Is there any reason for bringing these ideas together, other than an effort to bridge some of the central ideas in the two days of our conference?

I think there is very good reason just at present. I started to think about the theme for this address around July 7, the time of the bombings on the London transport system. In the context of those terrible events, we heard comments about the extent to which those responsible were ‘beyond the pale’, and the need to reinforce the ‘war on evil’. We also saw discussions about the extent to which the terrorists’ Islamic religious commitments played a role in motivating them.

It was widely denied that there is any necessary link between Islamic commitment and terrorist activity. Nevertheless, we have seen some tension between liberal, western policy and some Islamic communities and countries about values and lifestyle. This tension may have become especially salient since the invasion of Iraq by western-led forces, but it was noted by one sociologist well before that, soon after the attacks of 11 September 2001. He argued that ‘it would be an error for sociologists and political analysts to concentrate on revisions of economic and political theory while not paying equal attention to the moral tensions between Islamic and Western cultures’, and suggested a need for ‘bilateral studies of Western and Islamic conceptions of morality and standards of right and wrong’.1 Recently we have seen the beginnings of some effort at dialogue amongst Muslim and other community leaders. In this address I explore some issues about our need for dialogue.

####Conflict, interests and positions####

There is no reason to suggest that Islam has any more association with violence than any other religion. The literature on group processes suggests that any group commitment can lead to attacks on members of other groups. In the case of group religious commitments, the effects can be especially pronounced because of the depth and resilience of the commitments, and the fact that they are often associated with explicit value commitments. It is the case of tension between Islamic communities and others which is especially salient at present, and which raises some questions for public policy, including policy on freedom of speech as one notable example. In the context of such tensions and policy issues, we see traditional ethical problems about toleration. How far do we tolerate others’ values or activities when we see them posing a threat to our own values or welfare? There are new forces at work to highlight the problems, and new practical implications, but still we see some of the same issues as always.

If these issues have a new salience, it is partly because the approaches to public policy that have become dominant in the West have some special problems dealing with the issues. In Australia and a number of other major western countries we have seen a trend toward policy formation informed by public choice theory, neoclassical economic theory and increased use of markets to allocate resources.

This family of approaches to policy formation and implementation have both a descriptive and a normative dimension. Their descriptive or explanatory dimension revolves around assumptions that individuals’ self-interest is the mainspring of human behaviour, and as a corollary these approaches have a normative aspect oriented toward individual preference satisfaction as a criterion for success and correctness. Such views have major difficulties, which are well known, but have not been enough to overcome the forces in their favour.2

The difficulties include the clear fact that some preferences are vicious or imprudent, as well as problems of interpersonal comparison. From an ethical point of view, they also seem to be enmeshed in versions of utilitarian thought that have difficulty with what many of us consider clear ethical imperatives: duties of good faith, honesty and the like.

In the present context, however, they have another sort of difficulty, the difficulty of dealing with commitments and differences that revolve around religion and doctrine rather than around matters of individual preference satisfaction.

This is not to deny that economic and political factors are a major part of the explanation for events in the Middle East. There are plausible arguments that events of the past few years have been fuelled by western policies and support for oppressive, undemocratic regimes in a number of Islamic countries – including previous regimes in Iraq – and that these policies have to a significant extent been prompted by economic considerations, including the need for reliable access to oil supplies.3

Nevertheless, it also seems plausible to suggest that the values and ethical conceptions of participants in the conflicts of recent years have played a significant part in the events. It is possible to analyse and explain how individuals come to be suicide bombers,4 but it is difficult to imagine explanations which allude only to considerations of self-interest, and which fail to consider the values and commitments of those individuals.

One place where this difficulty has emerged is the advice given some years ago by Fisher and Ury, well-respected and insightful writers about negotiation, the advice to concentrate on ‘interests’ rather than on ‘positions’.5 We can construe ‘interests’ here in terms of preference satisfaction: peoples’ interests are satisfied just to the extent that their preference satisfaction is maximised in the long run. On the other hand, ‘positions’ have to do with such nebulous and difficult ideas as beliefs about what an outcome ought to be, or what a party deserves, or what commitments they have.

