Reverse Versus Radical Discourse: A Qualified Critique of Butler and Foucault, with an Alternative Interactive Theorisation

ABSTRACT

This article explores the concept of reverse discourse, as suggested by Foucault and Butler. It is argued that Butler's concept of subject formation is overly determinist, as is Foucault's of discourse. Following Scott's critique, it is argued that there is a strong and a weak conceptualisation of dominant ideology. Discourses are in competition for authority, where dominant ideology is the discourse of more powerful decision-makers, while subaltern ideologies persist. This leads to a more interactive theory of structural constraint and the conditions of possibility for radical action. Social actors can change power relations by reproducing dominant discourses while reversing implied power-authority relations – reverse discourse. Alternatively, more radically, they can resist dominant ideology by attempting to build consensus around subaltern ideology, which is incommensurable with dominant ideology. Reverse discourse has the advantage over radical critique in that it reproduces the natural-order-of-things. However, it has the disadvantage of reproducing reifying norms.

KEYWORDS:

• Reverse discourse
• radical discourse
• dominant ideology
• resistance
• Judith Butler

In social theory the concept of reverse discourse has emerged as an important conceptual tool for the analysis of strategies of resistance (Lilja 2021). The intellectual origins of the concept lie in the work of Foucault and Butler. In this article I wish to qualify the concept, by arguing that it is one of two strategies of resistance. While I fully accept that Foucault and, especially, Butler have left us a valuable conceptual tool, I argue that they underestimate the possibility of more radical resistance, due to an implicit intellectual legacy that overplays structural linguistic-cum-cognitive determinism of social subjects and underplays creative agency. The article opens with a short qualified critique of Butler, which is followed by an illustrated comparative discussion of reverse versus radical discourse, which is theoretically different from their perspective.

In terms of structure, the first sections critique Butler and Foucault. Following this, the article explores the distinction between a strong and weak sense of the concept of dominant ideology, arguing that Butler/Foucault represent the strong version, which is theoretically problematic. This moves us to a weak version of dominant ideology, endorsed by this article, where the primary source of structural constraint is the response of the interacting other, to performances of authority. The discussion moves on to illustrate unusual examples of the counter-hegemonic discourses; including astrology, homeopathy and extra-terrestrial spacecraft. This takes us to the contested edges of authority, where it often fails to elicit confirmation from other. It doesn’t fail because of internal constraint, or strong dominant ideology, as in Butler/Foucault, but weaker dominant ideology, whereby social action fails as interaction to enrol the powerful others. Yet, simultaneously that same action may succeed in enrolling less powerful actors, including fans of astrology and UFO spotters. The article concludes with examples of feminist social actors, choosing between radical critique and a more conservative strategy of power.

The article has a developmental structure: it opens as critique but gradually develops into an alternative conceptualisation of dominant ideology and the constraints upon agency and creative thought. With respect to the critique, the readers should understand that, while the focus is upon Butler and Foucault as named theorists, the objectives are wider. The target is overly determinist and subject-oriented, as opposed to interactive, ways of thinking about the creative agency, external constraint and dominant ideology. So, in a sense Butler and Foucault are used as ideal types of a prominent way of thinking in social theory. Consequently, I may, at times, overdraw dichotomies of perspective. This means that, no doubt, statements can be found, made by these authors, which do not correspond exactly to the ideal types constructed. I apologise for this, and ask the reader’s indulgence, in the interests of moving social theory forward toward new critical perspectives on reverse and radical discourse.

The contribution to social theory of this alternative framework is that we move away from over-determinist theories of the social subject. Actors are not characterised as epistemic dupes, as suggested by most theories of dominant ideology. Yet this is not voluntarism, where the individual is unconstrained. The theoretical significance of performative failure is used to give an account of dominant ideology in a weak sense. Interactive failure is not absolute, social actors create epistemic communities where their order-of-things has resonance. The implications of this are that there are two strategies of contesting dominant ideology. The first is reverse discourse, where actors use dominant ideology in subversive ways, which has the significant strategic advantage of appealing to the natural-order-of-things of powerful actors. The second strategy is radical discourse, where an alternative epistemic interpretative horizon is put forward as reasonable. However, this will almost invariably be rejected as infelicitous, as unreasonable, by dominant authorities. Consequently, radical actors then have to engage in a process of enrolment, slowly drawing powerful others into this interpretative horizon, as the natural-order-of-things. Radical discourse contests the reifying essentialisms that are used to shore up dominant ideology, and therefore constitutes more profound resistance. However, the chances of failure are much higher than reverse discourse. So, in many instance, reverse discourse makes greater strategic sense than radical discourse, although it is normatively less satisfactory.

Butler on the social subject and reverse discourse

In History of Sexuality (1981, 101) Foucault describes the proliferation of nineteenth century discourse of perverse sexualities, including “inversion”, “pederasty” and “psychic hermaphrodism”. He argued that this proliferation “made possible a ‘reverse’ discourse: homosexuality began to speak on its own behalf, and demand that its legitimacy or ‘naturality’ be acknowledged, often in the same vocabulary, using the same categories by which it was medically disqualified”. The concept of reverse discourse is developed further by Butler in The Psychic Life of Power (1997a), which is grounded in a conceptualisation of the social construction of social subjects.

Butler conceptualises reverse discourse relative to subjection, which is influenced by Foucault’s article “The subject and power” (Foucault 1982). Foucault’s article opens with the following assertion: “My objective, … has been to create history of the different modes by which, in our culture, human beings are made subjects. My work has dealt with three modes of objectification which transform human beings into subjects”. (Foucault 1982, 208).

Echoing these words with an important theoretical difference, Butler writes that “[s]ubjection signifies the process of becoming subordinated by power as well as the process of becoming a subject”. (Butler 1997a, 2). Note the replacement of objectification with the abstraction power. Butler argues that central to the creation of subject position is both the Althusserian process of interpellation and Foucault’s theorisation of “discursive productivity” (Butler 1997a, 2). Interpellation refers to Althusser’s (1970, 86) example of a passer-by who is hailed by a police officer’s authoritative voice and responds, whereby “recognition is proffered and accepted, and interpellation – the discursive production of the social subject – takes place”. (Butler 1997a, 5). Butler acknowledges that in Foucault’s later work he insists that “the subject is not ‘spoken’ into existence and that the matrices of power and discourse are neither singular nor sovereign in their productive action”. (Butler 1997a, 5). Contrary to Foucault, for Butler the subject is discursively constituted, which leaves no conceptual space for the individual to transcend discourse. Butler argues that the subject, rather than “the individual, ought to be designated as a linguistic category … ” (Butler 1997a, 10). Furthermore, that it makes “little sense” to treat the individual as “an intelligible term” (Butler 1997a, 11). Rather, the social subject is intelligible “only to the extent that they are, as it were, first established in language”. (Butler 1997a, 11). This leads to an account of the subject that she admits to being “inevitably circular” (Butler 1997a, 10), whereby the subject presupposes the discourse that constitutes the social subject. The social actor expresses itself as a subject through the categories of meaning by which it is constituted. Following Althusser, Butler argues that the mastery of language is the subjection to “ruling ideology”, whereby mastery and subjection are in a “paradoxical simultaneity” (Butler 1997a, 116). This means that any attempt at agency only takes place in the language of subjection, and therefore “the agency of the subject appears to be an effect of its subordination” (Butler 1997a, 12). There is no individual in-itself; rather a social subject that is discursively constituted by linguistic categories. Consequently, Butler argues that “social categories signify subordination and existence at once”. (Butler 1997a, 20). Within this framework, “[p]ower is both external to the subject and the very venue of the subject”. (Butler 1997a, 15).

