Ideology as Relativized A Priori:  On the Mind’s Relation to the Social World 

I.

In this article, our aim is to determine what account of cognition (its nature, structure, and relation to the world) the notion of ideology presupposes. We begin with the widespread idea that ideologies are “worldviews.”⁶ The notion of a “worldview” is a somewhat odd one, for it has at least two incompatible readings. On the first reading, worldviews are interpreted as the subject’s account of how the world is, which is naturally analyzed as a system of beliefs. Thus, several theorists characterize ideology as a “set of beliefs” which are arranged into a “network” or “system.”⁷ For instance, Eagleton calls ideologies “ideas and beliefs,” Stanley focuses on “ideological belief,” while in his influential work Shelby writes that ideologies are “beliefs,” which are not “isolated,” “but always [located within] a network of other beliefs.”⁸ Elsewhere, Shelby writes that ideology is “a widely held set of loosely associated beliefs and implicit judgments.”⁹ Haslanger remarks that one could read such claims as asserting that ideologies are “propositional,” comprising a set of connected propositions believed by the subject.¹⁰ This evokes what we might call a doxastic approach to ideology, on which ideology is entirely analyzed in terms of belief.¹¹

The main difficulty for doxastic approaches is to make sense of the widespread assertion that ideologies are located in the “background” of cognition;¹² that they “lie at the base of our understanding”;¹³ or again, that they don’t “just consist of shared beliefs [but are] a source of beliefs.”¹⁴ This tension is rooted in the two incompatible readings of “worldview.” On the first reading, as we just saw, a worldview is the subject’s account of how the world is. But on a second reading, a worldview does not consist in the subject’s beliefs themselves, but explains why subjects form the beliefs they do. On this reading, we might say, for instance, that it is because a subject has a particular worldview that they have formed a particular belief about a situation: a worldview therefore does not consist in the beliefs that a subject has, but in that which gives rise to these beliefs.

We have called the two readings “incompatible,” but proponents of doxastic approaches might attempt to reconcile them by appealing to the Quinean idea according to which beliefs form “webs,” where beliefs are connected by evidential relations, such that beliefs closer to the center of the web are less easily given up in the face of recalcitrant experience.¹⁵ For instance, Geuss writes that the beliefs which “constitute” an ideology are “central to the agents… in Quine’s sense, i.e. the agents won’t easily give them up.”¹⁶ In this way, ideologies can be said to be located deep in the base of cognition in the sense that they are comparatively robust, i.e. that subjects are reluctant to abandon them; and furthermore, that subjects generate more particular surface-level quotidian beliefs by inference from these deeper ideological beliefs. One would then interpret the claim that ideologies are the “source” of beliefs as the claim that they are an inferential source, and ideologies would be both beliefs as well as the source of beliefs.

But even this kind of account would be susceptible to the general problem that plagues doxastic approaches, namely that, as Haslanger puts it, ideology concerns “the very tools that” subjects have “in order to think.”¹⁷ Ideology cannot simply consist of beliefs which inferentially produce or epistemically support other beliefs, but must, in some sense, make any belief possible at all: it must be a “tool” for constructing beliefs in general. Doxastic approaches cannot make sense of this, for they posit that ideologies are, ontologically, simply beliefs, and not whatever is prior to them. It follows that an adequate theory of ideology cannot have the flat structure posited by doxastic approaches: as Srinivasan suggests, “ideology [cannot] be thought of in the familiar terms of belief [and] evidence.”¹⁸

In response to this problem, commentators have rejected doxastic approaches to ideology in favor of what Dotson and Sertler call “framework approaches.”¹⁹ On framework approaches, ideology is identified with that which produces beliefs rather than with the beliefs themselves, and the production of belief is understood in hermeneutic terms.²⁰ Hall, for instance, argues that ideology consists in “mental frameworks—the languages, the concepts, categories, imagery of thought, and the systems of representation, which [allow us to] make sense of, define, figure out and render intelligible the way society works.”²¹ This idea is taken up by many contemporary theorists of ideology, who posit a “conceptual array” or “conceptual apparatus,”²² a “framework,”²³ “sets of concepts” or “conceptual scheme,”²⁴ a “framing,”²⁵ “conceptual tools,”²⁶ “schedules of salience that operate like meaning-making devices,”²⁷ or “conceptual repertoires”²⁸ as the components of ideology.

