Critical Republicanism and the Discursive Demands of Free Speech

A growing body of literature in feminist philosophy exposes the way in which occupying a particular group identity inhibits an affected agent’s ability to engage in communicative exchange effectively. These accounts reveal a fault in standard liberal defences of free speech, showing how, if free speech is a goal worth pursuing, then it must involve both a concern about the legitimate limits of state interference and of the effect of social norms on an agent’s communicative capacities. Building on the emergence of a ‘critical’ branch of neo-republicanism, this article argues that such speech-related injustices are best understood as a feature of structural domination that threatens the agency of those affected. Recalibrating our understanding of free speech along critical republican lines thus secures discursive agency in our communicative exchanges in a way that both ensures democratic legitimacy and realises equal status for all.


Introduction
The idea that members of certain groups are systematically disadvantaged in their status as speakers has, over the last several decades, gradually taken hold in debates surrounding freedom of speech. Those working in feminist philosophy of language, for instance, have written about the effect pervasive violent pornography has on the ability of women to express sexual refusal. 2 Social epistemologists such as Kirstie Dotson, 3 meanwhile, have outlined the mechanisms by which oppressed groups are subject to 'epistemic violence', such that their attempts at engaging in linguistic exchange fail due to the (intentional or unintentional) pernicious ignorance of the audience. 4 Building on a long history of black thought, 5 intersectional feminists have drawn attention to the way that racial hierarchies, and the norms upon which those hierarchies depend, profoundly inhibit the ability of black individuals to be taken seriously as equal agents engaged in discursive exchange. 6 Having a secure status as a speaker is important for some of the familiar reasons why free speech is important. When we express ourselves through words, we are making an active contribution to the realm of discursive exchange with our fellow human beings. Speaking with others allows us to influence, to convince, to argue, to make our feelings known, and to call attention to those things that have wronged us. When our attempts at speaking are misunderstood, distorted, silenced, or ignored, we experience the effects of a lapse of recognition in this shared discursive game. When this loss is systemic, when it occurs in a regular and socially entrenched way, we suffer a profound loss of agency, as our capacity to influence the world around us becomes limited in ways that are outside of our control. Importantly, when this loss occurs at the level of political engagement, our attempts to better our position will also be consistently undermined. 7 If an individual's social position can have a profound effect on their capacity to express themselves through words, then the liberal idea that free speech is best preserved through non-interference in the 'free marketplace' is thrown into serious question. 8 As these feminist analyses show, if we care about free speech, we must also pay critical attention to the conditions under which communicative exchange takes place, including those norms and practices that unjustly distort the 'free marketplace' to the exclusion of some of society's most vulnerable members.
Beyond offering a defence for a free speech right free from state control, however, the political theoretical literature has yet to offer a substantive solution to those social norms that systematically inhibit the speaker status of members of affected groups. With this gap in mind, this article argues for a critical republican account of free speech. My motivation for drawing on the republican literature is twofold. First, the neo-republican revival, with its emphasis on freedom as non-domination, 9 shows how the demands of freedom depend not just on mitigating domination on a vertical plane between the individual and the state, but also in the horizontal, intersubjective relations that citizens share with one another. This means that, in addition to the robust set of formal arrangements that citizens require in order to enjoy freedom, they must also live in a society in which the social norms through which they engage in their relations with others satisfy the demands of intersubjective freedom. 10 The second motivation for developing a neo-republican approach to free speech stems from the theory's stated purpose as both a method of social critique and as a prescribed institutional remedy that depends on the support of the informal actions of individual citizens. In contrast to existing philosophical accounts of speaker injustice, then, neo-republicanism provides an attractive and rich set of resources with which to construct viable institutional and social responses to threats to the speaker status of individuals. 11 Given republicanism's firm commitment to securing the conditions for an independent, sovereign citizenry, as well as its concern with social relations of equal intersubjective standing, the theory's omission of any substantive treatment of free speech is somewhat surprising. It appears, then, that the neo-republican revival has missed an important trick in regards to how we conceptualise free speech that, if addressed, should aid in constructing a much-needed reformulation of how we understand what the ideal of free speech actually requires. One reason for this, I suggest, stems from the faith that those within the tradition have in the power of contestatory democracy to shift social norms over time. On Pettit's influential account, a well-functioning democracy ensures that members of marginalised groups have the opportunity to contest and shift unjust norms and social arrangements via the process of deliberation. Relying on deliberation to shift social norms also has the added benefit of securing citizens against excessive and arbitrary state interference in the pursuit of social justice. In short, then, while republicans indeed express concern about the effect of 'social' forms of domination for limiting the freedom of individuals, they are wary of those forms of interference that aim to remedy such domination from the 'top-down', rather than from citizens themselves.
As several theorists 12 have pointed out, however, the effectiveness of this contestatory approach will itself be inhibited by those social norms that undermine the ability of members of certain groups from engaging in political deliberation on an even footing to more dominant groups in society. In rescuing the neo-republican project from this charge, then, recent developments in a 'critical' form of republicanism aim to respond to both state and social domination without betraying the tradition's core commitments to freedom. This paper thus adds to this debate by highlighting a further way in which Pettit's approach overlooks important threats to freedom that must be safeguarded against if we are to secure non-domination for all.
My argument will proceed as follows. The next section outlines and critiques Philip Pettit's recent distinction between 'unhindered' and 'protected speech' 13 showing how, while it mounts an important challenge to the prevailing liberal framing, it nonetheless does not go far enough in addressing the normative conditions required for one's speaker status to be sufficiently protected. In the Discursive Injustice and the Limits of Pettit's Approach section, I unpack one considerable source of speech distortion which Pettit's account fails to consider: discursive injustice. As Quill Kukla's 14 framework shows, our capacity to deploy standard discursive conventions can be severely inhibited by the existence of norms that track social disadvantage in systemic ways. This phenomenon, I argue, should trouble those republicans who, like Pettit, stress the role of speaker status in the enjoyment of equal status more generally.
