Power and self-formation

Ok.  Acknowledging that this line of the thought of the blog is probably even more associative, under-researched, and crankish than normal, I want to review where I’ve gotten to with this series of posts, then plough on.

I’ve organised the project that I’m pursuing here into two rough levels.  At one level, we’ve got the formal modelling resources of standard, mainstream economics: as I’ve argued at length before on the blog, I think these resources are generally unfairly maligned in what one could call a ‘critical theory’ space, and I’m very happy to make use of them.  That said, clearly these resources have limits – and one of those limits is their unitary understanding of the self.  This unitary model of the self means that there are core features of social life that this kind of modelling apparatus is intrinsically never going to be able to capture.  Within economics there’s a whole vast research programme – behavioural economics – attempting to bring our model of the self as economic actor closer to empirical reality.  But that’s not the line of inquiry I’m pursuing here.  Instead, I am – no doubt perversely – pursuing here an even more abstracted and a prioristic line of thought, using the resources of Brandomian neopragmatism to construct a picture of the socially-constituted self within a complexly pluralistic social world, basically by means of armchair philosophising.  Now, my aspiration is for this framework to be sufficiently abstract and slimline that it doesn’t constitute an a priori philosophical anthropology (which would be bad) but simply provides a lot of analytic tools to think with (which would be good).  I understand, of course, why people might be sceptical about this project – but in any case, rightly or wrongly, that’s the project I’m pursuing here.

Now, having outlined the Brandomian-Hegelian apparatus that I am taking as the core of this level of the project, I’ve been trying to push to extend that apparatus in three discrete directions.  First, I want to be able to theorise normative pluralism within the Brandomian-Hegelian apparatus, in a way that takes us more or less as far as possible from the Lukácsian vision of totality.  Second, I want to make space for a more psychoanalytic-inflected understanding of the way in which the self is mediated and constituted by occupying other selves’ perspectives.  Third – and importantly – I want to bring in power and coercion.  Brandom, as I said in the last post, basically rules power and coercion out of his work’s domain towards the start of ‘Making It Explicit’ – Brandom is proposing a sanctions-based account of the institution of normativity, but the sanctions he is interested in are exclusively normative sanctions.  This distinction can be articulated in more political-economic terms within the framework of Elinor Ostrom’s metatheoretical apparatus.  For Ostrom, political-economic institutions are structured by deontic statements – statements that can be articulated using a deontic normative vocabulary.  In Ostrom’s vocabulary, norms are deontic statements the ‘enforcement’ of which simply involves normative sanctions: social actors comply with the norms because of the “warm glow” associated with compliance, and the shame associated with non-compliance.  Rules, by contrast, are deontic statements that are ‘enforced’ by material sanctions of some kind: sanctions that go beyond normative attitudes and their psychological consequences.

In this Ostromian vocabulary, what I want to do now on the blog is start thinking about rules, not just norms.  Just as norms are instituted by normative attitudes, so rules are instituted by normative attitudes plus ‘material sanctions’ – i.e. exercises of coercive power.  That coercive power doesn’t have to be violence – but violence is (so to speak) both a limit case and a central case of coercive power.  You can have normative attitudes without norms – many normative attitudes do not successfully participate in the collective social practice of instituting norms.  Similarly, you can have coercive power without rules – many forms of material sanction, coercion and violence are not involved in the institution of rules, they are “just” coercion and violence.  But I want to start to think about the way in which ‘material’ sanctions – coercive power – enters into the social context which co-constitutes both our meanings and our selves.

I guess I should flag at this stage that I’m aware that I’m moving in a fairly cavalier fashion between different senses of “power”.  Of course, “the force of the better reason” is a kind of force.  Epistemic authority is a kind of power.  Moreover, I suspect that I’m eventually going to end up in a place where different senses of power are mutually implicated.  Nevertheless, I am, in the first place, interested in, so to speak, “crass power” – coercive power, violent power, the kind of power that Brandom is talking about when he discusses “beating people with sticks”.  I don’t think we can think about political economy without thinking about the power to wield violence, and the power that derives from violence – and I don’t think we take the poststructuralist or social constructionist challenge seriously unless we see this kind of power as a major part of the social context that constitutes our meanings and our selves.  So that’s the line of thought I’m pursuing here.  And I want to start with a fairly crass typology, pivoting off of Steven Lukes’ work on power, which I summarised a few posts ago.  I want to ask: how is coercive power involved in the constitution of the self?  And I want to suggest a range of different categories of answer.

Answer one: Coercion has nothing to do with the core of the self at all.  From this perspective, the self exists prior to and independent of coercive power.  Coercive power comes from the outside.  It may break the self, it may occlude the self, it may coerce the self, but at some fundamental level it does not make or remake the self.  This kind of perspective provides a very robust ‘critical standpoint’, in which coercion can be critiqued from the standpoint of the underlying self that resists it.

Answer two: Coercion does not make the self, but it may reveal or realise it.  This, I take it, is one of the options made available by the ‘three-dimensional’ accounts of a self divided between real interests and mistaken preferences that I discussed in my post on Steven Lukes.  One (notorious) way in which this approach can inflect is the Rousseauian argument that people can be “forced to be free”, because their real (hidden) selves have chosen to follow rules that their false (manifest) selves are transgressing – the coercive enforcement of compliance with those rules is therefore really ‘chosen’ by the true underlying self.  I think it’s obvious why we might want to be wary of this approach – but of course as soon as we move into any kind of account of a divided self, this sort of scenario is put on the table.  More on all this one day, no doubt.

Answer three: Coercion does make the self, by bringing the self into line with community norms that are then internalised.  This is, basically, an account of the role of coercion in socialisation, and the role of socialisation in the formation of the self.  In my own view, for what it’s worth, some version of this story must be right.  I don’t think it’s really possible to bring up small children without seeing that this is a process that is frequently coercive: small children have extremely strong preferences that you sometimes simply cannot allow them to act on, because you really do know better than them what they should and shouldn’t do.  Moreover, the process by which you ensure that they do not, or do, act in such-and-such a way is, inevitably, sometimes coercive.  I want to be extremely clear here that I am not in any way advocating violence in childcare.  But of course we coerce small children – we limit their freedom of movement, we constrain them in ways they don’t want to be constrained, and so on and so forth.  And the basic model here is that this is part of an ongoing process of socialisation via which the (adult) self is (eventually) formed.

As I say, then, some version of this account must surely be right.  But we need to take great care in the generalisation of this account – because of course (like the Rousseauian picture discussed above) it can very easily provide warrant for the most grotesque acts of violence, understood as crucial to instituting ‘moral values’ in the recipients of the violence.  This, in fact, is the flip side of the Enlightenment’s understanding of itself as “man’s emergence from his self-imposed nonage”.  On the one hand, this Enlightenment self-understanding insisted on the independence of the individual, and on the fact that power and authority were instituted by individuals, rather than being natural phenomena to which we must defer.  That’s the perspective of the ‘Enlightened’ self, and is one of the core ways in which Enlightenment philosophy is ‘critical’.  But the understanding of humanity as emerging from “self-imposed nonage” carries with it the idea that huge swathes of humanity may still be within their “nonage” – and if this is so, then the process of self-formation may still fall grossly short for most people.  Much of humanity may still be in a state of childhood in which extremely far-reaching coercive sanctions are appropriate because no full moral self yet exists.  This kind of violence is required and justified, both to protect the civilised (those who have attained moral maturity) against the uncivilised (who cannot be trusted to not themselves engage in arbitrary violence, by virtue of their lack of real norm-following self-control), and as the pathway via which civilisation (Enlightenment) may one day be achieved for the ‘wards’ of Enlightened government, on the model of the child’s movement towards adulthood.  Needless to say, this ideological framework was one part (though only one part) of the ideological apparatus that justified Enlightenment liberal despotism: in the domination of women by men, the poor by the wealthy, slaves by masters, and the vast colonial periphery by the imperial core.

In short, there is good reason to be very wary about this ‘paternalistic’ understanding of the role of coercion in self-formation.  But, as I said, something in this formal space clearly needs to be available to us, in some form.

Answer four: coercion shapes the self, but not in a way that brings the self into compliance with community norms.  On this picture, coercion is experienced, but the coercion is not part of a paternalistic educational or socialisation process – at least not one we recognise as legitimate. It is simply something that happens, something that is done to the self, and the self is shaped by what happens to it.  I think it’s extremely clear that this happens all the time, and we need to be able to theorise it.  We may wish some awful event had not taken place; but the fact that it did take place is, nevertheless, constitutive of who we are.  This is, in my view, the kind of space we need to be able to think in if we are to theorise the ‘fourth dimension’ of power, in the terminology I discussed in my post on Lukes.  Here we do not endorse the coercion, but the coercion is nevertheless constitutive of the self that refuses to endorse it.

Finally, answer five.  I’m not entirely sure how to shorthand this final answer, but I would characterise it as an enthusiastic endorsement of the constitutive role of coercion in self-formation – not just because coercion is a means to an end via which moral progress takes place, but because coercion is itself a realisation of self.  Fascist social theory (though not only fascist social theory) occupies this ideological location.  Nietzsche is an example of this attitude.  From this perspective, the social world is coercive in its core, and ideological attitudes that aspire to limit or abolish coercion are themselves engaged in a kind of false consciousness, or self-deception – lying to themselves about their own true nature.  The correct social-theoretic attitude, from this perspective, is to embrace and endorse the fact that the self is made by violence and perpetrates violence – coercion is, in the end, the stuff of life.

I strongly reject this final perspective.  But I think it is a perspective that needs to be worried about when you are engaged in the kind of project that I am engaged in – that is, pursuing a Foucault-adjacent investigation into the role that coercion plays in the social construction of our analytical categories and our selves.  I think a lot of theorists who go down the kind of road that I’m advocating end up in (or start off in!) this kind of broadly Nietzschean space.  (And I don’t think Foucault is an exception to this.)  Just as the account of Hegelian recognitive relations I’ve been working through on the blog aims to strongly distance itself from Lukácsian totality, so the account of the role of coercion in the social constitution of meaning and self that I’m trying to begin to work through now aims to strongly distance itself from Nietzschean ideas of the will to power, slave morality, and so on.  I’ll say more about this down the road, hopefully.  For now, I just wanted to start off with this typology, as some grist for the mill.  I’ll see whether I can do anything useful with any of this down the road.

Poststructuralism and pragmatism – some very programmatic thoughts

The goal of this blog post is just to ‘externalise’ some thoughts that I don’t have time to really work through properly right now.  I’m currently working on another longish blog post about the relation between Derrida and Brandom – and I am in some despair at how far these issues are from the ones I’m nominally meant to be pursuing on the blog.  So I want to try to situate that (presently non-existent) blog post within the larger argument to which it aspires to contribute.

