2. Meta-Ethical Foundations

Reflective realism is concerned, in the first instance, with certain aspects of our moral and political outlooks, or with certain questions about our moral and political outlooks taken as a whole. A moral and political outlook in this context can be thought of as something consisting partly of a stock of moral and political concepts together with dispositions to apply those concepts to particular cases and dispositions to respond in various ways to cases where those concepts are applied – where responses are to be broadly construed and involve desiring, emotional responses, and action. The possession of such moral and political concepts, and the dispositions to apply them, may depend on the possession of others’ concepts and dispositions to apply them which also partly constitute a broader theoretical outlook. One typically comes to acquire a moral and political outlook from the broader culture one is raised in. Possessing a moral and political outlook is essential to navigating one’s social world.

An importation feature of an outlook is the extent to which it is itself reflective: there are aspects of the outlook which concern itself and its relation to other things, in particular other outlooks, and its own history. It is a feature of most contemporary moral and political outlooks that they are, to a significant degree, reflective.Footnote ⁶ Thus, contemporary moral and political outlooks contain the very seeds of philosophical puzzlement that reflective realism is concerned to understand and, where possible, resolve. Two initial sources of puzzlement are these: first, that reflection reveals tensions and conflicts within our own moral and political outlooks and between our outlooks and those of others who share our social and political world, and second, that reflection reveals that our own moral and political outlooks are not completely transparent to us.Footnote ⁷ Reflective realism may help to resolve the former puzzlement by helping us to understand the sources of conflict in our outlooks and to better understand both the desirability and feasibility of resolving these conflicts. Reflective realism may help to resolve the latter puzzlement by making the contours of our outlooks clearer by tracing certain aspects of our outlooks to their sources. These sources of puzzlement are likely to arise with any degree of reflection on our moral and political outlooks. The distinctive responses to these sources of puzzlement offered by reflective realism are motivated by a third source of puzzlement.

This third and deeper source of puzzlement arises at a deeper level of reflection, a level of reflection that reveals that a certain kind of objectivity exhibited by other aspects of our outlooks – those aspects of our outlooks concerned with describing and explaining the world as it is in itself – is lacking in the case of our moral and political outlooks. Whereas these other aspects of our outlooks can plausibly be taken to answer to an objective reality, to the way things are in themselves, to the way things are independently of any point of view on them, our moral and political outlooks are revealed in reflection to lack this feature. If they answer to anything at all, they answer to something more local, more parochial, to some feature of our social worlds, and not to the world as it is in itself independently of any point of view on them. Moreover, while those other aspects of our outlooks can plausibly be taken to be independent of our more or less local or parochial points of view – they could be shared by rational enquirers occupying different points of view – our own moral and political outlooks are revealed in reflection to be tied up with points of view that partly constitute the social world they are about. Our ethical thought, and ethical reality itself, is thus revealed in reflection to be non-objective in contrast with the more scientific aspects of thought and the reality it is about. This is a deeper source of puzzlement, in part, because it is more unsettling. Unlike the existence of conflicts and the lack of clarity in our outlooks, which do not immediately call into question the idea that they may answer to ethical reality, the non-objectivity of our moral and political outlooks calls this idea into question and calls for an immediate response on our part if we are to continue to be committed to these aspects of our outlooks. Reflective realism itself is thus motivated by the recognition of the non-objectivity of our moral and political outlooks, and it seeks to offer a response, where possible, to the philosophical puzzlement that this gives rise to.

It may sound as though reflective realism is motivated by a contentious meta-ethical position concerning the non-objectivity of our moral and political outlooks. There is a sense in which this is so. But it is important to recognise how minimal the assumption of non-objectivity is and to recognise that, in some way or another, the non-objectivity of our moral and political outlooks is something that is already implicitly recognised in modern moral and political outlooks insofar as they already incorporate a significant degree of reflection. In this sense, reflective realism is a response to a kind of philosophical puzzlement that is already there. There are various ways of characterising this element that is already there in our moral and political outlooks. Williams characterises it in terms of ‘a problem that has recurred in European thought since historical self-consciousness struck deep roots in the early nineteenth century: a problem of reflection and commitment’ (Williams, Reference Williams2006, p. 192). It is related to Max Weber’s thought that the modern world has become entzaubert. We do not have the same confidence in our moral and political outlooks that traditional societies had, largely as a consequence of this disenchantment and the degree of reflection we incorporate into our outlooks. I am suggesting that, at its heart, the problem is one that arises when reflection reveals the non-objectivity of our outlooks. Of course, this is not to say that there have not been many attempts to provide objective foundations for our moral and political outlooks. But the felt need to find such foundations just reveals the extent to which reflection has unsettled our commitments.