More elaborately, Haslanger claims that ideology is “subdoxastic.”²⁹ For her, it constitutes “a language, a set of concepts, a responsiveness to particular features of things (and not others), a set of social meanings,” but also a “network of semiotic relations” or “semiotic net,” or a “web of meanings, symbols, scripts, and such.”³⁰ She writes that an ideology is a “technē,” a term she characterizes as a “placeholder” for the “clusters of concepts, background assumptions, norms, heuristics, scripts, metaphors (and so on) that enable us to interpret and organize information and coordinate action, thought, and affect.”³¹ This “framework” or “scheme” or “web” or “net” is necessary for the formation of beliefs: it provides a space of possibilities in which particular beliefs are formulable. Framework approaches thus substitute the flat structure of doxastic approaches for a layered approach: they theorize ideology, no longer as a mere set or network of beliefs (the first reading of “worldview”), but as that which makes their production possible (the second reading). Ideology, on framework approaches, functions by providing an “interpretation” which allows the formation of beliefs.³²

Framework approaches to ideology are illuminating in two respects. The appeal to concepts, and to hermeneutical resources more generally, allows us to start making sense of the way ideologies enable belief. And the language of “frameworks,” “schemes,” “nets,” and “networks” gives us an idea of how ideologies can generate a system of beliefs. But without further specification, they fall short, especially of this second aim. For on some formulations, such as that of “sets” or of “clusters” of concepts, it is not obvious how ideology can do more than supply building blocks for groupings of beliefs, that is, how it can produce beliefs that cohere into a comprehensive and integrated whole. But on other formulations, such as that of “hermeneutical schemes” or “conceptual schemes,” the opposite problem arises, and the resulting system of beliefs risks being construed as perfectly internally consistent, devoid of any cracks or contradictions. Moreover, the fact that framework approaches identify ideology with hermeneutical resources risks making it difficult to distinguish between the wholesale adoption of an ideology, and the mere use of associated hermeneutical resources. It seems possible, for example, to employ fascistic concepts without thereby seeing the world through a fascistic worldview, if for instance one is a scholar of fascism. But it is not clear how a framework approach which equates ideology with hermeneutical tools can make sense of this distinction.³³ These problems are compounded by the fact that, even though the idea of a framework makes intuitive sense, none of the locutions surveyed here amount to a complete account of frameworks. In what follows, we show that an appeal to the Kantian a priori goes a long way towards formulating a rich framework approach that explains how ideology enables belief while avoiding these two problems.

Before formulating our Kant-inspired framework account of ideology, it will be useful to review Kant’s account of cognition. Kant’s famous Copernican Revolution consists in the revelation that the world is not immediately given to us, but is endowed with its form and shape by dint of our own cognitive activity: instead of presupposing that “all our knowledge must conform to objects… we suppose that objects must conform to our knowledge,” thereby making “the spectator revolve and stars… remain at rest.”³⁴ Objects of knowledge, in other words, are shaped and conditioned by the workings of the mind.

In the first division of the Critique of Pure Reason, Kant famously argues that the subject necessarily produces the categories of the understanding and forms of intuition (space and time) as part of the activity of generating what he calls “cognitions,” which we can think of roughly as isolated truth-apt beliefs. In order to generate cognitions, intuitions (roughly, sense-data) and concepts, including both the categories and empirical concepts (“table,” “chair,” etc.), must be “synthesized,” or combined into one discrete unity. We can therefore think of the production of cognitions, following Kant, as an activity: “we can represent nothing as combined in the object without having previously combined it ourselves,” he writes, and this “is an act of the spontaneity of the power of representation.”³⁵

To indicate that synthesis is a condition of possibility for the production of meaningful cognitions, Kant calls it “transcendental a priori.” But in the Appendix to the Dialectic of the first Critique, Kant advances another a priori condition on cognition beyond this, which he refers to as the transcendental principle of systematicity: our acquisition of knowledge is guided by the ideal of integrating each individual proposition, to which we are separately committed, into a cognitive unity.³⁶ The regulative idea of systematicity is a necessary part of the Kantian a priori: Kant claims that, without this idea and the principles of reason that derive from it, not only would we have “no coherent use of the understanding, and lacking that, no sufficient mark of empirical truth,” but even more strongly, that “no experience would be possible” at all.³⁷ Kant reinforces this point in the Critique of Judgment, where he argues that without the unifying ideal of systematicity, all we would be left with would be a “disturbingly unbounded… heterogeneity of natural forms,”³⁸ or, as Allison puts it, an “empirical chaos,”³⁹ from which no “possibility… of a whole”⁴⁰ could be conceived.

