text
stringlengths
0
1.71k
absolute rule, or an assessment of its consequences. Pacifists
have usually regarded the use of violence as absolutely wrong,
irrespective of its consequences. This, like other 'no matter what'
prohibitions, assumes the validity of the distinction between
acts and omissions. Without this distinction, pacifists who refuse
to use violence when it is the only means of preventing greater
violence would be responsible for the greater violence they fail
to prevent.
307
Practical Ethics
Suppose we have an opportunity to assassinate a tyrant who
is systematically murdering his opponents and anyone else he
dislikes. We know that if the tyrant dies he will be replaced by
a popular opposition leader, now in exile, who will restore the
rule of law. If we say that violence is always wrong, and refuse
to carry out the assassination, mustn't we bear some responsibility
for the tyrant's future murders?
If the objections made to the acts and omissions distinction
in Chapter 7 were sound, those who do not use violence to
prevent greater violence have to take responsibility for the violence
they could have prevented, Thus the rejection of the acts
and omissions distinction makes a crucial difference to the discussion
of violence, for it opens the door to a plausible argument
in defence of violence.
Marxists have often used this argument to rebut attacks on
their doctrine of the need for violent revolution. In his classic
indictment of the social effects of nineteenth -century capitalism,
The Condition of the Working Class in England, Engels wrote:
If one individual inflicts a bodily injury upon another which
leads to the death of the person attacked we call it manslaughter;
on the other hand, if the attacker knows beforehand that the
blow will be fatal we call it murder. Murder has also been committed
if society places hundreds of workers in such a position
that they inevitably come to premature and unnatural ends. Their
death is as violent as if they had been stabbed or shot .... Murder
has been committed if thousands of workers have been deprived
ofthe necessities oflife or if they have been forced into a situation
in which it is impossible for them to survive .... Murder has been
committed if society knows perfectly well that thousands of
workers cannot avoid being sacrificed so long as these conditions
are allowed to continue. Murder of this sort is just as culpable
as the murder committed by an individual. At first sight it does
not appear to be murder at all because responsibility for the death
ofthe victim cannot be pinned on any individual assailant. Everyone
is responsible and yet no one is responsible, because it appears
as if the victim has died from natural causes. If a worker
dies no one places the responsibility for his death on society,
308
Ends and Means
though some would realize that society has failed to take steps
to prevent the victim from dying. But it is murder all the same.
One might object to Engels's use of the term 'murder'. The
objection would resemble the arguments discussed in Chapter
8, when we considered whether our failure to aid the starving
makes us murderers. We saw that there is no intrinsic significance
in the distinction between acts and omissions; but from
the point 'of view of motivation and the appropriateness of
blame, most cases of failing to prevent death are not equivalent
to murder. The same would apply to the cases Engels describes.
Engels tries to pin the blame on 'society', but 'society' is not a
person or a moral agent, and cannot be held responsible in the
wayan individual can.
Still, this is nit-picking. Whether or not 'murder' is the right
term, whether or not we are prepared to describe as 'violent'
the deaths of malnourished workers in unhealthy and unsafe
factories, Engels's fundamental point stands. These deaths are
a wrong of the same order of magnitude as the deaths of
hundreds of people in a terrorist bombing would be. It would
be one-sided to say that violent revolution is always absolutely
wrong, without taking account of the evils that the revolutionaries
are trying to stop. If violent means had been the only way
of changing the conditions Engels describes, those who opposed
the use of violent means would have been responsible for the
continuation of those conditions.
Some of the practices we have been discussing in this book
are violent, either directly or by omission. In the case of nonhuman
animals, our treatment is often violent by any description.
Those who regard the human fetus as a moral subject will
obviously consider abortion to be a violent act against it. In the
case of humans at or after birth, what are we to say of an
avoidable situation in which some countries have infant mortality
rates eight times higher than others, and a person born
in one country can expect to live twenty years more than someone
born in another country? Is this violence? Again, it doesn't
309
Practical Ethics
really matter whether we call it violence or not. In its effects it
is as terrible as violence.
Absolutist condemnations of violence stand or fall with the
distinction between acts and omissions. Therefore they fall.
There are, however, strong consequentialist objections to the
use of violence. We have been premising our discussion on the
assumption that violence might be the only means of changing
things for the better. Absolutists have no interest in challenging
this assumption because they reject violence whether the assumption
is true or false. Consequentialists must ask whether
violence ever is the only means to an important end, or, if not
the only means, the swiftest means. They must also ask about
the long-term effects of pursuing change by violent means.
Could one defend, on consequentialist grounds, a condemnation