When I was a young man, the ‘science wars’ were raging.  Those ‘wars’ were in part about the status of scientific truth, and in part about the role of ‘postmodernism’ and ‘poststructuralism’ within our intellectual culture.  I have a lot of thoughts about those science wars – but here I want to think aloud about poststructuralism.  In terms of my own personal canon, I saw the two major philosophical figures within ‘post-structuralism’, whether rightly or wrongly, as Derrida and Foucault (though of course there were many other important figures).  These were both ‘critical’ philosophers – and their critique seemed, to me, to be directed in two directions.  On the one hand, they were critics of the ‘subject’.  Poststructuralism was interested in ‘social constructivism’ – it saw many philosophical categories as socially constructed, and it saw the self as socially constructed.  In contrast to a liberal philosophical and political imaginary which saw the pre-social and self-possessing self as ontological bedrock, poststructuralism saw the self as produced by larger forces, in some sense.  But poststructuralism’s critique was also directed towards accounts of those larger forces.  The ‘structuralist’ intellectual milieu from which poststructuralism emerged and against which it defined itself was, obviously enough, preoccupied with social structure.  Poststructuralists saw structuralist analysis as often dogmatically reifying or hypostatising an internally divided social context.  This critique had a political dimension – for poststructuralism, an excessive emphasis on structure was also associated with dogmatic forms of (often Marxist) leftism that sought to insist on specific rigid social categories as the only legitimate form of analysis of the social.  The poststructuralist critique thus pointed in two directions: left and right; micro and macro.  It rejected a liberalism that sought to ground itself in a naive concept of a non-socially-constituted individuality, but it also rejected a leftism that insisted on lockstep ‘grand narratives’ in which rigidly defined social structures were the engine of history.  And poststructuralism therefore had enemies on both ‘wings’ that it attacked.  On the one hand, many right-wingers saw poststructuralism as an attack on all that is good and virtuous in Western civilisation (these fights were a clear and immediate precursor to the fights over ‘cultural Marxism’ we are enduring today).  On the other hand, much of the ‘traditional’ left saw poststructuralism as a form of hated liberalism, which sought to dissolve concrete analysis of real social structures into an undifferentiated goop of ‘constructivist’, i.e. subjectivist, gibberish.

In my twenties, though I wouldn’t have put things like this (and probably wouldn’t have been able to), poststructuralism seemed pretty great.  And, if I’m honest, I still think a lot of it is pretty great.  The poststructuralist project, articulated in these terms, still seems to me to be basically a good and exciting one.

There were problems, however – and from my point of view the biggest problem was one of… how to put this… signal to noise ratio.  To be less generous: poststructuralism had a problem with bullshit.  Poststructuralism’s critics were often absurdly ungenerous in their engagements with key poststructuralist texts, sometimes attaining something close to functional illiteracy in their willed inability to parse straightforward philosophical arguments.  On the other hand, I think all but the most strident defenders of poststructuralism would have to admit, looking back, that the intellectual movement generated a huge amount of profoundly opaque verbiage.  It’s very hard, in my view, not to see a lot of that verbiage as straight-up obscurantism.  The poststructuralists had ahold of something, they had important things to say – but exactly what they had to say was often unclear even to them, and it was common for even the most serious and careful figures within the tradition to resort to absurd hyperbole and bluster rather than slow down and try to spell their claims out in – inevitably more banal, less exciting, and more difficult – careful detail.

Looking back, then, my attitude to poststructuralism is pretty close to the analytical Marxists’ attitude to Marxism.  (Of course I also share the analytical Marxists’ attitude to Marxism.)  That is to say: I think there is much of value in this tradition, but unfortunately what is of value needs to be extracted from a vast cloud of handwaving, obscurantist bluster.  What I would ideally like, again by analogy with analytical Marxism, is “poststructuralism without the bullshit”.

Now, shifting gears slightly: Brandom is of course much too nice a person to characterise ‘A Spirit of Trust’ as an exercise in “Hegel without the bullshit” – but I think it’s pretty clear that that’s what it is.  I am, of course, strongly in favour of this enterprise.  Moreover, because Hegel is such an important intellectual figure, whose influence on subsequent European philosophy is so substantial, Brandom’s no-bullshit elaboration of Hegel’s apparatus also gives us a lot of resources for cashing out in a much more analytic, legible, and precise idiom the arguments of later figures (including poststructuralist figures!) who draw on Hegelian resources.

As I said a few paragraphs ago, for me the two main figures here are Derrida and Foucault.  The post I’m trying to work on in the background tries to elaborate in greater detail than I have to date the degree of positive overlap between ‘A Spirit of Trust’ and Derrida.  I really think there are a lot of commonalities there – and that thinking through those commonalities can help us to understand both thinkers better.

But here are the points I really want to make in this post.

First: although Derrida is, as it were, my ‘first love’ within the poststructuralist canon, his preoccupations are really quite distant from my current interests.  Derrida is – like Brandom – in the first instance a philosopher of language and of meaning.  This is why there is so much overlap between the two thinkers: both of them are, at base, obsessed with the issue of the determination of conceptual content by a complex and evolving social context.  And I am very interested in this too!  Indeed, I am much more interested in this issue than I want to be – I am apparently helplessly drawn to these theorists, and I can’t stop thinking and writing about them.

But what I want to be interested in – the domain of the intellectual project I am actually trying to pursue here – is political economy.  And here the figure who is most relevant is Foucault, not Derrida.  For sure, Derrida sometimes writes about political-economic issues – in ‘Spectres of Marx’, and in ‘Given Time’, for example.  But really he is interested in other issues.  Foucault, by contrast, is at base a political philosopher: Foucault’s engagement with the political-economic canon is central and ongoing.  Both Derrida and Foucault are interested in the way in which the ‘social context’ constitutes thought and self – but when Foucault talks about the social context he is, often, interested in actual political-economic history.  To put things maximally crudely: Derrida is interested in language; Foucault is interested in power.

So if I want to draw my interest in “no-bullshit poststructuralism” into this ‘psychodialectical’ side of the project I’m pursuing here, really I ought to be thinking about Foucault much more than Derrida.  But here’s the problem: precisely because there is so much overlap between Brandom’s and Derrida’s interests, Brandom’s “no bullshit Hegel” is much more amenable to the reworking of Derridean themes than it is to Foucauldian themes.

I think this is most clear on the issue of coercion.  When it comes to many Derridean preoccupations – iterability, the constitutive role of the signifier in the formation of conceptual content, and so forth – I think it’s fairly clear how those preoccupations overlap with the preoccupations of ‘A Spirit of Trust’ (or, for that matter, ‘Making It Explicit’).  But when it comes to Foucault’s most obviously distinctive preoccupations – power, punishment, sexuality, etc. – the Brandomian-Hegelian apparatus is largely silent.  Not only is it silent – in important ways it specifically rules these issues out of bounds.  Thus, for example, as I keep saying, near the start of ‘Making It Explicit’ there is a discussion of coercive sanctions (“beating people with sticks”) that mainly serves to foreground the fact that these are not the kinds of sanctions that will play a fundamental role in Brandom’s apparatus.  On the contrary, Brandom is interested in normative sanctions, not in any form of coercion.

So when we come to one of the central Foucauldian problematics or issue-spaces – the role of power in the formation of knowledge, and in the formation of selves – Brandom’s apparatus is unfortunately largely silent.  I want to claim (though I have not made the case) that Brandom’s apparatus can vindicate much of the Derridean project.  One of the characteristic poststructuralist moves is to say “such-and-such a phenomenon, excluded from accounts of such-and-such a philosophical category, is really constitutive of that category”.  When it comes to Derrida’s key favoured ‘excluded categories’ – such as textual signification – I think it is clear how the Brandomian apparatus can assist in the ‘cashing out’ of that argument.  But when it comes to power?  To violence and coercion?  What does the Brandomian-Hegelian apparatus have to say on this set of issues?  I don’t think the answer is quite “nothing” – but clearly there is a lot of work involved in pursuing this line of thought, within this metatheoretical framework, relative to pursuing the Derridean line of thought.

In previous posts in this series I’ve basically said that the project I’m pursuing on the blog is operating at two levels.  One is the level of ‘traditional’ formal political economy: using the marginalist apparatus to analyse political-economic institutions and behaviours.  I haven’t spent much time on this side of things, on the blog – maybe one day I will.  But, for what it’s worth, I think there’s an obvious resource available for pursuing the study of power and coercion within this framework: bargaining theory.

At the metatheoretical, Brandomian, neopragmatist level, though, I feel like I am still really only beginning to think about how to think about power and coercion.  I guess the main thing I want to do in this post is just flag this fact – and put down a marker that this is something I want to keep worrying or working away at, over the coming months and years.

Preferences in the divided self

Ok.  In the last post I summarised Steven Lukes’ typologies of different understandings of power, and of the relationship between interests and preferences.  In the post before that I outlined the way in which some of the different bits of the project I’m pursuing here fit together at a metatheoretical/methodological level – specifically how the ‘homo economicus’ apparatus I broadly endorse fits together with the social constructivist pragmatism that I also endorse.  In this post I want to map those two previous posts onto each other, and in doing so start the slow process of bringing power into the metatheoretical apparatus I’m elaborating here.

Lukes, recall, draws a distinction between three different accounts of power.  In ‘one-dimensional’ accounts, per this typology, researchers simply look at which actors within a political system are best able to realise their preferences.  In ‘two-dimensional’ accounts, researchers look at how the ‘rules of the game’ that structure the political system determine whose preferences even enter into contestation within the political system.  Both of these approaches simply take social actors’ preferences as given – where they differ is in how much attention they pay to the way some social actors’ preferences can shape the ‘rules of the game’ to suppress other actors’ preferences.

I think both of these approaches can clearly be accommodated within a basic ‘homo economicus’ analytic framework.  (Lukes is interested in more qualitative kinds of political science and social theory, but I’m at least in principle interested in pursuing more formal approaches.)  In ‘one-dimensional’ accounts modellers can specify the political framework, and then model negotiations between social actors within that framework.  In ‘two-dimensional’ accounts modellers can look one level up, at how the ‘rules of the game’ are themselves shaped by social actors’ preferences.  I’m no political scientist, but within the more popular end of political science works I’ve read, I guess Bruce Bueno de Mesquita’s negotiation models are an example of a ‘one-dimensional’ approach, while Acemoğlu and Robinson’s accounts of how political actors’ interests produce more ‘inclusive’ or ‘extractive’ rules of the game might count as a ‘two-dimensional’ account (see also Douglass North, or of course a whole range of other institutional theorists, including many public choice theorists).

