APPENDIX E
SCIENCE, POLITICS AND ETHICS
IN
THE LOW DOSE
Keith Baverstock, the senior radiation
advisor to the WHO was sacked because he drew attention to the matters I have
raised in this report. This is his paper on the issue, published in a peer
review journal.
Science, Politics and
Ethics in the Low Dose
Debate
KEITH BAVERSTOCK
University of Kuopio, Finland
SCIENCE, POLITICS AND ETHICS IN THE LOW DOSE DEBATE 89
MEDICINE, CONFLICT AND SURVIVAL, VOL. 21, NO. 2, 88 – 100 (2005)
ISSN 1362-3699 print/1743-9396 online
DOI: 10.1080/13623690500073380 # 2005 Taylor & Francis Group Ltd.
90 K BAVERSTOCK
The roles of science, ethics and politics are identified in respect of the risks of
exposure to low-dose radiation. Two case studies, the epidemiology of the United
Kingdom nuclear test veterans and the risks to civilians associated with the military
use of depleted uranium, are considered in the context of their ethical framing,
scientific evaluation and political resolution. Two important issues for the present
and future, the safe management of UK radioactive waste and the future of nuclear
power, in which the science of low dose effects will be crucial and where the ethical
issues are much more complex, are introduced. Specific consideration is given to the
potential hereditary effects of ionising radiation in relation to the current state of
radiobiological knowledge. It is concluded that for science to be useful in public
health policy making there needs to be some reform from within the profession and
the political imperative for freely independent scientific institutions.
KEYWORDS Depleted uranium Ethics Genomic instability Nuclear power
Nuclear weapons tests Politics Radiation risks Radioactive
waste
Introduction
A life without any risks whatsoever would be boring and some would say
totally uncharacteristic of human nature, so we must accept that risk is a
part of life. But how much, of what nature, and how caused, are important
issues not to be dismissed lightly.
Alice Stewart identified a risk to children from the exposure of pregnant
women to diagnostic X-rays in the 1950s, which was to prove to be pivotal
in transforming our perception of risk from low-dose ionising radiation.(1)
It was by any standards a remarkable piece of dedicated scientific investigation.
It caused alarm and concern in the radiological protection and medical
communities when the result was first published in 1956, when Alice was a
relatively newly qualified doctor. The initial ‘establishment’ diagnosis was
that there was a mistake, but as the evidence was consolidated and a similar
result was reported with a larger number of cases and controls, the personal
criticism started. It was not to be until the 1970s that Alice’s claims were
vindicated. The International Commission on Radiation Protection (ICRP)
recently published a report (2) that devotes great attention to Alice’s
contribution (the Oxford Survey of Childhood Cancers is the biggest study
on this issue ever mounted), but generally in a highly critical tone, with an
eventual rather grudging acceptance that Alice was right.
Alice’s professional life illustrated a phenomenon that can be seen when
someone, as the great British geneticist CH Waddington noted, not a
member of the ‘dominant group’ in scientific society, claims a discovery.
First, attribute the ‘discovery’ to a simple error born of lack of experience;
then, when the claim to the ‘discovery’ is not withdrawn, attribute mental
instability to the discoverer; and finally, to point out that this ‘discovery’
was not a discovery at all but had in fact always been known. Now the ICRP
has added a ‘post-final’ stage which, in essence, notes that the basis for the
discovery was in fact highly suspect and the discoverer the beneficiary of a
great deal of luck in that although the correct result was obtained it was not
by a scientifically valid method. I will return to the issue of flawed
epidemiology later. But let there be no doubt that Alice overturned what
was the established view of radiation as being rather a benign agent with
great benefits to humankind, and the process she started in the 1950s is still
changing our view of this particular agent. So in this sense Alice lives on.
