Cause climate problems? Then take the lead on a fix
As Adam Vaughan writes, it is now clear that we have to reduce our greenhouse gas emissions to zero (27 April, p 20). But the question of when depends on who we mean by “we”.
The Intergovernmental Panel on Climate Change special report shows that global emissions should reach net zero by about 2050. But this doesn't mean that the US, UK and other developed nations should set 2050 as their target dates. These countries have contributed far more than their fair shares to climate change. The principle of global equity that is enshrined in the Paris Agreement on climate change requires those who have caused the problem to take the lead in solving it.
That means developed nations should reduce their emissions to net zero well before 2050. In the UK, only Extinction Rebellion and the Green Party – with target dates of 2025 and 2030 respectively – have faced up to this.
The origins of language may not be with hunters (1)
Watching people, in particular young children, concentrating on using their hands for a fine motor skill while sticking out their tongue lends credence to the possibility that language began with hand movements (4 May, p 34). I wonder whether speech developed as our ancestors began to live in groups, with some members going out to hunt and the rest making clothing and fashioning tools. These home-based activities could have loosened the tongue as the hands worked and enabled tongue anatomy to evolve, eventually leading to speech.
The origins of language may not be with hunters (2)
Since gathering, rather than hunting, provided the majority of calories for early communities, why is such a focus for the origins of language placed on suggestions of what may have aided our ancestors to hunt, rather than to gather? Some propose singing for protection as the origin of language. Maybe this started when the gatherers 鈥� probably mostly female 鈥� were foraging, rather than just for protection overnight.
Information would need to be transmitted about which plants were safe or which areas should be explored on a daily basis, not just for sporadic hunts.
How early humans could have reached Luzon island
If another human species has been discovered on the island of Luzon in the Philippines, how did it get there, and when? There were five or six periods of low sea level in the mid-to-late Pleistocene (13 April, p 9). In the two most recent of these, levels were about 120 metres below those of the present around 20,000 years ago, and 100 metres lower about 65,000 years ago.
This would allow someone to walk most of the way to Luzon from Malaysia or Flores via Borneo, Palawan and Busuanga. The main obstacle would have been the Mindoro Strait. This narrow, submarine canyon is much too deep to have ever been dry. But at this time, it would have been only about 65 kilometres wide. On a calm day, someone could have paddled across on a log.
I would put my money on Homo luzonensis being a close relative of Homo floresiensis, rather than having evolved as an isolated island species.
Welcome to our new and even more bizarre masters
You report that DeepMind's artificial intelligence was unable to add up seven 1s, and that it concluded that 17 × 4 is 69 (13 April, p 12). That is artificial stupidity.
Until now, artificial stupidity was the responsibility of the human programmers. Now, the machine is stupid on its own.
I predict that further advances in artificial intelligence will yield ever worse artificial stupidity, for intelligence and stupidity are linked. When the robots awake, they will make unusual mistakes.
Apply the precautionary principle to mobile safety (1)
Chris Stokel-Walker says there is little evidence that banning Chinese company Huawei from supplying 5G mobile technology is the right approach (4 May, p 11). But when deciding this, we shouldn’t be seeking evidence to test a hypothesis. We should use the precautionary principle: don’t do something if it might be disastrous. The question is not whether Huawei has mishandled data in the past, or whether it has done things for the Chinese state. It is whether the company can be trusted not to do so in the future.
Apply the precautionary principle to mobile safety (2)
What is the motivation of the US in trying to stop Huawei selling 5G infrastructure to the UK and others? There is a body of opinion that it is to get US companies to dominate the market, and to enforce the spread of US spyware, not Chinese. The US National Security Agency has with US-made devices that route data on all networks. The US would find it harder to put backdoors in kit from elsewhere.
Backdoors have not been found in Huawei kit. Its are , but so, unfortunately, are those of others who sell hardware.
As you say, sensitive data should be encrypted. But this still leaves communications open to traffic analysis 鈥� tracking who talks to whom, rather than what they talk about.
Please choose to read this to try to fathom free will (1)
Tom Stafford argues that the idea that free will doesn’t exist is based on misguided intuitions of what it means to be a biological machine (6 April, p 34). Is it fair to assume that any scientific model of the mechanism of a free will decision should ultimately be expressible in mathematical form?
I understand this to be the case for scientific models generally. Mathematics, however, appears only to have concepts of direct deterministic inference or of randomness, no blend of which fit the bill for modelling free will.
A scientific model of free will would appear to require the building of a model that takes identical inputs on each run, maintains no state between runs, uses no purely random sources and yet produces results that are not fully predictable.
Perhaps maths can’t contain such a model. But what would science be without maths? Or free will without science? This has been driving me up the wall for the past 25 years, since I got my masters in maths, specialising in quantum field theory, general relativity and all that mathematical physics goodness.
I am now training to be an Anglican priest, so all this free will stuff is appearing again in a slightly different context!
Please choose to read this to try to fathom free will (2)
Stafford confuses complexity with freedom. He admits that “our thoughts are caused by our brains, our environment and our history”. But he maintains that, because this causal mix is unique to each individual at each moment, human behaviour is so difficult to predict that it is elevated above that of digger wasps.
Yet so long as human behaviour is totally caused by any number of factors in the physical world, the notion of human free will, the intuitively real sense that we alone can make up our minds, remains illusory.
Living in a human society requires us to hold everyone responsible for their behaviour. But that is a practical necessity, not proof of individual freedom.
Hacking a car may not be an innocent foible
Chris Baraniuk concludes that it is one thing to hack an artificial intelligence controlling a vehicle in a car park, and quite another to hack an AI controlling a weapon (27 April, p 34).
But what about a weapon such as a driverless car that is doing 50 kilometres per hour on a city street, for example?
Single-gene mutations may be detectable by smell
Discussing whether cystic fibrosis carriers might recognise each other, Richard Harris says “it seems extremely unlikely that one mutated gene among tens of thousands would be detectable by smell” Letters, (20 April). But there are examples of just this.
For example, trimethylaminuria, also known as fish odour syndrome, is caused by a mutation in the FMO3 gene (6 December 1997, p 25). Body odour is affected by changes in the ABCC11 gene, which also controls the type of earwax you produce (23/30 December 2017, p 67).
Cystic fibrosis is caused by a mutation that results in, among other things, a build-up of thick mucus in the lungs and digestive system. It isn't inconceivable that lung infection caused by such a build-up could be detectable by odour.
Up-to-date magnetic units would be so much clearer
Alongside Daniel Cossins's interesting article on fast radio bursts, you describe the comparative magnetic field strengths of various bodies, but sadly use the antiquated non-SI unit, the gauss (11 May, p 34).
The tesla is the modern SI unit of magnetic flux density. You showed a comparison of the field strength of commercial magnets with neutron stars and magnetars.
These values would have been much clearer if expressed as 10 millitesla, 100 megatesla and 100 gigatesla respectively.
For the record 鈥� 25 May 2019
鈥� Wet, wet, wet: It is the class of space rocks that includes Itokawa that may have brought half as much water as is in Earth’s oceans (11 May, p 18).
鈥� Mind like a sieve: The reason that Harald Helfgott’s prime number-finding algorithm is purelytheoretical is that he hasn’t optimised it (11 May, p 11).
鈥� Quite crowded: There are 10 million bacteria per square centimetre of human skin, on average (13 April, p 28).
Want to get in touch?
Send letters to 快猫短视频, 25 Bedford Street, London WC2E 9ES; please see terms and email at