Imagine three coffee houses in Toronto. Everyone is sitting around happily when something disturbs the peace.
At Starbucks 1, a militant dressed in black rushes up the window and breaks it with a stick. People are shocked, and frightened – they experience this as an outbreak of violence.
At Starbucks 2, a militant dressed in black comes in and buys some coffee. Since the militant wants to sit down, the staff tell a hobo, who had come in to sit in an armchair and rest, that he’ll have to leave – when he protests, they become belligerent, badger him and raise their voices. Eventually he leaves, but he feels coerced, threatened – he experiences this as an outbreak of violence.
At Starbucks 3, a homosexual couple come in, holding hands and kissing. People are shocked, and frightened for their children who have to watch this – they experience this as, in a somewhat archaic phrase, ‘violence against nature’, and feel violated and threatened.
Of these various perceptions of ‘violence’, which ones are right? Which of these events are really violent? Suppose we sufficiently understand the psychology behind this type of experience: how do we get to objective applications of the word?
Now strictly, we can define words however we like – I could use the sounds ‘vi-o-lence’ to designate lizards. But usually we already know some things that fix the meaning of the word being defined. The main ‘already known’ thing about ‘violence’, for me, is the principle that non-violence is a precondition for many valuable relationships, in particular political dialogue. Violence is what we exclude, in order to open up a space where disagreements can be aired safely and legitimately.
This principle looks nice on paper. But can it be given any objective meaning?
The simplest sort of objectivity is when there’s a feature of the world, outside people’s present experience, which ‘corresponds to’ that experience. For instance, most people tend to think that spatial arrangements are objective, so if I perceive things as being arranged in a line, my perception is accurate if they are, in themselves, arranged in a line.
How far this works is a vexéd philosophical question, but it seems very unlikely that it would work for violence. If my last post is correct, violence is defined in terms of embodiment, and embodiment is defined psychologically. No physical fact intrinsically corresponds to it.
So strict objectivity isn’t available; but we’re often fine without strict objectivity. For instance, colours are not strictly objective: real objects don’t have redness or blueness, but surfaces that reflect and absorb different combinations of wavelengths under different conditions, which different eyes interpret in different ways. But this doesn’t stop us talking objectively about colours. How do we manage this, and can the same strategy be applied to violence?
It helps that we understand fairly well how colour-experiences arise from facts about light and retinas. This allows us to identify a set of ‘normal conditions’ and label variations on this as ‘errors’. For instance, if an object looks green in white light, but looks black in red light, we say that it’s ‘really’ green, and ‘only looks’ black, because it’s ‘normal’ to see something in (approximately) white light.
But this requires the ‘normal conditions’ to be sufficiently common and to ensure sufficiently close agreement. With other things, like preferring certain tastes, there is some agreement (chocolate tastes nicer than ink) but just as much disagreement, which is evenly spread and doesn’t go away. Consequently, we can’t really talk seriously about objectively better or worse tastes.
So the question becomes, is experiencing-as-violence more like colour, or more like taste-preferences? Equivalently, does everyone experience themselves as embodied in roughly the same ways, or in irreducibly varied ways? And it seems that it’s half-and-half.
On the one hand, experiencing yourself as embodied in your limbs, torso, head, and surrounding space is more or less universal. And just as colour is relatively ‘objective’ because of the determinate physiology of the human retina, feeling violated when one’s body is harmed or controlled is relatively objective because of the physiology of pain receptors, vital organs, and skeletal muscles.
On the other hand, perceptions of non-bodily violence seem to be irreducibly varied, because it’s largely culture that makes us feel embodied in external things, and it does so in a great range of unpredictable ways, as the story of the Three Starbucks illustrates.
This means that ‘objectively real’ violence will, on any reasonable standard, include bodily violence. But regarding other sorts there are three options:
1) Some, but not all, experiences of non-bodily violence are accurate (some, but not all, ways of experiencing oneself as embodied beyond one’s body are appropriate).
2) All actions that are perceived as violent really are violent; nobody’s experience should be disqualified or overruled.
3) No actions count as violent unless they violate someone’s body, even if someone experiences them as violent.
How can we decide between these different definitions? Recall the principle ‘non-violence is a precondition for political dialogue’: we have good reason to reject definitions of violence that render this principle either contradictory or unworkable. I would argue that definition 1 renders it contradictory, and definition 2 unworkable.
Definition 1 basically says that prior to entering political discourse, we must first accept a certain system of rights and rules, refrain from breaking them, and moreover accept, as consistent with non-violence, any ‘defensive’ bodily violence taken to enforce them.
But such sets of rights and rules are exactly the sort of thing that political dialogue is concerned with. So to define violence in this way amounts to saying that ‘doing things our way is the precondition for discussing which way to do things’. It presupposes an answer to the question it’s supposed to allow us to debate. And it mystifies the constant threat and frequent reality of law-enforcing violence, makes it appear non-violent and pre-political.
What about definition 2? This sounds good on the surface – very tolerant, very respectful of differences. But it’s impossible to apply fully – consider the Three Starbucks. Can we make a rule that stops people violating property rights when that would be felt as violent, but also stops people asserting them, even verbally, when that would? And a rule which simultaneously avoids violating the sensibilities of homophobes and homosexuals? And all varities of religious and secular wacko? And then make this walking-on-eggshells rule a precondition for legitimate disagreement?
Even worse, violence legitimises counter-violence: if a cartoon or a broken Starbucks window can be ‘violence’, that mitigates bodily violence against blasphemers and vandals.
So if we want a meaningful, consistent definition, my vote goes to definition 3: violence is only that which attacks someone’s body, in any of various ways – harm, pain, restraint, intrusion. What affects noone’s body may be wrong in various other ways, but is not violence.
It follows from this that the events that initially prompted my reflections, Saturday’s protests in Toronto, were, except in a very few minor cases, non-violent. This isn’t a complete vindication of the black bloc – their actions might be open to many other forms of criticism, but not this one, which happens to be particularly rhetorically effective.
But the fact that “violence!” was people’s reflexive response indicates that by and large, people are using a version of definition 1: violence is what attacks either the body, or any other object in which people feel embodied, as long as that feeling of embodiment is in accord with the law. This means that our judgements of ‘violence’ are already highly political, even ideological, and this undermines the coherence of the oft-repeated principle that non-violence should be a precondition for political dialogue.