Radiation Risks
We know that exposure to ionising radiation does present serious risks to
health. Exposure to, say, one gray (Gy) in periods of a few hours will
produce health consequences that are directly observable and attributable to
the exposure and we understand what causes what happens as a result of
scientific investigation. But when we try to study what effect such a dose
might have when spread over several years, measurement of the effect in a
population, through epidemiology, becomes much more problematic. Some
findings point to a risk, others are inconclusive, and still others seem to
indicate a beneficial effect. Chance has started to play a role. In these
circumstances science comes into play again, by constructing models of
what might be happening and extrapolating those models on assumptions
that are judged or believed to be reasonable and realistic. An example is the
linear no-threshold (LNT) model, upon which radiological protection is
based. I believe that LNT is credible and realistic (as well as an appropriate
modus operandi for radiological protection) on fundamental grounds;
nevertheless, that is a belief and not knowledge. Others believe otherwise.
However, societal decisions concerning risk acceptability are essentially
political. In democratic societies such decisions are taken by elected
governments, or bodies nominated by them. Most governments make strong
claims to be basing their decisions on risk on the best available scientific
opinion, but we must be aware that this opinion involves beliefs as well as
knowledge.
Beliefs of another kind are also relevant in this context, which can
collectively be called ethical considerations. These almost always ‘frame’ the
risk issue, influencing the perceived importance of various aspects of the risk
debate; they therefore also impact on the science. Risk is a simple sounding word but it is quite a complex concept. It involves both the degree of impact and the frequency of both detrimental and beneficial effects, which sometimes leads to ambiguity and therefore misunderstanding. Uncertainty is usually entrained in any risk assessment
and has to be identified and addressed. In policy making risks have to be
traded, mitigating one risk may enhance another, one detriment might be
accepted to obtain a benefit elsewhere, economic cost in mitigating one risk
might exacerbate another, and so on. Good policy seeks the best
compromise in a very complex web of scientific and social issues, but my
thesis is that it requires above all sound and honest science and careful and
sensitive ethical framing.
The aim of this paper is to explore, initially with the help of two relatively
simple examples, how science, ethics and politics have inter-played, one
with another, in practice. In each case I will first outline the setting for the
risk issue, then discuss the ethical and scientific considerations, positing
what might be the correct political outcome, and finally I will describe what
has in fact happened. I then introduce two issues for the future where the
scientific, ethical and political dimensions are much more complex.
Two Case Studies
The examples I wish to use to illustrate my points are firstly the health of the
United Kingdom test veterans: Some 25,000 people, mainly young men,
served their country by providing the backup and support for the UK
weapons testing programmes in Australia and the South Pacific in the 1950s.
For more than the past 25 years many have felt that their health has been
adversely affected but many do not receive compensation for their injuries.
The second example is the use of depleted uranium (DU) weapons. DU has
been used in many battle theatres since the Gulf war of 1991. Although DU
is acknowledged to be radioactive, its specific activity is low and it is
therefore not thought to present a serious radiation hazard.
I cannot, in the time available, give exhaustive consideration to either one
of these, let alone both, so I will have to be selective, but I also hope to be
even-handed. I consider each in turn and try and draw some conclusions at
the end.
The UK Test Veterans
In 1983, following pressure from veterans associations, it was decided to
mount an independent epidemiological study on the UK test veterans, and
the National Radiological Protection Board (NRPB) was funded to
undertake the work by the Ministry of Defence (MoD). The study was
designed to compare the test veterans with a similar cohort of servicemen
from the three services whose tours of duty took them to tropical areas but
not the nuclear tests in Australia or the Pacific. There have been three
analyses of the survey, the latest published in 20033 together with a fuller
report.(4)
There was a strong ethical element in mounting this study, as it was
considered at the time that if there was reasonable concern that an adverse
health consequence would accrue from any occupational exposure, the
possibility of resolving the concerns through epidemiological study should
be considered. It is certainly the case that no one thought that the study was
intended to add to scientific knowledge about radiation as a cause of cancer;
the survey had a purely socio-ethical justification.
