Critical Realism and Social Science

Although Glynos and Howarth distance themselves from Roy Bhaskars (early) version of critical realism, their approach shares a great deal with it. In particular, the strong opposition between nature and human society, and their related qualified sympathy for a hermeneutic approach to social scientific research.

Partly as a result of this rather central challenge coming from within critical realism, there has occurred, since Ian and I wrote this book, a flood of new work in which critical realists engage with one another about a great range of issues, but also reach out to consider the relationships between this approach and others formerly seen as incompatible with it.

The issues that have received the greatest attention through the first decade of this century can be roughly organized into six topics:

1.What should critical realists make of Bhaskar's ‘spiritual turn'?

2. Can critical realism help with long-standing issues in methodology and social theory?

3. Can critical realism help span the division between the social and natural sciences?

4. What is the relationship between critical realism and Marxism (and other substantive approaches)?

5. What is ‘critical' about critical realism? Critical realism, morality and ‘emancipation'.

6. Does critical realism have anything to offer for feminist social science?

What Should Critical Realists Make of Roy Bhaskar’s ‘Spiritual Turn’?

It has to be admitted that Bhaskar’s writings from Dialectic onwards are a daunting prospect for the reader, but some realists have done us the service of taking on the task of subjecting them to serious critical scrutiny. One of these is Gary Potter (2007), who patiently takes apart Bhaskar’s new arguments for (and definitions of) ‘God’. Also, the authors of a collection of very thoughtful essays all engage critically with Bhaskar’s newer writings, in particular his arguments for a ‘meta-reality’ beyond the reach of science, and only accessible by the self-realizing reflection of human individuals (Dean et al. 2006). For some, Bhaskar’s later work threatens to subvert the valuable role of critical realism in its earlier role as ‘underlabourer’ for the social sciences, while for others the threat is deeper, abandoning the earlier notion of critical and emancipatory activity as collective struggle against specific forms of oppression in favour of a ‘New Age’ spiritualist concern with individual self-realization.

My own response has been one of dismay at the transformation of critical realism as a pluralistic collaborative effort to sustain explanatory and critical work in the social sciences into its opposite: a metaphysical-religious belief-system. The provisional and fallible, often mundane work of social scientific research might test the patience of one anxious for ‘emancipation’, and so the temptation to the certainties of religious faith can well be understood. Those who succumb to the temptation are, of course, entitled to their share of its comforts and consolations. My main point is that the shift to metaphysical system-building and theological speculation should not be confused with critical realism in its original incarnation as a critical philosophy of social science. The latter has a crucially important role still to play in illuminating issues in the social sciences, and should not be abandoned simply because one of its most talented and influential originators has done so.

Can Critical Realism Help With Long-Standing Issues in Methodology and Social Theory?

Some of the most fruitful developments ofcritical realism have come less from philosophers, and more from social theorists reflecting on their own substantive research experience. Examples include Sayer (2000) on the importance of integrating spatial relations in social theory, drawing on his research in urban sociology and political economy; Mellor (1998), drawing on her research also in local and urban social processes and (see also Hutchinson et al.

One of the most perplexing and long-standing questions in social theory - and one spanning both disciplines and rival meta-theoretic traditions - is the nature of social structures and the relation of ‘structure’ to ‘agency’ (Lopez and Scott 2001; Scott 2001). The essays in Dean et al. (2006) address this topic in rather different ways, and the book concludes with an interesting debate between the authors in which this becomes a focus. Several contributors accept the usual critical realist view that social structures are real, and should be conceptualized as distinct from the activities of human agents whose actions may be either enabled or constrained by them, but which also (intentionally or not) reproduce or transform them. However, Wight argues that this ‘dualism’ of structure and agency (Archer 1995) leaves an explanatory ‘gap’. His proposal is to draw on Bourdieu’s notion of ‘habitus’ as it was a means of representing the way socialization into routine performances embeds structural determinants into the non-reflective practice of individuals. However, Dean, also working within a broadly Marxian critical realism, argues against both Archer’s and Wight’s conception of ‘structures’ as external to agents.