The assertion that the social subject can only come in touch with the world-out-there through speech is not the same as arguing that the world-out-there is made of speech. As Butler argues with respect to the human body, while we might accept “the proposition that the body is only knowable through language, the body is given through language, it is never fully given in that way … ” (Butler 1997b, 5 – italics original). The language through which the body is known “does not fully or exclusively form it” (Butler 1997b, 5). While linguistic social construction means the social body is made by language, it is not made of language (Butler 1997b, 4), which is an important point, as Butler’s social constructivism does not suggest the total absence of materiality. In sociological theory terms, while the social subject, the cognitive being, is formed through speech, the social object, which the subject understands through language, is not a linguistic entity in-itself. Both the physical body and the world-out-there have a physical facticity that exceeds language. However, the cognising subject is discursively constituted.

Since being a social subject implies subjection to discourse, thus domination, one might expect social actors to refuse subjection. To explain social subjects’ complicity in self-domination Butler grafts an ontological-cum-psychoanalytic account onto the Foucauldian and Althusserian paradigms (Butler 1997a, 87). The subject’s attachment to its possibility of existence is likened to a desire for and passionate attachment to the norms constituting subject formation. Ontologically, the being-in-the-word of the social subject is constituted through the discourse that subjects it. Therefore, the only vehicle for the social subject to express its being-in-the world is through the reproduction of the language of subjection. The social subject “pursues subordination as the promise of existence”. (Butler 1997a, 2). The subject has no being-in-the-world outside the language that constitutes its sociality, therefore “to affirm one’s existence is to capitulate to one’s subordination – a sorry bind”. (Butler 1997a, 79). Hence, the desire to be, rather than not to be, binds the social subject into expressing its being-in-the-world through the discourse that defines and oppresses it (Butler 1997a, 7). The social subject is “the compelled consequence of a narcissistic attachment to one’s continuing existence”. (Butler 1997a, 113). When hailed, the social subject responds out of an understanding that the “I” that grounds its existence is dependent upon its constitution as a subject in the norms that are constitutive of meaning (Butler 1997a, 107). By being given a name the social subject is given the possibility of social existence (Butler 2021, 2).

Once given a social existence the social subject craves it. Butler uses Hegel’s account of the master-slave (bondsman) dialectic as symbolic of this craving. The slave can only exist by giving the fruits of his labour to another, to the master. The slave substitutes him/herself for the body of the master and comes into existence through the other (Butler 1997a, 36). The slave has a desire to labour driven by an “absolute fear” (Butler 1997a, 39), which comes from the realisation that the slave’s only chance of agency is by working on an object that belongs to another (Butler 1997a, 39).

Because Butler interprets the social subject as constituted through discourse, denying non-discursive individuality, this begs the question: how can such social subjects resist? The source of resistance lies in the process of replication. Speech is repeated, and in every repetition, there lies the possibility of slight deviation. Butler writes: “The subject is compelled to repeat the norms by which it is produced, but that domain establishes a domain of risk, for if one fails to reinstate the norm ‘in the right way’ … ” (Butler 1997a, 28). Iterability creates the possibility of “opposing and transforming the social terms by which it [the social subject] is spawned”. (Butler 1997a, 29). The possibility of imperfect repetition creates the possibilities: “[p]ractices of parody” that appear “as derived phantasmatic, and mimetic – a failed copy as it were”. (Butler 1990, 200). Thus, resistance can be achieved through “subversive repetition” (Butler 1990, 201). That subversion can take the form of parody or mimicry or, as I argue later, can take the form of using the logic of existing discourse to subvert the power relations of the status quo.

Critique of Butler

The concept of reverse discourse is significant for understanding resistance. This is qualified critique, I fully accept that reverse discourse is a significant strategy of resistance, while I argue that there is also a more radical strategy whereby the social subject rejects dominant discourse. The radical strategy is not taken sufficiently seriously by Butler because, like Althusser, whom she draws upon, Butler is overly structurally determinist. This has two inter-related aspects: linguistic determinism and an overly strong concept of dominant ideology.

Butler argues that the social subject is constituted through discourse, which is a form of subject-oriented linguistic determinism. For Butler, as social subjects, our categories of thought are constituted through the language we speak. Consequently, there are no individual thoughts beyond language. To clarify, this refers to the position of the social subject, not the object understood/interpreted, which is external to the cognising subject. As Butler makes clear in her 1997b essay, she is not taking the extreme (and obviously false) position that the world-out-there has no essence beyond discourse and I am not suggesting she does. In short, I am not suggesting that she denies materiality, and the alternative account that follows is also socially constructivist. It is the social subject, but not the object, that is linguistically determined. While this subject-relative linguistic determinism constitutes a coherent philosophical position, empirical observation trumps philosophical argument. While all, so-called, empirical facts are theory dependant; a theory does preclude certain events, as outside the conditions possibility. As I will demonstrate, there are forms of social behaviour and phenomena that exist, which should be precluded by Butler’s theorisation of the social subject as entirely discursively constituted.

The social subject, language and thought

As argued by Bloom and Keil in their review of debates on the mind-language relationship, in empirical linguistic cognitive science literature there is a spectrum of opinion on the relationship between the self and language (Bloom and Keil 2001). At one end of the spectrum are linguists for whom thought or concepts follow language, while at the other end concepts and thoughts come first, with language as a conceptual tool that facilitates the expression of thoughts. At the language-determines-thought end were thinkers like Whorf (1956) and Sapir (1921), who argued that very different languages, such as English and Hopi, lead to fundamental differences in thought processes (Bloom and Keil 2001, 352). At the other end of the spectrum is Fodor (1975) and Pinker (1994).

In considering this spectrum, we are not exploring the weak claim that language can have an effect on thought, the question is: are humans capable of thought independently of the words we have been taught? (Bloom and Keil 2001, 354). If Butler is correct that the subject is constituted through discourse, such a theory precludes the existence of a thinking subject (as distinct from physical body) outside of language. It follows that, outside discursive constitution, the concept of the individual who thinks for themselves, non-discursively, is meaningless (Butler 1997a, 10). This is strong language determinism. However, within contemporary cognitive linguistics, the Sapir and Whorf view of language determinism has fallen out of favour, especially since Pullum’s (1989) article “The great Eskimo vocabulary hoax”, which critiqued linguistic determinism using detailed anthropological case studies that revisited the examples used by Sapir and Whorf and demonstrated that they misrepresented differences.

