Abstract
Political realists claim that politics should be regulated by a distinctive political normativity, one that does not rely on external, pre-political moral standards. It is in this sense that they distinguish political realism from ‘political moralism’, regarded as an approach that understands political theory as applied ethics. Importantly, realists’ anti-moralism is not motivated by the conviction that moral considerations do not play any role in the political realm. Rather, the target is the externalism of the normative resources on which moralist theories typically ground their conceptions of legitimacy. In contrast to moralists, some realists have argued for the need to elaborate internal theories of legitimacy, ones that develop normativity internally – that is, from within the political context under evaluation. This commitment entails the effort to reconnect legitimacy to the beliefs and attitudes of people subjected to the political power. In these accounts of legitimacy, critique is typically exercised internally, by means of self-reflection and ideology critique. Contra some realists, I argue that some forms of externalism are desirable and compatible with internalist accounts of legitimacy, and I show how to accommodate them within internalist theories of legitimacy. They are desirable because they strengthen the emancipatory potential of internal reflection and, in some cases, even stimulate it, as when, because of political circumstances, internal critique cannot be exercised. In addition, they prevent internalist approaches from falling into descriptive accounts of de facto acceptance of power. These forms of external critique can be compatible with internalist accounts of legitimacy, provided that they do not contrast with what I call methodological and substantive internalism. Accordingly, in the article, I discuss which stances of external critique can and should be accommodated within internalist accounts of legitimacy, without renouncing to the realist character of the approach.
Keywords
1. Introduction
Political realists claim that politics should be regulated by a distinctive normativity – that is, by standards that derive from a proper understanding of the political practice rather than being grounded on external, universal moral values (Galston 2010; Rossi and Sleat 2014; Sleat 2013; Williams 2005). 1 It is in this sense that they distinguish political realism from ‘political moralism’ (Williams 2005), regarded as an approach that understands political theory as applied ethics and derives its normative standards from pre-political moral considerations.
Realists’ anti-moralism is motivated by the conviction that politics is characterised by ineliminable constitutive features, such as conflict and disagreement about, among other things, world views and moral commitments, which makes politics a distinctive sphere of the human experience (Galston 2010; Stears 2007; Williams 2005). The distinctiveness of its nature, realists continue, calls for the distinctiveness of its normativity – that is, for the search for political standards that are anchored in the constitutive features of politics. For this reason, political realists refuse to ground political theory on pre-political moral assumptions (Rossi and Sleat 2014); or, in other words, they give politics priority over morality (Sangiovanni 2008).
The rejection of moralism is not motivated by the conviction that moral considerations do not play any role in the political realm. 2 Rather, the target of realists’ criticism is the externalism of the normative resources on which moralists typically ground their conceptions of legitimacy – that is, the conviction that, in order to assess the legitimacy of the political authority, we must look for standards (moral ones) that are deemed to have antecedent authority over political considerations and should therefore be applied as evaluative tools to political regimes (Scheuerman 2018; Sleat 2013). In contrast, a distinctive political normativity develops from within the political realm and proceeds through an understanding of the political context under assessment (Floyd 2011).
Some realists, therefore, argue for the need to elaborate theories of legitimacy that develop normativity internally – that is, from within the political context under evaluation. Moreover, realists argue, the notion of legitimacy should be reconnected to the beliefs and attitudes of people subjected to the political power (Horton 2010a, 2012), although remaining a normative notion able to criticise the status quo.
In this regard, I show, realist theories of legitimacy are internal in a double sense. First, they derive their normativity from within the political realm, rather than giving in to the alleged antecedent authority of the moral sphere. Call this methodological internalism. Second, they are internal in the sense of elaborating normative standards of legitimacy that take the beliefs and attitudes of the subjects to political power in each and every specific political context as their starting point. Call this substantive internalism. By denying any difference between different types of internalism, they have consequently rejected any type of externalism.
In contrast to some realists, I argue that there is a place for external critique within a realist theory of legitimacy. Specifically, some forms of externalism are not only compatible with realist theories of legitimacy, as long as they do not undermine internalism, but also fruitful where they enrich the normative import of the approach. Indeed, I argue that while methodological internalism must be preserved in any theory that claims to be a realist one, substantive internalism can be complemented with some forms of external critique, provided that certain conditions are met. 3
To develop my argument, I proceed as follows. In section 2, I offer a summary of realist legitimacy and illustrate why realists emphasise the need for internally elaborated standards rather than employing externalist approaches to legitimacy. In section 3, I argue that some forms of externalism might be desirable for realist theories of legitimacy: first, because they strengthen the chances to enhance political progress by offering new political scenarios as political possibilities; second, because they reinforce the normative impact of the theory. In section 4, I specify what holding a belief in legitimacy amounts to and lay out two prominent cases of disagreement between internal and external perspectives. Finally, in section 5, I argue that some forms of externalism can be accommodated into a realist notion of legitimacy; that is, I specify which stances of external critique are admissible within an internalist account of legitimacy.
2. Realist political legitimacy and the two types of internalism
In his posthumous essay ‘Realism and Moralism in Political Theory’, Bernard Williams sketches a novel approach to political legitimacy. He defines a political realist approach, in contrast with ‘political moralism’ – that is, the idea that political theory is applied ethics (Williams 2005, chap. 1). It is, in sum, a commitment to anti-moralism (i.e. the denial that morality has antecedent authority over politics) and, consequently, to internalism (i.e. the conviction that standards for legitimacy should stem from within the political realm). 4
Briefly, this is Williams’s argumentation. 5 He starts from the ‘“first” political question in Hobbesian terms’, namely the question of ‘securing of order, protection, safety, trust and the conditions of cooperation’ (Williams 2005, 3). It is first because without a solution to it, no other questions can be posed (Williams 2005, 3). Legitimacy comes into play because, indeed, not all the solutions to the first political question are acceptable: meeting the Basic Legitimation Demand (BLD), Williams argues, is what discriminates between legitimate (LEG) and illegitimate states (Williams 2005, 4). The concept of the BLD sets the requirements for a state to be legitimate. First, ‘the state has to offer a justification of its power to each subject’ (Williams 2005, 4). However, as the second requirement, the acceptance of such a justification cannot be produced by the same power that needs to be justified (Critical Theory Principle (CTP)); in other words, manipulated beliefs do not count as markers of legitimacy. 6 Third, the regime has to make sense as an intelligible authoritative order, given the historical and cultural circumstances (Williams 2005, 10–11). Accordingly, although ‘the legitimations appropriate to a modern state’ are tied to liberal values (Williams 2005, 9), ‘there manifestly have been, and perhaps are, LEG non-liberal states’ (Williams 2005, 4).