To overcome this oversight in the republican literature on free speech, the A Critical Republican Concept of Free Speech section presents a critical republican lens through which to understand and address threats to speaker status of this sort. Arguing for a form of republicanism that is capable of being employed in critical social analysis, Dorothea Gädeke's 15 proposed critical republican framework proves conceptually significant for the present purposes in two ways. First, her argument re-centres the normative core of non-domination from discursive status to normative authority, thereby stressing the importance of co-authorship in the construction of discursive norms. Conceptualising status as normative authority, rather than discursive status, highlights the asymmetry in status that gives rise to discursive injustices in the way described by Kukla. In addition, Gädeke offers an account of domination that is structurally-constituted in a way that exposes those systemic threats to speaker status not explained by Pettit's proposed distinction. To conclude, I provide a brief sketch of recommendations for a critical republican account of free speech.

Pettit's 'Two Concepts' of Free Speech
Pettit's argument in 'Two Concepts of Freedom of Speech', 16 provides an implicitly republican framing of the issue of freedom of speech, as set against the backdrop of the university setting. The specific, and timely, practical question he aims to address in his discussion can be summarised as thus: when an academic is afraid to speak her mind, lest she is interfered with by university administration, can we really say that she enjoys freedom of speech? 17 Unsurprisingly, Pettit's answer to this question is a firm 'no', and he builds up to his conclusion in the following way. First, and seemingly expanding the standard neorepublican distinction between negative freedom as non-interference, and nondomination as freedom from the threat of uncontrolled 18 interference, 19 he offers a comparison between what he calls 'unhindered' and 'protected' free speech. 20 'Unhindered' speech follows the standard liberal position on the issue, that I enjoy freedom of speech when no one censors my speech or prevents me from expressing whatever I wish when I choose to express it. In other words, 'you get to say whatever you might want to say, regardless of the option you prefer'. 21 On this account, we enjoy free speech to the extent that our speech options are not interfered with, even when we decide not to exercise them. Unhindered speech thus draws a lot of parallels with a Berlinian 22 understanding of 'negative' liberty, and represents the most prevalent framing of free speech within both the theoretical literature and in legal and political understandings of free speech.
Under unhindered speech, then, we are only free to exercise speech so long as others, whether that be other individuals, the state, or non-state regulators, allow me to do so. In contexts where such an account of free speech applies, your speech choices can thus easily be constrained, either overtly or covertly, by the actions of others, which may include 'removing, replacing, or misrepresenting the options before you'. 23 Even where others do not interfere with our speech in these ways, the overarching threat of potential future interference will necessarily limit what we feel able to say publicly. To use an example, consider a society in which journalists are not legally constrained from publishing harsh criticism of the government. However, while publication of such material is permitted, they face the very real threat of sanction from both the government and their employers when their work is perceived to question the power of the state. So, while the state does not explicitly ban or censor such speech, journalists are not formally protected in saying what they wish, and so must curtail their speech and self-censor in different ways.
Pettit's second, and preferred, concept of free speech is what he terms 'protected' speech, 24 where there exists tangible limits on the ability of others to limit your exercise of speech. These limits come in two forms: first, we may put in place certain obstacles that mean that others no longer have an option to interfere with your speech; and second, we can arrange things so that those who do interfere with your speech face a penalty of some sort for doing so. 25 In order to secure against further arbitrary interference, these limits are designed and monitored in a transparent and democratically-controlled way. Pettit outlines three advantages of endorsing the concept of protected speech over unhindered speech. The first, and most important, is the status benefit associated with enjoying formal security to speak one's mind free from the threat of interference from others. Here, rules that constrain others from interfering with my speech provide me with a public expression of my independence. That the exercise of my speech is protected from interference sends the message to others that I am valued as a speaker, on an equal footing with those others with whom I share society. Individuals who enjoy protected speech in this way thus do not have to tread carefully for fear of saying the wrong thing and triggering the wrath of colleagues or administrators. They can walk tall and speak forthrightly. 26 The second benefit attached to protected speech concerns the way in which it allows for the enfranchisement of one's silence, something which Pettit has explored in earlier work. 27 On this view, once suitable speech protections are in place, then we can communicate our approval for a particular topic by remaining silent. Were we not suitably protected, our expression of disapproval might put us in danger of sanction from others, meaning that the communicative intentions of our silence are rendered unclear. A third benefit of protected over unhindered speech is that it allows speakers to be held properly responsible for the words that they say. Here, protection means that we can be assured that what others say really was freely chosen by them. These second two benefits of protected speech reflect the consequentialist aims of free speech. That is, in order to ensure that the laws and rules of society track the 'avowable interests' of citizens, then the process of public deliberation must take place in a context that allows citizens to make contributions without threat of sanction or censorship. In other words, then, the freedom promised by the republican political project can only be made possible where citizen speech is suitably protected in the way that Pettit describes.
We can thus see the republican commitments of this distinction between unhindered and protected speech. The central take-away of Pettit's formulation of unhindered speech is that it exposes the paradigm association of free speech with maximum choice as normatively insufficient. A society in which no laws or regulations regarding speech were to exist, and I were thus able to 'enjoy' speech in an unhindered way, I would be caught in a state of fear of reprisal or sanction for what I say from those who hold a greater level of power or influence than I do. As with many challenges to equating freedom with noninterference, the neo-republican revival has served to successfully challenge the myth that I am 'more free' the fewer formal restrictions there are on my choices. At heart, this challenge stems from republicanism's recognition of a particular tension between the independence and interdependence of human beings. As Pettit elucidates, 28 I enjoy nondomination when I have a robust and secure status, one that provides me with the independence of a liber ('free person'), where I do not have to express deference in my relations with others and am recognised as an equal discursive agent capable of answering to and being held responsible for reasons. As to the interdependence element, republicans also recognise that this enjoyment of status depends on the quality of our relationships with fellow citizens. In a world in which we enjoy only the unhindered sort of speech, then, we are not protected in our status as discursive agents but are instead subject to those uncontrolled forces that work to silence, discount, distort, and compel our speech, often before we have even chosen to exercise it.
