Universal guilt
David Papineau suggests one consequence of the branching universe is that you should feel guilty about “the passengers your other self has killed” in a car crash that did not quite happen in the actual world, but might have done (22 September, p 7). This cannot be right.
For me to feel guilty about the crash that occurred, I have to be the person in the other branch where that is what happened. But I am not, the crash didn’t occur, and “I” am the lucky driver in this branch. I may feel guilty about driving in a manner that might at some time cause death and destruction, and resolve to be more careful in future, but I can’t split my identity in such a manner as to take credit (or blame) for events that happen in other branches, because it isn’t me that participates in those events.
Probably (if that is the appropriate word) the criterion of personal identity needs to be couched in terms of looking back from the latest node in the branching structure. It is a sum only over such histories as lead through here (but not to other present and future branches that spring from points in our common pasts).
Atheism is a belief…
There has been a subtle twist in the threads of letters here about religion and god(s). It’s time to point out that atheism is not a denial of something. It is a belief: a belief that religion per se is utter (noun substitution) swill. As the science fiction author Robert A. Heinlein said, “If you can’t express it in numbers it’s opinion, not science; ‘faith’ strikes me as intellectual laziness.” A far more lucid discussion is available at .
Under pressure
In the article “Noisy neighbours” (11 August, p 28) Hazel Muir says that the average distance between molecules in the atmosphere on Mars is about 120 times greater than on Earth.
I accept that the mean free path in the Martian atmosphere is about 120 times longer than in Earth’s. However, there is no way that the mean free path can be translated or explained as being the “distance between molecules”; it is the average distance a molecule travels between collisions.
The ratio of average distance between molecules is roughly the cube root of the density ratio of 120 to 1.
Who boldly goes?
If people were serious about colonising space or the planets (Letters, 29 September, p 22) they would not send couples, but a party of healthy young women and a sperm bank. Every child born would have a different father, thus minimising the dangerous genetic bottleneck otherwise caused by a small founder population.
Back to the source
You replied to Marc Tanenbaum that “Religion is a feature of the natural world, and as fit a subject of inquiry for science 鈥� and for New 杏吧原创 鈥� as any other” (22 September, p 24). So you could presumably squash the doubters once and for all by interviewing God. It should be no problem for an omnipresent deity and a magazine of your reputation to arrange such a meeting.
For the record
鈥� Feedback misnamed Sara Batts (6 October). Sorry.
Proper treatment
In past weeks you have printed articles on the side effects of neurological drugs; on those of untested drug combinations; on the inadequacies of the US Food and Drug Administration (FDA) guidelines; on dichloroacetate (DCA), an untested substance that some cancer patients wish to take; and letters from such cancer patients desperate to hold their diseases in check.
These articles define the problems of medications. Developing drugs, especially cancer drugs, is hard. So how do we ethically decide and regulate what is safe, and what is efficacious? How to square the circle, and get such medicines to people who will benefit from them?
First, the drugs should be as safe as possible. Of course we must “first do no harm”, yet, as Paracelsus remarked half a millennium ago: “the dose decides if something is poisonous”. In short, everything is toxic, and Millie Kieve reminds us (15 September, p 24) that only after many people take a drug in the real world do some side effects become noticeable.
The feature “Toxic cocktails” notes that drug mixtures may be unpredictably disastrous in a some people (l September, p 44). Expect and insist as we may that drug companies make all test data accessible, this risk will not disappear no matter how “adequate and extensive” FDA guidelines become. We will still not be able address all combinations of circumstances (genetic, environmental, behavioural, ability of the therapist) that define future patient populations.
For efficacy, we need credible and statistically relevant clinical trials. Take cancer: people vary, and cancer varies too. So just finding appropriately uniform groups of sufferers to participate in a trial is a challenge.
The drugs must show a statistically valid effect, and acceptable toxicity, before the FDA (and EMEA, the European Medicines Agency) allow them to be sold. “Hope” is good for the immune system, and so bad for cancer; but snake-oil salesmen prey on patients who are so desperate that they are prepared to take untested substances. Classical “therapies” may be known not to have a great beneficial effect, but we can only say that because they have been tested.