The point is that these approaches are entirely compatible with traditional formal economic modelling: models in which social agents are taken to act rationally given preference functions that are – crucially – unitary and internally consistent.  All these kinds of modelling approaches centrally rely on taking the relevant social actors to have well-behaved, consistent, unitary preference functions.  I don’t (contra Lukes) think it’s necessarily the case that, for this reason, ‘homo economicus’ intrinsically identifies preferences with interests.  It is perfectly possible for economic theorists to claim that the preference functions they ascribe to social actors express those social actors’ real interests better than the social actors themselves understand them.  This is one of the worries that Austrian economists have about technocratic economics oriented to welfare maximisation: they worry that mandarin economists believe their welfare apparatus can enable them to paternalistically ‘maximise wellbeing’ regardless of whether the social actors in question agree about what constitutes their wellbeing.  So this ‘homo economics’ apparatus isn’t (contra Lukes) intrinsically incompatible with drawing a sharp distinction between interests and preferences.  Nevertheless, this apparatus does make it difficult to theorise the distinction between interests and preferences – this distinction intrinsically falls outside the formal apparatus in question.

If we want to theorise something in the space of a distinction between interests and preferences, then – or even just conflictual preferences of any sort – we need a non-unitary conception of social actors.  This is where we need to move beyond ‘homo economicus’ – stepping into the moral psychological space that I discussed two posts ago.  In terms of my own metatheoretical apparatus, this is where I want to step into a ‘psychodialectical’ space.  (I desperately need a better term for this apparatus than ‘psychodialectical’.  What I’m trying to capture is that this is a psychoanalytically-inflected version of Brandom’s deflationary neo-pragmatist Hegelianism.  I guess I’ll stick with ‘psychodialectical’ for now.)  In the remainder of this post, I want to begin to think about how the ‘third’ and ‘fourth’ dimensions of power that I discussed in the last post can be articulated within a ‘psychodialectical’ framework.  I want to do this in two stages.

Stage one: the divided self.  In order to talk about the ‘third dimension’ of power that interests Lukes, we need to be able to talk about social actors as having conflictual preferences.  I think the broadly Brandomian ‘psychodialectical’ apparatus I’ve been working through here gives us several resources to do this.  Brandom of course is most centrally interested in semantics – but as the blog has discussed at some length recently, Brandom’s semantics is in a sense a special case of his broader theory of action, and it’s ‘practical rationality’ that I’m mostly interested in here going forward.  (That is to say: the goal on the blog going forward is to apply the Brandomian apparatus to problems of ‘practical reason’ that inhabit the same broad problem space as the narrower ‘instrumental’ concept of practical reason that animates homo economicus.)

First resource: the de dicto / de re distinction.  Social actors can take themselves to be committed to one thing, while we take them to be committed to something else.  This is true in semantics – but it is also true in terms of practical intention.  You think the implications of your action are X but I (take myself to) know that the implications of your action are really Y.  This is the space of ‘unintended consequences’.  There’s a strong case that this kind of division within the self can be folded into a ‘homo economicus’ account by virtue of limiting social actors’ knowledge – and obviously there’s a vast literature looking at the consequences of relaxing the ‘perfect knowledge’ assumption of traditional formal economic modelling.  Still, it is a way in which the self can be seen as divided.

Second resource: Preferences unacknowledged to self.  This is where we come to the real psychoanalytic dimension of the apparatus.  One of the claims of the Brandomian-Hegelian apparatus is that we can come to know ourselves by seeing our actions – that just as ‘meanings’ don’t quite fully exist prior to expression (because the semantic content in question is determined by the broader social context, and therefore cannot really exist independent of that context) so ‘preferences’ don’t quite fully exist prior to action (because what the preferences consist in is, likewise, partly determined by the actual substance of the action).  There are a range of different claims bundled together here that I haven’t worked through yet, on the blog – and I think it would be easy to end in with some deep philosophical confusions by muddling up those claims.  So I want to tread a little carefully here.  But I think it’s reasonable to say, as a first pass, that this element of Brandom-Hegel’s apparatus is at least adjacent to the Freudian idea of the unconscious.  That is to say: we may not know our own intentions until we see the actions they result in.  Moreover, we may not see our own intentions correctly even then – others may have a better sense of the intentions that are animating our actions than we do ourselves.  Again flagging that this all needs to be worked through more carefully, I think this kind of space gives us the resources to talk about a ‘deeper’ divergence between our sense of our own commitments and those commitments themselves.  We may refuse to accept what our own commitments really are – we may be attributing to ourselves more palatable motives than the motives that are really animating our own actions.  This, then, is a second kind of way in which intentions can be divided.

Third resource: We may simply have conflictual preferences.  We are large, we contain multitudes.  Normative pluralism applies within the self, as well as between subcommunities.  In navigating my social world, I am constantly navigating between incompatible commitments, all of which I legitimately hold.  The practical side of the process of the ‘synthetic unity of apperception’ attempts to wrangle the self-contradictions of the self into manageable proportions, but this is a Sysiphean task.  So at this very basic level, we do not have well-behaved preference functions.

So – once we enter this moral-psychological or psychodialectical space, we are dealing with a massively more complex understanding of the self’s preferences.  These three different forms of non-self-identity within preferences can of course combine in complicated ways.  Consider a given decision.  I may have incompatible preferences that I somehow have to navigate between.  I do so, and I make a decision.  But what is that decision – what preference does it express?  I have a view about what preference it expresses.  But it may turn out that that view is doubly wrong.  First, I am wrong about the likely consequences of my action – I am incorrect about what I have committed myself to.  Moreover, even if I were correct about what I have committed myself to, I may be lying to myself about what desire or goal this commitment really expresses – the unconscious may be at the wheel.  Somebody assessing my action can therefore in principle draw a very complicated ‘map’ of my preferences: “They think they have committed themselves to action A in order to achieve consequence B. In fact they are wrong about the implications of their action – really they have committed themselves to action C.  Moreover, they are wrong about the desire guiding their action – they really intend to achieve consequence D.  Furthermore, although that desire is what is guiding their action, they are internally conflicted, and their overall preferences are still on balance committed to E.”

This is all of course highly abstract – but I hope it resonates with how we actually think about psychology.  At least, something in this broad space is how I think about things.  I often have fairly complicated assessments of what’s going on in people’s heads (including my own), and something of those complicated assessments is, I think, captured by this broad formal space.  In fact, things can get significantly more complicated than this, once we introduce other Freudian concepts like double-investment, and so on.  But leave all that aside for now.  The point is that we here (even in this relatively simple and schematic form) have a vastly more complicated account of preferences, and the relation between preferences and action, than ‘homo economicus’.

And – I want to say – all this gives us a lot of resources for ‘cashing out’ in slightly more elaborated form some elements of the ‘third dimension’ of power that preoccupies Lukes.  Lukes is interested in the distinction between preferences and interests.  I would argue that putting things in terms of this bipartite typology is a bit misleading.  Rather than talking about interests versus preferences, I think it is more elucidating – and more symmetrical – to simply talk about the self being divided, in the way I just have.  When we consider anyone’s psychology, from this kind of moral-psychological or psychodialectical perspective, we find a complexly divided self.  We can identify multiple preferences – multiple categories of preference – all at work at the same time.  And then we have a set of ‘interpretive’ decisions to make.  When we consider a person’s psychology, with its many different, incompatible desires and commitments, we have to decide: which of these desires and commitments are likely to be dominant at any given time?  Separately, which of these are the ‘real’ self, and which are more incidental?  There is not necessarily a ‘fact of the matter’ here.  We may choose to treat some relatively ‘minor’ element of self as the most normatively or psychologically important – and attempt to foster that element of self, so that over time it becomes more influential in practice.  And which elements of self we regard as most important is going to depend on our own normative judgements.

When theorists of ‘interests’ talk about real, underlying interests, then, they are (as Lukes says) making value judgements about which elements of a divided self they are choosing to regard as ‘core’, and which they are choosing to regard as more ‘incidental’.  There is no socially-transcendent fact of the matter here – this is a normative judgement that is itself a product of social location and commitments.  But of course, as I have laboriously discussed on the blog for years now, from a Brandomian neopragmatist perspective this is true of all our judgements.  So thinking in this way doesn’t rule out this kind of assessment of the self – in fact, we cannot but make this kind of judgement when we are engaging with people’s psychology.  But we need (I claim) an apparatus that (at a metatheoretical level) treats these kinds of judgements ‘symmetrically’.  There is no metaphysical fact of the matter as to which self is the true self.  This is a judgement that needs to be made as part of the ‘messy retail business’ of asking for and giving reasons.

And once we think in this way, I want to claim, we can see why the distinction between the ‘third’ and ‘fourth’ dimensions of power is somewhat arbitrary.  The self is always divided, and we are always making judgements about which elements of the divided self are part of the ‘real’ self, and which we should disregard.  In making such judgements we are (in something like Lukes’ vocabulary) drawing a distinction between ‘real’ preferences that are core to self, and ‘external’ preferences that can be sloughed off.  But this is, always, a matter of judgement.  Indeed, the core self is constituted as the core self by virtue of these judgements (and by the actions that form the judgements).  The self overall is socially constituted – and the distinction between the ‘core’ self that lies ‘beneath’ external influence, and the self shaped by external influence, is itself likewise socially constituted.  (Of course, this doesn’t contradict the idea that the self is also material, biological, etc. etc. – I hope the fact that these are simply different categories of claim has been well-established enough at this point on the blog.  Socially instituted normativity is fully compatible with a material world which we all inhabit and are made from, our material actions instituting that normativity.)

So – I think the Brandomian/Hegelian/psychodialectical apparatus we’ve got up and running here allows for a quite involved ‘cashing out’ of Lukes’ analysis of the relationship between ‘interests’ and ‘preferences’, via a complicated account of the internal division and social constitution of the self.

That’s one issue.  The other issue is power.  Lukes is interested not just in the divided self, but in how the preferences of the divided self are shaped by power.  I guess I’ll leave that issue for my next post.

Steven Lukes on power and interests

Back on the ‘note taking’ end of things – I wanted to put up some thoughts on Steven Lukes’ ‘Power: A Radical View’, the recent third edition of which I just read.  I’m going to try to do so very quickly, with apologies for possible loss of legibility.

The core of Lukes’ book, then, is the first chapter in which he outlines three accounts of power – ‘one-dimensional’, ‘two-dimensional’, and ‘three-dimensional’.  

One-dimensional accounts of power – which Lukes associates with empirically-oriented mid-twentieth-century pluralist political scientists like Robert Dahl – analyse power by looking at which actors within the political system are best able to have their preferences determine outcomes.  This kind of power is studied by looking at actors operating within a political decision-making framework, and seeing whose influence is greater.

Two-dimensional accounts of power – which Lukes associates with the work of Bachrach and Baratz – accept the basic Dahl picture, but then add the fact that the ‘rules of the game’ that structure decision-making can be structured in such a way that many social actors’ preferences never find expression within the political system, and thus certain outcomes can be excluded before any overt political decisions are made.  This kind of analysis needs to look beyond the empirical study of political decision-making, to study the decisions (or ‘non-decisions’) that are implicit in the institutional structures within which decision-making takes place.

