1. Introduction
There is a huge amount to agree with and admire in Rainer Forst’s ambitious and wide-ranging new volume, The Noumenal Republic. This makes the task of the critic unenviably difficult. Needless to say, Forst’s singularly theory-driven approach leaves him vulnerable to perennial objections from those who favour a more plural, empirically driven, bottom-up approach to social critique (McNay Reference McNay2020). The critical point I press in this article may seem to align with this sort of criticism in that it highlights the limits of interpreting politics so single-mindedly as a question of justification. But my critical point is not, in fact, a methodological one. My problem is not with the theory-driven nature of Forst’s approach but rather with the substantive way in which he theorises the concept of justification as a foundational term capable of mediating empirical and normative critical analysis. My contention is that the critical potential of his theory of justification will only be as plausible as the conception of agency that underwrites it and that this conception is not yet fully convincing.
In Chapter 6 of this volume, entitled ‘Normativity and Reality: Toward a Critical and Realistic Theory of Politics’, Forst argues that the concept of justification, rightly understood and correctly applied, allows us to ‘bridge the gulf’ between normative theory and the empirical study of social and political relations (Forst Reference Forst2024: 130). He goes on to suggest that the concept is uniquely capable of mediating between the ideal and the real, between abstract principles and the messy, situated realities of social–political life. His aim in the chapter is to show how the concept can perform this mediating function without reducing either side of the above binaries to its opposing term, thereby allowing for a method of critique that is both grounded in the granularity of existing social relations while at the same time remaining transformative and emancipatory in its aims. ‘What is essential to this project’, Forst writes, ‘… are mediating terms that enable us both to analyse political orders and their dynamics descriptively and to develop normative reflection along the very same conceptual lines of analysis as in the descriptive work. The mediating term I have in mind is the concept of justification’ (120).
In what follows, I first clarify Forst’s claims about reasons and the role they play in agency, before asking how far they can support his broader contention that justification can function as the mediating term he takes it to be. I then go on to argue that his account of reasons is problematic in two respects. In the section ‘Justification as a mediating term of analysis: the problem of motivating and explanatory reasons’, I argue that Forst collapses the distinction between motivating and explanatory reasons, while overplaying the role of motivating reasons in his account of power and agency. The result is an explanatory framework that downplays or excludes factors such as racism and class oppression, thereby limiting its ability to account adequately for the workings of power. In the section ‘Justification as a mediating term of analysis: the problem of motivating and normative reasons’, I critically examine Forst’s use of ‘justification’ to refer both to normative reasons and to the motivating reasons instilled by power, arguing that this terminological move rests on a guise-of-the-good theory of agency that may struggle to account for expressive, intentional action. I suggest that the overly intellectualised model of agency to which Forst subscribes, coupled with his personalist conception of power, weakens his claim that justification can mediate between normative and empirical registers of analysis.
2. Justification as a mediating term of analysis: the problem of motivating and explanatory reasons
In this section, I argue that Forst errs in collapsing the distinction between motivating and explanatory reasons, as well as in stipulating that all power operates through the manipulation of the former. Forst’s argument that justification should be the central term of empirical political analysis rests on the premise that ‘the real reality of politics, of the cave as well as of a perspective free (or, better, partially free) from the cave, is the reality of justifications’ (Forst Reference Forst2024: 120–21). He argues, ‘thinking realistically about politics requires political scientists to find ways to analyse and understand the justificatory dynamics that give rise to normative orders and that stabilize or destabilize them’ (124). But when Forst claims that ‘normative orders in modern societies rest on complex narratives of justification that support existing power structures and enable individuals or groups to exercise power within those structures’ (ibid.), he is not speaking about ‘good justifications from a reflexive or even a Platonic viewpoint…’. ‘[R]ather by justifications in an empirical and descriptive sense’, he writes, ‘…I refer to what makes people effectively think in certain ways and view reality in a certain light’ (121). It is justifications in this empirical–descriptive sense that ‘…constitute political reality as a reality of thought and action’ (ibid.).
