The task of a critique of violence may be defined as setting out its relationship to law and justice. Because violence in the succinct sense of the word is attained only by a cause (however effective) when that cause intervenes in moral relations. The sphere of such relations is characterized by the concepts of law and justice. So far as the first of these is concerned, clearly the most elementary, fundamental relationship in any legal system is that of end and means. It is also clear that violence is initially to be sought only in the field of means, not of ends. Stating these facts implies more than may appear (plus of course different things) regarding the critique of violence. The fact is, where violence is a means, a criterion for its critique might quite simply seem given. It becomes glaringly apparent in the question whether, in specific instances, violence is a means to just or unjust ends. Its critique would therefore be present by implication in a system of just ends. However, it is not. For what such a system (assuming it was secured against all doubt) would imply is not a criterion of violence itself as a principle but one for instances of its being used. It would remain an open question whether violence generally, as a principle, was ethical even as a means to just ends. Answering that question calls for a more detailed criterion, a distinction within the sphere of means themselves, irrespective of the ends those means serve.
Dismissing this more precise critical problem typifies a major direction in philosophy of law as possibly its most striking feature: natural law. This no more sees a problem in using violent means to just ends than a person will have a problem with the ‘right’ to move his body in the direction of the goal he is working towards. In this view (which supplied the ideological basis for terrorism in the French Revolution), violence is a natural product, a kind of raw material, the use of which is quite unproblematic, except where such violence is misused for unjust ends. If according to the political theory of natural law people forgo all their violence in favour of the state, this happens on the premiss (which Spinoza, for instance, sets out expressly in his Tractatus Theologico-politicus) that the individual, in and of himself and prior to concluding so sensible a contract, also exercises de jure any violence that he possesses de facto. It may be that these views received a late boost from Darwin’s biology, which in a thoroughly dogmatic manner, in addition to natural selection, sees violence alone as not simply the original means but also the only one up to performing all of nature’s vital ends. Post-Darwinian popular philosophy has frequently shown how small a step it is from this dogma of natural history to the even coarser dogma of philosophy of law whereby the violence capable of fulfilling natural ends virtually unaided is for that reason also quite legitimate.
This theory of natural law, according to which violence is a natural given, is diametrically opposed to the positive-law theory of violence as a product of history. If natural law can assess all existing law only on the basis of a critique of its ends, its positive counterpart can assess all emergent law only on the basis of a critique of its means. And if justice is the criterion of ends, legitimacy/lawfulness is that of means. However, this conflict notwithstanding, the two schools come together in a shared basic dogma: just ends may be attained by justified means, justified means employed for just ends. Natural law seeks through the legitimacy of ends to ‘justify’ means, positive law seeks through justified means to ‘guarantee’ the legitimacy of ends. The antinomy would prove irresolvable if the shared dogmatic premiss is false, if justified means on the one hand and just ends on the other are in irreconcilable opposition. However, no understanding of this problem could ever arise until the circle had been breached and mutually independent criteria for just ends as well as for justified means drawn up.
The realm of ends (hence also the question of a criterion of justice) can be ruled out for the time being so far as this study is concerned. By contrast, the question of the justification of certain means that constitute violence is central to it. Principles of natural law cannot decide it; they will lead only to bottomless casuistry. For if positive law is blind as regards the absolute nature of ends, so is natural law as regards the contingent nature of means. On the other hand, positive legal theory is acceptable as hypothetical base from which to launch this study because it fundamentally distinguishes types of violence, irrespective of instances of the use of violence. That distinction falls between historically recognized, so-called ‘sanctioned’ violence and violence for which there is no sanction. If the following considerations proceed from it, the implication cannot of course be that given violences are classified according to whether or not they are sanctioned. For in a critique of violence its criterion under positive law cannot be applied but instead only assessed. The question at issue is what consequences (regarding the nature of violence) flow from the fact that such a criterion or such a distinction is at all possible so far as violence is concerned – in other words, what is the point of the distinction? Because that such a positive-law distinction does have a point, is perfectly justified in itself, and can be replaced by no other distinction will emerge soon enough; at the same time, light will thus be cast on the sphere that represents the only place where that distinction can take place. Briefly, if the criterion that positive law erects for the legitimacy of violence can be analysed only in terms of its point, the sphere of its application must be criticized in accordance with that sphere’s value. The standpoint for such a critique needs to be found outside positive philosophy of law but also outside natural law. How far only legal discussion pertaining to the phil osophy of history can supply this will become apparent.
The point of distinguishing between lawful and unlawful violence is not immediately obvious. Most emphatically, the natural-law misapprehension needs to be rejected, according to which that point consists in drawing a distinction between violence for just ends and violence for unjust ends. It has been suggested already that positive law demands that any violence produce proof of its historical origins, which in certain conditions will constitute its legitimacy, its sanction in law. Since acknowledging legal violences manifests itself most palpably in a fundamentally resistless bowing to their ends, a hypothetical principle of classification of violences must be founded on the existence or absence of a general historical acknowledgement of those ends. Ends that lack such acknowledgement may be called natural ends, the others legal ends. And in fact the varying function of violence, depending on whether it serves natural or legal ends, can most graphically be demonstrated on the basis of specific legal circumstances. To simplify matters, the following remarks should be seen as relating to those currently obtaining in Europe.