Thus, Kant’s account of cognition works as follows. The subject continuously generates experience by synthesizing intuitions and concepts: it is because the subject is perpetually unifying the manifold of disparate sense impressions that these can take the form of discrete representations that can then figure in truth-evaluable cognitions or judgments with objective purport. Moreover, the particular cognitions that the subject forms are not collected by them into “a raw chaotic aggregate”⁴¹ or mere heap of disparate appearances, but must themselves be integrated into a coherent whole or a unity: isolated cognitions must be continuously made to cohere with the general cognitive framework, governed by the ideal of systematicity, by which the subject takes anything to be sensical or meaningful at all. Each individual representation, in other words, manifests as significant to us only against a cognitive frame or background—one which we produce—that bestows on experience in general meaning and value. At both the level of each isolated experience and the general systematization of all experiences, Kant claims that the world shows up as intelligible to us only through the form conferred on it by our own activity of reasoning. To the extent that this activity constitutes a condition of possibility for the generation of a meaningful system of beliefs, and since Kant takes it to ultimately legitimate the a priori categories and forms of intuition,⁴² we call it the “a priori activity” of the mind.⁴³

We propose to appropriate elements of Kant’s account of cognition by understanding ideology as a form of a priori activity.⁴⁴ On our account, ideology is not a type of object, such as the set of beliefs posited by doxastic approaches, but it is a type of activity. It is perpetually performed by the subject to generate isolable beliefs and to organize them into a system whereby the whole becomes meaningful.⁴⁵ To anticipate our arguments in Section II and Section III, this kind of activity, unlike Kant’s a priori, can be performed in different ways: to say that a subject is in the grips of a particular ideology is to say that they are continuously interpreting and systematizing experience in a particular, contingent way; hence, on our account there is more than one way to synthesize and systematize experience into a meaningful whole.

Our account entails that an ideology is a “worldview” in the second sense mentioned earlier: it plays a hermeneutic role, and thereby explains why subjects hold the particular beliefs they do. As such, our account falls within framework rather than doxastic approaches. As we saw, while framework approaches are often not developed into systematic accounts, “sets” or “schemes” of concepts play a central role in all of the accounts we surveyed. But for us, ideologies consist not simply in sets or schemes of concepts, but in particular ways of generating an intelligible sense of how the world is. In other words, we move backwards from the concepts themselves to the activity that employs them.

This allows us to improve on existing framework approaches by overcoming the two difficulties we presented for them.⁴⁶ Firstly, it allows us to account for the difference between using and applying particular concepts, on the one hand, and adopting the attendant ideology, on the other. There is a difference, for instance, between a scholar of fascism and a fascist, one which cannot be accounted for merely by pointing to the concepts they use. Instead, marking this distinction requires referring to how these concepts feature in the subject’s cognitive ecology. On our account, whether a subject adopts a particular ideology is a matter of how they systematize their experiences and beliefs; it is not only a matter of which particular concepts make up their conceptual repertoire. The subject may well apply a particular concept in an isolated fashion, but if the concept does not play a unifying role in their cognitive activity of integrating new experiences within their preexisting cognitive system, it will not count as structuring the subject’s thought, such that the subject will not be said to adopt the attendant ideology. Our account, by positing that ideology consists, not in the set of concepts or conceptual scheme that the subject uses, but in how the subject systematizes experience using certain concepts, allows us to mark the distinction between attending to and adopting a particular ideology.

Secondly, our account allows us to improve on existing framework approaches by accommodating the fact that ideologies and the beliefs they produce are neither disjointed nor fully coherent. Some expressions of framework approaches, which appeal merely to “sets of concepts” and other aggregates of “conceptual tools,” suggest that ideologies play an important role in the genesis of beliefs, but they cannot account for the fact that ideologies produce unified and comprehensive systems of belief. Our account, by contrast, posits that ideological activities, qua a priori activities, are activities of unification and systematization into a single whole. Other expressions of framework approaches, which invoke “frames” and “schemes” as loci of ideology, risk construing ideology as fixed or static, exhaustive, and internally coherent, when it is clear that ideologies are to some extent malleable and in flux, and that they usually are suffused with internal tensions. On our account, however, a priori activities of the mind perpetually work to integrate disparate elements into a systematic whole. It follows that ideological beliefs can, at any point in time, admit of all manner of fissures and disharmonies, for the activity of systematization does not require that disparate representations already be coherent before the subject begins to work towards their coherent integration. Instead, the systematic ideal of unification provides a standard against which ideology can be said to have cracks and contradictions; by appealing towards the way in which it tends towards an integrated whole, we can explain how it fails to reach that goal.