To understand why social–political life is, on Forst’s view, constituted by the ‘reality of empirical justifications’, we need first to understand how he thinks power operates on agents. For while ‘the real reality of politics…is the reality of justifications’, it turns out that, for Forst, the real reality of power consists in how justifications affect motivating reasons. Forst begins with the following definition: ‘Power is the capacity of A to motivate B to think or do something that B would otherwise not have thought or done’ (Forst Reference Forst2024: 122). ‘If B is an agent, the secret of power is to affect his or her agency internally… by ‘giving’ B a motivating reason to act in a particular way [or belief or feel something]’ (ibid.).Footnote 1 He writes:
[…] to be the subject of power means that the justifications that make you think or act in a certain way are determined in relevant ways by others: if the determining in question is unintentional, I call that an effect of power, if it is intentional, then it is an exercise of power (133).
So, ‘…the real site of the resources for and the exercise of power is the cognitive realm, that is the realm of justifications, since A needs to be able to affect and change the realm of motivational reasons for B’ (122). In sum, ‘…to have and exercise power means being able to influence, use, determine, occupy or even seal off the space of [motivating] reasons for others’ (ibid.).
Importantly, Forst intends this account of power to be normatively neutral.Footnote 2 By this, he means that A may influence or colonize B’s realm of reasons or justifications through ‘a good speech, a sermon, an ideological world description, a lie, a threat, an order or an act of seduction’ (Forst Reference Forst2024: 123). In seeking to clarify the spectacularly diverse sorts of motivating reasons at issue, he again stresses the point, stating that these reasons are ‘for the most part not reflected on or critically tested, but they are nevertheless effective reasons, reasons that “work”’ (ibid.).
The argumentative sequence just reconstructed is worth considering more closely. Forst moves from the claim that political reality is constituted by justifications, to the claim that power operates by affecting agency internally, to the further claim that it does so by shaping the motivating reasons in light of which agents think and act. Thus, the empirical usefulness of ‘justification’ as a mediating concept depends on a substantive thesis about agency: that power becomes socially real by entering, shaping, or foreclosing the space of an agent’s motivating reasons via the justifications it provides. My worry begins here. If motivating reasons are only one part of what explains action, belief, and political subject-formation, then justification may still be a crucial mediating term, but not the comprehensive one Forst needs it to be.
For the sake of my present purposes, then, I bracket the question of what happens when we think normatively about these same empirical or motivating reasons, those that ‘work’ in making people think, act, and feel as they do. Instead, I focus on Forst’s conception of motivating reasons, for it raises questions about the independent plausibility and critical potency of his account of justification as a mediating term of analysis.
On my reading, Forst largely follows contemporary orthodoxy in holding that a motivating reason is a reason on which someone acts – a reason which guides their action, or, as we might put it, a consideration in light of which they act (Scanlon Reference Scanlon1998; Dancy Reference Dancy2000; Schroeder Reference Schroeder2007; Alvarez Reference Alvarez2010). In calling them ‘reasons that work’, Forst suggests that these reasons cause real effects in the empirical world – in the form of actions, beliefs, attitudes, feelings, and so forth – which can, to a certain extent, be observed and studied empirically. The primary contrast here is with the concept of a normative reason, which, if I am reading Forst correctly, he takes to be a reason for someone to act in a particular way. Forst seems to follow the contemporary orthodoxy on this question as well, which is well captured by T. M. Scanlon’s characterisation of a normative reason as ‘a consideration that counts in favour of’ someone’s acting, believing, or feeling in a certain way (Reference Scanlon1998: 17).
In my view, Forst’s use of ‘normative reason’ is best read as deliberately thin: A normative reason is a consideration that counts in favour of an action, belief, or attitude. For when he uses the term ‘normative reason’ and its equivalents, he neither assumes nor seeks to establish by definition anything in particular about how we come to distinguish between better and worse normative reasons, or what the grounds or scope of good (or better) normative reasons might be. Of course, Forst has a tremendous amount to say about these questions: reasons that can be generally and reciprocally justified to others considered equal normative authorities count as a special class of ‘good’ normative reasons in the moral–political sphere. This claim, however, is supported by further argument and is not presented as following from the broad definition of normative reason he employs.