The NRPB initially accepted from the MoD, apparently without
independent verification, the primary data in the form of names of
servicemen attending the tests and the rather sparse dosimetric data. It
was recognised from the outset that there was incomplete ascertainment,
especially of RAF personnel, and after the first analysis in 1988, further
veterans were found by the veterans’ organisations and notified to the
NRPB. These persons were put in a separate category called ‘independent
responders’ and have not been included in the main analysis but have been
analysed separately.(5) It is now established that there might well be a
shortfall of 15 per cent on the ascertainment of the test veteran population.
From a scientific point of view and contrary to the claim made by the NRPB,
this shortfall raises the prospect of a serious flaw in the methodology.5,6 The
exposed and control populations can no longer be guaranteed to be free of
bias as there can be no such shortfall in the controls, who were not
ascertained from a finite and defined population (they were simply 22,000
service personnel not having served in the test areas). The NRPB claims that
in spite of the shortfall the population is representative.
The veterans’ associations maintain that the independent responders
should be included in the main analysis. Sue Roff maintains that the NRPB
analysis has missed 30 per cent of the cases of multiple myeloma and thus an
analysis missing 15 per cent of the veterans cannot be representative.(7)
There does not have to be any deliberate wrong-doing for there to be a
serious bias problem here. Records compiled in the 1950s may well be
incomplete some 20 to 30 years later, especially if there were no
compelling reasons to keep the records in good order. If the loss of the 15
per cent of records was associated in any way with the health outcomes
being studied, then the fact that there is not a comparable loss in the
controls (for the same reason) immediately introduces bias. One reason for
the loss of records may well be their relocation in connection with a claim
for compensation or diagnosis of illness or death. When all in this
population have died there will be some 25 per cent of deaths due to
cancer. At the last analysis, almost 23 per cent of veterans had died, with
seven per cent from cancer, that is, less than the 15 per cent missing from
the study population. Thus it is entirely possible for the missing 15 per
cent to conceal an excess of cancer deaths.
A second issue concerns the results of the survey, which found an excess
of all leukaemia, excluding chronic lymphatic leukaemia (which is not
thought to be associated with radiation) in the veterans population when
compared with the controls but not when compared with the general
population.(3) This results from a large deficit of leukaemia in the controls
for which the NRPB has not found an explanation, so attributes it to
‘chance’. Now there is an ethical issue here. Having chosen at the outset to
compare the veterans with a control population, it is unacceptable both
scientifically and ethically to ‘move the goal posts’ when the result is
known. Even more interesting, but not disclosed by the NRPB (although
they claim to have known it), is the fact that in the veterans population,
the excess leukaemia risk, compared to the controls, appears to be
concentrated in those who served in the Pacific.(6) Those attending tests in
Australia seem to have a similar leukaemia incidence to the controls. Of
course, as the populations are subdivided so the statistical significance of
the result declines, and it becomes more difficult to define risks as
attributable.
The lack of dosimetric data also is a factor. Some duties, such as
decontaminating planes and vehicles, are likely to incur higher doses than
others such as servicing the canteens. Lumping together the exposed and
unexposed in the absence of individual dose assessments, however crude,
will ‘dilute’ any exposure-related excess of disease.
The correct political conclusion is that the NRPB survey, for a number of
reasons, is deficient and that the data have not been exploited to the full
extent that is possible to resolve the impact on health of the test veterans.
Further work needs to be done.
The present political position is that, according to the NRPB survey, the
veterans have not had their health damaged by their participation in the tests
and thus the MoD is able to conclude that there is no case for compensation
for injuries that are claimed as due to radiation.
In this case it is clear that the science and the associated ethics (of
recognising the need for an independent study) have been perverted for
political ends. It is sad that the NRPB, which should be an independent and
technically competent body, was complicit in this process.
Depleted Uranium
Depleted uranium (DU) has a lower specific activity than naturally occurring
uranium, which contains greater quantities of two other uranium isotopes,
U-234 and U-235. Technically it is a waste arising from the enrichment
process that produces U-235 for weapons and civil nuclear power plants. In
1991 it was used in the Gulf War as a weapon. It is not its radioactivity that
makes it effective as a weapon but its density.When delivered as a bullet to a
hardened target, a DU munition will have sufficient momentum to penetrate
armour and buildings. After penetration the bullet fragments and burns,
causing the release of an oxide smoke consisting of very fine particles.