By introducing the making sense category (MS), Williams adds a further aspect to the internalism of his realist approach. By recalling Max Weber’s thought, he highlights the connection between legitimacy and subjects’ beliefs, although in different terms from Weber. In a nutshell, Williams refuses to collapse legitimacy onto de facto acceptance of power, so that not just any belief in the legitimacy of power is a marker of the regime’s legitimacy. 7
This aspect has been taken up by some scholars in the contemporary realist literature. For example, John Horton argues that legitimacy should be reconnected to the attitudes and beliefs of people subjected to political power (Horton 2010a, 2012). Similarly, Philp remarks that taking the context into account means to consider (among other things) the ‘aspirations, capabilities and dispositions of actors’ (Philp 2010, 468) and Geuss argues that we should start ‘from an account of our existing motivations’ (Geuss 2008, 95). 8
In what follows, I argue that both in Williams’s and in some contemporary scholars’ reflections, the commitment to internalism should be understood as split into two different, although related, commitments. The first is what I call methodological internalism – that is, the claim that standards of legitimacy should not be derived from pre-political moral considerations. This aspect is strictly connected to the realist commitment to anti-moralism. Methodological internalism pertains to Williams’s arguments against political moralism and his claim that questions of legitimacy arise from the attempt to solve the first political question. Horton’s argument in favour of ‘decoupling justice and legitimacy’ (Horton 2012) also falls into the category of methodological internalism. 9 As Horton explains, this commitment is motivated by a double concern. First, in moralist theories of legitimacy, there is a risk of ‘descriptive inadequacy’ when setting the premises for developing a normative account (Horton 2010b, 433). In a nutshell, standards are expected to regulate the critical realm and should therefore be thought of in relation to a disenchanted picture of what that realm amounts to. Failing to do so leads to the second concern, namely ‘normative irrelevance’ (Horton 2010b, 433). That is, by setting the bar too high, we lose the capacity to offer normative guidance in political circumstances that are far from ideal.
Importantly, though, anti-moralism does not imply that we cannot appeal to moral values in the political sphere. Sleat argues that realist political theory can appeal to moral considerations, but this does not imply that those considerations are deemed to have antecedent authority over political considerations. Questions of legitimacy, he continues, arise not because there are pre-political moral issues that are pressing us to probe the legitimacy of the regime; rather, they arise from within politics because ‘there is a crucial difference between political rule and successful domination’ (Sleat 2014, 322). Remarkably, he continues, ‘realist theory can draw upon moral considerations when making judgements regarding legitimacy without having to assume that morality has antecedent authority over politics’ (Sleat 2014, 322).
I call the second type of internalism substantive internalism. It is substantive in the sense that it refers to the values that subjects of a given political authority hold. Those values depend on the historical and cultural circumstances of the political community under assessment, yet also on the contingent aspects of that community (e.g. economic resources; social, political, natural events). Williams’s reflection on the MS category pertains to substantive internalism. Williams illustrates the contingency of the substantive standards for legitimacy by the use of an equation: LEG + Modernity = Liberalism (Williams 2005, 9). Liberalism is a contingent response to the question of legitimacy in the sense that it is valid only because (or when) the historical and cultural circumstances of modernity are in place: ‘The satisfaction of the BLD has not always or even usually, historically, taken a liberal form. Now and around here the BLD together with the historical conditions permit only a liberal solution’ (Williams 2005, 8). The overall message is that what makes sense to us as a justification for political power affects what the regime is required to do in order to be legitimate.
This is where Sleat’s acknowledgement that we can appeal to moral considerations while still remaining realists is prominent. In fact, the values that condition the substantive requirements of legitimacy can be – and normally are – moral. For example, the fact that, under certain historical circumstances, we value ideals such as freedom, justice and respect for life implies that we will not be satisfied with a notion of legitimacy that does not enhance, or even undermines, those ideals. Remarkably, though, the relevance of those values for identifying the standards of legitimacy is not determined by the assumption that those ideals are pre-politically valid. Rather, it is because we believe in those values, here and now, that they determine our requirements of legitimacy, here and now.
As previously mentioned, I take the two types of internalism to be strictly connected, although they represent two different stages of internalism. Sleat goes in a similar direction when he argues that in order to assess the appropriateness of a value for the political realm, we need to respond to ‘two different but related questions’: ‘“Is it a value that is appropriate for the political realm?”; and then, “Is this a political value for us?”, which is to ask the further question as to how far it might be a value that fits or belongs to our social world’ (Sleat 2016, 254). Williams’s introduction of the MS category is certainly functional to expounding the anti-moralist commitment of his realist approach. In fact, the MS claims that since we cannot ground legitimacy on pre-political moral considerations, we ground it on an understanding of the specific political context at stake. However, the MS goes further than what anti-moralism would require qua anti-moralism, which is to avoid the commitment to pre-political moral considerations. The requirement to look at what people happen to believe in given certain circumstances is not the same as saying that we should not lean on pre-political moral commitments. 10 In fact, as Sleat argues, ‘What counts as political values for us or any other society will always be a subset of all possible political values’ (Sleat 2016, 271; emphasis added).