Pettit's proposed concept of protected speech thus captures several important benefits not provided by the liberal 'unhindered' kind. In particular, the publicly salient nature of protected speech behaves with its own expressive force, in that it provides public recognition of the crucial relationship between being able to express oneself and the enjoyment of an equal status. So, quite apart from the instrumental benefits of speech, including its necessary place in processes of democratic deliberation and the contestation of arbitrary power, regulations that protect speech naturally serve to counteract those distortions of the free marketplace that unjustly prioritise the voices of some over others. Crucially, the neo-republican project of dismantling asymmetrical relations of power reflects a primary normative concern with protecting one's status as an equal interlocutor in the eyes of others. This relationship between one's speaker status and one's discursive status thus cannot be overstated. Under unhindered speech, the speech component of discursive status struggles to be realised in a world where others might choose to arbitrarily interfere with my speech. Recognising others as discursive subjects requires that we can attribute their engagement with discursive norms to their own true thoughts and feelings. But when we live in a world with unrestrained threat of sanction on one's speech, we do not know whether or not others are speaking under duress or as a way of flattering or placating potential interferers. And so, under unhindered speech, while we may formally possess a wide range of speech options, we cannot be said to enjoy nondomination: our status as equal interlocutors is undermined and we cannot 'walk tall' and look others in the eye. 29 As a potential alternative to the prevailing liberal stance on free speech, however, Pettit's proposed account of protected speech seriously overlooks the effect of pervasive social norms on the communicative capacities of certain individuals. As he argues, suitable protections in the academy will mean that academics and students alike will be able to 'walk tall and speak forthrightly', 30 and as thus will no longer self-censor or keep schtum when it comes to the kind of expression they wish to make. Again, the benefits here are not just about securing a greater range of speech options but are to imbue each individual with a robust speaker status.
The status that this confers, however, will not be possible unless we also take seriously the ways in which discursive agency is limited by unjust social norms, such as those that perpetuate racism and sexism. This general concern that Pettit's broader neo-republican framework is incapable of responding to structural or social forms of domination has been of growing interest in recent years. Critics here point to the limited way in which contestatory democracy alone can dismantle unjust status hierarchies, without additional support aimed at combating those norms, attitudes, and structures that allow such hierarchies to continue, even where formal equality has been secured. 31 The purpose of this present analysis can thus be said to add to that debate. In contrast to several previous critiques of Pettit, however, I do not think that his account is inherently incompatible with analyses of social domination. Rather, and as I suggest here, advancing a critical republican framework simply builds upon previously-underdeveloped aspects of the neorepublican project in a way that is sensitive to emerging debates on social domination and oppression in political theory.
The insufficiency of Pettit's proposed account of free speech, and of his lack of concern for social domination more generally, will be examined in the following section. Here, I show how a successful republican account of free speech must also address the threat of social norms as they give rise to the phenomenon of discursive injustice, something which both undermines the emancipatory potential of free speech and, in turn, prevents those who suffer from it from enjoying free and equal status in the rich, republican sense.

Discursive Injustice and the limits of Pettit's Approach
I have so far stressed the importance of the relationship between speaker status and discursive agency. When we express ourselves to others, we are making them aware of our status as interlocutors, capable of understanding reasons and making claims on other people in accordance with those reasons. And, as Pettit points out in his concept of protected speech, the contextual conditions that make this recognition of discursive status possible ought to be a key concern in the pursuit of freedom.
In their analysis of the crucial link between speech and status, Lance and Kukla 32 provide a model for understanding a specific set of second-person calls that are considered directly dependent upon the speech-acts that institute them. Here, calls such as requesting, entreating, and ordering, and the speech-acts that make them possible, are considered appropriate or inappropriate depending on the context in which they take place, on the role relations involved between parties, and the power relations at play between them, among other factors. A 'call', on this account, will thus depend upon and make use of existing normative contexts and roles for their production, and call into being new relations and statuses that revise the normative structure of social space and action. 33 To explain how such normative statuses are invoked when a call is made, Lance and Kukla 34 distinguish between the 'input' and 'output' of a speech-act. The 'input' includes those conditions that make the act possible, such as the entitlement of the speaker to perform such an act, and the myriad surrounding contextual conditions that provide the act with the appropriate shift in normative status. The 'output' is the new status that the speech-act has created. For example, when a friend requests that I water her plants while she is abroad, the input required to make the request includes an appropriate existing relationship between us, that I live within a reasonable distance from her house to carry out such a favour, that she actually is going abroad, and that she has plants that need watered, among many other things. When she makes the request, a normative shift takes place, providing me with a range of appropriate responses. Provided a speech-act is felicitous, then, the call that one makes to me when they address me with a particular utterance draws me in to a normative status with them. The kind of speech-act one uses thus narrows the range of potential kinds of uptake that can result.
Important for the account presented here is that different sorts of calls place different sorts of obligations upon those to whom they are addressed. Requests, as opposed to orders, offer addressees the freedom to decide to fulfil the action to which the request is concerned. Were I to agree to carry out my friend's request, it would then be normatively appropriate for her to express gratitude for my help. In contrast to requests, orders place on addressees an obligation to comply. Once someone issues an order to me, the normatively appropriate response is one of compliance. If I do not comply, the orderer can hold me accountable in some way appropriate to the order itself. More specifically, the speech-act involved in an order is only considered successful when my compliance was motivated by the reason instantiated by the order.