We can think of reflective realism, then, as being motivated by a philosophical development of a kind of philosophical puzzlement that is already there in our moral and political outlooks. The problem of reflection and commitment arises when we notice that our moral and political outlooks cannot be explained in directly vindicatory terms, that is, in terms of answering to a reality that is independent of our local and parochial points of view on that reality. Unlike in the case of change in our outlooks brought on by scientific discovery, where we can see, on reflection, how our outlook can explain the appearances of our previous outlook, and reveal more of the reality that is already there, in the case where we reflect historically on our moral and political outlooks, we do not see them as explaining the appearances, or revealing more of what was already there. Although we can present arguments within our new outlook against the old outlook, the arguments are not part of the story of the transition. Rather, the story of the transition is the story of how we came to have an outlook within which those arguments are compelling (Williams, Reference Williams2006, p. 190). It is partly the story of how the social and ethical reality that our moral and political outlooks are concerned with came to be our social and ethical reality.

Reflection may reveal the non-objectivity of our moral and political outlooks and thereby call into question their authority. As Williams writes:

This is what threatens our commitment to our own moral and political outlooks and may undermine our confidence in them. Reflection itself may have consequences for our dispositions to go on applying the moral and political concepts we have (dispositions which characterise our confidence in the concepts) and thus bring about a change in our outlooks themselves.Footnote ⁹

Since this is the most important motivation for reflective realism, let me expand on these points. Consider what happens when we take up a reflective point of view on our own outlook in relation, say, to some past outlook. We know that our outlook may differ in radically different ways to other outlooks. It will likely involve concepts that the other outlooks lack and vice versa. Now, we may raise a kind of explanatory question about our outlook, namely, how we came to have that outlook.Footnote ¹⁰ And the answer to this question may say something about this outlook in relation to the earlier outlook. Now, for many aspects of our own outlook, those aspects related, perhaps, to a modern scientific understanding of the world, we can explain our coming to have these features of our outlook, in relation to some earlier outlook, in terms of our outlook allowing us to correct some aspects of that earlier outlook as being mistaken in taking something as a feature of reality that is not a feature of reality – as including concepts for things which are not features of reality and using these concepts as if they described features of reality – or as being incomplete in the sense of not capturing some feature of reality – as lacking concepts for things that are features of reality. Examples which come to mind are the concepts of phlogiston and élan vital. This kind of explanation is directly vindicatory, since the latter outlook is directly vindicated with respect to the former. There are some important presuppositions of these explanations, the most important being that the outlooks are in the same game, so to speak, such that it makes sense to speak of the former being mistaken or incomplete in relation to something that the latter outlook overcomes.

What is the contrast meant to be when it comes to moral and political concepts? Well, the basic point here is that, for the most part, vindicatory explanations will not be available. To find a vindicatory explanation of our own outlook in relation to some past outlook would be to find some aspect of that earlier outlook that was mistaken in the sense that it included concepts for things which, from the point of view of our new outlook, are revealed not to be features of reality, or to find aspects of the earlier outlook that failed to capture features of reality that our outlook can. Now, the task should not be confused with the much easier task of findings points of difference, additional concepts in our outlook, things that look mistaken in terms of our outlook (they thought slavery was right, we think it is wrong). What’s missing in the moral and political case is any sense that there is a single thing, a single ethical reality, that both outlooks are outlooks on such that the latter can be said to correct mistakes in the former with respect to that reality and/or capture more of that reality relative to the former. Notice that this is not to say that the outlooks do not each aim to be outlooks on a single thing (it is not to say that they are relativised to ‘us’). It is just that the single thing that they each aim to be outlooks on can only be thought of, it seems, within one or the other outlook, is only ‘there’, so to speak, from the point of view of one or the other outlook. This undermines the possibility of a directly vindicatory explanation. As A. W. Moore puts it, the basic idea here is ‘that the best reflective explanation of our having the ethical knowledge we have, unlike the best reflective explanation of our having the scientific knowledge we have, cannot directly vindicate that knowledge: it cannot directly reveal us as having got anything right’ (Moore, Reference Moore and Thomas2007, p. 25).Footnote ¹¹ A direct vindication would involve giving the kind of explanation above in terms of improving on the earlier outlook.