Both of these approaches to power are looking at the relation between the preferences of social actors and the outcomes of a political system.  The focus is: which preferences exert influence, and why?  The one-dimensional account is more narrowly focused on key actors within a political system; the two-dimensional account is more broadly focused on the way in which social actors’ power can be used to shape that system.

Three-dimensional accounts of power – which Lukes is advocating – are focused on the formation of preferences themselves.  Lukes is interested in the way in which power can not just determine whose preferences are influential, but what those preferences even are.  The core of Lukes’ book is his criticism of analyses of power that do not take account of this ‘third dimension’.  As Lukes says in a later (second edition) chapter, then, his book isn’t so much about power, as it is about a particular category of the exercise of power: using power to shape opinions and preferences.

In a later chapter, Lukes considers the argument that really there are four dimensions of power.  Lukes isn’t fully convinced, and since these are just typologies obviously there is no objective fact of the matter.  But I can see a use for a four-part typology, so I’m going to roll with it.  From this perspective, we can distinguish two different ways in which power can shape opinions and preferences.  A ‘three-dimensional’ account distinguishes between the real underlying self (perhaps identified with a core human nature, or with basic interests) and the elements of self that have been imposed or manipulated by the external forces of power.  A ‘four-dimensional’ account sees the underlying self itself as constituted by power relations.  From this ‘four-dimensional’ perspective, external power relations enter into the very depths of the self; there is no self that can be understood independent of the analysis of power.  Orthodox Marxist ‘false consciousness’ accounts are plausibly understood as an example of three-dimensional analysis in Lukes’ sense – as is Chomsky’s ‘propaganda model’. For these theoretical approaches, the self has ‘real’ interests, and is then tricked by ideology or propaganda into adopting preferences that run counter to those interests. By contrast, there are a range of more ‘constructivist’ theorists – possibly Bourdieux, and certainly Foucault – who argue that the very core of the self is formed by power; there is no ‘real self’ that can be extracted from the process of power-sodden socialisation. Those approaches are examples of four-dimensional accounts.

So – Lukes is focused on the role of power in shaping preferences.  As he puts it:

Extremely crudely, one might say that the liberal takes people as they are and applies want-regarding principles to them, relating their interests to what they actually want or prefer, to their policy preferences as manifested by their political participation.  The reformist, seeing and deploring that not everyone’s wants are given equal weight by the political system, also relates their interests to what they want or prefer, but allows that this may be revealed in more indirect and sub-political ways – in the form of deflected, submerged or concealed wants or preferences.  The radical, however, maintains that people’s wants may themselves be the product of a system which works against their interests, and, in such cases, relates the latter to what they would want and prefer, were they able to make the choice. (42-3)

This third option is the one that Lukes endorses – but it raises two key problems.

The first problem is the ‘four-dimensional’ issue I just discussed: what if power is sufficiently central in shaping people’s wants and preferences that it simply doesn’t make sense to consider a counterfactual version of the person for whom those power influences were not operative?  What if the relevant counterfactual is not a version of that person who is “able to make the choice”, free of the baneful influence of power, but is, rather, a totally different person, because the power influences in question are constitutive of the person’s very identity?  This doesn’t seem like an uncommon or implausible scenario, to me.

The second problem is as follows.  Assume we can do away with this ‘fourth dimension’ problem somehow.  On what basis can we then claim to know what a person’s real preferences are – their underlying preferences, in the counterfactual scenario in which power has not influenced them? If the person in question disagrees with us about our assessment of their real preferences, who are we to say we know better?

This latter question is the main problem that preoccupies Lukes.  Lukes is critical – or at least very wary – of the orthodox Marxist category of ‘false consciousness’.  This kind of analysis – where the analyst claims to know people’s interests better than they do themselves – seems at best to risk paternalism, and at worst to justify full totalitarianism.  If the analyst can know people’s real preferences, which the people in question are blind to, why should those people be granted any ‘voice’ in the political process at all?  Can’t a revolutionary or technocratic cadre simply make decisions for the mass of people, based on an assessment of people’s real interests – and simply disregard dissent, on the grounds that dissent expresses (faulty) preferences, not (real) interests?

Lukes is alive to this risk.  But he also thinks that the liberal approach – to simply identify interests with preferences – is inadequately attentive to the ways in which power genuinely can shape preferences, in ways that genuinely do misalign people’s preferences with their real interests.  At base, for Lukes, if we are to do justice to political reality, we sometimes need to be able to say that people are simply wrong about their interests.

The relationship between interests and preferences is therefore clearly central to Lukes’ project.  In the second chapter of the book, he typologises ways in which theorists can relate these two categories.  In Lukes’ words:

There are alternative ways of conceiving of agents’ interests.  One way is the purely subjective way of straightforwardly identifying them with preferences…

These preferences may, Lukes writes, be “overt or covert”.  In either case:

Behind the equation of interests with preferences, overt or covert, lies the Benthamite view that everyone is the best judge of his or her own interests. (86)

Much more could be said about this.  The broadly subjectivist, utilitarian value theory that comes out of Bentham and informs so much of the tradition of economics, has very important internal schisms with respect to how it understands preferences – and different strands of thought within this tradition can arguably be identified with different positions within Lukes’ typology.  But that can all wait for another day – let’s just follow Lukes in putting this all in one box for now, where that box is labelled “Subjectivist liberalism: interests = preferences.”

Lukes continues:

An alternative way of conceiving interests is to see them as the necessary conditions of human welfare: what individuals generally need in order to live lives that are satisfactory by their own lights, whatever those lights may be. (86)

This approach is exemplified by (though of course not only by) Sen and Nussbaum’s ‘capabilities approach’.  The idea here is that we don’t claim to know better than the social actor in question what they really want.  Nevertheless, we want to acknowledge that social actors’ preferences may be a function of power – with different power relations, those preferences might be different.  Therefore, the argument goes, we need to expand the options available to social actors, by expanding their capabilities.  Social actors can then choose how to live their life – but from a much broader range of options.  This capabilities approach thus aims to walk a middle path between a pure ‘subjectivism’, which effectively refuses to acknowledge that preferences might be a product of circumstances (and therefore might be radically different if the available options were radically different), and a paternalist ‘objectivism’, which aspires to simply speak for social actors, by claiming to know their preferences (or interests) better than they do themselves.  I am, as I’ve said before on the blog, sympathetic to this capabilities approach (which no doubt inflects how I’ve characterised it), but that isn’t really the point here. 

Lukes continues:

A third way of conceiving interests is to see them… as constitutive of wellbeing: that is, comprising the leading of such a life itself…. Here one’s interests are given by the content of leading a worthwhile life. (87)

In other words, this approach claims that there is, at least potentially, a fact of the matter concerning what living a worthwhile life involves.  It is in the interest of people to live such a life.  And this is true regardless of their preferences.  Preferences can, on this account, be fundamentally misaligned with interests – people can be blind to their own best interests, and blind to how best to pursue the life they ought to live.

This final attitude is the most ‘objectivist’ – the observer can claim to know better than the person being observed what their interests really are.  Of course, this ‘objectivism’ doesn’t intrinsically lead to paternalism – other commitments are required for that.  But it is clear that it provides fertile ground for paternalist approaches.

Ok – so this is Lukes’ typology.  I’ve summarised his typology of forms of power, and his typology of accounts of interest.  These are closely related because, as I said above, Lukes’ central preoccupation is how to give an account of power that can adequately reckon with the role power plays in forming preferences.  I think it’s fair to say that Lukes’ account is motivated by his belief that we need to be able to talk about interests being distinct from (and a sounder basis for political action than) preferences, without falling into a dogmatic paternalism or totalitarian indifference to social actors’ wants.  I don’t really have a view about the extent to which Lukes succeeds in that project – the main function of Lukes’ book, as I see it, is to map the intellectual terrain.

I’m going to stop there, I think.  As I said up top, the purpose of this post is really just to summarise or take notes on Lukes’ book.  Having done that, though, I’m also interested in the issue of how the metatheoretical apparatus I’ve been laboriously developing on the blog intersects with these typologies.  Maybe I’ll try to address that issue in the next post.

From homo economicus to psychodialectics

Ok – in the last post I began footling around in the psychoanalytic or at least psychological or moral psychological dimension of the project here.  In this post I want to cover much the same ground in a slightly different way.  I don’t promise greater clarity, but hopefully taking different bites at this content will, over time, at least give me more resources for thinking all this through.  In this post, then, I want to organise things in a facilely ‘dialectical’ way, beginning with ‘homo economicus’ and then looking at how critiques of this simple model of the human animal can lead us to and through other theoretical perspectives in the social sciences broadly understood.

Homo economicus, then, is the much-maligned model of the human animal that sees humans as narrowly and rationally self-interested.  This is an instrumental sense of rationality – it is an entirely different concept of rationality from that which motivates (for example) Brandom’s metatheoretical project.  Within the methodological framework of ‘homo economicus’ we assume that humans act in their own self interest, and that their actions are guided by an instrumentally rational sense of the expected outcome of those actions.

Now, critics of economics as a discipline like nothing better than to criticise this approach, on the grounds that it is manifestly an unrealistic model of humanity.  And those critics aren’t wrong – it is indeed a manifestly unrealistic model of humanity.  I am, however, a defender of homo economicus.  I’ve written about this on the blog before, and I don’t want to recapitulate everything here, but basically I think homo economicus in fact is a pretty good heuristic for how people behave in economic life a large proportion of the time.  Workers don’t strike for lower pay; bosses don’t do their best to lose money; in general people do in fact pursue their economic self-interest, and whatever additional complexity one adds to one’s understanding of human behaviour, this is a good baseline model for thinking about the behaviour of most economic actors most of the time, in my view.

Moreover, as I have written before on the blog, I think the centrality of this model to normative economics is well justified by Dennis Robertson, in his 1954 lecture ‘What does the economist economise?’  There Robertson writes:

There exists in every human breast an inevitable state of tension between the aggressive and acquisitive instincts and the instincts of benevolence and self-sacrifice.  It is for the preacher, lay or clerical, to inculcate the ultimate duty of subordinating the former to the latter.  It is the humbler, and often the invidious, role of the economist to help, so far as he can, in reducing the preacher’s task to manageable dimensions.

I agree with this argument as a justification for the central role that ‘homo economicus’ plays in economic analysis.  We want political-economic institutions that don’t rely for their reproduction on unrealistically high degrees of altruistic behaviour on the part of economic actors.  Incentive-compatible institution design is extremely desirable, and homo economics gives us an extensive apparatus for thinking about incentive compatibility.  For these two reasons, I think homo economics has a lot to recommend it as a methodological heuristic. 

