, can we
predict when Google or Microsoft
are going to go bust?
Because they surely will
eventually.
Okay, so those are the kind of
questions, and I want to lead
into all of this, this is
biology but I want to lead into
it via cities because we are
facing an extraordinary
challenge, and this is
addressing the question of how
biological our cities and can we
understand them that way.
We are facing this extraordinary
challenge that we live in this
exponentially expanding universe
of socioeconomic quantities And
just to give you an example,
we've gone from a few percent
being urban to over 80% being
urban in just 200 years.
The world's crossed the halfway
mark.
We're going to go to somewhere
close to 80% by 2050.
China's building several hundred
new cities in the next 20 to 30
years, and indeed, this
statistic ought to freak
everybody out that every week
from now to 2050, on the
average, one and a half million
people are being urbanized.
Therefore, every two months,
there's a New York metropolitan
area coming onto this planet.
So by the beginning of December,
there's another New York
metropolitan area.
By the beginning of February,
there's another one.
By the beginning of April,
there's another one and so on,
inextricably for the next 30-40
years.
Now, can I use expletives in
this talk?
>> Sure.
>> Okay, how the [###] are we
going to deal with that?
[LAUGHTER]
This is an extraordinary
challenge.
So I want to get into that, and
it is related to energy.
Okay, so, the other aspect to
recognize is that all of the
tsunami and problems we're
facing from global warming to
the environment to health
problems to crime to resource
problems, water, energy, and so
on, all are driven by
urbanization, by people in
cities.
So that's the problem, but
cities are also the solution
because they suck all of you
into Madison.
Cities suck in people, and they
are the sources of ideas,
innovation, and wealth creation.
So, this is what cities are
represented by.
The good of cities, this is
what, in fact, attracts people
to cities.
All these good things, culture,
music, goods and so on, driven
and participating in an ever
expanding, exponential expanding
economy, but if this is the
thermodynamic system, which it
is, it produces entropy.
So it produces socioeconomic
entropy.
Incidentally, side comment, if
you look in any of the standard
economics books, you never see
the word entropy, but what is
more amazing is we've looked in
five of the standard texts, only
in one did the word energy
appear.
That's a side comment.
Provocative comment.
So here's socioeconomic entropy.
Okay?
You're familiar with all this.
And the question is, is that
what New York and San Francisco
looked like 2050?
Or that?
Or like this?
Or this?
This is what we want.
Or even like that?
Beautiful.
Or like that in a hundred years.
Or that.
And most importantly, to
maintain that which is a social
buzz of a city.
The interaction of people.
So, here's the question, given
this extraordinary role of
cities, we need a science of
cities.
And I'm going to use energy as
the major piece of that.
So the question is, can we have
a science of that, and is that
another version of this?
Well, if it is, that would be
good, actually.
If this were another version of
that because you ask any
question that has a quantitative
metric associated with it, like
how many trees are there of a
given size, how many leaves are
there on a given branch of a
given size, how much energy is
flowing through each one, how
big is the canopy, what is the
mortality rate, what is the
growth rate, how far apart the
trees of given size, etc, all
those can be answered in a
quantitative conceptual
framework with a bunch of
mathematical formulas which,
in a coarse-grained way,
agrees with data from forests
all across the globe.
And the question is, can we do
the same for this?
Sorry, oops.
I went backwards.
So, well here's some of the
commonalities, metaphorically
at least, between social
organizations and biological
organisms which we're all
familiar with, and I'm going to
show you just very briefly some
data that substantiates the idea
that we understand a little bit
about generically the way
forests work.
So what this is, is just the
plot of the number of trees of a
given size.
The theory, which I'm going to
talk about momentarily, the
theory predicts that it goes as
the inverse square of the
diameter of the tree trunk.
So the number of trees of a
given size goes to the inverse
square of the diameter of the
tree trunks which means if you
look at trees that are twice the
size of any others, there's a
quarter of the number, two
squared, a quarter of the
number.
If it were three times, it'd be
eighth and so on.
So, you get the idea.
And there's data and what is
important about it, not only
does the data fit the theory,
but you can see it's just over a
period of 30-40 years, this
forest, which is a tropical
forest, has changed
dramatically.
Lots of trees have died, new
ones have grown.
Record turnover.
But this law has remained
robust, and there's data into
the '90s and 2000s that continue
with that.
I'm going to miss this.
So here's the kind of framework
that I want to invoke.
Here's a cartoon of said
ecosystem.
I could have done the city.
