[Applause]
so let’s start with a very important
question when should we attempt to
engineer shared eco systems now this is
a hard question and it’s one best
addressed by the local communities that
are likely to be affected and when you
go to people and you ask them to share
their concerns and criticism of a
particular proposal you get some people
that are enthused and they ask how they
can help you get some people that have
specific technical concerns that they
want addressed and then you have some
who question the very premise of the
idea that is they want to know how can
we justify our apparent hubris in trying
to engineer a complex ecosystem that we
don’t fully understand now I deeply
respect this particular concern because
I believe that we are morally
responsible for all the consequences of
our actions we are equally responsible
for the consequences of choosing to do
nothing and many ecological problems
around the world are severe enough that
the moral cost of doing nothing may be
too great if you look at Hawaii you have
invasive mosquitoes spreading bird
malaria that are threatening to
eradicate the last few iconic native
bird species you have an invasive fungus
that’s attacking the most abundant tree
in the hawaiian forest and of course you
have species such as rats doing what
they do everywhere which is driving many
native species to the brink of
extinction and over the edge that’s the
natural world what about human effects
here in New England our fondness for
beautiful forested meadow environments
has led us to inadvertently maximize the
number of ticks and of the white-footed
mice responsible for infecting them with
pathogens that cause Lyme disease and
worse so we in the sculpting evolution
group have been working on addressing
all these problems together with the
local communities who know these
environments
and we’ll have to live with the
consequences of whatever they decide to
do but what’s really been striking about
some of these conversations is how many
people have said you know this town hall
discussion about whether we should
genetically engineer the wilderness was
a lot more calm and reasoned than
discussions about most cultural topics
nowadays and that suggests that there is
something wrong so let me try to explain
from the perspective of an ecological
and evolutionary engineer what might be
wrong with our social ecosystems now the
first rule of evolution in the
laboratory when you’re trying to direct
it for a particular purpose is you get
what you select for which is not
necessarily what you want now this is
important to society because of numbers
now please raise your hand if you like
numbers all right I love numbers because
numbers are solid reliable anchors to
reality without numbers were lost in the
fog of unquantifiable uncertainty we
don’t know how important anything we do
is unless we can measure it the problem
is not everything is equally easily
quantified and that introduces bias so
for example we want to maximize human
wellbeing but that’s hard to measure
whereas GDP is a beautiful single number
so we have economic policies that
promote things like flexible labor
markets now please raise your hand if
you were raised in the Boston area some
of you here please raise your hand if
you live within blocks of your extended
family a few within blocks of a trusted
friend from childhood Wow some people
are multiply lucky here how about a
trusted friend from college as well a
few but not very many that is what we’ve
been doing is maximizing growth and the
cost of that flexible labor market is
measured in severed human relationships
and it’s those relationships which are
hard to measure which really make up the
social fabric of society
so I suspect that a lot of the
underlying cause of our problems is due
to that fraying social fabric loss of
trust in one another and only the most
visible of those consequences the most
measurable are things like the rising
deaths from despair
so in short economic policy is broken
along with many things like it that
focus on numbers quantifiable things
like GDP as proxies for hard to measure
things like human wellbeing and that may
have worked once but right now it’s no
longer working another lesson from the
natural world is that evolution has no
moral compass that is to say that which
replicates wins becomes more abundant by
definition regardless of the cost to the
greater system if you look out there in
the natural world more than 40% of all
species are thought to be parasites
ecosystems aren’t optimized for anything
and this is relevant to us because
capitalism amounts to a clever way of
harnessing evolution to satisfy consumer
preferences anyone who can offer better
goods and services that people want at
progressively lower prices is rewarded
with resources to keep on doing that and
you have a virtuous arms-race cycle to
satisfy consumer preferences more and
more effectively this is how we create
wealth and it works very well in some
industries consumer electronics clearly
but capitalism doesn’t seem to be so
efficient in others in the past 50 years
the cost of housing if you control for
size and quality has gone up by 50% the
cost of primary education has more than
doubled with no obvious improvement in
learning the cost of college and of
healthcare have both increased by more
than a factor of 10 again no really