Archer’s insistence on the duality of structure and agency was pitched against what she calls the ‘elision’ of structure into agency in the most influential alternative view, that of Giddens’s ‘duality’ of structure. For Giddens (as for Dean), social structure enters into the constitution of the agent and thence into the agent’s social practice, but is also the outcome of the agent’s practice. Structure is thus both the medium and outcome of social practice. Stones (2005) gives a qualified defence and development of structuration theory which places phenomenology and hermeneutics at the centre of the interpretation of social practices, so emphasizing the interdependence of structure and agency that is the strength of the notion of structuration. However, Stones’s advocacy of ‘strong’ structuration also acknowledges the significance of larger scale ‘external’ social structures. This debate continues! See, for example, Elder-Vass (2007a, b), King (2007), Porpora (2007), Varela (2007).

Can Critical Realism Help Span the Division between the Social and Natural Sciences?

My own essay (Benton 1981, reprinted in Archer et al. 1998) was an early attempt to show how this could be done by indicating the great variety of natural sciences (geology, physical geography, meteorology, physiology, evolutionary biology, biogeography, organic growth and development and so on), each with its own specific methodological approaches and problems, forms of explanation and modes of evaluation. The absolute ontological and methodological division between natural and social science is much harder to defend in the face of this diversity. Numerous writers have drawn in various ways on critical realism to deal with both philosophical and substantive questions concerning the relationship between society and nature.

Andrew Collier's Critical Realism (1994a: ch. 8) argued against nature/society dualism in critical realist social theory, and in his ‘Value, Rationality and the Environment' (1994b), he made use of the Marxian distinction between use value and exchange value to explore the relationship between the sorts of rationality involved in market transactions and the degradation of the environment. His Being and Worth (1999) argued for an objective environmental philosophy derived from the theology of Saint Augustine, and his subsequent In Defence of Objectivity (2003), similarly made a case for both a view of nature as real, independently of human experience or knowledge, and a related view of the objectivity of the value of (non-human) nature. In a tribute to Collier's philosophical work Kate Soper (2004) and I (2004) endorsed Collier's realism about nature, though arguing for the significance of the availability of cultural resources for human recognition of its meaning and value. The question of the objective reality of values will be taken up a little later in this chapter.

Since her brilliant What is Nature? (1995), Kate Soper has continued to develop and defend her version of humanist philosophy in the light of strong environmental concerns, as well as conducting both philosophical and empirical investigation of consumerist culture and alternative sources of pleasure and happiness. In contesting the common assumption that abandoning a consumerist lifestyle involves privation, she has introduced the influential and politically important concept of ‘alternative hedonism' (see for example, Soper 2000, 2004; Soper, Ryle and Thomas 2009). Since his path-breaking Ecology, Policy and Politics (1993), John O'Neill has continued to relate his philosophical work to questions of deliberative democracy and the possibility of reasoned judgements in fields such as environmental policy, where qualitatively different considerations have to be weighed against one another. This is contrasted with the way in which capitalist markets reduce such considerations to a single, quantitative dimension (see, for example, O'Neill 2007; O'Neill, Light and Holland 2008). Peter Dickens, too, since his major contributions in the 1990s has gone on to explore new developments in capitalist political economy (such as the so-called ‘knowledge economy') and their consequences for the subjective life of workers and consumers (for example, Dickens 2004). Together with James Ormrod, he has taken critical realist concern with the relation between humans and nature literally into outer space in the very original and timely Cosmic Society (2007). Mary Mellor has continued to develop her synthesis of green, feminist and socialist thought, with a focus on the significance of women's work both outside and within the market for both social and natural sustainability, as well as for the envisioning of radical alternatives (Hutchinson et al. 2002; Mellor 1998). Her most recent focus is on local communities and sustainability in an urban context, as well as continuing highly original work on the nature of money and finance (Mellor 2010).

In line with the critical realist concern with questions of ontology as prior to epistemology and methods, this exploration of the relation between the social and natural sciences and with environmental issues has highlighted the age-old topic of human nature. Against the tendency of most post-structuralist thinkers and many ‘social constructionists' to dismiss questions about ‘human nature' as mistakenly ‘essentialist, many critical realists have seen such questions as necessary to social science.