Shifting from methodological to substantive internalism without recognising their difference does not come without problems. First, it makes it unclear to what extent subjects’ beliefs should matter for a realist conception of legitimacy. Second, this shift has a problematic implication for the normative import of Williams’s approach. Williams specifies that the MS category is not normative until we apply it to our own political society ‘because what (most) MS to us is a structure of authority which we think we should accept’ (Williams 2005, 11). However, legitimacy is instead a normative concept when applied to others too: ‘The idea of “LEG” is normative for us as applied to our own society; so it is also normative in relation to other societies which co-exist with ours and with which we can have or refuse to have various kinds of relations: they cannot be separated from us by the relativism of distance’ (Williams 2005, 14). The problem with this formulation, as I see it, lies in the fact that – if legitimacy at least partly depends on satisfying the MS category – it is unclear what asking the question of what makes sense to subjects of other regimes, with different cultural and historical circumstances, would normatively entail.
This ambiguity poses the problem of external critique, both from an internal and an external standpoint. These two aspects come together, as two sides of the same coin: ultimately, the problem is how to deal with external critique impacting internal stances of legitimacy, yet this problem can be investigated from the external standpoint (how to address internal stances) or from the internal standpoint (what to do with external perspectives). Williams himself argues that ‘conceptual complications multiply when one is concerned with a different case, that in which a style of legitimation that was accepted at one time is still accepted in some places but no longer accepted in others’ (Williams 2005, 70). He takes the example of the violation of human rights and defines the problem in terms of two questions that must be addressed: first, ‘what is actually happening?’ (Williams 2005, 70) and, second, ‘what, if anything, can we do about it?’ (Williams 2005, 72). Contextually responding to these questions provides some indications for how to react, yet Williams does not provide a more specific answer even with regard to the specific cases he brings up as examples, other than, first, warning that we might be in front of a case of legitimacy despite the violation of rights and, second, highlighting the role of external sources of information as making ‘a positive contribution against secrecy, the control of information, and the suppression of criticism’ (Williams 2005, 74).
Yet the same issue manifests when addressing it from the internal standpoint – that is, what should be the due impact of external critique on internal stances of legitimacy. Paul Sagar seems to allude to this issue when elaborating his internalist account of legitimacy. He claims ‘that the society may come into contact with outside values’ and changing its practices might look like a ‘realistic fable’ (Sagar 2018, 133): ‘Change can come about due to contact with different ways of organizing society, the sheer knowledge that difference is practically possible leading to the invalidation of inequalities previously legitimated through a belief in their necessity or naturalness’ (Sagar 2018, 133). Sagar argues that ‘contact with such outside values, however, would immediately render them candidate inside values’ (Sagar 2018, 133). In doing so, Sagar points out a way to concede some form of external influence to transform political practices that have been internally believed to make sense so far.
However, Sagar expresses a hope more than a prescription – he calls it ‘a realistic fable’ indeed (Sagar 2018, 133). The point, instead, should be more radical: if we can say that political realism would benefit from introducing some form of external critique, then the ‘fable’ turns into a theoretical commitment. If we cannot, then we should be content with hoping that subjects will be reflexive and informed enough to revise their acceptance of the status quo.
3. Why externalism might be desirable
Political realism has been frequently attacked for being a status quo–biased and conservative approach. However, the recent literature on the topic has shown that this need not be the case: while some ways of implementing the realist commitments can lead to conservative theoretical outcomes, there is nothing in the realist commitments per se that necessarily implies conservatism (Finlayson 2017). The self-reflection triggered by the CTP and the notion of ideology critique included in recent realist proposals open up theoretical space for conceiving of change as triggered by purely internal mechanisms of self-reflection. In the following, I argue that some forms of externalism can be desirable for internalist accounts of legitimacy, as they support the internal critical effort that realist theories already include in their normative setting. In fact, internal critique might lack a sufficiently strong tool to actually enhance self-reflection and political change (as in the example of secrecy, control of information and suppression of criticism that Williams recalls).
Take the example of vindicatory genealogy as recently developed (Prinz and Rossi 2017, forthcoming; Rossi and Argenton, forthcoming; Rossi 2019). In brief, genealogical critique should rely on epistemic standards, rather than moral ones, in order to distinguish between vindicatory narratives (ones that are consistent with the practices in place and must be preserved) and narratives that should be debunked (ones that are epistemically suspicious and should therefore be rejected). Problematically, though, it seems unclear who has the burden of doing genealogical critique: either the critique must be undertaken by the theorist, in which case it needs to be clarified where the theorist’s perspective stands vis-à-vis the subjects’ perspective, or the genealogical critique must be undertaken by subjects, in which case questions about the reliability of epistemic standards, information at disposal and freedom of criticism arise. Take the example of North Koreans facing the death of Kim Jong-il in 2011: despite the oppressive character of his regime, people were desperately crying over his death on the streets of Pyongyang, as the BBC reported. Such spontaneous reaction to the leader’s death shows how emotional reactions resulting from feelings of memberships and devotion to the charismatic leader trump rational considerations about how the leader actively contributed to the worsening of his people’s living conditions. This example puts under question the possibility for internal critique to develop, not only because open critique is repressed in totalitarian regimes but, more radically, because critique requires one to internalise evaluative standards that are independent from the ones delivered by power. The theoretical problem that this example highlights is that there are political circumstances in which the epistemic standards necessary to either vindicate or debunk legitimation narratives might not be at one’s disposal.
Introducing the external perspective can bring about some advantages both at the level of the political practice and at the level of the normative impact of the theory. The first issue concerns the possibility of actually enhancing progressive change in the political practice, with regard to both the values held by individuals and the institutional setting. At the level of people subjected to political power, knowing how things work in other political contexts can be of help in reassessing which states of affairs are desirable and possible: the introduction of the external perspective has the potential to bring about new political scenarios to enrich the set of political possibilities that might either be reachable directly from the status quo for a given political community or represent long-term objectives to work towards in order to make political progress. Besides, the external perspective brings about emancipatory potential by unveiling manipulative attitudes exercised by political power. Manipulation can affect both the interpretation and knowledge of relevant facts and the endorsement of certain values and standards. Such interpretation and endorsement can also be affected by applying an internal critique, at least to some extent; however, as previously noticed, it is difficult to expect that in regimes in which manipulation is highly widespread and pervasive, subjects can easily come up with independent or, at least, epistemically nonsuspicious standards that are able to push the internal critique of the same dynamics of power that affect those epistemic standards in the first place.
