APPENDIX Q 
THE NEW ENGLAND JOURNAL OF MEDICINE 
must) i)e stopped. My first objective is to formulate al- 
ternatuc, plausible major premises of such syllo- 
gisms. 
Principles of prohibition must pertain either to the 
product or to the process of the research in question. 
I'he line between the two may be hard to draw, but 
under one or the other can be listed all the general 
principles that seem remotely tenable. 
Set A: By-product (n>1-i!<3) 
(1) Research should not be permitted when it aims 
at (or is likely to result in) the discovery of 
knowledge that is wrong for human beings to 
possess. 
(2) Research should not be permitted when it aims 
at (or is likely to result in) the discovery of 
knowledge that is not wise to place in hum.an 
hands. 
(2a) When there is any probability that the 
knowledge developed will be used with 
very injurious consequences. 
(2b) When there is moderate probability that 
the knowledge developed will be used with 
very injurious consequences. 
(2c) When there is high probability that the 
knowledge developed will be used with very 
injurious consequences. 
(3) Research should not be permitted when it 
aims specifically at the development or perfec- 
tion of instruments for killing or injuring hu- 
man beings. 
Set B; By-process 
(-1) Research should not be permitted when it is not 
conducted openly, for all to examine the ongo- 
ing process and results. 
(5) Research should not be permitted when its con- 
tinuation is unfair, either to the subjects of the 
experiirientation or to those otherwise involved 
in trie research. Research is unfair when, 
through coercion or deceit, or in some other 
way, the rights of those involved are not re- 
spected. 
(b) R.esearch should not be permitted when its con- 
duct (as distinct from its pioduct) presents risks 
so great as clearly to outweigh the benefits rea- 
sonably anticipated. Risks here must be under- 
stood to include all hazards that that process of 
inquiry entails, of which there are two large cat- 
egories: risks of “misfire” (i.e., achieving re- 
sults different from and more dangerous than 
those sought) and risks of “accident” (i.e., un- 
foreseen mishap during the process). 
(6a) When the risks are essentially to persons 
involved in the research: 
(6a i) the subjects of experimentation: 
(6a2) the researchers and ilieir associates. 
(6b) Where the risks are essentially to others 
than tliose involved in the research. 
These exhaust the alternative principles that are at 
all reasonable, or arguably tenable, for the prohibi- 
tion of research. Phrasing (and the degree of specifi- 
city) may vary, but every plausible candidate, I con- 
tend, will fall under one or another of the six kinds dis- 
tinguished. 
As an illustration: Robert Sinsheimer, professor of 
biophysics at the California Institute of Technology 
and an acute critic of recombinant DNA research, has 
suggested some possible answers to the question, “For 
what specific purposes might we wish to limit in- 
quiry!’” He proposes, or at least entertains, several 
candidates.' 
(a) To preserve human dignity. “We should not do 
experiments that involuntarily make of man a means 
rather than an end.”' But. of coarse, human subjects 
in medical experiments are means. Sinsheimer surely 
intends to emphasize, with Kant, that one must not 
treat human beings as means only, but also as ends — 
for which reason research committees do rigorously 
insist that their participation be truly voluntary and 
inform.ed. This is but a more elaborate statement of 
principle (5), above, demanding fairness. 
(b) To avoid “involuntary physical or biological 
hazard.”' Sinsheimer recognizes, of course, that one 
cannot avoid all such hazards. He wants us all to be 
very sensitive to the level of danger. This is Isut anoth- 
er formulation, less precise, of principle (6), above, 
addressing the balance of risk and benefit. 
(c) Cost. “[One] asks if the primary consequences 
of the inquiry, that is, the knowledge to be gained, 
is worth the expenditures of talent, time and re- 
sources.”' But refusing to support research is one 
thing, prohibiting it is another. When the research en- 
terprise involves heavy expense, one is right to insist 
that the worth of the knowledge to be gained be very 
carefully estimated. Protecting free inquiry does not 
entail irrationality in the expenditure of resources. 
And it is true that institutional refusal to fund a re- 
search project often has the effect of blocking that pro- 
ject in that context. Many research activities, howev- 
er (work with recombinant DNA technology being 
one important example), do not require very large in- 
vestments of institutional or governmental funds. Al- 
though one may conclude, therefore, that .for some 
projects the object does not justify the expenditure, it 
is essential to see t.hat cost cannot serve as grounds for 
prohibition. Some research is simply not worth do- 
ing, but reasons for not troubling to do certain things 
ourselves must not be taken as reasons for keeping 
other persons from doing them. 
Some principles, very different from those stated 
above, are procedural, requiring special machinery for 
the approval of research protocols in certain areas. 
Approval (some say) must be given by the appropri- 
ate bodies, with appropriate membership, deliberat- 
ing with appropriate care. Others say that the deci- 
sion to permit research of certain kinds may be made 
only by some larger community (city or state) through 
some democratic voting procedure. Such regulation, 
although awkward, is being tried in some quarters. 
But, even wlicn feasible and fitting, these procedural 
Appendix Q--2 
