New Scientist - 19.10.2019

26 | New Scientist | 19 October 2019

Views You r le t te r s

Editor’s pick

We are still aware of some

doubts on consciousness

21 September, p 34

From Dave Hulme,

Stockport, Greater Manchester, UK

Michael Graziano says we might be

able to begin to develop artificial

visual consciousness with existing

technology, but that it will take a

lot longer to build a machine with

a stream of consciousness. I have

a sneaking suspicion that we will

develop an artificial human-like

brain before we crack consciousness.

We could then task it with cracking

the problem for us, and hope we

understand the answer.

From Guy Inchbald, Upton upon

Severn, Worcestershire, UK

The idea of an “attention schema”

as a “self-reflecting mirror” that is

the brain’s representation of how

the brain represents things, so

that consciousness isn’t so much

an illusion as a self-caricature, is

beguiling and probably correct. But

Graziano is mistaken in suggesting

that he knows how to solve “the

hard problem” of consciousness.

The objective existence of a dynamic

self-caricature is one thing; the

subjective experience of that

caricature is quite another.

The integrated information

theory that Graziano mentions is

honest about the divide. What his

team has actually done is bring the

hard problem into sharper focus.

From Ben Haller, Ithaca, New York, US

I welcome Graziano’s statement

at the beginning of his article that

“instead of trying to grapple with

the hard problem” of consciousness,

he takes “a more down-to-earth

approach”. But he then discusses an

attention schema and sensory and

verbal capabilities. All he is really

entitled to claim is that a machine

having these would be attentive,

not that it would be conscious.

From John Theophilus,

Milkwall, Gloucestershire, UK

Graziano’s article on consciousness

was enlightening, especially the

proposal to build a machine that

reflects the author’s consciousness

model and see whether it displays

conscious behaviour. My question

is: how would you know whether

it was or wasn’t conscious?

We have difficulty determining

whether a human in a coma or with

locked-in syndrome is conscious.

We do have personal experience

of what it is like to be a conscious

human, but we have no idea what it

is like to be a bat. How could we tell

what a machine was experiencing?

And if we did think it was conscious,

would it be moral to switch it off?

Look on our works, ye

mighty, and despair

Almost the Last Word,

14 September

From Ben Walsh, London, UK

I was interested in the various

responses to the question of

how long any traces of human

civilisation would last beyond our

species’ sudden and catastrophic

demise. If extraterrestrials were

heading to Earth, wouldn’t they

encounter our array of orbital

satellites before getting anywhere

near the surface? If so, then surely

this raises a new question: how

long would the evidence of our

technological achievements,

currently found in near-Earth

orbit, last?

From Judith Hanna, London, UK

Hillary Shaw suggests that aliens

might find “fossil tunnels” in

Earth’s crust and that some of this

persists from 4.4 billion years ago.

The geology that is that ancient

isn’t intact crust, but three sub-

millimetre zircon crystals found

embedded in 3.3 billion-year-old

sediments in the Jack Hills in

Western Australia. So I doubt that

tunnels would be preserved that

far into the future. Geochemical

traces of our air and water

pollution might, however, survive

within similarly tiny crystals.

Don’t expect more than

we are prepared to give

28 September, p 12

From Judith Graham,

Millbrook, Ontario, Canada

Brazil is burning its forests to

make room for farms. Canada

and the UK have already cleared

vast tracts of forest for the same

purpose, which is why the

situation in Brazil is so critical

globally. Instead of vilifying

developing nations, we should

either pay them to maintain their

existing ecosystems or rewild our

own lands. Expecting more than

we are prepared to give doesn’t

improve the global situation.

Pint-sized data firms may

be perfect ransom victims

13 July, p 9

From Robert Willis, Nanaimo,

British Columbia, Canada

Chris Stokel-Walker’s overview

of the increasing threat of

ransomware missed a major

cause of the ramp-up in attacks.

Small and medium firms, at least

in the US, are outsourcing data

operations to small outfits.

The likes of small medical

or dental offices may forgo the

major cloud storage companies

due to the cost of their services.

This has opened up a sub-market

for smaller firms to offer cheaper

off-site backup and storage.

Some of these service providers

lack the skills, software or technical

knowledge to ensure the security

of their clients’ data. They are

perfect ransomware victims. Such

companies may pay a ransom to