Oxford
WELL established in the US, toxic tort litigation has finally arrived on the
British legal scene. 鈥淭ort鈥濃�攍oosely translated from medieval
French鈥攎eans twisted or wrong, and is the law of civil wrongs. Toxic tort
cases concern allegations that the polluting activities of industrial facilities
have harmed individuals or their possessions. In 1993, Vivien Hope and Elizabeth
Reay sued British Nuclear Fuels (BNFL) in the High Court. They claimed that
radiation exposures received by Hope鈥檚 father and Reay鈥檚 husband while working
at BNFL鈥檚 Sellafield nuclear plant in Cumbria in the 1950s and 1960s caused
acute lymphatic leukaemia in the Reay鈥檚 daughter, Dorothy, who died at 11
months, and predisposed Vivien Hope to non-Hodgkin鈥檚 lymphoma (she later
recovered). In 1995 Andrew and Irene Graham alleged, also in the High Court,
that their cattle had been poisoned by chemicals manufactured by Re-Chem and
that the animals suffered illness, birth defects and infertility.
Key to both these claims was what lawyers call the 鈥渃oncept of
causation鈥濃�攃ould Hope and Reay and the Grahams show that their misfortunes
were caused by the actions of BNFL or Re-Chem? Determining the facts in Hope and
Reay, the judge, Mr Justice French, assessed the validity of an epidemiological
study carried out by the late Martin Gardner, professor of environmental
epidemiology at the University of Southampton. In the Graham case, the judge, Mr
Justice Forbes, weighed up the symptoms of fat-cow disease and assessed the
toxicological properties of polychlorinated biphenyls, dioxins and furans. Even
though the judges were able to show they understood the scientific
issues鈥攖heir judgments described principles of nuclear physics and organic
chemistry at some length鈥攏either explicitly addressed the different
approaches that scientists and lawyers take to burdens of proof, the threshold
of which determines when a scientific theory is robust or legal causation is
鈥减谤辞惫别苍鈥�.
Advertisement
The differences between the two disciplines become clear when courts turn to
epidemiological studies. Epidemiology examines the frequency and types of
injuries and diseases in human populations where the community is the primary
focus of concern. Proceedings in tort law focus on the individual, attempting to
determine whether one party鈥檚 actions (the defendant) are attributable to the
alleged harm suffered by another party (the plaintiff).
Epidemiological studies start with a hypothesis and then formulate its
inverse, attempting to falsify 鈥渢he null鈥� rather than proving the initial
hypothesis. Having borrowed extensively from statistics and probability, the
science seeks to reduce the possibility that the null hypothesis will be
disproved by chance. It employs a range of statistical tools and creates the
possibility of false negatives and positives. And here lies the problem. For
many epidemiologists and other scientists developing new hypotheses, the theory
should be as robust as possible and conventionally the probability of false
positives is about 5 per cent, with false negatives often 20 per cent. For
plaintiffs, and their lawyers, a theory which permits false negatives may harm
innocent individuals. As the American philosopher Carl Cranor has pointed out,
scientists primarily seek to prevent false positives from erroneously adding to
scientific knowledge and mistakenly chasing research chimeras. But public health
advocates are typically concerned to prevent false negatives, to avoid treating
a toxin as non-toxic.
Conversely, plaintiff lawyers argue that if the burden of proof in civil
cases is truly a 鈥渂alance of probabilities鈥� (better than 50:50), then requiring
scientific certainty effectively elevates every tort action into a 鈥渃riminal鈥�
one. In a criminal trial the evidence must be proved 鈥渂eyond reasonable doubt鈥�.
In both the Hope and Reay and the Graham case the plaintiffs鈥� lawyers argued
that their scientific theories of causation had only to be proven on a 鈥渂alance
of probabilities鈥� for their clients to be compensated. In both cases the judges
declined to address this point of principle, concluding instead that neither
explanation of causation had even reached this lower threshold.
American lawyer Peter Huber cautions against overly credulous 鈥渏unk science鈥�.
Following these arguments could result in compensating individuals before
any scientific link is 鈥減roven鈥� to the satisfaction of scientists. And yet, if
we ignore them, we are subjecting toxic tort plaintiffs to higher standards of
proof than legal convention allows. Plaintiffs and their lawyers may then be
correct in asserting that some victims are not being compensated for bearing the
burden of the industrial and technological decisions we as a society willingly
take.
One answer might be to look beyond the current adversarial legal process to
science courts and environmental ombudsmen. Another could be to implement
interim judgements where any compensation is returned (either by the individual
or society as a whole) if the science later indicates that no direct casual link
between the symptoms of an individual and the industrial activity ever existed.
Whatever solution is adopted, we should acknowledge that these are mixed
science-policy decisions, where all participants need to be explicit about the
implications of their views.