I spent yesterday’s post defending Rousseau’s concept of the sovereign and the general will: today I want to attack it. In particular, I want to attack the apparently very unequal relationship he presents between the sovereign and the subjects (that is, between the group-as-political-unit and the individuals who compose it).
Firstly, “a public decision can impose an obligation on all subjects towards the sovereign…while, conversely, such decisions cannot impose an obligation on the sovereign towards itself; and hence it would be against the very nature of a political body for the sovereign to set over itself a law which it could not infringe.”
Secondly, “as the sovereign is formed entirely of the individuals who compose it, it has not, nor could it have, any interests contrary to theirs; and so the sovereign has no need to give guarantees to the subjects, because it is impossible for a body to wish to hurt all of its members…But this is not true of the relation of subjects to sovereign…subjects will not be bound by their committment unless means are found to guarantee their fidelity.”
So on the one hand, the sovereign need not bind itself to any guarantees, and cannot do so anyway; on the other hand, individual subjects can be thus bound, and must be. Even when Rousseau grants that subjects are entitled to expect the sovereign to not interfere with those parts of their life that affect nobody else, he swiftly adds that “the sovereign alone is judge of what is of such concern.”
Note, I am still assuming that certain problematic issues have been solved – I’m assuming that agreements are unanimous and expressed by direct votes (or, rather, if unanimous, by consensus-decision-making). Even in this case, though, why is it that this group-individuals relationship, which is after all one of identity (the group is the individuals), seems so skewed? I want to argue that it shouldn’t be – and that Rousseau’s reasons for holding it to be are contradictory.
So firstly, Rousseau says that while “a public decision can impose an obligation on all subjects…while, conversely, such decisions cannot impose an obligation on the sovereign.” Now this is trivially true, because he specifies that he is speaking of public decisions – i.e. decisions made by the sovereign themselves. So if the group can make the decision, it can un-make it. But the real question is whether private decisions can impose obligations on the sovereign – can I, just by deciding to do X, impose an obligation on the rest of society, for example to allow me to do X, or prevent others from preventing me?
That is, are there any personal freedoms or rights that are specifically social, and yet which do not depend on society deciding to endorse them? Rousseau never poses the question in these terms, but from this neglect we can conclude that his answer is ‘no’. But I think there are things within his system that should lead him to say ‘yes’.
Firstly, note that he does say this: “since the…sovereign, owes itsbeing to the sanctity of the contract alone, it cannot commit itself…to anything that would derogate from the original act of association.” But his only examples are external acts – “it could not…alienate a part of itself or submit to another sovereign.”
Secondly, note that although he talks constantly of ‘the general will’, he never mentions ‘the general intellect’. But it seems that you can’t have one without the other: to determine what I desire, what I decide, which options I prefer, I need to think about it, and the way I think about matters can change what desires I have towards them. At least, I would think so.
Now, if an individual intellect works primarily by analysis and synthesis, by distinguishing and connecting, in short by the relations of idea, then the general intellect should work primarily by the relations of different people’s ideas, i.e. by communication. That’s fairly plausible, isn’t it: if the group is to form a single ‘artificial person’ then they must be able to discourse and dialogue so that this person can have an ‘artificial thought-process’ to direct its ‘artificial limbs’.
But then it follows that anything which sets limits to this thought-process is an intimate injury to that collective person, a sort of destruction of part of its brain. Hence, arguably, social rules which specify certain opinions or ideas as inadmissible for expression (or even for belief) should be, by Rousseau’s lights, inadmissible and automatically illegitimate.
The details of this argument, and ‘how far it should extend’, are tricky (and very familiar) arguments that I’ll ignore. The point is, Rousseau’s system does not in principle pose any impediment to holding at least one individual freedom (free thought and expression) to be an entirely fundamental ‘first principle’ of political systems. The fact that Rousseau doesn’t say this is his own choice.
The second issue I want to look at is the question of securing obedience by individuals to collectively-decided rules. Here Rousseau has no hesitation in affirming the group’s right to harm, imprison, or kill individuals. But the legitimation of this actually follows two opposite lines.
On the one hand, the offender is punished because “it is tacitly implied in [the social contract] that whoever refuses to obey the general will shall be constrained to do so by the whole body” – i.e., because they agreed to it. Here the emphasis is on the individual’s ongoing participation in the social contract. On the other hand, they are punished because “by violating [society’s] law, he ceases to be a member of it; indeed, he makes war against it.” Here the emphasis is on the individual’s removing themselves from the social contract.
Obviously the significance of this depends heavily on whether individuals have a right to defect from the social contract that they have entered. If they do, after all, then why would criminals not simply defect immediately prior to committing their crimes? But if they don’t, can it really be fair to hold people, at the age of, say, 60, to an agreement they made at the age of 18? Hell, I would feel a little harsh for holding a 25-year-old to their 18-year-old word. People do, after all, change. In important ways, people are not the same individual at all points in their life, so we shouldn’t give a youth the freedom to rob all freedom from an aged person in the future.
Moreover, I argued last time that the social ‘contract’ isn’t really a contract in the sense of an exchange of promises, but more like a collective act of permission, granting the right to balance harms and benefits between people as well as within them. This might strengthen the impression that it should be something that can be withdrawn from – if someone can give permission, why can they not take it back? Perhaps Rousseau is relying on his doctrine that in entering the social contract, each member gives up ‘all their rights’ to the group – but I argued that he has very weak grounds for holding this. Yet if the permission can be withdrawn, how are the costs of society to be imposed on anyone, since individuals can, if pushed, slip out of the social contract?
This is somewhat paradoxical, isn’t it? I don’t know how to decide for one side or the other, but I do think that there may be ways to stop the conflict from arising. That would be to design the system of penalties so that the legitimacy of inflicting the penalties is independent of whether someone is ‘still a party to’ the social contract.
This would involve two sides. On the one side, it must be legitimate to deprive a non-member of the benefits and privileges of membership – to simply refuse to provide them with certain services, refuse to help them in certain ways. This (along with the accompanying social costs) would be potentially a strong deterrent, given how dependent individuals are on society’s help, and would be unquestionably legitimate (even if this legitimacy encounters limits, e.g. providing life-saving treatment etc).
On the other side, it would be madness to say that certain acts – such as physical assault, restraint, and injury – can never be naturally legitimate. Most obviously, they seem to be legitimate when directed at those who have themselves committed equally severe violations of others. Only a tiny minority would deny that it can be right to use violence to defend yourself or others from violence. If that’s the case, then the justification of severe punishment becomes much easier when the crime is equally severe, in particular when it’s violent crime.
So a system of penalties that used refusal of benefits for lesser crimes and ‘naturally just’ counter-violence for violent crimes would avoid the need to argue over whether and for how long people give up their rights to society. It also happens to be what I’ve argued for on-and-off for the last year or so.
Of course it carries certain constraints. It would make it impossible, for instance, to have non-violent crimes punished violently – to have the death penalty for adultery, say. But that seems to me a good thing. Admittedly there are some tricky questions about how the boundaries of, for instance, ‘violence’ are drawn (e.g. does poisoning count? if so, what else?) But the rough principle – no violent penalties for non-violent crimes – seems to be a good and valuable one.
So the upshot of what I’ve been arguing here is that Rousseau’s picture, where the group enjoys a marked superiority over individuals, does not follow necessarily from his principles or from the notion of the general will/sovereign. A view which includes, as entirely fundamental principles of political life, things like a right to free speech and a strict restriction on violent punishments (such as imprisonment), fits just as well, if not better.