And this is the lens that I want
to look at it through, and all
of these equations actually are
really to do with energy flow.
Okay.
So, another concept I want to
talk about very briefly is the
concept of scalability that a
system needs to be scalable if
it is a system that is going to
be resilient and evolvable and
adaptive to change.
It needs to be scalable just in
the sense that we are scalable.
This is us.
We have scaled over a range of a
hundred million in size and
mass.
We go down to something that
sits on the palm of my hand all
the way up to something that's
much bigger than this room, and
we're all pretty much the same
thing.
We may look different, but in
terms of, in a coarse-grained
average way, in terms of our
life history and our physiology,
we're all pretty much the same
thing, and I'm going to show you
that in a moment.
But even going further in going
from the molecular levels, I
could have put the genome here,
this is what the cartoon of the
molecules that produce your
energy, all the way through
mitochondria to cells to
multicellular organisms like
this that produce things like
this, houses, and produce things
like this so that, in fact, what
goes on here scales all the way
through this to keep this going.
All of these have to be scalable
because they are highly complex
and continuously evolving and
continuously adapting.
And so there has to be a kind of
scalability, and out of that
comes the idea that you can't
have this being arbitrary.
There have to be emergent laws
that govern this, and I think
the next slide, ah, yes.
So, in a similar way to the way
we scaled in a much shorter time
frame in terms of our growth,
and one of the things that I
will talk about briefly in a
moment is about growth and the
fact that the same theory, the
same theoretical framework,
which I will come to in a
moment, which purports to
explain the organization,
dynamics, growth of forests that
I talked about earlier applied
to us, to mammals, gives rise to
equations like this or pictures
like this for the,
what is called a growth curve.
This is the weight as a function
of age.
This is us.
It happens to be a rat version
of us.
And the line there, the solid
line is a prediction from the
theory and the points are the
data, and you see it's very
good.
And this can be done for any
organism, and if I had time, I
would show you lots of other
wonderful fits.
But the point is that we can
understand growth, and I will
come to that in a moment, but
the important thing I want to
stress now, moving to social
systems, is if this were taken
over to socioeconomic systems,
mainly that the organism grows
quickly and then stops like we
did, I mean one of the amazing
things about biology is that we
eat, we grow, and then we go on
eating but we don't grow, and
the theory explains that.
I don't have time to go into it.
Maybe we can discuss it later.
And it's all to do with the
input of energy and the
distribution of energy and the
scaling of the networks that are
inside us.
But the point is that this would
be very bad if it were a
socioeconomic system because
this is our image of
socioeconomic systems, and I
showed you a picture before of
the economy, always open and
expanding.
And I want to come back to this
in a moment.
So, here's a picture of the
extraordinary scaling in
biology.
So, let me spend a minute on it.
What you see that's plotted here
is the most fundamental
quantity, certainly from a
physicist's view, in biology
That is your metabolic rate.
How much energy do you need per
day to stay alive.
The 2,000 food calories a day
that you eat to stay alive, and
here it is plotted on the
vertical axis against the
weight, the mass of the organism
on the horizontal axis, and it's
plotted in a Byzantine way.
So I can put mice and elephants
on the same graph, and the
Byzantine way is that instead of
being linear, it goes up by
factors of 10.
One, 10, 100, 1,000 watts.
One, 10, 100, 1,000 kilograms.
So we can get everyone on the
same graph.
And what you see is something
extraordinary.
That there's an extraordinary
regularity when we're dealing
with maybe one of the most
complex and diverse phenomena in
the universe.
Something unbelievably simple,
ridiculously simple has emerged,
and this, at some level,
astonishing because we believe
that each one of these
organisms, each subsystem of
that organism, each organ, each
cell type, each genome has
evolved with its own unique
history and its own unique
environmental niche.
So, you might have expected from
that natural selection viewpoint
that if I plotted anything
versus the size of an organism,
it'd be all over the map
representing, manifesting the
historical contingency involved
in all of this.
No, that's not what you see.
You see something ridiculously
regular, and that regularity has
some extraordinary features.
For one, the slope of it is
approximately three-quarters,
and that three-quarters is less
than one.
One would be a slope,
a line like this.
And one is what you might
naively expect if you thought
everything were pretty much the
same at the most naive level.
You'd double the size of an
organism, double the number of
cells, therefore doubling the
amount of energy needed to keep
it alive.
That's not what you see.
You see something less than one
which means, in that language,
double the size of an organism,
approximately, instead of
needing twice the amount of
energy, you only need, roughly,
75%.
