October 22, 2017

Mellanby versus theory and fact

Pages:  1   2   ALL 

This letter to Kenneth Mellanby follows the earlier exchange, “What makes Kenny run?”.

It was published in The Ecologist Vol. 8 No. 5, September–October 1978.


Edward Goldsmith replies:

I must thank you for taking the trouble to answer my criticisms, nevertheless more criticisms must follow for I do not find your arguments at all convincing.

You say that you have done research yourself and directed further research on the subject of energy and agriculture, pesticides, hedgerows etc, and that the results have been unexpected and, by implication, not in keeping with what one would have expected on the basis of what you call “dogmas” and I refer to as “basic ecological principles”.

First of all, I am not at all sure that it is by means of “research”, as it is currently understood, that the accuracy of many of your statements can be verified. What sort of research, for instance, is going to tell you that the Australians should not set aside large areas of wilderness unless, as you told them “they have first established that there is a demand for it”? This statement merely expresses your belief in the virtues of the market economy to which, you imply, any ecological considerations should be subordinated – a belief that no amount of scientific research is likely to dispel.

What sort of research can justify your view that burning straw and spraying Paraquat over our crops are justified because they save labour? Again these statements surely reflect your belief that our energy-intensive agricultural system is here to stay, a belief that cannot be confirmed or invalidated by conducting scientific research in a laboratory. It must be examined by considering energy-intensive agriculture in the light of a holistic model of the biosphere, and this would undoubtedly reveal that few of the conditions required to sustain this type of agriculture are likely to obtain in the next decades. With regard to the other statements that I have criticised, research carried out by you that might appear to confirm their validity can be presumed to be wrong, because they conflict

  1. with the basic ecological principles that you refer to as “dogmas”
  2. with all the mutually consistent empirical evidence that has been acquired over the last few decades on these subjects and which confirm the validity of these principles.

Let us first consider this notion of basic ecological principles which you refer to as “dogmas”. (I shall deal with the relativity between diversity and stability at a later date). Are basic physical laws dogmas too? Is the law of gravity a dogma? Is the second law of thermodynamics a dogma? Do you suggest that we should not have any laws and that every situation should be judged entirely on its own merits?

This would certainly create a paradise for scientists. Think of all the experiments that would be required if each one were only regarded as providing information on the specific white mouse it was carried out on, rather than on all white mice and perhaps even on other mammals such as man. Yet this is what you are suggesting.

As knowledge builds up and we learn how things work, it becomes possible to predict behaviour without having to carry out so many experiments and observations. If you put an earthworm in a maze it will take a long time to find its way out, a rat will get out more quickly, a man quicker still. The reason is that as we move from simple to more complex forms of life, learning ability increases and fewer experiments are required for finding out how things work. In fact one can formulate a basic principle or “dogma” to the effect that the number of observations required to understand how something works is inversely proportionate to one’s knowledge of it. Otherwise there would be no reason for acquiring any knowledge.

This knowledge, like all information used in nature, is organised hierarchically to constitute a model on the basis of which responses are mediated and monitored. A model is, of necessity, hierarchically organised, being made up of basic principles that are differentiated into less and less basic ones, dealing with increasingly less general and more specific things. The most basic principles tend to be treated as a prioris.

A primitive tribe, for instance, will not question the basic principles underlying its world view which provides it with a model on the basis of which its behaviour is mediated. We do not behave any differently, and the basic principles of the world-view of industrialism, which has been swallowed hook, line and sinker by most of our scientists, are also regarded by us as a prioris – that all considerations must be subordinated to economic ones, for instance.

Epistemologically speaking, no basic principles, not even the law of gravity nor the second law of thermo-dynamics are a prioris. Each is simply the best hypothesis that we can formulated for explaining important observable patterns on this planet. A hypothesis can of course be wrong. Without being wrong, a hypothesis can sometimes be advantageously replaced by another that takes into account certain factors that the previous one did not. In this way, Newton’s physics were replaced by Einstein’s.