Nevertheless, the critics of homo economics do have a point.  Indeed, they have more than one point.  The first point is that economic actors just do frequently act in ways that break with the rational self interest heuristic of homo economicus, and it would be nice to have the analytic resources to think about all the many kinds of human behaviour that fall outside that narrow model.  To this end, a vast modern research programme in behavioural economics has developed, which aims to study how humans actually behave in economic practice, without relying on the heuristics of traditional neoclassical economics.  I am in general quite sceptical of behavioural economics as a research programme: I think its flagship claims are often weakly evidenced and vulnerable to replicability problems, and I worry that it is often, in practice, simply developing an alternative philosophical anthropology that lacks even the robustness of homo economicus as a methodological heuristic.  Nevertheless: obviously I shouldn’t dismiss this research programme on these grounds!  And I should do more reading than I have to date in behavioural economics.

The second set of worries about homo economicus interest me more.  This is a set of worries associated with institutional economics.  Institutional economics (or at least a lot of  institutional economics) focuses on the ‘rules of the game’ that structure economic life.  One way to do this is for the economic modeller to simply fiat the rules of the game, and then let loose homo economics agents within this exogenously researcher-specified institutional modelling environment.  But what if we want to think about the ways in which economic actors institute those rules of the game themselves?  There are substantial research programmes that aspire to address this question using the toolkit of traditional formal economics – i.e., using homo economicus.  I don’t want to dismiss those research programmes either, and indeed I want to spend a lot more time studying them.  Nevertheless, it seems to me prima facie plausible that homo economics doesn’t provide us with all the resources we might want for thinking about the ways in which economic actors institute the rules of the game that then, in turn, constrain their actions.

Here, then, we might want to turn to the other side of the dichotomy that Robertson articulated in the passage I quoted above.  Robertson distinguishes between the acquisitive and the self-sacrificing elements of human behaviour: selfish and altruistic dimensions of the human animal’s psychological makeup.  Focusing on the latter side of this dichotomy might lead us into the space of moral psychology: thinking about how the faculty of sympathy, and our social nature, inclines the human animal to non-self-interested behaviour in day-to-day economic action.  Crudely put, we might want to turn away from the Adam Smith of ‘The Wealth of Nations’, and spend more time with the Adam Smith of ‘The Theory of Moral Sentiments’.

And this is worth doing!  Let’s just fiat that we’re going to do it!  Indeed, a huge amount of what I’ve been doing on the blog over the last few years is working within a Brandomian pragmatist apparatus that sees human beings as intrinsically social – as intrinsically invested in occupying the perspectives of other humans, or intrinsically involved in specular social mediations, such that the self is constituted via its relation to the other.  One day in the medium term I hope to work through ‘The Theory of Moral Sentiments’ again with this apparatus in hand, and tie Adam Smith’s moral philosophy more closely to this contemporary neopragmatist apparatus.  But in any case – I endorse this broad category of move.

But now my question is: how exactly are we to interpret this kind of move?  One obvious way to interpret this kind of move is in terms of Robertson’s dichotomy.  Here we would see the human animal as possessing two broad categories of psychological disposition: those oriented towards individual self-interest, and those oriented towards sympathetic, cooperative, and other-oriented sociality.  As we move into the realm of ‘moral psychology’, we are (in Robertson’s terms) moving into the terrain of the preacher “lay or clerical”.  The question now becomes: what is the internal psychological make-up of the human animal, of any specific human animal, and how do these two orientations interact?

This perspective can in turn lead to a huge space of social-theoretic and moral psychological argument.  For example, one can think here of (often but very much not exclusively) conservative discourses whereby humans become moral agents by being embedded within the appropriate ethical community, such that the influence of the self-interested dimensions of human psychological faculties are controlled by the other-oriented elements.  Here there is a common narrative available: the process of socialisation renders humans full moral agents by shaping their psychology, via the other-oriented faculties of sympathy, in such a way that they can master their self-interested or appetitive drives sufficiently to achieve individual-level self-control, a precondition of full participation in the broader political space of collective self-government.  This of course isn’t only a conservative discourse – but in my view this kind of moral psychological reasoning looms particularly large in conservative political theorising, frequently justifying as it does the refusal to extend full rights and liberties to the huge swathes of humanity that are seen as lacking this moral-psychological capacity of full self-mastery, and who thus have not earned the right to participate in collective self-governance.

In my last, very muddy, post on this topic I suggested that this kind of narrative can easily be read into our out of Hegel’s (or Hegel-Brandom’s) reflections on the ‘moral valet’, and the distinction between particularity and universality.  From this (I am claiming, broadly conservative) perspective, agents who do not adequately participate in the broader moral community, and who do not permit their appetites to be adequately regulated by that participation, are, by virtue of this fact, not full moral agents, and thus do not fully enter into the proper space of political recognition.  I obviously disagree with this perspective, but I want to spend more time thinking about it, going forward, equipped with the moral psychological apparatus that I’m trying to articulate here.

In any case.  We are considering here a scenario in which we take there to be two broad categories of psychological impulse in the human soul or psyche: on the one hand, appetitive self-interest; on the other hand, other-oriented sociality.  Homo economics focuses simply on the former, with a correspondingly instrumental understanding of rationality; moral psychology aspires to incorporate both, but with an emphasis on the latter as the true locus of our humanity – included a non-instrumental faculty of reason.

But now enter psychoanalysis.  I want to suggest that the characteristic move of psychoanalysis is to occupy the terrain of moral psychology, but to analyse moral psychological categories – that is to say other-oriented categories – in terms that fold these categories back into an account of the psyche in terms of self-interest.  That is to say, psychoanalysis has an account of the psychic economy – a libidinal economy (though I have discussed the equivocal status of the category of ‘libido’ in psychoanalysis on the blog before) – that analyses the ‘moral faculties’ in terms of ‘baser’ instincts.  Psychoanalysis, moreover, does this in two ways.  First, it claims that the ‘moral faculties’ are constituted by the internal vicissitudes of those baser instincts.  Second, and relatedly, it claims that apparently other-oriented dimensions of the self are in fact driven by the gratifications or satisfactions they can provide to the self.  Thus, the moral faculty can be understood as providing its own characteristic gratifications, and we follow the guidance of that moral faculty to the extent – and only to the extent – that those gratifications ‘outweigh’ other gratifications within the internal psychic economy.  For this reason, the category of narcissism plays a complicated double role in Freud’s account of the psychic economy.  On the one hand, narcissism is a disorder that is characterised by excessive orientation to self over other.  On the other hand, narcissism can simultaneously function as a kind of master category, because every element of the psychic economy – including those that are mediated via others, or that are other-oriented in some sense – must, for Freud, provide gratifications to the self in order to function at all.  In this specific sense, for Freud, we are all narcissists, always – our actions are always in this specific, quasi-technical sense selfish or self-oriented.

Now there are many objections that are directed at this Freudian account.  One common objection is that the way the Freudian apparatus is set up, it is effectively irrefutable.  Whatever human behaviour we’re talking about, the Freudian apparatus can always come up with some complicated account of the internal psychic economy that can explain it in Freudian terms.  I think this is a reasonable objection to Freud and psychoanalysis – and I think it’s obvious that in practice psychoanalytic explanation frequently does operate in the kind of hermetic interpretive circle that an irrefutable apparatus makes available.

At the same time, I have a hard time joining the throngs of people who dismiss Freud and psychoanalysis wholesale on this kind of basis.  The reason for this is that it seems obvious to me that this kind of Freudian account of the psyche has all kinds of incisive psychological insights.  When I read Freud I frequently think “oh my god, this is full of shit” – but I also frequently think “oh wow yeah, that’s insightful”.  More specifically, the overarching idea that when we look to explain human behaviour we should look to see what gratifications that behaviour provides within the psychic economy – that self-abnegating behaviour is typically also self-satisfying in some complex way… well, I almost can’t imagine trying to understand human behaviour without this kind of approach.  It seems to me to be just straightforwardly true that when one looks into people’s motive structures, one finds complex, often multi-layered gratifications that are explanatorily clarifying about what’s driving people’s actions.  I don’t want to say that everything can be folded into the psychoanalytic framework, but it seems to me to be clearly valuable.

Now, the epistemic status of this kind of claim is a bit uncertain: basically I’m just saying that core elements of the Freudian apparatus seem right to me, based on my own experience of life.  I can see why we might not want to permit this kind of judgement the same kind of epistemic status as other forms of social-scientific claim.  On the other hand, I think it’s true – and, moreover, that judgement is based on experience (rather, than, say, gnostic insight), so you can hardly say it’s completely non-empirically grounded, even if a lot of that experience is difficult to fold into some of the most influential understandings of scientific evidence.  On the blog at least, this is good enough reason for me to pursue this line of thought.

But now comes another criticism of the Freudian/psychoanalytic apparatus.  This criticism is that the psychoanalytic apparatus evacuates real normativity in the way it reworks moral psychology.  And I think this criticism comes in two forms.

The first form is the idea that if our other-oriented behaviour is driven by the gratifications it provides within the internal psychic economy, then isn’t it really simply selfish behaviour, not other-oriented – and doesn’t this evacuate the moral content of altruistic actions or attitudes?  A version of this line of thought is articulated by the novelist David Foster Wallace (though not in reference to psychoanalysis) in his essay on Joseph Frank’s Dostoevsky book.  There Wallace writes:

Is the real point of my life simply to undergo as little pain and as much pleasure as possible?  My behaviour sure seems to indicate that this is what I believe, at least a lot of the time.  But isn’t this kind of a selfish way to live?  Forget selfish – isn’t it awful lonely?  …  But if I decide to decide that there’s a different, less selfish, less lonely point to my life, won’t the reason for this decision be my desire to be less lonely, meaning to suffer less overall pain?  Can the decision to be less selfish ever be anything other than a selfish decision?

Here the line of thought is that if altruistic, other-oriented behaviour is driven by the gratifications associated with an internal psychic economy – the pleasures and pains of the psyche now understood in a broadly psychoanalytic rather than a rational choice homo-economicus form – then this evacuates the other-oriented (i.e. moral) dimension of the behaviour.

I think this line of thought is a mistake, for two reasons.  First, understood on a broadly psychoanalytic model (which I’m suggesting is, in this dimension, compatible with the Brandomian-Hegelian apparatus I’ve been working through) the ‘internal’ self is never entirely ‘internal’ – the psychic economy really is mediated via others, and therefore other-orientation is baked into the human psyche at a depth level.  This doesn’t mean the individual gratifications aren’t real, and aren’t driving behaviour – but they can’t (or shouldn’t) be seen as fully ‘negating’ the other-oriented dimensions of self with which they genuinely are bound up.

Second, even if we don’t care about this first point… what does it matter?  If I behave well because it gratifies me to behave well… this is fine.  It doesn’t negate the good act that it was done because I took satisfaction in it.  The model of altruism here is that altruism has to be ‘pure’ if it is to be real – that it cannot be guided or shaped by personal gratifications or satisfactions.  If one takes this attitude, one is indeed flung into an ‘alienated’ moral psychological abyss whereby all human behaviour seems to be fundamentally amoral, as soon as one is honest about the gratifications in play – but this is due to a faulty theory of morality and moral psychology, not due to the intrinsic awfulness of any given human’s behaviour.

