Max Weber, Foucault, Legitimacy, and the rewriting of the past: with a cameo by Ludwig Erhard.
Today’s post is on the opening pages of Weber’s (1918) “Politics as Vocation” (and I am using Gordon C. Wells’ translation modestly adjusted.) It’s all a bit tentative. But here goes.
The state is defined as follows by Weber: “The state is the human community [Gemeinschaft] that, within a defined territory— and the key word here is “territory”— (successfully) claims the monopoly of legitimate physical violence [physischer Gewaltsamkeit] for itself.” The emphasis on a defined territory suggests that Weber here presupposes something like the Westphalian system. In fact, that’s also clear from the fact that there is an agnosticism about how such legitimacy is constituted. (I return to that below.)
Weber’s definition also implies — and this seems less frequently noted — that states can also exercise illegitimate (or non-legitimated) violence outside their boundaries against those who have not subordinated to it. Since the lecture itself repeatedly draws attention to the Great War and its aftermath that is not just logical point. Of course, the lecture is not a contribution to just war theory, so for the moment I just mark this fact.
In addition, it seems to me that for Weber it is conceptually impossible for the state to engage in illegitimate violence in its own territory. On the view I am ascribing to him states cannot commit crimes against their own people. However, state bodies and state agents can engage in unauthorized and so illegitimate violence. This occurs when they use force without right.*
For, after his definition of the state, Weber goes on to say,
The specific characteristic of the present is that the right to use physical force is only granted to any other associations or individuals to the extent that the state itself permits this. The state is seen as the sole source of the “right” to use force.
So, the state is a source of legitimacy and authorizes or delegates this legitimacy to subordinate bodies and institutions.
The previous paragraphs may seem troubling even if one is a human rights skeptic. Surely we want to say that even legitimate states can be criminal in character. There is a loophole here. And it is one that Ludwig Erhard, and later Foucault, pick up one and uses. But in order to explain it, I need to note a few other things first.
When I claimed that Weber presupposes the Westphalian system, which conceives of state as unitary agents within a boundary, I don’t mean to suggest that for Weber that’s the only way violence and political life can be organized. He explicitly acknowledges that states are only one of many forms political life can be organized throughout history.
But lurking in Weber’s definition is also the idea that states presupposes human community; and so Weber does attribute a kind of autonomy of (historically and materially condition) civil society to the very existence of states and social life more generally. This becomes absolutely clear from an aside later, while criticizing the attempt to find a guilty party that caused the Great War, where he notes that it was the structure of society [Struktur der Gesellschaft] which generated/fathered the war [den Krieg erzeugte].
And in fact, one reason why I invoke the spirit of Westphalia here is that while Weber is vague on the details, he accepts that how states are organized is up to the underlying community, or at least those that gain control over it. This becomes clear when he comments
“What the revolution has at least achieved is that its leaders have taken the place of the legally established authorities, have established themselves, through usurpation or election, in positions of executive power over the political personnel and the administrative apparatus and derive their legitimacy, whether justifiably or not, from the will of the governed.”
In context ‘the revolution’ is a reference to the development of the modern state. It’s important to note that for Weber ‘the will of the governed’ is not the only way a state can derive legitimacy. (He is not Locke.) In fact, the state authorities can claim all kinds of sources for its legitimacy. And state forms can be of many kinds (imperial, parliamentary, etc.)
However, what all states have in common (beyond a defined territory) is that in them force and right come together; and that through the exercise of legitimate force the state organizes a relationship of dominion [Herrschaftsverhältnis] “by people over people.” Again, force is not the only manner by which such dominion or rule is effected. But it is distinctive of relationships of dominion and subordination in states.
As an aside, in virtue of the fact that it is distinctive of the state that it exercises physical violence, power within it is sought. This is also why politics is fundamentally a high stakes (including ethical stakes) activity for Weber. Politics is a mechanism by which physical violence is organized and directed in a context of uncertainty and opacity about the effects. The significance of this fact for political life is the main theme of the lecture, but here I just mark it.
Be that as it may, legitimate dominion presupposes that those that are subordinate to power, “must submit to” acquiesce in, “the claimed authority of those” who have dominion. So, from a functional perspective, all legitimate force ultimate rests in this bottom up acquiescence by the ruled. The German ‘fügen’ is usually translated as submitting. But it can also mean to ‘comply with.’ And that also indicates that the acquiescence involved need not involve approval, explicit consent, good opinion, or particular psychological states. It can involve a relatively thin notion of passive obedience.
There is a circularity lurking here: the use of legitimate force flows or is delegated down from the state to its agents (not just the army or police, but also other engines of physical control (psychiatry, medicine, education, jails, etc.), but this legitimacy is ultimately constituted from the bottom up by the acquiescence of those that are on the receiving end of it. Here I am not going to try to explain away the circularity, but rather show how it enables the loophole I promised earlier.
Above I noted and explicitly quoted that according to Weber revolutionaries and usurpers can turn out to hold power legitimately. Once the ruled comply with their newly acquired power, the usurpers can tell a story by which they turn out never to have been usurpers. This is not original with Weber, we find something like this in Hobbes (Leviathan 29, who likens it to effects of witchcraft)), David Hume, Treatise, 3.2.10.19) and doctrines like the Mandate from Heaven. That is, legitimate authority can write the story of its own legitimacy and this necessarily involves what from afar looks like the rewriting of history. (In Hume, the rewriting is done not by the usurper but later posteriority!)
The preceding paragraph describes the conceptual structure of the loophole I hinted at. Because it is also, thereby, possible that an entirely legitimate state can retroactively deny the legitimacy of preceding authorities. (This happens after revolutions and the successor state refuses to honor the debts of the prior state.) But it cannot simply do so without throwing a shade on itself by pointing to its violent character (because that is intrinsic to the exercise of state authority).
More subtly, it can also retrospectively claim that it was formally a legitimate authority, but that it failed to represent the underlying community. On Foucault’s reading this is what Erhard executed in 1948. I quote Foucault’s diagnosis (although Erhard’s own words are also worth recalling), presented on 31 January 1979, translated by Graham Burchell, Lecture 4, The Birth of Biopolitics:
A state which violates the basic freedoms, the essential rights of citizens, is no longer representative of its citizens. We can see what the precise tactical objective of this kind of statement is in reality: it amounts to saying that the National Socialist state, which violated all these rights, was not, could not be seen retrospectively as not having exercised its sovereignty legitimately. That is to say, roughly, that the orders, laws, and regulations imposed on German citizens are not invalidated and, as a result, the Germans cannot be held responsible for what was done in the legislative or regulatory framework of Nazism. However, on the other hand, it was and is retrospectively stripped of its rights of representativity. That is to say, what it did cannot be considered as having been done in the name of the German people. The whole, extremely difficult problem of the legitimacy and legal status to be given to the measures taken [under] Nazism are present in this statement. (81-82)
For the present purposes what matters here is not the endorsement of human rights by Erhard and ordoliberals (something their critics sometimes ignore). And what’s also irrelevant for present purposes is to what degree the rights of representativity actually amounts to anything outside this context. I am not mentioning it as something that follows from Weber’s account (which is much thinner). But as Foucault suggests, the purpose of this retrospective analysis is to allow for political accountability of actions done under the former regime, while maintaining some legal continuity of the state (and which may seem to suggest that where violent actions were backed by law these would have been legitimate even from the perspective of later generations).
As regular readers know, Foucault (somewhat unusually) underestimates what Erhard was up to here. As I have argued if we go to Erhard’s own argument, he was claiming that from the perspective of the 1948 present, there had been no legitimacy of the German State previously at all. So, Erhard was not re-founding the German state (in the way Lincoln did near the end of the civil war), but founding it. But the reason why Foucault misrepresents this episode is that he inscribes (and I endorse this) his history of Ordoliberalism into the reception of Max Weber (which is the theme of the fifth lecture), and so he is alert to the ways in which the functionality of legitimacy allows for such retrospective stripping, or rewriting of the past in all kinds of ways.
*While there are non-trivial differences, there is a surprisingly Lockean flavor to this.