So why not mix together some untested stuff 鈥� including, if you will, DCA 鈥� and “put together the best of the bunch” as John Davidson suggests (8 September, p 26)? Because that way we will never know if the treatment is better or worse than the disease. Davidson may be a science graduate, but he slept through a lot of the good bits: “I bet that… the survival advantage… would be better than the… standard chemotherapy,” is a bet he will lose, because he will not compare survival advantage with and without his mixture. When he does we shall know. The rest is silence.
Not-so-super bug
James Kingsland suggests that Clostridium difficile is “resistant to all but two antibiotics” and likens it to MRSA (29 September, p 37). In fact, antibiotic resistance remains fairly rare with this organism, unlike with MRSA. Resistance to the two antibiotics commonly used (metronidazole and vancomycin) is rare, and other antibiotics such as rifampicin can also be used. The problem in hospital patients is repeated relapses rather than true resistance: patients may respond if the same antibiotic is given a second time.
C. difficile is troublesome not because we are running out of effective antibiotics but because many hospital in-patients now acquire it when given antibiotic courses for other infections. The development of newer antibiotics will therefore not help to stem C. difficile; in fact newer, broad-spectrum antibiotics may encourage its spread. C. difficile certainly continues to prove very challenging for both patients and clinicians, but in quite different ways to MRSA.
Ecology and diversity
You argue that “we need to do more than just put a price on nature’s head” (15 September, p 3) and that we could “use market forces to safeguard ecosystems” (p 6). These are not new ideas, and perhaps we need to be far more inventive.
One of the problems is the frequent use of the term “biodiversity” in the limited sense of numbers of species. You go a little beyond this to recognise ecosystem and species interactions, but not far enough.
The operative word is “diversity” in nature, and that is what we should be defending. The conservation of diversity and its sustainable and equitable use is the basis of the 1992 Convention on Biological Diversity. Its definition includes “variability among living organisms and the ecological complexes of which they are part”.
Diversity at all levels of biological, taxonomic and ecological organisation is the fundamental basis for human health and sustainability. For example, soil quality depends on diversity of soil communities; agriculture and horticulture depend on genetic diversity. Losses in nature occur because human impacts decrease diversity at all levels.
Perhaps if we did justice to the convention, more people would understand the fundamental importance of diversity in nature, and perhaps nature conservation would then be recognised as a necessity, not a luxury.
Kiss miss
You report a study that surveyed college students about kissing and found that many people stopped being attracted to someone after the first kiss (8 September, p 22). You then speculated that this was because they had swapped saliva and found immune system incompatibilities.
I would fail my students if they measured one thing (not liking someone after kissing them) and concluded without further evidence that it was because of another thing (genetics). At no point did the study actually study saliva and gene swapping.
One might lose interest in someone simply because they were a bad kisser, not because of genetic incompatibility. The study also failed to take into account all the cultural factors associated with kissing.
For a cogent overview of the sloppiness of many studies in evolutionary psychology, Anne Innis Dagg’s is worth a read.
Managing beavers
Jane Karthaus states that “The beaver is a protected species and cannot therefore be shot or lethally trapped” (22 September, p 25). This is not true. Throughout the countries of the European Union, member states have been able to make arrangements to suit their traditions and needs.
In Latvia, as in Sweden, the beaver is regarded as a game animal. There are prescribed seasons when the animal may be trapped or shot, and quotas are set by the relevant authority.
In Bavaria, Germany, managers employed by a non-governmental organisation funded by the state respond to complaints from the public as they arise. If, for example, beavers are consuming farmers’ crops or impeding field drainage, a live trap is set and the captured animal removed. If no new home can be found for the beaver it is destroyed.
In France, the beaver is totally protected. There, people who complain are given advice on how to mitigate damage, and compensation is paid where justified.
Karthaus states that landowners and farmers are “the primary stakeholders”. This is so, but as the UK foot and mouth epidemic of 2001 showed, tourism is much more important to the UK’s economy than either agriculture or forestry. As the owner of a holiday cottage I am increasingly aware that visitors come to stay because there are beavers here at Bamff.
Landowners’ and farmers’ organisations have resisted the return of the native beaver to Scotland largely because they apparently share Karthaus’s misapprehension that the animal is totally protected and cannot be managed at all.
Now you see it…
It seems Graham Lawton’s contention that the brain “blanks out” saccades (22 September, p 35) is not entirely true.