Nevertheless if a single piece of research yields a result that is not in accord with basic principles that have been found to hold good over a long period of time, it must be regarded as exceedingly likely to be wrong. All the more likely, as observations, even in controlled laboratory conditions, are notoriously unreliable.

First of all, as Heraclitus said, “You cannot step into the same river twice”, the reason being that it is always changing. So too are the white mice that you study in your laboratory and the conditions in which you examine them. No two experiments can ever be carried out in identical conditions. Nor do experiments on things carried out in isolation from other things, provide very much knowledge about the real world, because things in the real World do not occur in isolation.

That is why the possibility of Minamata disease was not predicted. It was not then known that organic mercury would be changed, by the marine ecosystem, into the terribly toxic di-methyl mercury, because this did not happen in test tubes. The things that you study in your laboratory do not in fact exist outside laboratories any more than do phoenixes and unicorns.

This is one of the basic lessons of ecology. The biosphere is an organisation and as such is not just made up of the sum of its parts. How these parts are organised is equally important and this your laboratory research will never tell you, because it is designed precisely so as to isolate the thing you are studying from all other things with which they are normally interrelated and which are wrongly regarded as irrelevant for the purposes of the study.

What is more scientists often make mistakes. Que Choisir?, the French consumerist magazine, recently provided 32 laboratories with a specimen of human faeces, into which they had introduced various micro-organisms of a kind that are not commonly found in human faeces. Thirty-one of the laboratories failed to identify them and provided totally false analyses.

Scientists also cheat, as was revealed by a series of articles published last year by the New Scientist. Even if they do not falsify the results of experiments, these are often set up in such a way that they will yield the desired results. It was in this way that Dr Dolphin attempted to demonstrate that workers at Windscale had a lower cancer rate than anyone else in the country, in defiance of known principles of cancer induction by low levels of radiation.

What he did was conveniently avoid including in his study, workers who had left Windscale to work elsewhere, who had retired or who had died, (possibly of radiation-induced cancer). The same was true of the famous experiments invariably quoted by Shell Chemicals to demonstrate the harmlessness of DDT. I shall quote Wurster on this subject.

“Several studies of the physiological effects of DDT, Aldrin, Dieldrin and Endrin, have involved human subjects [Jager, 1970; Hayes et al., 1971]. These studies were deficient in experimental design, failed to consider the most relevant parameters and were more concerned with levels of CH storage than with physiological or biochemical effects. They establish only that under current environmental conditions, excluding accidents and suicides, members of the general population are not dying of acute CH insecticide poisoning, nor are they suffering overt toxic symptoms. Long-term, chronic effects were inadequately studied.

“To be more specific, the investigations by Hayes et al. [1971] and those conducted in the Shell laboratories [Jager, 1970] had only men in their samples; women, children, infants and foetuses, were not studied. The small numbers of men involved were completely inadequate to evaluate biological events (such as carcinogenesis or mutagenesis) that may occur once in many thousands of individuals. Periods of exposure were too short to detect biological effects involving induction periods that may be many years or decades. Emphasis was given to reviewing the men’s attendance records at work and many of the other simple blood and other routine tests performed were largely irrelevant. When two of 22 men who were being fed high dosages of DDT became severely ill after months on this diet, they were dropped from the experiment and excluded from the data, with the conclusion that ‘at no time was there any objective finding to indicate a relationship between illness and DDT storage’. [Hayes et al., 1971]

“It is unlikely that these tests on men could have detected behavioural changes, hepatic enzyme induction, carcinogenesis, mutagenesis or other effects that might be anticipated in man because they occurred in experiments with laboratory animals. The authors concluded, nevertheless, that exposure to these CH insecticides involved no ilI-effects on human health – a conclusion that has been widely quoted by the pesticide industry. It seems remarkable that although hundreds of millions of people have been exposed to these substances for more than two decades, their effects have been so inadequately tested by such primitive studies on such a small number of men!”

The fact is that observation and measurements that conflict with our knowledge of a particular subject are usually wrong, as our everyday behaviour should make clear. If you put a stick into the petrol tank of your car and it reveals it to be empty, and you then get into your car and drive off quite happily for a couple of hundred miles or so, do you assume that your car has learnt to function without petrol? Of course not. You know damned well that the measurement you made was wrong.