So far as the individual as legal subject is concerned, the European legal situation is characterized by the tendency not to admit the natural ends of such individuals in all cases where such ends might, in a given situation, be more appropriately pursued by violence. In other words, in all areas in which individuals might more appropriately pursue ends by violence, the European legal system insists on establishing legal ends that can only, in fact, be brought about in this way: namely, by legal violence. Indeed, it goes further, insisting that even areas in which natural ends can, in principle, to a great extent be given their heads (education, for instance) – that even such areas be restricted by legal ends so soon as those natural ends are striven for with an excessive measure of violence, as that legal system does in laws governing the limits of the educational right to inflict punishment. It can be said to be a general maxim of current European legislation that all natural ends pursued by individual persons inevitably come up against legal ends when pursued with a greater or lesser degree of violence. (The contradiction posed by the right to practise self-defence in this respect will no doubt find its own explanation in the course of the ensuing considerations.) It follows from this maxim that the law sees violence in the hands of individuals as threatening to undermine the legal system. As threatening to frustrate legal ends and law enforcement, perhaps? Oh no; because then what would be condemned is not violence pure and simple but only violence employed for unlawful ends. A system of legal ends cannot survive, it will be said, if somewhere natural ends may still be striven for with violence. However, in the first place that is merely a dogma. To counter it, one may have to consider the surprising possibility that the interest of law in monopolizing violence so far as the individual is concerned may be accounted for not so much by any intention of safeguarding legal ends, far more by the intention of safeguarding law itself. The possibility that violence, where it does not lie in the hands of the relevant law, constitutes a threat thereto, not because of the ends it may pursue but through its very existence outside the law. More drastically, the same supposition may be evoked by recalling how often the figure of the ‘great’ criminal, however repellent his ends may have been, has elicited the secret admiration of the people. Only one thing can account for that: not what he did but the violence to which it bears witness. In this case the violence that present-day law seeks to take from the individual in all fields of activity really does assume threatening proportions, stirring up, even in defeat, the feelings of the majority against the law. Through which function violence can with reason seem to constitute such a threat to law and how greatly law can live in fear of it must be apparent precisely where, even under the current legal system, deploying violence is still admissible.
This is primarily the case in the class struggle, where it takes the form of the workers’ guaranteed right to strike. Aside from actual countries, organized labour is nowadays the only legal subject possessed of a right to violence. Of course, against this view it may be objected that a cessation of activity (not doing anything, which is what a strike ultimately is) may not be described as an exercise of violence in the first place. No doubt such thinking made it easier for governments to grant the right to strike, once this could no longer be avoided. The grant is not without restrictions, however; it is not absolute. Not doing something, even withdrawing a service, where it simply equates to a ‘breakdown of relations’, may constitute a wholly non-violent, uncontaminated means. And as in the view of the state (or of the law) the labour force’s right to strike does not include both a right to violence and a right to evade violence as wielded indirectly by the employer, so of course from time to time a strike may occur that corresponds to this and is intended simply to proclaim a ‘rejection’ of or ‘alienation’ from the employer. However, the violence factor certainly does enter into such a cessation of activity (taking the form of blackmail) where the cessation occurs in a state of readiness, in principle, to resume the activity as before on certain conditions, whether these have nothing to do with that activity or simply modify some superficial aspect of it. And in this sense the right to strike, from the viewpoint of the workers (which is opposed to that of government), constitutes the right to use violence to attain certain ends. The fact that the two views clash comes out very clearly when we look at the revolutionary general strike. In this the workers will invariably appeal to their right to strike, while the government, dismissing such an appeal as improper on the grounds that the right to strike was not meant ‘that way’, will enact its own special measures. For it is quite at liberty to say that simultaneous strike action in all firms, when the special provocation allowed for by the legislator is not universally present, is unlawful. In that difference of interpretation, what comes out is the substantive contradiction in the legal situation whereby the government acknowledges a kind of violence towards whose ends, as natural ends, it is at times indifferent but in an emergency (the revolutionary general strike) downright hostile. The fact is, paradoxical though this may seem at first sight, even conduct adopted in exercising a right will still, in certain circumstances, deserve to be called ‘violence’. Moreover such conduct, where it is active, may merit the name ‘violence’ when using a right in its possession to overthrow the legal system that bestowed that right; where it is passive, on the other hand, it is no less deserving of the same name if, in the sense of the argument unfolded above, it constitutes blackmail. That is why we see evidence only of a substantive contradiction in the legal situation, not of a logical contradiction in law, if in certain circumstances the law resorts to violence to counter strikers who deploy violence. For it is here, in strike action, that the government fears more than anything that particular function of violence that this study seeks to identify as offering the sole reliable basis for a critique thereof. The fact is, were violence (as it seems to be at first) simply the means of securing immediately whatever happens to be being striven for, the only way in which it could achieve its purpose would be as pillage [raubende Gewalt]. It would be quite incapable of establishing or modifying circumstances in a relatively stable fashion, Strike action, however, shows that it can do so, that it is capable of establishing and modifying legal relations, much as this may offend any sense of justice. There is a temptation to object that such a function of violence is random and occurs rarely. A look at martial violence will refute the objection.
