Values and Strategies of Argumentation in Everyday Life, Politics and Social Science

Values and Strategies of Argumentation in Everyday Life, Politics and Social Science

Andrew Sayer, Department of Sociology, Lancaster [email protected]

April 2008

AbstractThe paper discusses some difficulties in intervening in popular political debate. These have to do with strategies of argumentation, types of explanation of social phenomena, and values. It compares social scientific and popular political forms of reasoning, commenting on ‘the belief in a just world’, differences between social and individualistic explanations, and between explanation and justification. It assesses GeorgeLakoff’s claim that political debates are won on the basis of appeals to voters’ values, not policies, and argues that values should be treated as within the scope of reason, and as related to understandings of human needs and well-being.

Introduction

Popular political discourse is a disturbing spectacle foracademics, particularly social scientists, for its explanations of social phenomena and its diagnoses of social problems are generally simplistic, and often incoherent, prejudiced, or simply false. Worse, argumentsare often ‘eristic’, that is, conducted purely in order to gain victory over ‘opponents’ and persuade others by any means that works, regardless of whether the argumentsare logical or illogical, empirically supported or not, or involve ad hominem/feminam points. Eristic discourse is driven by self-interest rather than an interest in finding the strongest argument, regardless of whether it favours our self-interest. Defeat is signalled by popularverdict, which may be influenced simply by any sign of

1

weakness (e.g. ‘looking ruffled’) or uncertainty on either side. Indeed, in the worst cases, it may be flattering popular political discourse to call it argument at all, for much of it is at the level of image,style, ‘conviction’ and ‘put-downs’. It might be called the ‘boxing model’ of political debate, in that the primeaim is to land the knock-out punch. Even when the arguments are put sincerely, because their advocates genuinely believe them to be right rather than merely a way of scoring points and winning, political arguments have to be stripped down to simple points, ignoring qualifications, if they are to gain any purchase, for patience is in short supply. Those who have complex arguments are at a disadvantage, regardless of their quality, and of course political issues are often highly complex. On their home ground, academics are accustomed to having generally respectful, captive audiences, and ample time or thousands of words to play with. Not surprisingly, they are likely to find the prospect of intervening in popular political discourse daunting. In addition, political arguments are often conducted betweendifferent value positions, and as such may seem difficultor even irresolvable. It is common for both lay people and academics to regard values as ‘merely subjective’ andbeyond the scope of rational deliberation, so that the discovery of value differences brings argument to an end.As Hilary Putnam notes, ‘that’s a value-judgment’ functions not only as a conversation stopper, but as a thought stopper (Putnam, 2002). Some, like George Lakoff,argue that we should realise that political arguments aredecided on values, not on policies, and put values first in framing political debate (Lakoff, 2004), in which case, careful assessments of political problems and policies miss the point.

In this article, I want to explore some of the sources ofdifficulties of political argument, first by examining some typical problems that emerge in debates between academics and lay people or politicians. Through examplesI shall try to show that the difficulties for academics in making interventions into popular debates derive not only from the fact that the former tend to be more

2

complex and qualified and hence unappealing in contrast to the simplicities of populism, but from differences in explanations of behaviour and its outcomes and the way they are evaluated. A key difference lies in the attribution of responsibility. Lay or popular explanations of behaviour tend to be strongly individualistic and voluntaristic, attributing all or most of the responsibility for behaviour and its outcomesto individuals or groups, whereas, academic, especially Left academic, explanations of the determinants of behaviour and its outcomes tend to have a strong social component, according to which, what people are like and what they do is profoundly shaped by their social circumstances.1 Social determinants are themselves complex, and individuals do indeed also have some responsibility for their character and for many of their actions, and indeed for their outcomes, so any explanation which tries to do justice to this constellation of elements is bound to be complex, at least compared to explanations which simply credit or blame individuals or groups. These are simultaneously empirical matters and moral and political matters, for they involve attributing responsibility, and hence creditor blame, for good and bad circumstances. I argue that voluntaristic and individualistic explanations support and are supported by ‘the belief in a just world’, that is the view that, broadly speaking, people get what they deserve and deserve what they get. Again, this belief hasthe advantage of simplicity, both as an empirical view ofthe world and a moral-political stance. I then turn to the question of values in political debate, as posed by George Lakoff, arguing first that his approach in his popular book, Don’t Think of an Elephant (2004), implies a misleading fact:value dualism, and obscures our understanding of such debate by presenting value-judgements as lacking a rational – or reasonable - basis.(As we shall, however, Lakoff has developed a much more sophisticated analysis of values in his more academic 1 As Norman Fairclough has pointed out (personal communication), popular political arguments do often attribute credit or blame to abstract entities like ‘globalisation’, though perhaps less so where individual behaviour is concerned. As always, commonsense is a complex and often inconsistent assemblage of ways of thinking.

3

work.) Then, drawing upon the work of philosophers such as Charles Taylor, Hilary Putnam and Andrew Collier, I argue that political arguments implicitly or explicitly appeal to beliefs about flourishing and suffering and about who or what is responsible for them, and insofar asthey do so, they involve a kind of reasoning about values. Although in practice, the kind of reasoning is usually crude, I argue that it implies a more hopeful view of the possibilities of political discourse in whichdiagnosis and value-judgement are fused and amenable to rational discussion.

Strategies of argumentation: eristics and complexity versus simplicity

These issues are very much about strategies of argumentation. ‘Strategy’ of course implies instrumentalism, and hence could be taken to imply an eristic outlook. In an ideal kind of discussion, or in anideal speech situation, as Habermas termed it, the only force or motive would be that of finding the better argument or account, and participants would be willing tochange their minds if others’ ideas proved to be superior. Academic discourse aspires to this ‘disinterested’ form of argument, though of course it often falls short of the ideal, while eristic argument promotes and is supported by ‘systematically-distorted communication’ and is interest-driven (Habermas, 1979). In popular discourse, this distinction is often referred to as one between discussion and argument, the former more open and exploratory (needing to be had), the lattercombative (needing to be won). All argument or debate, including that of academic discourse, involves strategy, but it need not be eristic and involve subordinating the means to the ends, or deceit; it may simply involve a desire to be clear and not to be misunderstood, and to dojustice to the arguments and claims one sincerely thinks are strongest. To be sure, we are also likely to enjoy persuading and perhaps ‘defeating’ others, but presumablywe want to deserve the status that comes with ‘success’ in debate through having worked out the best position,

4

and to prefer that to having won that status through someform of deceit or mere appeal to authority, or similar fallacious kind of reasoning. We may also sometimes defend minority positions if we believe them to be right,even though it brings us disapproval.