So there's this extraordinary
economy of scale as you get
bigger, and this is called
sub-linear behavior.
So it turns out that if you look
at any physiological variable
that you can measure or any life
history event that you can think
of, it all has this kind of
character.
And I'm going to show you a
couple of examples.
There's heart rate, a very
mundane kind of quantity, versus
body size.
I'm not going to explain it in
detail, but what you see is,
again, a very simple scaling
law.
This is your thinking.
This is your white matter to
gray matter in your brain.
Again, a very simple law
evolving.
This is your genomes, and what
you see from this is that you
have this regular behavior and
the other remarkable thing is
all the slopes of those, like
that three-quarters, are simple
multiples of one-quarter.
Four plays this amazing role in
these scaling laws of
extraordinarily complex
phenomena.
What is underlying them, and
this is work I did with
colleagues at the Santa Fe
Institute, and I'll tell you who
they are in a moment, the idea
is how in the hell can this be
that all of these phenomena have
simple scaling laws, and why is
it that the scaling laws of the
mammals are the same as for
birds, fish, crustacea, insects,
and so on?
How can that be?
Well, what could be universal
among them?
What could be universal is
somehow the supply of energy and
resources through networks has
to be a commonality among them,
and the idea here was, is that
it is the mathematics, the
universal mathematics and
geometric topological properties
of those networks that transcend
design that are constraining
natural selection.
That's the idea.
And I don't have time to go into
that other than to say that if
you look at these networks,
that's your brain, that's your
white to gray matter of the
brain, that's your lungs, that's
a tree, that's a little thing,
that's inside an elephant, but
they're ubiquitous.
And they have these, the
postulate is they have these
properties, and if you use those
properties, put them into fancy
mathematics and use Bill
Hamilton's principle of energy,
least action kind of ideas, out
pop all of these marvelous
scaling laws, including the
number four, the one-quarter,
which happens to be, for your
information, the dimension that
we live in three plus one.
And just to add to that, the
plus one has to do with the fact
that these networks have a
fractal-like behavior.
That's a kind of tangential
remark.
So, here's others.
This is down at the
intracellular level.
This is the mitochondrial level
and so on.
So here's that growth again.
So, the idea is we have this
theory.
The claim is we have this
wonderful theory based on
networks, the mathematics of
networks, and that from that we
can understand these scaling
laws, but also, if you like, we
understand the network.
We can understand the structure
of the network within each of
the organisms.
And you can use that to go back
to this growth and use this kind
of idea that if you eat, you
metabolize.
That energy goes through the
network, controls the network
which maintains the cells that
are there, replaces ones that
are dead, and grows new ones.
And then, from that, you can
mathematize it, and that gives
rise to this equation which
actually invokes all of the
properties of the network and
gives rise to these kinds of,
what we call, sigmoidal, meaning
they stop growing, curves.
Okay.
So here's kind of a summary of
this part.
We have these amazing non-linear
quarter power scaling laws.
They represent an economy of
scale.
The one-quarter is intimately
related to flows of energy in
networks.
One thing I did not emphasize is
the pace of life systematically
slows with size.
So that, for example, heart
rates decrease systematically
when in to these quarter power
scaling laws.
The rate of diffusion of oxygen,
say, across membranes decreases
with size.
Organisms live longer
systematically.
And so on and so forth.
Everything to do with time slows
down the bigger you are.
Growth is sigmoidal.
It reaches a stable size of
maturity.
And to emphasize again, the idea
is this is all based on
mathematics and properties of
networks.
So, and this is a sustainable
system.
It's sustainable,
phenomenologically, because it's
been around a couple billion
years or more.
Part of that sustainability, I
believe, is actually invoked in
the fact that things stop
growing.
They stabilize and most
organisms, not all, we can
discuss this, most organisms
spend most of their mature life
in a stable size configuration.
Not all and there are
exceptions, and we can discuss
that.
But this leads to a sustainable,
resilient system.
And the question is, I'm sorry.
Yes, so this is the motley crew
that I work with.
Some biologists, some
physicists, some chemists.
But I then took the work over
into social organizations, and
there's another motley crew.
The top guys did all the work.
All the guys at the bottom are
famous and did none of the work.
[LAUGHTER]
And the guys in the middle
started off with me and then
moved on to other things.
Okay, so here's the question,
the first question about whether
cities and companies are
biological.
The first thing is, do they
scale?