To conclude, we have begun to show why we might analyze ideology in terms of a priori activity, and more generally, to motivate the move from the flat doxastic approaches to a layered and dynamic framework approach which imbues ideological cognition with a transcendental structure, one which can elegantly explain how surface-level thoughts or beliefs are made possible by the subject’s interpretation of the world.

II.

But our account of ideology as a priori activity faces an immediate challenge: Kant famously takes the a priori elements of cognition, such as the categories, forms of intuition, and ideal of systematicity, to be “absolutely independent of all experience.”⁴⁷ For this reason, he deems them “necessary” and “universal,” and as such, unchanging and absolute.⁴⁸ Moreover, he argues that these necessarily give rise to a particular hierarchical form of cognition, such that we collectively build to one system of knowledge, which necessarily takes the single form of a specific taxonomy of concepts, over the course of human history.⁴⁹ Thus, for Kant, the material world only supplies the manifold of intuition received by sensibility without shaping or influencing the mind’s a priori workings in any respect. So cognition in this sense is “prior” to experience not only in the sense that it’s required to have any experience at all, but also in that only its content, not its structure, is informed by empirical conditions.

But one of the central commitments grounding the classic conception of ideology has been that consciousness must be conceived as conditioned by the world in a much more substantive way. Thus, in The German Ideology, Marx and Engels insist that “man… possesses ‘consciousness,’ but… not inherent, not ‘pure’ consciousness.”⁵⁰ Rather, “the production of ideas, of conceptions, of consciousness, is… directly interwoven with the material activity and the material intercourse of men,” such that “ideology and [its] corresponding forms of consciousness… no longer retain the semblance of independence.”⁵¹ Here, Marx and Engels criticize conceptions of consciousness on which it is unaffected or unconditioned by material reality. They argue that consciousness is instead “interwoven with” material reality. The tradition of ideology theory has retained from them this claim: that the form consciousness has depends at least partly on the material world, rather than being independent from it.

Adopting this view of consciousness requires modifying what is usually taken to be a central feature of Kant’s account of cognition. We discuss the first step of this modification in this section; we turn to a second step—concerning its specifically material dimension—in Section III. We proposed in Section I that ideologies are forms of a priori activity. Given what we have just said about the interwovenness of ideology with material reality, this means that the a priori must come to depend (in some sense) on material features of the world. Since there does not exist a single absolute and unchanging ideology, the a priori activities of the mind cannot be either. Thus, the a priori must be relativized to these features: since it must be responsive to the world, it cannot be simply given to consciousness, as Kant had argued, but must instead be made contingent, and, as we conclude below, potentially revisable.

As Friedman shows, the idea of what he calls the “relativized a priori” has an important place in post-Kantian philosophy of science.⁵² Reichenbach remarked that “Kant’s concept of a priori has two different meanings. First, it means ‘necessarily true’ or ‘true for all times,’ and secondly, ‘constituting the concept of object,’ that is, ‘order[ing] the perceptions according to a certain schema.’”⁵³ In other words, Reichenbach proposed to separate the apodictically certain from the transcendental. Drawing this distinction opened up the prospect of retaining Kant’s transcendental structure for cognition—conditions of possibility for belief could be posited—whilst rejecting the necessary and unchanging nature of these conditions. Indeed, “since deep conceptual revolutions or paradigm-shifts are a fact of scientific life… we are never in a position to take our present constitutive principles as truly universal principles of human reason—as fixed once and for all throughout the evolution of science.”⁵⁴

Thus, Friedman argues that we should understand Kuhn’s paradigms⁵⁵ as instances of the relativized a priori—as contingent, revisable constitutive principles that constitute the conditions of possibility for cognizing the natural world. He writes that “what we end up with, in this tradition, is… a relativized and dynamical conception of a priori mathematical-physical principles, which change and develop along with the development of the mathematical and physical sciences themselves, but which nevertheless retain the characteristically Kantian constitutive function of making the empirical natural knowledge thereby structured and framed by such principles first possible.”⁵⁶ This idea of the relativized a priori need not be restricted to natural science. For instance, the early Foucault’s notion of the “historical a priori,”⁵⁷ which is also drawn from Kant,⁵⁸ is a relativized a priori applied to the social world.⁵⁹

There is a long tradition, therefore, of relativizing the a priori: of retaining the idea that experience has constitutive conditions, whilst rejecting the idea that these conditions are necessary and universal. But both Friedman and Foucault operate with a different understanding of the a priori than we do: where they parse the a priori in terms of constitutive principles (Friedman) or discursive rules (Foucault), which are presumably static, we proposed to remain more faithful to Kant by following these constitutive conditions back further to the dynamic activity of the mind. Thus when we argue that the a priori should be relativized, we mean that the activity of sense-making in which the subject perpetually engages can be performed in different ways: contrary to what Kant had claimed, there is not one unique way of synthesizing and systematizing cognitions into a single whole.⁶⁰ So, an ideology is a particular contingent way of imbuing the world with meaning.