Introducing external critique also presents some advantages at the level of the normativity of the theory. In particular, it represents a further safeguard for the normative impact of the approach. Recall that these realist approaches insist on the connection between legitimacy and people’s beliefs. However, realists warn, legitimacy should not collapse onto de facto acceptance of power, thereby turning the notion into a descriptive and conservative one. As Beetham has argued against Weber, a regime is legitimate not because subjects believe it is but because it can be justified by people’s beliefs (Beetham 1991, 11). Further, as Horton points out, beliefs in legitimacy should be based on salient criteria: support of a regime due to habit, fear or expediency does not count as a justification for the regime’s legitimacy. Once again, in political contexts in which subjects commonly engage in self-reflection and political standards are continuously examined, internal critique possesses sufficient instruments for avoiding the possibility that de facto acceptance of power prevails over normative reasoning. However, in political contexts in which subjects are threatened by the political authority or political debates stop being central in political life, external critique might be necessary to exclude from the set of beliefs relevant for the assessment of legitimacy those that result from fear or uncritical habit.
These aspects are prominent for realist theories of legitimacy for two reasons. First, realist standards for legitimacy aim at being useful for assessing any type of regime, including nondemocratic ones. Normative standards for legitimacy, in other words, should not collapse onto democratic standards. Second, even within democracies, we may live with distortive ideological phenomena and in times in which political reflection is dormant.
4. The place for external critique: Preparing the field
In this section, I provide an account for how some forms of external critique can be accommodated within an internalist theory of legitimacy, namely one that commits to both methodological and substantive internalism. I argue that while methodological internalism cannot withstand the intrusion of any external perspective, substantive internalism is compatible with some modes of external critique. Substantive internalism being the focus of investigation, I first need to say something about what it means to reconnect legitimacy to people’s beliefs and attitudes.
4.1. Beliefs in legitimacy and positive governance
I claimed that for realists, the connection between legitimacy and subjects’ beliefs must be restored. Yet little has been said about what it means to believe in legitimacy. In what follows, I analyse what a belief in legitimacy consists in and what it means for a subject to hold such a belief. Yet before proceeding, a preliminary consideration is due. In this account, I assume there is a difference between a belief in legitimacy and legitimacy as justifiability in terms of people’s beliefs, which I spell out by borrowing some remarks from John Horton’s account. Horton (2012, 141) claims that ‘we have to restore the connection between political legitimacy and the beliefs and attitudes of those subject to it’. However, he warns, this does not equate either with saying that a state is legitimate so long as people believe it is or with endorsing a consent theory of legitimacy: ‘It is not a consent theory because “consent,” or “recognition” or “acknowledgement” as it might be better termed, is not itself the reason on this account why people accept a state as legitimate’ (Horton 2012, 141–42). Rather, people accept power’s legitimacy because they have reasons supporting their stance: ‘I consent to, or more properly recognize or acknowledge, the state as legitimate, because it meets the salient criteria of legitimacy that are practically operative.…The affirmation of legitimacy matters, but that affirmation is grounded on something other than the affirmation itself’ (Horton 2012, 142).
Similarly, I argue that a belief in legitimacy expresses subjects’ acceptance of power, which is grounded on reasons related to their positive assessment of the regime’s performance. 11 In a nutshell, it is because subjects perceive the positivity of the governance – according to their standards and systems of values – that they are able to accept the exercise of political power.
The very fact of linking legitimacy to positive assessment of governance implies that a belief in legitimacy includes at least two distinct elements that form the grounds upon which subjects acknowledge the state’s legitimacy. First, a belief in legitimacy (BL) includes a reference to facts: the regime does x, y, z (call this the factual component (FC)). Second, it includes a value judgement: x, y, z are positive features (call this the normative component (NC)). The positivity of x, y, z is based on values people hold, although they might not expound or be fully aware of them. 12 For example, an individual who positively views some policies promulgated by her regime (and therefore declares its legitimacy) regarding the distribution of resources does so because she believes in the value of justice or, if she is selfish, because she thinks that the regime has to provide her with some benefits (which probably equate with supporting principles of inequality).
Note that this structure requires the individual to be coherent in deriving the belief in the regime’s legitimacy from the premises about its positive performance (NC and FC). The scope of coherence is displayed on two distinct levels. The first level concerns the link between legitimacy and positive performance: the individual is inconsistent if she derives the regime’s legitimacy from a negative assessment of the regime’s performance. The second level concerns the link between values and positive assessment of the regime – that is, it is because people believe in certain values that they evaluate the regime positively when it performs according to those values. 13 In other words, an individual claiming at the same time to believe that the value of justice is a salient criterion for legitimacy and that a given unjust regime is legitimate would be expounding an incoherent belief. Coherence is relevant because expressing a belief in legitimacy ultimately means to authorise the rightful exercise of political authority. When individuals accept political authority even though their ‘true’ evaluation of the regime would lead to a different outcome, they end up supporting a regime that they would not support according to the criteria they consider salient for a regime to be legitimate.
4.2. Why do we need constraints?
So far, I have maintained that subjects believe a regime to be legitimate when they perceive that the political order is grounded on principles and values they share – that is, when they consider the regime to have positive features and to provide positive outcomes that reflect the values they hold. Surely, subjects can exercise self-reflection and revise the standards that they previously adopted.