Typically, when a tank is hit, up to four kg of depleted uranium oxide dust is
formed.
There is of course an ethical dimension to war of any kind, but putting
that to one side for the time being, the Geneva Convention seeks to
minimise the impact of war on civilian populations. Over time battlefields
usually return to civilian use and weapons that remain, such as unexploded
land mines and cluster bombs, are not supposed to be left to pose a health
risk to the civilian population. The question here is whether DU, either as
unburned metal or as oxide dusts, poses such a risk. If it does it should be
cleaned up or such weapons banned.
Natural uranium (NU) is ubiquitous in the environment and the chemical
properties of DU are identical to those of NU. DU metal buried in soil will,
over many years, dissolve and enter ground water where it may raise the
uranium concentration, perhaps by a few per cent. This is unlikely to pose a
public health hazard.
As far as the International Commission on Radiological Protection
(ICRP) and the International Atomic Energy Agency (IAEA) are concerned,
DU oxide dust can be treated like any other uranium oxide.(8,9) However,
although there is extensive exposure of workers in uranium mines to
naturally occurring uranium oxide dust, there is no natural analogue for
depleted uranium oxide. What we know from occupational exposure to
uranium compounds, including relatively soluble oxides, is that uranium is
chemically toxic. In fact, the ICRP regards uranium primarily as a chemical
hazard and not a radiological one, an exception being insoluble uranium containing
particles retained in the lung. However, the product of burning
of depleted uranium is a mixture of two oxides, one insoluble and the other
sparingly soluble. The more soluble oxide of depleted uranium, when in the
body, for example retained in the lung, results in the formation of the
uranyl ion. This, while soluble in tissue, binds avidly to DNA and proteins
and so is only slowly transferred from the lung tissue to blood, from where
it is transferred to other tissues, particularly the bone, before finally being
excreted through the kidney. Damage to the kidney from exposure to the
uranium is generally regarded as the principal toxic effect. However, recent
results have indicated that while it is in transit to the kidney, that is,
retained over long periods deep in the lung, it may give rise to genotoxicity,
mediated not by radiation alone but by its chemical properties in
combination with its radioactivity.(10) There will be a period, ranging
perhaps from months to years, where a slowly dissolving particle in the
deep lung is surrounded by cells containing uranyl ions. Typical particles
may emit an alpha particle once every few weeks, and thus there is the
possibility of a synergistic effect between a chemical carcinogen and
radiation. There is also the possibility, particularly important for lowspecific-
activity alpha emitters, of effects mediated by the bystander effect,
where cells not actually irradiated, but located close to ones that are,
exhibit radiation effects.
From articles published in 2003 it is clear that neither the ICRP (9) nor the
IAEA (8) have taken these three potential effects into consideration when they
assess the risk from inhaling depleted uranium dusts. It is also the case that
when the World Health Organisation were advised of these three potential
mechanisms they ignored the information in the preparation of a
monograph on the health effects of depleted uranium published in 200111
and subsequently suppressed the publication of a paper postulating these
three mechanisms.
In 1991 the United States forces discharged 300 tonnes of depleted
uranium in the area around Basra. More than 800 tonnes are said to have
been deployed by the US in the latest Gulf War. Given the arid and dry
climate that affects much of Iraq it seems likely that the resultant oxide dusts
will remain potentially dangerous, if re-suspended, for a considerable time.
This contamination presents a serious potential hazard to health for both
the Iraqi population and the coalition forces.