If the recognition of the non-objectivity of our moral and political outlooks is going to have consequences for our commitment to those outlooks, it is not simply going to be a matter of our recognising such non-objectivity. As Williams writes: ‘The relevance of cultural, psychological, or economic explanation to ethical values does not lie simply in its providing a challenge to them all collectively’ (Williams, Reference Williams2014a, p. 383). Reflective recognition of non-objectivity provides a general challenge to our moral and political outlooks, but it is another question whether that challenge can be met for certain aspects of our outlook. As Williams writes:

But he goes on to note that to react in such a way as to lose commitment to one’s outlook, to treat it ironically, is to make a mistake about what is required for commitment:

Why should we think that we could be entitled to our own moral and political outlook only if they were as objective as other aspects of our outlooks? There may be no global and direct vindication of our outlooks available. But this doesn’t mean that there cannot be more local vindications, or explanations of our having the outlook that we do, which would allow us to endorse it. Consider, for instance, what Williams says about the virtues associated with truthfulness. According to Williams the forms that such values take are going to be ‘culturally and historically various’ (Williams, Reference Williams2006, p. 192). Because of the non-objectivity of our outlooks, we are not going to be able to provide a vindication of these aspects of our outlooks that vindicates them over outlooks that did not contain exactly the values that our outlooks contain. Still, Williams hoped that we could provide a local vindication of some of the virtues associated with truthfulness in terms of the ‘universal human need of qualities such as accuracy (the disposition to acquire true beliefs) and sincerity (the disposition to say, if anything, what one believes to be true)’ (Williams, Reference Williams2006, p. 192). Such a local vindication would allow us to make sense of our dispositions to treat such virtues as being intrinsically valuable without thereby claiming that such concepts answer to an objective reality that is independent of our local point of view. Notably, Williams also thought that some virtues associated with truthfulness, such as a particular conception of authenticity, cannot be made sense of in such terms (Williams, Reference Williams2002, Ch. 8). The concepts associated with such virtues might be seen to fail to answer to anything at all, not even some aspect of our social world. This points to a positive project for reflective realism, namely that of finding which aspects of our moral and political outlooks can be indirectly vindicated through reflection and which aspects cannot. I will have more to say about how this might work below.

Before doing so I want to reflect on some implications of what I have said so far for political philosophy. The first implication is that the observations motivating reflective realism make engaging in first-order political philosophy as moral theory, while ignoring the issues raised by reflection, a less than fully truthful and authentic enterprise. Unlike other forms of realism, however, reflective realism can find a place for first-order theorising, as long as it is done in light of reflection. In his discussion of these issues in ‘Philosophy as a Humanistic Discipline’, Williams characterises ‘realism in political philosophy’ in terms of the connection between ‘the philosophical activity of reflecting on [our outlooks] at a more general level and trying to make better sense of them’ and ‘the historical activity of understanding where they came from’ (Williams, Reference Williams2006, p. 194). These are contrasted with ‘the first-order activity of acting and arguing within the framework of our ideas’ (Williams, Reference Williams2006, p. 194). It is clear that, for Williams, most moralistic political philosophy involves ‘arguing within the framework of our ideas’, that is, arguing within our outlooks, or making sense of our outlook from the inside. This may count, when some degree of reflection is involved, as an instance of the second kind of activity. But Williams’s point is that there’s more to the activity of reflecting on and trying to understand or make sense of our own outlook than this. It is this that ‘helps define … realism in political theory’. And it is this which connects realism in political theory with Williams’s view of philosophy as a humanistic discipline which must draw connections with historical and sociological understanding.Footnote ¹²

3. Methodological Implications

Much of the promise, then, of reflective realism as an alternative to political philosophy as moral theory comes from the ways in which it may resolve the philosophical puzzlement that arises from the recognition of the absence of directly vindicatory explanations of our moral and political outlooks. What would a political philosophy look like that took seriously the idea that there might be local or indirect vindicatory explanations of some aspects of our moral and political outlooks? What would a political philosophy look like that took seriously the idea that reflection may reveal that some aspects of our outlooks cannot be given even a local or indirect vindicatory explanation or may even be given an explanation that leads us to abandon that aspect of our outlook? While reflective realism may help us to better understand the tensions and conflicts in our outlooks, that is, may help us to better understand our outlooks themselves, its main promise lies in how it may transform aspects of our outlooks through reflection. My aim in this section is to explore the role of reflective understanding in critical reflection and reflective endorsement.