As Martin Reisigl and Ruth Wodak note, the English word ‘persuasion’ covers two meanings for which some other languages, like German, have separate terms. Überzeugen refers to the practice of convincing by rational argumentand aligns with Habermas’s ideal speech situation, while überreden refers to manipulative forms of persuasion that involve suspended rationality, such as flattery, deceit, demagogy or threat (Reisigl and Wodak, 2003).2 That decisions on what is true or valid are themselves fallible makes no difference to these distinctions, for even accepted truth claims that are later refuted may have been arrived at sincerely and for their own sake rather than merely because they were thought likely to bring their sponsors advantyages. (Those who want to argue that academic, indeed all, debate is always and only eristic need to ask themselves what that implies fortheir own views on this very matter: are they only putting such a view because they think it will bring themprestige, or because they sincerely believe it? If the former, we need not take them seriously; if the latter, they are guilty of a performative contradiction.3)

Good argument in the non-eristic sense is supposed to be oriented to finding the best, or most true, or most practically-adequate understandings of its subjects.4 2 See also the Aristotelian distinction between rhetoric and dialectic (Van Meeren and Grootendorst, 2004). I do not use the argument:rhetoric distinction, because like O’Neill, I regard them ascapable of being compatible, in the sense that rational argument can hardly avoid rhetoric, but that this does not necessarily compromise its rationality (O’Neill, 1998).3 This is a variant of the liar paradox: the statement ‘I am lying’ is true only if it is false, and false only if it is true.4 Again, those who refuse any idea of truth or possibility of empirical support or refutation for propositions invite performative contradictions as soon as they argue for such a position (there’s no such thing as the truth, and that’s the truth!), and they invite theory-practice contradictions as soon as they have to engage in

5

Academics are supposed to develop and apply skills of careful reasoning, observing a number of rules for rational dispute and constructive argument, and avoiding various fallacies, such as treating the contingent as necessary, generalising from a sample of one, the fallacyof composition, the ecological fallacy, and other non-sequiturs (see van Eemeren and Grootendorst, 1992, and Reisigl and Wodak, 2003). Academics become used to making‘tight turns’ in arguments, that is, taking a premise or argument that is commonly assumed to point in a certain direction, and showing that actually it does not do so orneed not do so, and can or should allow a move in a quitedifferent direction. This is most characteristic of philosophy, where relations of logical entailment have priority over other forms of association, but it is common in all academic argument. By contrast, in political argument, even where it is not eristic, references to controversial objects typically involve dubious topoi and carry a heavy baggage of associations, often pejorative, which make such tight turns difficult to make in a way that is likely to convince. This is mostobvious with common stereotypes such as ‘asylum seekers’ (in the UK) or ‘people on welfare’ (in the US). If the negative associations are well-established, they can be evoked without using explicitly-pejorative language. Those who use such terms disingenuously or naively can then say in their own defence that they haven’t disparaged anyone because they haven’t used pejorative language.

So how do those accustomed to careful, patient discourse,making tight turns and fine distinctions, and refusing such baggage (or attempting to do so), intervene in such debates without being misunderstood? When confronted withan eristic interlocutor, for whom strategy is merely the most effective route to victory, it is hard not to respond in kind, indeed it may be naïve not to. Academicsmay, for example, have important information about topicslike immigration, which is not known or ignored in the popular debates. However, one risk is that just in

practices in which what is the case is important – as in deciding whether (it is true that) a car is coming when crossing the road.

6

mentioning the subject they will attract more attention to the popular ways of understanding it and invite othersto twist their claims. (Even those who regard social scientific findings as ‘value-free’ cannot avoid some moral responsibility for how they use them, for the claimed value-freedom applies only to the information, not to the act of communicating it.) Particularly if we feel that a controversy is about matters which do not deserve to be treated as key issues in an election or other occasion for political debate, we may feel that it would be better ‘not to go there’, and to draw attention to other issues and frame our own agenda. However, this may have the unfortunate effect of leaving the fields in which we could intervene uncontested. And if the controversies in question have taken hold, the public mayrefuse our alternative agenda. Thus, for example, even those politicians who want to set fresh agendas free of racist associations may find that voters want to talk about asylum seekers or immigration, and on the basis of entrenched racist stereotypes. How can one argue against such beliefs simply and effectively, without compromisingone’s position and being misunderstood?

Explanation, responsibility and the belief in a just world The complexity of political issues and the consequent complexity of adequate analyses of them is not the only problem of engaging with populist political discourse. The difficulties also derive from certain pervasive ways of understanding the social world in commonsense thinking, centring on the issue of responsibility for people’s behaviour and circumstances. This is clearly important, for it affects how people assign credit or blame, and it figures strongly in political thought. Commonsense thinking is of course enormously rich and diverse, including many quite different and indeed often opposed ways of thinking, so I acknowledge that the tendencies I highlight below are not the only ways of thinking that any given individual uses. In general,

7

however, commonsense and Right wing understandings of social phenomena make extensive use of voluntaristic, often individualistic, explanations of people’s behaviourand circumstances, so that they are represented as whollyresponsible for their character, for what they do and howthey live, and for the consequences of their actions. Theeffects of social structures and discourses which sociologists in particular emphasize go unnoticed or are ignored. Even if it is certain groups which are blamed for their situation, rather than individuals, it can be represented crudely as simply their choice. There are also forms of spurious essentialism or determinism, in which behaviours are explained as simply products of ‘those kinds of people’ (‘that’s just how they are’), as implied by George Bush in his references to ‘evil-doers’.In some cases their genetic make-up may be invoked. Theselatter tendencies, which imply that such people could notbehave otherwise, are often inconsistently combined with attributions of blame and hence accountability for what they do, but then logical consistency is not a requirement of commonsense thinking.5 Although voluntarismand essentialist determinism are opposites with regard toresponsibility, they have in common a failure to acknowledge any social influence on behaviour and character. Again, this is not to deny that there are alsomore adequate forms of explanation in some kinds of popular thinking, particularly where there is some degreeof politicisation, indeed in the practical business of everyday life, individuals may flip between different forms of explanation without noticing.