That’s one objection to this psychoanalytic model.  The second objection is very similar, but subtly different – this is, at base, the same objection that I have been discussing for years now in connection to Brandom’s pragmatist apparatus: the objection that all we are dealing with here are normative attitudes, and that normative statuses cannot be explained in terms of normative attitudes.  Here we assume that normative statuses find their locus in pro-social behaviour (because on our pragmatist account normativity is socially instituted), and because even our pro-social behaviour is driven individual-level satisfactions and gratifications, there is no way that real normative statuses can influence our psychology or actions.  This argument – for all the reasons I have laboriously rehearsed on the blog going back years – is fundamentally a category error.  Normative statuses are not a category of normative attitude – they are instituted by normative attitudes.  So the fact that all our actions are (on the psychoanalytic model) driven by particular gratifications within the individual psychic economy is irrelevant to the question of whether they are guided by normative statuses: particular behaviours, with corresponding gratifications, are all there ever empirically are, and normative statuses emerge out of them, rather than take their place among them.

Nevertheless – and this is what I was trying to get at in the previous post in this series – if we confuse the distinction between normative attitudes and normative statuses with the distinction between selfish and other-oriented actions, then the psychoanalytic explanation of other-oriented actions in terms of gratifications within the individual (albeit socially constituted and mediated) psychic economy, may seem like an argument against normative statuses – that is, against normativity as such.  The key thing we need to understand is that these are orthogonal issues.  There is nothing in the psychoanalytic explanatory apparatus that need debunk normativity (or other-oriented altruism) provided we have an appropriate understanding of what normativity (and other-oriented altruism) actually consist in.  If we do have that understanding, then we should be able to in principle incorporate the entire psychoanalytic moral psychology into our framework without it causing any problems for our understanding of normativity, altruism, ethics, and so on and so forth.  (We may of course want to reject much of the psychoanalytic apparatus on empirical or scientific grounds – that’s another issue: I’m just saying that there is no good reason to reject it on philosophical grounds.)  To the extent that we see the psychoanalytic framework as corrosive of our understanding of normativity, this implies that we have a faulty theory of normativity.  ‘Particularity’ and ‘universality’ don’t reside in different faculties within the human psyche – everything is particular.  Universality is instituted by particularity, and needs to be analysed at an entirely different level.

Ok.  This is basically the destination I wanted to get to in this post, and I want to review quickly the (facilely) ‘dialectical’ route I’ve taken to get here.

We started with homo economicus – a model of the human animal as straightforwardly self-interested, with rationality understood exclusively in instrumental terms.  We acknowledged that obviously there is lots of human behaviour that this model doesn’t capture.  We then introduced a two-sided vision of humanity: the selfish instincts and the altruistic instincts.  From this perspective, homo economicus models one dimension of humanity, and the other dimension of humanity falls outside the discipline of political economy.  We then made the shift to moral psychology – the philosophical or scientific study of the other-oriented side of human behaviour and the human psyche.  I discussed some (frequently conservative) discourse that emphasises this perspective.  We then brought in psychoanalysis, as a form of ‘moral psychology’ that folds other-oriented behaviour into a psychic economy driven by individual-level gratifications.  I discussed the fact that it is easy to see this perspective as corrosive of our ability to understand ourselves as genuinely normative creatures.  I then used the Brandomian-Hegelian apparatus to argue that this last move is a mistake – if we understand normativity properly, as something instituted by (rather than in opposition to) ‘particularity’, then there is no contradiction between the psychoanalytic attitude and seeing human beings as frequently pro-social creatures guided by norms.  Adopting this synthesis does, however, mean giving up the ‘crass’ distinction between self-interested and other-oriented behaviour with which we began, at a philosophical level.  (Of course, it doesn’t mean giving up the ability to call any individual action selfish or altruistic at an empirical level.)  On the other hand, it makes available the theorisation of a non-instrumental sense of rationality.

This last synthetic position is the one I endorse, philosophically speaking.  I also think that this kind of philosophical position makes available (or, rather, makes available the philosophical warrant for) an orientation to political economy that is much more ‘psychological’ than the homo economicus perspective with which we began.  The best ‘psychological’ (by which in practice I guess I mean sociological) political economists – and here I’m basically thinking about Du Bois and Weber; there aren’t many people (at least that I’ve read) in this category – are great because of their ability to bind together the ‘hard empirical facts’ of large-scale economic structures with a deep insight into the psyche of the participants in those structures.  This ability to ‘hold in one thought’ economic structures and the psychological dynamics associated with those structures is a hard thing to do – and of course is not something that a political economy that makes use of ‘homo economicus’ as its central model of the human is even trying to do.  As I said, kicking off, I am a strong defender of homo economicus.  But from my (admittedly idiosyncratic) perspective the reason homo economicus is a good model is because it is a reasonable first-pass approximation of the much more complex sense of self-interested behaviour associated with the psychodialectical / moral psychological framework I’m trying to articulate or work through here.  And (again from my perspective) I think that’s relevant to thinking about how the different methodological elements of political economy as a larger discipline hang together.

Ok.  That will do for now, I think – hopefully I’ll be returning to this terrain yet again in future posts.

The social constitution of pre-social norms

I fear this post is going to be even more fumbling and inchoate and difficult to make sense of than usual, so I’m going to try to at least be brief.  In this post I want to return again to the problem space of the ‘moral valet’, in connection to the psychoanalytic (or perhaps I should say, more expansively, moral psychological) sub research programme I discussed in my last post.

Let me start by outlining a vision of moral psychology that I think is very widespread, though perhaps more frequently and comprehensively articulated or theorised by conservative thinkers.  The picture goes something like this.  Human beings have varied internal psychological faculties.  Some of these faculties are more ‘base’ or appetitive – features or legacies of our animal natures.  Some of these faculties are more social or moral – oriented to community.  For some theorists, of course, we also have faculties that can make contact with the divine – but this is a resolutely secular blog, and I’m not going to discuss this kind of religious thinking.  Focus, instead, on these two ways of thinking about our psychology.  From this (rather crass) perspective we have a ‘lower’ self and a ‘higher’ self – one animal, one moral.

Within our Brandomian-Hegelian framework, I think it’s very tempting to map this opposition – between appetitive elements of individual psychology and community-oriented elements of individual psychology – onto the Hegelian opposition between the ‘particular’ and the ‘universal’.  From this perspective, the moral valet interprets social actors as driven by the appetitive and particular – i.e. non-normative – dimension of self; the great-souled perspective sees the social actor as guided by norms that inhere in the community.  The difference between non-normative and normative therefore maps onto the difference between the appetitive and the social.

I think that aligning these two oppositions in this way is a serious mistake.  I think if we take the Brandomian-Hegelian apparatus seriously, and apply it ‘symmetrically’, then we should see these two oppositions as almost entirely orthogonal – or, at least, as only related in a very complicated way that does not permit any direct mapping between the two halves of each opposition.  (And this is in turn connected to the ways in which the opposition with which I began – between the appetitive/animal and social/moral – is a flawed framework.)

On the Brandomian-Hegelian account, normativity is socially instituted.  Our practices have normative content because they are embedded in a social network the complex interactions of which establish that normative content.  In this sense, normativity is intrinsically connected to sociality – we wouldn’t be normative creatures if we weren’t also social creatures.  

[NB: In earlier posts on this blog I have made an argument about why the preceding sentence isn’t strictly true as articulated here, but I’m just going to ignore that for now – it’s not relevant or important.]

But this fact (about the social institution of normativity) doesn’t mean that the elements of self that are clearly oriented to complex social interaction are ‘more normative’ than other elements of self.  The entire self is embedded in the networks of social practices that institute normativity!  We can attribute normative content even to elements of self that are not themselves part of the ‘higher’ ‘moral’ faculties!  We therefore should not conclude from the fact that normativity is socially instituted that the ‘less social’ elements of self are less normative – that would be, in my view, a kind of category error regarding the way in which normativity is social.

The same goes in reverse as well – we can (from any specific social-normative perspective) collectively choose to interpret complex social practices as in fact failing to follow norms.  We can, likewise, interpret ‘particularist’ elements of self that break with apparently locally-instituted norms as correctly following norms that those apparent locally-instituted norms disregard.  [And, if we follow the pluralist approach that I’ve been advocating, we can potentially drop the “apparently”s from that preceding sentence – this is all a matter of social perspective.]

So, for example, and to get very very slightly more concrete…  One of the ideological perspectives that one could take on the opposition with which I began the post – the opposition between the appetitive/animal and the social/moral – basically inverts the normative valence of the poles of this opposition.  There are various perspectives on moral psychology that see the pre-social self as more authentic than the social self, and see the ‘particularist’ elements of self as the best guide to action.  Looked at through the prism of the Brandomian-Hegelian framework I’m advocating, this kind of perspective is also compatible, at a metatheoretical level, with the social institution of normativity.  Here one would simply need to say that the normative content granted to all elements of self by the self’s involvement in the right kind of recognitive practices implies that the elements of self that reject the norms adopted by other members of that social network are in fact the normatively right ones – i.e. that the ‘appetitive’ or ‘particularist’ elements of self are the ones whose implicit normative content should be recognised by the broader recognitive community, even though they are not.  From this perspective, at a metatheoretical level normativity is socially instituted, but at a theoretical level the social elements of self are amoral and the asocial elements of self embody the appropriate guide to action.  I’m not advocating this position.  But I’m saying that it is compatible with the metatheoretical framework I’ve been advocating, and therefore this metatheoretical framework cannot be taken to align with a ‘vulgar’ distinction between appetitive/particularist/amoral and social/universal/moral.

This is relevant to psychoanalysis because one of the ways in which psychoanalysis is often understood is as a ‘debunking’ discourse that exemplifies the perspective of the ‘moral valet’.  From this perspective, psychoanalysis takes the ‘higher’ ‘moral’ faculties, and reduces them to the ‘lower’ ‘appetitive’ ‘amoral’ faculties.  Moral reasoning is explained in terms of the vicissitudes of libidinal instincts, and so forth.  It would be easy to see psychoanalysis as therefore ‘reducing’ norms to appetites, and ‘reducing’ the universal to the particular.