obvious improvements and outputs here so
what’s going on well you can say housing
that’s an effective limited land supply
growing population especially near the
coast
and you can excuse the others by saying
well if you have increased wealth like
we do now health and education seemed
perfectly reasonable things to spend the
excess on but then why is the cost of
infrastructure going up by so much if
you look at New York’s most recent
subway line 2.2 billion dollars per mile
that’s ridiculous
it’s an outlier but it’s ridiculous and
it’s not alone the trend is there and
this is strange because we have much
better technologies for doing everything
required for infrastructure and for
healthcare and everything everything
from better computers to better concrete
mixers we would expect all of these to
be getting cheaper not more expensive
and it’s not going to wages or even
really to benefits despite health care
so what’s going on well again in the
laboratory if you run a directed
evolution experiment for a while it
might work great for a first few
generations but then eventually you’re
gonna throw up some cheaters or in
parasites that just feed off of the
others and as we know if you look at
nature the longer you let it go the
worse this gets so my suspicion is that
capitalism and possibly democracy too
have really just been infested by a
proliferation of evolutionary cheaters
people have figured out more and more
ways of cleverly siphoning additional
resources off and this breaks the system
slowly doesn’t completely break it but
it makes everything more expensive and
it increases inequality and in turn that
contributes directly to political
polarization and anger just because
we’re stuck with increasingly higher
bills for the exact same services we
used to get for much cheaper so who’s
gonna pay the extra it’s a lot harder to
reach agreement so at this point in the
talk you’re probably asking what can we
do well again laboratory how do you get
rid of cheaters we have a very easy way
we apply superheated steam and
30 atmospheres of pressure and they’re
all gone it works great the same sort of
thing seems to work for society if you
look at history that is the reliable way
to reduce inequality and remove economic
rent-seeking is war famine pestilence
and mass death so we really need some
other solutions here and by the way you
should be depressed now I mean come on
these are some of our most eco systems
that we’re proud of I’m supposed to be
working in efficient in clever ways of
harnessing evolution and they’re not
working so if you’re an optimist like me
you’re probably thinking well there’s
got to be some human cultural ecosystem
that’s working really well and I’m a
scientist so I naturally think well
science science is the crown jewel of
our civilization it can be frustrating
and inefficient and let me tell you from
the inside it’s very frustrating and
inefficient it’s very much a matter of
one step back for two steps forward but
it reliably moves us towards a more
accurate picture of the world and that’s
amazing the problem is if you’re a
civilization not all knowledge is worth
having because science is vision just as
technology is power and it’s the dance
of science and technology that is the
engine that drives our civilization
without them it falls and billions will
die with them
well I’m increasingly concerned that we
might inadvertently solve our
rent-seeking problem suppose there is a
discoverable technology out there
potentially in our future that if
discovered and publicized would bring
down civilization you can probably
imagine some things like that but the
really scary ones are those that none of
us in this room or anywhere else has
anticipated if that exists then the
current scientific ecosystem is highly
likely to stumble straight into it
without ever seeing the danger
the reason is that in an age of costless
communication we still do almost all
research including publicly funded
research behind closed doors where
everyone is blind to what everyone else
is doing and most research is done by
teams of similarly trained specialists
who can’t reliably anticipate the
consequences outside their own narrow
field of specialty which means if you
have something that it requires a
synergy of this and that neither team is
going to know that the other is working
on that or even know of the possibility
of that kind of synergistic interaction
this is still very theoretical so you’re
probably not convinced yet so let me
give you an example five years ago no
one imagined that we might be able to
engineer entire wild species but if you
take CRISPR genome editing and you
combine it with gene Drive which is a
natural phenomenon it’s a strategy
whereby genetic elements ensure that
they get inherited put those together
and that’s exactly the capability that
we now have
if you engineer an organism so that it
has the edited changes that we want in
its genome and we encode the editing
machinery next to it then when that
organism mates with a wild one the
offspring will inherit one of each and
in those offspring the editing machinery
will convert the original version to the
edited version it’s a find and replace
because the next generation is
guaranteed to inherit and the next and
the next and the next so four years ago
my colleagues and I first envisioned
this possibility and its potential