By contrast, social scientists, particularly sociologists, tend to attribute causal responsibility primarily to actors’ social circumstances (including prevailing discourses), de-emphasizing individual or group (causal and moral) responsibility. Not surprisingly, they tend to be more Left leaning in their politics than non-academics. Of course, there are also some Right wing academics who tend to have individualistic and voluntaristic social theories, such

5 As Bourdieu argued, consistency is not always necessary for successful practice (Bourdieu, 2000).

8

as those provided by mainstream economics, which diminishawareness of social injustice precisely because they makeit appear that individuals are wholly responsible for their situations and argue that most or all action is self-interested. Thus arguments between Left and Right and interventions of academics in popular political debate tend to involve different kinds of explanation of social phenomena.

The Right often accuses the Left or progressives6 of either denying or condoning instances of anti-social behaviour in subaltern groups – ‘refusing to face facts’.If the Left concedes that such behaviour occurs but argues that they are products of social conditions, the Right can treat the concession as equivalent to an admission of defeat, for they are acknowledging the very things for which the Right attacks those groups. Alternatively they can easily appear to falsify the second claim by pointing to all the cases where individuals who have experienced poverty, misrecognition and social exclusion have not engaged in anti-social behaviour.7 It can thus ridicule the Left as saying that ‘society is to blame’, and for not only being unable to explain the behaviour of those - often the majority - whodo not indulge in such behaviour, but for apparently impugning the characters of these law-abiding and well-behaved members of disadvantaged groups. A more sophisticated but disingenuous strategy is to accuse the Left of demeaning those they seek to defend by denying them any individual responsibility, as if they were ‘dupes’ or ‘cultural dopes’; in effect, the Left is held to infantilise them, treating them as pure victims, again, consequently disrespecting those in the same context who do not develop anti-social behaviour. In response, the Left could argue that it is not inevitable that social disadvantages produce crime, but merely likely in severe cases, other things being equal, and that there6 I shall henceforth use ‘Left’ and ‘progressives’ interchangeably.7 This is what is known in the philosophy of science as ‘naïve falsification’ for it may merely identify a situation where the hypothesized relationship is blocked or overridden and hidden by other processes rather than a falsification. Citing aeroplane flight as a falsification of the law of gravity would be an example.

9

are likely to be exceptions, although even these aren’t simply a result of individual responsibility but of contingent factors such as belief systems.

While I think this kind of argument is correct, it’s highly vulnerable from an eristic point of view, for suchqualifications are likely to be taken as partial admissions of defeat, and those who make them may be advised ‘to stop digging a hole for themselves’. It’s notthat it’s a case of ‘death by a thousand qualifications’,for in an eristic context, even one or two qualificationsare likely to be read as an admission of defeat. By contrast, in the context of an argument or discussion seeking the best account, such qualifications would be regarded as constructive.

The Left’s argument about anti-social behaviour within disadvantaged groups is full of qualifications and tight turns, whereas the Right wing pathologisation is simple, and resonates with the individualising tendencies of commonsense. Few are likely to engage in the deeper reflection that shows it to be incoherent. This is typical of the situation in which progressives find themselves – apparently equivocating over qualifications and complications while their opponents deal in straightforward, easy to understand terms. Acknowledging social context, discourses and structures of domination and analyzing their effects produces more complex explanations than do voluntaristic accounts. For example,any adequate explanation of particular forms and instances of crime would have to include and integrate social conditions, social relations of gender, class and perhaps ‘race’, socialisation or subjectification, discourses, beliefs, and individual responsibility. It would seem, therefore, that a large part of the problem is that dealing adequately with matters of responsibilityis unavoidably difficult. And of course, as we see in theoretical debates about discourse, ‘the subject’, and structure and agency, social scientists disagree among themselves over what particular kind of social explanation is legitimate.

10

Not surprisingly, in view of these difficulties and dangers, progressives often choose to avoid any comment that might seem complicit in the pathologization of oppressed groups; in other words they may opt for a ‘don’t go there’ strategy. Thus, the ethical socialist, Richard Titmuss and his followers, were resolutely opposed to anything that might appear to reopen the debate about personal responsibility for social pathology. He refused all discussion of ‘problem families’, for instance (LeGrand, 2003). Pierre Bourdieu also discusses the difficulty of describing inequalities and oppression in a way which does justice to the effectsthey have on people without pathologizing them, that is without presenting descriptions of the very things which the Right despises and uses to justify inequalities (Bourdieu et al, 1999). Thus he and many other radical sociologists are wary of giving detailed descriptions of the appearance and speech of members of oppressed groups lest they invite class or racist contempt in a mostly middle class, white readership.8 These are moral-politicaldecisions made against the backcloth of wider cultural representations that individualise social problems and stigmatise the oppressed.

I respect these decisions, but it is important to recognise that in making a concession to this wider systematically-distorted form of communication by avoiding such matters, such authors have to distort theirown accounts, by evading issues which could and perhaps should be addressed. Poverty, exclusion, misrecognition, domination and lack of hope are generally not ennobling, and are likely to produce effects on people’s characters and behaviour which are problematic, such as self-abnegation, shame or drug dependency (Nussbaum, 2001; Sayer, 2005). These are part of the grounds for objecting to poverty, misrecognition and the like. However, because these pathologiesare often the triggers for Right wing contempt, which sees them as evidence of inherent failings (‘that’s just how they are’ and hence that their situation ‘is all theydeserve’) the Left may be tempted to ignore them and present somewhat idealised accounts of oppressed groups. 8 E.g. Skeggs, 1997. But see Charlesworth, 2000.

11

This has the additional advantage of enabling links to presentations of more attractive ‘role models’ of such groups that may help to reduce prejudices against them. However, this too can play into the hands of the Right, for it then has two options in response to these idealisations. On the one hand, it can falsify them by identifying examples of the pathologies which the Left has chosen to gloss over. (These pathologies are the stuff of tabloid news, ’reality TV’ and drama.9) Alternatively, it can pretend to accept these idealised accounts, adopting a spurious egalitarianism and affecting respect for those to whom they refer, arguing disingenuously that their moral qualities demonstrate that no harm is done by inequality – they are a ‘proud’ people or community. In this way the Right can oscillate between a disingenuous ‘noble savage’ rhetoric and one which stigmatizes the subaltern. Equivalents are common in sexist discourse in the form of flips between misogynist contempt and idealisation of feminine subordination and self-abnegation.