The injunction for negotiators to concentrate on interests rather than on positions can often be wise advice. If I approach a wage negotiation with a fixed view about a required pay increase, I may be hindered from seeing the greater benefits of improved leave or conditions. ‘Positions’ may reflect preconceived and ill-informed ideas about what will really be of benefit, and impair our ability to think of other ways to do things.

In some cases, however, the injunction to focus on interests rather than on positions may just distract attention from the real basis of conflict.6 In particular, where conflict is grounded in differences of religion or values, it may be misleading and counterproductive to construe it in terms of interests and to try to settle it by considering what will lead to the best outcome in terms of preference satisfaction: not because preference satisfaction is unimportant, but because the idea of preference satisfaction does not capture everything that is important.

####The importance of dialogue in public policy####

One way that economists or public choice theorists may try to take account of the fact that not only individual preference satisfaction is important, is by using the idea of side constraints. Certain rules are agreed as ones that have to be observed whatever the effect on preference satisfaction may be. Thus, for example, it might be agreed that contracts have to be enforced, or that slavery is unacceptable, or that certain human rights have to be maintained, whatever the effect may be on the preference satisfaction of most individuals. More generally, it might be agreed that certain means are not permissible for achievement of preference satisfaction for any individual or group. For example, it might be agreed that terrorism is beyond what is acceptable, even if terrorism might markedly improve the preference satisfaction of some user group.

This accords with many reported comments from politicians and others in the aftermath of terrorist attacks. Thus, for example, the BBC News ‘Have Your Say’ website7 records such comments in response to tube bomber Mohammad Sidique Khan’s video statement setting out his motives for the July 7 attacks in London:

There is never an excuse or justification for terrorism, no matter how just your cause or how many magic men in the sky you believe in. (Phil, Newcastle, UK)

This person is a confused and evil man. These are the words of someone who has lost their reason and human compassion. (David Burton, London)

Nothing can ever justify the sickening atrocity that this man perpetrated. (Jim, London)

We may be inclined to agree that terrorism ought to be ruled out as a means to any end, and it is not hard to envisage other means that might be ruled out for the achievement of any ends, however good. Unfortunately, however, we do not have a widespread consensus on what can be ruled entirely out of court. We might have thought that deliberate torture would be generally agreed to be unacceptable, but political defences of interrogation techniques at Abu Ghraib prison in Iraq and at Guantanamo Bay have shown us that even here the waters can be muddied by contention over what does and does not constitute ‘torture’, and there has also been subsequent academic discussion of the extent to which torture may on occasion be justified.8 Even some of the BBC website responses are equivocal. One says ‘Everyone is entitled to their own opinion.’ This seems like a descent into extreme relativism, which simply accepts that all preferences are equally legitimate, and equally merit satisfaction in the universal marketplace.

For most of us, though, that descent into relativism seems like a reductio ad absurdum of the family of views that revolve around preference satisfaction through market mechanisms as the touchstone for policy formulation. For most of us, the problem is how to decide on mechanisms that will regulate preference satisfaction in ways that are reasonable and effective. The approach which simply rules some preferences out of court and leaves all others equal, suffers from a lack of consensus regarding which are to be ruled out. It will also suffer from problems about accepting as equal all those preferences which are not ruled out entirely. Mill recognised the need to distinguish pleasures in their quality as well as their quantity,9 but that sort of distinction is hard for market-oriented public choice theorists to incorporate into policy formulation and implementation, because it carries the threat of elitism and the capture of policy processes by bureaucrats whose preferences and values are at odds with those of the wider community (à la Sir Humphrey Appleby, of Yes, Minister fame10). Instead, we are confronted with populist politics that emphasise ever-greater tax cuts to enable individual preference satisfaction at the expense of public goods and care for the poor and marginalised of the community.11