So what we said before was that
whales live in the ocean and
elephants have trunks and
giraffes long necks and we stand
on two feet and mice scurry
around, but in fact, at the kind
of 85% level, they're scale
versions of one another in this
non-linear quarter power
fashion.
In this mathematical fashion,
they're scaled versions of one
another.
Is that true of cities?
And is it the kind of universal
scaling as it is in biology?
So, are cities scaled versions
of one another?
Is New York a scaled up
San Francisco, which is
a scaled up Madison,
which is a scaled up Santa Fe?
Even though they look completely
different.
Well, you can only answer that,
at first light it's hard to
believe, every city feels so
unique.
But of course, you can only
answer that by looking at data.
But one of the reasons you might
think there are scaling
phenomena, there are these kind
of generic universal
similarities, is because cities
are, indeed, networks.
There's the obvious networks of
roads and electrical lines and
all the rest of it, like this,
and transportation systems, but,
more importantly, there's this
network which never occurs,
essentially, in biology.
It is the network of us
talking to one another.
It's us, social networks,
that is unique.
It's only been around on this
planet for maybe 10,000 years.
Maybe in the universe for all we
know for 10,000 years.
Who in the hell knows.
But, certainly, in our solar
system.
This is it.
It's new.
And it's this.
This is just people talking to
one another, how they interact
with one another.
So, oops, I seem to have lost
the slide.
So, I use this slide.
What is the city but the people.
Indeed, the city is not the
buildings and the roads and all
this other stuff.
It's us.
We are the city, and that stuff
is a manifestation of the
interactions between us.
And that's something that is
outside of biology but is
integral to biology because it's
driven by biology, and it's
driven, in large part, by
energy, and I'm going to talk
about that.
So, the other part of the
network is not just we interact
with each other but we prosper.
So, I'm going to not say much
more about that but go straight
to the data and ask the question
about scaling.
So this is a graph plotted in
the same way as those biological
graphs were, but this is a
mundane one.
The number of petrol stations,
I was working with European
colleagues, the number of gas
stations in a city as a function
of its size in various
countries.
And you can see, there's pretty
good evidence of scaling again,
there's a very simple line on
this plot, plotted this way.
Very much like biology, and like
biology, this would be linear so
that you don't, when you double
the size of a city, you don't
need twice as many gas stations.
In fact, what you find from
this, you only need about 85%
more gas stations.
The slope of this line is not
three-quarters as it is in
biology.
It's about .85 it turns out.
Most importantly, though, is
that they're pretty much the
same.
The slopes of these are the
same.
The same economy of scale occurs
in all these European countries
but it occurs in Columbia, it
occurs in Chile, it occurs in
China.
Anywhere you look across the
globe, you get the same scaling,
and if you look at any
infrastructural quantity that
you can measure, you get the
same scaling as this.
There's a consistent, always
roughly 15% saving every time
you double.
Okay?
>> Geoffrey?
>> Yeah.
>> Ten.
>> Good, perfect.
So this is kind of a mundane
quantity, and that's like
biology.
Looks just like biology
except it's .85
instead of .75.
But this is something that
doesn't exist in biology.
Wages, super-creative people,
and you ask, how do those scale
with the size of the city?
And here they are.
This is wages at the top, and
this is super-creatives,
so-called super-creatives by a
man named Richard Florida.
And what you see is there's more
fluctuations in the data, but
there's pretty good evidence of
scaling.
But, most importantly, the
scaling is different than in
biology.
The slope of these are bigger
than one rather than less than
one, and this is critical.
Namely, the slope of these you
see is roughly about 1.15.
So what that says is that if you
double the size, instead of
needing less per capita, you
have more per capita, and I will
come back to this in a moment.
So here's wages, super-creative.
Here's patents.
This is some crude measure of
the innovation of a city.
The slope of this is also about
1.15.
This is the crime in Japan, a
little bit bigger 1.15.
This is police, tax receipts,
construction, all about 1.15.
And there they are, all plotted
on one another, just a few of
them to show a universality of
income, GDP, crime, and patents.
Very different things but they
all scale in the same way with
quite a lot of fluctuations, and
the slope of that is roughly
1.15.
And we believe that underlying
are the universality of social
networks.
The way we interact with each
other is a kind of universal
phenomenon whether you're in
China, Japan, Columbia,
United States, anywhere.
The fundamentals of the social
interaction is universal, and
that is what is being manifested
in these scaling laws.
These scaling laws are for
multiple socioeconomic
quantities anywhere in the
world.
We've looked at data in Europe,
United States, China, Japan,
Latin America, and so forth.