My own Natural Relations (1993) was in part a response to normative questions about our treatment of non-human animals, but it also included an attempt to develop a naturalistic view of human nature on the basis of a full acknowledgement of our evolutionary kinship with other species (an approach I called ‘human/animal continuism'). On the one hand, this insisted on the importance for social science of recognizing human embodiment, our vulnerability to disease and bodily disability, our dependence on continuous interchange with the rest of nature, our distinctive patterns of reproductive behaviour, extended juvenile dependency and so on. On the other, it resisted the reductive tendencies of some biological determinisms (especially ‘sociobiology', but also the more sophisticated ‘evolutionary psychology') that seek to explain human social and individual life by direct and unmediated application of generalities derived from a gene-centred version of neo-Darwinism. My argument was that the emergent properties of culture and language in hominid evolution did, indeed, distinguish us from other species, but that the nature and significance of this difference was best understood through a comparison with those other species. In other words, culture and language might be better seen as evolved ways in which humans meet needs that they share with other species, rather than as ‘sui generis’ attributes, to be understood exclusively in their own terms. Of course, this did not rule out the emergence of radically new needs - as well as vulnerability to new kinds of harms and suffering - as a consequence of the emergence of such attributes (for example, what Maslow (1970) called ‘self-actualization' needs - that could not arise for a being with no sustained concept of self, and incapable of imagining unactualized possibilities for its further development).

Peter Dickens and Andrew Sayer, among others, have broadly shared this approach, but Kate Soper has argued for a much stronger sense of human uniqueness, partly in order to sustain an environmental politics that would depend on humans uniquely reflecting on and deliberately transforming their relationships to the rest of nature. Andrew Sayer, too, is more inclined to emphasize human distinctiveness than I have tended to do. In part, these differences of view about human nature may be related to differences in the way non-human animals are represented. Kate Soper, for example, refers to animal needs as ‘biological requirements’, ‘instincts' and the like, whereas my comparisons are with species that share with us relatively complex and flexible psychological capacities and learning abilities, various forms of sociability, parental care and so on (see’ for more detail on these exchanges’ Dickens’ Soper’ Sayer and Benton in Moog and Stones 2009).

What Is the Relationship between Critical Realism and Marxism (and Other Substantive Approaches)?

The development of critical realism as a distinct approach in the philosophy of social science took place during the 1970s, a time when Marxian perspectives of various kinds were very influential in sociology and related social science disciplines. However, the early work of Keat and Urry (1975), Roy Bhaskar (1979) and myself (1977, 2015) argued for realism in the social sciences independently of our varying commitments to Marxism. My own book was an attempt to show that the actual explanatory strategies used by Weber and Durkheim in their classic works were realist in character, even though the explicit accounts they gave of their methodological principles were often quite different. However, it soon became clear that Marxist historical materialism fitted the critical realist ‘model’ of explanation particularly well, and it is probably true that Marxists have made the most use of critical realism as a means of guiding and evaluating their research practice.

More recently the increased interest in critical realism, combined with declining commitment to Marxism, has resulted in a wide range of other theoretical approaches being subjected to the critical realism ‘treatment’. Rob Stones (2005) draws on critical realism in his critical re-working of structuration theory, Frank Pearce and Tony Woodiwiss (2001) and Jon Frauley (2007) read Foucault as a realist, while Frank Pearce has also made a strong case for Durkheim’s realism (2001, 2007) as has Ray Murphy with respect to Weber (Murphy 2002a). Chodos et al. (2007) argue for a minimally realist reading of Gadamer, but also use Gadamer in a partial critique of critical realism itself on the concept of truth.

Despite this variety in critical realist readings of non-Marxian theorists, there is a strong tendency for critical realism to oppose post-structuralist and post-modernist trends in philosophy and social theory in view of their supposed relativism, resistance to ontology and consequent lack of critical orientation. The edited collection by Lopez and Potter (2001) carried the provocative title After Post-modernism and was intended as an intervention against the posts’ anti-enlightenment stance. However, Stones (1996) while distancing himself from the more extreme versions of the ‘posts’, offered a more sympathetic account of their critique of ‘modernist’ theory from a critical realist standpoint. Sayer, too, argued for critical realist recognition of the force of some post-modernist themes, but also provided important critiques of what he terms ‘defeatist postmodernism’, including a very telling defence of some uses of forms of reasoning rejected by post-modernists/post-structuralists as ‘essentialism’ (Sayer 2000: ch. 4). Day (2007), too, argues for a more nuanced and sympathetic critical encounter between realism and post-structuralism (as evidenced in Pearce, Woodiwiss and Frauleys treatment of Foucault, mentioned above). It is also worth noting that the version of post-Marxism presented by Glynos and Howarth has much in common with critical realism.