The science here is clear. The hazard is not certain, that is, the risk is not
attributable, but it is not so speculative that it should be ignored. The ICRP
routinely uses essentially untested models to determine the risks from
internal emitters. I suggest that the science behind the postulated
mechanisms I have just described is somewhat harder than that underpinning
some of the ICRP models. But there is also an ethical issue in
connection with the role of the IAEA in responsibility for the safe use of
nuclear technology. Depleted uranium is a by-product or waste of nuclear
technology and thus should come under the safety mandate of the IAEA. But
its responsibility should not be limited to the radiation effects, but should
consider all the hazards associated with the material. Nor should the ICRP
consider that its responsibility is confined to radiation effects. Where there
is the possibility of an interaction between two carcinogenic processes the
non-radiation one should not be ignored. Perhaps the most serious violation
is that of the WHO, whose mandate to protect public health has surely been
compromised. In an ideal world the WHO would have alerted the IAEA and
the ICRP to the potential hazard of DU oxide dusts in Iraq.
The political situation as it should be is that, until there is clear evidence
that DU oxide dusts are harmless, either the weapon should be banned or
battlefields where it has been used should be cleaned up. Also there are
alternatives to depleted uranium for its military uses. Tungsten is almost as
good a penetrator but it does not break up and catch fire, so it is not as
effective at killing people as depleted uranium. We cannot therefore ignore
the possibility that the IAEA, ICRP and WHO are responding to political
pressure not to disclose the potential health consequences to either military
or civilians in the use of depleted uranium. In fact, Dr Thomas Facy
(personal communication) notes the discrepancy in the US between the
military attitude to the hazards of uranium and that taken by the civil
uranium industry, where effective and elaborate precautions are taken to
protect the workforce. Clearly there are double standards operating here.
Implications
These two examples illustrate how science can be distorted to achieve
political ends, in the first case to avoid paying compensation and in the
second retaining the military capability of DU. I could have given many
other examples but I chose two that are topical, one because of the relatively
recent report by the NRPB on the test veterans study, and the other because
the existing contamination of Iraq by DU has, within the past year or two,
been made even worse, and three international agencies have recently
endorsed a less than precautionary approach to the effects of DU on health.
All the organisations involved – the NRPB, the WHO, the ICRP and the
IAEA – claim ‘independence’ and technical/scientific excellence. At a
Nuclear Energy Agency workshop in Villigen, Switzerland, I even heard
Abel Gonzales of the IAEA, who is also a member of the ICRP Main
Commission, claim that the ICRP is a ‘scientific academy’. While it has to be
recognised that several highly qualified people are members of that
organisation, it hardly rates as that. But it is the case that the international
agencies (WHO and IAEA) and the NRPB do employ people who should be
scientifically and technically competent and trustworthy. So if it is not
incompetence, what is it that has led to such perverse political outcomes
when the science and the ethics are so clear, even if the science is not
sufficiently strong to produce irrefutable evidence to allow risks to be
objectively assessed – that is, it depends on judgement to a degree?
The fact that situations such as these exist has a profoundly negative effect
on policy making, by corroding public trust in science and technology. In
the UK this process has been underway since the early 1980s where
radiation is concerned. Previously, the Medical Research Council played the
role of a ‘referee’ between the pro- and anti-nuclear lobbies through the
application of objective scientific risk assessment. There was considerable
public (and ‘player’) trust in the result, and as a consequence there was not
the same social concern about the risks of low doses as there is today.
Politics, aided and abetted by some in the scientific community, can be said
to have poisoned the well that sustains democratic decision-making.
Two Problems for the
Future
At the present time we face at least two very crucial and inter-related issues.
The first is especially important to the UK: the future management of the
accrued radioactive wastes over the past half-century. The second is a global
problem but one for which there has to be national policies, namely the
future of energy supply and the role that civil nuclear power will play. I have
a particular interest in the first problem as a member of the Committee on
Radioactive Waste Management (CoRWM) set up recently to advise
government on a long-term management strategy that is both implementable
in the foreseeable future and commands public confidence.
Both these problems have very significant ethical dimensions relating to
environmental sustainability, equity and fairness to future generations and
to those in whose backyard existing and future wastes will end up. Both are
also extremely technically challenging, the first requiring a means of
protecting the environment from the release of radioactivity for very long
periods and the second technical ingenuity to ensure future generations
sufficient energy supplies. So far we have barely started to solve these
problems. Clearly, as far as the second problem is concerned, we have to
consider the future of nuclear energy, a non-greenhouse gas source, with an
abundant fuel supply but a largely unsolved waste problem, in which the low
dose issue is writ large.