For example, a subject deals with the experience of caring for an elderly parent. She must systematize this experience with all of her other experiences: those of caring for her children, of relating to the state, and so on. She might interpret caring for her parent as normal and natural, the proper way to repay her debt to them for having raised her, something she herself will one day be entitled to from her children and that does not concern the state or anyone else. Alternatively, she might interpret her care as one of solidarity with another person who has been abandoned by the state and larger community: not as something natural, but as something done out of the joint recognition that solidarity and mutual aid can fill in the cracks. In either case, she faces the need to unify all of her experiences, and she does so in different ways, thereby performing different actions, forming different explanations of why she needs to care for her parent, and so on. She may employ different concepts in each case, but this difference alone does little to account for what distinguishes one case from another: we need to appeal to the subject’s different ways of systematizing her experiences.

With the a priori relativized, we arrive at a developed framework account of ideology. As such, we can reap some of the benefits of framework accounts to illuminate ideology’s central features. For instance, despite often being identified with “false consciousness,” theorists have maintained that ideology is not false outright.⁶¹ Shelby writes that “a form of social consciousness may be ideological in ways that are not fully or accurately conveyed by simply calling [it] ‘false.’”⁶² Instead, it “misrepresent[s]… realit[y],”⁶³ it operates “through some form of masking,”⁶⁴ it is “misleading”⁶⁵ or “deceptive”⁶⁶ in some way; or in the most frequently used terms, it is an “illusion”⁶⁷ and a “distortion”⁶⁸ of reality. Yet there is also a way in which ideologies can be true. As Wills explains, “false consciousness gets things ‘right’ at the level of appearance, but it mistakes that appearance for a ‘deep’ or essential truth.”⁶⁹ As Eagleton puts it, “some at least of what we call ideological discourse is true at one level but not at another: true in its empirical content but deceptive in its force, or true in its surface meaning but false in its underlying assumptions,” such that “not all ideological language characterizes the world in erroneous ways.”⁷⁰ Jaeggi concludes that “this would mean that in the case of ideologies, we are dealing… with a peculiar inadequacy of the criterion of truth.”⁷¹

There is much to be said about what it might mean to say that an ideology is false but not quite, or true but not quite, and we return to this question in Section III. But the view that we have constructed so far, qua framework approach, already allows us to go some way towards understanding this feature of ideology. In constituting the conditions of possibility for belief, the a priori activity of cognition produces the propositions that may then be believed or disbelieved by the subject; that is, it creates what can show up as what Hacking calls a “candidate for truth.”⁷² It makes possible the very judgments that show up as true or false in the first place; it establishes what counts as “true-or-false.”⁷³ In this way, ideology concerns, not the dimension of truth, but the dimension of intelligibility.

Our account of ideology as (relativized) a priori, in virtue of being a developed framework approach, can illuminate several further features of ideology. Firstly, it helps us to see how ideology guides what is salient. Indeed, it is commonly claimed that, under ideology, “aspects of reality [are] masked or obscured,” or “inconspicuous and almost invisible”;⁷⁴ that ideology “guide[s] our attention in ways that occlude important and valuable features of the world.”⁷⁵ If we take consciousness to be conditioned by a priori activity, shifting our focus away from the dimension of truth to the dimension of intelligibility in the way we’ve just discussed, this allows for an explanation of why ideology guides salience: particular features which are recalcitrant to integration into the subject’s systematic worldview will show up to the subject as less intelligible, less sensical—or might even be unintelligible, nonsensical, or simply elude perceptual attention at all. Thus, the more intelligible features, those that make more sense within the subject’s worldview, will be more prone to rise to the level of attention or to show up as salient. What is less often noted by its theorists is that ideology might also make certain features of the world hyper-salient, if the a priori activities of sense-making make these features particularly conspicuous. (Note for instance just how salient gendered characteristics of individuals are under patriarchy.) Thus, in full generality, if ideology operates transcendentally, it can affect what of the world is apparent to the subject and what is not.

