To: Hon. Ray Thornton 
January 19, 1982 
Page 3 
(2) Risk Considerations 
In general, risk considerations have been poorly understood in the 
continuing r-DNA policy discussions, and risk assessment often not appre- 
ciated at all. The materials produced by the RAC Working Group (46 Fed . 
Register 59385-90) are a better treatment than one has come to expect. A 
risk should be considered in terms of three essential aspects: the prob- 
ability of its occurrence, the magnitude or severity of the hazard should 
it occur, and the distributional nature of the hazard (who bears the bur- 
den). There are four general categories of risk relevant to r-DNA work. 
The first is the unintentional abuse, which because of accident or general 
sloppiness, etc., results in an epidemic or environmental disaster. The 
second is intentional misuse, malfeasance, which is the application of the 
technology with malice and/or the training of persons who may use recombinant 
DNA techniques in such fashion. Third are the moral and ethical concerns, such 
as human interference with natural rules of evolution or whether specific 
experiments used to generate new knowledge are inherently immoral ( in the 
same sense that some Nazi experiments which were evil did produce knowledge 
of utility). Finally, there are the political concerns, including our reali- 
zation that access to new knowledge is very skewed and that economic interests 
play a disproportionate role in shaping the development of technological 
phenomenon. Related to this last category are the risks inherent in very 
competitive modes of scientific endeavor where struggles to be first with a 
new discovery (because of prestigious prizes and acclaim) have, as we shall 
see, led to the cutting of corners. 
Whereas risk may be thought of as a generally objective measure of harm, the 
notion of safety should be understood as being subjective. Safety is the 
level of acceptable risk. Thus, safety depends on the notion of acceptability, 
and presumably involves issues of knowledge and acquiescence on the part of 
those subject to the probability of harm, for instance in the local communi- 
ties. 
Repeatedly over the past several years, scientists have minimized the risks 
in recombinant DNA research. We should understand that they are only consid- 
ering those risks in the first category above, the technocratic view of risks. 
Other risks exist and need to be considered by NIH. 
The NIH Risk Assessment Program has only been in operation for 2 years. 
Unfortunately, both the Final Plan of the Risk Assessment Program announced 
by NIH near the end of 1979, and the Update announced a year later in 1980, 
contain two main flaws. 
[ 676 ] 