outcomes and we were initially elated
but then very concerned so we asked a
bunch of experts from diverse fields
from ecology to security to even
environmental NGOs to see what they
thought we should do and we concluded
that it’s not very physically dangerous
just because it’s slow it spreads over
generations and it’s easy to detect by
sequencing and once you detect it you
can counter it with another find and
replace but it’s very socially dangerous
just because well you can imagine the
head
if some scientist accidentally releases
something into the wild that will
eventually engineer that entire species
it would be pretty devastating to public
trust in science and as previously
mentioned we really need science so how
to stop that well we decided we would
publish our findings our assessment of
what this could do and a detailed list
of technical safeguards that should be
able to prevent that kind of accident
the next year the very same high profile
journal that published our warnings came
out with a report detailing the same
technology under a different name
without using any of the safeguards now
I have no explanation for the journal
but the researchers are brilliant
well-meaning scientists who hadn’t seen
our work weren’t even aware of the
concept of gene Drive because it’s
something for from evolutionary biology
which was not their field and they were
just focused on building a better tool
for laboratory genetics experiments that
could be used by others they never
imagined that it could be applied to
wild populations so there you have it a
powerful completely unanticipated
technology that despite our best efforts
was still independently developed and
publicized by researchers who were blind
to the consequences do we have any
barriers in place to prevent this from
happening for a more dangerous
technology I can’t think of many so at
this point the question can we engineer
social ecosystems is looking a little
bit more urgent that means I have a
confession for you when I say that I’m
an ecological engineer people laugh and
they left because we don’t know how to
do that it’s only because of tools like
CRISPR based gene Drive which let us
harness the strategies of cheaters to
alter populations and by those
populations the associated ecosystem
that’s the only way that we think we can
do it now and similar approaches so the
question is can we apply this to the
social setting what is the equivalent
and I suspect we can now I’m gonna focus
on science because I know the scientific
ecosystem best scientists to be
practicing scientists we need
recognition for what we do which leads
to continued funding so if you want us
to share our experimental plans in
advance with others so that society can
scrutinize them for potential dangers
and intervene early then you need to get
the scientific journals and the funders
to change the incentives now speaking
from experience now I can tell you that
none of them wants to be first it’s a
collective action problem so we need
some way to get the wagon moving without
them I think the answer is to use a
technology to harness it cheat in this
case the cheat is intellectual property
now patent trolls love to use
intellectual property to halt useful
products unless they get paid off it’s a
great way for them to siphon off
resources but if what you want to do is
change the incentives you want
scientists to do something differently
well how about you say you can’t do that
research unless you first publicly
disclose your plans this wouldn’t
normally work but we have an opportunity
here in the form of CRISPR based gene
Drive so here’s how it would work to get
together get together a patent pool of
relevant IP for building CRISPR based
gene Drive ignore commercial licensing
don’t touch that at all and just
regulate the research but offer anyone a
license for free provided that they
share their experimental plans in
advance this will be controversial
because it violates this long-standing
precedent which has very good reasons
behind it that patents should not
interfere with non-profit scientific
research in this case I think there is
very good reasons to make an exception
to that
and that controversy will lead to media
coverage yet it’s really hard to stand
up in public and assert that you have a
right to develop a technology that can
single-handedly engineer ecosystems in
secret without telling anyone what
you’re doing or allowing anyone to have
a look and make sure you’re using
safeguards so favorable media coverage
will then incentivize the funders and
journals to jump on the Nell moving
bandwagon and add their own incentives
if it can work for CRISPR based gene
Drive then you have a precedent and the
strategy might then be applied to
related fields involving technologies
with shared impacts and eventually to
most of the areas were concerned about
throughout the scientific ecosystem and
who knows if it works there perhaps it
could be applied to the other ecosystems
that we think may require a fix or –
let’s be fair this idea could fail it
could fail spectacularly in fact most
experiments in science do fail but
that’s why we would then have to take a
lesson from engineering and that is try
something else
until it works and we have to do that
because when we think that civilization
itself may be at risk the moral cost of
doing nothing is always too high thank
you
[Applause]