In this section, I am interested in how the external critique can be included in a theory of legitimacy that prioritises subjects’ beliefs. Recall the two levels on which political realism is internalist: methodological (meaning that political standards should be anchored in the constitutive facts of politics) and substantive (meaning that political standards should be elaborated starting from the beliefs of those subject to political power). It is with regard to this second level of internalism that I propose to introduce the external perspective as complementing the internal one. However, I argue, not all external interventions are desirable for an internally elaborated theory of legitimacy.
To introduce the problem, let us schematise the situation in which the internal (II) and the external (EI) individuals express their assessment of the legitimacy of a regime R. II and EI are here conceived of as fictional characters who represent the collective internal and external perspectives, respectively. While I use these fictional characters to make the methodological discussion less abstract, I in the end draw some conclusions for when external critique can (and should) be integrated into internalist accounts of legitimacy. In the fictional interaction between II and EI, four situations may occur. Two of them are cases of agreement between II and EI: they both agree on either the legitimacy or the illegitimacy of R. The other two (summarised in the below table) are cases of disagreement.
I start from the hypothesis that these two cases of disagreement are not equally bad. More specifically, there is something intuitively worse in telling an individual her regime is legitimate when she thinks it is not than in telling her it is illegitimate when she thinks it is. Understanding where this asymmetry comes from sheds light on the reasons why we need normative constraints on the interaction among individuals. Besides, it helps define the content of those constraints.
Recall what I suggested about beliefs in legitimacy. First, they assume a link between legitimacy and positivity of the regime’s properties; second, they include both normative and factual claims. We can visualise such a belief as follows: (1)
For EI to deny the belief in legitimacy, he needs to deny either NC or FC (or both). Note that denying either NC or FC implies denying BL – that is, the legitimacy of the regime and, accordingly, the individual’s perception of correspondence between her values and the ordering principles of her regime. This implication is relevant because what is at stake, when denying others’ convictions, is not just a disagreement about values. Rather, such a denial has profound implications regarding the experience of being a subject of a political power.
To deny NC means making a claim about facts – that is, asserting that regime R does not possess quality x. For instance, II might claim the following:
(2)
For EI to deny FC, he needs to provide evidence showing that the regime does not actually provide freedom – for example, showing that people do not have a right to free speech. Another option for EI is to deny NC – that is, to deny the positivity of x. By doing so, EI is not making a judgement about facts but rather a judgement about the values held by II that support II’s acknowledgement of the regime’s legitimacy. Following example (2), EI needs to claim that freedom is not a positive feature.
Although, at first glance, the denial of FC might seem easier, it presents some theoretical problems. First, when we speak of facts, it might be difficult to truly distinguish between those facts and our standpoint on those facts. Second, to individuate the threshold beyond which evidence allows for abandoning a certain belief remains problematic. After all, II will probably not abandon her belief in the regime’s legitimacy just because it failed to guarantee a certain performance once, although a continuous failure in guaranteeing that performance would lead her to question the regime’s legitimacy. The problem pertains to the ability to effectively establish an acceptable threshold.
The rejection of NC involves even deeper concerns because NC concerns the sphere of values held by subjects of a political power, the values that support II’s claim that the regime is legitimate. In fact, while the charge of being wrong when uttering FC might be imputed to limited rationality or a deficit of information, having one’s values contested might be perceived as an attack on substantial aspects of the person. Besides, if we assume there is no objective hierarchy of values to lean on (as political realism does when rejecting moralism), any judgement that claims the wrongness of NC risks being perceived by II as an imposition by EI.
In the following, I spell out the risks incurred when introducing the external perspective in the wrong way – that is, in a way that breaks the connection between legitimacy and people’s beliefs and attitudes. I argue that two problems might arise when imbuing substantial internalism with the external perspective: denying that legitimacy at least partly depends on subjects’ values (thereby reintroducing a form of externalism), and denying that being exposed to coercion gives subjects a right to ultimately decide upon their regime’s legitimacy. These reflections serve as a basis for carving out space for external critique and understanding what a correct introduction of externalism might look like.
4.3. Preserving the link between legitimacy and subjects’ beliefs: Problematic cases of external critique
Let us now analyse the situations displayed in the table. Recall that II and EI are fictional characters representing, respectively, individuals living in a political community and the external critique. Also note that a belief in legitimacy is grounded on the criteria that II considers salient for acknowledging the regime’s legitimacy. For EI to criticise II’s belief, we previously saw, she can appeal either to the FC or the NC contained in the claim that the regime is legitimate. For example, EI might claim that although II is persuaded that the regime provides freedom, it actually does not and, therefore, II should judge her regime to be illegitimate. Surely, not all claims that deny FC are necessarily problematic for preserving the connection between beliefs and legitimacy. Let us imagine three different cases. EI claims that the regime implemented some rules that violate the subject’s freedom. EI claims that the amount of that violation is unacceptable. EI claims that, according to the right understanding of freedom, the regime is actually not providing its subject with freedom.
Case (a) is not problematic, as it criticises FC on the basis of additional factual information that might not be at II’s disposal, without calling into question the link between II’s values and legitimacy. After all, limited information intrinsically characterises how we as human beings know things. Besides, manipulative regimes try indeed to hide relevant information, hence leading to a belief that is controversial in its coherence.
Case (b) can be problematic, depending on the reasons supporting the claim. If (b) is a consequence of (a), then it is not problematic, because the reason why II considers the violation of freedom as acceptable could be that she lacks complete information about the amount of such a violation. However, it would be problematic if EI claimed that II has an accurate understanding of the violation of freedom but the threshold beyond which II is no longer willing to acknowledge legitimacy is too high to be acceptable. This claim would imply that II’s formulation of the value of freedom is considered as invalid for grounding her acknowledgement of the regime’s legitimacy, which in turn would amount to denying substantive internalism tout court.
Case (c) is a problematic one because EI implies that there is a right understanding of freedom, typically his own, and that according to that understanding, the regime’s performance must be negatively assessed. This is problematic because it denies both substantive internalism and methodological internalism by assuming that there is a moral truth that reveals the wrongness of II’s political standards.
