New_Scientist_3_08_2019

3 August 2019 | New Scientist | 23

F

ULL disclosure: I am a
hypocrite. Two weeks ago,
I wrote a New Scientist lead
article about the urgent need to
find alternatives to flying. Last
Friday, I boarded a plane from
London to Pisa, Italy, for a
scientific conference that, at first
glance, would almost certainly
fail the “necessary” test.
In my defence, one of
the themes of the conference,
organised by the Foundational
Questions Institute (FQXi), was
the extent to which intelligent
agents control their actions – and
I did offset my emissions. What’s
more, I left the convention feeling
that something quite important
JOSIE FORDwas stirring in the Tuscan hills.

Comment

Richard Webb is executive

editor at New Scientist

Views

The columnist

Eating locally isn’t

always greener, says

James Wong p24

Letters

The food advice

I needed came

from a song p26

Aperture

Arctic wildfires emit

record-breaking

carbon levels p28

Culture

Utopians and realists

agree the economy

is broken p30

Culture columnist

Helen Marshall

explores sci-fi slants

on modern times p32

At its heart is the mystery of

life: how atoms and molecules

come together to make stuff that

can self-sustain, make decisions,

influence and exploit its

environment, and be conscious.

Physics can’t explain that. A few

months ago, physicist Paul Davies,

who presented at the conference,

proposed in these pages that

better, information-based

physical theories might be the

key to resolving the conundrum

(2 February, p 28).

Maybe. But the problems go

deeper. Our most basic theory

of reality, quantum mechanics, is

beset by the problem of conscious

observation, which in some way

seems itself to interrupt reality

and perhaps determine its

evolution – with potentially

nonsensical consequences.

My brain overheats when

confronted with these problems,

and I was glad to learn at the FQXi

conference that I am not the only

one. But it is becoming clear that

we will only make progress in

fundamental physics and other

fields by working out where the

“I” is in reality.

Ideas are out there. On

page 34 of this issue, for example,

cognitive scientist Donald

Hoffman presents his own

evolutionary theory of how our

perceptions necessarily cloud our

ability to see what is truly what.

Conferences such as FQXi

provide an opportunity for

researchers from different

disciplines to compare notes.

What struck me observing them

was how much work there is to do

even just to define concepts such

as agency, free will, consciousness

and intelligence, and how they

interrelate. Maybe physicists can

help, though one early attempt –

the integrated information theory

of consciousness – has come

under fire for allowing some

pretty strange definitions of

what counts as conscious.

For the very small amount it

is worth, I am sceptical we will

ever develop a comprehensive and

objective theory of first-person,

subjective experience. The lesson

from other realms of enquiry, for

example Gödel’s incompleteness

theorems in maths, is that logical

systems blow up when trying to

explain themselves.

But I am glad that people are

trying. As we develop artificial

intelligences and ask ourselves

about their limits and dangers, the

pay-offs of a better understanding

of this relationship between

conscious experience and the

physical world could be huge.

It will take a different

application of agency and

intelligence to solve climate

change (I will take the train next

time). But figuring out how we

relate to the world around us on

a basic level is necessary, too.  ❚

The ‘I’ in reality

To make progress in understanding the world, we need to

grasp what par t we play in making it, says Richard Webb