271 
to the lowest level that can reasonably be attained, that the level of risk 
to the public is extremely small, and that the potential benefits are sub- 
stantial and are significantly greater than those risks. This principle has 
been embodied in all new public health and safety legislation for at least 
the past 20 years, and pervades public opinion today." 
Now, in applying that principle, what I found very difficult in the 
discussion of the past two days has been the discussion about risk, because 
on occasion people talk about risk in terms of risk to a single laboratory 
worker, and in the next breath talk about risk of an epidemic to the public 
at large, without distinguishing between the two. I think a good deal more 
thought must be put into articulation of the risks, and while I am not as 
optimistic as some about quantifying them, nonetheless, talking about them 
in terms of risks relating to other things that we daily engage in, such as 
riding in automobiles and flying on airplanes. 
I also think that the risk of not conducting the research must be dealt 
with. I was enormously persuaded by Leon Eisenberg's recent article in 
Science magazine that there is as important an ethical dimension in failing 
to act as there is in taking improper action, and that the failure to seek 
out new forms of medicine, new forms of helping the public, can be far more 
dangerous than engaging in research that might do affirmative difficulty. 
The concern about misuse I find not very helpful to the discussion. If 
one were to be concerned about misuse, I would suppose we ought to ban the 
automobile and the airplane, and certainly the airplane. So that anything we 
have in our society can be misused, and if we were to be concerned about that 
we would do nothing at all. My favorite analogy there is a toxicology friend 
of mine who keeps pointing out that even water is dangerous if you inhale it. 
(Laughter . ) 
The difficulty then is to weigh benefits which are difficult, very 
difficult, to quantify, probably impossible, against risks which are equally 
uncertain. Indeed, when we talk about risks, I think we are actually talking 
more about degree of uncertainty than we are quantifying known risks. 
As long as the burden of proof is on the scientific profession to 
determine or attempt to determine to the best of their ability, the best 
of human ability, safety, we must proceed with caution, and for that reason 
I very deeply believe in the need for Guidelines of this nature and for the 
statement perhaps not of erring on the side of caution, but nonetheless 
proceeding on the side of caution. It largely then becomes in my mind — and 
I speak, of course, from the standpoint of a regulator and a lawyer — it 
becomes a matter of procedural due process, of fairness of making certain 
that it is a public procedure in which all of us can have confidence, includ- 
ing those of us, and I particularly point to myself, who have no scientific 
competence in this area. 
[ 475 ] 