As we saw in the previous section, one significant motivation for reflective realism comes from the absence of a certain kind of explanation of our having the moral and political outlooks we have. It comes from the absence of directly vindicatory explanations of our outlooks. Although this kind of explanation is not available there still remain important distinctions between the different kinds of explanations of different aspects of our outlooks that are available. A helpful, but potentially misleading, analogy here comes from the case of secondary qualities like colour. Reflection reveals that, in some sense, things are not as they appear to be when it comes to our colour concepts. Nonetheless, we can give a kind of vindicatory explanation of our disposition to apply colour concepts to physical objects. We can see how this aspect of our outlook answers to something – it tracks certain features of our physical environment which are there independently of us. Whatever we want to say about the metaphysics of colour and the truth and falsity of our claims about colour, recognising this fact about this aspect of our outlook does not undermine our commitment to it. This analogy is potentially misleading when it comes to aspects of our moral and political outlooks because there is unlikely to be anything that stands to our moral and political outlooks as certain surface reflectance properties stand to our colour concepts. If our moral and political outlooks answer to anything, it is going to be something more local than this. Nonetheless, there seems to be an intelligible distinction here between our moral and political outlooks answering to some aspect of our social world and its not doing so. This is something potentially revealed by reflection. We can ask, in reflection, whether the best explanation of our moral and political outlook having some feature is the feature answers to something in our social world. If it does, then, just as in the case of colour concepts, we may retain our commitment to it. This is what we might call reflective endorsement or vindication. If it does not, then, we may no longer be able to maintain our commitment to it. Reflective understanding, as we saw, is likely to have significant consequences for our commitment to certain aspects of our moral and political outlooks. As Williams writes: ‘If one comes to know that the sole reason one accepts some moral claim is that somebody’s power has brought it about that one accepts it, when, further, it is in their interest that one should accept it, one will have no reason to go on accepting it’ (Williams, Reference Williams2002, p. 231). Elsewhere, Williams mentions the kind of critical and debunking genealogies of Nietzsche (Williams, Reference Williams2002, Ch. 2). This is a common consequence of what we might call reflective criticism. I will now discuss each of these potential upshots of reflection, starting with the latter.

The basic idea here is that when we reflect on our own moral and political outlooks, we ask, from a reflective point of view, whether the best explanation of our having the moral and political outlook that we do is that it answers to something in our social world or answers to nothing. We already know that our moral and political outlook is not going to answer to objective reality, a reality that is there anyway independently of our local or parochial points of view on it. This much is recognised when we recognise that our outlooks are non-objective. But this does not mean that they answer to nothing. They may answer to something more local and parochial, something more intimately tied up with our local practices and parochial points of view. Sometimes the best explanation will be that some aspect of our moral and political outlook answers to nothing, that is, the best explanation is an explanation according to which our moral and political outlook does not even answer to something more local and parochial. Recognising this may lead us to give up this aspect of our moral and political outlook.

Notice that in attempting to answer the question of whether some aspect of our moral and political outlook answers to something rather than nothing we are under no pressure to try to understand what it answers to from a maximally objective point of view. As Williams writes:

In other words, we are not attempting to vindicate our outlooks from nowhere. The thought is, however, that in reflection we may be able to draw on features of our outlook that are at least somewhat less local and parochial than the aspect of our outlook that is under consideration. This may be enough to allow us to draw the relevant conclusion as to whether some aspect of our outlook answers to something or to nothing. Williams himself often appeals to fundamental or universal human concerns and interests in attempting to offer a local vindication of some aspect of the modern liberal outlook. What this reveals is that reflective realism is not committed to finding some point of view entirely outside of our moral and political outlooks from which to judge them.Footnote ¹³ It is not committed to drawing only on epistemic evaluation. However, the role that these considerations play in reflection distinguish the approach from more moralistic alternatives. The role played by an appeal to these more objective but nonetheless not fully objective aspects of our outlooks, is that of determining whether some aspect of our moral and political outlook answers to something in our social world or answers to nothing. The question, in the first instance, is one about the relation between our ethical thought and the reality it purports to be about – do our thoughts answer to that reality? We may appeal to more objective aspects of our moral and political outlooks, and our broader theoretical outlooks, to answer this question. In doing so, we are not claiming that aspects of our outlooks answer to these things, for then we would be claiming that they after all answer to objective reality. Rather, we are appealing to these things in order to answer the question of whether these aspects answer to some aspect of social reality or to nothing at all. The appeal to fundamental human interests and concerns may allow us to make sense of the idea that some aspect of our moral and political outlook answers to some feature of our social reality as opposed to answering to nothing. This is the sense in which the question is one about the relation between our ethical thought and the reality it is about. Importantly, the question is not one about whether some aspect of our moral and political outlooks promotes, serves, respects or obscures some independently recognised moral or political value – a point I will expand on below.Footnote ¹⁴