Those on the Left might do better to counterattack by arguing that while individuals have some responsibility for their behaviour, purely individualistic explanations of anti-social behaviour are incoherent, implying local concentrations of uncaused evil.The Left can cite cases where crime rates or other socialpathologies have risen following a worsening of social and economic conditions, and thus require a social explanation. Thus, they could point to the example of theSouth Wales valleys, which in the early and mid-20th century had a vigorous, proud working class culture and the most strongly organised labour movement in Britain. Now, with the closure of the mines and massive deindustrialisation, this culture has collapsed and the valleys are now sometimes locally referred to as ‘the heroin capital of Britain’. Such cases show that the causes of the behaviours which the Right interprets as 9 As official disapproval of racism develops, racism in such media tends to adopt more disguised forms, but their contempt for the so-called underclass remains uncensored and unrestrained. On the coexistence of political correctness about race and sexuality with class contempt, see hooks (2000).

12

products of defective individual character – or perhaps even genetics - are primarily social. Similarly, individualistic explanations of other problematic behaviours, such as terrorism, simply fail to account forwhy they crop up in particular times and places, usually in response to changed contexts.

While this is surely an effective, telling argument, likeany largely social explanation, it also opens up another flank which the Right can attack10 – the vexed issue of the relationship between explanations of behaviour and justifications. Explanations of such behaviour that referto oppression and the like are likely to be interpreted, often captiously, as justifications or excusals of the behaviour. Thus those who try to provide social explanations of, say, terrorism, are likely to encounter horrified reactions – or affectations of horror - that they are condoning or justifying it. This may be an understandable yet mistaken response11 to difficult philosophical issues about reasons and causes and free will. The differences and relationships between explanations and justifications are indeed complex, involving problems of the nature and extent of responsibility, though the distinction is not unusual in lay thinking (Butler, 2004; Sayer, 2005; Smiley, 1992; Williams, 2004). There are also unsatisfactory responses to it in social science, particularly determinist ones which reduce actors to dupes - that is, beings who do no more than reflect and relay external forces.12 Circumstances x may explain a certain behaviour y, but x does not justify y in the sense of warranting approval ofy as a reasonable or acceptable thing to do. People can do things for bad yet understandable reasons; to deny this would be to imply that lay knowledge and reasoning were infallible, and that people were always heroically

10 Sporting or military analogies seem particularly appropriate, for in eristic argument, many forms of attack create a vulnerability to counter-attack.11 That a belief can be 'understandable yet mistaken' of course already presupposes that to explain something we do not have to approve of it.12 The word 'ought' itself presupposes at least some autonomy and responsibility (Midgley, 1984).

13

moral. If we identify a bad reason as a cause of an action this helps to explain it but clearly does not justify it. The invasion of Iraq might be partly explained by western interests in its oil, but that does not justify it.

The distinction between explanation and justification actually requires us to acknowledge that individuals do have some degree of responsibility for their actions. Only then does it make sense to say that although we understand that certain things influenced someone’s actions, perhaps putting them under pressure, we think that they should nevertheless not have reacted as they did, although sometimes we may want to say it would have been a lot to expect them to have done otherwise, and that it’s easy for those in more comfortable circumstances to underestimate the pressures encouraging such reactions. It has to be said that this is something which sociology has often been reluctant to acknowledge either because of the use of reductive accounts of action(as in behaviourism or in post-modern approaches which assume ‘the death of the subject’), or else because of a tendency to treat individual reflexivity and responsibility merely as an uninteresting source of incidental noise mediating social pressures or mechanisms.13 The discipline of sociology was partly a product of a reaction against individualistic explanations, but while that was and still is needed, it does not warrant a flip to sociological reductionism or determinism. Although actors may do a great deal through habit and socialization they also frequently monitor their situation and conduct, and have to make decisions and deal with dilemmas (Archer, 2003: Billig et al, 1988). In their own behaviour, social scientists, like anyone, hold individuals to account for many of their actions; for example, they hold their students to accountfor the quality of their essays, rather than seeing this simply as a product of social circumstances. The challenge of social scientific explanation is to take

13 Insofar as this reflexivity produces varied responses to the same situations, it is an irritation to researchers hoping to find regularrelationships between behaviour and context.

14

account of this, instead of excluding or ignoring it or merely flipping from voluntarism to social determinism.14

Purely voluntaristic explanations of behaviour fit comfortably with what the psychologist Melvin Lerner termed ‘the belief in a just world’, that is, a world in which goodwill, effort and merit are rewarded and ill-will, lack of effort or merit are penalised, so that people get what they deserve (Lerner, 1980). Thus, as regards economic rewards, it implies that you are paid what you are worth and worth what you are paid, and that redistribution is therefore unfair, since it seems to penalise the virtuous and reward the deficient. The belief presupposes the absence or unimportance of either what philosophers call ‘moral luck’, in which unforeseeable contingencies interrupt the relations between intentions, actions and effects, or of mechanismswhich systematically allow domination, expropriation, deceit and misrecognition; the world therefore appears tobe morally ‘well-ordered’. This is not merely something that many people happen to believe, on the whole, but something that they want to believe. Who would not want tobelieve that well-intentioned efforts will be rewarded, and ill-intentioned actions frustrated, and to live in a society that ensures this? It is easy for normative beliefs about something so important to encourage wishfulthinking so that people imagine it to be true or mostly true as a description of the social world.15 However well-meaning it may be, it ironically condones and legitimizesinjustice, and the blaming of victims, for they merely get what they deserve (André and Velasquez, 1990). It is therefore double-edged, on the one hand encouraging virtuous behaviour, on the other concealing or legitimizing injustice. The belief in a just world is strongest amongst conservatives, and is common among subordinate as well as dominant groups. It also discourages the ‘unsuccessful’ from blaming others or 14 This is also a flip from 1st and 2nd person justifications to 3rd person explanations of actions (Manent, 1997).15 Lerner found that in experiments in which subjects observed people playing games in which they knew that outcomes depended wholly on luck, the observers nevertheless revised their opinions downwards of those who lost.

