Rough Theory

I’ve just been reading R.A. Wild’s Heathcote: A Study of Local Government and Resident Action in a Small Australian Town (1983). This work operates within a more traditional concept of politics than does Richards’ book, about which I wrote below: it focuses on “classical” political conflicts between governmental institutions and organised citizen political movements, and operates largely within a classical Weberian framework for understanding the trajectory of political institutions. Among other things, this approach leads to some strange anachronisms: it sounds historically out-of-synch, even for a book written in the early ’80s, for an author to relate with confidence that (p. 57):

Generally the modern state has become more bureaucratic, interventionist and centralised. Bureaucratic expertise has become necessary to administer the complex tasks of the state. Intervention has become increasingly positive and directive and has encompassed what were traditionally private areas of economic activity. Centralisation has occurred partly as a result of the need for national and international long-term economic planning.

The book therefore occupies a thought-space that is historically just on the cusp of the powerful reassertion of the market that would undermine the way in which sociologists had for decades conceptualised the secular trends of politics, economy, and society. This historical tension is evident in the economically liberal views the book documents in the Councillors, which Wild explores in relation to constraints on local government, without, however, calling into question Weber’s essentially linear historical narrative of the sweep of rationalisation through history, culminating in the dominance of formal, procedural rationality in the bureaucratic service of elite interests under the planned capitalist economy (see p. 59-63).

I tend to think it is quite important to keep Weber’s narrative in mind in attempting to develop a critical theory – although I would strip from that narrative its world-historical dimensions, and reinterpret those as a projection back into time of dynamics that are specific to capitalism – thus opening the path to reclaiming Weber’s work as a critical analysis of oppressive trends within a particular society, rather than an analysis of the unavoidable consequences of rationality per se. (Habermas has approached this same issue in another way – reinterpreting the essence of the world-historical process, and distinguishing the potential logic of that process, from the actual historical realisation of that process under capitalism – I’m sure I’ll have the opportunity to comment more on his narrative over time.)

Leaving aside the issue of how to appropriate Weber’s work for critical theory, I think that it is particularly important to remember his insights now, in the wake of a period of market triumphalism in which many analysts have forgotten the ways in which capitalism is not the opposite of state planning and bureaucracy, but instead generates its own pressures for planning, bureaucratisation, and regulation.

At the same time, in my opinion, Weber’s epistemology hinders his analysis, and often strains against the critical logic of his own analysis. I’ve mentioned the world-historical dimension of his narrative – the notion that the trends he is describing are on some level transhistorical – as one potential epistemological problem. Another – and one more central to Wild’s book – is Weber’s notion that his analytical categories are “ideal types”. As Wild explains the concept (p. 19, ftnt 1):

As ideal-type models and analysis are sometimes the cause of confusion, I shall comment briefly on the approach. Elsewhere I defined ideal-type modes as ‘hypothetical constructs developed from empirically observable or historically recognisable components with the aim of making comparisons and developing theoretical explanations’. The basic assumption behind ideal-type analysis is that a total reality in all its complexity cannot be known – cannot be objectively grasped by the human mind. In observing reality we are confronted with a chaos of sense impressions. Ideal types are used to help us abstract from this chaos and thereby make knowledge of the world possible through a framework of prior conceptualisation. Ideal types themselves do no generate knowledge or provide explanations, rather they are a framework for research and point us towards asking paticular questions and formulating useful propositions.

The method of reasoning used in this passage is actually quite common in the social sciences, and is not limited to sociologists who use ideal types, or even to social scientists who invoke relativist concepts. The passage is worth unpacking, to draw out its assumptions and non sequiturs.

Wild starts with a statement that would be quite difficult to disagree with, regardless of the reader’s epistemological or normative stance – namely, that “total reality in all its complexity cannot be known”. This very strong claim provides an aura of authority that can make it easy for the reader not to realise that the statements that happen to follow sequentially in the passage, nevertheless do not actually follow logically from the initial claim.

