QJM vol. 97 no. 12 © Association of Physicians 2004; all rights reserved.
Biologic |
Footnote
In a preface written in 1816, Coleridge blamed a man from Porlock for breaking his concentration in 1798 when he was composing Kubla Kahn. The editor of QJM must bear a similar, but inverted, burden of guilt—I had agreed with him that this series should run for 3 years exactly, or 36 columns. However, one contribution was not chased up by the editor, knowing that, like Coleridge, I was then under the influence of opiates. I had, in fact, proposed to finish Kubla Kahn, as a relief to you all from developmental biology, risk assessment, fish and the other odd issues that had caught my fancy in the last years. But now we will never know the details of what the Damsel did with the Dulcimer.
Those who have read this column regularly will realize that I am fascinated by the way in which organs have evolved, and how they form and grow. This has occupied much of my time, but in later years I have spent a lot of time in a parallel existence in medicine in drug, device, chemical and (more recently) GM crop regulation. I hope that the relevance of the former to the practise of medicine has become as apparent as the obvious need for the latter, but in both fields you have to deal with nonsense, from creationists to regulators who do not like to use the science base.
It is generally true that human beings look for patterns in almost anything they deal with on a regular basis. It is also true that this temptation sometimes leads to the imposition of patterns that do not exist onto reluctant data sets, and in particular circumstances, to the generation of bizarre hypotheses. Blake made much of the Golden Ratio, 
(1.618) in his startling illustrations, and others have certainly invoked it in the design of the Great Pyramid of Gaza. All nonsense, of course, (but see footnote on how to make phi).
(the capital form is at the start of the sentence—it doesn't mean anything different) has a startling relationship to the Fibonacci numbers. These, you remember, are made up thus: start with 1 and 1, and let each number be the sum of its predecessors. You get the sequence 1, 1, 2, 3, 5, 8, 13, 21, 34, 55 and by the 30th you are up to 832 040. Now, divide any number in the series by its predecessor, and as the number gets larger it tends to (55/34 = 1.6176). The scales on a pineapple are in spirals of increasing steepness, but all the pineapples you will ever see (as well as all those you won't—I have grandchildren and they like things Just So) have 5, 8, 13, or 21 spirals. It's the same for the kernels of a sunflower, whether clockwise or anticlockwise running.
So how do you prevent anyone making something of these types of data? That particular growth patterns might depend on a non-exponential but regular pattern of replication of numbers of cells will not seem remarkable to many (stem cell populations have many patterns of release of descendants), but explaining why some things are nonsense is not always easy. Are there any rules to differentiate between false argument and rational speculation, whether described in pointillist detail (the explanations of the Turin shroud) or broad-brush generalizations (Stonehenge was built by spacemen)? You realise that in this context I write in Seurat rather than in Ingres.
The first thing to say is that science moves on. The fact that evolution has occurred is supported by evidence from many sources, from Darwinian observation, the fossil record, from the persistence of successful adaptations in parallel structures and from molecular biology; in the latter field from a myriad of examples, some of which have been explored in this column. It has also been observed to be happening now. Those who suppose that modern geological features were created by Noah's flood have no new data. Of course, there are novel and brilliant ideas, but things mostly go with the weight of evidence.
Second, in any argument about a scientific point the protagonists go out of their way to disprove claims and are (mostly) glad to change their minds if new data give a better solution. Pons and Fleischman were foolish to announce cold fusion at a press conference, but were wrong to continue with their claim when others could not reproduce their results. A major failing of those who support bizarre ideas is that they do not seek confirmatory evidence by another route or routes, but rather discount data that do not fit their conception of events.
There are rules about research. It always enraged me, when a council member of the Royal College of Pathologists, that otherwise sensible people, talking about the need for 5 years training in diagnostic pathology, would suggest that a year of research somehow conferred ability in that field. It takes time to learn to do research (especially clinical research) and a real grounding is necessary—some recent catastrophes, including MMR, show this. Shermer uses the example of the search for intelligent extra-terrestrial life; a scientist will start from the null hypothesis that there is none and that evidence must be found before this large claim is made. UFO fans start the other way round and look for anything which might support the idea, from recollections of abductions by aliens (recalled under hypnosis) anecdotes of UFO sightings, conspiratorial denials involving government cover-ups (I was once asked to comment on a film of an autopsy of an alien—it's OK, it wasn't one) and poor quality photographs with supporting explanations of the reason for the problem (‘he moved’).
There is also a tendency to simply deny the current scientific position without producing a new explanation or to fail to supply an adequate refutation of the things that the current hypothesis explains well. If you do not believe that HIV causes AIDS, you have to have an explanation of what has happened to many haemophiliacs. It is disregard of the obvious that seems to me to come easy to many of the daffiest people I have met; it is failure to consider non-supportive data that has impressed me least when considering grant applications (in non-supportive evidence often lies supportive information).
Finally, there is the problem of faith. By this I do not refer to the philosophical systems we chose to support our existence but rather the social, political and ideological biases we all have that may influence our view of data. It seems to me that this problem is generally well dealt with by peer review, but in some fields at some times a damaging consensus may evolve. This prevents advance and results in misuse of resources—in train collision warning systems, the use of nuclear energy, the structure of the health service in the UK, in damaging attempts to ‘repair’ the environment and in the use of the precautionary principle, as examples. In 1988, 16 years ago, a leader in the Sunday Telegraph considering the danger of the politicization of environmental issues, commented on shoddy research, unscrupulous use of evidence to attract publicity and funding and the announcement of impending disasters which fail to materialize. As a leader in the New Scientist pointed out around the same time ‘politicians and overzealous consumerists should not get away with selling policies on the strengths of dubious science’.
Vale.
How to make a phi ()
Take a line AB and divide it by a point C such that the whole segment is to the large part as the large part is to the small: AB/AC = AC/CB. If we assign to the segment CB a unit length (CB = 1) and denote the length of AC by x, then
 |
 |
 |
, an irrational number, a string of digits that never ends or repeats. It bears the same kind of relationship to the regular pentagon that bears to the circle (it is the ratio of the diagonal to the side of that figure).
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||