One reason for the popularity of market-based mechanisms in public policy processes is the fact that they take responsibility for decisions away from individuals and place it in an impersonal, mechanistic, market structure. In doing so, they hold out the prospect of greater efficiency and achievement, as well as overcoming populist suspicions of elite bureaucratic values. In so doing, they identify many problems as ‘technical’ rather than ‘political’ or ‘ethical’ problems, and can try to segregate public servant functions into policy formulation and advice, on the one hand, and policy implementation or service delivery, on the other.12

However, there are major difficulties for these efforts. The last-mentioned distinction, between policy formulation and policy implementation, is susceptible to criticisms that attack the distinction between what is done and how it is done. We are very familiar with how these intermingle at the level of face-to-face, interpersonal interaction, and good managers are also aware that the outcomes of decisions are often dependent on how they are implemented. But even at a conceptual level, the distinction is hard to maintain. Perhaps the decision to invade Iraq is policy formulation. Does that mean that all the decisions about how to invade Iraq are policy implementation, rather than policy formulation? Clearly not. Perhaps the decision to restore South Australia’s credit rating is policy formulation. Does that mean that all the decisions about how to do so are policy implementation, rather than policy formulation? Clearly not. How to do things is just as much a matter of policy formulation as what things to do. The relevant distinctions have to do with levels of policy formulation, how wide and important the effects of decisions are, who is affected and to what extent.

The implication is that it is not viable to categorise some problems as technical problems only, to be solved by managers with technical managerial expertise only, and other problems as policy problems, to be solved by policy advisors with expertise in policy formulation rather than policy implementation.

The further implication is that it is not viable to draw a line between one sort of process, the process of policy formulation, and another sort of process, the process of policy implementation, and try to deal with them in different ways. In particular, for example, it is not viable to suggest that processes of policy formulation should involve the sorts of dialogue and discussion that are suitable for addressing ethical and political problems, while on the other hand processes of policy implementation should involve mechanistic routine, whether it be the routine application of handbooks of administrative procedure, or the mechanistic transactions of an idealised marketplace. Both formulation and implementation of decisions is likely to involve some technical issues, but also some issues that need to be addressed by encounter and dialogue with community members who are affected.

In summary so far, then, I am contending that public policy formulation and implementation at all levels is likely to require dialogue with community members, because the approaches to public policy which revolve around market mechanisms and individual preference satisfaction simply cannot account for such facts as that some preferences are more worthy than others and that it is not possible effectively to segregate policy formulation from policy implementation and service delivery. Acceptable service delivery no less than sound policy formulation will require dialogue with community members. It will require ‘voice’ and not only ‘exit’ behaviour, in Hirschman’s terms.13 In Elster’s terms, it will require the processes of the forum as well as the processes of the market.14

Dialogue, public policy and religion

Dialogue is important in a variety of public policy contexts, from local community policy implementation to international efforts at dealing with terrorism. The need for dialogue and problems with dialogue emerge in some of the other responses shown on the BBC website after Mohammad Khan’s video statement. While one respondent says ‘Your beliefs and ways are just something I will never understand’, and another ‘It makes me so furious watching that sickeningly evil person try to justify his carnage with a completely flawed argument’, yet another says ‘It’s depressing that no one seems to want to talk about problems but rather just kill each other’.

Some prominent theorists have recognised the importance of dialogue in social and political processes, and I shall not try to discuss all the important work that has been done.15 All I shall attempt is to identify some of the things we can say, without trying to explore them in full.