While welcoming this flourishing of non-Marxian uses of critical realist philosophy, it seems to me there remain strong reasons for holding on to the ‘affinity’ between critical realism and (some versions of) Marxism. The fruitfulness of the link is either argued for or exemplified in many studies, including the chapters by Albritton, Engelskirchen and Ehrbar in Frauley and Pearce (2007), the substantive social scientific work of critical realists including Bob Jessop, Peter Dickens, Mary Mellor and many more. Jonathan Joseph (2001) used critical realism to great effect in evaluating a range of different versions of Marxism and critical theory, and, together with Kathryn Dean, John Roberts and Colin Wight, co-authored an exemplary exploration of theoretical issues arising from the linking of critical realism with Marxism (Dean et al. 2006). Sceptical about the value of Bhaskar’s later work, they continue to find critical realism in its ‘underlabouring’ mode valuable as a means of correcting and resolving conceptual problems in Marxian historical research.

I have elsewhere argued for the continuing importance of the linkage between critical realism and Marxism for three main reasons. First, the approach to concept­construction and historical explanation pioneered by Marx and Engels was not only broadly consistent with the precepts of critical realism, but their own philosophical reflections on their approach had much in common with it. Although elements of realist explanation are certainly present in the work of other classic theorists, their development was limited by inappropriate philosophical considerations (Benton 1977, 2015). Second, much of the attention of Western Marxism in the twentieth century has been devoted to developing Marxian cultural and political theory, so that the old critique of Marxism as a form of ‘economic reductionism’ has no serious purchase - if it ever had. In view of this, it is now ‘safe’ to return to the analysis of the dynamics of capitalist political economy. Capitalism as a socio-economic form of life continues to have overwhelming causal importance in shaping the geographical distribution of economic activity, the life-chances of whole categories of people, the available policy­options for dealing with pressing economic, social and ecological problems and so on. In the wake of the neo-liberal ascendancy and capitalist globalization, this is even more inescapably so. It remains the case that the Marxian legacy offers the most fully developed and theoretically sophisticated critical account of capitalism as a whole system and its dynamics. Third, the commitment to ‘naturalism’ that has been a consistent theme in critical realism provides indispensable methodological support for research that investigates in a non-reductionist way the interrelations between human social, economic and personal life and the naturally given conditions and processes that sustain life, health and human flourishing. Arguably, it is current work (some of it mentioned above) enabled by both critical realism and contemporary Marxism that has taken us further with this aspiration than any other approach. The emergence of an emancipatory politics with any hope of addressing the growing crisis in the relationship between contemporary capitalism and our ecological life-support systems may depend on the kinds of understanding such research can achieve.

But this takes us on to a further set of questions about critical realism - and also, of course, to wider questions about the purposes and significance of social science itself.

What Is ‘Critical’ about Critical Realism? Critical Realism, Morality and ‘Emancipation’

As Ian and I noted in the first edition, one of the reasons critical realism has proved attractive has to do with the adjective ‘critical’. This implies that under the guidance of critical realism, social research can provide justifications for normative (ethical, moral, political) judgements about the states of affairs that are studied, and, through that, justifications for political actions to preserve or change those states of affairs. The concept that has been most used to link the critical realist account of social science explanation with morals and politics is that of ‘explanatory critique’. In the work of Bhaskar and Collier, especially, this concept applies to the way some social structures can give rise to false beliefs, and also to the way some of them may give rise to avoidable suffering or deprivation. If social structures can be shown to have these effects, then we can pronounce a negative judgement on them, and justify their removal/transformation. The impression is given that this overcomes the traditional philosopher’s claim that moral judgements cannot be deduced from mere statements of fact (to do so is held to fall foul of the ‘naturalistic fallacy’). But for the proponents of the notion of explanatory critique, the sense in which a normative judgement follows from a factual statement is not captured by the notion formal deduction. Rather, the point is that to say someone is starving but should not be fed just does not make sense. Of course, it is admitted that there could be exceptional circumstances in which it would make sense not to feed someone who is starving (there may be extremely limited food, and many people starving, so hard decisions about priorities would have to be made) - so the connection between starving and feeding has to be qualified by a ‘ceteris pαribus^, clause - that is, ‘other things being equal’, we should feed the starving.

