The Linear Hypothesis
It is very difficult to measure the incidence of cancer and other
ailments caused by low level radiation. This is because the rates are
very much lower than incidence of cancer from other (more usual)
causes. One way of estimating the incidence is to give animals high
doses of radiation, count how many cancers occur and assume that the
number at low doses is proportionately lower. This is called the
no threshold linear hypothesis.
The most thorough studies of the linear hypothesis have been
made by Professor Bernard Cohen of the the Physics Department of the
University of Pittsburgh in the US. He has found much evidence
against it and describes it in his article
Validity of the Linear No-Threshold Theory of
Radiation Carcinogenesis at Low Doses.
The Uranium Institute in
London, England has links to additional studies on its web page.
Center for Nuclear Technology and Society
at Worcester Polytechnic Institute offers much additional evidence
against the linear hypothesis.
The linear hypothesis is unsupported by any direct evidence, and
there is a lot of evidence against it. Here is some of it.
The rival to the linear hypothesis is the threshold hypothesis. The body
repairs damage to cells from ionizing radiation. According to the threshold
hypothesis radiation results in cancer only when there is enough radiation
to overwhelm the threshold mechanism.
- The Atomic Bomb Casualty Commission composed of Japanese and American
scientists has kept track of Hiroshima and Nagasaki survivors and estimated
how much radiation each received. According to the linear hypothesis,
these survivors should have had more cancer than has been observed.
- People living at high altitudes or at locations in India with high
natural thorium levels should have had more cancer than has been observed.
- People in areas with high emissions of radon from granite and other
rocks should have more cancer than has been observed. The EPA's estimate
of the number of lung cancers caused by radon is based on the linear
hypothesis and is almost certainly a large exaggeration. If so, the
very expensive remedial measures proposed are a waste of money that
would better be spent on getting rid of more serious threats to
One could mix these hypothesis and assume that there is some linear component
but that the major effect is a threshold effect. The above evidence suggests
that the linear component is very small.
The linear hypothesis was chosen for regulatory purposes because (a) it is
extremely conservative and (b) it isn't hard for the nuclear industry to meet
standards based on it. It is politically very difficult to change it, because
anti-nuclear organizations will accuse the regulatory authorities of
jeopardizing public health.
The high estimates of cancer from the Soviet Chernobyl accident are based on
the linear hypothesis.
Up to: Nuclear FAQ
Send comments to firstname.lastname@example.org. I sometimes make changes
suggested in them. -
The number of hits on this page since 1996 December 27.