15

resenting the system, and encourages them to make renewedefforts. The belief in a just world is stronger in some societies than others: strong in the United States where it is central to ‘the American dream’, leading the poor often to blame themselves for their situation (Macleod, 1995: Gilens, 1997); weak in more strongly politicised societies like France (Lamont, 2000). There is clearly a relationship of mutual reinforcement between the belief in a just world, and the tendency we noted earlier to adopt voluntaristic, individualistic or essentialist explanations of behaviour.

Again, this is not to deny the presence of other tendencies in popular thought; even individuals who interpret some situations in terms of a belief in a just world may also, in other situations, invoke luck, or eveninjustice in accounting for them.

Lakoff’s strategy and the nature of values

So far I have discussed the problem of how academics might engage with populist political discourse in a way which assumes that what matters is the adequacy of the explanations of matters of political concern, so that political debate appears as an argument between rival analyses. George Lakoff regards this as fundamentally misconceived as a representation of mainstream political discourse. In Don’t Think of an Elephant! Know Your Values and Frame the Debate (Lakoff, 2004) he argues that elections are decided primarily on values rather than policies, and that if progressives are to win against the Right, they need to realise this and stop arguing within value framesset by the Right and about policies, and argue instead from and for their own values. Thus, Lakoff criticised USDemocrats both for failing to realise this and for accepting rather than challenging the Republican way of framing debates and issues. He argued that the Republicans mobilise a frame of values involving a ‘strict father model’. This involves obedience to authority, self-discipline and self-reliance, the US as global moral authority, patriarchal families and

16

heteronormativity, respect for the wealthy as deserving and contempt for the dependant as undeserving. (Although he does not make the connection, the strict father model complements a voluntaristic folk sociology and a belief in a just world.) Rather than respond within this frame, progressives should have posed their own, and here Lakoffproposes the ‘nurturant parent model’ as an alternative, which he feels would be more appealing to voters if characterized effectively. This implies, that faced with difficult issues on which progressives would rather not fight, they should adopt the ‘just don’t go there’ strategy.

Lakoff’s critique was borne out by the pre-election debates between Kerry and Bush in 2004, where Kerry wouldinvariably answer any question by stating facts about thetopic or appealing to authorities, and then from this deduce and defend his policies. Bush, by contrast, usually began with bald statements of his beliefs or convictions, with little or no reference to contingent facts. Kerry’s tactics played into the hands of the Republicans who had portrayed him as indecisive (‘flip-flop man’). Ironically, although the Kerry strategy seemsmore rational in terms of justifying what needs to be done by reference to facts or what experts say, in sifting through evidence and allowing his values to appear to be dependent on it, he indeed came across as lacking conviction and not knowing what he thought. Bush knew what his position was, he had strong convictions, and he stuck to them (regardless of the evidence!).

Lakoff is a Professor of Linguistics but his book is intended for a popular audience, so it lacks the usual complexities, qualifications and measured claims that onewould expect in academic discourse. Not only is it partisan, in that it is written for progressives, but it doesn’t defend its own political standpoint; it assumes the readership already agrees with the author’s politicalposition and hence does not require arguments for it. It therefore comes across as largely instrumental, indeed itmight be accused of being somewhat eristic, for it tells progressives what to do if they want to win – namely,

17

frame the debate in terms of their own values - without providing much argument either for the prioritisation of value frames, or for the particular values favoured by progressives. Furthermore, even if we accept Lakoff’s factual claim that people vote primarily according to values, not policies, we might regard this with alarm - as a degenerate form of democracy in which political discourse has been thoroughly ‘dumbed down’, and hence a condition that should itself be challenged, rather than one that we should happily play along with. It seems to imply that progressives should drop their academic analyses and arguments and resort to populism, albeit a progressive populism. Thus, for those who favour progressive, Left politics, and want to start winning, the book is instructive from an instrumental point of view, but it is unsatisfactory from the point of view of a normative evaluation of what the nature of political argument should be.

On Lakoff’s account in Don’t Think of an Elephant, values appearto be beyond debate about circumstances and the way the world is, indeed there is an implicit fact:value dualism in his argument, implying that values are ‘merely subjective’ and beyond the scope of reason, which can deal only with the realm of fact. One way of interpretingLakoff’s recommendations of a value-based politics, then,would be to see individuals as having complex, internallydiverse value systems, so that one can perhaps persuade them to vote one way rather than another by evoking different parts of these value systems. The Right may play upon fears, aversions, xenophobia, self-interest andsense of individual rights and liberty, while the Left may play upon feelings of solidarity, need of others, andbenevolence. However, on its own, this seems to be reducible to an a-rational or irrational matter of ‘pressing the right buttons’. Lakoff not only omits to explain how values are within the scope of rational deliberation about people and the world, but his failure to defend his own value position lends weight to the viewthat they are not the kind of thing that is rationally defensible. People just have them, and Lakoff’s values happen to be ‘progressive’ in some indefensible sense.

18

However, this subjectivist, irrationalist impression is mistaken. In his earlier book Philosophy in the Flesh, co-authored with Mark Johnson, Lakoff argues that moral values are defined by a range of metaphors that “are grounded in the nature of our bodies and social interactions, and they are thus anything but arbitrary and unconstrained. They all appear to be grounded in various experiences of well-being, especially physical well-being. In other words, we have found that the sourcedomains of our metaphors for morality are typically basedon what people over history and across cultures have seenas contributing to their well-being.” (Lakoff and Johnson, 1999, p. 290; see also Lakoff, 2002). Clearly, Lakoff does not believe that values are ‘merely subjective’ and unrelated to objective characteristics ofthe world, including those of subjects themselves, but hedoes not counter this subjectivism in Don’t Think of an Elephant, thereby inviting us to wonder why one set of values should be preferred over another. Consequently, the aims of the book are likely to seem arbitrary for it doesn’t explain why we should prefer progressive values to conservative ones.