Speaking generally, if linguistic determinism of the social subject were true it would be impossible to think without language. Indeed, the common everyday experience of having an idea and then looking for words to express our thoughts should not occur. Such an everyday phenomenon would be outside the conditions of empirical possibility, as predicted by the theory. This phenomenon occurs because, in social theory terms, the social subject is a creative individual who cognitively exceeds the language available. Similarly, on a more ontologically fundamental scale, persons without language would be incapable of abstract thought, which is not the case. The absence of language is a relatively common phenomenon. Deaf children brought up in neglect and lack of education can grow up without linguistic skills. However, they still think abstractly, as evidenced by their ability to accomplish practical tasks, including solving practical mathematic problems without linguistic numeracy. The point here is not simply that the external reality exceeds speech, as affirmed by Butler (1997b), but that these children seem to understand the world out there without speech, which concerns their interpretative facility as social subjects. An interesting instance, recounted by Pinker (1994, 67) are the experiences of Schaller (1991), who taught sign language to a twenty-seven year old deaf Mexican immigrant. The deaf student had no form of language whatsoever – no sign language, no writing, no lip reading. Schaller recounts teaching the sign language word for “cat”. It was an epiphany, once the deaf student understood the idea that signs stood for concepts, “a dam burst, and he demanded to be shown the signs for all the objects he was familiar with”. (Pinker 1994, 68). The last words are significant: this individual was already a thinking being with concepts that he was familiar with. However, he had no signs for these concepts to facilitate interaction. Within a short time this social subject was able to tell Schaller his life story, which included isolation, disability and extreme poverty. Afterwards, Schaller taught many language-less students from forgotten corners of society and all were able to think abstractly – evidenced by solving mathematical problems and fixing mechanical things (Pinker 1994, 68). These language-less social subjects, who think without discourse, are outside the empirical conditions of possibility of Butler’s social theory. These individuals were familiar with concepts first; then they acquired signs as a linguistic tool to express prior concepts, which were already part of their interpretative horizon.

Contrary to popular misconception, sign languages are not pantomime imitations of spoken languages; rather, fully developed languages. An interesting further instance is Language de Signos Nicaraguense. Until the Sandinistas took over in Nicaragua, deaf people remained isolated from one another, thus without sign language. Under educational reforms, deaf children were brought together in schools, where teachers taught them lip-reading, with dismal results. However, because they were together (in playgrounds and school buses) these children developed their own sign language. In a relatively short period a spontaneous linguistic sign system congealed into Language de Signos Nicaraguense (Pinker 1994, 36). In addition to these examples, there are comparative cognitive tests of individuals socialised into languages that do not have certain concepts, who are then compared to speakers of different languages that have these concepts. For instance, speakers of a language that has a single word for blue and green are compared in cognitive tests to speakers of a language where blue and green are different words (Pinker 1994, 62–63). Overall, yes, the linguistic difference affects the speed of distinguishing objects in these colours (language colour distinction gives an advantage). So, lack of socialisation into words for concepts does create some internal constraint. Yet, those who speak a language without this distinction are able to complete tests that hinge on making distinctions between colours that they have no words for (Bloom and Keil 2001; Pinker 62).

Summarising the relationship between thought and language Bloom and Keil ask themselves the questions: “Does language influence thought? Obviously yes; this is why we use language in the first place. Does language have a dramatic influence on thought in some other way than through communication? Probably not”. (Bloom and Keil 2001, 363–364). They summarise the use of language as follows: “it is a tool for the expression and storage of ideas. It is not a mechanism that gives rise to the capacity to generate and appreciate these ideas in the first place”. (Bloom and Keil 2001, 364). Translated into social theory terms, an individual is well able to come up with novel ideas that exceed her linguistic community. Language is a conceptual tool for interacting with other members of the discursive community. As shall be explained later, the necessity to communicate with other imposes significant structural constraint upon the creative agency of the social subject. In other words, much of social constraint is external, from the necessity of interactive collaboration.

Two meanings of the concept of a dominant ideology

The ability of social subjects to significantly exceed linguistic exposure was evidenced in the Atlantic slave trade. In the sugar plantations slavers deliberately mixed slaves of different linguistic backgrounds, to make rebellion difficult. Yet, these slaves had to accomplish practical tasks and developed a simple naming language, where words stand for things, without complex linguistic grammars, which linguists refer to as pidgin. However, the children of these slaves did not simply copy their parent’s pidgin. Rather, they built upon pidgin, developing a fully formed language, which linguists refer to as creole (Pinker 1994, 33). As emphasised by the founder of variationist linguistics, Labov (1973), while considered primitive by dominant linguistic groups, Creole is a fully formed highly sophisticated linguistic system. In other words, these slave children were far from passive reproducers of received language. Interactively, they went from the disparate signifiers of their parents to a fully developed language of their own. These actual slaves, unlike the hypothetical (thought experiment) slaves of Hegel’s master-slave dialectic, did not speak the language set out for them by their masters. They did not internalise dominant discourse or reproduce it imperfectly or ironically; they created their own discursive field.

The fact that these slaves showed significant discursive agency is consistent with the work of Scott (1990). Based upon anthropological research of peasant and slave resistance, Scott (1990) critiqued Lukes’ (1974) third dimension of power. Scott argues that subaltern actors who appear acquiescent, frequently are so not because they have internalised dominant discourse, or are suffering from a kind of false consciousness, as in Lukes (1974) – or are interpellated, as in Althusser/Butler. Rather, these subaltern actors go through the rituals of dominant ideology as acquiescence to structural constraint. Slaves and highly dominated peoples do not passively internalise ideology. Rather, they are often fully aware of the discrepancies between the legitimating aspects of dominant ideology and the structural reality in which they live. They only reproduce dominant ideology as a performance, which can be termed public transcript (Scott 1990, 2). Contrary to the hypothetical master-slave dialectic, slaves reproduce the discourse of their masters only to get by, while they create their own counter-discourse among themselves. The theoretical point being that linguistic constraint does not come from the social subject’s constitution through discourse. Individuals may well transcend that discourse but when they do, they are still constrained by the necessity for interacting with powerful others, who will at the very least reject, and quite likely punish, the expression of any such alternative discourse. In other words, the problem of constraint is often external, from a non-collaborative interacting other.

Scott was influenced by Abercombie and Turner’s (1978) critique of the Marxist dominant ideology hypothesis. There are two ways to interpret the concept of dominant ideology. The strong version is the claim that the ruling class essentially determines the overall ideology of society, including the thought processes of the subaltern classes. The weak version accepts that there are many conflicting ideologies in society, with dominant and subaltern having different ideologies. The ideology of the powerful is dominant in the weaker sense that this ideology is what informs the practical knowledge of those in key decision-making positions, which does not preclude a subaltern ideology that co-exists with the dominant ideology, but is not often publicly expressed in the presence of the powerful. As it is the discourse of the powerful that is most visible, it is the dominant in that weaker sense.

In classic Marxism there is an ambiguity around the extent to which Marx and Engels thought the working classes internalised dominant ideology. Suggestive of the strong version of dominant ideology, Marx and Engels wrote: “the ideas of the ruling class are in every epoch the ruling ideas, i.e. the class which is the ruling material force of society, is at the same time its ruling intellectual force”. (Marx and Engels 1974, 64). However, Engels describes the actual working classes as follows: they “speak other dialects, have other thoughts and ideals, other customs and moral principles, a different religion and other polities than those of the bourgeoisie”. (Engels 1968, 124). This suggests the weak version, consistent with Scott. Moving forward in time, Gramsci’s (1973) account of hegemony and Althusser’s (1970) account of ideology are both typical of the strong version: both thinkers argue that the dominant class socialise the subaltern into bourgeois ideology. Butler’s account not only explicitly builds upon Althusser, it ties the subaltern class to dominant ideology even tighter by adding a subject formation psychological account of attachment and the underlying assumption of linguistic determinism of the social subject (Butler 1997a, 106–131).