so let’s start with a very important
question when should we attempt to
engineer shared eco systems now this is
a hard question and it’s one best
addressed by the local communities that
are likely to be affected and when you
go to people and you ask them to share
their concerns and criticism of a
particular proposal you get some people
that are enthused and they ask how they
can help you get some people that have
specific technical concerns that they
want addressed and then you have some
who question the very premise of the
idea that is they want to know how can
we justify our apparent hubris in trying
to engineer a complex ecosystem that we
don’t fully understand now I deeply
respect this particular concern because
I believe that we are morally
responsible for all the consequences of
our actions we are equally responsible
for the consequences of choosing to do
nothing and many ecological problems
around the world are severe enough that
the moral cost of doing nothing may be
too great if you look at Hawaii you have
invasive mosquitoes spreading bird
malaria that are threatening to
eradicate the last few iconic native
bird species you have an invasive fungus
that’s attacking the most abundant tree
in the hawaiian forest and of course you
have species such as rats doing what
they do everywhere which is driving many
native species to the brink of
extinction and over the edge that’s the
natural world what about human effects
here in New England our fondness for
beautiful forested meadow environments
has led us to inadvertently maximize the
number of ticks and of the white-footed
mice responsible for infecting them with
pathogens that cause Lyme disease and
worse so we in the sculpting evolution
group have been working on addressing
all these problems together with the
local communities who know these
environments
and we’ll have to live with the
consequences of whatever they decide to
do but what’s really been striking about
some of these conversations is how many
people have said you know this town hall
discussion about whether we should
genetically engineer the wilderness was
a lot more calm and reasoned than
discussions about most cultural topics
nowadays and that suggests that there is
something wrong so let me try to explain
from the perspective of an ecological
and evolutionary engineer what might be
wrong with our social ecosystems now the
first rule of evolution in the
laboratory when you’re trying to direct
it for a particular purpose is you get
what you select for which is not
necessarily what you want now this is
important to society because of numbers
now please raise your hand if you like
numbers all right I love numbers because
numbers are solid reliable anchors to
reality without numbers were lost in the
fog of unquantifiable uncertainty we
don’t know how important anything we do
is unless we can measure it the problem
is not everything is equally easily
quantified and that introduces bias so
for example we want to maximize human
wellbeing but that’s hard to measure
whereas GDP is a beautiful single number
so we have economic policies that
promote things like flexible labor
markets now please raise your hand if
you were raised in the Boston area some
of you here please raise your hand if
you live within blocks of your extended
family a few within blocks of a trusted
friend from childhood Wow some people
are multiply lucky here how about a
trusted friend from college as well a
few but not very many that is what we’ve
been doing is maximizing growth and the
cost of that flexible labor market is
measured in severed human relationships
and it’s those relationships which are
hard to measure which really make up the
social fabric of society
so I suspect that a lot of the
underlying cause of our problems is due
to that fraying social fabric loss of
trust in one another and only the most
visible of those consequences the most
measurable are things like the rising
deaths from despair
so in short economic policy is broken
along with many things like it that
focus on numbers quantifiable things
like GDP as proxies for hard to measure
things like human wellbeing and that may
have worked once but right now it’s no
longer working another lesson from the
natural world is that evolution has no
moral compass that is to say that which
replicates wins becomes more abundant by
definition regardless of the cost to the
greater system if you look out there in
the natural world more than 40% of all
species are thought to be parasites
ecosystems aren’t optimized for anything
and this is relevant to us because
capitalism amounts to a clever way of
harnessing evolution to satisfy consumer
preferences anyone who can offer better
goods and services that people want at
progressively lower prices is rewarded
with resources to keep on doing that and
you have a virtuous arms-race cycle to
satisfy consumer preferences more and
more effectively this is how we create
wealth and it works very well in some
industries consumer electronics clearly
but capitalism doesn’t seem to be so
efficient in others in the past 50 years
the cost of housing if you control for
size and quality has gone up by 50% the
cost of primary education has more than
doubled with no obvious improvement in
learning the cost of college and of
healthcare have both increased by more
than a factor of 10 again no really
obvious improvements and outputs here so