Moreover, our view can accommodate the fact that, in ideology, “description and evaluation intermingle,” such that “ideology is always already both an understanding and an evaluation” and ideology critique is inevitably “normatively significant.”⁷⁶ This “intermingling” of the descriptive and the evaluative derives from the fact that ideology shapes the boundaries of the modal: ideologies “stake out the field of possible actions”; they “determin[e] and limit… possibilities”;⁷⁷ they “distort… what is possible,… what is natural or decreed by God.”⁷⁸ This proceeds through mechanisms such as “naturalization—something socially ‘made’ is imagined to be something naturally or irreducibly ‘given,’”⁷⁹ or again “by creating the illusion that… relations (or their causes), which are actually the product of historically contingent human action or convention, are ‘natural’ and, thus, ineradicable, unavoidable, and unalterable.”⁸⁰ Our view of ideology as a priori activity can accommodate this feature: by giving rise to what is thinkable for the subject, it produces the form of the subject’s space of possibility. And in producing what it is possible for the subject to do, as well as what is possible to think, ideology inevitably has normative consequences. As Hacking claims, if the a priori “represents what is held to be thinkable, to be possible,” then relativizing it entails that “what may be deemed possible at one time may not be held to be so at another.”⁸¹

Thus an agent in the grip of a particular ideology may see things as necessarily related, or be able to conceive of no alternative to a situation. This partly explains how ideology has a restricting power on subjects: it renders contingent situations as more necessary, or loosely related facts as tightly fastened. Hence the liberatory potential of emancipation from ideology, where “the given becomes the contingent.”⁸² It should not be forgotten, however, that if ideology has the power to redraw modal boundaries, this is not restricted to making contingent facts appear necessary: it also makes necessary facts appear merely contingent. For instance, the systematic exhaustion of workers is not as contingent a feature of each of their lives as capitalist ideology would have it, but indeed can be recognized from a critical perspective to be a much more necessary consequence of their shared situation.

This brings us to a further feature of ideology that our account illuminates, one that we have already begun to discuss in Section I: that ideologies are located in the “background” of, or “deep” into, the agent’s consciousness.⁸³ Shelby explains that an ideology is “often held without full conscious awareness,” such that a subject will not be “aware that he is in [its] grip” and that it is “deeply entrenched.”⁸⁴ These two features are well-explained by our account. For if ideology organizes experience, thereby already imbuing it with given meanings, the experiences of a subject in the grip of an ideology is liable to have will not stand in significant tension with it. It follows that it will be difficult for the subject to be aware of the presence of the specific ideology at hand. Kuhn makes a similar point about scientists working within a particular paradigm: when in too flagrant a contradiction with the paradigm, anomalous features of scientific experience are sometimes hardly visible or apparent, tend to be assimilated into the paradigm as either a local or an inconsequential anomaly, or might not even be noticeable as such. It is only with the formulation of the new paradigm that the full significance of the anomalies comes into view. This begins to reveal that ideology has a self-confirmatory aspect.⁸⁵

Yet, unlike the paradigms of natural science and in virtue of their social character, ideologies are not just self-confirmatory because they are prior to the experiences they frame, but also because they materially produce their own truth. The subject thus genuinely does get corroboration of their ideology as they go along in the world, thereby explaining why it is so entrenched. But it is not readily apparent how the relativized transcendental picture we’ve presented so far can make sense of this way in which ideology is self-confirmatory. So, the relationship of consciousness to the material world is more complex in the context of social theory than, for instance, in that of natural science: it is not merely one of “world-to-mind,” but also one of “mind-to-world.”⁸⁶ In Section III and Section IV, we move to the question of this relation.

III.

In the previous section, we proposed to relativize the a priori. In this section, we take a further step in materializing it: we describe the way in which it is conditioned by material reality, and the way it conditions it in turn. This is necessary if our account of ideology is to accommodate the materialism of Marx and Engels mentioned in the previous section, or, as MacKinnon puts it, the way in which “mind and world [are] interpenetrated.”⁸⁷ There is, we show, a loop between the two. On the one hand, ideology produces its own material truth: ideology does not just shape experience, but also shapes the world. On the other hand, the material world produces ideological forms of thought; it does not simply supply the content of experience, but also shapes its structure.

Let us begin with the first step of this loop: the fact that ideology makes itself true, that it “makes real what it purports simply to describe.”⁸⁸ The idea is that ideology “structure[s] the material relations in which all people are forced to participate”; that it has “constitutive effects, bringing into existence new things or making things true.”⁸⁹ Let us illustrate this with what might be called family ideology, on which the proper location of care and reproductive labor is the private nuclear family.⁹⁰ Thinking in the terms of family ideology—an ideology that produces the belief, for instance, that the conditions for subsistence and flourishing are to be found within the confines of the private nuclear household—leads one, as a matter of course, to organize one’s personal life so as to bring about the existence of a family that could serve this function. The more individuals think and act this way, the more care and reproductive work will indeed be confined to the private nuclear family. Thus, the more widespread family ideology becomes, the more entrenched the family is as the way in which care and social reproduction are organized, and the truer family ideology becomes.