In just the same way, if you met a man who tells you that he has built a perpetual motion machine, you know that he is either lying or that he has failed to identify some external form of energy which is in fact responsible for the motion in question. In the same way if a BNFL scientist tells you that the cancer rate is lower at Windscale than anywhere else, and if scientists working for Shell Chemicals tell you that they have carried out experiments that have proved that DDT has no ill effects, you know that they have either made a mistake or their experiments have been wrongly carried out.

The same must be true of your particular research and that of your assistants. If it proves that modern agricultural practices do not cause erosion; that one does not encounter diminishing returns on fertilisers; that soil treated with Paraquat is an ideal habitat for wildlife; that hedges are of no ecological value; that low levels of pesticides are harmless; that we should go on using DDT and that millions of people in the Third World will die of malaria if we do not; and that growing exotic trees and clear-felling them at regular intervals is good conservation practice, then there must clearly he something wrong with it. To prove it right, you would have to reconcile your findings with the following recent theoretical and empirical material.

Back to top

Why DDT must be bad

There are good, in fact unanswerable theoretical reasons why highly biologically active, synthetic substances such as DDT should not be introduced into our environment. Consider that it has taken several billion years of evolution for the biosphere or world of living things, of which we are an integral part, to take on the shape that industrial man found it in and thereby provide an ideal habitat for man and the myriads of other forms of life that compose it.

During the course of this evolution, as Commoner puts it:

“The chemical, physical and biological properties of the earth’s surface gradually achieved a state of dynamic equilibrium, characterised by processes which link together the living and non-living constituents of the environment. Thus were formed the great elementary cycles which govern the movement of carbon, oxygen and nitrogen in the environment, each cycle being elaborately branched to form an intricate fabric of ecological interactions. In this dynamic balance, the chemical capabilities of living things are crucial, for they provide the driving force for the ecological cycles; it is the chemistry of photosynthesis in green plants, for example, which converts the sun’s energy to food, fibre and fuel.”

In other words the biosphere, or world of living things of which we are an integral part, can function as a self-regulating natural system and maintain its basic structure, on which the very survival of its living components depend, only if the critical interrelationships between all its components – at all levels of organisation, including that of the atom or the molecule – are maintained.

Now to say that the biosphere has a structure means that it displays order, organisation or negative entropy, and that the parts are subject to the influence of the whole, and hence can contribute to overall stability.

Another way of looking at this is to say that constraints limit their range of possible responses. The only situation in which there are no such constraints is that of total disorder, disorganisation or entropy. Thus as the primaeval dust began to organise itself into ever more complex forms, so has there been a corresponding increase in constraints. The evolutionary process can in fact be regarded as the development (or more precisely the accumulation) of constraints.

As Commoner points out

“. . . the chemical processes which are mediated by the biochemical system represent an exceedingly small fraction of the reactions that are possible among the chemical constituents of living cells. This principle explains the frequency with which synthetic substances that do not occur in natural biological systems . . . turn out to be toxic”.

He illustrates this principle thus:

  1. Of the approximately 100 chemical elements which occur in the materials of the earth’s surface, less than 20 appear to participate in biochemical processes, although some of those which are excluded, such as mercury or lead, can in fact, react quite readily with natural constituents.
  2. Although oxygen and nitrogen atoms are common in the organic compounds found in living systems, biochemical constituents which include chemical groupings in which – nitrogen arid oxygen atoms are linked to each other are very rare.
  3. Although the numerous organic compounds which occur in bio-chemical systems are readily chlorinated by appropriate artificial reactions and the chloride ion is quite common in these systems, chlorinated derivatives are extremely rare in natural biochemical systems.”

It is no coincidence that these chemicals are not found in living tissues. There is good reason for it. The organisation that is the biosphere has been able to evolve at the expense of eliminating possible reactions between these substances and living things. If any living systems once included them, then they have been eliminated by natural selection. As Commoner writes:

“the consistent absence of a chemical constituent from natural biological systems is an extraordinarily meaningful fact. It can be regarded as prima facie evidence that, with a considerable probability, the substance may be incompatible with the successful operation of the elaborately evolved, exceedingly complex network of reactions which constitutes the biochemical systems of living things.