My claim is that this is a category error.  I’m not entirely convinced that Hegel isn’t engaged in this category error – I don’t have a view on Hegel’s own text yet (though I’ll have to acquire one).  I’m not even fully confident that Brandom-Hegel isn’t engaged in this category error (though if they are the error is a subtle one).  But I do think it’s a category error.  If we take the Brandomian-Hegelian apparatus seriously, then all elements of the psyche can be understood as bearing normative content.  Moreover, ‘particularist’ psychological motivational structures just are how we follow norms.  We cannot divide the self up into particular and universal – personal and social – components, as if one bit of the self is particular and another bit is universal.  Rather, universality operates via and emerges from particularity.  Everything is (at least potentially) both particular and universal simultaneously.  An analogy to the broader Brandomian framework here would be the distinction between normative attitudes and normative statuses. It’s not that some of our normative attitudes are attitudes, and some of our normative attitudes are statuses. All we have, at a concrete empirical level, is normative attitudes. It’s just that those normative attitudes can institute normative statuses. Any knowledge of or interaction with normative statuses occurs via normative attitudes – but this fact shouldn’t be taken to debunk the idea of normative statuses. Similarly, at the psychological level, the fact that all we have are ‘particularlist’ psychological motives shouldn’t be taken to debunk the idea that we also have moral selves. It’s just that (as psychoanalysis argues) those moral selves are instituted via the complex interaction of different ‘particular’ psychological drives. In other words, this way of understanding the Hegelian-Brandomian argument is strongly compatible with psychoanalysis’s refusal to draw a sharp qualitative line between ‘higher’ and ‘lower’ faculties, instead explaining both in terms of the same basic instincts.

This has been an enormously inarticulate post – but hopefully it is at least broadly legible what kind of thing I’m getting at.  I’ll circle back around this terrain again in future posts, and hopefully attempt to improve on clarity.

Three subsidiary research projects

Or maybe subcomponent research projects, I’m not sure how to put it.  In any case, I recently concluded a long train of thought about Brandom’s ‘A Spirit of Trust’ on the blog.  I feel those posts laid out my basic interpretation of (and some minor tweaks to) the Brandomian-Hegelian apparatus.  All that’s now basically in the bag, from my perspective.  This is the core philosophical or social-theoretic apparatus that I endorse, on the blog, and this is the apparatus that I am largely going to be working within, moving forward.  But I want to note three trains of thought that emerge from that interpretation or that apparatus, and that I want to pursue over the coming months and years.

First, there’s the issue of the relation between the Brandomian-Hegelian apparatus, and liberal pluralist political and social theory.  I’ve already written about this a bit on the blog (it’s been one of the main topics of recent posts), but I’ve done so in my usual sloppy and under-informed way.  What I want to do over the next… however long, then, is to try to work through this line of thought a bit more carefully.  This means doing my homework: reading more liberal political theory, reading more Hegel, reading more Hegelians, and reading more secondary literature on Brandom.  This is all quite time-consuming, and time is precious, but that’s how it goes, at the end of the day.  Hopefully, if I do all that, I’ll be able to write up some of the kinds of things I’ve been saying on the blog (or at least something in their space) in a more formal academic style.  That’s “off-blog” work, so to speak – the blog is for thinking things through, not for nailing things down.  But I wanted to note it here as a research project I’ll be pursuing.

Second, there’s the issue of coercion.  Brandom basically brackets this question completely when thinking about the institution of norms: his theory of normativity is sanctions-based, but he is interested in normative sanctions – sanctions that gain their power from “the force of the better reason” – not ‘material’ sanctions.  [These categories are going to end up being mutually implicated, but I think it’s clear enough what I’m getting at.]  Clearly, though, in the real world, coercive power is central to shaping our normative worlds.  So I want to think this through, in the broad philosophical register of Brandomian-Hegelian neopragmatism.  This second research project is, in a way, the most important to me: it is where my interest in philosophical pragmatism and my interest in institutional political economy most overlap.  But (in my usual perverse and probably self-defeating way) for that reason I’m going to leave it till last – I will come back to this question at some later time yet to be determined.

Third, there is the issue of “psychodynamics”.  One of the consistent threads on the blog has been me making sort of gestural remarks about how the things I’m thinking about intersect with psychoanalysis.  These remarks are gestural because this line of thought is emphatically still in the “thinking things through” space.  But this is the line of thought I want to focus on in the short term, spending at least a post or two on it in the near future.

Crudely put, then, there are three lines of thought here that you could see as focused on the intersection of the Brandomian-Hegelian social theoretic apparatus with:

• liberal pluralist political theory
• loosely Marxist conflict theory
• loosely Freudian psychodialectics

Arguably all three of these lines of thought (and certainly the last two) push the Brandomian apparatus in a less ‘rationalist’ direction.  But I think at this point I have done enough to make the case that even very robust (though not maximally robust – see my post on the fragility of norms) forms of rationalism will end up being compatible with these lines of thought.  That is to say: I think I have the resources to make the case that rationalism is not thrown out the window when we use this apparatus to drill down a bit further into liberal normative pluralism, conflict theory, and psychodialectics.  Let’s see how it goes.

Three distinctions within Hegelianism

I feel like the series of posts recently concluded did quite a lot of intellectual heavy lifting, with respect to Robert Brandom’s interpretation of Hegel – at any rate, it felt difficult.  I’m going to relax a bit for a little while, then, on the blog, with some much more basic remarks.  In this post I want to talk about three ways you can slice up the Hegelian tradition.

In my usual manner, I’m doing this as a way to get my thoughts in order before I start on a round of serious reading.  It’s not that I’ve read nothing in the Hegelian tradition, but clearly I need to read a lot more.  So all the usual caveats apply.

That said, here are three ways to divide up Hegelianism.

First: there’s a distinction between more metaphysical or theological Hegelianism, and more pragmatist or social-theoretic Hegelianism.  Obviously the Brandomian apparatus this blog has been working through is a social-theoretic or pragmatist interpretation.  (Moreover, in my recent post on Derrida and ‘A Spirit of Trust’ I tried to extract a still more anti-metaphysical Hegel out of the already extremely down-to-earth and pragmatist Hegel than Brandom gives us.)  Obviously enough, I want to position myself on the pragmatist and social-theoretic side of this divide.

Second: there’s the distinction between ‘left’ and ‘right’ Hegelianism.  Again, my reading in the actual intellectual history here is patchy at best.  Still, as I understand things, this boils down to a disagreement about whether Hegel’s philosophy of history points towards a basically conservative vision of existing society as embodying the recognitive ideals Hegel advocates, or whether it critically points beyond existing society to some future unrealised society that would better instantiate those ideals.  One wants to be a bit careful how one thinks about all this (for reasons that I’ve discussed, though perhaps not clearly or explicitly enough, in previous posts on the blog – basically anti-teleological reasons).  In any case, I take it that in the broadest sense I am aligned with the ‘left Hegelian’ side of this divide – I’m interested in the critical potential of the Hegelian apparatus.

I think these two distinctions are extremely commonplace.  Maybe the third one is too, but I want to draw it anyway.

Third, then: there is a distinction in how ‘homogenising’ or ‘totalising’ we take the Hegelian apparatus to be.  Hegel is, in general, and speaking very crudely, interested in unity in difference, and difference in unity.  I want to say (in very rough and ready terms) that which of these two sides of the Hegelian story we emphasise (unity or difference) has a big impact on how we interpret his apparatus.

So: the Hegelian apparatus is, I think, frequently interpreted as pointing in something like a ‘totalitarian’ direction.  That is to say: all the complexity and internal difference or diversity of our societies is ultimately going to be folded into a unified ‘We’, and the individual can only truly be understood as a subcomponent of that ‘We’.  From this perspective the ‘We’ is privileged within “the ‘I’ that is ‘We’, the ‘We’ that is ‘I’”.  Hegelianism in this form is seen as one of the key expressions of the political idea of the subordination of the individual to the large-scale collectivity.  This is the charge that Popper levels at Hegel, for example, in seeing Hegel as an enemy of the open society.

But this interpretation of Hegel is also adopted by admirers of Hegel (though phrased in a more appealing or less hostile way, of course).  Moreover, it cuts across ‘left’ and ‘right’ Hegelian divides.  You can see the self as subordinated to the collectivity in conservative terms (provided we’re talking about existing social institutions).  At the same time, there’s a very significant tradition within Marxism that mobilises this homogenising idea of totality in a ‘left Hegelian’ way.  In my view, Lukács is by far the most important figure in this ‘totalising left Hegelian’ tradition.  Clearly I’m being a bit fast and loose here. Nevertheless: I want to align with (many of) Lukács’ critics, by claiming that Lukács adopts an incipiently totalitarian understanding of Hegelian totality.  That is to say: Lukács’ homogenising concept of the class consciousness of the proletariat (which is also the subject of history) is intrinsically hostile to pluralism within the society Lukács aspires to build – and that’s a bad thing.

[Interstitial note: These kinds of claims often make a lot of Marxists very angry – they insist that ‘totalitarian’ is a bogus, reactionary category, and that “totality” in the Hegelian sense doesn’t mean anything like this regardless.  In my view, though, there is pretty clearly a large left tradition that insists on the crucial political importance of the subordination of the individual to a collectivity that is theorised in homogenising, totalising terms.  I also think it’s pretty clear that Lukács is a major representative of that tradition. I want to reject this tradition.]

Critical Hegelianism (‘left Hegelianism’) doesn’t have to be this: it doesn’t have to have this ‘totalising’ understanding of the telos or ideal of the Hegelian system. In the last long series of posts on the blog, I made the case that the Brandomian-Hegelian apparatus can be deployed (with only very minor tweaks) in the service of a quite complexly elaborated account of normative pluralism within a highly internally diverse society.  This apparatus opens possibilities for ‘Hegelian’ critique that are quite different from those grounded in Lukácsian totality.  We could, for example, criticise social and political institutions precisely on the grounds that they suppress diversity.  And we can mobilise critical resources that are grounded not in ‘totality’, but in some specific slice – or slices – of social reality.

This is the line of thought, at any rate, that I’m currently pursuing on the blog.  One could argue – perhaps with some justice, I don’t know – that it’s perverse to characterise this as a ‘Hegelian’ line of thought.  But I think the apparatus Brandom develops in ‘A Spirit of Trust’ gives us very considerable resources for a ‘microfounded’ pragmatist theorisation of pluralism.  So I want to continue to explore this line of thought, at least for a while.

Breather

The last post concludes this line of thought on the blog concerning ‘A Spirit of Trust’.  There’s more that I want to say, eventually, about ‘A Spirit of Trust’, and about these themes, but for now I’m going to take a break from Brandom blogging.  Thank you for reading along!

The fragility of norms

Alright. In my most recent series of posts I’ve moved at breakneck speed through some of my thoughts about the second half of ‘A Spirit of Trust’.  I think I’ve now articulated, in some form, all of the critical remarks I want to make about that work.  In this post I therefore want to back up, and return to the problem space that structured my earlier series of posts on Brandom – the problem space opened by (among many other works) James Conant’s piece on Rorty and Orwell in the Brandom-edited volume ‘Rorty and his critics’.  I’m afraid that I’m writing this post extremely quickly, so I’m not going to go back and reread Conant, and I apologise if this significantly damages quality control here.  On the other hand, I think the core arguments I want to make are clear enough.