One obvious objection to this sort of argument is that descriptions of people’s unmet needs, suffering, delusions and so on are themselves implicitly evaluative, so the explanatory critique only derives value judgements from ‘factual’ claims that already embody the values that we draw from them. The issue of value-commitment is only pushed a step further back into the question of the concepts we use to describe social structures, states of affairs and so on. There are two main ways this retort can be addressed. The first is to pursue the ‘holy grail’ of a value-free description of social life, as proclaimed by positivism, while the second is to acknowledge the ineradicably ‘valuey’ nature of all meaningful characterizations of social life. The first option would, if taken to its limits, deliver descriptions that told us little or nothing intelligible about its topic. This is partly, but not wholly, because the self-understanding of social actors is ineradicably evaluative and as such (partially) constitutive of social life. So, the second option seems to be what is left: acknowledge that any worthwhile characterization of social practices, structures and so on will necessarily carry with it value-connotations.

Now, this is a problem for the concept of explanatory critique if it is held to of itself justify a value judgement, or, still further, to provide objective grounds for political action to transform the social structure or practice that is the alleged cause of suffering or delusion. The move to claiming the ‘reality’ or ‘objectivity’ of values, as Bhaskar and Collier, especially, are inclined to do, cannot be sustained by appeal to the model of explanatory critique (whether it can be underwritten by theology is a separate question). Of course, there is a sense in which values are ‘real’ and ‘objective’. They are really present as part of the substance of social life itself - they are the means by which people and societies make sense of what they do, form their identities, define their hopes and aspirations, make decisions in difficult times and so on. This is part of the substance of Andrew Sayer’s eloquent critique of the way much sociology fails to acknowledge the rationality in the ‘lay normativity’ of everyday life (Sayer 2009). However, to recognize that values are present in the social practices we study is also to recognize the immense diversity of value orientations that empirical social science discloses: we seem to be back with the value-relativism and abandonment of independent critical judgements about social life that the hermeneutic tradition has often been thought to imply!

But there are other problems with the idea of explanatory critique (see, especially, Sayer (2000) and Benton (see this work, pp. 134-7; Benton 2004). It might be, for example, that a causal mechanism that produces suffering or injustice also has other effects that favour human well-being. Even if we accept that the suffering it produces is bad, there would still be an issue about whether it should be abolished in the light of what we know about its countervailing good effects. Tourism might be an example here - we might think its negative effects on environmental sustainability would justify banning or reducing it, but this would have to be weighed against the positive values of contact with other places and cultures, and of individual liberty. This could be addressed by the ‘other things being equal’ qualification, but the difficulty with this is that other things never are equal. In the complex situations in which difficult choices have to be made, any action is likely to have consequences along more than one axis, and be open to a variety of different, often incommensurable, considerations. How do we weigh the loss of liberty involved in banning a popular activity, or the injustice involved in using market incentives to restrict enjoyment of it to those who can pay, against the benefits of reduced carbon emissions, increased cross-cultural understanding, increased affluence of sections of local communities, or social and ecological degradation and increased social inequality of tourist destinations? It is hard to see that the notion of an explanatory critique would get us very far with untangling these issues.

Sayer (1995, 2000) points out that ‘defeatist’ post-modernism gives up on the critical project in the face of this, but he argues persuasively for an alternative: give up on the unreflective adoption of a critical standpoint in favour of explicit engagement in normative argument. If the critical social theorist or researcher is at odds with a prevailing social practice and the ideas that sustain it, then she is required to be clear about her initial value position, and to engage in a critical dialogue with the disputed values. This would entail a much more developed normative discourse than critical realism has so far provided. Here, it seems to me, we have to find a way of thinking about normative discourse - concerning value commitments, the grounds for moral judgements, conceptions of the ‘good life' and so on - that avoids two opposite extremes. The moral relativism of ‘anything goes' is not only itself morally unacceptable as it dismisses and trivializes a central feature of human life but is also self-refuting, as it is itself a moral claim. The contrary position, that there are objective moral truths, is both philosophically indefensible (in secular terms) and potentially oppressive or, at least, liable to intolerance.

Between these opposite poles, it can be recognized that we can and do have reasoned arguments about moral questions - about the justification or otherwise of abortion, assisted suicide, same-sex relationships or the death penalty as well as broader questions to do with the moral ordering of our social relations and relations to non-human nature. People care, often passionately, about these issues, and in practice (if not in some versions of academic social theory) believe themselves to be right and their opponents wrong (even ‘evil'). Though actual public debate frequently falls short of mutually respectful reasoned argument, this is nevertheless possible (and desirable!). Actual moral disputes appeal to a variety of sources of evidence - including empirical evidence of a social scientific kind - as well as pragmatic considerations, moral principles, rival interpretations or applications of moral principles, more or less consistent or grounded general views of human nature, claims of inconsistency or hypocrisy, reference to widespread human emotions and moral sentiments and so on. Even if we suppose moral dispute conducted under the conditions specified by Habermas's notion of an ideal speech situation (Chapter 7 and also Chapter 8, pp. 134-7) and continued without limit of time, it is difficult to believe that a society-wide moral consensus could be reached. Irreducible, bedrock moral pluralism is, perhaps, the wisest assumption to make if we are to think seriously about the conditions under which, in a complex and cosmopolitan society, we are to order our relations with one another.