While all calls require the (at least partial) fulfilment of certain contextual conditions for the speech-act to be considered successful, there are many cases where an individual's attempt to employ standard communicative conventions becomes distorted in some way, such that the normative status created is at odds with what would ordinarily be expected. Along such lines, and building on the framework presented above, Kukla 35 advances an account of what they term 'discursive injustice', understood as a 'specific sort of discursive incapacity' 36 that takes place when a speaker's belonging to a particular group systematically prevents her from engaging with discursive conventions in the normal way. 37 In other words, a speaker who suffers from discursive injustice will be unable to perform certain speech-acts, despite the satisfaction of the required contextual conditions, because others have identified her as belonging to a particular group. In summary, discursive injustice occurs when our loss of control over our speech comes from our inability to mobilize conventions in the standard way, resulting in a failure of agency that tracks and enhances social disadvantage. 38 Building on the input/output pragmatic structure of speech-acts, Kukla here stresses the crucial role played by the audience in the dissemination of a speech-act. The performative force of a speech-act, and the shift in normative status that it creates, according to Kukla, relies not just on the satisfaction of contextual conditions on the part of the speaker, but also on the response of the hearer or addressee. It is therefore not possible to determine the performative force of a speech-act until some time after it has been completed.
Kukla's argument, then, works in the following fashion: 1. A complex web of practices, conventions, and norms mean that a speaker produces speech-act X in context Y. 2. When the speaker utters the appropriate words in the correct fashion in context Y, then she will be given uptake as having performed speech-act X. 3. Those who witness the performance of speech-act X by the speaker will respond in the normatively-appropriate fashion, thus completing the performance of the speech-act and providing it with the appropriate output.
In cases of discursive injustice, however, this process is distorted because the speaker is perceived as belonging to a particular social group. If I occupy a social position of this sort, then my attempts at employing discursive conventions may frequently fail in important ways, a process that Kukla refers to as 'a queering of the path between performance and uptake'. 39 If I am victim to this sort of distortion, it can have a profound effect on me, as, in such a scenario.
I have become, as it were, the victim of a kind of pragmatic breakdown, from my point of view; I cannot marshal standard conventions in the standard way, in order to act autonomously as a discursive agent. 40 Important to note here is the notion of entitlement to perform a particular speechact. Here, entitlement forms part of the conventional and ritualistic network of requirements that must be fulfilled in order for a speech-act to have appropriate uptake. For instance, where a speaker is not entitled to pronounce a couple as legally married, then the act clearly fails or is infelicitous. Entitlement is also bound up with perceived identity and membership of social groups. In cases of discursive injustice, then, formal entitlement does not always equate to normative entitlement, and as thus is reliant on the social norms and prejudices present in the context in which the utterance takes place. In an example used by Kukla, 41 a female manager's orders issued to the staff in a male-dominated workplace are perceived not as orders but instead as requests. In such an instance, the manager is formally entitled to issue orders, and the staff know that she is entitled to issue them. However, because of her gender, the male staff, who themselves have little experience with women as order-issuers in this context, do not perceive her call as generating the kind of normative obligations conventionally attached to orders. As they perceive her call instead as a request, then the corresponding uptake generated will not be an obligation to comply, but instead will offer them the voluntary choice of acting in accordance with the request, itself generating gratitude if they comply. When uptake unfolds in this way, and where the manager fails to express gratitude for the request being carried out, then the staff may perceive her as a 'bitch'. As Kukla notes Crucially, whether someone is performing an order or a request is not determined primarily by surface grammar, but by the way her act, in all material texture, is situated within a social narrative that takes place in a context that is structured by normative relationships of authority, friendship, and so forth, as well as a rich network of discursive conventions. 42 The manager in this scenario will have her discursive agency negatively impacted in important ways. She will naturally be thought of badly by her charges, and her competency will be considered low in her role as manager. Were this socially-constituted limiting of discursive agency expanded to other areas of the woman's social life, as it generally and unfortunately often is for women, then we can see how she is suffering not just from a discrete act of frustration, but from a systematic kind of injustice.
Another significant way in which one's social group status leads to discursive injustice is where an affected individual's utterance is perceived as an entreaty into a discursive game to which they are already entitled to participate. 43 As mentioned above, my entitlement to perform a particular speech-act depends on whether or not I hold the relevant recognised authority to do so, and provided certain ritualistic and conventional standards have been met. Part of these standards include being recognised as a relevant participant in the communicative exchange. As Kukla articulates In order to issue a speech act with a particular performative force, I have to be embedded within a normative relational space; I have to be recognizable, in situ, as a player in the relevant discursive game. 44 An entreaty, then, is a call to be included into a particular discursive game. Along such lines, the gendered social position occupied by women means that there are many contexts in which their contribution to a discursive exchange may be given uptake as an entreaty to the exchange, rather than as the participation in something to which they are already entitled. This is a common phenomenon in professional settings, where a woman's expert opinion can often be undervalued or dismissed. 45 Kukla's analysis of discursive injustice ought to be of particular concern for republicans for the following reasons. First, and as Pettit points out, if one's status as a speaker holds fundamental importance for one's status more generally, then it makes sense that systematic distortion of one's utterances severely limits discursive agency in a way that prevents one from enjoying freedom as non-domination. Quite simply, if non-domination requires securing the ability to stand on an equal discursive footing with our fellow citizens, then communicative frameworks that disadvantage certain social groups such that they cannot employ speech conventions on a par with the rest of society will necessarily be in direct conflict with the enjoyment of freedom in the rich, republican sense. So, if republicanism is to be an effective source of critical social analysis, it must be capable of identifying and responding to the kinds of norms and structural forces that prevent certain individuals from engaging in effective contestation in the first place. If social change is to come from the 'ground-up', as republicans have long endorsed, then it is imperative that we examine those roadblocks that undermine the capacity for marginalised voices to be heard and understood.
