STANFORD UNIVERSITY 
STANFORD, CALIFORNIA 94303 
DEPARTMENT OF BIOLOGICAL SCIENCES 
4 January 1982 
Dr. William Gartland 
Director, Office of Recombinant DNA Activities 
National Institutes of Health 
Building 31, Room 4A52 
Bethesda, Maryland 20205 
RE; Proposed Revisions, NIH Guidelines 
Dear Bill, 
As you might expect, I prefer the "RAC Proposal: to the "Gottesman Proposal" and the 
"Gottesman Proposal" to the Current Guidelines. However, I am not entirely happy 
with the elimination from both proposals of any reference to large scale 
operations. As my views about large scale work are complicated, I will return to 
that subject at the end. 
In considering the whole issue, I hope that RAC members will ask themselves, 
"Suppose that the Berg letter had never been written, the Asilomar Conference had 
never been held, and the NIH Gulldellnes did not yet exist. Could anyone justify 
introducing mandatory Guidelines at the present time, on the basis of rational 
concerns?" I can't imagine how. 
Certainly, no one can deny the possibility that some day some artificial recombinant 
will harm somebody or something. But that fact by itself does not imply that a 
system of prior review is desirable unless we can see some way of anticipating the 
danger and Identifying the culprit in advance. The typical RAC scenario, repeated 
may times during the last year I was a member, went something like this: Dr. X was 
to clone in bacterium Y, which is not an approved host. She submits data showing 
that bacterium Y (like all known natural species) is limited in its ability to 
spread, multiply and take over the natural world; and also that her laboratory 
strain (like other known laboratory strains) is at a competitive disadvantage to the 
wild strain, when tested in the wild. RAC reviews the data and grants a lowering of 
the required containment level. A lot of paper has been pushed, but has this really 
helped in identifying any potential hazards? "Dreaming the impossible dream" may be 
an admirable quest; but "predicting the unpredictable risk" is not a very practical 
charge for a committee. Yet RAC's mission has evolved more and more in that 
direction. 
Now to large scale work and related matters: It is unfortunate that RAC's consensus 
on the desirability of a major reduction in regulation coincides with an explosive 
Increase in the rate of industrial development of recombinant DNA technology. This 
certainly invites the interpretation by the public (or at least by journalists, who 
may register public opinion either present or future) that the reduction is 
contrived for the benefit of private economic Interests. RAC might respond with a 
"go slow" approach, maintaining an image of caution by adopting a more conservative 
option such as the "Gottesman proposal." I do not favor such a response, which 
[ 632 ] 
