188 
The first is, Do these plants also contain substances toxic to animals 
that might be likely to eat the plant or its fruit or its seeds? 
Secondly, Does the plant have a competitive advantage over other plants 
and become a pest in itself? Taking as an obtuse example, suppose we were 
able to create the nitrogen-fixing grass plant. Well, will it become an- 
other crab grass, or Johnson grass? I think we need to test these organ- 
isms in a controlled manner under conditions where they can still be con- 
tained before we release them to the biosphere, where once they are out 
it may well be too late. 
DR. FREDRICKSON: Dr. Callahan? 
DR. CALLAHAN: For me this problem comes down to a rather classic case 
of how does one establish proper policy bias? By bias I mean an inclination 
to go one way or the other. It seems to me here is the perfect example, at 
least as a non-expert listening to arguments. I feel my problem is not to 
somehow stand back and be objective, because I don't know where the base for 
objectivity lies. In particular cases, yes, but the base for objectivity 
concerning the possible benefits and possible harms it is very hard. One 
has to be imaginitive there, and hence the question is, What ought to be 
the policy bias? 
Now, I have heard nothing in two days to suggest that there is any 
moral obligation to do this research at all, that somehow we ought to do it, 
that somehow we would be remiss to society for the future not to do it. I 
have heard lots of arguments that it would be morally desirable to do it, 
morally permissible to do it, morally commendable to do it, but not morally 
obligatory to do it. 
'i 
I think there can be no moral obligation to do this research. It seems 
to me that there is, however, a moral obligation to do no harm. That is a 
much easier case to establish, that one ought not to hurt other people, or 
run the kinds of risks which put people in danger. 
Now, I think, given the uncertainty in this case, given the fact that 
one ought to do no harm, but that one may try to do some good, one probably 
ought to incline in the direction of caution. Namely, the benefit of doubt 
should go to the worriers, and not to those who are optimistic that problems 
will be solved. If it seems to be a toss-up of technical opinion, if it 
seems to be an unresolvable argument whether the worriers or the non- 
worriers are right, I would say give the bias to the worriers. 
It seems to me as one goes thorugh the particular elements of the 
guidelines that might be a workable principle to maintain. 
Now, I might, finally, having said all that, say I think that the 
guidelines are going in the right direction. I would not want to push the 
[329] 