With respect to the second problem we do have a choice as to whether to
solve it or not, but that option does not exist for the first problem, since we
already have the waste. This must be managed in some way, even if this
involves continuing to store it on the surface, at some considerable cost,
which takes resources from other possible uses, and where the waste is
vulnerable to accidents and terrorist action.
An issue common to both problems is that the ethical framings through
which individuals see these problems can be very diverse, even directly
opposed. This exposes the need for some serious risk trade-offs. In the case
of the future of nuclear power, the health risks of low dose exposure have to
be traded against those of global climate instability, so long as carbon based
fuels are the only viable alternative to nuclear. A similar issue arises in
respect of the stocks of plutonium in the UK. This material can be seen
either as a threat because of its potential to make weapons or as future fuel.
Here the risk of misuse now or in the future has to be traded against the
benefits that could accrue to future generations if the plutonium is available
to be used as a fuel.
Clearly, the solutions to these major challenges are not only a matter for
scientists and technologists but require, in a democratic society, the close
involvement of the public and those who have a particular interest, whom
we call stakeholders. Nevertheless science and technology has a crucial role
to play. We should not be forgiven by future generations if we fail to use, to
the full, the best scientific and technological knowledge known to man. The
failure of good science to prevail in the two examples above and in the many
others I could have quoted, has eroded the trust of the public and
stakeholder communities in science and technology to the extent that there
has grown up a phenomenon called ‘cognitive relativism’. This believes that
there are no truths and no best solutions to problems that have a strong
scientific and technological element. In the view of a cognitive relativist risks
from low dose radiation are a matter of belief, not reality.
Cognitive relativists would advise us to find solutions to the challenges of
future energy supply and nuclear waste management primarily on the basis
of public opinion. In the UK it is estimated that less than ten per cent of the
population claims to know anything about nuclear waste. It would seem to
me that relying on that approach would be little better than tossing a coin to
choose between options.
Clearly we must do better than that, but it is a real problem to find a way
to ensure that any solution to the waste problem is safe and satisfies the
legitimate requirements of democracy. This problem has been made
significantly worse by the experience of the test veterans and by the way
the DU issue has been handled, as I have just presented to you and by many
other examples in a similar vein. Sadly it is the case that some of those
scientific and technically based organisations that have taken on the
responsibility of being ‘independent’ technical bodies have misused science
in a way that overrode the strong ethical issues in which the problems being
tackled were framed by society. Done once this can be seen as an accident,
but when it is done repeatedly we know that it is deliberate political
interference. It becomes increasingly difficult for society to trust these
bodies, and ultimately those who set them up, namely politicians, with the
overridingly important task of protecting public health, and the door is open
to cognitive relativism.
Back to the Science
Even without this problem we would, as scientists, have an extremely
complex task in solving the radioactive waste problem. I want to focus on
only one of many uncertainties in predicting future risks from radiation
exposure, that is, the future genetic consequences. The phenomenon of
radiation induced genomic instability was only discovered just over a decade
ago,12 which, because of the large target size for the effect, is important at
low doses. Somewhat more recently the transmission of a form of instability,
mini-satellite DNA mutation along the germ line13–17 has been revealed. So
far mini-satellite mutations have not been linked with a specific health effect
and they occur spontaneously at a relatively high frequency in any case.
Does this mean we can ignore this effect in projecting over many
generations the risks of exposure to low doses of ionising radiation?
One feature of this phenomenon that concerns me is the lack, in the
studies with mice,(13) of ‘dilution’ of the effect in the ‘grandchildren’ of the
irradiated mice, that is, the mutation frequency is the same in the two
offspring generations. There are two implications:
. there is likely to be no fading-out of this effect over generations in the
future as would be expected in classical genetic effects;
. a mechanism that is presently wholly unknown must be involved.
These implications must decrease considerably the confidence we can have
about the health of future generations after exposure to radiation.
