Animal experimentation: morally acceptable, or just the way things always have been?

Following
the announcement last week that Oxford University’s controversial Biomedical
Sciences building

is now complete and will be open for business in mid-2009, the ethical issues
surrounding the use of animals for scientific experimentation have been
revisited in the media—see, for example, here ,
here,
and here.

The number
of animals used per year in scientific experiments worldwide has been estimated
at 200 million
—well in excess of the population of Brazil and over three times that of the United Kingdom. If we take the importance of an ethical issue
to depend in part on how many subjects it affects, then, the ethics of animal
experimentation at the very least warrants consideration alongside some of the
most important issues in this country today, and arguably exceeds them in
importance. So, what is being done to address
this issue?


In the
media, much effort seems to be devoted to discrediting concerns about animal
suffering

and reassuring people that animals used in science are well cared for,
and relatively little effort is spent engaging with the ethical issues. However, it seems likely that no amount of
reassurance about primate play areas and germ-controlled environments in
Oxford’s new research lab

will allay existing concerns about the acceptability of, for example, inducing heart
failure in mice

or inducing Parkinson’s disease in monkeys—particularly since scientists are not currently required to report exactly how
much suffering their experiments cause to animals
. Given the suffering involved, are we really
sure that experimenting on animals is ethically justifiable?

In
attempting to answer this question, it is disturbing to note some
inconsistencies in popular views of science. Consider, for example, that by far the most common argument in favour of
animal experimentation is that it is an essential part of scientific
progress. As Oxford’s oft-quoted Professor
Alastair Buchan reminds us, ‘You can’t make a head injury in a dish, you can’t
create a stroke in a test tube, you can’t create a heart attack on a chip: it
just doesn’t work’. Using animals, we
are told, is essential if science is to progress. Since many people are apparently convinced by
this argument, they must therefore believe that scientific progress is
something worthwhile—that, at the very least, its value outweighs the suffering
of experimental animals. And yet, at the
same time, we are regularly confronted with the conflicting realisation that,
far from viewing science as a highly valuable and worthwhile pursuit, the
public is often disillusioned and exasperated with science. Recently, for example, people have expressed

bafflement that scientists have spent time and money on seemingly trifling
projects—such as working out the best way to swat a fly
and discovering why knots form—and on telling us things that we already know: that getting rid of credit
cards helps us spend less money
,
and that listening to very loud music can damage hearing. Why, when the public often seems to despair
of science, do so many people appear to be convinced that scientific progress is
so important that it justifies the suffering of millions of animals? 

A pervasive
view is that experiments on animals are necessary in order to find cures for
diseases, and that it is better that animals suffer than that humans
suffer. This view is somewhat confused:
there are doubts about how relevant the results of animal experiments are to
humans (see, for example, here
and here),
and many animal experiments do not attempt to save lives, nor even to discover something worthwhile—for example, a
recent experiment in Japan involved applying electric currents to the nerves of
conscious, temporarily paralysed cats in order to observe the extent to which the cats’
pupils dilated in response
. Even ignoring these worries, is it acceptable
to use animals to find cures for diseases in order to avoid human suffering?

For many,
the answer to this question is likely to be a confident ‘yes’. However, the view that it is acceptable to
inflict suffering on animals in order to spare humans is a particular
application of the more general view that it is acceptable to inflict suffering
on one type of being in order to spare another type. This more general view is widely reviled in
modern times, and when applied within the human species, is associated with
uncivilised and morally corrupt cultures: consider the subjection of Blacks to
Whites in Apartheid South Africa, and the more recent subjection of Whites to
Blacks in Zimbabwe. Is there good reason to suppose that
discrimination on grounds of species is morally acceptable in a way that  discrimination on
grounds of race is not?  Well, in fifteen
years of studying philosophy, I have never encountered a good argument for the
conclusion that so-called speciesism is morally acceptable. Animals may be different from humans in many
respects, but it is their similarity to humans in a particular respect—their
capacity to suffer—that grounds the ethical debate surrounding animal
experimentation.

If it
really is unjustifiable to cause suffering to animals in the name of science,
why are relatively few people concerned with trying to stop it? Why do so many people believe that it is
acceptable, and even essential? The
answer is no doubt due in large part to the fact that humans have always
exploited animals, and that it can be difficult to feel moral outrage in
response to a familiar, ingrained practice. In essence, it is likely that many believe that animal experiments are
justified because they are currently permitted, and because many apparently
respectable people believe that they are justified. Possibly, even, many people recognise that such experiments are horrific, but convince themselves that they must be morally acceptable because it is too disturbing to confront the possibility that they are both unacceptable and permitted.  Similar factors can induce people to accept all
sorts of peculiar, often unjustifiable, things: consider that, during Apartheid South Africa, many white people who had grown up with the regime viewed it as acceptable.  There is, moreover, psychological evidence for the fact that people are far
more likely to accept a state of affairs if it has always been the case than if it is presented as a change from the current way
, even
if there is no independent reason to prefer the current situation (for an overview
of this evidence, and a strategy for avoiding such status quo bias, see here). In addition to such bias, it also seems likely that many
people are unwilling to entertain the possibility that animal experiments are unethical because they do not wish to associate themselves with activists who campaign for an
end to animal experiments, whom they view as mostly disruptive and unreasonable. Whilst this is largely an unfair stereotype, a few such activists are undeniably disruptive
and unreasonable.  Even so, we should take care not to allow our dislike of disruptive and unreasonable people to obscure our view of the important ethical
issue to which they call attention. Recall that, ten years ago, those who campaigned to raise awareness of
climate change were also regularly dismissed as disruptive and unreasonable. Today, however, climate change is recognised
as one of the most important ethical and scientific issues of our age. Could the issue of animal experimentation
receive similar recognition in years to come?

Edit: This post also appears on the Overcoming Bias blog, where it is followed by a lot of discussion.

  • Facebook
  • Twitter
  • Reddit

2 Responses to Animal experimentation: morally acceptable, or just the way things always have been?

  • Andre Menache says:

    No-one is suggesting that complex human health problems can be solved in a petri dish. Rather, it is time that scientists such as Professor Alastair Buchan demonstrate whether animal experiments can actually deliver good science, i.e. are they evidence-based? The vast majority of systematic reviews comparing treatment outcomes in animals and people show a huge discordance in the results obtained between the two. The choice today is between incomplete human data that is relevant to the species in question (humans), or complete animal data that is irrelevant to the species in question, when studying human disease.

  • Dave says:

    Nice post.

Authors

Subscribe Via Email

Affiliations