Having reconstructed Forst’s notion of a motivating reason and his contrasting conception of a normative reason, let us now turn to his understanding of the connection between motivating reasons and empirical social analysis. Recall that for Forst ‘thinking realistically about politics requires political scientists to find ways to analyse and understand the (factual) justificatory dynamics that give rise to normative orders and that stabilize or destabilize them’ (Reference Alvarez, Way, Zalta and Nodelman2024: 120). What underpins this claim, I contend, is the contention that motivating reasons – engendered or influenced through the exercise or effects of power – are explanatory of the beliefs, behaviours, world views (etc.) of those subject to power. This is why, as Forst argues, ‘we fail to understand how power is exercised either individually, collectively, or within structures if we do not understand the justifications that guide people’s actions, as those who have and those who are subject to power’ (123). The notion of justification employed in this passage is clearly an empirical one, referring to those motivating reasons ‘that make you think or act in a certain way’ insofar as these have been manipulated or ‘determined in relevant ways’ by justifications supplied by others (133). It follows that, on Forst’s view, an empirical analysis of orders of justification is at the same time an analysis of power, because the motivating reasons that power produces, shapes, or forecloses are taken to explain the beliefs and actions of those subject to it. Justification, in its empirical–descriptive sense, thus provides the lens through which the internal effects of power on agency can be traced.
I agree with Forst that we can very often explain what a person does (or believes, etc.) by citing their motivating reasons. For example, we might explain why someone voted for a particular candidate by pointing out that they believed the candidate would improve public healthcare and that they care deeply about that issue. The person’s belief and desire together function as their motivating reason, which explains their vote in a way that is both intelligible and revealing. Given the strong, intuitive link between motivating and explanatory reasons, it is little surprise that motivating reasons are often simply called ‘explanatory reasons’ in the literature and treated as such (Alvarez and Way Reference Alvarez, Way, Zalta and Nodelman2024). Simply collapsing the distinction between the two, however, as Forst appears to do in order to argue that power can be comprehensively explained by analysing justificatory dynamics that operate unfailingly through motivating reasons, strikes me as hasty. I contend that the reasons that motivate people’s actions are not always the reasons that explain them, a point that is increasingly acknowledged in the contemporary literature on reasons (Baier Reference Baier1958; Alvarez Reference Alvarez and Leist2007, Reference Alvarez2009, Reference Alvarez2010; Hieronymi Reference Hieronymi2011). An example will help illustrate.
Bob storms the U.S. Capitol on 6 January, guided by the belief, induced by lies, that the 2020 election was stolen.Footnote 3 Of course, this belief is decidedly false: The 2020 U.S. election was conducted fairly and securely, and there is a mountain of publicly available evidence that proves that.Footnote 4 Still, there are two things that seem true about Bob’s attack on the Capitol that any account of reasons (or justifications, to use Forst’s preferred term) should be able to recognise. One is that Bob was motivated to act by the (putative) fact that the election was stolen. If you ask him why he did it, he will cite the stolen election as his motivating reason. The other is that his attack on the Capitol is, in fact, explained by something else: not the (putative) fact that the election was stolen, since there is no such fact, but the real fact that Bob falsely believes the election was stolen. So, here we seem to have two quite different reasons: one that motivates Bob – the putative fact of the stolen election; and one that explains his action – the veridical fact that he believes, falsely, the election was stolen. We can distinguish, that is, between the reason that explains Bob’s action and the reason that motivates him to act.
Accordingly, I think it is wise to follow the recent literature by maintaining a tripartite distinction between kinds of reasons: normative, motivating, and explanatory, corresponding, respectively, to whether a reason favours, motivates, or explains an action (Dancy Reference Dancy2000; Alvarez Reference Alvarez and Leist2007, Reference Alvarez2009, Reference Alvarez2010; Hieronymi Reference Hieronymi2011). Although in many cases the same reason may fulfil more than one of these roles – or even all three – it seems reasonable to assume that there are also cases where favouring, motivating, and explaining reasons (or some subset of these) diverge, as in the Bob example. This possible divergence is important, for it calls into question Forst’s assumption that the motivating reasons instilled or influenced by power are sufficient to explain the exercise and effects of power, where these include the actions of those who are subject to it. We can, and should, agree with Forst that motivating reasons are the considerations in light of which agents act, and that such reasons are often shaped by relations of power. But we need not follow him in treating these reasons as being fully explanatory of the actions they motivate across the board.