Furthermore, such theoretical considerations are confirmed empirically. Thus mercury is one of those 80 elements not found in living tissue. There is at least one good reason for this – biochemical systems have evolved a system of enzymatic catalysis in which sulphur-containing groups play a crucial role. These react with mercury introduced into a living system and enzymes are inactivated, often with fatal results.

There is also a good reason why synthetic nitroso-compounds in which nitrogen and oxygen atoms are linked do not occur either in living tissue. They appear to interfere with the reactions involved in the orderly development of cells, and often give rise to cancer and mutations. So too, synthetic organochlorine compounds such as DDT and PCBs are excluded from living tissue. They are often very toxic or produce long term damage such as cancer.

In general, the more the environment changes as a result of man’s activities, and the less it resembles that in which we evolved, the less efficiently will our normal behavioural mechanisms enable us to adapt to it. Thus, while the human liver is capable of detoxifying those chemicals that it has learnt to detoxify over millions of years of evolution, it is incapable of detoxifying chemicals to which man has not been exposed during this period and which are thereby likely to cause biological damage.

Perhaps, Professor Mellanby, you disagree with these principles? If not how do you explain that your research shows things to be otherwise?

If there are theoretical reasons why DDT and other such substances are harmful, there are also very good empirical ones. The amount of material that points to the biologically damaging effects of DDT is immense and growing the whole time. I cannot review it all here. I shall simply look at recent material – theoretical and empirical, that casts light on those aspects of the use of pesticides that you have stressed in your various statements – and that is relevant to the subject of forestry and conservation.

Back to top

The effect of sub-lethal levels of pollutants

Since receiving your letter I have obtained copies of many of the papers read at the recent Royal Society meeting on the sub-lethal effects of pollutants on marine organisms. These papers all tend to confirm the increasingly accepted thesis that exposure to low levels of many pollutants over a long period can often be as harmful as high levels over a short period. Let me quote three paragraphs from Waldichuk’s paper “The Assessment of Sublethal Effects of Pollutants in the Sea”.

“In general, the harm in bio-accumulation of such substances as metals and organo-chlorines is usually associated with consumption of sea food by humans. Levels of mercury at 0.5 ppm and higher in the tissues of organisms that bio-accumulate this metal have not been considered seriously harmful to the organisms themselves. However, some laboratory research demonstrates that this may not be true. When fish containing as little as 0.1 ppm mercury in their muscle tissue, are subjected to a torque in a rotating cylinder, they have greater difficulty in compensating for it than the control fish [Lindahl & Schwanbom 1971]. Cadmium apparently affects calcium metabolism and this has adverse consequences on the otolith and the equilibrating mechanism of fish [Rosenthal & Alderdice 1976].

“Anderson [1971] demonstrated that concentrations as low as 20 ppb of DDT produced a condition response (propellor tail reflex) in trout. His experiments showed that the learnability of fish could be affected by very low concentrations of chlorinated hydrocarbons. This could have rather unfortunate consequences on the ability of fish to return to their home streams, if they were exposed to even low concentrations of DDT during the imprinting period in their juvenile stage. Kleerekoper [1976] has indicated how behavioural responses can be influenced by multiple types of alterations in the aquatic environment with a gradient in metal concentration superimposed on a thermal gradient. Chemo-receptors can be affected in organisms by pollutants and this could be of great significance to fish, not only in homing to their parent streams on a spawning migration, but also in searching for food and possibly in avoiding predators. Equilibrium in fish can be affected by uptake of mercury [Lindahl & Schwanbom 1971] and cadmium [Rosenthal & Alderdice 1976].

  • Twitter
  • Facebook
  • Digg
  • Reddit
  • StumbleUpon
  • Diaspora
  • Identi.ca
  • email
  • Add to favorites
Back to top

Pages:  1   2   ALL