Some familiar strategies of political debate clearly do appeal primarily to our values, either by appealing to our fears and resentments or to ‘our better nature’. For example, a possible way of responding to racism, sexism, homophobia and the like is to denounce them in various ways as unethical, for example, as racist, sexist, etc. This has now become a common form of official response toracist discourses. It has the advantage of simplicity andappealing to the audience’s ‘better nature and motives’ of respect for others and openness and generosity. Presumably Lakoff would regard it as a more effective strategy than ones discussing particular circumstances and policies, because it appeals directly to values rather than matters of fact. Yet, while the value strategy may make some headway, it is unlikely to be sufficient, especially if misrepresentations and individualised explanations of subaltern groups remain dominant in popular culture, for values are not

19

independent of perceived facts. It’s hardly a strong explanation of the position of subaltern groups to imply that it derives wholly from the deliberately unethical prejudices of others, as if social structures and unintended consequences had nothing to do with it.

It is not that people intend to unethical: for example, empirical studies show that few people are willing to refuse any notion of equality of some sort, and most racists will say they “believe in equality but . . . “ (Myrdal, 1962; Billig et al, 1988). A common strategy in moral tales in literature and drama is to portray a member of a stigmatised group as having moral qualities that demand recognition. Certainly, prejudices often involve a strong element of ‘othering’, in which groups project qualities they fear and dislike onto others, ‘purifying’ themselves in the process, in effect, saying,‘we are virtuous, they are bad’. But even then, they do not claim that the virtues they think they monopolise andwhich their others lack are only good for themselves. Thus if they see themselves as hardworking and self-reliant and their others as idle and irresponsible, they are not assuming that hard work and self-reliance are only good for their own group; rather they are assuming that they are good for everyone, but that they happen to be the only ones who have such virtues. Unless this were the case, they would have no reason for disparaging thosewho they considered to lack those qualities. Furthermore,in treating their values as universal in scope, they expose their assumptions about who does and doesn’t matchup to their standards to the possibility of falsification. It may be possible to show them that members of stigmatised groups are in fact virtuous, as they imagine themselves to be (Sayer, 2005). Thus people are not generally prejudiced against some groups because they want to be unethical, but because they believe theirothers are unethical or otherwise deficient; indeed they are likely to believe their negative feelings about theirothers have a justifiable moral basis. (We sometimes meetpeople who at first seem thoroughly generous and moral, only later to discover that they hold racist beliefs or other forms of prejudice.) Where this is the case, moral

20

appeals are unlikely to work on their own, for it needs to be shown that the value-judgements are mistaken about their objects. This reminds us that values are about something, they have cognitive content, and are susceptible to argument in relation to their referents. Merely denouncing prejudices as unethical in some respectis likely to have less effect than demonstrating that their empirical assumptions are false and based on ignorance. Thus, in the case of asylum seekers, trying tomake others more aware of their humanity – they are not ciphers but ‘real people’, moral beings like any other - and the horror of the conditions which they are seeking to escape, may have more impact than denunciations of racism, which may merely prompt jibes about ‘political correctness’.

In considering such possibilities one may have difficultydistinguishing facts from values, but that it is the point: in matters that concern well-being, or needs, or flourishing and suffering, the dualism collapses. In making sense of the social world we use what philosopherscall ‘thick ethical concepts’, indicated by terms like ‘cruel’, ‘kind’, ‘considerate’, ‘selfish’, ‘supportive’, ‘sexist’ and the like, which transcend the fact:value dualism. They are simultaneously descriptive and evaluative, and the two cannot be separated without loss of meaning. The more moral values are divorced from descriptions or understandings of states of affairs, the more arbitrary, irrational, moralistic (or in some cases the more hysterical), they seem, appearing as empty exhortations or denunciations (Putnam, 2002). We should stop de-rationalising them by divorcing them from what they are about. They are the product of evaluative judgements of matters affecting our well-being – and sometimes that of non-humans, and we argue about them by reference to what they are about, by trying to ‘do justice to’ their objects (Sayer, 2006). More precisely, values are sedimented valuations of things that have meaning for us, and have become attitudes or dispositions, which we come to regard as justified or worth having, usually because they bear upon our well-being and that of the things and relationships that we

21

care about. They differ from mere preferences or inclinations in that to some degree we have reflected upon them, and not simply absorbed them, though often they are only rarely articulated; we may find ourselves thinking ‘that kind of behaviour is not right’, or ‘I like that person’s values’ without being able to say why very precisely. Full justifications tend to be left to philosophers. They become habitual, but they are habits of thinking to which we have become committed or emotionally attached to and give us our character. They merge into emotional dispositions and they recursively guide and inform the evaluations we make of particular things, as part of our conceptual and affective apparatus. The acquisition of values in everyday life lies between the two extremes of passive osmosis and considered judgment.

Values thus represent both determinants and products of particular evaluations of specific things; that is, they both guide particular evaluations and are confirmed, and occasionally modified, by them. Values are not merely subjective or a priori, but are themselves to some extent the product of interactions, and discursively and culturally mediated. In turn, our evaluations are not necessarily solely a function of our values, indeed they would be problematic if they were because they would thenbe indifferent to what was being evaluated and put us at risk. Values themselves may sometimes be weakened or strengthened by having to evaluate novel situations, which may disturb the sediments, though this does not generally happen easily, precisely because they are things to which we have become committed and part of our characters. Just as theory-laden observations need not betheory-determined and can register anomalies, so value-laden evaluation need not be indifferent to the qualitiesof the thing or situation being evaluated. Values form complex networks of difference and generally include tensions and contradictions, though sometimes we may notice these and try to resolve them.

Judgements can only be made in terms of available cultural discourses, but they can be susceptible to

22

challenge from within. Thus, for example, the assumption of capitalist culture that material wealth is the key to happiness and well-being is open to empirical challenge (Lane, 1991). Non-eristic political debate – and often eristic debate - is very much about finding ways of life that promote more inclusive forms of flourishing. Hence, understandings of the nature of contemporary society, of what kind of lives people are leading, and of the effectsof policies, should not be regarded as something separatefrom matters of values, but part of what those values areabout. These assessments of needs, etc., are of course subjective in the sense that they are made by subjects, but they are not simply matters of pure ‘subjective’ values, unrelated to the world, but assessments of objective states and possibilities. The assessments may be fallible, of course, but in that respect they are no different from any other judgement or descriptive or explanatory claim. Their fallibility implies there is something objective (in the sense of independent of theirbeliefs) about which they can be mistaken (Collier, 2003). Some aspects of flourishing may be difficult to define and evaluate, but some, such as the ability to live without stigma or threat of physical violence, seem straightforward. Awareness of the former should not temptus into acceptance of the absurd idea that flourishing and suffering are no more than products of collective wishful thinking. We cannot help but adopt an evaluative orientation to our condition and the world around us because we are needy beings capable of suffering as well as flourishing.