This brings us to the second step of the loop. The more care is organized around the institution of the private nuclear family, the more intimately subjects come to cognitively associate care with the nuclear family. Thus, the way that subjects cognize care and social reproduction becomes shaped in the form of the family, according to ideas of (heterosexual, monogamous, cohabiting, lifelong, procreation-oriented) “romantic” and (unconditional, property-conferring) “filial” love; of care as “duty” to the members of one’s own family and as very much not one to others; of “home” as not simply housing, but the place to which one “belongs”; of the line between safety and danger as the line around the family (“stranger danger”); and more. This way of making sense of social reproduction focuses the subject’s attention on caring for and sharing resources with their parents, spouse and children, refracting away from other possible recipients of care. Thus, the couple or the nuclear form comes to appear in the mind as the primary, and often only, possible site for sustenance. The more family-shaped the material social world is, the more family-shaped the structure of cognition becomes.

This second step of the loop is the main epistemological insight of the German Ideology. There, Marx and Engels propose to “explain the formation of ideas from material practice,” to “show that circumstances make men just as much as men make circumstances.”⁹¹ As we indicated in Section II, they oppose what they call idealism, where consciousness is “inherent” and “pure,” that is, has an “independent existence”; and argue instead for materialism, the view that “from the start the ‘mind [Geist]’ is… ‘burdened’ with matter.”⁹² There, we established that the a priori can no longer be conceived, as Kant held, to be given necessarily to consciousness, but must be (in some sense) responsive to the world. We can now precisify this point, by turning to its specifically material dimension.

As we saw in Section II, our view of ideology as a priori entails that we should interpret the materialist thesis as stating not only that the content of thought is partly determined by the material world, but, more radically, that its structure is too. Thus, what depends on a subject’s material conditions is not simply the choice of which option to believe in a series of well-formed alternatives—of incompatible, already expressed judgments about the world, such as “there is a duty to care for one’s children” versus “there is no such duty”—but the very way in which those alternatives are formed in the subject’s consciousness (e.g. the conceptualization of meeting others’ need for care as an individualized “duty”). In the terms we have been employing, materiality conditions not simply which beliefs an agent adopts, but which propositions are even candidates for belief: it conditions the subject’s a priori. This is the way, we suggest, in which “material life structures understanding”:⁹³ the a priori, relativized as we propose, is not constructed by the subject in separation from the world, but is, at least partly, shaped directly by the material conditions of their existence.

The materialization of the relativized a priori allows us to add another dimension to our understanding of the three features of ideology above: the fact that they are entrenched, that they are both true and false, and that they are unconscious. To begin, it explains how ideologies become so entrenched. Subjects in the social world find themselves needing to act in accordance with these ideologies, which then shape the social world in turn, in such a way that their own cognition—as responsive to the world they live in—comes to accord with the ideology, and so on. This explains their persistence and apparent inexorability. For instance, we live in a world where the architecture of private housing, tax and inheritance law, and more make it materially advantageous to organize one’s care life in a family shape. Provisions for elderly people and childcare, the resources required for emotional and mental flourishing, and material sustenance are not easily available outside of the family. Houses are built to nuclear family dimensions: one kitchen, one dining room, a couple-sized main bedroom with child-sized smaller bedrooms. Financial arrangements outside of government-sanctioned family ties (marriage, partnerhood, parenthood) are significantly more complex. And the predominance of the nuclear family strengthens the bonds of dependence between its members, making it harder to establish relations of care outside of its confines. These material constraints make it hard to think outside of the family form, which then entrenches these constraints further.