So Wild moves from “total reality in all its complexity cannot be known”, to the much more controversial and contestable claim that “in observing reality we are confronted with a chaos of sense impressions”. A wide variety of perspectives, from sociobiology to theistic approaches to critical theory, would disagree with this second claim, albeit for different reasons, while having no real difficulty with the first claim. Wild then argues that ideal types “help us abstract from this chaos and thereby make knowledge of the world possible through a framework of prior conceptualisation” – a statement whose logic suggests that our most fundamental analytical categories somehow pre-exist and are unshaped by our interactions with the world around us, but are also fundamentally random. In this formulation, there is no check – whether social or natural – on the ability of a particular anaytical category to arise, to appear persuasive, or to succeed in orienting our actions in the world. Analytical categories in this approach are “mind play” – all approaches as intrinsically plausible as one another, with only convention dictating which categories pique our interest at a particular time.

In practice, very few social analysts – and least of all Weber – actually practice their academic work in a manner consistent with these philosophical claims. Instead, they procede very much as though they are making interpretive claims about a non-chaotic reality: they make arguments in favour of their interpretation, and against others’; they refer to supporting evidence; they even occasionally cede ground to a more convincing analysis. Where the explicit epistemology would seem to support an “I’m okay; you’re okay” relativism, analytic practice suggests an awareness of shared normative standards to which we can appeal when assessing the worthiness of our analytical categories. This is indicated quite clearly in the passage quoted above, where Wild also describes ideal types as “constructs developed from empirically observable or historically recognisable components” – a formulation that invokes a common ability for empirical observation that sits in clear tension with a “chaos of sense impressions”.

So why does practice suggest a tacit non-relativist epistemology, while the explicit theory talks in relativist terms?

I think there are several levels of answer to this question and, as always, I won’t be able to do full justice to any of them here.

One issue – on which I have written before – is that Wild (and other social scientists invoking similar explicit epistemologies) has a very absolutist concept of truth. Reading between the lines of his passage, to be “true”, something must be “objective”, “total”, and, a bit more tacitly, it must be abstract and sit outside the chaotic detail of the real world. While some approaches might appeal to religious or natural truths that claim to meet these criteria, a social theorist who posits that profound changes have taken place over time in the structure of human society, culture, and even fundamental categories of perception and thought, cannot logically aspire to truths that are objective, total, and abstracted from concrete historical experience.

There are several ways social scientists try to thread their way around this dilemma.

Possibly the most common approach is simply to ignore the issue – to continue to analyse the forms of perception and thought that characterise earlier eras as historically specific, culturally determined, etc., while not turning this same lens on our own analytical categories. When asked, most historically-sensitive social scientists would claim that they are as historically and culturally embedded as their objects of analysis – but there is nothing internal to their analysis itself that explains how they could potentially escape the relativist trap and make, to use Habermas’ vocabulary, truth claims about their work. There is also the minority who do seem to maintain that, while the object of analysis is culturally and historically determined, the social scientists’ own categories represent “discoveries” that transcend social, cultural, and historical limitations. I believe it was Marx who described this approach as the claim “that there used to be history, but now there is none”.

Another is the approach adopted in Weber’s explicit epistemological statements, as well as in a variety of other (sometimes far less sophisticated) relativist approaches: to accept that historically-sensitive social scientific research cannot attain objective, total truth – and therefore to repudiate the notion that social science is in the “truth game” at all. This position is seldom consistently maintained – tensions between explicit epistemological expositions, and the actual logic of the social analysis (such as in Wild’s passage above) are more the norm than the exception.

A third is most recently expressed in Habermas’ work (although many of the social scientists who draw inspiration from Habermas seem loathe to follow his line of reasoning to the end). Habermas provides a sweeping historical/evolutionary narrative of the emergence of reason in history – effectively, a very sophisticated way to explaini why it could potentially be the case that “there used to be history” – people used to think and perceive the world very differently from how we do – but “now there is none” – we can make and assess potentially valid truth-claims about the individual, social, and naural worlds.