In addition, EI might deny the validity of II’s claim about the regime’s legitimacy by denying NC. As mentioned before, NC regards the sphere of values II appeals to when expressing her judgement about the regime. This case might partly overlap with (c) mentioned above but could also represent a fourth case, (d), in which EI claims that x (freedom, in the example above) is not the right salient criterion on which to ground the regime’s legitimacy. In this case, EI claims that II’s judgement of the regime’s legitimacy must be rejected because it is based on the wrong criteria.
So far, I have defined cases (b), (c) and (d) as problematic – that is, worthy of further discussion to establish the limits of the external critique vis-à-vis the realist commitment to the link between legitimacy and subjects’ values. I discuss the legitimate space for intervention by EI more in section 5.1. However, something can already be said with regard to case (d). The intervention of EI in case (d) is not acceptable, because it violates methodological internalism, the idea that the standards for political legitimacy depend on fulfilling pre-political moral values that are treated as universally valid and possessing antecedent authority. In claiming (d), EI assumes that there is a universal (moral) truth to appeal to in order to make a judgement about the political legitimacy of a regime. Therefore, EI’s position being a case of methodological externalism, her stance cannot be integrated within a realist theory of legitimacy.
4.4. The ruled facing the rulers
The second case of disagreement occurs when II claims that her regime is not legitimate, while EI claims that it is. This case does not only risk severing the relationship between legitimacy and subjects’ values but also risk missing something important, namely that it is up to the ruled to face the rulers’ political power backed by coercion. After all, the fact of experiencing the exercise of political power is not a neutral element in evaluating the regime’s legitimacy, and, by definition, such an experience pertains only to II. Linking legitimacy to people’s values responds to an experiential need – namely, the need to think that coercion is not randomly exercised, but rather it is functional to realising a vision of the political order that is shared by rulers and ruled. This aspect is contained in Williams’s category of MS and in Max Weber’s work when he provides the definition of legitimate domination: Domination will thus mean the situation in which the manifested will (command) of the ruler or rulers is meant to influence the conduct of one or more others (the ruled) and actually does influence it in such a way that their conduct to a socially relevant degree occurs as if the ruled had made the content of the command the maxim of their conduct for its very own sake. Looked upon from the other end, this situation will be called obedience.… The merely external fact of the order being obeyed is not sufficient to signify domination in our sense; we cannot overlook the meaning of the fact that the command is accepted as a ‘valid’ norm. (Weber 1978, 946)
This case of disagreement is different from the earlier case because it affects the subject’s experience of the relationship between the regime’s use of coercion and the positivity of such use. In the first case, the individual is persuaded of her regime’s positive performance and therefore of its legitimacy; hence, she believes that there is correspondence between her values and the ordering principles imposed by the political authority. Although she might be wrong in assessing the positive properties of the regime, she does not experience the feeling of being arbitrarily ruled: according to her system of values, the regime has a right to coerce her because it does so in accordance with principles that she shares. In the second case of disagreement, instead, II claims that her regime is illegitimate – that is, it is exercising coercion without right, according to her perspective, because the political order does not match with her values. When EI claims, in contrast, that the regime is legitimate, she breaks the relationship between perceived positivity of the regime and the regime’s legitimacy, so that the painful experience of being arbitrarily ruled – according to the internal perspective – does not have any place in the evaluation of the regime. Put differently, the negative evaluation of the regime by II is not a valid element for assessing the regime’s legitimacy. EI’s position cannot therefore be included in the relevant assessments of the regime’s legitimacy, because it denies the relevance of substantive internalism tout court for an internalist account of legitimacy.
5. Accommodating external critique
We previously saw that introducing external critique within an internalist account of legitimacy can be beneficial. In what follows, I argue that it is also possible, provided that some constraints are respected. In other words, not just any type of external critique can be accommodated in an internalist account of legitimacy: while methodological externalism should be rejected, substantive internalism allows for (some modes of) interventions from outside, without renouncing the internalist character of the approach.
5.1. External critique facing the internal perspective
In what follows, I argue that there are three modes of external critique that can be accommodated within an internalist account of legitimacy because they do not undermine methodological or substantive internalism, while two others cannot.
First, the external critique cannot be included when it substitutes value x – on which the internal judgement about legitimacy is grounded – with another value. In terms of the fictional representation of the interaction between EI and II, EI is not allowed to claim, for instance, that freedom, instead of justice, should be the salient criterion for acknowledging the regime’s legitimacy. Two reasons support the introduction of this constraint. First, as discussed in cases (c), (d) and partly (b) in section 4.3, failing to do so would amount to denying methodological internalism, which is a fundamental commitment for any realist account of legitimacy. Second, substituting the value held by subjects would amount to denying substantive internalism, for it would deny that the salient criteria individuated by II are a fundamental element of the legitimacy assessment.
Second, external critique cannot be appealed to as a universal or more objective standpoint with regard to the dimension of values (as happens in cases (c) and (d)). As previously illustrated, political realism rejects both the assumption that there is any objective hierarchy of values and the commitment to grounding political standards upon alleged pre-political moral truths. Appealing to universal values would undermine methodological internalism.
