For the record
鈥� We quoted Steen Stender as saying that Denmark has seen a 20 per cent fall in deaths from heart disease since the 2003 ban on trans-fats (6 January, p 13). He clarifies that there was a 20 per cent fall between 2000 and 2005.
Free will and evolution
In his discussion of free will, John Searle makes a classic error in attributing purpose to evolution (13 January, p 48). He argues that if we live in an entirely deterministic world, why would evolution have endowed us with expensive features such as big brains, since “all decisions are fixed by deterministic neuron processes”, and thus having the big brain “makes no difference”.
Complexity can evolve even in deterministic systems, as is proven by genetic programming techniques. If bigger brains make a difference to a particular life form’s ability to pass on its genes, then they can evolve. Even though the creature could not have made any other decisions than it did, it could make better decisions, using all the power of the complex network it possesses.
Evolution is blind, and cares not that this might produce the cruel illusion of free will for very complex organisms. In such a system, it only makes no absolute difference to a big-brained philosopher who sees the big picture.
From Eric Norton
John Searle maintains that the hypothesis that free will is an illusion “runs dead counter to evolution”, since without free will “conscious rational decision-making… plays no role in the survival of the organism” and therefore we could not have evolved our big brains.
But all animals respond to external stimuli in their environments, and big brains allow their possessors to assimilate more information about their environment and model it mentally in more sophisticated ways. This in turn allows them to respond more appropriately to their environment, which would certainly be an evolutionary advantage.
In any case, it’s not as if we don’t have the ability to choose, it’s just that with enough knowledge of us and the information we are working on, it would be possible to predict the choices we will make. That’s hardly revolutionary 鈥� our friends and loved ones have a pretty good chance of predicting how we’ll choose to take our tea, for example.
Aylesbury, Buckinghamshire, UK
From Peter Smith
John Searle says that without free will “the enormous time spent educating the young to make rational decisions makes no difference”. But the learning that I have experienced is surely among the causes of the actions that I take, even if those actions are not a matter of free will.
Even in a strictly deterministic view, I cannot take an action that is inconsistent with my state of knowledge of the world. In apparently making a free choice to look before crossing the road, I may in fact be blindly following a learned rule but, either way, without the relevant education I would be swiftly removed from the gene pool.
Leicester, UK
Manipulating placebos
The interview with psychiatrist Patrick Lemoine highlighted some intriguing features of the placebo effect, notably the impact of the doctor-patient relationship (16 December 2006, p 42). It would be interesting to know more about the effect of this on drugs trials.
Complementary therapies well-regarded by the public notoriously fare badly in clinical trials. The usual explanation for this is that the public is deluding itself.
It is often overlooked, however, that in such trials “efficacy” is measured by the proportion of patients receiving the therapy who recover, relative to the proportion who did so after receiving a placebo: the bigger the ratio, the more effective the therapy is deemed to be. Thus a complementary therapy that cures just as high a proportion of patients as a conventional therapy can appear “useless” simply because the trial also produced a relatively high placebo response.
The corollary is also true: that a relatively useless conventional drug can be made to look very impressive simply by minimising the placebo response rate. Perhaps drugs companies should spend less time on quantum chemistry, and more on how to create thoroughly unsympathetic doctor-patient relationships.
Gamuts of colours
We read about lizards that “can see all the colours we can see plus ultraviolet” (6 January, p 36). There is a lot of beauty in the colours we see, but the article vastly understates the information available in the spectrum of light if only we could see it. Taking an average digital camera as example, the addition of an extra primary colour to the normal three might seem like an increase in performance by a third, but actually the number of colours resolvable by the camera would multiply by 256 (given the standard colour resolution of a “channel”).
In the field of spectroscopic imaging we work with cameras having hundreds of primary colours, often including wavelengths invisible to the eye. With the aid of computers, surprisingly rich information can be extracted from the images. I think the perception that we can see all colours worth seeing leads us to underrate the capabilities of a tetrachromatic animal. The same perception may be the most important obstacle to a more widespread use of readily available spectral imaging technology in many potential applications.
Gas garbler
George Monbiot does his argument no favour by trying to slate the Toyota Prius using a quadruply invalid comparison with the most frugal of the range of Peugeot 205s (23/30 December 2006, p 25). The 205 was a “supermini” and the Prius is a saloon. He compares just highway fuel consumptions, ignoring the fact that the Prius, unlike other cars, maintains its low consumption in urban driving. The 205 was a diesel and the Prius runs on petrol, which produces about 13 per cent less CO2 per unit volume of fuel. Worst of all, Monbiot compares the Prius’s consumption in miles per US gallon with the 205’s in miles per imperial gallon.