In its current formulation, however, Pettit's preferred account of protected speech overlooks those norms that distort one's speaker status in the way presented above. 46 There are two potential reasons for this. First, Pettit argues that the normative thrust of non-domination is a concern for the discursive status of all individuals in society, and it is this concern that drives his endorsement of protected speech over unhindered speech.
Discursive status refers to the status we enjoy when we are recognised as discursive agents capable of giving and being addressed according to reasons. Under protected speech, rather than unhindered speech, we are thus secure in our status as speakers. We do not speak (nor remain silent) out of fear or deference, and, in that sense, others can attribute our speech to us, thus holding us responsible for what we say.
In the case of the female manager, however, her inability to employ standard communicative conventions is not the result of a denial of discursive status, and yet she nonetheless suffers from a severe lack of agency in a way that tracks social disadvantage. In this case, the sexist norms at play in society, and, in particular, her male-dominated workplace, are such that there is a 'queering' in the way that her utterances receive uptake. And, while none of us will ever fully be in control of the kind of output our speech-acts have, the kind of systemic failure at the heart of this case does indeed reflect a certain kind of domination-based injustice. A similar phenomenon is at play in Jennifer Hornsby's account of 'disempowered' speech, where she explains: To the extent to which someone is disempowered as a speaker-to the extent to which there are things that that person cannot do with speech-her right to free speech is to that extent unexercisable. 47 Can a republican free speech framework incorporate a concern of this sort? I argue that republicanism, suitably calibrated along critical lines, offers an attractive conceptual framing of free speech that overcomes the limitations both of the liberal paradigm and of Pettit's 'protected' approach. Once we re-frame free speech debates to include discursive justice, then our conceptual resources for understanding both the normative and political implications of speech exclusion become apparent. Incorporating discursive injustice into a critical republican framework thus provides an alternative to the paradigm liberal framing by taking seriously the role of systemic power differentials in speech exchange. The result is a re-centring of the central relationship between discursive status and one's position as an equal speaker.

A Critical Republican Concept of Free Speech
One of the clearest and most compelling accounts of critical republicanism can be found in the work of Dorothea Gädeke. Gädeke's argument for critical republicanism begins with an important question: what is distinctive, from a normative perspective, about the republican conception of freedom as non-domination? 48 According to Pettit, the crucial distinction between liberal unfreedom as interference and republican unfreedom as domination comes from an understanding of domination's effect on equal standing. 49 This close connection between standing and non-domination stems from the discoursetheoretical commitments underlying Pettit's account. As we have seen, Pettit argues that freedom depends in large part on the relation between individuals as discursive subjects. In order to say that we enjoy equal status, on this account, we must be treated as capable of exchanging and giving reasons to which we are then held responsible. On this account, the practical norms within which we engage in discursive exchange must therefore uphold our status as equal interlocutors. In an asymmetrical relation of domination, then, the dominated will not have discursive control in the way required in order to make freedom possible. This condition has serious consequences for one's ability to command respect from others. When one does not have discursive control, their contributions to the space of reasons will be deemed unreliable, and they will not be able to engage in reason-giving in a way that gives rise to a symmetry in standing. The status one has as an equal interlocutor is thus at the centre of Pettit's account of non-domination. To use the famous slavery example, the badness of being in a state of slavery (even when under the control of a benevolent master) can be described as a denial of one's voice. When I am under the control of another, my words are not taken as my own, I cannot be considered responsible for them, and so I am considered untrustworthy in the eyes of others as a credible exchanger of reasons. And, without such trust and credibility as an interlocutor, I cannot be said to enjoy equal standing among others.
According to Gädeke, 50 however, Pettit's focus on discursive control lacks an important degree of critical self-reflexivity. Under Pettit's practical rationality, individuals with equal standing are recognised as capable of engaging in and being held responsible for the norms to which they are subjected. In Gädeke's alternative argument from practical justification, one that draws heavily on a Habermasian account of discourse theory, equal status must also require that agents are considered the authors of the norms to which they are subjected. 51 What implications might Gädeke's account have for our understanding of nondomination? Gädeke recognises that, under Pettit's proposed aims, a stronger discourse-theoretical account of normative validity is not necessary. This, as Gädeke argues, is a crucial mistake for those concerned with the status afforded to those who are secure from domination. On this point, Gädeke writes that being subjected to norms whose validity we cannot challenge means that we lack a fundamental discursive power, namely the power to shape and interpret our relations to others and thus our own normative status. 52 In light of this oversight in Pettit's account, Gädeke thus proposes that we re-centre the normative core of non-domination from discursive control to normative authority. This shift reveals important shortcomings in Pettit's account once we apply it to relevant examples. In the slave case, we saw that, in Pettit's version, the asymmetry in standing suffered by the slave undermined how others recognised them as interlocutors capable of giving and being held responsible for reasons. Instead, on Gädeke's account, the state of the slave is such that they are denied authorship of the rules and norms to which they are subjected. And it is this state, according to Gädeke, that explains the problematic nature of dominating relationships. 53 The critical potential of Gädeke's revision of the normative core of non-domination from discursive control to normative authority proves especially salient when we examine the issue of discursive injustice described above. In the example of the female manager, the 'queering' of speech that takes place, which consequently limits her discursive agency, occurs despite the fact that she enjoys discursive control in the way that Pettit's account demands. On Pettit's version of protected speech, there are suitable obstacles in place that limit the opportunity for others to interfere with the speech options of individuals, and which provide them with the ability to 'walk tall' among others. Presumably, the manager in question works in a company that provides a range of formal protections against gender discrimination. It is also considered a central part of her job role to issue orders and demands upon her staff, and to provide suitable consequences for those who fail to comply. It would thus be inaccurate to suggest that the 'queering' of output of her speech was due to a lack of discursive status. The norm that renders her speech as requests rather than demands, then, reflects a lack of normative authority in this context. She is recognised as a giver and exchanger of reasons, and is considered to be held responsible by the norms of the workplace and according to her role. But her gendered social position denies her the normative authority to act as an author of the norms of this role and context, thereby undermining her status. And it is only when we adopt Gädeke's critical account of republicanism that this issue comes to light.