Nevertheless, there remains an important place for reasoned normative discourse, even if we do not assume that consensus will be the end result. Sometimes, indeed, consensus will be reached; sometimes rival parties will gain in respect for one another's views and so find some form of compromise, or mode of coexistence; sometimes a ‘second order' consensus might be reached about a framework within which continuing conflict can be contained, or put to constructive purpose.

The previous two paragraphs take us a step beyond ‘critical' social theory and its justification. Many critical realists speak of some versions of social theory as ‘emancipatory', and this is one important link between critical realism and Marxism. However, other orientations to social theory - linked to social movements other than the labour movement, such as feminism, anti-racism and the green movement - also see a link between critical social theory and a wider project of emancipation. As we saw above, the critical realist notion of societies as ‘open systems' implies that social life includes numerous interacting structures and practices, such that changing one structural cause of a specific form of injustice or suffering will have consequences

that ripple out beyond the point at which intervention is made - and might well undermine whatever good is done, or generate further problems. For example, setting financial targets for health service trusts to ensure that beneficial new investment is efficiently used may have the unintended effect of distorting medical priorities and harm patient care.

These are well-known problems facing particular policy initiatives conceived and given effect within a broad ‘taken-for-granted’ framework of power relations, institutional structures and so on. However, the notion of emancipation is generally taken to involve a more profound and extensive transformation of social relations. It is usually taken to be implicit in the overthrow of a regime of domination and subordination that is in some sense at the core of the functioning of a social order. Classically, of course, it was applied to the overthrow of slave systems, but in its Marxian version, it was applied to the relation between capital and wage-workers. In feminist thought the fundamental oppressive relation is the power wielded by men over the lives of women and in the case of anti-imperialist struggles it is the oppressive and exploitative rule of an imperial power over the people of a subordinated nation or culture. Because the relation of power that is challenged by an emancipatory struggle is viewed as central to the functioning of the established order, emancipatory movements carry, implicitly or explicitly, a vision of a qualitatively different whole mode of social life.

However, Sayer (1995, 2000) takes to task much critical social science, especially as reconstructed by critical realism, for performing only part of the necessary critical work. To criticize a particular structure, or, even more seriously, a whole pattern of social power relations and associated institutions as a source of suffering, injustice or unmet needs carries weight only if there is an alternative way of ordering society that is both feasible and desirable. That is, a mode of life that can be brought into being, and that can reasonably be expected to ameliorate the negative aspects of the status quo without producing ills that are even worse than the ones criticized. In short, our system might be bad, but it might be less bad than any of the alternatives! Sayer is critical of the notion of emancipation often relied on by critical realism as the replacement of unwanted determinations by wanted and needed ones. Although this does acknowledge that emancipation cannot mean a leap into some absolute, unconditioned ‘freedom’, it does not adequately address the diversity of needs, aspirations, values and interests that would have to be taken account of and realigned in any fundamental transformation of society. Nor does it seem to recognize the role of the obligations and commitments that flow from human social relations, nor the constraints that follow from a qualitative re-ordering of human and practical relationships to the rest of nature. This is, of course, not a critique of the vision of radical emancipatory change as such. Rather, it is a call for a more complex and nuanced understanding of what would be involved.