As characterised by Scott, the dominated are highly creative social actors, who perform two scripts. In situations of unequal power, front-stage performance corresponds to the ideology of the dominant, while the back-stage performance corresponds to the ideology of the subaltern. For Scott, linguistic determinist and strong versions of dominant ideology portray social subjects as too passive, underestimating the creative capacity of individuals to resist dominant discourse with alternative discourses – in essence, such theorists hold an “over-socialized” (Wrong 1961) concept of the self, where the social subject becomes a cultural, or epistemic, dupe.

The intersubjective basis of structural constraint

Analogous to Levinas’ critique of Heidegger and Husserl, in social life there is not a social subject in relation to the world-out-there, as suggested by any narrow focus upon subjectivity, humans are intersubjective beings (Guenther 2013, xv). Social actors are interactive beings, whose sources of social constraint lie both internally and externally to them, in the process of interaction. Internally, consistent with aspects of Foucault and Butler, social actors are constrained by the discourse that they have internalised as the natural-order-of-things. This is a significant source of constraint but, as argued above, social actors are also creative beings with the capacity to think beyond the discourse that they have internalised through socialisation. In creative mode, social subjects’ problem is to find others willing to validate alternative subaltern discourses. Following Arendt’s (1958, 247) view of natality, every person born is a unique individual with the potential for individual originality. However, their individual originality is confronted by the necessity of interactive collaboration.

As is convincingly argued by Butler, gender is performative, where each iteration, or act of structuration, is a recreation of the social construction of social reality. While Butler correctly identifies imperfect iteration as a source of social change, what is relatively lacking from her model is the interactive significance of iterative failure. If we return to Austin’s famous example of performatives, it is worthy of note that the example is actually about iterative failure:

Suppose … . I see a vessel on the stock, walk up and smash the bottle hung on the stem, proclaim “I name this ship the Mr. Stalin” and for good measure kick away the chocks: but the trouble is, I was not the person chosen to name it … . We shall agree

1. that the ship was not thereby named;

2. that it is an infernal shame

One could say that “ … . that my action was void or without effect, because I was not a proper person, had not the “capacity”, to perform it” … . (Austin 1975, 23–24)

Austin went through the correct authority performance of naming the ship but failed to realise agency power-to to name ships because his actions were infelicitous. Let’s imagine that Queen Elizabeth came along an hour later, performed exactly the same actions as Austin. In contrast, everyone applauded, and the ship was thereby named. In contrast to Austin, Queen Elizabeth has agency power-to because her actions were deemed felicitous.

The distinction between felicity and infelicity is a judgement made by others. The reproduction of authority presupposes others who confer the status of felicity, which makes it an external constraint. Moving into the language game of sociology, what Butler describes as subject positions are essentially authority positions (Haugaard 2020, 27–33). The social actor performs a certain role because they desire whatever empowerment comes with authority. When, for instance, a discriminated-against minority group find themselves unable to apply for positions of power, in management or politics, their social actions to occupy such positions of authority are deemed infelicitous by others, thus rendered outside the conditions of possibility for those subaltern actors.

Typical examples of authority (the Queen of England or President) suggest that authority is always a property of high-status political hierarchies, which is not the case. Authority is part of everyday life. With the exception of slaves, prisoners or other instances of extreme domination, everyone has circumscribed authority. The most basic act of authority is the right to be the author of your own interactions. In the classic account of slavery, the slave was someone who did not have the authority to speak for herself. In real life a close approximation to this state of total absence of authority is given by Primo Levi, in his account of being slave labour in Auschwitz. In Italy he was a chemist and a “citizen of Jewish race” (Levi 1991, 4), but when he laboured in the chemistry laboratory of Auschwitz, the fact he was of “Jewish race” precluded him from being a citizen. Not fully aware of his loss of authority, in the laboratory Levi asked one of the Aryan cleaners for something but they did not reply. Further, symbolic of his lack of authority, when the cleaners swept the floor, they swept his feet (Levi 1991, 168). Levi’s absence of authority to speak and to be heard was tied to a specific racist discourse formation. His request, as a professional chemist, for, say a test tube, would have been felicitous in most other imaginable discursive circumstances in the twentieth century. Current social movements, including Black Lives Matter, are claims to legitimate authority on behalf of persons for whom claims to authority are deemed infelicitous. By creating a social movement these social actors create a community of likeminded persons for whom their claims to authority are deemed felicitous. Butler (2015) emphasises the democratic emancipatory aspects of contemporary social movements, which I don’t deny but don’t consider inevitable (for reasons outside the scope of this article). Rather, I wish to emphasise a different aspect, which is their power-cum-authority potential as communities of epistemic resonance.

In the weak sense of dominant ideology, those in power do not accept the claims to authority of less powerful groups. The overall view of society is one of competing epistemic groups, jostling for command authority positions of power. This contrast with the strong dominant ideology thesis, where dominant elites control the thoughts and desires of society, therefore there is a single overwhelming epistemic world view, in which everyone is socialised.

Systems of thought, discourse formations and external structural constraint

Foucault suggests that there is radical discontinuity between systems of thought, for instance, the Renaissance (1500–1650), was displaced by the Classical (1650–1800), and later the modern system of thought (1800–present) (Foucault 1970). Foucault’s concept of systems of thought, as a totalising phenomenon, is reminiscent of the strong dominant ideology thesis, as social actors appear trapped within epistemic systems of thought. I will use examples of alternative discourses, to explain how society is characterised by competing epistemic interpretative horizons, some of which confer greater authority than others.

The Renaissance was characterised by resemblances and metaphor, (Foucault 1970, 17). Contrary to the implications of Foucault’s theory of discontinuity, this discourse did not end in 1650. Today, on the internet, we learn from the Astrological Project that one of the key principles of astrology is metaphor and aphorism: “Every astrological chart promises rich insight, yet taunts with complex symbolism. Down the centuries of astrological writings, two teaching aids maintained a key presence – Metaphor and Aphorism”. (Astrology Project 2020: Core Principles vii). Within the contemporary discourse formation of astrologers there are clearly highly precise debates, suggested by the following error: “it is … erroneous to conflate zodiacal and planetary symbolism with the meanings of the houses. In being a unique component with a unique informing tradition, the houses cannot be “naturally” elemental (fiery, earthy, etc.) or “naturally” zodiacal (Arian, Taurean, Geminian, …) or “naturally” planetary (Martian, Venusian, Mercurial …)”. (Astrology Project 2020: Core Principles ii).

This astrological website is current; the quotations from lectures given at astrological conferences. These authors/presenters have authority within an epistemic community. However, imagine making the statement that “houses are not naturally either zodiacal or planetary” at a meeting of the Max Planck Society for astrophysics. The reaction would simply be a stunned silence or laughter (infelicity). What has changed since the seventeenth century is that astrology no longer has high status authority within the discursive fields of astrophysics; yet astrology has not ceased.