At this year’s SIGGRAPH conference in San Diego there was a demonstration of a . This was simply a vertical bar of LEDs flashing a pattern corresponding to a succession of vertical scans of an image. Turning one’s head, or even deliberately moving one’s eyes, revealed nothing, but if a saccade occurred in sync with the flashing of the bar, a spooky image 鈥� a photographed face, perhaps, or a line drawing 鈥� could momentarily be seen.
Interestingly, there seemed to be little distortion of the image when it appeared, indicating that saccades have a pretty constant sweep rate. The demonstrator told me that they had done considerable work with eye-monitoring electrodes to decide on the display timing.
I am surprised you omitted the “two noses” illusion, which is about the easiest of the lot to demonstrate. Simply cross your index and middle finger and then run them up and down the bridge of your nose. The illusion that you have two noses is very hard to dispel.
Looking at the chimeric faces you printed (p 38) I discover that, unlike most people, I find the top picture much happier and warmer looking. The bottom picture looked like a false smile to me. Friends and family all thought the bottom picture happier, as expected. Is there something a bit odd about those of us who see it the other way round?
I tried the “half smile” experiment half expecting to get opposite results to usual on that too, but didn’t.
Unfair to atheists
Amanda Gefter’s review of John Cornwell’s reply to Richard Dawkins’s The God Delusion simplifies the position of today’s prominent atheists to the point where it is turned into a caricature of what they actually profess (22 September, p 53).
The position of Dawkins and others is not that science is “ultimately capable of explaining everything about the universe”.
It is that science is the only method so far devised with the ability to explain the explainable.
When religious thinkers try to accurately describe our world they end up with either the neutered and redundant definition of religion 鈥� as art or ethos or pantheism 鈥� that Gefter’s review mentions; or an incorrect description of our world 鈥� for example, any based on an interventionist god. Dawkins has gone to great lengths to explain just what sort of religion he rails against. Unfortunately, religious leaders are unwilling or unable to be so precise. They have no real-world basis on which to formulate such a definition, and the result is an ever-shifting definition of “religious experience”.
Gefter has fallen into the trap set by theists who claim that a belief in science is a form of faith. It is not. Science, properly applied, is a self-correcting mechanism for seeing the world as accurately as possible.
Religion, regardless of its outer form, is an attempt to inject an unmeasurable force to explain what some people are uncomfortable not being able to know.
The ability to accept that some things are unexplained is a demarcation line between scientific and religious thought. Its lack is what causes religiously minded people of all types to invent a heavenly host of ad hoc explanations for the natural workings of our world, rather than perform the heavy lifting required to tease out supportable answers from an often uncooperative cosmos.
Data at your fingertip
“Imagine being free to forget all of your passwords and use your fingerprint to log in to your online bank, eBay and email accounts” (22 September, p 30). Imagine injuring your finger.
Even if the system can cope with scar tissue, you will find yourself locked out of your computer and online life until the bandages come off. And what about severe burns that destroy the fingerprint permanently?
I’d rather stick with passwords.
You write that “encrypting the fingerprint using conventional cryptography and then transmitting it is not an option…” This could be misunderstood. In fact, the only convincingly secure solution at present is based on extended use of conventional cryptography.
This includes public key cryptography, authentication and protocols such as secure time-stamping and non-repudiation 鈥� a process which protects against a communicating entity denying that it participated in the exchange of information. In such a solution the fingerprint template is protected in the same way as secure transactions between financial institutions. These are expensive, state-of-the-art solutions affordable only for large-scale implementations by large organisations, but they are secure.
For now, there seems to be no easy way to design a sufficiently secure scheme that treats biometrics like passwords.
The attack mentioned in your article revealed a simple way to bypass one proposed solution 鈥� the fuzzy vault. Considerable research is going on and the future may show us some useful alternatives.
Universal guilt
Like all ideas that require quantum logic, the logic of many worlds runs afoul of itself. Suppose we accept the premise of many worlds that every possible quantum contingency resolves into every possible quantum outcome such that each unique outcome generates its own world.
Thus, as David Papineau suggests, every game is both won and lost by every team involved. Surely, then, the quantum physics team cannot be on the winning side of their game in more than half of all worlds. How do our would-be quantum heroes pretend to know which world we are in? In broad strokes, by the logic of quantum theory, quantum theory both is and is not valid. Whoops!