Dean et al. (2006) take issue especially with aspects of Bhaskar’s later work on Dialectic and the ‘spiritual turn’ (Bhaskar 1993, 2002). In their view, emancipation has to be understood as a creative and collective project of social transformation, whereas, for them, Bhaskar’s later work reduces the notion to one of individual ‘self-emancipation’ through contemplative activity. Joseph (2006), especially, comments that emancipatory practice should be understood in terms of the possibilities given by concrete, historically given circumstances, rather than presented in abstract, universalizing concepts. Kathryn Dean, however, recognizes that Bhaskar, as well as Marx, do envision an alternative mode of social life in which distinctively human potentials might be realized. There is, then, a response to the challenge posed by Sayer - an attempt to specify the outlines of a possible form of human society in which the frustrations and ‘absences' of our current capitalist way of life would be transcended. Dean argues that humans are instinctually underdetermined, so that it is only through being nurtured in a specific culture that their potentials can be actualized. She draws on Aristotle's conception of ‘eudaimonia' - a state of social being in which virtue and practical wisdom (‘phronesis') are nurtured by the practices of citizenship. This implies a public sphere in which citizens exercise judgement in their deliberation on the common good. Divested of Aristotle's unjustified exclusions from citizenship of women, slaves and others, this account of the ‘good life' provides a contrast with the individualizing and depoliticizing character of the processes through which individual selves are formed under capitalism.

The growing recognition by critical social scientists that criticism is seriously weakened if we can say little or nothing about feasible alternatives has led to a resurgence of interest in utopian thought - whether in the shape of literary fiction, film or philosophical-political speculation. Interestingly, most of the commercially popular representations of the future in literature or film are decidedly dystopian - to the point that it has been said it is easier to imagine the end of the world than the end of capitalism. However, social theorists and philosophers such as Ruth Levitas, Kate Soper, Martin Ryle, Andre Gorz, John O'Neill, Pat Devine, myself and others have persisted in facing up to the challenge of imagining future societies which might enable human potential to be realized on the basis of social justice, conviviality and respect for the non-human world. In this, they can draw on a long history of utopian thought from at least as far back as Aristotle, and including the diggers and levellers of the English Civil War, the artist and mystic William Blake, designer and revolutionary William Morris and many more. All of this heritage of thought about alternative modes of social being does, of course, bear the mark of its time and place, and it is a useful reminder of how difficult it is even for the most imaginative thinker to escape from the prejudices and assumptions of her or his own historical moment. Nevertheless, theoretical imagining, drawing on these past works, as well as study of attempts in practice to ‘prefigure' possible futures in social experiments such as cooperative enterprises, communes, local exchange and trading schemes and so on is necessary if the horizons of our own political culture are to be opened to possible alternatives - alternatives so far systematically excluded from our public discourse. While by no means all of this imaginative work is carried out by critical realists, it is arguable that critical realism has much to offer, both in sustaining the rationality of such work, and in providing criteria by which to assess its outcomes.

Does Critical Realism Have Anything to Offer for Feminist Social Sciences?

Feminists influenced by critical realism include Kate Soper and Mary Mellor, whose work has already been mentioned. Caroline New (1998, 2003) has argued for the value of a version of critical realism in defence of feminist research in the face of some versions of post-structuralism that criticized feminist research and political mobilization based on an assumed common interest among women.

Critical realist economist Tony Lawson, too, sparked a lively controversy in the journal Feminist Economics (1999, 2003b) by arguing that a critical realist turn to social ontology among feminist theorists would have much to offer for the emancipatory potential of that discipline. The ontology he outlines is one of human society as open, structured (in the sense that structures are not reducible to the activities of human individuals), dynamic, or ‘processual’ and predominantly comprising internal relations. The view of emancipation he offers acknowledges the importance of ‘differences’ (of gender as well as others) but situates these in the context of a deeper sense in which there are commonalities shared by all humans in the necessary conditions for our flourishing. These include our need to develop species-specific competences such as language. On this account, emancipation would amount to the emergence of a society in which ‘the flourishing of each is a condition of the flourishing of all and vice versa’ (Lawson 2003a: 125). In the course of this argument, he also notes a convergence of critical realism as he interprets it with feminist standpoint epistemology (see Chapter 9). If we, following critical realism, think of society as an open system, it follows there is no single privileged point of access for understanding - the interests or ‘standpoint’ of the investigator will necessarily shape the specific phenomena and associated causal mechanisms that they select as research topics.