The possibility of a martial law, a ‘law of war’, rests on precisely the same substantive contradictions in the legal situation as that of a right to strike – namely, that legal subjects sanction instances of violence whose ends, for those doing the sanctioning, remain natural ends and as such may, in an emergency, clash with their own legal or natural ends. Martial violence, though, addresses its ends very directly – and it does so in terms of pillage. Nevertheless, it is conspicuous in the extreme that even (or in fact precisely) in primitive conditions, where constitutional relations scarcely exist as yet, and even in instances where the victor has taken unassailable possession, some sort of peace ceremony is very much a requirement. Indeed, the word ‘peace’, in its signification as correlate of the signification ‘war’ (there is another signification, of course, a quite different one, equally non-metaphorical and political: Kant’s ‘Eternal Peace’), denotes almost an a priori peace, a necessary sanctioning of victory, any victory, that has nothing to do with any other legal relations. Such a peace will consist in the new conditions being acknowledged as a new ‘law’, regardless of whether or not those conditions require de facto some kind of guarantee that they will last. In other words, if conclusions may be drawn from martial violence as an original prototypical form of any kind of violence directed at natural ends, all such violence possesses a law-establishing character; it ‘lays down the law’, so to speak. More will need to be said later about the implications of this finding. It throws light on the aforementioned tendency of modern law to take away every kind of violence, even that directed at purely natural ends, at least from the individual considered as legal subject. In the large-scale offender, such violence confronts the law with the threat of establishing a new law or legal system – a threat that, despite its impotence, still in significant cases makes the nation tremble today, just as it did in primitive times. Government, however, fears such violence simply as law-establishing, having to acknowledge it as law-establishing where foreign powers oblige it to grant them the right to wage war, classes the right to strike.
In the last war criticism of martial violence became the jumping-off point for an impassioned critique of violence in general, which shows one thing at least: violence is no longer either exercised or tolerated naively. However, it was not only as a law-establishing phenomenon that it came in for criticism; perhaps even more devastatingly it was judged in another function as well. The fact is, a twofoldness in the function of violence is characteristic of militarism, which only universal conscription enabled to emerge. Militarism is the compulsion to make generalized use of violence as a means to state ends. Recently this compulsion to use violence has been condemned as emphatically as or even more emphatically than the use of violence itself. In it violence is seen in a quite different function than in its simple use for natural ends. That function consists in the deployment of violence as a means to legal ends. For the subordination of citizens to legislation (in the case in point, to the law governing universal conscription) is a legal end. If that first function of violence is described as law-establishing, this second function may be called law-upholding. And since conscription is a perfectly straightforward case (in other words, in no way different in principle) of the use of violence to uphold the law, truly effective criticism of it is by no means as simple a matter as the rhetoric of pacifists and peace activists makes out. Actually, it is just like criticism of any legal violence (criticism of legislative or executive violence, say) and cannot be performed at all in connection with a lesser programme. It is also (unless the intention is to proclaim a positively childish anarchism) not of course performed by refusing to acknowledge any kind of personal constraint and declaring, ‘What people like is allowed.’ Such a maxim merely rules out any consideration of the ethical and historical realm and hence of any kind of meaning of action – more: consideration of any kind of meaning of reality itself that cannot be constituted if ‘action’ has broken out of its arena. What is perhaps more important is that not even the so often attempted appeal to the Categorical Imperative, with its unquestionable minimal programme (act in such a way that you invariably use humanity, both in your person and in everyone else’s, as end, never merely as means) – that not even this is sufficient in itself for such a critique.2 Because positive law, where it is aware of its roots, will have no hesitation in demanding that the interest of humanity should be acknowledged and promoted in the person of each individual. It sees that interest in the representation and upholding of a fateful order. That order (which basically claims to safeguard the law) cannot of course be immune from criticism. But equally, any challenge aimed at it will lack force if it is issued only in the name of a formless ‘freedom’ without being able to describe that higher order of freedom. However, it will be entirely impotent if it challenges not the legal system itself, root and branch, but individual pieces of legislation or legal customs that the law does then admittedly take under the protection of its might, which consists in two things: that there is only one fate, and that precisely what exists (and especially what threatens) belongs inviolably to its order. Because law-upholding violence is one that threatens. Moreover, the threat it wields does not have the connotation of deterrence that uninformed liberal theorists put upon it. Deterrence in the precise sense of the term would imply a certainty that conflicts with the nature of threat and that is also not achieved by any piece of law, the ‘long arm’ of which there is always some hope of escaping. All the more threatening does that make it appear, in fact – like fate, which after all determines whether the offender falls victim to it. The deepest significance in the uncertainty of legal threat will emerge only after subsequent consideration of the sphere of fate from which it originates. There is a valuable pointer in that direction in the area of penalties. Among these, since the validity of positive law was first called into question, the death penalty has provoked more criticism than any other. Superficial though its arguments have been in most instances, its motives have been and still are ones of principle. Its critics have felt (possibly without being able to say why, probably without wishing to feel that way) that to challenge the death penalty is to mount an assault not on a sentence, not on specific pieces of legislation, but on law itself at its very root. The fact is, if violence, violence as crowned by fate, constitutes that root, that origin, the obvious assumption is that in the highest violence (that over life and death) in which it appears in the legal system the origins of that system erupt imposingly into what exists and there become appallingly manifest. It is in line with this that in primitive legal situations the death penalty is also applied to crimes (offences against property, for instance) to which it seems wholly ‘disproportionate’. So the point of the death penalty is not in fact to punish lawbreaking but to strengthen the new law. For in practising violence over matters of life and death, law as such finds greater reinforcement than in any other legal consummation. At the same time, though, something rotten in the law announces its presence in this very context to (first and foremost) the more delicate sensitivity, because that sensitivity knows itself to be a million miles from conditions in which fate in its majesty would have revealed itself in such a consummation. However, reason must seek the more determinedly to approach such conditions if it would bring its critique of both law-establishing and law-upholding violence to a conclusion.