To repeat, my argument above does not deny that motivating reasons have explanatory force. They often do. But they explain in a distinctive, participant–perspectival way: They render an action intelligible as something the agent took herself to be doing in light of some particular consideration. A third-person analyst can reconstruct that perspective and cite the agent’s motivating reason as part of an explanation. But this does not mean that the motivating reason exhausts the explanatory story. In Bob’s case, the putative stolen election explains his action only as the consideration under which the action appeared intelligible and compelling to him; a fuller explanation must also cite the false belief, the propaganda that helped produce it, and the wider social conditions that made him vulnerable to that propaganda. My objection to Forst is therefore not that motivating reasons are explanatorily unimportant, but that they do not necessarily explain or explain enough for the purposes of a comprehensive critical theory of power.
This point might seem like mere hairsplitting, but I believe that it illuminates a weakness in Forst’s methodological framework – focused on the concept of justification – that is too important to ignore. For once a social analyst discovers other, perhaps more powerful explanatory reasons for Bob’s actions – not the putative stolen election (his motivating reason), but instead his false belief in the stolen election, for example – they will naturally be moved to seek out further explanations as to why and how Bob acquired his false belief, etc. In the first instance, the analyst might appeal to the politically motivated propaganda that led Bob to believe as he did. She might then appeal to further factors that explain Bob’s vulnerability to such propaganda, which may include historical, cultural, and political dynamics that go far beyond Bob and his mental states. Here, for example, the analyst might appeal to factors such as working-class decline or the perceived loss of social status relative to other groups – which are not, taken just as such, motivating reasons for anybody to do anything. This is important. On Forst’s account, none of these larger, structural social forces or dynamics – racism, class oppression, economic decline, etc. – can count as direct expressions or manifestations of power because none are motivating reasons for the agent, that is, none are considerations in light of which the agent herself acts, or forms beliefs, etc.
The problem, then, is that most of us do very much want a conception of power that allows us to view racism, class oppression, economic decline, social humiliation, etc. as expressions or manifestations of power, brought about either by its exercise or its effects. In this way, my first objection – that, contra Forst, motivating reasons are not necessarily explanatory reasons – gives way to a second and stronger objection: that by insisting that power operates exclusively through the manipulation of people’s motivating reasons (via the empirical justifications it provides), Forst excludes explanatory factors we would naturally want to include in giving a full account of the exercise and effects of power upon agents.
We can state the point slightly differently by starting from the observation that most accounts of acting for a motivating reason assume that the agent stands in an epistemic relation to that reason. This relation is often taken to be that of belief: to act for the motivating reason that p, the agent must believe that p (Alvarez and Way Reference Alvarez, Way, Zalta and Nodelman2024). One advantage of having an account of power that allows us to cite structural racism, say, as the explanation for the behaviour of many people (as well as institutions and systems) is precisely that it does not require us to postulate any specific epistemic relation between an individual agent and the effects of power that are expressed or enacted or reproduced in her behaviour. Lois McNay makes a similar point in her contribution to this same Author-Meets-Critics exchange, ‘Justification: Mediating Term or Homogenising Abstraction?’, where she writes that ‘…justification critique remains caught in an essentially personalist or interactionist account of power as domination that fails to reckon with the ways in which some forms of structural coercion actually operate’ (McNay Reference McNayforthcoming).
In sum, by stipulating that all power operates by way of individual motivating reasons, and then problematically equating motivating and explanatory reasons, we are left with a reductive explanatory framework that excludes, a priori, explanatory factors we would naturally want to include. This will be especially true of structural factors that do not seem to involve any specific epistemic relation between an individual and the effects of power that are expressed or enacted or reproduced in her behaviour. By the same token, a broader and more multivalent conception of power would not pressure us to exclude from our explanatory framework factors that are not reducible to operations that affect individual motivating reasons directly, whether engendering them, influencing them, or limiting what they can possibly be.