Astronomy and physics have gained status by being taught at universities whose qualifications are endorsed by the state. However, there is such an institution as the Astrology University (2020), with a four-year degree course, which mirrors a typical BA. However, the authority of that degree is not recognised in powerful decision-making circles as a felicitous qualification.

While the powerful in society generally do not recognise the authority of a BA in astrology, according to the Pew Research centre, 29% of the population in the USA believe in astrology (Pew 2018), which is probably a conservative estimate because astrology has low status. In the survey, 41% believe in physics. If you equate physics to astronomy, over three hundred plus years after the supposed end of the Renaissance era, astronomy is a dominant ideology (weak sense) relative to astrology, true, but astrology is very far from dead as a popularly shared ideology, 41% versus 29%.

The contrast between astrology as popular discourse and dominant ideology is illustrated by the following incident. President Ronald Reagan was known for taking advice from First Lady Nancy Reagan. She believed in the veracity of the discourse of astrology and hired the services of Joan Quigley, a prominent astrologer. However, when the use of the services of an astrologer in the White House was made public, this appointment was deemed infelicitous: “For the most part, Nancy Reagan was mocked. The New York Post had a headline saying, ‘Astrologer Runs The White House’, and one joke suggested a Cabinet post in charge of voodoo be created”. (Los Angeles Times 2016). In this example, Nancy Regan had the agency to go against dominant discourse but she was infelicitous, thus externally structurally constrained.

Foucault once observed that “Medical statements cannot come from anybody: their value, efficacy, even their therapeutic powers, and, generally speaking, their existence as medical statements cannot be dissociated from the statutorily defined person who has the right to make them”. (Foucault 1989, 51). This is both correct and incorrect. It is correct in the qualified sense that authority status position directly impacts how felicitous a statement is perceived to be with a specific audience. It is correct that among the people for whom Western Medical science represents the discourse of truth, medical statements can only come from medical doctors. However, the field of medicine stands in contrast to the field of homeopathy, in a roughly analogous way to astronomy and astrology. Like astrology, homeopathy also follows the Renaissance principles of similitude: similia similibus curentur (Hahnemann 1833, 48–49). While Hahnemann (the founder of modern homeopathy) is no longer read by medical students, his principle of similia similibus curentur would be widely respected by practitioners of homeopathy. Sales of homeopathic remedies were $1.2 billion in 2014 (Washington Post 2015). Clearly there are a lot of consumers in the USA who accept medical statements from homeopaths. Foucault is mistaken to suggest that medical statements can only come from doctors, if the reference is to popular discourse. However, he is correct in the qualified weaker sense that medical statements made by doctors are felicitous within the discourse community of powerful decision-makers – thus dominant discourse in the weak sense.

It is worth pointing out that these examples furthermore illustrate the point that subaltern discourses are not necessarily closer to the truth, more normatively desirable or democratic, than dominant discourse, as suggested by Butler (2015), although full discussion of this point remains for another article.

To have authority means to be treated as persons with the authority to speak and be taken seriously. This entails two external structural constraints. One is the status authority position (Foucault’s point). The other is the status of the epistemic logic that they wish to reproduce. Surprisingly, in a further qualification of Foucault’s point, even high-status authority position does not confer automatic authority, if contrary to dominant discourse. To give an academic example, Professor Avi Loeb is Frank B. Baird, Professor of Science at Harvard University and chair of Harvard’s department of astronomy, which is arguably one of the highest possible scientific authority positions in the world. Yet, he cannot pronounce upon astronomy as he wishes. Loeb is an immensely prolific astrophysicist who “has produced pioneering and provocative research on black holes, gamma-ray bursts, the early universe and other standard topics of his field” (Billings 2021). In 2017 the astronomy community observed an enigmatic interstellar object, Oumuamua, which was a strange oblong shape, shiny like metal, with an apparent capacity to accelerate, and no evaporation tail. These are not the properties typical of any known asteroid. Loeb and his co-author Shmuel Bialy, a Harvard postdoc, published a paper in the Astrophysical Journal Letters arguing that Oumuamua could be a light-sail “floating in interstellar space as debris from advanced technological equipment”, or indeed “Oumuamua may be a fully operational probe sent intentionally to Earth vicinity by an alien civilization”. (Shmuel and Loeb 2018, 4–5). Loeb has followed this up with a book, entitled, Extraterrestrial: The First Sign of Intelligent Life Beyond Earth (Loeb 2021). While Loeb’s hypothesis is popular with the mainstream media and especially with UFO spotters, among the scientific community, this was received as an entirely infelicitous pronouncement. Whenever Loeb shares the extraterrestial hypothesis with other astrophysicists “Ninety-nine per cent of the time, I’d get this silence … . On Twitter, one scientist described the hypothesis as insulting” (Moshakis 2021). Loeb observes that “‘most of the people who attacked … hadn’t even looked at my paper, or read the issues, or referred to the items we discussed’” (Moshakis 2021). They simply do not engage with his work because, Loeb observes: “‘There is a taboo on the subject’, ‘It frightens credible scientists from discussing it. [Furthermore,] I’ve noticed a chilling effect on some people who have worked with me’” (Moshakis 2021). Despite having this highest authority possible in the discourse community of astrophysics, Loeb doesn’t have the authority to speak about the possibility of extraterrestial civilisations. Of course, literally he can speak: he is not internally restrained by his subject position. Loeb is externally restrained by the necessity of having to interact with the peer scientific community. Of course, UFO spotters consider Loeb felicitous but they are not powerful decision-makers. Thus, dominant discourse is only dominant in the weak sense, as there is a discursive community of less powerful actors who consider it felicitous.

These examples of infelicitous pronouncements illustrate the nature of external constraint upon knowledge production. These social subjects are not internally restrained by hegemonic ideas, but they are externally so. Relative to the critiques of Butler and Foucault earlier, they can step outside a dominant system of thought (Foucault), and/or they are individuals who go beyond their subject position – beyond simply exploiting the gaps in reiteration of speech (Butler). Relative to the earlier examples, of children inventing speech, they share the radical novelty of being able to think outside dominant discourse. However, they are still constrained externally by finding a discursive community to enrol into their alternative vision. In the case of Loeb, as an academic astronomer, enrolling non-academic UFO spotters is not real empowerment, because the academic community that matters to him are other astrophysicists. In contrast, the Nicaraguan deaf children and the slaves developing Creole, discussed earlier, were a collective community without the linguistic conceptual tools to communicate but they had a ready-made social relevant community of confirm-structuring others with whom they could collaborate. These social actors were mutually enabled. However, no doubt, in the case of Creole their speech was considered “primitive” and “uncivilized” by higher status, thus infelicitous, by more the powerful (colonisers or members of more standard English/Spanish speech communities). These variations by minority linguistic communities are perceived of as “primitive” by the dominant linguistic community, which is way deeming them infelicitous and of low authority. Yet, as argued by Labov (1973), Creole is highly sophisticated in terms of internal grammar or syntax and perfectly tooled for the purposes of mutual communication. Furthermore, argot and Creole speakers exploit their rejection by the hegemonic linguistic community to enable them speak their own counter-hegemonic discourse more freely, as they won’t be understood by the powerful. In other words, hegemonic exclusion followed by the creation radically different subaltern alternative can have its enabling aspects, although it will take many years for such discourse to gain recognition as felicitous from more powerful decision-makers.