In a very much more nature-hostile combination than in the death penalty, in a blend that is almost eerie, these two types of violence are present in another institution of the modern state: the police. A deployer of violence for legal ends (it has power of disposal), the police force is at the same time itself authorized to set that violence within wide bounds (it also has power of decree). The demeaning quality of such an authority, felt by few for the sole reason that its powers rarely stretch to the grossest interventions (though they may of course switch all the more blindly to the most vulnerable areas and against thoughtful folk from whom the laws do not shield the state), lies in the fact that in it there is no separation of law-establishing from law-upholding violence. While the former is required to reveal itself in victory, the latter is subject to the restriction that it abstain from setting itself new ends. Police violence is emancipated from both conditions. It is law-establishing in that its function, typically, though not the promulgation of laws, is each and every decree that it enacts with legal entitlement. And it is law-upholding for it places itself at the disposal of those ends. The claim that the ends of police violence are always identical with or even simply connected to those of the rest of the law is utterly untrue. On the contrary, so far as the police are concerned, ‘law’ basically marks the point at which the state, be it through impotence, be it because of the immanent coherence of every legal system, can no longer guarantee, through the legal system, the empirical ends that they [the police] wish at all costs to attain. That is why the police intervene ‘for security reasons’ in countless cases where no clear legal situation exists, if indeed they do not, without any reference to legal ends, accompany the citizen as a brutal encumbrance through a life governed by decrees or quite simply keep him/her under surveillance. In contrast to law, which in a ‘decision’ pinpointed by place and time acknowledges a metaphysical category through which it lays claim to criticism, consideration of the institution that is the police lights on nothing intrinsic. The violence of the police is as amorphous as its phantom manifestation (nowhere graspable, everywhere in evidence) in the lives of civilized countries. And if the police everywhere look the same, even down to the details, the fact remains that we must not, ultimately, forget that the spirit of the police force is less devastating in absolute monarchy, where it represents the violence of the ruler in whom legislative and executive absolute power are combined, than in democracies, where the existence of a police force, lacking any such superior context, suggests the greatest degeneration of violence conceivable.
All violence, seen as means, is either law-establishing or law-upholding. If it claims neither label, it forgoes, of its own accord, all validity. The consequence, however, is that any kind of violence, seen as means, participates even in the most favourable case in the problematic nature of law generally. And even if its significance cannot be foreseen with any certainty at this stage of the investigation, after what has been said law already appears in so ambiguous a moral light that the question positively rears its head: is there no other means of settling conflict between human interests than violence? Above all, the question forces one to note that wholly non-violent conflict settlement can never add up to a legal contract. The fact is, however peacefully such a contract may have been entered into by the contracting parties, it still leads in the end to possible violence. For it gives each party the right to proceed against the other by laying claim to violence of some kind, should that other party breach the contract. Not only that: like the outcome, so too does the origin of every contract point to violence. Violence may not, as law-establishing, be directly present therein, but it will be represented therein to the extent to which the power guaranteeing the legal contract is itself of violent origin – if it was not lawfully inserted in that contract by violent means. As awareness of the latent presence of violence in a legal institution fades, that institution will decline. Nowadays, it is parliaments that exemplify this. The reason why they present the woeful spectacle so familiar to us all is they have failed to retain an awareness of the revolutionary forces to which they owe their existence. In Germany especially, the latest manifestation of such violences likewise passed off without consequence so far as parliaments were concerned. Parliaments lack any sense of the law-establishing violence represented in them; no wonder they fall short of resolutions that would be worthy of such violence, cultivating instead (in compromise) what they take to be a non-violent manner of conducting political affairs. The result, however, is still
a product that, no matter how much it spurns all naked violence, nevertheless remains true to that way of thinking. This is because the striving for compromise is not motivated from within but comes from outside in the form of a counter-striving; it is because, however voluntarily a compromise is accepted, its coercive character cannot be disregarded. Every compromise, basically, is seen as a ‘second best’.3
Significantly, the decline of parliaments may have alienated as many minds from the ideal of a non-violent way of settling political conflict as war attracted to it. Pacifists on the one hand are opposed by Bolshevists and Syndicalists on the other. The latter two factions have been devastatingly and for the most part tellingly critical of present-day parliaments. Desirable and delightful though a parliament of high standing may be, comparatively speaking, any discussion of what in principle shall be non-violent means of political agreement must leave aside parliamentarianism. Because what parliamentarianism achieves in matters of life or death can never be anything but legal systems that, in terms of their origin and outcome, are touched by violence.