what’s going on well you can say housing
that’s an effective limited land supply
growing population especially near the
coast
and you can excuse the others by saying
well if you have increased wealth like
we do now health and education seemed
perfectly reasonable things to spend the
excess on but then why is the cost of
infrastructure going up by so much if
you look at New York’s most recent
subway line 2.2 billion dollars per mile
that’s ridiculous
it’s an outlier but it’s ridiculous and
it’s not alone the trend is there and
this is strange because we have much
better technologies for doing everything
required for infrastructure and for
healthcare and everything everything
from better computers to better concrete
mixers we would expect all of these to
be getting cheaper not more expensive
and it’s not going to wages or even
really to benefits despite health care
so what’s going on well again in the
laboratory if you run a directed
evolution experiment for a while it
might work great for a first few
generations but then eventually you’re
gonna throw up some cheaters or in
parasites that just feed off of the
others and as we know if you look at
nature the longer you let it go the
worse this gets so my suspicion is that
capitalism and possibly democracy too
have really just been infested by a
proliferation of evolutionary cheaters
people have figured out more and more
ways of cleverly siphoning additional
resources off and this breaks the system
slowly doesn’t completely break it but
it makes everything more expensive and
it increases inequality and in turn that
contributes directly to political
polarization and anger just because
we’re stuck with increasingly higher
bills for the exact same services we
used to get for much cheaper so who’s
gonna pay the extra it’s a lot harder to
reach agreement so at this point in the
talk you’re probably asking what can we
do well again laboratory how do you get
rid of cheaters we have a very easy way
we apply superheated steam and
30 atmospheres of pressure and they’re
all gone it works great the same sort of
thing seems to work for society if you
look at history that is the reliable way
to reduce inequality and remove economic
rent-seeking is war famine pestilence
and mass death so we really need some
other solutions here and by the way you
should be depressed now I mean come on
these are some of our most eco systems
that we’re proud of I’m supposed to be
working in efficient in clever ways of
harnessing evolution and they’re not
working so if you’re an optimist like me
you’re probably thinking well there’s
got to be some human cultural ecosystem
that’s working really well and I’m a
scientist so I naturally think well
science science is the crown jewel of
our civilization it can be frustrating
and inefficient and let me tell you from
the inside it’s very frustrating and
inefficient it’s very much a matter of
one step back for two steps forward but
it reliably moves us towards a more
accurate picture of the world and that’s
amazing the problem is if you’re a
civilization not all knowledge is worth
having because science is vision just as
technology is power and it’s the dance
of science and technology that is the
engine that drives our civilization
without them it falls and billions will
die with them
well I’m increasingly concerned that we
might inadvertently solve our
rent-seeking problem suppose there is a
discoverable technology out there
potentially in our future that if
discovered and publicized would bring
down civilization you can probably
imagine some things like that but the
really scary ones are those that none of
us in this room or anywhere else has
anticipated if that exists then the
current scientific ecosystem is highly
likely to stumble straight into it
without ever seeing the danger
the reason is that in an age of costless
communication we still do almost all
research including publicly funded
research behind closed doors where
everyone is blind to what everyone else
is doing and most research is done by
teams of similarly trained specialists
who can’t reliably anticipate the
consequences outside their own narrow
field of specialty which means if you
have something that it requires a
synergy of this and that neither team is
going to know that the other is working
on that or even know of the possibility
of that kind of synergistic interaction
this is still very theoretical so you’re
probably not convinced yet so let me
give you an example five years ago no
one imagined that we might be able to
engineer entire wild species but if you
take CRISPR genome editing and you
combine it with gene Drive which is a
natural phenomenon it’s a strategy
whereby genetic elements ensure that
they get inherited put those together
and that’s exactly the capability that
we now have
if you engineer an organism so that it
has the edited changes that we want in
its genome and we encode the editing
machinery next to it then when that
organism mates with a wild one the
offspring will inherit one of each and
in those offspring the editing machinery
will convert the original version to the
edited version it’s a find and replace
because the next generation is
guaranteed to inherit and the next and
the next and the next so four years ago
my colleagues and I first envisioned
this possibility and its potential
outcomes and we were initially elated