Perhaps the first question to answer is; ‘are these observations reliable’?
The phenomenon of mini-satellite mutation has been seen in the children of
Chernobyl exposed fathers (14,17) and in the two generations of offspring from
fathers exposed to weapons testing in Kazakhstan. (16) An extremely closely
related phenomenon, tandem repeat mutations, has been seen in mice.(3)
However, mini-satellite mutations have not so far been observed in the
children of the survivors of the atomic bombings in Japan,(18) in Chernobyl
clean-up workers, (19) or in a study of radiotherapy patients.(20) There could be
reasons for this lack of uniformity of observation and we should recall that
we assume a genetic risk in radiological protection largely on the basis of
studies in mice and not direct observation in humans. I think we have to
assume that at least under some conditions the phenomenon of germ-line
transmission of mini-satellite mutations exists. We should, therefore,
exercise caution even though we do not know how seriously the
phenomenon impacts on health, as the effect may be irreversible and
potentially with major consequences.
I contend that phenomena such as this, and genomic instability in general,
can be understood if we assume that the genome is a dynamically stabilised
or self-organising entity.(21) For life to have evolved over more than three
billion years requires astonishing robustness. Yet, when we look at living
systems we see incredible complexity, requiring a very high degree of
organisation both spatially and temporally that surpasses anything that can
be man-made even in the simplest of living creatures. Usually qualities of
robustness and complexity do not go hand in hand. It is as if a living
organism is an object with the simple robustness of a steam traction engine
and the complexity of a Formula One racing car. Perhaps we find it difficult
to comprehend such an object because we have so far not really studied
dynamically stabilized objects very thoroughly. Another such system is the
climate, which while capable of producing extreme conditions, is in fact
extraordinarily stable given its possibilities. Yet we have the ‘butterfly wing
effect’, in which it is said that a flap of a butterfly’s wing in Hong Kong can
cause a hurricane in the Caribbean. In other words the system is both robust
and sensitive. The climate is a dynamically stabilised system. Very little
attention has been given to the possibility that the genome is such a system.
The point here is that as scientists we must be open to possibilities not so
far conceived. Perhaps it was Alice Stewart’s ‘inexperience’ in her early
professional years that allowed her to take seriously results that her peers
would probably have rejected as artefacts. This was Waddington’s message
when he derided the ‘conventional wisdom of the dominant group’, which
he coined as the acronym COWDUNG.(22) There is, however, a genuine
paradox here; we need ‘stability’ in the knowledge base to make
scientifically informed policy, but we also need to move forward in our
scientific understanding of nature in order to ensure that policy remains
realistic. How to achieve the most beneficial compromise is a true challenge
for the scientific community.
What should not be such a challenge is how to use our existing scientific
institutions to better serve the policy-making process. There are clear
aberrations here and they can be corrected with the appropriate
determination of the scientific community and the appropriate political will.
Acknowledgements
This is the transcript of the Alice Stewart lecture to the 20th Anniversary
Low Level Radiation and Health Conference, Edinburgh, July 2004.
References
1. Stewart A, Webb J, Giles G, Hewitt D. Malignant disease in childhood and
diagnostic irradiation in utero. Lancet 1957; ii: 447.
2. International Commission on Radiation Protection. Biological Effects after
Prenatal Irradiation (Embryo and Fetus). In: Valentin, J., ed. Annals of the
ICRP,, Publication 90. Oxford: Pergamon, 2003.
3. Muirhead CR, Bingham D, Haylock RG, et al. Follow up of mortality and
incidence of cancer 1952–98 in men from the UK who participated in the UK’s
atmospheric nuclear weapon tests and experimental programmes. Occup
Environ Med 2003; 60: 165–72.
4. Muirhead CR, Bingham D, Haylock RGE, et al.. Mortality and cancer incidence
1952 – 1988 in UK participants in the UK atmospheric nuclear weapons tests and
experimental programmes. Didcot: National Radiological Protection Board,
2003: 129.