Is non-violent conflict settlement possible at all? Undoubtedly. Relations between private individuals abound with examples. Accord without violence occurs in every situation where the culture of the heart has placed pure means of agreement at man’s disposal. The fact is, lawful and unlawful means of every kind (all of them, of course, constituting violence) may as pure means be set against those deploying no violence. Courtesy, affection, love of peace, trust (one could go on) are their subjective prerequisite. Their objective appearance, however, is governed by the principle (the colossal scope of which is not here up for discussion) that pure means are never those of direct but always those of indirect solutions. So they never relate immediately to the settlement of conflicts between individuals but only by way of the matters at issue. The most objective context of human conflicts over property is where the field of pure means opens up. That is why technique in the broadest sense of the term is its most characteristic sphere. Its most profound instance is possibly discussion, seen as a technique of reasonable agreement. The fact is, in discussion non-violent agreement is not simply a possibility; the exclusion of violence on principle is quite expressly verifiable in terms of a significant circumstance: namely, in terms of the fact that lying goes unpunished. Possibly no legislation on earth originally penalizes it. It follows that there is a realm of human agreement that is so non-violent as to be wholly inaccessible to violence: the proper realm of ‘communication’, which is language. Not until late on and in a peculiar process of decline did legal violence nevertheless invade this realm by making fraud a punishable offence. What happened was, while originally the legal system, trusting in its victorious violence, was content to hammer unlawfulness wherever it happened to appear, and fraud, having as such nothing of violence about it, was exempt from punishment on the principle ius civile vigilantibus scriptum est [‘civil law is written for the vigilant’] or Augen für Geld [‘eyes for money’] in Roman and Old Germanic law respectively, the law of a later time, lacking faith in its own violence, no longer felt itself to be as formerly a match for all comers. On the contrary, fear of violence and distrust of itself characterized its loss of confidence. Law began to set itself ends designed to spare law-upholding violence stronger manifestations. In other words, it turned against fraud not through considerations of morality but from fear of the acts of violence [Gewalttätigkeiten] that fraud might provoke its victim into committing. Since such fear conflicts with the essentially violent nature of law as derived from its origins, ends of that kind are inappropriate to the justified means of law. In them, not only does the decline of law’s own sphere find expression; so too, at the same time, does a lessening of pure means. The fact is, in forbidding fraud law restricts the use of wholly non-violent means because these might generate reactive violence. The intended direction of law also played a part in granting the right to strike, which goes against the interests of the state. Law concedes that right because it holds in check violent acts that law is afraid to counter. Previously, workers were resorting immediately to sabotage, setting fire to their factories.
To induce people to reconcile their interests in a peaceful manner without becoming involved in any kind of legal system, there exists, ultimately (all virtues aside), an effective motive – one that more often than not provides even the stiffest will with that pure means instead of a more violent one. It lies in fear of the disadvantages to both parties that threaten to arise from a violent confrontation, whatever its outcome. In countless cases of conflict of interest between individuals, such disadvantages are clear to see. It is different when classes and nations are in dispute, for here the higher systems that threaten to overwhelm victor and vanquished alike remain hidden from most in terms of feeling and from almost all in terms of understanding. In this context, looking for such higher systems and the shared interests corresponding to them, which supply the most lasting motive for a politics of pure means, would take us too far.4 So let reference simply be made to pure means of politics itself as being analogous to those governing peaceful dealings among individuals.
So far as class struggles are concerned, in them (under certain conditions) strike must count as a pure means. Two essentially different types of strike, the possibilities of which have already been considered, need to be described in greater detail at this point. It was Sorel who (more on grounds of political than purely theoretical considerations) first drew a distinction between them, comparing and contrasting them as political and proletarian general strike. The contrast extends to the way in which the two relate to violence. Of partisans of the former he says, ‘Strengthening the violence of the state5 is the basis of their ideas; politicians (for which read: moderate Socialist politicians) are already, in their present organizations, laying the foundations for a strong, centralized, disciplined violence that will not be deterred by opposition criticism but will have the wit to impose silence and enact its mendacious decrees.’6 ‘The political general strike demonstrates how the state will lose none of its strength, how power passes from the privileged to the privileged, how the mass of the producers will swap masters.’ Unlike this political general strike (the formula of which, incidentally, appears to be that of the late German Revolution), the proletarian version sets itself the sole task of destroying the violence of the state. It ‘excludes all ideological consequences of any possible social policy; its adherents see even the most popular reforms as bourgeois’. ‘This general strike very clearly proclaims its indifference to the material gains of conquest by stating that it seeks to do away with the state; the state was actually […] the ground of existence of the dominant groups benefiting from every enterprise, the burdens of which are borne by all.’ While the first form of withholding labour amounts to violence, occasioning a purely external modification of the conditions of labour, the second, being pure means, is wholly non-violent. The reason is that it occurs not in any state of readiness to resume work after superficial concessions and some sort of modification in the conditions of labour but in a determination to resume only a quite different kind of labour, one not imposed by the state – a total upheaval that this type of strike not merely causes but actually brings about. For the same reason, while the first of these undertakings is law-establishing the second is by contrast anarchistic. Taking up things said occasionally by Marx, Sorel repudiates programmes and Utopias of whatsoever kind – in short, all legal prescriptions for the revolutionary movement: ‘With the general strike all these fine things vanish; the Revolution is seen as revolt pure and simple, with no place reserved either for sociologists, or for those elegant amateurs, the social reformers, or for the intellectuals who have made a profession of thinking for the proletariat.’ Nor can this profound, ethical, genuinely revolutionary conception be countered by any consideration that seeks to brand such a general strike as violence, given its potentially disastrous consequences. Granted, it can rightly be said that the economy of today, viewed as a whole, can be likened far less to an engine that stops running when abandoned by its stoker than to a wild animal that goes berserk as soon as its tamer turns away; even so, the violence of a course of action may be judged neither by its effects nor by its ends but only in accordance with the law of its means. State violence, of course, thinking only of effects, will oppose just such a strike (unlike partial strikes, most of which are really acts of blackmail) as alleged violence. Incidentally, how far so strict a conception of the general strike as such is likely to reduce the development of actual violence in revolutions is something that Sorel goes into with some highly ingenious arguments.