But what more can be said about the nature and use of dialogue? Is not the call for dialogue a well-worn cry? A proponent of market processes may say the position I have stated is nothing more than the well-worn advocacy of paying attention to customers’ wishes. There is no point in contending that more dialogue is required unless there is more to be said than that. If there is more point to the suggestion than that, it is to identify the fact that the sort of dialogue required involves more than just paying closer attention to others’ preferences and encouraging them to articulate those preferences more fully. It involves exploration of value commitments and other commitments, as well as of preferences and inclinations. Worthwhile dialogue is inhibited by policies that constantly increase the role of market mechanisms and by populist politics which emphasises the satisfaction of individual preferences without regard to the nature or worth of those preferences. Undoubtedly, the devices and processes of the forum have their own pitfalls, just as those of the market do16, but the implication ought to be that we devote time and attention to improve processes of voice and dialogue, not that we reject them in favour of market processes.

The sort of dialogue that is required must go far beyond the mere exploration of one another’s preferences and inclinations. It cannot even content itself with exploring ‘interests’, but has to get to grips with others’ ‘positions’, as well as their interests. Indeed, in many of the sorts of conflicts and policy processes we are concerned with, we might enjoin participants to ‘focus on positions, not interests’: quite the opposite of the approach that Fisher and Ury showed to be appropriate and successful in commercial negotiations. Why focus on positions? Because in many of the sorts of conflicts and policy processes we are concerned with, what are important are the concerns people have and the decisions they make and the actions they carry out which emerge from values and ethical commitments, and there is evidence that these are bound up with people’s positions and social identities.17

What more can we say about the sort of dialogue that is needed? One thing is that it cannot stand on extreme relativism, because it has to aim at some agreement. There are areas of life where we can agree to go our separate ways, but in the sorts of conflict and policy process we are thinking about, some joint action is required, whether it be agreement about government processes, royalty payments for oil production, sources of energy, gay marriage or industrial relations institutions.

On the other hand, however, it also seems unlikely that the sort of dialogue required can assume that there is a single correct solution to be unearthed by processes of technical or scientific enquiry. To accept this, we do not have to believe that there is a supernatural realm accessible only through processes of faith or mystical revelation. We know quite uncontroversially that there are ‘ways of seeing’ that reflect cognitive differences amongst individuals which nonetheless are not distinguished by the fact that some are true and others false. Wittgenstein drew our attention to the ‘duckrabbit’ picture that can be seen either as a duck or as a rabbit.18 In his discussion of ‘seeing-as’ and ‘seeing aspects’, he notes that ‘there are here hugely many interrelated phenomena and possible concepts’, and it would be a mistake to over-simplify discussion. Nevertheless, we can note that the idea has subsequently been the subject of extensive and detailed investigation by cognitive psychologists, and that some of this investigation has used the concept of ‘framing’ for a related family of ideas.19 For our purposes an important feature of framing was clearly identified in empirical study by Kahneman and Tversky, where they showed that different cognitive ‘decision frames’ can significantly effect people’s behaviour, even though the differences between the different frames—or ‘ways of seeing’—cannot be distinguished on the basis that one is true and the other false. One well-known experiment showed that people’s attitude to risk was significantly affected by whether they were led to frame a decision as letting some people die or as saving the remainder.

The idea of ‘seeing as’ does not seem to have been taken up so widely as one might have expected in the philosophy of religion.21 The possibility of exploring it in detail must also be put aside here. However, it is enough to make sense of the idea that dialogue amongst people of different religious commitments may be dialogue amongst people who share real differences in how they see the world, but differences where there is no fact of the matter which could even in principle be used as a basis for investigation and enquiry to aim for ultimate agreement. If I see a glass as half full, and you see it as half empty, our different ways of seeing it may be real and may significantly affect our behaviour, even though there is no way to say that one of us has a true belief and the other a false belief.