Williams develops these ideas in a particular way in relation to political concepts of power and coercion and this particular development is what has received the most attention from critics. In Truth and Truthfulness Williams makes use of what he calls the ‘Critical Theory Principle’ and a similar principle is appealed to in his influential work on legitimacy and the basic legitimation demand. This is Williams making good on the idea that political philosophy might connect better with the ‘concerns which have animated neo-Marxist and neo-Hegelian critics of our society, concerns which have, so far, been expressed in more archaic philosophical forms’ (Williams, Reference Williams2014a, p. 124). The critical theory principle is developed, at least initially, in relation to critical reflection on the moral and political outlooks of others. But it also obviously has implications for critical reflection on our own moral and political outlooks.

Contemporary realists have picked up on this aspect of Williams’s position and have discussed it in relation to ideology critique (Prinz and Rossi, Reference Prinz and Rossi2017; Rossi, Reference Rossi2019). It is important to emphasise at this point in the discussion, however, that ideology critique can be understood as being both broader and narrower than the kind of critical reflection central to reflective realism. Ideology critique is typically concerned with the relation between power and authority of various kinds and the moral and political outlooks of those over whom that power and authority is exercised. As we will see in a moment, reflecting on the role of power and authority in relation to our moral and political outlooks is of central importance to reflective realism. However, reflective realism can be motivated by a broader set of concerns. There may be other factors at work in explaining how we came to have the moral and political outlooks we have, and these may raise puzzles as much as concerns about power do. At the same time, ideology critique is often motivated by a more general kind of scepticism about morality, and a general concern about power.Footnote ¹⁵ This broader motivation for ideology critique makes the prospects of a local vindication of aspects of our moral and political outlook look unlikely from the outset.

Let us consider in more detail how Williams thought the Critical Theory Principle is meant to work. As sympathetic critics of Williams have pointed out, Williams seems to be concerned in his discussion of the Critical Theory Principle with the moral ideal of freedom or autonomy, what he calls the ‘most basic sense of freedom, that of not being in the power of another’ (Williams, Reference Williams2002, p. 231). Williams clearly thinks, in both Truth and Truthfulness and ‘Moralism and Realism in Political Theory’ that there is an important normative or evaluative distinction between being under the power of another when being so accords with one’s moral and political outlook and when it does not. He thinks that conditions in which it accords with one’s outlook only because one’s outlook is itself a product of coercion by that power, are morally on a par with conditions where it does not accord with one’s outlook at all. Both are conditions of unmediated coercion. In Truth and Truthfulness Williams speaks of unmediated coercion as being unjust (Williams Reference Williams2002, p. 231). In ‘Moralism and Realism’ Williams speaks of illegitimacy due to the basic legitimation demand not being met, that is, when a legitimation of rule is simply not accepted or acceptable, or it is accepted only because the outlook according to which it is accepted is a product of coercion and power. For Williams, then, the Critical Theory Principle, and reflection on our moral and political outlooks in general, plays an important role in distinguishing cases of unmediated coercion from cases where political power is appropriately mediated by the moral and political outlooks of those over whom it is exercised.

When Williams further specifies how to apply the Critical Theory Principle, he writes:

I will suggest a way of understanding what Williams is saying here in a moment. But first I want to head off a potential misunderstanding. Rossi and Prinz, in discussing this claim, accuse Williams of importing a moral claim here: ‘That test is hypothetical, so to see whether it succeeds we need to unpack its conditional: “If they were to understand that they came to hold this belief as a result of a violation of their basic freedom, then they would have reason to give it up”, or something of that sort’ (Prinz and Rossi, Reference Prinz and Rossi2017, p. 357). There is no reason to think this is how Williams understood the critical reflection test, however. He did not take it to be ‘a heuristic to discover this sort of freestanding moral flaw’ (Prinz and Rossi, Reference Prinz and Rossi2017, p. 357). Rather, what the tests ask is whether the legitimation of the exercise of power and authority over them would make sense to them if they were to understand how they came to hold the legitimation. This is how the test is applied. What the test reveals, if the legitimation story does not pass the test, is that the exercise of power and authority is nothing more than an instance of unmediated coercion. This latter thought needn’t be available to those to whom the test is applied. But it may be available to the theorist applying the test.