5. Muirhead CR, Kendall GM. UK nuclear-test veterans. Lancet 2003; 362: 331–2.
6. Baverstock K. The 2003 NRPB report on UK nuclear-test veterans. Lancet 2003;
361: 1759–60.
7. Roff SR. Under-ascertainment of multiple myeloma among participants in UK
atmospheric atomic and nuclear weapons tests. Occup Environ Med 2003; 60:
e18.
8. Bleise A, Danesi PR, Burkart W. Properties, use and health effects of depleted
uranium (DU): a general overview. J Environ Radioact 2003; 64(2–3): 93–112.
9. Valentin J, Fry FA. What ICRP advice applies to DU? J Environ Radioact 2003;
64(2–3): 89–92.
10. Miller AC, Stewart M, Brooks K, Shi L, Page N. Depleted uranium-catalyzed
oxidative DNA damage: absence of significant alpha particle decay. J Inorg
Biochem 2002; 91: 246–52.
11. World Health Organization. Depleted Uranium: Sources, Exposure and Health
Effects. Geneva: World Health Organisation, 2001.
12. Kadhim MA, Macdonald DA, Goodhead DT, Lorimore SA, Marsden SJ, Wright
EG. Transmission of chromosomal instability after plutonium alpha-particle
irradiation. Nature 1992; 355: 738–40.
13. Barber R, Plumb MA, Boulton E, Roux I, Dubrova YE. Elevated mutation rates
in the germ line of first- and second-generation offspring of irradiated male
mice. Proc Natl Acad Sci USA 2002; 99: 6877–82.
14. Dubrova YE, Nesterov VN, Krouchinsky NG, et al. Human minisatellite
mutation rate after the Chernobyl accident [see comments]. Nature 1996; 380:
683–6.
15. Dubrova YE, Plumb M, Gutierrez B, Boulton E, Jeffreys AJ. Transgenerational
mutation by radiation. Nature 2000; 405: 37.
16. Dubrova YE, Bersimbaev RI, Djansugurova LB, et al. Nuclear weapons tests and
human germline mutation rate. Science 2002; 295: 1037.
SCIENCE, POLITICS AND ETHICS IN THE LOW DOSE DEBATE 99
17. Dubrova YE, Grant G, Chumak AA, Stezhka VA, Karakasian AN. Elevated
minisatellite mutation rate in the post-Chernobyl families from Ukraine. Am J
Hum Genet 2002; 71: 801–9.
18. Kodaira M, Satoh C, Hiyama K, Toyama K. Lack of effects of atomic bomb
radiation on genetic instability of tandem-repetitive elements in human germ
cells. Am J Hum Genet 1995; 57: 1275–83.
19. Livshits LA, Malyarchuk SG, Kravchenko SA, et al. Children of Chernobyl
cleanup workers do not show elevated rates of mutations in minisatellite alleles.
Radiat Res 2001; 155(1 Pt 1): 74–80.
20. May CA, Tamaki K, Neumann R, et al. Minisatellite mutation frequency in
human sperm following radiotherapy. Mutat Res 2000; 453(1): 67–75.
21. Baverstock K. Radiation-induced genomic instability: a paradigm-breaking
phenomenon and its relevance to environmentally induced cancer. Mutat Res
2000; 454(1–2): 89–109.
22. Waddington CH. Tools for Thought. London: Jonathan Cape, 1977.
(Accepted 11 December 2004)
Keith Baverstock is at the Department of Environmental Sciences, University of
Kuopio, Finland. From 1971 to 1991 he was senior grade scientist at the MRC
Radiobiology Unit, Chilton and from 1991 to 2003 was Head of the Radiation
Protection Division of the World Health Organization (Europe). His research
interests are in the biological and physicochemical bases of the effects of ionising
radiation on health. Recently he has explored theoretically the possibility of treating
the genome as a complex adaptive system and has developed the dynamic genome
concept, which arises from new developments in radiobiology, in particular
radiation – induced genomic instability.
Correspondence: Department of Environmental Sciences, University of Kuopio,
70211 Kuopio, Finland; email: 5keith.baverstock@uku.fi4.