On the other hand, an egregious case of violent neglect, less ethical and cruder than the political general strike (more like a blockade, in fact), was the doctors’ strike experienced by a number of German cities. This showed a repellently unscrupulous use of violence – positively depraved on the part of a professional class that had for years, without the least attempt at resistance, ‘secured death its prey’, only to abandon life of its own volition at the first subsequent opportunity.
More clearly than in recent class struggles, it is in the thousands of years of the history of nation states that means of non-violent agreement have emerged. Only occasionally does the business of diplomats consist in reciprocal negotiations directed at modifying legal systems. In essence, their job is (entirely on the analogy of agreement between individuals) to work peacefully from case to case, without benefit of treaties, towards settling conflicts on behalf of their countries. It is a delicate task and one performed in a more resolute fashion by courts of arbitration. In principle, however, this mode of solution enjoys a higher standing than arbitration because it goes beyond any kind of legal system, thus also beyond violence. The result is that, as dealings among individuals have done, dealings among diplomats have given rise to peculiar forms and virtues that, though now superficial, were not always so.
In the whole area of the violences that both natural and positive law provide for there is none that escapes the thorny problems (as outlined here) facing every kind of legal violence. However, owing to the continued impracticality of any notion of an anyhow conceivable solution of human tasks (not to mention a release from the influence of all previous existential situations in the history of the world) that rules out completely and on principle any kind of violence, the question inevitably arises: are there other types of violence than those envisaged by the whole range of legal theory? So too does the question as to the truth of the basic dogma shared by those theories: just ends may be attained by justified means, justified means employed for just ends. How would it be if any kind of fateful violence, in employing justified means, inherently came into irreconcilable conflict with just ends, and if at the same time a violence of a different kind were foreseeable that could not of course be the justified nor the unjustified means to those ends but did not at all act as means thereto but in some other fashion instead? That would help to explain the curious and at first discouraging experience of the ultimate insolubility of all legal problems (comparable only, perhaps, in terms of its hopelessness, with the impossibility of making a conclusive ruling with regard to ‘right’ and ‘wrong’ in emerging languages). The fact is, decisions about the legitimacy of means and the justice of ends are never made by reason but only, in the former case, by fateful violence and in the latter by God. Such an insight is rare only because the ingrained habit persists of thinking of those just ends as ends of a potential law – that is to say, not simply as generally valid (which follows analytically from the defining property of justice) but also as capable of being generalized, which could be shown to conflict with that property. For ends that are just, merit general recognition, and hold good universally in one situation are not just and have neither of those qualities in any other, no matter how similar the situations may be in other respects.
One direct function of the violence at issue here can be seen from everyday experience. A person will be driven by anger, for example, to very obvious outbreaks of a violence that does not relate to a pre-established end as means thereto. It is not means but manifestation. And the fact is, that violence has some thoroughly objective manifestations in which it may be subjected to criticism. To start with, such manifestations are very significantly present in myth.