Even if Williams did not think of the critical reflection test as being a ‘heuristic to discover a sort of freestanding moral flaw’, there is a question about what role Williams took political and moral claims to play in critical reflection. He certainly didn’t think of such claims as playing the role that they are taken to play by proponents of the ‘new ideology critique’ (Aytac and Rossi, Reference Aytac and Rossi2023, pp. 1216–1218). He did not think of reflection as involving the application of moral values to reveal the ways certain aspects of our outlooks undermine or distort those values. Nor does it seem that he thought that critical reflection revealed a purely epistemic flaw in one’s outlook. It may do. Critical reflection may reveal that one has no reason to hold some belief. And this may rationally lead to its abandonment. But I think that Williams thought that more often what reflection would reveal is that some aspect of our moral and political outlook fails to answer to something that we can recognise from a more general level. Recall the connection Williams draws between ‘the philosophical activity of reflecting on [our outlooks] at a more general level and trying to make better sense of them’ and ‘the historical activity of understanding where they came from’ (Williams, Reference Williams2006, pp. 194). This suggests that the conclusion of reflection is either that some aspect of our outlook makes sense to us or that it does not make sense to us. Whether it makes sense to us or not is a matter of whether we can make sense of it from ‘a more general level’. Among the things we can recognise at the more general level are fundamental human desires and needs, including the need to live together, things that are pre-moral or pre-political.

Although Williams says some specific things about considerations that might bear on the application of the principle, it is clear that, at a very general level, the issue here is one of whether the best explanation of some aspect of one’s moral and political outlook is one that reveals it to answer to something about one’s social world, or to answer to nothing. In cases where it is revealed that the best explanation of one’s having some outlook is that it is a mere product of political power it will have been revealed to answer to nothing. In the context of the critical theory principle, what we are interested in is a special case of debunking explanation. Not only do we conclude that this or that aspect of our outlook does not answer to anything. We conclude, moreover, that the best explanation of our having it is that our having it merely serves the interests of the advantaged. Upon recognising this we will no longer be able to truthfully endorse this aspect of our outlook.

Edward Hall, in a very sympathetic discussion of the critical theory principle, has argued that the principle, as developed by Williams, will either be overinclusive or underinclusive (Hall, Reference Hall2020, p. 161). However, Hall’s criticism depends on too narrow or literal an understanding of Williams’s position. Hall attributes to Williams the view that it would be sufficient to fail the critical theory principle if it were recognised that the advantaged in society would have an interest in perpetuating some legitimation story which advantaged them even if it were not true (Hall, Reference Hall2020, pp. 161–62). But this obviously shouldn’t be a sufficient condition if we are understanding the critical theory principle in terms of an inference to the best explanation of one’s commitment to a particular feature of one’s moral and political outlook. It is not obvious that it would be sufficient to conclude that one’s moral and political outlook answered to nothing just because one recognises that one is committed to that aspect of one’s outlook only because of the teaching of the advantaged and recognises that they would have an interest in so teaching even if it were not true. Hall notes that there are good interpretive reasons for not attributing this understanding to Williams in any case (Hall, Reference Hall2020, p. 162). We have just seen that there are good theoretical reasons too.

The other side of Hall’s dilemma is this: that the critical theory principle is underinclusive. He suggests that on an alternative understanding of the principle the reflective question is just one of whether the claims of the advantaged are to be taken as true or authoritative. But the worry about this is that it will only capture the easy cases involving ‘legitimation stories that rest on clear falsehoods, deceptions, and myths’ (Hall, Reference Hall2020, p. 164). The idea here is that reflection involves imagining what would happen if those with certain moral and political outlooks had those outlooks corrected in the direction of recognising certain ‘plain truths’ about their situation. This may deal with the easy cases, but it leaves us unable to say anything about cases where we suspect that an outlook might fail the critical theory test, but is not maintained by clear falsehoods, deceptions, or myths (Hall, Reference Hall2020, pp. 164–65). Hall provides aspects of neoliberal ideology as a candidate (Hall, Reference Hall2020, p. 164). But, again, it isn’t obvious that this is a consequence of the critical theory principle if we understand it in terms of an inference to the best explanation of one’s commitment to a particular feature of one’s moral and political outlook. It may well be that if those who are disadvantaged by neoliberal ideology but nonetheless accept aspects of it were to reflect on this feature of their outlook and ask whether the best explanation of their being so committed might not be one that entails that their outlook answers to nothing and is merely an illusion. It is notoriously difficult to say anything systematic about which considerations might bear on an inference to the best explanation in any particular case.Footnote ¹⁷ We can understand Williams as pointing here not to considerations that provide necessary or sufficient conditions for a debunking explanation to be the best explanation. Rather, one can understand him as pointing to the kind of considerations that often do bear on such an inference.