3. Justification as a mediating term of analysis: the problem of motivating and normative reasons
In this section, I return to Forst’s understanding of the relation between motivating reasons and normative reasons. More specifically, I examine Forst’s use of ‘justification’ to refer not only to normative reasons but also to the motivating reasons instilled or shaped by power, a move that underwrites his claim that the term ‘justification’ can mediate between empirical and normative analysis. I argue that this controversial usage tacitly relies on a guise-of-the-good view of motivating reasons, according to which such reasons appear justificatory in a normative sense to the agents who act on them. I question the adequacy of the guise-of-the-good view by considering cases of expressive action – particularly those driven by emotion – that challenge the assumption that all intentional action involves agents acting on motivating reasons that appear normative to them. These cases suggest that power may operate on agents not only through the manipulation of their reasons but also through affective channels that bypass reason altogether, thereby complicating Forst’s account and raising doubts about justification’s capacity to serve as the comprehensive mediating term he says it is.
As we have seen, and as Forst frequently points out, motivating reasons need not be normative reasons. The actual, motivating reason in light of which a person acts may not, in fact, favour the thing they do. Moreover, people do not always act for the normative reasons that favour their doing something even when they do what those normative reasons favour. Take, for example, those who do their moral duty for the motivating reason that it makes them look good or feel superior. And yet despite the ways motivating and normative reasons frequently diverge, Forst nevertheless persists in calling the empirical motivating reasons instilled or engendered by power ‘justifications’. It is precisely this move that sets him up to theorise ‘justification’ as a term capable of mediating between both empirical and normative registers of social analysis.
The question is why Forst takes the unusual step of calling the empirical, effective motivating reasons instilled or influenced by power ‘justifications’, when this term is usually reserved for normative reasons, on the assumption that only reasons that count in favour of an action – that is, normative reasons – can justify it. Forst’s purely empirical or descriptive use of the term ‘justification’ is apt to generate confusion, and he often has to stop and remind the reader that the empirical justifications to which the analyst must appeal in order to understand the exercise and effects of power need not be normative in any genuine sense. In other words, these ‘justifications’ do not have to be or provide ‘good reasons’, as he puts it. We can state the point more carefully by saying that the empirical justifications that explain the actions of those subject to power and its effects need not involve reasons that actually count in favour of the actions so explained.
What motivates Forst’s loaded, sometimes misleading, use of the term? The answer, I think, lies in the kind of agency his mediating project requires. Forst’s explicit clarification is that he is using ‘justification’ in an empirical and descriptive sense, to name what makes people think, act, and view the world in certain ways (Forst Reference Forst2024: 121). But this clarification raises a further question. Why call these empirical factors justifications at all, rather than causes, incentives, pressures, affects, or ideological mechanisms? If the term is to mediate between empirical and normative analysis rather than merely shift between two unrelated meanings, the empirical factors in question must have a reason-like form from inside the agent’s own practical point of view. They must operate not merely as causes of conduct but as considerations in light of which agents think, act, feel, or orient themselves in the world.
This is where, I think, Forst’s account tacitly relies on a weak guise-of-the-good view of motivating reasons. I do not mean that Forst explicitly endorses that doctrine, nor that he is committed to the strongest version of that view. My claim is only that, for Forst’s empirical ‘justifications’ to count as justifications, the motivating reasons shaped by power must appear to agents as considerations with some justificatory force. They need not be good reasons, as Forst himself emphasizes, nor need the agent take them to be morally sound, reflectively defensible, or all-things-considered rational. They must, however, appear as considerations that count in some respect in favour of the relevant action, belief, feeling, or orientation. This, therefore, echoes the weak Davidsonian thought that, when an agent acts for a reason, there is, from the agent’s point of view, ‘something to be said for the action’ (Davidson Reference Davidson1963: 691). Darwall (Reference Darwall1983) and Scanlon (Reference Scanlon1998) defend related versions of this position.
The textual basis for attributing this weak guise-of-the-good view to Forst is, I think, substantial. He defines power as the capacity to motivate B to think or do what B would not otherwise have thought or done and then specifies that, if B is an agent, power must affect B’s agency internally by ‘giving’ B a motivating reason (Forst Reference Forst2024: 122). He locates the exercise of power in the cognitive realm of reasons or justifications and describes speeches, sermons, ideological world-descriptions, lies, threats, orders, and seductions as ways of influencing or colonising that realm (122–3). He then insists that such reasons may be unreflected and untested, while still being effective reasons, reasons that ‘work’ (123). Taken together, these claims suggest that, for Forst, power does not merely produce behaviour from the outside. It operates by shaping what can show up to agents as normative or justificatory reasons for action.
