80 • Alternatives to Animal Use in Research, Testing, and Education 
satisfy the demands of justice (25 , 3 7, 3 9). Since it 
only requires that individuals with interests be 
given the same consideration, but in its summa- 
tion of interests allows the claims of any one indi- 
vidual to be overridden by the sheer weight of 
numbers on the other side, it seems to sanction 
a tyranny of the majority that permits violations 
of individual rights. This may not, however, under- 
mine the utilitarian case for animals if animals have 
doubtful standing as right-holders. 
Some commentators have suggested that there 
may be an acceptable double standard in morals, 
consisting of a nonutilitarian principle for agents 
with standing as persons and a utilitarian rule for 
handling individuals with interests but not rights 
(21,37). The use of different rules for different 
kinds of individuals is already well established. 
Rules that would be objectionably paternalistic if 
applied to adults are admissible if restricted to chil- 
dren. The dangers are that inconsistent standards 
might hold for the same individual or that differ- 
ences between the two classes of individuals might 
be arbitrary. 
The suggestion that the adult-child and human- 
animal distinctions are comparably rational and 
justifiable (21) is superficial for two reasons. First, 
it does not seem to be arbitrary to distinguish be- 
tween the adult and the child, because human soci- 
ety understands that children may be intellectually 
and experientially unable to make wise choices. 
Thus, society can choose for children that which 
society believes is in their best interests . The prob- 
lem with the human -animal distinction is that an 
animal may in fact be able to make and communi- 
cate a decision that expresses the animal’s self- 
interest: It wants no part of any scientific proce- 
dure that results in pain or distress. Even if the 
animal could not make or communicate a decision, 
it may be arbitrary to distinguish between such 
animals and humans who are similar in their in- 
ability to make such decisions (the profoundly men- 
tally handicapped), allowing society to use the 
former but not the latter as research subjects. 
The second difference between the adult -child 
and human-animal distinctions relates to the pur- 
pose for distinguishing between two groups. The 
first distinction is permissible because it allows so- 
ciety to protect the interests of the child, while 
the purpose of the human-animal distinction is to 
allow society to ignore, or at least diminish, the 
interests of the animal. 
The device of a double standard is often used 
to explain the sharp differences in the constraints 
governing the treatment of animals and humans 
as experimental subjects. For animals the stand- 
ard is humane treatment, which forbids unneces- 
sary suffering but otherwise allows experiments 
that harm and even kill the animal. That same rule, 
proposed for human subjects, is generally consid- 
ered unethical. There are many experiments in 
which perfectly reliable results can only be ob- 
tained by doing to a human what is now done to 
an animal. Nevertheless, without the subject’s in- 
formed consent— indeed, sometimes even with it— 
such experiments are absolutely impermissible, 
no matter how beneficial the consequence might 
be. They would violate the rights of the human 
subject. 
The proscription against unnecessary suffering 
is best understood as a corollary of the principle 
of utility. Since suffering is a bad consequence, 
there is an initial utilitarian onus against behavior 
that would produce it. Such treatment calls for 
justification. To meet this burden, a bare appeal 
to some offsetting good consequence will not be 
sufficient. The principle of utility, as formulated, 
is comparative . It requires that an action or policy 
have better consequences than any available alter- 
native. Among the alternatives will be uses that 
do not involve animal suffering. If one of them has 
consequences at least as good as or better than 
the one proposed, the suffering will be unneces- 
sary . Other things being equal, then, it should prove 
harder to establish necessity than the contrary, 
since the former must rule out all the alternatives 
while the latter need find only one. 
Necessity is a relation between a means (an ac- 
tion or policy) and an end (its objective). Restricted 
necessity takes the end as given— that is, not sub- 
ject to evaluation— and asks only whether the 
course of action suggested is an indispensable 
means to that end. For example, in an LD S0 test 
for toxicity that uses 40 rats as subjects (see chs. 
7 and 8), if no alternative procedure using fewer 
or no rats could get the same results with the same 
reliability, that test would be necessary in the re- 