My suggestion here is that differences in religious commitments and other commitments may be like those differences in seeing things one way rather than another: genuine cognitive differences but not differences that reflect true or false beliefs. The point is important in the context of establishing dialogue amongst people with different religious or value commitments. When we are considering dialogue that involves exchanges amongst different people about the ways they see things, or frame things, we do not at present have methods of ethical assessment that allow us to evaluate people’s behaviour.22 If we are considering dialogue in the form of argument, we have some established criteria which rule out some processes as unacceptable. To some extent, we justify these criteria by reference to the fact that they affect the likelihood of the dialogue resulting in people accepting true positions or beliefs and rejecting false positions or beliefs.23 We are potentially in difficulty if we need to consider dialogue as exchanging points of view with differences that do not hinge especially on their truth or falsity. Take the example of a barrister setting out to convince a jury. We can accept without demur that it is wrong to lie or to conceal information. We can even accept that ‘distortion’ is wrong, if that includes revelation of some facts but withholding others. But is it ‘distortion’ to refer to something as ‘making a mistake’ rather than as ‘committing a crime’, when each description is a true one?24

In dialogue that partly involves people engaging with one another’s ways of seeing, we may have to evaluate dialogue processes in essentially ethical terms: not just by reference to their effects on attaining true or false beliefs. It may be that the criteria for evaluating participation in dialogue are not only outcome-oriented but also process-oriented: it may be that they do not revolve just around truth or falsity, but around efforts to genuinely apprehend and engage with others’ ways of seeing things.25


If this sketchy account has any merit, then one implication is that we need to examine more fully the nature of dialogue, the criteria we may use to evaluate it, and standards we may use to guide our own participation in it. I think that at present we are not clear about how to approach dialogue that to some extent involves interaction amongst people’s different ‘ways of seeing’ or ‘framing’. We are familiar with dialogue that revolves around people’s different preferences. We have well-established principles and standards to govern processes of bargaining and exchange, and quite well-developed ideas about fairness, even if we often fail to implement those principles and standards in contexts where different parties have varying advantages and power. We also have quite well-established methods for dialogue and discussion where differences hinge on truth and falsity. Principles of scientific enquiry and academic encounter can often be given their justification by reference to what processes are most likely to turn up true rather than false propositions. Even rules of courtroom procedure can often be given some justification in similar terms. However, we have less well-established approaches to forms of dialogue that are neither about preferences, to be dealt with in market processes, nor about distinguishing true from false propositions, but which are about commitments people have which involve different ways of seeing the world.

1 Davetian, B (2001) ‘Moral tensions between western and Islamic cultures: the need for additional sociological studies of dissonance in the wake of September 11’, Sociological Research Online, 6:3, U45–U53.

2 For discussion which is still pointed and relevant, despite being published over ten years ago, see Self, P, Government by the market? The politics of public choice, London: Macmillan, 1993.

3 For some forthright statements of such views see eg Ghannoushi, S (2005) ‘Whose side is th e US on?’, A l j az e era we b s it e , accessed 3 Sep temb er 2005,

4 See eg Bond, M (2004) ‘The making of a suicide bomber’, New Scientist, no 2442, 15 May.

5 Fisher, R and WL Ury, Getting to YES: negotiating agreement without giving in, Boston: Houghton Mifflin, 1981.

6 For some discussion see Provis, C (1996) ‘Interests vs positions: a critique of the distinction’, Negotiation Journal, 12:4, 305–323.

7 BBC News (2005) ‘Have Your Say – Tube bomber video: your reaction’, accessed 3 September 2005,

8 Mirko Bagaric and Julie Clarke, ‘Not enough (official) torture in the world? The circumstances in which torture is morally justifiable’ (2005) 39 University of San Francisco Law Review 1–39; summarised in Mirko Bagaric ‘A Case for Torture’, The Age, Melbourne, 17 May 2005.

9 Mill, JS Utilitarianism, liberty, representative government, London: Dent, 1968, pp 7–8. Originally published 1859–1863.

10 For discussion of the extent to which themes in Yes, Minister were explicitly intended to reflect the ideology of public choice theory, see Borins, SF (1988) ‘Public choice: “Yes Minister” made it popular, but does winning the Nobel Prize make it true?’ Canadian Public Administration, 31:1, 12–26.