But should this kind of queering be of concern for non-domination? Gädeke stresses the crucial importance of normative authority by invoking the example of the slave. On Pettit's account, the problematic status of the slave is such because he is denied the status of an interlocutor in the exchange of reasons. On Gädeke's richer account of normative authority, however, the slave suffers from a lack of status not only because they are not recognised as someone to whom certain norms can be addressed, but because they are 'denied the normative power to challenge the rules of their interaction'. 54 On Gädeke's account of practical justification, then, speakers must not just be recognised as credible speakers, but as co-authors of the norms to which they are subjected. In the case of the female manager who suffers discursive injustice, then, she is denied normative authority in the way outlined by Gädeke's account, and her voice is limited in a way that reflects domination.
Gädeke's proposed shift from discursive status to normative authority also has significant implications for how we understand how power gives rise to domination. On Pettit's account, power is a kind of disposition possessed by those who have the capacity to arbitrarily interfere with others. 55 When one is dominated on this account, they need only be subjected to another who holds this disposition, no matter if they choose to use it to interfere with their choices. Those who are dominated on this account often do so along group-specific lines. For instance, when women are not sufficiently provided with the protections necessary to withstand gender-based power from men, then we say that they suffer from domination. According to Gädeke, however, this dispositional account of power presented by Pettit overlooks important structural features of how power gives rise to domination. On Pettit's account, power is understood in terms either of actual actions, or of anticipated future actions. However, this action-based understanding of power fails to align with our status-based understanding of domination due to the fact that it conceives of domination in terms of interactions. In order to fully grasp the role that power plays in domination, then, we must instead look at what it is about the social standing of the powerful that allows them to dominate. In other words, from where does this power come?
To overcome this problem, Gädeke argues that, in order to make sense of capacities to interfere, we must bite the bullet and conceive of all domination as a structurally-constituted form of power. 56 To illustrate, Gädeke points out that, in the paradigmatic case of slavery, the capacity to interfere stems from a structurallyconstituted system of legal norms, upheld by a complex set of social norms and practices. In other words, the source of the interference in slavery does not depend on a particularly extreme instance of individual power over another. Instead, the capacity of the slave master to interfere with the enslaved is a robust one, in the sense that it is a durable capacity that is constituted and reproduced by norms and practices which create the very positions of power and disempowerment that we refer to as slavery. 57 Conceiving of domination in this structurally-constituted way means that we can now describe it as 'the power structure that puts the dominator and the dominated into asymmetrical positions to one another', 58 in other words, as a denial of equal status. Importantly, this structural form of power also works for those examples of domination that do not rely on legal structures. To use Gädeke's example, racialized brutality and implicit bias among US police, whilst explicitly illegal, nonetheless is based on a set of norms and social practices that continue to reproduce dominating power relations that systematically disempower blacks. And it is precisely this sort of informal, structurallyconstituted form of power that gives rise to discursive injustice under Kukla's account.
Once we conceive of domination as always structurally-constituted, we immediately see a problem with Pettit's endorsement of protected speech when faced with the realities of discursive injustice. Important for Pettit's argument for the benefits of protected speech over unhindered speech was the status benefit. Under protected speech, on this account, we can say that we have free speech when we live under laws, and supportive norms, that protect the exercise of one's voice in a publicly-recognised way. As we saw above, this account of speech is clearly connected to Pettit's understanding of discursive status as constituting the normative core of non-domination. The asymmetrical and systematic power relations between men and women in the manager case is, we can say, the source of the domination in that scenario, and not the fact that, say, the men outnumber or are physically stronger than the female manager.
Why is Pettit's account of protected speech insufficient for identifying cases of discursive injustice like the one described in the manager example? Pettit's concept of protected speech recognises the need for a sphere of protection as a speaker in order to enjoy equal status. In order to have one's words reflect your own true thoughts and opinions, we must be able to say what we truly think without fear of excessive penalty or sanction. Without this, we cannot enjoy discursive agency as we do not stand in a symmetrical relationship with others in society. In the female manager example, however, the woman is recognised as an agent capable of normative reasoning with her staff and is entitled to issue certain normative claims upon them, relevant to her capacity as their manager. She is also provided with the formal protections required to do this. Despite this, however, her discursive agency as it relates to her status as a speaker continues to be systematically undermined, and this is something that Pettit's account of protected speech severely overlooks.
In addition, Gädeke argues that, on her account of domination, we must also look to how those who are dominated are systematically disempowered. 59 This disempowerment can be described as the set of norms and social practices that both empower the dominator and disempower the dominated at the same time, and occurs even in those cases where there does not exist a direct, interpersonal relationship of domination between two or more persons. We can say that the manager suffers domination in this case as she does not enjoy normative authority in the way that she should. Again, while there may be institutional and workplace regulations in place that protect her when she chooses to speak (discursive control), her inability to use communicative conventions in the correct way represents a lack of normative authority. As noted by Kukla, however, not all instances of communicative failure constitute discursive injustice. The contextual conditions of a speech situation, including the role that one's perceived identity plays in the resultant output of an utterance, are such that there are a variety of ways in which failures can occur. Where such failures reflect a discursive injustice, however, we can say that a group of speakers has been systematically disempowered in their speech in a way that undermines discursive agency and which thereby tracks social disadvantage. However, as we see from Kukla's example, the female manager's inability to use speech in a conventional way does reflect a structure of power that disvalues the role of women in positions of authority, particularly in the workplace. In this case, there is a structural and gender-based asymmetry of power in place that systematically empowers the voices of men whilst disempowering the voices of women. As we can see, while Pettit's dispositional account of power cannot adequately explain the structural nature of this asymmetry in status, Gädeke's critical framework exposes the inherently structurally-constituted nature of this kind of discursive injustice. Pettit's account understands power as a capacity to interfere, which is not what is going on in cases of discursive injustice of this sort. Under discursive injustice, no actual or potential interference needs to take place. Instead, the dominated suffer from an asymmetry in status which systematically empowers the speech of the powerful whilst disempowering their own speech.