A fourth – and this is what I would like to work toward, and what I believe a fully self-reflexive critical theory would actually require – would seek to embed our analytical categories within our own social experience (would move beyond the claim that there is no longer history), and would explain how to move beyond the notion that we perceive “a chaos of sense impressions” without, however, appealing to biological or theistic understandings of how perception and thought come to be structured. This approach would bear a number of analytical burdens, among them moving away from the concept that valid truth claims can be made only with reference to timeless, abstract, transhistorical, asocial, absolute values, and moving equally away from the concept that no truth claims can ever be made (e.g., because timeless, abstract, etc., valules are impossible) – and moving toward a concept of “truth for us” – of socially-generated and shared forms of perception and thought that can validly perceive and make judgments about our society precisely because these judgments are integrally related to the society we create.

There are many other criteria a fully-self-reflexive critical theory would need to meet (not least among these, providing an analysis of why relativist and absolutist approaches to truth are so common in this historical period), but proposing an alternative concept of truth is a pivotal step. Understanding truth as the non-random, but still socially-embedded and non-absolutist, collection of things we are collectively teaching ourselves at this historical moment, begins to overcome the subject-object divide that haunts most analyses, and points to the need for a theory of how our subjectivity – our forms of perception and thought – are constituted by, constitute, and are in many ways coterminous with the various “objectivities” of our society – the nature of the economy, the state, civic institutions, gender relations, race relations, family structures, etc.

It’s always a bit disappointing not to be able to “walk the talk” in this blog, but there are limits to the medium, as well as advantages. I’m sure this theme will recur, and I’ll hopefully be able to give a clearer sense of where I would like to go, and how I might attempt to get there, in future entries.

I have been reading Manuel Castells’ The Rise of the Network Society, 2nd ed., 2000. In this post, I don’t intend to undertake a systematic and comprehensive review of Castells’ ambitious work, but I would like to comment on Castells’ historical periodisation, and the theoretical presuppositions that appear to underlie it.

Early in the work, Castells outlines his basic theoretical framework, distinguishing it from post-industrial theory, and situating it against a broadstrokes account of the historical relationship between the state and technological development. I’ll quote Castells’ description of his project at length (p. 14-15):

This book studies the emergence of a new social structure, manifested in various forms, depending on the diversity of cultures and institutions throughout the planet. This new social structure is associated with the emergence of a new mode of development, informationalism, historically shaped by the restructuring of the capitalist mode of production towards the end of the twentieth century.

The theoretical perspective underlying this approach postulates that societies are organized around human processes structured by historically determined relationships of production, experience, and power. Production is the action of humankind on matter (nature) to appropriate it and transform it for its benefit by obtaining a product, consuming (unevenly) part of it, and accumulating surplus for investment, according to a variety of socially determined goals. Experience is the action of human subjects on themselves, determined by the interaction between their biological and cultural identities, and in relationship to their social and natural environment. It is constructed around the endless search for fulfillment of human needs and desires. Power is that relationship between human subjects which, on the basis of production and experience, imposes the will of some subjects upon others by the potential or actual use of violence, physical or symbolic. Institutions of society are built to enforce power relationships existing in each historical period, including the controls, limits, and social contracts achieved in power struggles.

So here we essentially have three very familiar categories – translations of the economy-civil society-state triad that seems to permeate the self-understanding of capitalist societies and, not coincidentally, appears as a core analytical concept in the works of many social theorists (Habermas’ three orientations for communicative action provide another example). After defining these concepts, Castells continues in a more explicitly Marxist vein (p. 15-16):

The product of the production process is socially used under two forms: consumption and surplus. Social structures interact with production processes by determining the rules for the appropriation, distribution, and uses of the surplus. These rules constitute modes of production, and these modes define social relationships of production, determining the existence of social classes that become constituted as classes through their historical practice. The structural principle under which surplus is appropriated and controlled characterizes a mode of production.