What is interesting to
substantiate that is this graph
which is very recent data which
is the number of cell phone, the
amount of people you call on
your phone, on the average, as a
function of city size.
So this is a kind of way of
tapping in to the network, the
social network, and if this
theoretical framework of the
mathematics of that social
network is the right one, the
slope of this should be about
the same as these other
networks.
And indeed, we could have
plotted this on top of this, and
it would fit right on top of it.
So this is the number of people
you talk to, on the average, as
a function of your city size.
So, I'm going to miss this.
That's what I said a moment ago.
And here's kind of a summary of
this.
The good, the bad, and the ugly.
If you double the size of a city
or if you look at a city that's
twice as big as another city,
then systematically across the
globe, and you double the size,
you can go from 50,000 to
100,000 or 5 million to 10
million, it doesn't matter where
you begin to double the size,
systematically, income, wealth,
patents, colleges, creative
people, police, crime, AIDS,
flu, all of these things, the
good, bad, and the ugly all go
up together to the same degree.
And, at the same time, you save
about 15% on the infrastructure.
Big cities are good, and the
bigger ones are better still
both at the collective level.
They save on infrastructure per
capita.
I had a graph which I flipped
through.
They save on carbon footprint
per capita.
And at the individual level,
everybody is attracted to cities
by the first lot of those,
income, wealth, number of
patents, colleges, and the
general buzz of city, and ignore
the other bad things.
The fact that they're coming
along with it, hand in glove,
there's more crime, disease, and
so forth.
Okay.
So, the network dynamics is the
one that dominates this again,
and the networks dynamics has
the following interesting
phenomenon that if it's
sub-linear scaling then we have
this pace of life slowing with
size, as I mentioned, but if
it's super linear, bigger than
one like we see in socioeconomic
systems in cities and driven by
the social networks, then the
theory tells you the pace of
life systematically increases
with size.
And so, the pace of life in
New York is systematically
faster than it is in Madison,
which is systematically faster
than it is in Espanola, which is
a tiny town near Santa Fe
in a systematic way.
And here's some whimsical data
to show you that.
We've looked at many things.
On the left is some data on
heart rate versus body size.
The one on the right is walking
speed.
I show you walking speed.
People have actually measured
walking speed perfect.
And you can see it
systematically rising, and it's
in reasonably good agreement
with the data.
But why?
Because walking, actually,
walking in a city is a social
phenomenon.
Okay, so I'm going to miss this
out, I think.
Yes.
So, this is kind of a summary,
and I'm going to finish off in
the last few minutes with the
last part of this.
So, unlike biology, super linear
scaling, socioeconomic, which
has to do with wealth creation
and innovation and ideas,
dominates, and so you get this
kind of 15% rule, which I'm not
going to talk about its origins
here, that also leads to a
systematic increase in the pace
of life.
And the last thing that I'm
going to talk about which is
very satisfactory is that if
this is true, which it is, and
it comes from these networks,
then I'm going to talk about
growth and the idea of
open-ended growth following from
this because we're going to use
the same kind of growth equation
that, again, based on energy
equivalent of the flow of energy
through these systems leading to
maintenance and growth, and
that's what we had before for
us.
The sub-linear scaling leading
to boundary growth, and if it's
super linear scaling, there's a
cartoon on the left, this leads
to actually faster than
exponential growth.
And that's very satisfying.
That's great because that's what
we see, and that's what we
apparently love or have loved
for the last couple hundred
years.
And that's great.
However, it has, if you believe
the theory, this has a fatal
flaw.
And the fatal flaw is denoted by
this line.
And it's called, in the
technical language, a finite
time singularity.
And to put it in very simple
language, what it says is that
in some finite time, this damn
growth curve will go to infinite
size, and that's obviously
impossible.
So this is kind of a Malthusian
argument.
Namely, at some stage, you're
going to run out of whatever it
is that's keeping the system
alive, the resources, the
energy, whatever it is, and the
theory tells you what happens.
It says as you go through this
so-called singularity,
you stagnate and collapse.
That's terrible and we need to
avoid that, and we have avoided
it.
And this is how we avoid it.
We avoid it for the very reason
that Malthus and Paul Ehrlich
were rejected.
That is, they did not take into
account the fact that we
innovate ourselves out of these
problems.
So I want to finish on this
note.
So here's the situation.
This is, you start at some point
and you start growing and you
would hit this singularity and
collapse.
So somewhere along here you
better change something because
this growth was done in the
paradigm of whatever the major
innovation is.
It could be the discovery of
iron.