Mythic violence in its archetypal form is manifestation of the gods, pure and simple. Not means to their ends, scarcely manifestations of their will, primarily manifestation of their existence. The Niobe legend contains an outstanding example of it. Granted, the action of Apollo and Artemis might appear to be simply a punishment. But their violence sets up a law rather than punishing infringement of an existing law. Niobe’s arrogance invites her undoing not because it breaks the law but because it challenges fate – to a fight in which fate must be victorious and only in victory, possibly, reveals a law. How little such divine violence in the classical sense constituted the law-upholding aspect of punishment is shown by the hero legends, where the figure of the hero (Prometheus, for example), acting with dignified courage, challenges fate, fights it with varying degrees of success, and is left by the legend not without hope of some day bringing humanity a new law. It is this hero, actually, and the legal violence of the myth he incarnates that even today ordinary folk, in their admiration of the large-scale wrongdoer, seek to recall. In this way, out of the risky, ambiguous sphere of fate, violence overtakes Niobe. Destructive it is not – not really. Despite inflicting a violent end on Niobe’s children, it stops short of the life of their mother, leaving her behind, merely more guilt-laden than ever as a result of the deaths of the children, an eternal mute bearer of guilt and boundary stone marking the line between men and gods. If such direct violence in mythic manifestations seeks to be shown to be intimately related if not identical to the law-establishing kind, this leaves law-establishing violence with a problem in that the latter was described above, in the account of martial violence, as being purely indirect in nature, acting only as means. At the same time, this link promises to shed more light on fate, which lies at the basis of legal violence in every instance, and in broad terms to bring critical discussion thereof to a conclusion. The fact is, the function of violence in establishing law is in a sense twofold: while this operation seeks to achieve what is established as law as its end, using violence as its means, in the moment of establishing what it is aiming at as law it does not repudiate violence but only now, strictly speaking, turns it (directly, this time) into a law-establishing agency; this it does by establishing as law not an end that is free of violence and independent of it but one that is necessarily and intimately bound up with violence, calling it power [Macht]. Establishing law equates to establishing power, and to that extent it is an act of direct manifestation of violence. Justice is the principle of all divine end-establishment, power the principle of all mythic law-establishment.
The latter principle finds a hugely momentous application in constitutional law. The fact is, it is in the area of constitutional law that the kind of drawing of frontiers undertaken by the ‘peace’ that settled all wars in the age of myth is the archetypal phenomenon of law-establishing violence altogether. Here it becomes abundantly clear that what is to be guaranteed by all law-establishing violence is power more than the most lavish gain in terms of possessions. Where frontiers are established, the opponent is not simply destroyed; what happens is that, even if the victor commands vastly superior violence, the opponent is granted rights. Moreover, in a diabolically ambiguous way these are ‘equal’ rights: for both parties to the agreement it is the same line that may not be crossed. What comes out here, in all its awful originality, is the same mythic ambiguity of laws that may not be ‘overstepped’ as Anatole France lampoons when he writes: they ban rich and poor alike from sleeping under the arches. It seems, too, as if Sorel touched on a truth not just of cultural history but of metaphysics when he surmised that in the early days all law [Recht] was the prerogative [Vorrecht] of kings or magnates – in short, those in power [die Mächtige]. So it will remain, mutatis mutandis, as long as it exists. For from the standpoint of violence, which is the only thing that can guarantee law, there is no equality but at best equally great violences. However, the action of setting bounds is important for the recognition of law in another respect as well. Rules and defined bounds remain (or at least they did in primeval times) unwritten laws. A person may overstep them unawares and so fall into moral debt. Because every intervention of law that is occasioned by infringement of an unwritten, unknown rule is called not punishment but a state requiring expiation. Yet however unfortunately such intervention may strike the unsuspecting, in the eyes of the law this is not chance but fate once again portraying itself here in its systematic ambiguity. It was Hermann Cohen who in a brief survey of the classical notion of fate described as ‘a realization that becomes inescapable’ the way its ‘very decrees […] seem to cause and occasion this stepping out of line, this apostasy’.7 The modern principle that ignorance of the rule does not protect the offender from being punished still bears testimony to this spirit of law, as too the battle for written law in the early years of the classical political unit should be seen as a rebellion against the spirit of mythic dictates.
Far from opening up a purer sphere, the mythic manifestation of direct violence reveals itself as deeply identical to all legal violence and turns the presentiment of its problems into the certainty of the corrupting effect of its historical function, destroying which thus becomes a duty. Ultimately, that very duty once again poses the question of a pure direct violence that might be able to call a halt to mythic violence. As in all areas God opposes myth, so divine violence opposes mythic violence. In fact, it denotes the opposite of the latter in every respect. Where mythic violence is law-establishing, divine violence destroys law; where the first sets bounds, the second wreaks boundless destruction; where mythic violence apportions blame and calls for expiation simultaneously, divine violence expiates; where the former threatens, the latter strikes; where one is bloody, the other, albeit lethal, kills without bloodshed. As an example of this violence, compare the Niobe legend with God’s judgement against Korah and his fellow rebels.8 This strikes privileged Levites, strikes them unheralded, without warning, nor does it stop short of their annihilation. Yet at the same time there is an expiatory quality about it; there is no mistaking a profound link between the bloodless and expiatory character of this violence. For blood is the symbol of bare life [das blosse Leben]. It cannot be set out in greater detail here, but the dissolution of legal violence goes back to the guilt of bare, natural life that delivers the person living it, all innocence and misfortune, up to expiation, which then ‘expiates’ his or her guilt – and also, no doubt, redeems the guilty party, though not from any guilt: from law. For with the passing of bare life the sway of law over the living lapses. Mythic violence is blood violence over bare life for its own sake; its divine counterpart is pure violence over all life for the sake of the living person. The first calls for sacrifices, the second accepts them.
This divine violence shows itself not through religious tradition alone: it also occurs (in one consecrated manifestation, at least) in the life of the present. That which stands outside the law as educational violence in its perfect form is among its manifestations. In other words, these are defined not by God himself exercising them directly in miracles but through those moments of bloodless, smiting, atoning consummation. Ultimately, by the absence of any kind of law-establishing. To that extent, while there is justification for referring to that violence as destructive, it is so only relatively, with regard to property, law, life, that sort of thing, never absolutely in terms of the soul of the person living.