Here it is vital not to confuse different meanings of ‘objective’, as so often happens on the Left. Objective (1) in the ontological sense of pertaining to objects, things that exist regardless of what someone thinks aboutthem (oxygen and exploitation can exist regardless of whether anyone notices them), is different from objective(2) in the epistemological sense of ‘true’, concerning truth claims like ‘6 million people were killed in the Holocaust’, and from objective (3) in the sense of ‘value-free’. There are corresponding opposite meanings of subjective (Sayer, 2000). Thus a value-laden

23

statement, like ‘torture causes suffering’ (subjective 3)is not necessarily subjective in the sense of untrue (subjective 2). Acknowledging the value-laden character of discourse does not therefore require us to abandon theconcepts of objectivity in the first and second senses.

This critique of subjectivist conceptions of values is not wholly to reject Lakoff’s position in Don’t Think of an Elephant or to defend the ways of doing politics that he was attacking. It implies a different position from either, one in which we do more to relate the abstractions of political discourse and the particular practices and forms of organisation in question, to particular experiences and hence their implications for well-being and ill-being. So instead of treating the nurturant parent model simply as an alternative to the strict father model, and hoping it will be more attractive, we can argue why it’s more attractive, in terms of enabling people to experience an expanded form of flourishing. For particular cases, for example, responses to immigration or class or gender, this will require arguments going into descriptions and explanations of what these involve.

24

Arguing about valuations

Part of the problem with the privileging of values in theframing of discourses, is that their intimate relationship to understandings of matters of fact is unnoticed, indeed they are treated as if they were entirely different. It also fails to challenge the fact-value dichotomy and the customary argument that ought can not be logically deduced from is. These ideas have been attacked by a large number of philosophers (e.g. Anderson, 2004; Collier, 2003; MacIntyre, 1998; Nussbaum,1993; Putnam, 2002; Taylor, 1967; Williams, 1985). Like Charles Taylor, I would argue the value frames are also implicitly descriptive, suggesting a certain general viewabout human needs and nature and about what enables people to flourish or suffer (Taylor, 1967; see also Nussbaum, 1993). Taylor argues that even though particular terms, phrases or sentences may be largely value-free when taken in isolation, the frames implied bythe larger discourses of which they are part create what he calls a ‘value-slope’, which tends to ‘secrete’ certain values and encourage the reader or listener to evaluate the phenomena being described in a certain way. Thus an account of economic matters which centres on consumers choosing what to buy is likely to create a positive response in that it highlights individual liberty and autonomy, while one which centres on job insecurity and unemployment is likely to secrete more critical responses because it suggests failure to satisfypeople’s need for security. (That ‘secretion’ does not have the logical force of deductive reason does not matter;that ‘the wire is live’ doesn’t logically entail ‘the wire is dangerous’ or ‘you ought not to touch it’ is not a good reason for ignoring the normative judgments.) While readers or listeners are of course free to disagreewith the secreted values, they might be expected to justify doing so. Two options are then open to them. Theycan respond by attempting to undermine the account, for example, denying that there is any job insecurity or thatunemployment causes any harm. Alternatively they can adopt an overriding strategy, that is, concede that the

25

secreted values have some warrant, but argue that other concerns override them; for example, argue that while economic insecurity is a problem, it is a lesser evil than possible alternatives, such as guaranteed employment, which might inhibit economic development. (Obviously, the argument might not end there!)

To some extent, this is what already happens, at least inthe more open and less eristic kinds of political debate.Moreover, we can evaluate the relative merits of different frames by assessing their explanatory power. Aswe saw, a wholly individualistic view of society, as assumed in Right wing accounts, runs into difficulties inexplaining why particular kinds of behaviour correlate strongly with social conditions which are not of individuals’ choosing. A wholly sociologically determinist theory which allows no space for individual responsibility implies theory-practice contradictions in that we cannot function in social life without attributing, and accepting, such personal responsibility.An approach which integrates social influences and scope for reflexivity and responsibility can explain things which neither of these one-sided theories can. We should therefore, other things being equal, prefer the values itsecretes to those secreted by the reductionist theories (e.g. either that individuals should be held wholly responsible for their actions, or that we should blame social conditions only, never individuals.) This may be difficult, where eristic debates are concerned, but sometimes even slogans may succeed in challenging certainframes and suggesting others. Tony Blair’s slogan: ‘Toughon crime. Tough on the causes of crime’, was actually a brilliant example. While no slogan can be expected to capture complex arguments, they can at least set people thinking in productive directions – in this case thinkingabout crime in a way which implied both individual responsibility and social causes. (The only pity was thatit wasn’t acted upon!).

Taylor’s approach might be described as a methodology forreasoning about matters of values, but again, the values themselves imply claims about matters of fact, for

26

example, what causes suffering. The frames we favour simultaneously imply what seems to us to be the best available understanding of human needs and capacities anda system of related values. We can also respond to arguments made from within framings which we find inadequate by undermining them. Even where we accept the values that a certain account secretes, there may still be grounds for refusing them by means of an overriding argument.

Conclusions

Commonsense and Right wing thinking are resistant to social forms of explanation not only because of the latter’s complexity, or the former’s eristic tendencies, but because of their implicit theories of behaviour and society. Commonsense accounts often use voluntarist, individualistic or essentialist explanations, which naturalise and normalise contingent social circumstances,and consequently blame and credit people for social circumstances for which they have little or no responsibility. The dialectic of responsibility and susceptibility to social influence forms a kind of front-line of confrontation in debates between Left and Right. In an eristic context, the Left’s use of mainly social explanations makes it not only vulnerable on account of the consequent complexity of its arguments, but because they tally neither with commonsense interpretations of actions and outcomes in terms of individual responsibility, or indeed with the belief in a just world, both of which are central to the Right’s understanding of the social world.