Radical discourse

For social actors who hold a minority, low-status, discourse formation as a language of truth, and who wish to initiate social change, there are two possibilities. First they can try to create an epistemic community within which their views are taken seriously. As that epistemic community expands, their authority and the status of their utterances will gain authority. This epistemic discourse community will be considered radical, as outside dominant ideology. However, over time, they can attempt to enrol (Callon 1986) powerful decision-makers and, in so doing, make their discourse dominant.

When Mary Wollstonecraft wrote A Vindication of the Rights of Women in 1792, she set out on the lonely path of uttering views that few contemporaries were willing to confirm-structure (Haugaard 2020, 40–43) as felicitous. Wollstonecraft admitted that she herself found it “difficult to deny the central presumption of her age, that women possess natures different from men” (Landes 1988, 13). Yet, she was able to overcome the dominant discourse of her age. Her problem was with the reception of those radically different ideas – external constraint. Vindication has three themes running through it, as follows: (1) that women should be equally educated, (2) that gender differences between women and men are socially constructed,  and (3) that women should have the same political rights as men. Interestingly, among radical circles point one, concerning education, was well received (Janes 1978, 293–294), but the other two claims were ignored as infelicitous. “Demands for political participation by women or for changes in women’s social behavior were regarded as unessential and absurd” (Janes 1978, 293). With regard to political rights, Wollstonecraft refrained from spelling out the detail because she knew such demands were infelicitous: “Had Wollstonecraft argued specifically for the franchise, equal access to the professions, abolition of discrimination based upon sex, positions consistent with the book’s argument but not developed, … her first readers might justly have thought her mad” (Janes 1978, 302). In other words, Wollstonecraft developed a radical discourse but spelling it out in detail was infelicitous. At that time, only the request for access to education was heard as felicitous, while “modern readers light upon” (Janes 1978, 302) the socially constructivist and political claims, as felicitous speech acts. The reason that modern readers react differently is that radical feminists have, over two-hundred plus years, through a process of “enrolment” (Callon 1986) and made the socially constructivist account of gender a comparatively (but not uncontested) dominant discourse.

Reverse discourse

When Foucault writes of a strategy of reverse discourse as using the “same vocabulary, using the same categories by which it was medically disqualified” (Foucault 1981, 101) he is pointing out pragmatic advantages of using already dominant discourse for unexpected ends. Going beyond Butler, the advantage of using reverse discourse is that reproducing existing discourse increases the possibilities of a felicitous response from the more powerful. Typically, social actors approach the world using the natural attitude, whereby the natural-order-of-things is reproduced, affirmed as felicitous, almost as reflex. This is the theoretical conceptual equivalent of the internal constraint described by Butler, except that it is not as determinate as in her account of subject formation through language. It is not as determined because the natural attitude can be suspended, and often is in critical situations or for the purposes of critique (Haugaard 2020, 84–86). In this theorisation reverse discourse is a strategic use of the familiar natural-order-of-things of the more powerful to lessen the possibility of rejection through infelicity by working within the internal constraint of others, using their natural-order-of-things to change power relations. When an actor uses the taken-for-granted order of things of others, it is more likely that their performance will be regarded as felicitous. In Butler’s theorisation of reverse discourse mimicry and irony are central, while in this theorisation they are is not. Reverse discourse is the use of the logic of existing dominant discourse for the purposes of changing power relations. The natural attitude of the powerful is appealed to in order to make them re-evaluate existing relations of domination.

A paradigm instance of the less powerful using an existing discourse for the purposes of initiating social change is Martin Luther King’s “I Have a Dream” speech of 1963:

… we’ve come to our nation’s Capital to cash a check. When the architects of our republic wrote the magnificent words of the Constitution and the Declaration of Independence, they were signing a promissory note to which every American was to fall heir.

… 

It is obvious today that America has defaulted on this promissory note insofar as her citizens of color are concerned. Instead of honoring this sacred obligation, America has given the Negro people a bad check; a check which has come back marked “insufficient funds”.

But we refuse to believe that the bank of justice is bankrupt. … So we have come to cash this check … . (King 1963)

By using the “magnificent words” of the Constitution and Declaration of Independence as a “check”, King is not propounding a new discourse of rights. Rather, he empowers a subaltern group by using the already dominant discourse of US politics, to make a claim that will change power relations. The image of the cheque bouncing is an image of the powerful political class reneging on their own discourse.

This is not reverse discourse as parody or mimicry. Rather, the strategy is to exploit the epistemic structural constraint of the more powerful others by using the logic of local reason to reverse status quo power relations. Within a local language game, social actors find it difficult to uphold blatant contradiction relative to their taken-for-granted practical knowledge. When powerful social actors are confronted by performative contradictions, relative to a discourse that they hold in high esteem (their local language of truth), then the less powerful has leverage upon them. This may sound like I overestimate reason. To be clear, this is local reason: the logic of specific discourse formation, and can be wildly different, depending upon context – not Enlightenment reason, economic, scientific reason, rational choice reason or anything of the sort.

Essentially, I am arguing that reverse discourse uses the assumed natural-order-of-things of the practical knowledge of the dominant, or dominant ideology (weak sense), to subvert existing power relations. Reverse discourse accepts the external epistemic constraints imposed (what is felicitous or infelicitous) and strategically works through the logical implications of that discourse in order to reverse the power relations of the status quo. This strategy recognises the significance of constraint but not solely internally to self but from the interacting other. This is portrayed as one of two choices (reverse versus radical discourse). Parody and mimicry are not necessarily, though they can be, part of a technique of reverse discourse. This is qualified critique, not a refutation.

Using an existing discourse that has high authority status for unexpected ends is a highly effective strategy, but not without drawbacks. This strategy reproduces, thus validates, the fundamental underlying principles of the local discourse formation. It is an act of confirm-structuration to the status authority of the dominant discourse and, consequently, a disconfirming of the epistemic validity of contrary subaltern discourses. For this reason, it can be characterised as a conservative strategy, rather than a radical one. If the underlying discourse has egalitarian elements that are normatively desirable (as in the King example) the strategy makes sense and has few drawbacks. However, if the underlying premises or assumptions are not normatively desirable, reverse discourse is a short-term strategy of acquiescence, which reinforces domination in the longer term. Contrary to Butler (2021), I would argue that this is the case even with parody and mimicry, as the very act of reproducing essentialist categories validates a reifying thought process.

Critique of reification and essentialism

One of the most effective ways of stabilising a discourse formation is to reify it. Social structures are socially constructed. Exposing this social constructedness is central to critique because once social actors perceive that structures are made, they can also be unmade (Foucault 1988, 37–38). Once social structures are by convention only, structures are arbitrary unless they can be shown to justifiable normatively. So, for instance, the structures of democracy are conventional though not arbitrary, as they can be justified normatively (see Haugaard 2020, 89–94). However, such structures are exceptional; many social structures, once understood as socially constructed, cannot withstand justification, especially social structures that confer unequal authority power. So, for instance, the status authority associated with male and female gender roles is unequal in patriarchal societies and, when understood as socially constructed, appears arbitrary, as normative justification is impossible. To avoid such social deconstruction the powerful have an interest in arguing that the social structures that confer unequal authority are not social constructions. Reification of social structures takes many forms, including essentialism, making sacred, or transcendental truth claims (see Haugaard 2020, 94–140).