That said, there are cases in which an external critique can be introduced into an internalist account of legitimacy. First, EI may show the consequences of choosing value x over y and to highlight the implications of such a choice. Take the example of China’s capitalism. In the last decades, China has enormously increased its economic power by structuring the political and the social order in accordance with capitalist principles. This happened, however, at the expense of the workers’ rights and health, as well as enormous costs for the environment. External critique could object that maximising capitalist achievements has undermined the security of citizens, as well as the global environmental sustainability. While external critique cannot delegitimise the capitalist values around which the political and social order is settled without undermining substantive internalism, it can advance the need for a more balanced system of standards in which capitalist growth does not overwhelm basic security and human rights. This mode of external critique does not violate methodological internalism, because it does not impose any pre-political moral value in the evaluation process. Besides, it does not violate substantive internalism, as it engages dialogically with the internal perspective – that is, it takes the internal standpoint as a valid starting point for further normative investigation. Besides, as anticipated in section 3 on the advantages of externalism, the interaction between internal and external perspectives has emancipatory potential because it can enhance self-reflection both over standards the application of which might lead to controversial outcomes that were not clear before and over standards that might have turned into contradictory ones over time. 15 In this sense, workers can raise their voice and defend their rights by comparing the working conditions of other capitalist societies and realise that a more sustainable form of capitalism is possible, thereby making external values ‘internal candidate values’ for their own society. This mode of external critique has the emancipatory potential of dislodging beliefs in the inevitability of a given structure of power, learning different ways of being subject to a political order, and fostering political change towards more emancipatory scenarios (Sagar 2018, 133–36).
Second, EI is allowed to show that value y may perform the same function that x does according to II, while presenting further advantages – for example, being more feasible or more morally sophisticated. This mode of external critique has the potential to broaden the set of political possibilities by shedding light on political settings that were previously unknown, thereby offering new possibilities for political progress. In addition, it can help understanding which are the circumstances that need to be changed in order to make new political outcomes possible. 16 In this case, external critique does not violate either methodological internalism, because it does not introduce any pre-political moral standard, or substantive internalism, because it enhances further normative investigation within the set of values held by subjects of political power. Take the example of a liberal democracy in which, as Williams recalls, what makes sense is a political order in which human rights are enforced and citizens are treated as equals. While this is a straightforward requirement for liberal democracies to be legitimate, the substantive conception of what a right is, and what are the respective duties of governments, might vary greatly according to whether rights are conceived of as either negative or positive, and to whether duties are conceived of as a matter of justice or as a matter of charity. For instance, ‘if we make the reasonable assumption that governments ought not simply to refrain from killing, maiming, or taking the property of their people, but should also take effective measures to ensure that others do not engage in such wrong-doing, then “negative” rights require “positive” actions on the part of government, including the redistribution of wealth through taxes’ (Buchanan and Powell 2018, 281). A shift in the conception of rights and duties has enormous impact on how to conceive of progress in political societies (Buchanan and Powell 2018, chap. 9).
Finally, EI is allowed to deny the validity of FC when such a denial does not contain any implicit criticism of the values held by II (case (b) in section 4.3). Take the example of a regime self-proclaiming to be democratic by virtue of their having regular elections, but employing several forms of coercion undermining the actual freedom to vote. External critique can reject its alleged democratic character, and the resulting claim to legitimacy, by pointing out that substantive freedom is actually denied in the political practice. This mode of external critique helps guarantee the coherence of the belief in legitimacy, as it criticises any final belief that leans on fact-insensitive premises (FC). Besides, it helps unveil at least some forms of manipulation, the ones exercised when falsely claiming that the regime delivers a certain performance. It does not violate either methodological or substantive internalism because it does not introduce external moral standards and does not involve any contestation of the values at stake in the belief.
In sum, external critique can and should be accommodated within an internalist account of legitimacy when it denies the validity of factual claims, when it highlights unknown normative consequences of choosing a certain standard and when it suggests modification of internal standards in a way that does not deny the validity of the internal values but rather develops a constructive response starting from those values.
5.2. Inadmissible internal stances
In what follows, I argue that there are three further cases of admissible intervention by the external critique. These cases concern situations in which II herself is undermining the tightness of an internalist account of legitimacy that claims to offer a normative perspective by still emerging from her beliefs and attitudes.
The first is a case of uncommunicative attitude. In the fictional interaction between external and internal individuals, it arises when II elaborates beliefs that are not assessable by EI. We can picture a situation such as this by imagining an individual saying, ‘My regime is legitimate’, full stop. It is a case in which subjects accept political power for reasons such as habit or fear. 17 In other words, it is a case in which the internal perspective fosters power acceptance without clarifying which are the salient criteria on the basis of which II proclaims the legitimacy of the regime. In this case, the external critique can reject the internal stance of legitimacy because the internal support to the regime is not grounded on any salient criteria as is required by an internalist account of legitimacy. In this case, it is the uncommunicative attitude of the internal perspective that breaks the link between legitimacy and subjects’ values.
The second is a case of incoherent reasoning. The external critique can reject internal stances when the belief in legitimacy manifests incoherence – that is, when the final judgement about the regime’s legitimacy is grounded on a negative assessment of the regime’s properties. 18 For example, II might maintain, ‘Feature x is a salient criterion for legitimacy; the state does not possess feature x; the state is legitimate’. External critique does not violate either methodological or substantive internalism, because it does not at all contest the values held by subjects, but rather the subjects’ formulating the final judgement about the regime’s legitimacy in a way that is contradictory vis-à-vis their own system of values.
The third is a case of normative suicide that can occur as a response to internal disagreement – that is, disagreement about values and world views among groups of individuals subjected to the same political power. It occurs when II’s belief includes a value x that, by definition, contradicts the very concept of legitimacy – that is, the concept that political authority is justified in exercising power if it delivers positive performances according to the subject’s perspective. An example would be an individual maintaining, ‘Ignoring the perspective of a minority is a positive feature; the regime ignores the perspective of the minority; the regime is legitimate’. Note that considering this stance inadmissible does not equate with claiming instead that the regime must be responsive to the perspective of the minority in order to be legitimate. A regime might be deficient in responsiveness to minorities for various reasons at the level of the political practice, which we are not investigating here. Yet treating unresponsiveness as a salient criterion for legitimacy is first of all conceptually incompatible with an account that claims to link legitimacy and people’s perspective. Note also that considering this stance inadmissible does not even imply that internal individuals should employ criteria of altruism, fairness or equality in order to assess the regime’s legitimacy. On the contrary, individuals can be selfish, although without denying that every single group of citizens has an interest in requiring correspondence between their own values and the ordering principles of the regime.