A Critical Republican Response to Discursive Injustice: A Brief Sketch
In this article, I have explored one way in which standard liberal accounts of free speech are insufficiently sensitive to the effects of social norms on one's status as a speaker. I have also shown how, while Pettit's recent formulation of 'protected' speech makes a considerable improvement on the prevailing 'unhindered' accounts endorsed by liberal theory, his agential understanding of domination seriously overlooks the effect of social norms on the communicative capacities of affected groups. In short, by shifting our understanding of domination to include a concern for normative authority, we can go some way to identifying and responding to those systemic threats to speaker status that significantly threaten the agency of certain individuals.
So, what does a critical republican conception of free speech actually look like? I argue that such an account must contain two strands: one, as a lens through which to conduct critical analysis, and two, as a guide for institutional and individual behaviour. On the first count, and following Pettit, the paradigmatic liberal framing of free speech conflicts is rejected as offering a normatively insufficient understanding of free speech as an issue of securing maximum choice of speech options. Instead, free speech as an ideal is considered valuable to the extent that it secures a particular status for speakers, and the status benefits of speech can only be realised once certain conditions obtain. This requires that we critically examine, on a case-by-case basis, the surrounding contextual conditions, including institutional and regulatory mechanisms, networks of norms and practices, and the relevant authority relations in each speech scenario for their role in securing the statusspeaker element of non-domination. This will require both an analysis of the way in which individual agents interact with one another, including the way that they endorse or (knowingly or unknowingly) perpetuate norms that contribute to discursive injustice and an account of how structural social arrangements give rise to discursive injustice in the first place. Ensuring that we as individuals are aware of, and attempt to correct for, discursive injustice in our exchanges with others is thus only one part of the puzzle.
What we want to avoid, then, is the belief that we can 'reason' our way out of discursive injustice simply by exercising virtue when engaging in communicative exchange with members of oppressed social groups. The large-scale shift in norms and attitudes that inform speaker status requires examining both interpersonal and structural relations for their role in propping up social norms. The critical republican account of free speech presented here, then, recognises that speaker status is a function of one's social standing more generally. In contrast to the liberal atomist agent found in traditional understandings of free speech, a critical republican account of free speech acknowledges that one's ability to behave as an equal, discursive agent is only as good as the norms, practices, and attitudes through one's speech operates.
In addition, and moving beyond Pettit's account of discursive status, we must assess speech conditions for their ability to uphold the normative authority of speakers. Discursive norms must be subjected to reflexive critique to ensure that speakers are recognised not just as addressees of relevant norms, but that they are the co-authors of the norms themselves. Norms and practices that prevent members of particular groups from employing communicative conventions in the standard way are identified here as threats to normative authority by severely inhibiting the possibility of those affected from challenging the validity of these discursive norms. A critical republican account of free speech, as one concerned with the close relationship between speech and status, will also conceive of domination as structurally-constituted. The kinds of formal and informal protections proposed as ways of securing one's speaker status, then, will aim to eradicate those norms that simultaneously and systematically both empower certain speakers whilst disempowering others. The structural account of power presented here also allows us to distinguish between problematic cases of discursive injustice that require our critical attention, and those cases of communicative 'queering' that constitute a frustrating, yet non-problematic, aspect of the lives of all speakers.
The full extent of the problems found in Pettit's proposed account are revealed when we consider how we might tackle discursive injustice using his theory. For the most part, the queering of communication in a way that tracks social injustice is a function of certain norms in society that undermine the equal standing of affected groups. For women in traditionally 'male' professions, a woman can be disempowered as a speaker, despite enjoying equal formal opportunity in the workplace, employment rights, and rights securing her against sexual discrimination and harassment. Women here may also work with colleagues who fully endorse equal rights for women and are explicitly committed to securing employment opportunities for women in that particular profession. However, as Kukla's analysis illustrates, women in such positions still frequently suffer the effects of discursive injustice, mainly due to the ongoing persistence of social norms that attach certain expectations to women, often in hidden and implicit ways. Here, Pettit's proposed 'bottom up' solution to shifting norms of sexism will only go so far. The problem with relying on this method of social change, as we have seen, is that women's contributions to contestation will necessarily be skewed because of their relative social status, thereby severely limiting the potential for their claims to be taken seriously. As Pettit is concerned with those things that limit the potential for individuals to secure themselves against domination, he should also take issue with disempowered speech. This need not commit him to securing against such domination using legal intervention. Instead, a republican proposal ought to commit to identifying and implementing those arrangements that provide for all citizens the opportunity to engage in the collaborative effort of norm contestation, critique, and ongoing development. While the state has a clear role to play here in providing policies and avenues through which such contestation can take place, there are many ways that they can do so without falling prey to the problem of state domination itself.
One major problem, then, will come from identifying where particular norms are actually a result of discursive injustice, and which simply reflect the regular sort of queering. While marking a clear distinction between the two will not always be so simple, we can say that by shifting our understanding of free speech from an 'unhindered' account to one that aims to preserve speaker status, we are at least engaging in a kind of systematic critique that has been seriously lacking in discussions about free speech. Highlighting that such a phenomenon exists, and using it to tentatively guide our analyses, will hopefully bring to light many kinds of hidden speech injustices that have long existed under the radar of social critique.