Reification by essentialism is a strategy whereby some essence that transcends social construction is said to be inherent to a subject position. Essentialisms are often presented in positive terms, inviting the less powerful to endorse them, as in following quotation from a nineteenth century women’s magazine

Man is strong – woman beautiful. Man is daring and confident – woman is diffident and unassuming. Man is great in action – woman in suffering. … . Man has judgment – woman sensibility. Man is a being of justice – woman of mercy. (Quoted in Alexander 2006, 238)

Every male essential virtue is balanced against a female one. While dominating, the author presents him/herself as even-handed, claiming that the two sexes are by nature essentially different. If men and women are really so essentially different, superficially, it appears entirely fair and reasonable that women should not make decisions concerning justice and politics, while they should, concerning the caring professions, where their (supposed) abilities lie.

Radical critique attacks these kinds of reifying essentialism, by demonstrating that these categories of thought are socially constructed. It was precisely these kinds of essentialist arguments that Wollstonecraft was attempting to deconstruct. As she observes: “[w]omen are, in fact, so much degraded by mistaken notions of excellence … ” (Wollstonecraft 2014, 33). The thrust of her critique of essentialism is to show that not only is this kind of supposed essential virtue not a compliment; it is form of domination (see Wollstonecraft 2014, 171).

Reverse versus radical feminist discourse, an example

In dealing with essentialism there are two strategies of resistance. The radical one is to deconstruct the essentialism, showing it to be a social construction: that which was made can be unmade (Wollstonecraft’s strategy). In contrast, the reverse discourse strategy is to accept the essentialisms but reverse the power implications of their evaluative logic. Such a strategy will have the short-term advantage of not encountering infelicity associated with deconstructing the dominant discourse, while simultaneously shifting a balance of authority power. However, longer term, the logic of essentialism is reproduced.

The contrast between the strategic advantages of radical versus reverse discourse lies at the core of the complex reaction to Gilligan’s critique of Kohlberg (Auerbach et al. 1985, 149–150; Davis 1992, 219–202). While this debate took place in the 1980s, it is a paradigmatic example of the dilemmas faced by feminist critique, therefore still prescient. To be clear, I am not using the theory, which is outdated. Rather, Gilligan is under sociological observation as an exemplar of a social actor confronting strategies of power.

Under the influence of Piaget, Kohlberg (1981) developed a theory of six moral stages of cognitive development. Put simply, the most advanced stages entail the capacity for the application of abstract moral principles in a highly reflexive way, where rules can be adjusted in light of deeper principles. In contrast, the lower stages tend toward repetition of conventional rules, in an either unreflective stimulus-response or reward and punishment manner. Kohlberg undertook qualitative research to ascertain the extent to which children and young adults developed along this scale. Typically, they were told a story, which involved a moral conundrum, and asked to solve the problem, giving moral justification. The most famous example concerned the moral dilemma of a husband (Heinz), with a dying wife, who could not afford a medication developed by the local pharmacist. The children were asked should Heinz steal the medicine and were prompted to justify whichever response they made (steal or not steal) morally. The children’s justification revealed their stage of cognitive development. The ethnographic data showed that the majority of children and young adults never reached stage six reasoning and it was noted, in some studies, that girls tended more towards stage three reasoning, where the good is identified with what pleases.

In her feminist critique Gilligan argues that the problem with Kohlberg’s reading of females as less able for stage six reasoning, and more adept at stage three, is that “the very traits that have traditionally defined the ‘goodness’ of women, their care for the sensitivity to the needs of others, are those that mark them as deficient in moral development” (Gilligan 1977, 484 – italics added; see also 1982, 18). Rather than contest these (supposed) traditional traits as reified social constructions used to dominate women, Gilligan valourises these traits. This is classic reverse discourse. She reproduces the categories of meaning associated with traditionally defined female virtues, while changing their value. She argues for two moralities: one based upon rights, the other upon responsibilities and care. She contrasts the responses of a girl (Amy) and boy (Jake), both eleven. Asked whether Heinz should steal the drug, Jake argues that Heinz should steal, as follows:

For one thing, a human life is more than money, and if the druggist only makes $1000, he is going to live, but if the Heinz doesn’t steal the drug, his wife is going to die (Why is life worth more than money?) Because the druggist can get a thousand dollars later from rich people … . But Heinz can’t get his wife again. (Gilligan 1982, 26)

When it is pointed out that stealing is normally considered morally wrong, Jake replies that the law is “man-made”, and that “the laws have mistakes, and you can’t go writing up law for everything you can imagine” (Gilligan 1982, 26). Jake is clearly able to reason from abstract principles, and understands how rules are only an approximation for principles of justice.

In contrast, Amy’s responses appear at a lower cognitive stage. Asked if Heinz should steal, she responds as follows:

Well I don’t think so. I think there might be other ways besides stealing it, like if he could borrow the money, or make a loan or something, but he really shouldn’t steal the drug – but his wife shouldn’t die either. (Gilligan 1982, 28)

While on superficial reading Amy appears less sophisticated than Jake, Gilligan explains that the difference in approach comes from a different social ontology. Jake’s conception of reasoning is from the perspective of an autonomous individual, looking for impartial justice. In contrast, Gilligan argues that Amy’s social ontology represents a thick concept of the self, situated in a complex web of particular relationships where moral judgement is based upon understanding the particularities of the position of other; a process of care and responsibility where the effective emotions are inseparable from cognition (Blum 1988, 475–476), which is not a lower cognitive stage of development.

As a strategy of critique of domination, validating the experiences and reasoning of women is a laudable task and resonated with a diverse spectrum, from feminists to non-feminists (Auerbach et al. 1985, 149–150; Davis 1992, 219). It resonated with groups of women who wanted to increase the authority of women in (so-called) “caring professions”. For instance, writing from the perspective of nursing, Harbison observes “Gilligan’s gender-related theory may also be particularly appropriate for nurses, given the female domination of the profession and provides a defence for those attributes of caring and sharing which traditionally have not been highly valued by dominant ideology” (Harbison 1992, 204). Similarly, Rhodes observes that social work “has always been a largely woman’s profession … ” (1985, 103) and that Gilligan’s “‘responsibility’ mode of thought embodies many of the basic ideals of social work practice” (Rhodes 1985, 101), as “caring for others is considered the foundation of all social work” (Rhodes 1985, 102). In these examples, social actors use reverse discourse to increase the authority of female nurses and social workers.

The effectiveness of reverse discourse lies in its appeal to existing discourse. It does not encounter a reaction of infelicity because it reproduces the taken-for-granted, natural-order-of-things. However, it also reproduces social structures that essentialize male-female distinctions. Having made the case for the value of the social-work ethic Rhodes notes that this ethic has its roots in Christian reform associations of the nineteenth century: “Women’s moral superiority were preached in churches, … . [and] identified with Christian piety, moral integrity, compassion, self-sacrifice, and maternal concern” (Rhodes 1985, 103). This suggests endorsing the essentialist compliments that Wollstonecraft regarded as domination.