The difference can be tracked in terms of the distinction between exclusion and non-inclusion: in the case of a normatively suicidal individual, the salient criteria included in the assessment of legitimacy require one to exclude some individuals from receiving a given benefit; in the case of the selfish individual, the standards require one to provide a given benefit to herself. Surely, we would want to claim that both behaviours are morally wrong, but only the former case is a case of normative suicide, as it explicitly grants the fact that some individuals’ perspective is ignored.
In concluding this section, let me dispel a possible doubt. The notion of normative suicide might seem to be implicitly assuming a priori the value of political equality, with which an internalist notion of legitimacy is supposed to be uncomfortable (because of methodological internalism). However, it is one thing to argue that politics should guarantee an equal say to everyone; it is another to argue that each and every individual would claim to have a say. To claim the former equates with saying that a standard of equality needs to be applied to politics. To claim the latter, in contrast, equates with making a descriptive evaluation, namely that people generally have an interest in determining how the political order should be set. Hence the prohibition on committing normative suicide does not aim at preserving political equality, at least in the sense of a universal objective standard. Rather, it aims at preserving the grounds on which a realist notion of legitimacy is elaborated – that is, the connection between people’s beliefs (about governance’s positivity) and legitimacy.
5.3. Qualifying external perspective
Finally, an objection can be raised. Accommodating the external perspective within an internalist account of legitimacy serves two different purposes. One is to increment the emancipatory potential of a political community by offering further normative reasoning on the values and standards shared by the internal perspective (section 5.1). The other is to prevent situations in which the internal individual herself is undermining the possibility of an internalist account of legitimacy that has a normative impact (section 5.2) – that is, one that includes stances of critical self-reflection and does not collapse onto de facto acceptance of power. However, some could argue, the discussion of the advantages of introducing the external perspective illegitimately assumes that the external perspective is a qualified one – that is, that it does not introduce further normative problems in criticising the problematic stances of the internal perspective. The objection is fair and requires us to introduce a further standard for the external perspective, one that applies independently of the mode of external critique at issue. External critique must avoid committing what I call – symmetrically to normative suicide – ‘normative homicide’, a case in which EI holds a position contrasting with the underlying idea that legitimacy is connected to the positivity of governance as perceived by the subjects. Racist, fascist and imperialist positions, to take an example, violate the idea that governance should be beneficial for – or, at least, not detrimental to – subjects and therefore are not admissible into the stances supported by the internal perspective.
This requirement captures something important that is not ruled out by the other cases of intervention by the external critique. Imagine a situation in which II grants legitimacy on the basis of criteria such as order, stability and protection from horizontal violence. External critique, as discussed above, is allowed to show that value y is more effective than x at achieving the objective desired by II and suggests that repression of freedom is better equipped to guarantee order and stability. By the only analysis of the cases in which the external perspective should be integrated into an internalist account of legitimacy, it seems that such a perspective must be accommodated by the internal one. However, suppression of freedom being a value that enhances normative homicide, external critique must not be accommodated within an internalist account of legitimacy. In sum, the prohibition on committing normative homicide prevents internal critique in cases in which, paradoxically enough, the intervention of the external critique enhances regression rather than progress.
6. Conclusion
In this article, I have argued that some modes of external critique can and should be accommodated within a realist and internalist account of legitimacy. They should be integrated because they strengthen the emancipatory potential of internal reflection and, in some cases, even stimulate it, as when, because of political circumstances, internal critique cannot be exercised. Although internal criteria remain the ultimate source of standards for legitimacy, the inspirational role of the external critique can enhance the transformation of ‘outside values’ into ‘candidate inside values’ (Sagar 2018, 133). In addition, these modes of external critique prevent internalist approaches from falling into descriptive accounts of cases in which political power is de facto accepted by subjects.
External critique can be integrated as long as it presents certain characteristics. To illustrate them, I started by distinguishing two types of internalism that characterise realist understandings of political legitimacy. While external critique cannot contrast with methodological internalism, it can complement substantive internalism by providing the internal perspective with new normative cues.
In conclusion, let me highlight that these reflections affect the role of the political theorist – namely, what the theorist is supposed to do when facing internal and external stances regarding the legitimacy of a regime by elaborating an overall judgement about the legitimacy of a given regime. While showing the compatibility of political realism with some modes of external critique is only a first necessary step to address cases of disagreement, further analysis is required to establish how to treat different stances of legitimacy in concrete political cases. Still, some preliminary indications can be drawn, by returning to the table of cases of disagreement exposed in section 4.2. In this article, we saw that some modes of external critique, the ones compatible with realist commitments, should be integrated into an internalist account of legitimacy. What does this mean, in terms of the assessment of political regimes? It means that the overall judgement of a regime’s legitimacy does not purely depend on the perspective of subjects to the regime, yet rather their stances should be filtered by the standards highlighted in the discussion above on the constraints on the interaction between EI and II. Expressions of support to the regime that engage with uncommunicative attitude, normative suicide and deep incoherent reasoning should be rejected, yet also those that are unable to respond to compatible modes of external critique. Surely, the methodological character of these indications needs to be further tested against real stances of disagreement, where assessing the unacceptability of political support is more complicated. In addition, this approach should be complemented with an analysis of the role of the theorist as being not only a spectator of disagreement, yet rather a participant in the interaction. To borrow an expression used by Swift and White, the political theorist is in this case a ‘democratic underlabourer’ and can ‘help her fellow citizens make their political choices’ (Swift and White 2008, 54); or, with Philp, she has a duty to ‘keep alive values and issues in the context of political actions to which politicians should be alert’ (Philp 2010, 482).
Footnotes
Acknowledgements
I am grateful to Rainer Forst, Darrel Moellendorf and the participants to the Kolloquium at Normative Orders for their comments on an earlier version of this article. My thanks also to Afsoun Afsahi, Robert Jubb, Matt Sleat, and Enzo Rossi for their feedback on previous drafts.
