I’m writing this with a bit of trepidation, as it’s likely to be unpopular. I am hugely sympathetic to the political standpoint of the people whose arguments I’m targeting here. But the philosopher in me can’t bear to witness sloppy reasoning and logical fallacies, even when called into the service of a good and rightful cause. Philosophy is prior to politics, and the pursuit of truth should be prior to any convictions or ideology to which we are committed. It’s for this reason that I want to explain why one very frequently advanced claim is simply false, the product of a basic error in reasoning. This is the claim we most often see in discussions of rape and other sexual offences: that engaging in certain forms of activity or behaviour does not increase one’s likelihood of being a victim of these types of crime.
I have seen many comments along the lines of: “most people who are victims of rape are not drunk when it happens, so it’s not true that being drunk increases your chances of being raped”. Or similarly: “most people are raped by somebody they know, often in their own home, so walking late at night in a deserted area doesn’t make it more likely that you will be raped”. Now, I fully understand and feel the force of the sentiments that motivate this kind of argument. The worry about victim blaming is a legitimate one, and the problem of victim blaming is real and persistent. There are still many people who think that if a woman behaves in certain ways – gets drunk, wears certain clothes, walks home late at night – that she is somehow responsible or blameworthy if she is attacked. That she somehow had it coming, or ought to have expected it. This has serious implications for all manner of questions – how we punish convicted rapists, how we go about tackling rape as a societal problem, etc. And of course, it also has a serious impact on those who are assaulted. Not only are they victims of the first injustice, that of the assault; they are victims of a second injustice, when they are blamed for what happened to them, with all the suffering and pain this causes. So I fully agree that we need to avoid blaming victims for the crimes they have experienced. Unfortunately, the aforementioned argument is not the right way to go about this. It is an invalid argument, built upon a basic error of reasoning, so that its conclusion does not follow, and is probably false.
To be specific: in logical terminology, the error of reasoning that occurs in this argument is the fallacy of denying the antecedent. The fallacy occurs in this form:
If P, then Q.
Therefore, not Q.
This conclusion - not Q – does not follow from the premises; therefore, this argument is not valid. The premises cannot be held to support the conclusion. (That doesn’t mean that the conclusion is definitely not true, either; it just means that we can’t deduce that it is true, from the premises given).
The argument given above, invoked to avoid the danger of victim blaming, can be laid out in this form:
If you are drunk, then you are more likely to be the victim of rape.
Most victims of rape are not drunk.
Therefore, it is not the case that drunk people are more likely to be victims of rape.
Now, it might not be immediately apparent why this is a fallacious argument. This is pretty much the argument that one often hears, and so it may on the face of it seem valid. However, we only have to tweak it slightly and the error becomes obvious:
If you are drunk, then you are more likely to crash your car.
Most people who crash their car are not drunk.
Therefore, it is not the case that drunk people are more likely to crash their cars.
Hopefully now it’s clear why this argument doesn’t work. The fact that the majority of people who are in car accidents were not drunk does not mean that driving while drunk does not increase one’s likelihood of a crash. This argument has the exact same structure as the previous one, and yet we can see that it is invalid; the premises do not support the conclusion.
In the case of drink driving, we know that as a matter of fact, the conclusion is false, because the empirical evidence tells us that the first premise is true – it is a fact that being drunk increases your chances of crashing your car. Now, I should make it clear that I am not stating that the first premise in the argument about rape is known to be true. That too is a question that empirical evidence will support one way or the other. The facts will tell us if it is indeed the case that being drunk or walking home alone late at night or wearing high-heeled shoes or whatever increase one’s likelihood of being victim of an attack (intuitively, some of those seem more likely factors than others).
But the point is this: the fact that most people who are raped were not drunk does not in itself tell us anything about whether getting drunk increases your chances of being raped. We know it is the case that the vast majority of rapes that take place occur within the home, or between people who know each other; and so in those cases, drunkenness is not a determining factor. But from that it just doesn’t follow that if you are wanting to minimize your chances of being the victim of rape, not getting drunk won’t help you. It is an empirical question – but it seems a plausible hypothesis to think that in any given situation, holding everything else equal, being drunk will make you more vulnerable that you would otherwise have been. And if that’s so, then to reject the claim that being drunk increases your chances of being victim of crime is just a mistake, and is an example of putting our politics ahead of the facts of the matter.
It’s crucial to note what is being said here, and what is not. Saying that being drunk makes it more likely that you will be a victim of a crime does not entail anything about whether you are therefore blameworthy for that crime or responsible for it. But to avoid that conclusion, we need to develop a coherent and intuitively compelling theory of blame and responsibility (and this would hopefully be able to make some important distinctions between causal responsibility and moral responsibility).
Equally, saying that being drunk makes it more likely that you will be a victim of a crime does not entail anything about what is the best way to go about eliminating that crime – whether you understand best to mean morally appropriate, most effective, or something else. I completely agree with those people who say that rape prevention far too often places the onus on women to avoid being victims of rape, rather than on targeting men’s behaviour. And I also agree that this is a dangerous move to make because it increases the tendency towards victim blaming. I happen to think that women have a right to get as drunk as they please, wear as short a skirt as they like, and walk home through dark and deserted areas without being assaulted; none of those factors could ever justify somebody assaulting them. But what I do have a problem with is people moving from that essentially moral position to the factual claim that being drunk or walking home alone doesn’t increase your chances of being attacked, because that just doesn’t follow.
I worry about all of this because the perfectly legitimate desire to avoid victim blaming then often becomes an outright attack on people who are trying to give factual advice about how to avoid being the victim of a crime. I can see why analogies with leaving one’s windows open and burglary seem crass and offensive to many people. But as the proponents of those kinds of analogies often stress, it is not a moralized claim; it’s a purely factual one. The empirical question of what makes a certain event more likely to occur is entirely distinct from the moral question of what makes a person blameworthy or not. Telling women not to get drunk probably will, everything else being equal, reduce their chances of being raped. Whether or not this is the morally right way to go about reducing those chances is an entirely different question.