Sandra Harding (1999), Drucilla Barker (2003) and Fabienne Peter (2003) in their responses raise several issues. Barker agrees that critical realism has important methodological insights into the failings of ‘mainstream’ economics and notes the convergence between Lawson’s version of critical realism and standpoint epistemology. However, on the key claim that critical realism can provide a general account of human nature (a ‘philosophical anthropology’) that would overcome the divisions of perspective that flow from human difference, she is sceptical. For post-structuralist and post-colonial thinkers (she particularly cites Donna Haraway), knowledge is always ‘situated’, so that claims to the effect that there is some underlying or potential unity that transcends the differences that we experience should always be treated with scepticism - Whose interest does this ‘universalizing’ claim express? Who is to be emancipated by the proposed course of action? And so on. Both Harding and Peter call into question critical realism’s account of science. Since Bhaskar’s early transcendental arguments for the social ontology of critical realism rested on the assumption of successful scientific explanation, they argue that the feminist critique of science as a social practice driven by power relations, most notably those whereby women’s contributions are excluded or marginalized, is something that critical realism cannot take into account. Like Barker, Peter argues against critical realism's moral realism, and its claims to know what universal needs humans share.

This is clearly an ongoing debate, with a number of unresolved issues, but a few comments can be made here. First, the claim that critical realism assumes the success of scientific explanation and so cannot effectively criticize its unjust institutionalization has some force, especially against Bhaskar's early use of transcendental arguments in relation to physical science. However, it is also true that Bhaskar's view of science theorized it as a social practice in a way that was absent in earlier ‘mainstream' philosophies of science. Unfortunately, he did not take that insight far into a critical enquiry into the history and sociology of scientific practices and institutions of the sort that would meet the requirement of feminist critics. But Lawson's version of critical realism explicitly draws on a range of familiar experiences, and is not dependent in the way Bhaskar's was on the argument from scientific success.

Lawson's proposal that a ‘philosophical anthropology' linked to critical realism could indicate human commonalities that might transcend the difficulties arising from systematic differences and divisions of interest poses many important but challenging issues. We saw above that serious critical social research - especially where it is held to sustain an ‘emancipatory' project - stands in need of a defensible account of an alternative to the status quo that would be both feasible and desirable. Since one of the standard conservative responses to utopian visions is that they ‘do not take account of human nature', it seems likely that any plausible vision of a future society radically different from today's would have to be backed up by an account of human nature such that our irreducible differences, psychological dispositions, emotional repertoires, learning abilities, primary attachments, organic needs, developmental adaptabilities and so on would find a field for their exercise and satisfaction. So, in this sense, a ‘philosophical anthropology' almost certainly would be a necessary element in any fully developed argument for an emancipatory project. However, Peter is surely right to resist the claim for objective truth on behalf of any one version of such a philosophical anthropology. For any such claim to remain uncontested would certainly risk oppressive applications, so all proposed alternative visions should accommodate the ‘potential contestedness of needs' (Peter 2003: 99).

This does not mean, however, that the effort to use our imagination, evidence (from a variety of disciplines) and reasoning in constructing accounts of human nature is worthless, or, indeed, that any one of them is as good or bad as any other. The hope is that open and inclusive discussion might take us towards better versions, ones in which the perspectives of hitherto powerless or marginalized life experiences were represented. This might also play a part in legitimating such accounts, and the alternative social visions they carry, among sufficiently diverse and inclusive coalitions for a broadly based emancipatory struggle. But there are two further provisos. While a shared broad view of a common human flourishing might be necessary to inspire the politics of a movement for change, there is no requirement for such a movement to overcome or down-play internal differences or divisions. Lawson's critics take him to be arguing that mobilizations necessarily depend on a notion of a shared, common interest. In fact, many of the most successful social movements have involved very broad coalitions in which significant participation has come from activists who do not stand to benefit directly from the changes they seek: the world-wide struggle against apartheid, coalitions in favour of the extension of the franchise to women and other excluded groups, solidarity campaigns on behalf of political prisoners and many other examples could be cited.

My second proviso raises questions about the ‘humanist’ assumptions of the main advocates of the emancipatory objectives of critical social science. Two examples illustrate the point. One of these is the animal rights movement, a leading pioneer of which saw it as the logical next step beyond the emancipation of women, ethnic minorities and subjected nationalities (Singer 1975). The other is the international campaign in defence of the tropical moist forests. True, these are often presented in terms of the ecosystem services and potential medical resources that lie in the biodiversity of the forests. However, the deepest sources of anxiety and anger stem from a moral and aesthetic love of these unique and irreplaceable natural formations - ones that most activists have never visited - as well as solidarity with their indigenous inhabitants. My suggestion, here, is that the unsustainability of our current global social and economic system has provoked a requirement that our envisioning of alternatives has to put at its centre a qualitatively different social, economic and moral mode of human life in relation to those other species with which we (still) share the planet. Mere ‘humanism’ is no longer enough.

Ted Benton, May 2010.