Shifting the framing of free speech claims from one of maximal non-interference to one of speaker status comprises the central aim of a critical republican approach to free speech. The task of securing such a speaker status for all members of society, however, will be a complex and wide-ranging process. Beyond social critique, there are several practical ways in which discursive injustice can be curtailed without falling victim to arbitrary state interference. Chiefly, removing the threat of discursive injustice requires changing the norms that lead to the systematic disempowerment of the speech of historically-oppressed groups. Debates about how to shift attitudes and norms that undermine the equal status of oppressed groups have occupied feminists and race theorists for centuries, and I do not have space to fully explore their arguments here. Instead, I will briefly examine several different tracks that such a project must take in the context of speaker status specifically, at least in the first instance.
The first route to eradicating pervasive discursive injustice comes from ensuring that affected groups enjoy full and equal access to a set of basic civil rights and liberties that provide them with the opportunity to exercise their voice in political and civil matters effectively. While perhaps a rather obvious point, the continued involvement and contribution of women in local and national politics, as well as positions of influence in civil society more broadly, has over time been shown to shift societal perceptions of citizenship away from the historically white, male paradigm towards one which is reflective of the actual identities of society's members. Re-defining what it means to be a citizen will have a direct impact upon how one's discursive contributions are received in public deliberations, thus allowing previously-excluded groups to engage in the re-negotiation of deliberative norms in a way that secures speaker status for all.
A second way in which discursive injustice is combatted is by supporting measures that publicly counteract those harmful and demeaning stereotypes about affected groups that undermine their position as equal speakers. Representations of women and minority groups in the press, media, and entertainment industry that rely on such stereotypes directly perpetuate social norms that undermine such groups' speaker status. These representations need not satisfy the legally-defined category of 'hate' speech, but will often perpetuate stereotypes of women as, for example, objects of desire and men as suited to positions of power and authority. The subtle ways in which these stereotypes feed into broader social understandings of men and women clearly will have an impact on how offscreen women are perceived as discursive agents. Conversely, the positive representation of men and of white individuals on screen and in the media also indirectly empowers individuals from such groups as speakers in a vast range of social contexts. In addition, and apart from combating the most egregious forms of hate speech and propaganda, one of the tasks of contesting exclusionary speech norms will also require intervention and exposure at the level of norm enactment. In order to be successful, this requires that individuals actively reject attempts at norm enactment where they arise in both our everyday interactions and when we encounter them in the media and online. Where the source of norm enactment comes from speech itself, contestability can come in several forms. For example, and as articulated by Langton, 60 a speaker's authority can be questioned in a particular speech scenario, thereby potentially halting the process of norm enactment. Questioning or undermining speaker authority works to weaken the credibility of the words spoken, towards the aim of limiting the force they have to shift or entrench social norms.
How one is perceived as a speaker will also be influenced by non-intentional, structural features of the social world that attach certain stereotypes and meanings to particular roles in society. Building on Pettit's example of free speech within the university, it is clear that the norms and customs of the academy determine the level of empowerment one has as a speaker within their university. Pettit, in his defence of protected over unhindered speech, argues that the excessive power held by administrators and upper management unjustly impinge upon the free speech of academics by causing them to self-censor for fear of sanction or expulsion from their workplace. As I have shown throughout this article, however, even where such protective measures are in place, we cannot confidently say that all academics are equally free as speakers. It is widely known that, in the vast majority of disciplines, women and people of colour are underrepresented in the academy. Because underrepresentation is itself a key feature of persistent discursive injustice, then one of the main avenues for eradicating discursive injustice in the academy must be to increase representation from such groups. Jennifer Saul, in her influential paper 'Implicit Bias, Stereotype Threat, and Woman in Philosophy' (2013), explores the issue of underrepresentation of women in academic philosophy by reference to the psychological phenomena of implicit bias and stereotype threat. As she argues 'there is a tradition in philosophy of associating reason, objectivity, and philosophical thought with maleness, and emotion, subjectivity, and the non-philosophical with femaleness'. 61 If it is the case that objectivity and reason are closely associated with what it means to be a 'good' philosopher, then contributions that female philosophers make to communicative exchanges or debates in the academy will be skewed by the effect of discursive injustice. This also holds for other forms of social life: societal perceptions of the ideal citizen, manager, politician, and so on will, implicitly or explicitly, influence how historically excluded groups who now occupy those roles will be taken as interlocutors when they attempt to exercise speech.
The problem, however, is that the phenomenon of discursive injustice can discourage women from doing one of the main things that will serve to change social norms and attitudes surrounding women in such roles: taking up or pursuing such roles in the first place. Ensuring that women take up such roles requires not just removing barriers to entry, then, but by actively recruiting and encouraging them to take up such roles. Crucially, women must be involved in the process of shaping the meaning of these roles, rather than simply fulfilling the expectations imposed upon them to fit the predetermined mould of what such a role-occupier looks like.

Conclusion
Those who study the effects of unjust social norms on an agent's ability to use words effectively have long expressed frustration with the standard liberal account of free speech. As this article has shown, the rich conceptual tools offered by the neo-republican revival can go some way to re-framing what it means to enjoy freedom of speech. Here, I enjoy speech not when there are limited external obstacles on my speech, but when my status as a speakeras an equal discursive agentis reliably protected. The solution proposed by Pettit, however, proves insufficiently sensitive to those social norms that distort a speaker's words such that their speech is disempowered. In order to combat this phenomenon of discursive injustice, then, a critical republican account of free speech is endorsed. The critical republican account of free speech presented here provides a method of analysis that exposes structural barriers to equal speaker status, while at the same time limiting the potential dominating effect of arbitrary state interference in matters of speech.