It could be the discovery of
coal.
It could be oil.
It could be the invention of
computers.
The invention of IT, but
something major that has a kind
of paradigm shift and changes
everything.
So, this suggests that the way
out of this dilemma of collapse
is that you innovate somewhere
here, and you start over again.
You reset the clock, and then
you can go on merrily.
And, of course, you're going to
hit another finite time
singularity and collapse unless
you innovate again.
So you have to keep innovating.
So there's a kind of theorem if
you like.
But if you want to continue to
have open-ended growth, you have
to have cycles of innovation.
Well, people have been saying
things like that for a long
time.
Here's the difference, going
back to what we said: as you
grow and you get bigger, the
pace of life increases.
So, first of all, the pace of
life is increasing.
The second thing is, it turns
out the theory says yes you can
do this, but the time to go from
here to here necessarily has to
be, and systematically, a lot
shorter than the time from here
to here.
So that you can innovate, you
have these cycles of innovation,
but they must come quicker and
quicker.
Okay?
So, the image is we're on this
treadmill that is going faster
and faster, which is difficult
to begin with, but at some
stage, you've got to jump from
that treadmill onto another
treadmill that's going even
faster.
And then very soon, you've got
to jump onto another one, and
you've got to kind of keep doing
this so you have this kind of
double acceleration.
And the question is, do you
suffer a kind of global heart
attack from that?
[LAUGHTER]
So, is this sustainable?
Obviously not, ultimately.
And this is not mine at all.
This is someone, I don't know
who he is, but he did it and he
showed it was great.
Oh, this is something I saw in
an English newspaper a week ago
which I thought illustrated the
point a little bit.
So here's something, some of you
may have heard the idea of
singularity from this, what I
consider, little bit looney work
of Ray Kurzweil.
I think it's looney, but he did
some wonderful things of
collecting data.
I don't agree with a lot of
this, but what he's plotted here
is that here's the paradigm
shift.
That is, how long it took to
create each one of these
paradigms versus how long ago it
happened.
Okay?
And what you see is it takes
shorter and shorter times to
make these shifts, and it's
happening faster and faster.
And he drew this.
All kinds of people have done
things like this.
I'm going to finish up.
And, whoops, and interestingly
enough, that straight line, if
you take this theoretical
framework that I just told you
about, this continuous
innovation and jumping from one
treadmill to another, it
predicts, almost exactly, that
orange line.
So that's data kind of
supporting it.
So I'm going to finish with
that, and I'm going to finish
with one last image which is
this just to give you a sense of
things in terms of energy.
Our metabolic rate, I didn't
point that out on the graph.
You remember that graph of
metabolic rate?
Our metabolic rate in watts is
about 90 watts.
A little bit less, women versus
men.
But roughly it's about 90 watts.
We're a light bulb.
One of these bloody light bulbs
is equivalent to one of us.
So, turning a light bulb off,
it's like your mother telling
you if you turn it off, you help
starving people in Sudan or
somewhere.
So it's ridiculous having all
these lights on, frankly.
They're us.
We are extraordinarily
efficient.
That 1,800-2,000 food calories a
day is unbelievably efficient.
It may seem like a lot of food,
but it's just a light bulb.
However, and incidentally, that
metabolic rate is the metabolic
rate we should have for a mammal
of our size.
That's what we should have.
Okay?
And if you add in activity,
hunting and gathering, which is
the way we evolved or versions
there of, that number changes
from 100 to about 250, the
amount of energy we do to hunt
and gather and so on.
And that also scales the active
metabolic rate for mammals.
So that's what we "should have."
Then we started talking to one
another.
We discovered economies of scale
to form communities, and then we
started to innovate.
And once we innovated,
we created all of this [###]
around us, all this wonderful
quality of life and standard of
living that we'd like to have
for everybody.
And we can ask, what is our real
metabolic rate?
What is our social metabolic
rate of having cars, and lights,
and buildings, and all the rest
of the stuff?
That number goes from about
90 watts to 11,000 watts.
And you can turn it around, this
has been done on this graph as I
did here, and ask, how big is
it?
What big an animal are we?
This is about a dozen elephants.
Each one of us in this room and
there are seven billion people
of us on this planet who all
want to be like us, and there's
three billion more coming in the
next 30-40 years, and they all
want to be like us
using 11,000 watts.
So this is an enormous problem,
and I cannot answer the
question, what is the future
and what does it bring?
But I can say
that I'm a pessimist.
So I'll finish on that.
[APPLAUSE]