While critical reflection on the moral and political outlooks of others does not have consequences for whether they continue to endorse aspects of their outlook, reflection on our own outlook is likely to have significant consequences. We may find that the explanation of our coming to have some outlook does not support our endorsing it because we cannot see it as truthfully answering to anything in our social world. Reflection may reveal that some aspect of our outlook is illusory. And not merely in the sense that our colour concepts are revealed to be illusory. This much is already granted by non-objectivism. Rather, reflection may reveal that some moral or political concept, unlike our colour concepts which at least answer to something without that thing being exactly as it seems to be, answers to nothing. Insofar as we are concerned with truthfulness, with having our outlooks answer truthfully to something beyond them, we will be disposed to abandon this aspect of our outlook.

Now, reflective understanding need not have negative consequences for our moral and political outlooks. In some cases, some aspects of our understanding will survive reflective scrutiny. Williams has explored the way that reflective understanding can vindicate aspects of our moral outlook, leading us to endorse those aspects of our outlook, perhaps with slight modification.Footnote ¹⁸ This is ultimately what happens, according to Williams, when we reflect on the values of truth and truthfulness. We can reflectively endorse these values, contrary the position of ‘deniers’ like Rorty, who would have us take an ironic attitude to all of our commitments, even our commitments to truth and truthfulness themselves (Williams, Reference Williams2002). Being able to reflectively endorse certain aspects of our moral and political outlook is essential to avoiding the kind of alienation that may be induced from the reflective standpoint, the kind of alienation embraced by thinkers like Rorty. We have already seen why this position is not unavoidable. Reflective understanding can also lead us to resolve tensions in ways that we may not have been able to by thinking within our outlook, it may also make alternatives outlooks available to us that otherwise wouldn’t have been.

4. Realist Themes

In this section, I explore two realist themes from the recent literature and situate reflective realism in relation to them. Doing so will help to clarify certain aspects of reflective realism and to distance reflective realism from the more problematic aspects of other forms of realism.

Perhaps the central realist theme is the ‘rejection of the priority of the moral over the political’ theme (Horton, Reference Horton2017, p. 493). This theme is announced early in Williams’s ‘Realism and Moralism in Political Theory’. He characterises the enactment and structural models of political theory as representing ‘the priority of the moral over the political’ (Williams, Reference Williams2005, p. 2). He says that on both models ‘political theory is something like applied morality’ (Williams, Reference Williams2005, p. 2). And he defines moralist theories as theories which ‘make the moral prior to the political’ (Williams, Reference Williams2005, p. 3). Williams introduces a sub-theme of this theme when he characterises his alternative, political realism, as ‘an approach which gives greater autonomy to distinctively political thought’ (Williams, Reference Williams2005, p. 3). It is clear enough what this theme is, at least in broad outline, and why it should be attractive to anyone who thinks that there could be more to political philosophy than what one finds in the tradition of political philosophy as moral theory and applied moral theory. There is a question, however, about what Williams meant by speaking of ‘an approach which gives greater autonomy to distinctively political thought’ (Williams, Reference Williams2005, p. 3). Reflective realism provides an interpretation of what this means in terms of the kind of historical, social and political thought that goes into achieving a reflective understanding of our moral and political outlooks. If this theme supports a distinctive realist thesis, from the point of view of reflective realism, it is this: that aiming at a reflective understanding of our own moral and political outlooks in relation to our political circumstances is a worthy philosophical enterprise, and one that may have philosophical advantages over the traditional approach to political philosophy as moral theory.