11 For some relevant recent discussion of populism in modern Australian politics, see Uhr, J, Terms of trust, Sydney: University of NSW Press, 2005.

12 See eg Hughes, OE, Public management and administration: an introduction (3rd ed), London: Palgrave Macmillan, 2003, p 155. For some comments on the distinction between ‘technical’ and ‘political’ problems, see Nachi, M (2004) ‘The morality in/of compromise: some theoretical reflections’, Social Science Information, 43:2, 291–305.

13 Hirschman, AO, Exit, voice and loyalty: response to decline in firms, organizations and states, Cambridge, Mass: Harvard University Press, 1970.

14 Elster, J (1986) ‘The market and the forum: three varieties of political theory’ in J Elster and A Hylland (eds), Foundations of social choice theory, Cambridge: Cambridge University Press, pp 103–132.

15 A good deal of the relevant literature is commented on in Thompson, J, Discourse and knowledge: defence of a collectivist ethics, London: Routledge, 1998.

16 Very recently, Mark Latham’s diaries bear on pitfalls of such political processes: The Latham Diaries, Melbourne, Melbourne University Publishing, 2005.

17 See eg Monroe, K (2003) ‘How identity and perspective constrain moral choice’, International Political Science Review, 24:4, 405–425, and Harré, R and L van Langenhove (eds) (1999) Positioning theory, Oxford: Blackwell. The latter say ‘In this technical sense a position is a complex cluster of generic personal attributes, structured in various ways, which impinges on the possibilities of interpersonal, intergroup and even intrapersonal action through some assignment of such rights, duties and obligations to an individual as are sustained by the cluster’ (p 1).

18 Wittgenstein, L, Philosophical investigations (2nd ed) (trans GEM Anscombe), Oxford: Basil Blackwell, 1958, II, xi, pp 193–214.

19 See eg Neisser, U, Cognition and reality, San Francisco: WH Freeman and Co, 1976, pp 57–9.

20 Tversky, A and D Kahneman (1981) ‘The framing of decisions and the psychology of choice’, Science, 211, 453–458. For an exposition see eg Solso, RL, Cognitive psychology (4th ed), Boston: Allyn and Bacon, 1995, pp 428–9.

21 The notable exception seems to be John Hick: for some references and critical discussion, see Scott, M (1998) ‘Seeing aspects’, International Journal for Philosophy of Religion, 44:2, 93–108.

22 As an example of the ramifications of the point, we may note that John Rawls comments on conditions of his well-known hypothetical original position which are required if the putative reasoning people engage in there is to yield appropriate rules, saying in part that ‘such things as threats of force and coercion, deception and fraud must be excluded’ (Political liberalism, New York: Columbia University Press, 1993), but it is hard to say what constitutes deception and fraud when we are considering differences in ways of seeing.

23 In particular, for example, we may emphasise that fallacies like ad hominem or ad baculum tend not to lead to true conclusions, even though there are other grounds for criticising them in the ways they hurt others or fail to respect them: see eg Browne, MN and S Keeley, Asking the right questions: a guide to critical thinking (6th ed), Upper Saddle River, NJ: Prentice-Hall, 2001, p 97; Cederblom, J and DW Paulsen, Critical reasoning: understanding and criticizing arguments and theories (5th ed), Belmont, Calif: Wadsworth, 2001, pp 174–5 and Engel, SM, With good reason: an introduction to informal fallacies (2nd ed), New York: St. Martin’s Press, 1982, pp 166–9.

24 I have discussed these issues at greater length in Provis, C, Ethics and organisational politics, Cheltenham and Northampton: Edward Elgar, 2004, chs 8 and 9.

25 The point here is essentially similar to a point that can be made about the criteria for ethical evaluation of negotiation: see Provis, C (2004) ‘Negotiation, persuasion and argument’, Argumentation, 18:1, 95–112.