So: the scenario we’re considering is the scenario that Orwell presents at the conclusion of his dystopian anti-totalitarian work 1984.  Our protagonist, Winston Smith, has engaged in resistance against a totalitarian state, has been captured by agents of that state, and is undergoing torture.  The goal of the torture is not just to break Smith, but to transform his beliefs and values.  Orwell’s preoccupation, in this thread of the book, is the question of whether a truly totalitarian state is capable of reshaping the values of its citizens (or subjects) such that a critical standpoint against the state’s official values is, in effect, impossible.  Conant is interested in this scenario because it raises a question that has frequently been raised against pragmatists: is it really possible for pragmatists to hold fast to the values we want to endorse, in a scenario where a powerful social context imposes different, evil values?  And what does this question imply for the legitimacy of pragmatism as a philosophical position?

Now, this question has a number of different elements, and I think some of those elements are often treated together in discussions of these issues, in a way that can be confusing.  I want in this post to separate out several different issues.

First issue: if we adopt a pragmatist stance (i.e. claiming that values are instituted by social practice) is it possible for us really to hold values at all?  Isn’t normativity intrinsically and necessarily a phenomenon that must transcend specific social context and attitudes?  Doesn’t pragmatism abandon true normativity from the outset, by the very nature of the doctrine?

This is the objection to pragmatism that I think Brandom has relentlessly and extremely convincingly responded to, across more or less his entire body of work.  In Brandom’s vocabulary, this is the objection that pragmatism reduces normative statuses to normative attitudes, and thereby evacuates the substance and meaning of normativity altogether.  This is, in other words, the objection that pragmatism necessarily takes a reductionist or small-souled or alienated attitude towards values and normativity in general.  I take it that Brandom has extremely robustly addressed this objection to the pragmatist project, and I don’t intend to devote any more time to it in this post.

But now come a series of slightly different objections.  I think these objections easily collapse into the first, and defenders of pragmatism like Brandom can therefore easily take it that these objections are adequately responded to by responding to the first objection.  But I don’t think this is correct – I think these objections are different, and need to be tackled using different theoretical resources.

Second objection, then.  This objection grants (at least for the sake of argument) that values can be instituted by social practice, but worries that, if we grant this premise, we have effectively signed over our ability to evaluate what is right and wrong to the most powerful forces in society. The worry is that, for pragmatism, might makes right. Here the line of thinking goes: social practice determines values; power can determine social practice; therefore power can determine values.  The worry, in other words, is that Orwell’s antagonist, the state torturer O’Brien, is correct when he argues that ‘the party’ can decide what is true and false, what is good and bad.  The worry is that pragmatism does not give us the ability to locate a ‘critical standpoint’ that falls outside the power of society.

One common response to this worry is to abandon pragmatism in favour of some social-context-transcendent normative standpoint: God; an innate individual faculty of reason; the materiality of the body (which is not socially determined); and so on and so forth.  Another approach is to make a transcendental argument about the norms to which we are committed by virtue of engaging in normative social practice at all.  This kind of argument is often classified under the heading ‘pragmatist’, because it does, after all, ground our norms in social practice.  Habermas is making this kind of argument in ‘The Theory of Communicative Action’, and Brandom-Hegel is making this kind of argument in one key thread of ‘A Spirit of Trust’.  But this is not the kind of pragmatism I want to endorse.  In my view, there is no reason to believe that a transcendental argument can in fact derive norms that are substantive enough to serve as the kind of critical standpoint that can serve as a bulwark against the Orwellian totalitarian scenario we’re considering.  So I want to reject this approach too.  I want to claim that the norms with which we can resist our social context need to be derived from the “messy retail business” of everyday ground-level normative social practice, not any kind of transcendental condition of possibility of those practices.

So how can this be done?  Here my argument is simple: this can be done because the “totalitarian worry” has a vastly too simplistic vision of sociality.  Consider again Saul Kripke’s interpretation of Wittgenstein, which concludes that the substance of rule-following must be determined by ‘society’ or ‘the social’.  As Brandom argues in ‘Making It Explicit’, Kripke’s argument here simply black-boxes ‘the social’, treating it as an undifferentiated homogeneous category to which the individual can (and should) defer.  But ‘the social’ isn’t like that.  There is (to coin a phrase) no such thing as society – not in the sense of a single homogeneous “society” that can be treated as a single unit.  (We need to reject Durkheim and Rousseau as well as Kripke.)  Rather, ‘the social’ or ‘society’ is near-unimaginably complex, constituted by countless different subcomponents.  In ‘Making It Explicit’ Brandom initially draws this distinction in terms of an “I-We” versus “I-Thou” understanding of sociality.  But the reality is that an “I-Thou” model of sociality is also too simplistic, because the relevant unit of analysis is not really individuals, but rather social practices.  Every society is made up of many different individuals; and every individual is made up of (or rather engages in) many different social practices, all of which have normative content – and that normative content may differ widely between social practices even when those practices are carried out by the same individual.

Moreover, we can here draw on the concept of “pragmatic projection”, which plays a major role in ‘Between Saying and Doing’ but which Brandom doesn’t make much use of in his other works.  In pragmatic projection, we take some feature of a practice and extrapolate it to novel circumstances in which the practice currently does not imply.  Brandom derives this concept from Wittgenstein, who uses it to explain how we can construct new concepts by analogy from the application of old concepts.  Henry Staten, in his work on Wittgenstein and Derrida which I discussed a couple of posts ago, also makes heavy use of this Wittgensteinian concept (though not under the same name).  For Staten, this concept is crucial to understanding the ways in which Wittgensteinian sociality is always open to future uncertainty: we cannot treat ‘the social’ as a set of rails running off to infinity, in part because every practice can in principle be applied in lots of different ways, and the implications of that practice can be applied in contexts that have not yet been considered, in completely unpredictable ways.

I think that all of this is right.  But one of the things this means is that we can find “normative kernels” within even the most apparently unpromising and banal everyday practices.  If I take the normative principles that I find in such-and-such an everyday practice, and apply them to such-and-such a novel context, I may derive normative guidelines for that new context that are entirely novel, and that do not arise out of any of my existing normative commitments.  For this reason, the vast complex of social practices that we all engage in every day are always throwing out new normative possibilities, which can in principle serve as critical standpoints that can be levered against current social practice within the relevant domain.

The point I’m making here is that our sociality is immensely, basically incomprehensibly complex.  This is so in very obvious ways: every society contains countless shifting subcommunities (some of which may be very fleeting), and each subcommunity can in principle institute its own norms, in a way that can create critical (countercultural) norms that stand in tension with the larger social norms that subculture members also institute or observe.  Moreover, because the institution of normativity is diachronic, not just synchronic, the relevant subcommunity can in principle have only one member in the present moment: figures from the past, and imagined figures in the future, can in principle serve as the community of recognition that allows critical norms to be instituted in the present.  Finally, and as I was just saying, this all applies even ‘below’ the level of the individual – even individual social practices contain normative potential that can be mobilised against a larger social context.

What all this means is that if one has an appropriately complex model of the institution of norms by social context, there is essentially zero risk that a homogeneous “society” can suppress critical standpoints, because no imaginable society is even close to homogeneous in the relevant sense.  Real sociality isn’t even in the same galaxy as this “one-dimensional” vision of sociality. Our societies- all societies – are irreducibly plural.  If power wishes to suppress this pluralism, the suppression must take place, at least in substantial part, by force – but that force will not suppress the availability of critical norms, it can only suppress their institutional expression.  In other words, if we adopt the model of sociality I’m advocating, this absence of critical potential simply isn’t something that we need to worry about.

Now comes a third objection to pragmatism: relativism.  if we reject the idea that our norms emerge from some source other than social practice, and moreover if we reject the idea that there is a transcendental argument that implies that any and every norm-instituting social practice must institute these specific norms – well, doesn’t this imply that there is no necessity to the norms that we endorse? Does this mean that we could, in principle, pick any norm for wholly contingent reasons? And doesn’t this imply an “anything goes” relativism?

I think there will be lots of people for whom this argument is compelling.  But in my view this argument is a kind of category error.  The key point here is that one does not select one’s norms for no reason at all – one selects one’s norms as a rational and moral agent inhabiting the space of reasons.  The argument that the Brandomian/pragmatist approach, once it is untethered from transcendental arguments for any specific norm, amounts to an anything goes relativism, seems to me to require the tacit premise that “messy retail level” reasons for commitments aren’t real reasons.  From this anti-pragmatist perspective, the only reason that counts as a real reason is a reason that everyone ought to accept, by some universal feature of our nature as sapient creatures.  But I object to this premise – I just don’t think that this kind of necessary universal bindingness really is a necessary attribute of good reasons.  So from my perspective, the day-to-day give and take of the contingent messy retail business of asking for and giving reasons just is where our reason resides: all of it.  From my perspective, this denial of contingent messy retail business reasons the status of ‘real reasons’ is itself a sort of denial of the real substance of rationalism.

Again, I don’t expect this argument to be compelling to everyone.  But for me, this argument is the core of my pragmatism.  Criticisms of this argument, from my perspective, amount to rejections of pragmatism, and an insistence that real reasons – good reasons – must be social-practice-transcendent in some sense (even if that sense is a practice-grounded transcendental argument).  

Now comes the final objection to pragmatism that I want to discuss here: the argument that pragmatism implies a high fragility of norms.  Let’s grant that social practice really can institute genuine normativity.  And let’s grant that it can provide the resources to explain how social actors can reject the dominant norms of their social environment, even though (on our account) those norms are instituted by the social environment.  And let’s grant that this kind of pragmatism isn’t a form of relativism, because it can give an account of why reasons for adopting one belief or norm rather than another really are good reasons.  Let’s grant all this.  The final objection runs something like the following: “Ok, but doesn’t this imply that any given set of norms, which are on your account fully contingent and social, can simply be abolished?  Doesn’t it imply that norms – like the humans and communities that institute them – can be wiped off the face of the earth?  Isn’t it in principle the case that a society, by removing the social conditions that produce a norm, can destroy the norm itself?”

And my answer to this is basically: yes, my position does imply this.  Norms, like people and communities, are fragile.  There is no guarantee that they will survive the vicissitudes of history.  In fact countless norms have not survived those vicissitudes.  Humanity has existed for a very long time.  There will be many many religions that once existed, and which meant the world to the people who believed and practised them, that have been wholly destroyed and forgotten – they are unreconstructable.  No human being will ever again believe in those gods, or in the specific norms associated with their worship.  Like the contents of so many lost texts – and so much oral tradition – these norms are no longer part of the collective resources of humanity.  This conclusion is simply one of the consequences of a properly historicist and pragmatist understanding of what our norms are, where they come from, and how they perish. Why should our norms be any different?

In this fundamental sense, then, there is no guarantee that a totalitarian state – or any other social structure – cannot successfully abolish the norms we value.  My final claim is that this is simply how things are.  Our norms, like our communities and our lives, are fragile and contingent and will one day be swept away by remorseless history – but, like our communities and our lives, they are no less real or meaningful for that.