The editor writes:
鈥� The piece did confuse US and imperial gallons. Sorry. The Prius’s highway consumption of 51 miles per US gallon is 61 miles per imperial gallon; and lists it as achieving 60 mpUSg (72 mpg) in city driving.
Diminishing deities
I believe “good science will come” from the Biologic Institute’s efforts to find experimental proof of intelligent design, but not in the way that it intends or believes (13 January, p 18).
Starting with a premise or hypothesis, one devises experiments that attempt to negate the hypothesis or premise, demonstrating that it does not hold true for all cases, maybe not even for any. Starting with a bad premise simply means that it will take longer to come to the realisation that it is a false start; the premise must therefore be refined or revised and the process begun again. Thus they are inevitably destined to lose their theistic beliefs.
Imagine replacing all instances of “I don’t know the answer yet” with “God did it”, and continuing to labour to uncover the facts. Over time, we will inevitably chip away at the mountain of “God did it” assertions 鈥� we have been quietly doing so for centuries 鈥� but now we will be able to publicly state that what was once thought to be an act of God is now in fact (fill in your own discovery here).
Thus we will gather tangible and mounting evidence of the continued erosion of God’s claimed ability. At present, we make no formal note of the fact. I wonder how many believers will be willing to face daily despondency as yet another cherished “God did it” claim perishes before the unstoppable juggernaut of the search for truth.
Battery of possibilities
The vanadium flow battery has its attractions for electrical energy storage, but it is certainly not the only system that is practical (13 January, p 39). A quite large nickel-cadmium installation with a capacity of about 7 megawatt-hours is still, as far as I am aware, working in Alaska.
There are other technologies, too. Japan has over 100 large sodium-sulphur systems, with capacities up to 70 megawatt-hours, supporting its electricity grid. There are now three such installations in the north-east US. There is also a pilot ZEBRA (sodium-nickel-chloride) installation of 100 kilowatt-hours in Ontario, Canada.
All these systems have their pluses and minuses, but all will be needed to help the ageing and weary electric grid systems on every continent.
Wrong kind of kiss
I have just read the very enjoyable “Mind the mistletoe” (23 December 2006, p 42). I was surprised, however, to read in a sidebar that “The ‘Malay kiss’, which consists of nose rubbing and sniffing is the ‘French kiss’ among… Maoris”. The hongi is a common Maori greeting, but consists of a nose-press, and sometimes an exchange of breath, rather than nose rubbing and sniffing. It is a formal greeting, not an erotic “kiss” as comparison to a “French kiss” might suggest. And the plural of “Maori” is “Maori”.
Belief or not
Roy Sablosky writes to claim that atheism is not itself a belief (13 January, p 19).
But, to quote the Oxford English Dictionary, an atheist is “one who denies or disbelieves the existence of a God”.
The correct term for someone who does not believe in the existence of a god, but does not assert the non-existence of a god, is “agnostic”.
From Dave Minter
Roy Sablosky’s letter reminds me of a signature line that I saw on a contribution to an online bulletin board.
It read: “If atheism is a faith, then not playing chess is a hobby.”
London, UK
…and complexity…
John Searle’s argument on free will fails to see the elephant in the room. Seeking indeterminism, he looks for it only at the quantum level. This assertion grossly neglects the findings of chaos theory and complexity theory.
These have demonstrated, over and over again, that life is full of complex systems that have deterministic rules at the lowest levels, but whose large-scale conditions are intractable. Their future states are indeterministic, and a small variation in initial conditions causes a huge difference in the outcome.
Thus, a macro-scale system such as our brains can 鈥� and most probably will 鈥� have indeterministic states, even though deterministic rules govern its lower constituents. There is plenty of room for free will in such a system.
…and machines
Let us hypothesise that we can build an intelligent autonomous machine 鈥� and take an engineering view of the task. In order for the machine to interact with the complex external world we need to provide it with a model of the world using the finite processing power we have available. Newtonian dynamics can do a pretty good job of explaining why things happen in this world, but we need to allocate computing resources wisely, so practically we have little choice but to build a model that approximates to what happens.