Lakoff’s advice to the Left to use a strategy which emphasizes values and argues from within its own value model has some advantages, but it is important not to allow it to support a fact:value dualism which divorces values from positive understandings of the world, and consequently de-rationalises debates about values. For example, the belief in a just world is not reducible to aquestion of values, for it transcends the normative-

27

positive distinction and implies a particular account of the way the social world works. While it may be a good strategy in political argument to appeal to people’s ‘better natures’ that is, to their humane values, rather than their ‘base motives’ of selfishness and xenophobia, it is unlikely to be sufficient, for on its own it doesn’t identify the social causes of problems, but just calls for a more moral and benevolent individualistic approach. It may fail to identify structures of domination, mistaken beliefs, and more generally the way in which social contexts and structures generate problems. This, after all, is the usual failing of mainstream social democratic political ideas. It may therefore fail to make much headway against the belief that the world is mainly just. At the same time, while itis important to expose the social causes of injustice, asa normative ideal the belief in a just world should not to be rejected, for it can motivate emancipatory political change.

Our values depend on what kinds of being we take people to be, and on what we take the nature of the social worldto be, and particularly on how we think about needs, flourishing, suffering, and responsibility for behaviour.When we argue about values we imply ideas about these at least indirectly. Changing these ideas can thus change values, and there are rational ways of proceeding in engaging with political values, even if the pressures of eristic debate tend to block them. When we talk about someone or some group ‘becoming politicised’ we generallymean not just that they become concerned about politics. We also usually mean that they come to see their situation not simply in terms of their individual fortunes but as part of a social process, both in the sense of realising that others share their situation and,most importantly, that their suffering or concerns are not products of their own lack of responsibility or of bad luck but are caused largely by particular social arrangements which could be otherwise. This is surely also what political argument is about. Although the kind of reasoning about moral political values in political discourse is usually crude it is often important and can

28

lead into a more considered, patient, open and less eristic discourse where people, including experts in relevant fields, discuss issues. I realise that these suggestions do little to resolve the problems of complexity versus simplicity and eristic tendencies in political argument, but I hope to have illuminated some of the other sources of difficulty.

AcknowledgementMy thanks, with the usual disclaimers, to Norman Fairclough and Ruth Wodak for excellent comments on a draft of this paper.

References

Anderson, E. (2004) ‘Uses of value judgements in science:a general argument,with lessons for a case study of feminist research on divorce’ Hypatia, 19 (1), pp.1-24André, C and Velasquez, M (1990) ‘The Just World Theory’,Issues in Ethics, Vol.3 (2)Archer, M. (2003) Structure, Agency and the Internal Conversation, Cambridge: Cambridge University PressBillig, I. et al (1988) Ideological Dilemmas, Beverley Hills,CA: SageButler, J. 2004 Precarious Life, London: VersoBourdieu, P. 2000, Pascalian Meditations, Cambridge: PolityBourdieu et al (1999) The Weight of the World, Cambridge: PolityCharlesworth, S. 2000, A Phenomenology of Working Class Experience, Cambridge: Cambridge University PressCollier, A. 2003, In Defence of Objectivity, London: RoutledgeVan Eemeren, F.H. and Grootendorst, R. (1992) Argumentation,Communication, and Fallacies: A Pragma-Dialectical Perspective, Hillsdale, NJ: Lawrence ErblaumVan Eemeren, F.H. and Grootendorst, R. (2004) A Systematic Theory of Argumentation, Cambridge: Cambridge University PressFairclough, N. (2003) Analysing Discourse: Textual Analysis for Social Research, London: RoutledgeGilens, M (1999) Why Americans Hate Welfare: Race, Media and the Politics of Antipoverty Policy, Chicago: Chicago University Press

29

Habermas, J 1979 Communication and the Evolution of Society, London: Heinemannhooks, b. 2000, Where We Stand: Class Matters, New York: RoutledgeLakoff, G. (2002) Moral Politics: How Liberals and Conservatives Think, Chicago: University of Chicago PressLakoff, G. (2004) Don’t Think of an Elephant! Know Your Values and Frame the Debate, White River Junction, VT: Chelsea Green Publishing CoLakoff, G. and Johnson, M. (1999) Philosophy in the Flesh: The Embodied Mind and its Challenge to Western Thought, NY: Basic BooksLane, R. E. 1991, The Market Experience, Cambridge: Cambridge University PressLe Grand, J. (2003) Motivation, Agency and Public Policy: Of Knights, Knaves, Pawns and Queens, Oxford: Oxford University PressLerner, M. (1980) The Belief in a Just World: a Fundamental Delusion, New York: Plenum PressMacleod, J. 1995, Ain't No Makin' It: Aspirations and Attainment in a Low Income Neighbourhood, 2nd edn, Boulder Co: Westview PressManent, P. 1998, The City of Man, transl. By Marc A. LePain, Princeton, New Jersey: Princeton University PressMyrdal, G. 1962 An American dilemma:   the Negro Problem and Modern Democracy, New York: Harper RowNussbaum, M. C. 1993, 'Charles Taylor: explanation and practical reason', in Sen, A and Nussbaum, M. C. (eds.) The Quality of Life, Oxford: Clarendon Press, pp. 232-241Nussbaum, M.C. (2001) Upheavals of Thought, Cambridge: Cambridge University PressO’Neill, J. (1998) ‘Rhetoric, science, and philosophy’, Philosophy of the Social Sciences, 28 (2) pp.205-225Putnam, H (2002) The Collapse of the Fact-value Dichotomy, Cambridge MA: Harvard University PressReisigl, M. and Wodak, R. (2003) Discourse and Discrimination: Rhetorics of Discourse and Discrimination, London: RoutledgeSayer, A. (2000) Realism and Social Science, London: SageSayer, A . (2005) The Moral Significance of Class, Cambridge: Cambridge University PressSayer, A. (2006) ‘Language and significance – or the importance of import: implications for critical discourseanalysis’, Journal of Language and Politics, 5:3, pp.449-471

30

Skeggs, B. 1997, Formations of Class and Gender: Becoming Respectable, London: SageSmiley, M. 1992 Moral Responsibility and the Boundaries of Community, Chicago: University of Chicago PressTaylor, C. (1967) ‘Neutrality and political science’, reprinted in Ryan, A. (ed.) 1973, The Philosophy of Social Explanation, Oxford: Oxford University Press, pp. 139-70Williams. G. 2003, ‘Blame and responsibility’, Ethical Theoryand Moral Practice, 6, pp. 427-445Williams, G. 2004, ‘Praise and blame’, Internet Encyclopedia of Philosophy: www.utm.edu/research/iep

31