Such an extension of pure or divine violence will of course, particularly at present, provoke the fiercest attacks, and it will be countered with reference to the fact that, logically (according to its own deduction), it would also, under certain conditions, allow men to employ lethal violence against one another. That is not permitted. For to the question ‘May I kill?’ the fixed and final reply is the commandment ‘You shall not kill.’ That commandment precedes the crime, as God ‘prevents’ it from occurring. But of course, as surely as it may not be fear of punishment that leads to its being adhered to, the commandment remains inapplicable, incommensurable, in the presence of the done deed. Regarding the deed, the commandment passes no judgement. Accordingly, neither the divine verdict on the crime nor the reason for that verdict can be known in advance. That is why those who condemn every violent killing of a human being by a fellow human being are wrong to base their condemnation on the commandment. This constitutes not a criterion for such a verdict but a guideline for action aimed at the person or community acting, who must wrestle with the commandment in solitude and who in truly appalling cases have to shoulder the responsibility that comes from ignoring it. Such were the terms in which Jewry saw the situation too, when it explicitly refused to condemn murder committed in self-defence.
However, those thinkers fall back on a further theorem on the basis of which they may even imagine they can justify the commandment itself. This is the principle of the sacredness of life, which they either apply to all animal or even plant life or restrict to human life alone. In an extreme case illustrated by the revolutionary killing of the oppressor, their argument goes like this: ‘If I don’t kill, I shall never establish the empire of justice […], thinks the intellectual terrorist […]. We, however, believe that, higher still than the happiness and justice of a particular existence is existence as such.’9 That last principle is wrong, obviously; it is even dishonourable. But just as obviously it reveals the obligation to stop trying to find the reason for the commandment in what the murder does to the victim rather than in what it does to God and to the murderer. It is false and ignoble to say that existence is superior to just existence, if existence is simply meant to mean bare life – and that is the meaning assigned to it in the comment cited above. But it does contain a massive truth if existence [Dasein] (or better: life [Leben]) – words whose double meaning, entirely on the analogy of that of the word ‘peace’, is to be teased from their referring, both of them, to two spheres – means the unalterable state of ‘man’. If, that is, the principle is trying to say that the non-existence of the human being is something more appalling than the (necessarily ‘bare’) yet-to-be existence of the just human being. It is this ambiguity that gives the said principle its ostensible quality. The fact is, man is not at all identical with the bare life of man: he is no more identical with the bare life in him than with any other of his conditions and qualities – not even, indeed, with the uniqueness of his physical person. Man may be sacred (or in fact that life in him, which is the same thing in life on earth, death, and the afterlife), but his condition, his bodily life, vulnerable to his fellows, is not. So what, in essence, is the difference between his life and that of animals and plants? And even if animals and plants were sacred, surely they could not be so for the sake of their bare life; they could not be so in it? The origins of the dogma of the sacredness of life might well be worth investigating. Possibly – no, probably it is a recent development, the latest aberration of a weakened Western tradition of looking for a lost sacredness in cosmological obscurities. (The antiquity of all historical injunctions against murder in no way contradicts this, for they are based on other ideas than those underlying the modern theorem.) Finally, it gives pause for thought that what is here termed ‘sacred’ was regarded by the mythic thinking of old as the designated bearer of guilt: bare life.
The critique of violence is the philosophy of its history. ‘Philosophy’ of that history because only the idea of its outcome makes possible a critical, discriminating, decisive attitude to its temporal data. Examination purely of what is nearest to hand is able to perceive no more than a dialectical toing and froing of the twin forms of violence as law-establishing and as law-upholding. Their regular fluctuation is due to the fact that, indirectly, all law-upholding violence itself eventually weakens the law-establishing aspect represented in it by suppressing all inimical counter-violences. (Reference has been made in the course of the investigation to certain symptoms of this.) This continues until either new violences or those previously suppressed triumph over the violence that has underpinned law hitherto, thus establishing a new law destined to decline in its turn. Interruption of a cycle that is in thrall to mythic forms of law, the suspension of law coupled with the violences on which it depends as they on it (ultimately, the violence of the state) will give rise to a new era of history. If the dominion of myth is already, in the present age, broken in places, that new era is not such an unimaginably distant prospect that a word against law would take care of itself. However, if violence is assured of its continued existence as something pure and direct, even beyond law, that proves both the possibility of and the manner of revolutionary violence, by which name the highest manifestation of pure violence by humanity should be called. However, it is neither equally possible nor equally urgent for humanity to decide when in a specific instance pure violence was real. For only mythic violence, not divine violence, will be recognizable with certainty as such, except in effects that defy comparison, because the expiating force [Kraft] of violence is not obvious so far as humanity is concerned. Pure divine violence is free once again to adopt any of the everlasting forms that myth has bastardized with law. It is able to appear in true war exactly as in the divine court of the many on the criminal. But all mythic violence is reprehensible, the violence that establishes law, which may be termed the deciding kind; likewise reprehensible is the violence that upholds law, the managed violence that serves it. Let divine violence, the insignium and seal, never the means of sacred execution, be called the disposing kind.
[1921]