but then very concerned so we asked a
bunch of experts from diverse fields
from ecology to security to even
environmental NGOs to see what they
thought we should do and we concluded
that it’s not very physically dangerous
just because it’s slow it spreads over
generations and it’s easy to detect by
sequencing and once you detect it you
can counter it with another find and
replace but it’s very socially dangerous
just because well you can imagine the
head
if some scientist accidentally releases
something into the wild that will
eventually engineer that entire species
it would be pretty devastating to public
trust in science and as previously
mentioned we really need science so how
to stop that well we decided we would
publish our findings our assessment of
what this could do and a detailed list
of technical safeguards that should be
able to prevent that kind of accident
the next year the very same high profile
journal that published our warnings came
out with a report detailing the same
technology under a different name
without using any of the safeguards now
I have no explanation for the journal
but the researchers are brilliant
well-meaning scientists who hadn’t seen
our work weren’t even aware of the
concept of gene Drive because it’s
something for from evolutionary biology
which was not their field and they were
just focused on building a better tool
for laboratory genetics experiments that
could be used by others they never
imagined that it could be applied to
wild populations so there you have it a
powerful completely unanticipated
technology that despite our best efforts
was still independently developed and
publicized by researchers who were blind
to the consequences do we have any
barriers in place to prevent this from
happening for a more dangerous
technology I can’t think of many so at
this point the question can we engineer
social ecosystems is looking a little
bit more urgent that means I have a
confession for you when I say that I’m
an ecological engineer people laugh and
they left because we don’t know how to
do that it’s only because of tools like
CRISPR based gene Drive which let us
harness the strategies of cheaters to
alter populations and by those
populations the associated ecosystem
that’s the only way that we think we can
do it now and similar approaches so the
question is can we apply this to the
social setting what is the equivalent
and I suspect we can now I’m gonna focus
on science because I know the scientific
ecosystem best scientists to be
practicing scientists we need
recognition for what we do which leads
to continued funding so if you want us
to share our experimental plans in
advance with others so that society can
scrutinize them for potential dangers
and intervene early then you need to get
the scientific journals and the funders
to change the incentives now speaking
from experience now I can tell you that
none of them wants to be first it’s a
collective action problem so we need
some way to get the wagon moving without
them I think the answer is to use a
technology to harness it cheat in this
case the cheat is intellectual property
now patent trolls love to use
intellectual property to halt useful
products unless they get paid off it’s a
great way for them to siphon off
resources but if what you want to do is
change the incentives you want
scientists to do something differently
well how about you say you can’t do that
research unless you first publicly
disclose your plans this wouldn’t
normally work but we have an opportunity
here in the form of CRISPR based gene
Drive so here’s how it would work to get
together get together a patent pool of
relevant IP for building CRISPR based
gene Drive ignore commercial licensing
don’t touch that at all and just
regulate the research but offer anyone a
license for free provided that they
share their experimental plans in
advance this will be controversial
because it violates this long-standing
precedent which has very good reasons
behind it that patents should not
interfere with non-profit scientific
research in this case I think there is
very good reasons to make an exception
to that
and that controversy will lead to media
coverage yet it’s really hard to stand
up in public and assert that you have a
right to develop a technology that can
single-handedly engineer ecosystems in
secret without telling anyone what
you’re doing or allowing anyone to have
a look and make sure you’re using
safeguards so favorable media coverage
will then incentivize the funders and
journals to jump on the Nell moving
bandwagon and add their own incentives
if it can work for CRISPR based gene
Drive then you have a precedent and the
strategy might then be applied to
related fields involving technologies
with shared impacts and eventually to
most of the areas were concerned about
throughout the scientific ecosystem and
who knows if it works there perhaps it
could be applied to the other ecosystems
that we think may require a fix or –
let’s be fair this idea could fail it
could fail spectacularly in fact most
experiments in science do fail but
that’s why we would then have to take a
lesson from engineering and that is try
something else
until it works and we have to do that
because when we think that civilization
itself may be at risk the moral cost of
doing nothing is always too high thank
you
[Applause]
Please follow and like us: