Normativity, Causation and Explanation Revisited

Levi has put down some initial comments on my last post (here and here), and I feel that I really must clear up what appear to me to be some obvious misunderstandings of my claims. Fortunately, I think that the major misunderstanding Levi puts forward allows me to clarify some of the less clear points I made about causation and its problematic status within Levi’s variant of OOO. It also appears that I need to say some more about my own political pretensions in order to stave off the objection that I demand an appeal to ‘transcendent’ norms or that my approach ignores the reality of the political situation. Here we go then.

1. Varieties of Force

The major objection Levi has to my account of Latour is my characterisation of the first of the two moves I identified in his position (although Levi has yet to say much about the second, and I think he’ll find it equally problematic). I described this in two ways:-

1) The collapsing of the distinction between might and right.

2) The reduction of normative force to causal force.

Now, I repeatedly qualified the second way of spelling out the Latourian move because I expected that it would produce problems for Levi. I talked about a ‘loose’ causal force, or a generalised causal force, but was not really clear enough. Luckily, Levi’s objection to this second characterisation lets me make this issue clearer. Levi thinks that although the first of these characterisations is right, they are not equivalent,  and Latour does not endorse the second.

The reason he takes Latour to deny the second is the fact that Latour denies that we can have any a priori conception of what force is, or of what counts as a force. It is thus anathema to Latour to reduce one form of force to another. To quote a bit from Levi:-

First, we cannot reduce normative force to causal force. To reduce normative force to causal force is to decide a priori what constitutes a force or an actor. However, we do not know a priori what counts as a force. For Latour we only come to know another actor through the resistance it creates in another actor. “No force can, as it is often put, ‘know reality,’ other than through the difference it creates in resisting others” (1.1.5.3). To take a highly simplified example from psychoanalysis– I would have to write a much lengthier post to articulate this point –we come to know the force of the moral law through the guilt that it produces in us. This is not a causal relationship, at least not in any ordinary sense, but it is a relation of force between two entities.

The last sentence of this quote demonstrates what I feel is the real confusion in Levi’s position, if not in Latour’s itself. We have here supposed non-causal interactions between entities that are described in terms of the exertion of a force by one entity which produces an effect on another entity.

To demonstrate my problem in mad lib form (a genuine novelty for me): complete the well known phrase ‘[blank] and effect’.

It seems to me that the causation is synonymous the production of effects. It seems as if Levi is limiting causation to what might more adequately be thought of as a kind of causation. Leaving aside the issue of precisely what distinguishes this species, one might suggest that we can just name the genus ‘force’ and take causal and normative forces to be species of this genus. This would seem to alleviate the disagreement.

The major problem with this suggestion is that what Levi means by ‘force’ (which is what I meant by ‘generalised causal force’, or ‘loose’ causal force before), namely, the production of an effect upon one entity by another, is not what is at issue when we talk of normative force.

This is best demonstrated in relation to the notion of the force of reasons. When we demand reasons for adopting some theoretical position or performing some practical action, what we want are not just any reasons, but good reasons. This is another way of saying that we demand forceful reasons. However, although there is a sense in which the invocation of such reasons should cause us to either adopt that position or perform some action, the goodness of such reasons, or their force, cannot consist in this causal effect. Bad reasons sometimes ’cause’ others to adopt incorrect positions or perform unwarranted actions. We should not see this as undermining the very notion of ‘rightness’ or ‘correctness’ here (which should not be confused with truth), but rather as telling against the idea that ‘goodness’, ‘rightness’, or the forceful character of such reasons can be thought in terms of the effects they do produce. A good practical reason should make me act, but whether it does make me act is something entirely independent.

Another way of putting this is to ask what we assess when we assess whether a reason is a good one or not. Do we consider the causal effects that other instances of giving this reason (or similar reasons) have had on others? Sometimes we do. However, we do not do it all the time, nor should we. Such considerations are at best side considerations, they provide subsidiary and probative reasons in argument (e.g. “Kant was swayed by this argument? Maybe I should not be so hasty in rejecting it”). To think of it in different terms, is there ever a situation in which claims about the effect the giving of a particular reason had on anyone (or everyone) entail that it is thereby a good reason? Never.

If our assessment of the force of reasons, or the force of norms, is independent of our assessment of the effects produced by their invocation, then they are not the same. This notion of force has nothing to do with the notion of an effect.

Thus, even if we accept Levi’s terminology, and treat causation as a species of force, normative force is not a true species of this genus, because it is not to be understood in terms of effect, but in terms of correctness. Or to put it in the original terms, it is not to be understood in terms of might (the capacity to produce effects), but in terms of right.

We might thus rechristen Levi’s general notion of force (which is meant to apply to both causal and non-causal forces) effective force. This is precisely what I was trying to get at when talking about ‘generalised causal force’ and the like in my last post. The irony is that, by treating normative force as a kind of effective force, Levi has simply enacted the very reduction that I was suggesting Latour carries out. This point aside though, Levi’s overall division of force into causal and non-causal force remains problematic. This is because the notion of effective force is what is appealed to by the notion of explanation in general.

To condense the point at the end of my last post drastically: if we can create hybrid networks of explanation that appeal to both causal and non-causal forces, then we must have some general idea of how these forces relate to one another. However, it appears that there is no such account to be found in Latour. All we are provided is this general notion of force which we can only understand in terms of the very role it is supposed to play in explanation. My real concern (which one can read more about here and in the last post) is that no good sense can be made of this supplementary dimension of non-causal force that is required for the positing of hybrid explanatory networks, and moreover that it cannot be integrated with the more ordinary notion of causal force. If this is the case, then the very structure of explanation (which is supposed to motivate our adoption of the Latourian account) is undermined by this general metaphysics of ‘force’.

2. The ‘Existence’ of Norms

Moving beyond this point, there is another conflation involved in Levi’s response. This is evident in another quote:-

Insofar as norms exist within Latour’s ontology, it follows that he cannot undermine the structure of normativity as it functions in reasoning. This structure is a real actor within communicative relations. The question, for Latour, is not whether or not these sorts of norms exist, but rather of how they perform in trials of strength among actors.

Here Levi justifies the idea that Latour has not reduced normative force to causal (or effective) force by pointing out that Latour maintains the existence of entities like values and norms. Now, I didn’t deny that Latour allowed for the existence of such entities, in fact, I emphasized it. My point was that Latour follows the slightly odd strategy of both stripping these entities of their normative force and then situating them in networks where they nonetheless exhibit some generalised kind of causal or effective force on other actors.

Now, it should be clear that this is not an effective defense against the above critique. But more than that I think it is important to emphasize why I find this Latourian strategy with regard to norms so odd. To do this it is helpful to consider a few more quotes from Levi:-

Pete’s worry seems to be that Latour can provide no place for normativity in political struggles, but we have already seen that norms are real actors alongside of a host of other real actors.

—

Insofar as norms exist within Latour’s ontology, it follows that he cannot undermine the structure of normativity as it functions in reasoning. This structure is a real actor within communicative relations. The question, for Latour, is not whether or not these sorts of norms exist, but rather of how they perform in trials of strength among actors.

—

While I am all for independent norms that distinguish between reasoned discourse and abusive rhetoric, within the space of discourse itself, the appeal to these norms has very little effect. The issue then is not one of whether or not these norms exist, but rather with how these norms can be promoted within the public sphere.

The idea which repeatedly crops up here is that the role of norms in political struggle is exhausted by their existence as actors in networks, and the effects they produce within these networks. Moreover, the role of norms in political discourse is similarly exhausted by such effects.

Now, I am not denying that there are real networks of interactions between entities through which the political situation is determined. Neither am I denying that our behaviour, insofar as it involves the endorsement and propagation of norms, is an immensely important part of this. Similarly, I am not denying that political discourse itself is something which happens, and the particular results of which can be understood in terms of how certain entities interact so as to produce a certain outcome.

I am instead claiming three things:-

1) All of these interactions which are productive of the outcomes of political discourse and the wider political situation are causal, and moreover natural insofar as they are causal.

2) The normative and internal dimension of political discourse, and thus of genuinely collective political action, although manifest in these causal processes, is nonetheless distinct from them.

3) The terms that we necessarily deploy within such discourse are not necessarily the best terms with which to put together a properly external account of the causal processes in which they insist.

My own position is almost the inverse of Latour’s insofar as I claim that norms as we ordinarily talk about them don’t exist (they are pseudo-beings) but insofar as they are deployed as practical reasons within such talk they can nonetheless have force.

On the one hand, I hold that we cannot but talk about our desires and preferences in the context of individual practical reasoning, but that there is no entity like a desire or a preference which plays a role in the real causal determination of our actions as individuals (a claim familiar from EM). On the other hand I hold that we similarly cannot help but talk of norms in the context of collective practical reasoning, but that there is no entity like a norm which plays a role in the causal determination of how we act as a group.

I find Latour’s approach odd precisely insofar as it reduces the internal dimension of discourse, which is a matter of collective reasoning upon what we should do as a group, to the external dimension of how it is that we come to do what we actually do, and at the same time demands that we give this explanation using the terms that we are bound to deploy within that very internal dimension (such as ‘reason’, ‘idea’, ‘theory’, ‘norm’, ‘value’, etc.). These are classic examples of what I have elsewhere called empirico-normative hybrids (see here).

3. Solidarity Vs. Strategy

That clears up to a certain extent what I think, but it still doesn’t quite respond to all of Levi’s worries regarding why I think it. In order to clear this up it is important to get a grasp on the difference between individual and collective practical reasoning (or what I previously called private and public practical reasoning) and its relation to the idea of strategic communication. I think the best way to get at this to pick apart a further misundertanding Levi has, this time with regard to what I meant by ‘private practical reasoning’. In response to my claim that the Latourian elision of the distinction between might and right undermines all forms of reasoning other than private (individual) practical reasoning, he writes:-

This is an odd charge to advance against Latour that betrays a very weak understanding of his ontology. One of the points that Latour constantly emphasizes is the necessity of actors enlisting other actors in order to increase their degree of reality or resistance. Now an actor that is purely private, that is purely individual, is an actor that is tremendously weak in trials of strength. In order for an actor to establish itself in the world, it must enlist all sorts of human and non-human actors to increase that strength.

This gets the wrong end of the stick entirely. By ‘private’, I did not mean isolated. Individual practical reasoning is just reasoning about what one should do, not what one should do in isolation from others. It is helpful to illustrate the point I was trying to get across by appealing to a hypothetical example.

One could be a purely self-interested individual, who nonetheless desires things that he can only get through using others, through making ‘alliances’ with them, perhaps even by forming ‘alliances’ with dominant cultural norms. The example which springs to mind is that of terribly hypocritical televangelists, who ‘ally’ themselves to Christian moral norms in order to fill their pockets. They are terribly powerful and networked actors, and they are very vocal in the public or collective political discourse, but it is entirely possible that they are simply using their ‘allies’ for their own private ends. It is thus also possible that their engagement in the public discourse is not a sincere one, but a merely strategic one – just another way in which to leverage ‘alliances’ in furthering their private ends. This example could equally be extended to cynical self-serving politicians, and a whole host of others.

Such examples give us a good view of the strategic use of communication – the apparent engagement in collective reasoning upon what it is that we should do, both in the sense of what ends we should aim at and the means we should deploy to achieve them, while manipulating the debate to further one’s own private ends.

The idea is that genuine political discourse involves both a certain form of sincerity, but a certain form of solidarity.The former consists in acknowledging the normative force of the discourse, of being willing to be bound by the outcome in principle (although it is quite possible that no agreement will be reached). The latter consists in the acknowledgement of the collectivity of the discourse, of the fact that it is positing both ends and means for a putative ‘we’ of which we are included.

Sincerity is only possible on the condition that one seeks to give, and to get, good reasons for action. As we have noted above, one cannot simply assess such reasons in terms of their causal efficacy. My point was that this condition is only meetable under the Latourian scheme in individual instrumental reasoning, i.e., reasoning upon how to best achieve my own private ends. It might even be possible to sincerely engage with others in a discussion about how one should achieve one’s ends (e.g., the preacher might be able to have a relatively sincere discussion with his agent about how best to subvert the public discourse). The point is that this position does not allow for a genuinely collective discourse, by precluding anything like solidarity.

This is because solidarity is only possible on the condition that there are good reasons (i.e., forceful reasons) which motivate us independently of our individual desires and preferences. There must be the possibility of good reasons or collective action in and of themselves. Brandom’s analysis of practical reason gives us a good model here. He claims that the inference from “I am a Bank teller” to “I should wear a tie”, is a good one, simply because of the norm “Bank teller’s should wear ties”, which is here taken to be implicit in our practices (which is an issue for another day). Importantly, we do not need to read this as a disguised piece of instrumental reasoning – “Because of the effective force of the norm, if I do not want to get fired I must wear a tie, and I do not want to get fired, therefore I must wear a tie”. One can transpose this to the ethical and political case quite easily: “Performing action X would result in someone’s death, therefore I should not perform it” (because “One should not kill”, rather than “The law against killing is so strong that if I do this my other goals will be negatively affected”), and “The proposed judicial reforms undermine a persons right to meet their accuser, therefore we should resist them” (because “One should have the right to meet one’s accuser”, rather than “I might someday wish to meet my accuser” or “Allying myself with Habeas Corpus will put me in a good position with political party X”).

While the Latourian position does indeed allow that such norms as “Bank teller’s should wear ties” exist, and that they have effective force (if one does not wear a tie, one will be fired), it strips them of their normative force, which is to say, their status as good reasons for action. This does not mean that the position holds that everyone in fact reasons in a purely instrumental fashion, or that there is no solidarity. Indeed, people do reason as if norms have force, and groups who have solidarity gain certain kinds of effective force in virtue of this.

The problem is that if we genuinely and sincerely believe the Latourian position, then we are precluded from such solidarity, because we thus fail to see how intrinsic reasons for collective action can motivate us in anything other than an effective way. We may have mock-solidarity, if we desire to engage collectively, or prefer to do so. But shoring up reasons for collective action on the basis of personal motivations does not a genuine politics make. This is because one is still engaging in political discourse in a strategic way, albeit it at one level removed. This becomes evident when one’s own particular kind of solidarity is brought into question. Here one has no reason not to fall back into strategic communication, doing all one can  to see that one’s own preferred position (marxism, anarcho-syndicalism, etc) comes out on top. Of course, one’s preferences may change, and they may even be changed by discourse, but there is  no good reason for them to change, from an internal perspective, even if there are a whole set of networked causes, from an external perspective.

4. Transcendental Normativity and Rational Politics

I must now deal with the idea that I am in someway appealing to ‘transcendent’ norms. Levi positions this in terms of an opposition between transcendent ahistorical Platonic style norms (precisely what Latour rejects) and immanent norms that are generated by historical processes. His concern is that in denying the Latourian approaches situation of norms as effective entities that have real power within historical development, I can only then appeal to transcendent norms that are entirely isolated from history, and moreover, that such an appeal betrays a dangerous lack of concern with the real concrete forces that shape our lives.

Now, I want to defend myself in two ways. Firstly, by showing that although I do appeal to some ahistorical norms, these are not thereby transcendent, but rather transcendental. Secondly, by showing that I neither take all norms of action to be transcendental, nor am unconcerned with the kind of detailed analysis of the real structure of the social sphere (and the genesis of normativity that includes) provided by thinkers such as Foucault, Deleuze, Spinoza and Nietzsche. The proviso to all this is that I do not yet have an entirely fully formed political philosophy, I have some initial ideas based on my other philosophical considerations, but I hope that I can be forgiven a lack of completeness in my ideas here.

As I have tried to outline in other posts (notably here and here), I take it that there is a fundamental link between normativity and rationality. To be able to be bound by norms, to be a subject, consists in being bound by certain fundamental norms, the norms of rationality, which dictate both the structure of rational debate in general, but also the structure of interpretations or the norms we are already governed by, and the institution and determination of new norms. Being a subject does not consist in having a certain special kind of Being (such as Heidegger’s Existenz), but is itself a kind of normative status one takes on insofar as one engages in rational interaction with others (which includes more than just discourse), indeed it is simply what it is to count as taking part in such rational interaction.

The fundamental norms of rationality that one is bound by are not transcendent, but are rather transcendental insofar as their force is guaranteed by our inability to rationally reject the fact that we are bound by such norms. I have elsewhere called the fact of this inability to extricate ourselves from the norms of rationality the primary bind. It is our subjection to this primary bind, a kind of fundamental responsibility, which constitutes us as subjects. I take this to be a reworking of a fundamentally Kantian position.

However, it is important to differentiate this Kantianesque position from another one in the vicinty of  it – Levinas’ position. Levinas also holds that the subject is fundamentally constituted by its responsibility, and he also holds that the working out of this responsibility – ethics – is prior to ontology. Now, my position is different from Levinas insofar as I do not think that the we can derive the whole content of ethics and politics from the primary bind. The primary bind is most definitely a condition for ethics, insofar as we can only be bound by any norms through being bound by the norms of rationality, but its working out – what I call fundamental deontology – is not thereby equivalent to ethics.

My own positive approach is something more like Habermas’, insofar as I understand it (I really need to read the Theory of Communicative Action). I think that we can derive certain ahistorical imperatives which function as conditions for the possibility of ethics and politics from an immanent description of the structure of rationality itself, but this is far from the completion of ethics or politics. This means that I retain a fundamental commitment to both the enlightenment political ideal, and to the notion of democracy as such, insofar as I take the ideal of democracy to express the notion of a genuine process of collective reasoning upon what we should do as a group. This is to say, insofar as I take it to involve the ideals of sincerity and solidarity I outlined above. This does not imply that there is anything like a perfect democracy, just that the construction of any actual democratic system is guided by such ideals.

In short, I think what we can get out of fundamental deontology is not an ethics or politics per se, anymore than we can get a scientific account of the world out of it. Rather, just in the way that it gives us norms governing how we should proceed in elaborating and revising a scientific account of the world, I think it can give us norms governing our elaboration and revision of ethical and political perspectives – demanding that they proceed in accordance with an ideal of rational discourse on how we act toward each other and with eachother, respectively.

Now, any political discourse, in order to be an effective political discourse must have a proper grasp of the real conditions of the political situation, of the resources it has to work with and what is achievable, rather than pursuing pure utopianism. It is here that real analysis of social structures is as indispensible to social policy as real analysis of environmental systems is to environmental policy.

Now, I suspect that I differ from Habermas precisely insofar as I extend this to the domain of discourse itself. I think that we must not only attend to the ideals underlying how we should determine how we act collectively, but also to the real social structures underlying how we do determine how we act collectively. This ties in to another point that Levi made which is not entirely true, but does reveal something interesting:-

Pete calls for a naturalistic conception of reason

This is an illuminating half-truth (and I am not accusing Levi of deliberately mischaracterising me here). What I think is demanded is a transcendental account of the norms of reason, both individual and collective, on the one hand, and a naturalistic conception of how we actually make the decisions we do, both as individuals and in groups. These correspond to what I have been calling the internal and external perspectives upon rationality, respectively.

As noted above, I am opposed to Latour’s approach precisely insofar as it gives us neither of these two things, but provides us with a sort of empirico-normative hybrid that is unfit for either purpose.

Now, I do not have much to say about the domain of individual action (this is the province of naturalistic psychology). But with regard to the collective domain I think there is the possibility of naturalistic approach to collective reasoning that we have as yet not managed to articulate. I actually think that Foucault provides some very interesting tools for pursuing this project (and his antipathy toward Habermas could be seen as the fact that they are each taking a different route in the description of collective rationality, one ideal and one real, one normative and one quasi-natural). The important point is that to achieve this we must rigorously separate out the internal and external perspectives on the social, so that the one does no hold the other back (again, I’ve written more about this here).

5. Conclusion: Rationality Hacking

To conclude I will address a worry that many will be having at this point: If we have an ideal conception of collective rationality on the one hand, and a real conception on the other, how do these two conceptions interact? How do they do anything other than undercut one another and create confusions?

I think explaining at least one way in which this is the case is a good way of presenting my own peculiar conception of what philosophy is, and also what its traditional nemesis – sophistry – has become. (It should be obvious that I am not making any accusations of sophistry toward anyone specific here).

I will introduce this by way of a metaphor, and one must understand the terms of the metaphor for it to work. The word ‘hacker’ has become quite hackneyed in common usage, especially given the ludicrous and magical abilities conveyed on ‘hackers’ in the pop cultural domain, but there is still something interesting behind it. I am particularly interested in the distinction between ‘white hat’ and ‘black hat’ hackers (itself a metaphor taken from classic cowboy westerns). Both kinds of hackers probe computer systems, either by going through a program’s code or the more specific configurations of a particular computer set up or network, with the purpose of uncovering vulnerabilities. They differ in what they do with these vulnerabilities. ‘White hat’ hackers seek to patch the vulnerabilities, to make the computer system stronger. ‘Black hat’ hackers seek to exploit these vulnerabilities for their own ends.

I suggest that we think of the concrete structures of collective rationality (systems of information distribution, mechanisms of socialisation which create and maintain certain conceptual capcities, networks of deference and authority through which ‘expertise’ is articulated and disputed, hard-coded governmental structures which translate ‘public’ opinion into policy through a series of different levels) as being like a computer system, both in terms of having a specific setup (the elaborate hard coded social structures) and there being certain general programs which run throughout it (the conceptual capacities instituted and maintained throughout the system).

If we do this, then we can see that the philosopher is a rationality hacker. He analyses the both the conceptual structures that are propagated in the system, and the structure of the system itself. He performs two functions, both analogous to that of the ‘white hat’ hacker: a critical function and a constructive function. The former involves protecting against the misuse of existing conceptual structures, reigning in the over-extension of reason. He ‘patches’ concepts through improving their definitions, and clarifying their relations to other concepts. The latter involves replacing existing and defective conceptual machinery, constructing new concepts, be they metaphysical concepts or otherwise. In this sense I take both Wittgenstein and Deleuze to be right, but to be talking about two different sides of philosophy’s activity.

On the other hand, there is the sophist, who is the equivalent of the ‘black hat’ hat rationality hacker. He analyses the structures of individual and collective rationality to find weak points that he can exploit for his own ends. It does not matter here whether his ends are noble or not, whether he is protagoras or callicles. The sophist can only see collective reasoning from a strategic point of view, as another resources to be exploited, be it through tthe cynical formation of ‘alliances’ or otherwise. The contemporary form of the sophist is the marketing executive (as Deleuze noted in WIP) and the PR consultant. Those who see discourse as a matter of providing reasons that people will buy, rather than reasons that are genuinely good reasons.

If we are to be more than mere sophists, then we must have an ideal or reason above and beyond our own ends (no matter how noble those ends are). This ideal must guide our both our critical analysis and our constructive production of concepts and theories. Transcendental normativity is thus an essential complement to a naturalised conception of rationality. The difficulty is a matter of keeping these two different things separated, while nonetheless allowing them to inform one another, otherwise we can lose both our ability to make claims about what we should do, and undermine our explanations of how it is that such decisions are made and enacted.