This reading also explains why Forst thinks normative critique can arise from within the ‘cave-world’ of empirical justifications. If empirical justifications were merely causal determinants of conduct, then normative questions about good, better, or worse justifications would have to be introduced from outside the empirical analysis. The mediating function of justification would then collapse into equivocation: ‘justification’ would mean one thing in empirical explanation and something quite different in normative critique. But if empirical justifications already function for agents as apparent normative reasons – as considerations in light of which action, belief, feeling, or world-orientation appear in some sense as ‘good’ – then the same phenomenon can be studied descriptively as an effective motivating reason and normatively as a putative justification whose validity can be criticised. This, I take it, is the agency-theoretic assumption that enables Forst’s concept of justification to move between empirical and normative registers.
The difficulty is that this assumption is doing substantial philosophical work, and yet Forst does not defend it as fully as one might wish. A weak guise-of-the-good view may be plausible in many cases; indeed, much intentional action does seem to proceed in light of considerations that appear, to the agent, to count in favour of what she does. But it is not obvious that all intentional action has this structure. Consider the possibility that an agent acts intentionally without acting in light of any consideration that appears to her to count in favour of the action. It has been argued that negative emotions such as anger or frustration can lead us to act out in ways that we do not regard as good or reason-favoured in any sense (Alvarez and Way Reference Alvarez, Way, Zalta and Nodelman2024). Think of someone who impulsively smashes an object out of rage (Hursthouse Reference Hursthouse1991). In such cases, there may be no consideration the agent could cite as having appeared to count in favour of what she did. We might instead regard such actions as expressive or arational actions: intentional actions, perhaps, but not actions done for a reason in the ordinary sense (Bennett Reference Bennett2021).
This does not settle the issue against Forst, of course. A defender of the guise-of-the-good view might try to redescribe such actions as reason-guided under some very thin description: The agent wanted to vent, retaliate, express anger, or release tension. But that is precisely the point. Forst’s account requires an argument for why this redescription should always be available and why we should prefer it to the alternative view that some intentional action is affectively driven without being reason-guided in the strict sense. Without such an argument, it remains possible that power affects agency internally not only by supplying, shaping, or foreclosing motivating reasons, but also by organising affective dispositions – fear, anger, resentment, shame – in ways that do not pass through the agent’s space of reasons. If that is right, then justification remains a crucial object of empirical and normative analysis, but it is not the comprehensive mediating term Forst claims it is.
4. Conclusion
I have argued that Forst may be overplaying the role of motivational reasons in agency, while at the same time tying these reasons too tightly to explanatory reasons, on the one hand, and to normative reasons, on the other. All of this in an effort to get the term ‘justification’ to function as a conceptual bridge between normative and empirical analysis of ourselves as political beings who wield and are subject to power. Putting aside, for a moment, the critical diagnostic capacity of his account of power – about which I have raised doubts above – its independent plausibility will rest, in part, on the plausibility of the account of agency that underlies it. I have provided reasons for thinking that this account is not yet fully convincing.
To be clear, nothing I have argued here is intended to count as a criticism of Forst’s methodological orientation. I do not object to the attempt to formulate a single principle or concept that can mediate between empirical and normative modes of analysis, nor do I believe that any such attempt would necessarily be reductive in a bad sense. On the contrary, it strikes me as highly plausible that this mediating effort may be vital to the prospects of renewing critical theory by shoring up its most basic foundations and making good on its central promises. Forst’s way of tackling this challenge is inspiring. Indeed, I take his theory to be the best, most comprehensive, normatively grounded programme of critique currently on offer. My critical purpose, then, has only been to suggest that Forst’s substantive understanding of what justification is relies on a conception of power and agency that are not yet fully plausible and convincing in their own right, and that this, in turn, weakens the critical bridging function the term ‘justification’ is supposed to perform.
Acknowledgements
During final manuscript preparation in October 2026, I used ChatGPT (OpenAI, GPT-5.6, accessed via chatgpt.com) for limited editorial assistance with cosmetic issues of grammar, clarity, and sentence-level phrasing. The tool was not used to generate any of the paper’s content.