Gilligan’s response to the (supposed) fact that men are better than women at stage six reasoning is not the only avenue of resistance. It is plausible to argue that the reason these women are less accomplished at stage six reasoning is a consequence of domination. Stage six reasoning presupposes the autonomy of having meaningful choices. Being autonomous is a consequence of having a significant authority. In contrast, relational, case-by-case reasoning is typical of social actors in less powerful positions or oppression (Auerbach et al. 1985, 154; Davis 1992, 224). This reading is supported by some of Gilligan’s interviewees, for instance:

As a woman, I feel I never understood that I was a person, that I could make decisions and I had a right to make decisions. I always felt that that belonged to my father or my husband in some way, or church, which was always represented by a male clergyman. (quoted in Gilligan 1982, 67)

I will not adjudicate which reading of the (supposed) propensity of women to reason differently from men is correct as a truth claim (personally, I consider it suspect). As a sociologist, I simply want to observe that there are two contrasting strategies of power and resistance. The strategy of reverse discourse does not run the same infelicity risk as the more deconstructivist power approach. Reverse discourse doesn’t challenge the discourse of supposed essential differences, while the second, more radical strategy does. While more effective in the short-term, the unintended effect of the reverse discourse is to reproduce socially structured concepts that, longer-term, are part of the discourse of domination. For instance, as pointed out by feminists Auerbach et al. (1985), if Gilligan’s theory were introduced into an organisational context, short-term, it would make work-relations humane. However, longer-term, the unintended effect could be to reinforce the idea that women are best suited for middle-management, where dealing with people is key, while upper management, where autonomous reasoning is required, is preserved for men. In short, “[w]omen will be in there ‘relating’ to people, while men will continue to run the corporation” (Auerbach et al. 1985, 159). To fight negative essentialisms with positive essentialisms means reproducing essentialism, which is reifying, and has dominating implications. Reverse discourse falls foul of the critique that Arendt made of Marx’s turning upside down of Hegel’s dialect. The problem with the “‘turning upside down’ of philosophical systems or currently accepted values … is in the nature of the operation itself, that the conceptual framework is left more or less intact” (Arendt 1958, 17).

The opposition between reverse and radical discourse are ideal types, which are, furthermore, on a scale. Actual social interaction often has elements of both reverse and radical discourse. As we saw with the example of Wollstonecraft’s radical ideas, she did, in practice hold back with elucidating the full implications of her radical theory, because she did not wish to appear, to contemporaries, “mad”. In other words, she was careful not to be too infelicitous, which represents some backing away from radical discourse, more towards the reverse discourse technique of reproducing the dominant ideological natural-order-of-things, but only to a qualified extent.

Conclusion

We began with Foucault’s description of reverse discourse, where existing dominant discourse formations are strategically used to shift power relations. For Butler reverse discourse is the sole strategy of resistance because, according to her account, social subjects are constituted through discourse, which, in turn, is consistent with the strong dominant ideology thesis. I argued that Butler is overly determinist. Social subjects are significantly more creative; capable of thinking outside dominant discourse than her theoretical position suggests. Yet, I fully accept that the tacit interpretative horizon of social subjects is hugely significant in giving strategic advantage to not contesting the natural-order-of-things, which makes reverse discourse a significant avenue for contesting dominant power. Unlike Butler, because subjects are not discursively constituted, social actors can step outside internal constraints of the natural attitude, and create new meanings, which open the strategic possibility for a significantly more radical strategy of resistance, transcending local discourse. In that case, in the place of internal constraint, subjects are externally constrained by the need to interact. In order to gain agency power-to they must perform authority successfully. Consistent with Scott, less powerful social actors don’t necessarily internalise dominant ideology, they often reproduce it strategically to gain immediate agency. In that case, reverse discourse, rather than radical resistance, involves an awareness of radical alternatives but a pragmatic decision to use dominant discourse because such a strategy will be more immediately effective.

Foucault’s account of systems of knowledge is consistent with the tradition of strong dominant ideology. However, contrary to Foucault, the world-view of the Renaissance did not disappear in 1650 (a stronger version of dominant ideology); astrology and homeopathy live on but with lower authority (weak dominant ideology). Foucault argued that certain authority positions pronounce the truth, which is a position that is qualified. First of all, there are many discourse formations vying for dominance, hence authority is typically only lost relative to a single discourse formation, albeit that of powerful decision-makers. Second, even positions of high authority, such as Professor of Astrophysics at Harvard, are subject to the external constraints of performing felicitously, as defined by local discourse. A Harvard professor of astrophysics who overcomes the internal restraints of the dominant interpretative horizon, and consequently pronounces upon extraterrestrials, loses authority within their discourse community; although, they gain agency among less powerful discourse communities, including UFO spotters. Overall, social life is characterised by a multiplicity of epistemic communities, each wishing to become dominant, in weak sense of having greater authority than the rest. As the examples of astrology and homeopathy imply, subaltern discourses do not necessarily have a greater claim to truth, democracy or normative desirability.

Social actors who wish to change power relations have two strategies. As theorised here, reverse discourse is a relatively conservative strategy. The discourse of the powerful is reproduced to demonstrate performative contradiction: to prove that the powerful have issued a cheque they have reneged on it or by revaluing reified categories but without challenging essentialist ways of thinking. Short-term, this is a highly effective strategy because discursively felicitous relative to the dominant natural-order-of-things. Where the discourse of the dominant has implicitly normatively desirable principles, this strategy of resistance has few unintended costs. However, if the discourse of the dominant includes damaging essentialisms, then this is a short-term strategy that reproduces essentialism, even if reversing their value. With Gilligan we saw that reverse discourse reproduces the reifying conceptual logic of dominant discourse of essential difference.

The more radical alternative to reverse discourse is to establish an alternative discourse formation. However, at first this will be perceived as infelicitous by the dominant. Consequently, the subaltern has to adopt a longer-term strategy of enrolling powerful actors, to make subaltern discourse dominant. The radical ideas espoused by Wollstonecraft in 1792 took approximately two hundred years to become felicitous among the powerful. Currently, Professor Avi Loeb is infelicitous among the powerful in academia but maybe, by the twenty-second or twenty-third century, Harvard University will establish the Avi Loeb professorship in extraterrestrial studies; who knows?

Theoretically, this account moves structural constraint from the socialisation, or discursive constitution of the social subject to the problematic of felicitous interaction. Of course, internal restraint is still present as the natural-order-of-things and associated reifications. However, dominant ideology is not dominant in the sense of rendering social subjects epistemic dupes. Rather, social life is characterised by competing ideologies that vie for positions of authority. Ideology or discourse is not some systemic reified entity that is external to actors, subjectifying them, as in Foucault (1982) or Althusser’s account of interpellation. Rather, it is the product of creative agents, acting to gain new forms of authority or to preserve status quo power. This makes social actors significantly more creative than in the Butler/Foucault model, yet the theory does not fall into the trap of voluntarism.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Mark Haugaard

Mark Haugaard is Professor of Political Science and Sociology at the National University of Ireland, Galway, Ireland. He is the founder editor of the Journal of Political Power, published by Routledge, and the book series, Social and Political Power, with Manchester University Press. He has published extensively upon power and his most recent publications include: The Four Dimensions of Power: Understanding Domination, Empowerment and Democracy, 2020, Manchester University Press and “The four dimensions of power: conflict and democracy”, Journal of Political Power, 14:1: 153-175.