So here we have them: the forces and relations of production.

Castells uses these concepts to explain why he believes the emergence of new forms of information technologies *should* yield a qualitatively new historical period (which he analyses under the term “mode of development”), as well as to explain divergences between states and regions within historical periods. Castells thus tries to account for technological influence on society, and especially on culture, without invoking a reductionist technological determinism. His path is nevertheless a very familiar one within Marxist theory, relying on the concept that technology serves as a driving force within history, generating potentials that are then either hindered or facilitated by social relations.

There are a number of paths into a critique of these concepts – Castells is a nuanced and thorough theorist, and he raises and address many potential problems himself. In some ways, for me, it was watching the ways in which Castells strains against the limitations of the forces-and-relations-of-production framework – the ways in which he makes points that surpass what his analytical framework can easily grasp – that prompted me to write this post.

I have to admit that I always find the forces-and-relations-of-production rubric very frustrating – not that there aren’t interesting things that have been done, and could still potentially be done, with it. But it still misses what I believe were the most interesting questions Marx posed. I realise that it’s always problematic to attempt to divine Marx’s “true intent”, and I realise that the forces-and-relations-of-production approach is probably the dominant way to read Marx’s work. But it seems to me the forces-and-relations-of-production reading entails a really unfortunate flattening of Marx’s work – a flattening that is problematic, not because it misinterprets Marx, but because it sidesteps what seem to me to be fundamental questions for any contemporary critical theory.

Leaving Castells aside for the moment, the forces-and-relations-of-production account assumes that what motivated Marx’s theory was the desire to reveal the existence of class domination – that the primary critical goal of Marx’s work was to demonstrate that, in spite of its idealistic rhetoric of equality and freedom, bourgeois society was in fact characterised by its dark underbelly of class domination. In this reading, Marx’s intellectual achievement was to reveal the existence of class domination that bourgeois society denied – thus empowering the working class to grasp its true historical potential and achieve emancipation. This is the textbook reading of Marx (usually supplemented by the claim that Marx believed that bourgeois class domination would necessarily be overthrown by the workings of capitalism itself – with or without “agency”, depending on the theoretical tendency), and has made its way, with variations, into a wide range of Marxist-inspired theories. What interests me for present purposes is the vision of social critique inherent in this interpretation.

This vision of social critique – one that sees the primary purpose of social criticism as revealing that domination exists, and demonstrating how dominant discourses and ideologies can work to conceal awareness of domination – is actually quite common in various theoretical traditions, from the forms of Marxism discussed here, through to postmodernist approaches. I don’t contest that this sort of critical unveiling or deconstruction is an important task – I argue only that this is not the only possible way to approach social critique, and that it risks a systematic blindness to certain kinds of critical questions that can be important for a social critic to ask.

Moving back to forces and relations of production: while this reading of Marx has been incredibly productive for political movements and academic analysis, I think it misses a central point: I believe that what interested Marx was not revealing the *existence* of class domination (I suspect he thought the existence of class domination was rather self-evident), but instead explaining how, *given* the self-evident existence of class domination, bourgeois society could still find the *rhetoric* of liberty and equality *socially persuasive*. In other words, I suspect that Marx was primarily trying to explain how you could get these very strange, self-contradictory statements like the classic formulation by Rousseau: “Man is born free; and everywhere he is in chains!” If empirical observation so easily reveals the ubiquitous existence of the chains, then isn’t it peculiar that there is such a widespread belief in the existence of “natural” freedom?

It may be that I am wrong in this reading of Marx – certainly I wouldn’t be the first to read something into his text that isn’t there… But I will say that, if this *isn’t* the question Marx was trying to answer, it should have been 😉 More importantly, I believe that similar questions should be at the top of the agenda for critical theorists today. And, unfortunately, while theories grounded in analyses of the forces and relations of production may be useful for many purposes, I suspect very strongly that they cannot take us to the root of these kinds of questions…