My view is that Williams may have meant nothing more and nothing less than this in introducing the distinction between these different ‘approaches to political theory’, although, as I noted earlier, it is not my aim here to offer an interpretation of Williams. Reflective realism is an approach which ‘gives greater autonomy to distinctively political thought’ (Williams, Reference Williams2005, p. 3). It gives greater autonomy to such thought because of the need to reflect on political power in coming to a reflective understanding of our own moral and political outlooks. It also gives special attention to ‘distinctively political concepts, such as power, and its normative relative, legitimation’ (Williams, Reference Williams2005, p. 77). It gives special attention to these concepts as a consequence of aiming at reflective understanding and not because the view is characterised in terms of the priority of legitimacy over other political concepts, or in terms of the priority of the political value of order and stability over other political values, or in terms of a distinctive political normativity.

The other realist theme I want to discuss here is that realism has a distinctive approach to political values. Several realists have hitched the realist wagon to this theme. What is distinctive about realism, they claim, is the approach it takes to political value. On one characterisation, the distinctive realist claim here is that political arguments must begin ‘from where a given political community is’ (Galston, Reference Galston2010, p. 396). Rather than attempting to articulate moral and political ideals that may have no resonance whatsoever with the population they are meant to be ideals for, the realist starts with moral and political ideals that so resonate. There are various ways of spelling out such an approach. I will focus on the approach developed in Hall (Reference Hall2017) here.Footnote ¹⁹

Hall attempts to develop an approach to political value which he finds implicit in Williams’s work. The basic thought is that ‘[p]olitical values such as liberty and justice have a thin universal element as they relate to universally shared human experiences’ (Hall, Reference Hall2017, p. 286). In the case of liberty, Williams thinks that there is a pre-political concept of freedom that gets developed into a different political concept in different historical circumstances. Williams has sketched how the political value can be constructed out of the pre-political concept. Hall is keen to emphasise the idea that ‘political philosophers must not attempt to define, but to construct, a political conception of liberty from the non-political conception of freedom’ (Hall, Reference Hall2017, p. 287). Such an attempt at construction may indeed provide a better reflective understanding of the political value of liberty than some attempt at definition or analysis. This gives rise to what Hall calls the ‘realism constraint’. According to Hall, the constraint holds that ‘we cannot clarify the nature of various political values in any meaningful manner before we consider the historical and political question of what their elaboration requires “now and around here”’ (Hall, Reference Hall2017, p. 288). This seems plausible enough if we are taking the constraint to be a constraint on reflective understanding. Perhaps it is true that we cannot clarify the nature of our moral and political outlooks without attempting such a construction. But Hall clearly takes the constraint to imply more than this, since he takes it to conflict with a commitment to ideal theory in political philosophy (Hall, Reference Hall2017, p. 289). The conflict is this: that most political philosophers are happy to work within our moral and political outlook and to develop an ‘ideal theory’ of that outlook. One could understand the realism constraint as claiming that only values constructed according to the realism constraint have a role to play in our political theorising. I do not think that Hall wants to make such a strong claim. In the end, it seems, he only wants to claim that one approach to political value, a viable alternative to the approach taken in traditional political philosophy, is one that goes via such a construction. As he writes:

This is something with which reflective realism can wholeheartedly agree. But then Hall goes on to say things like this, ‘Realist political theory requires an imaginative grasp on what might actually make sense of people here and now’ and ‘political values must be constructed in such a way that they have purchase on the unique historical and political situation in which we actually find ourselves’ (Hall, Reference Hall2017, p. 296). What is the force of the ‘must’ in this claim? Is it that if we are going to have an indirectly vindicatory explanation of some political value, then it must be constructed in such a way that it has a purchase on the unique historical and political situation we actually find ourselves in? If there is a uniquely ‘reflective’ realist approach to political value, then, it is just that realism is concerned with questions about what the appropriate attitude to take towards our own moral and political outlooks is, and that an attitude of endorsement may only be justified if we can provide a vindicatory explanation, perhaps in the form of a construction of the value from something more primitive and pre-political given the social and historical context.Footnote ²⁰ Realism would differ from moralism in this respect, as the moralist may take no interest in critical reflection. A secondary consequence, and it seems to me that this is what Hall wants to draw attention to, is that the political values that do survive reflection will be more realistic in the sense of being true to the facts. Those aspects of our liberal outlook that depend on some kind of transcendental assumption of autonomy may be hard to reflectively vindicate. The aspects of our liberal outlook that can be reflectively endorsed may, in the absence of support from such an assumption, be more realistic.