We will also include a simplified model of the machine itself, so that it can predict which actions would be good or bad for itself and act appropriately. Our machine can then accumulate experience of past events and learn to make the best decisions using its internal models.
The states at any one time of the internal models of the world and of itself are available and so may be communicated to other intelligent machines in order to learn more efficiently. Should our machine use English to communicate it may well use the word “conscious” to describe the ability to model itself, and “free will” to describe the fact that actions taken are in agreement with what the internal model predicts it should do in its own best interests. That is, I think I have free will because I do what I think I should do.
Free will…
It is logically inescapable that free will, as usually defined, is an illusion; yet John Searle thinks it is odd that evolution would produce this illusion when it has no survival value (13 January, p 48).
The ability to carry out “what if” mental simulations is clearly an advantage to the higher animals. But this creates a design problem: two sets of mental processes need to co-exist. One attends to the here and now, and the other periodically roams around a simulated mental world.
The sensations involved in sight and sound are therefore needed to avoid confusion between real and simulated experiences. Hunger and pain are similarly needed to focus mental energy on real problems rather than being dissipated in unnecessary simulations.
The feeling of free will when a simulation process produces a decision is real enough, but the simulation is determined 鈥� as is everything else in our lives.
It may seem a mystery that the ability to carry out parallel mental simulations leads to actual awareness and sensations, but how else would mental processes be represented in a simulation of mental processes?
From David Fremlin
Free will is an experience, one which nearly all of us share. It is so important that absence of the experience is not only intensely distressing but is taken as a symptom of mental ill health. One has to suppose that the experience, like hunger, has a neurobiological counterpart; for all I know, there are identifiable neurons which are active when I feel that I am exercising my free will.
I do not know whether dung beetles have synaptic processes which can be called an experience of free will, though any sympathetic observation of these extraordinary animals must suggest the idea. Why should there not be beings as intellectually superior to us as we are to dung beetles?
Colchester, Essex, UK
From Brian Adams
In discussing the paradox of free will John Searle gives an example based on a choice of two items on a menu in a restaurant. Perhaps the answer is that both choices are made and the “consciousness” making the choice bifurcates into two closely parallel universes. This gives the illusion of choice but supports the hypothesis that free will is an illusion.
Liss, Hampshire, UK
No wonder drug
It is indeed scandalous that promising anti-cancer agents such as dichloroacetate (DCA) go begging for support simply because they are cheap and unpatentable (20 January, p 3 and p 13). You have done a great service in bringing this information and perspective before the public.
However, after you published online your first article on this proposed anti-cancer treatment (17 January), my medical information service was deluged with demands from desperate patients for what you call a “too good to be true” wonder drug. We had to inform them that DCA had never been tested in humans, only in cell lines and experimental animals, and that it was totally unavailable to today’s patients.
You did explain that it is too early to draw therapeutic conclusions, despite the promising lab work. But the magazine headline “Cheap, safe drug kills most cancers” implies that DCA is known to destroy actual tumours in humans. This continues to generate waves of unwarranted expectation among many patients and has already resulted in severe disappointment for people seeking a solution to life-threatening cancers.
It should also be pointed out that DCA is a by-product of the water chlorination process and a well-known environmental pollutant. It has been shown to be carcinogenic in rodent models and is also genotoxic, hepatotoxic and teratogenic in animals, all at doses well below what would seemingly be necessary to achieve a therapeutic effect in cancer patients. There are worthwhile anti-cancer drugs that are carcinogenic. But it would have been good to inform readers of this.
Design needs rules
Ed Douglas’s thoughtful article was as revealing in what it didn’t say about improving design for sustainability as in what it did (6 January, p 31). What seemed to be missing in the enthusiasm for changing the world by design was a recognition that this will get the support only of “virtuous” manufacturers, not the ExxonMobils of his example.
In describing Japan’s electronic waste recycling scheme, in which the cost of recycling goods must by law be incorporated into the retail price, Douglas also gave us an excellent illustration of how regulation is needed 鈥� and how state regulation actually creates profitable new industries. The Anglo Saxon world tends to see all regulation as a cost burden. Until it wakes up to the new realities, residents of the UK will continue to produce enough garbage to fill London’s Albert Hall every 2 hours and 鈥� shockingly 鈥� 99 per cent of all materials used in the US economy will still finish up on the scrap heap within six months.
Getting better at design, though valuable, is not enough. Enterprising new designers need to lobby their governments for more and better regulation, and so do the rest of us.