It also allows us to make deeper sense than we could in Section II of how ideology is both true and false, or neither quite one nor the other. MacKinnon, as we have already seen, describes ideology as a “myth”; but later she also writes that it is “hardly a myth.”⁹⁴ Jaeggi calls this a “paradox”: “How can an ideology be at once true and false?… If ideologies are supposed to be simultaneously true and false, might one not claim just as convincingly (or even more so) that they are neither true nor false?”⁹⁵ The quasi-truth and quasi-falsity of ideology, she responds, are not independent, but are instead “interpenetrati[ng]” or “entangle[d],”⁹⁶ such that ideology “is as much a separation from the full truth as from the mere lie.”⁹⁷

This can be gleaned from the example we raised above. On the one hand, under present conditions, it is true that many individuals would do better (in some sense) to enter into nuclear-familial relations than to abstain from them. In such conditions, participating in the nuclear family is clearly in their individual self-interest, and can therefore present itself as a form of the good. On the other hand, critics have argued that the nuclear family, because of its function of privatizing these forms of care, is very often a site of intense stress and scarcity, thereby making it especially unlikely to fulfill even its self-interested promise; for many less likely indeed, they argue, than forms of social organization where such care is collectively organized. Moreover, since the pervasiveness of the family form functionally hinders the availability of such material and emotional care outside of it, it is clear that this form of social organization cannot be in the general interest. In this sense, then, family ideology’s pretense to the good is false.⁹⁸ And the truth and falsity of family ideology are not separate but interwoven. For the more entrenched the family (and therefore the truer the associated ideology) is, the more scarcity-ridden and exclusionary it (the falser the associated ideology) becomes. And the more stress and scarcity people find themselves under—the falser it is—the more prone they will be to seeking it out for themselves: the more true it becomes.

Finally, a materialist theory of ideology allows us to make sense of a deeper way in which ideologies are unconscious. We saw that if ideology is a priori activity, it will be seemingly remotely located in the subject’s cognition (Section I), and it will be self-confirmatory in a way that makes it hard to discern (Section II). But theorists of ideology speak of it as unconscious in a way that goes beyond these two points: Geuss describes ideology as the “unconscious determinant of [subjects’] consciousness,” MacKinnon as “thought that is socially determined without being conscious of its determinations,” and Lukács as “a class-conditioned unconsciousness of one’s own socio-historical and economic condition.”⁹⁹ This way of conceiving the unconsciousness of ideology can be traced back to Engels: “The real motive forces impelling [the subject in the grip of ideology] remain unknown to him; otherwise it simply would not be an ideological process,” since “the so-called thinker [in the grip of ideology] works with mere thought material… and does not investigate further for a more remote source independent of thought.”¹⁰⁰ Subjects in the grip of ideology are unconscious of the fact that their way of organizing experience is conditioned by the material world. This creates the illusion that their thought is the product of their own reason, the result of their “pure consciousness.”¹⁰¹ The content of their thought thus appears to be generalizable and universalizable rather than particular and historical: ideological forms of consciousness have the “form of universality,” and wrongly appear to be “the only rational, universally valid ones.”¹⁰²

It remains to be explained how, precisely, the account of material determination by social arrangements, practices, and processes that we have begun to construct in this section is supposed to work. That is what we will now turn to.

IV.

Material determination is sometimes understood as determination by “structures,” “facts,” “institutions,” or the “material world.” For instance, it has been claimed that ideologies are determined by “the causal structure of the social phenomena,”¹⁰³ by “coercive social institutions,”¹⁰⁴ by “the material structure of society as a whole,”¹⁰⁵ or by “the material world.”¹⁰⁶ These claims could be read as parsing material determinants as things or facts: it is how the world is that determines consciousness. In fact, we have sometimes used similar terminology. However, the terms that Marx and Engels regularly invoke in the German Ideology suggest a different sense of material determination. They write that consciousness is conditioned or determined by the “life-process” and “production of life” and “material behavior” and “material activity” and “practical activity” of men, which is a social enterprise.¹⁰⁷ This is because Marx is adamant that we are continuously making and remaking the world as it is: there is no social world but what we do. So, to say that consciousness is materially conditioned is to say that it is conditioned by practice; not by how things are but by what people do.¹⁰⁸

Of course, what we can do, especially individually, is constrained by class, but also by race, gender, and so on. Marxists insist that the form these constraints take can, in part, be explained by the role they play in the overall mode of production.¹⁰⁹ This is immediate in the case of class, but it also holds, for instance, of the family form: according to Marxists, the private nuclear family exists in the form that it does because, on the one hand, it enables inheritance, which is necessary for the reproduction of the bourgeois or capitalist class, and on the other hand, it privatizes social reproductive labor, including but not limited to the production of future workers and the sustenance of existing ones.¹¹⁰ The practices of kinship-making that are available to people and which shape how they come to cognize care, are, in part, products of the political economy. More generally, the social practices which condition consciousness are themselves constrained and shaped by the mode of production.

Yet questions remain about how social practices can condition consciousness in the way Marx and Engels describe. In places in the German Ideology, they describe thought as the “reflexes and echoes of [the subject’s] life-process,” “the direct efflux of their material behavior,” “phantoms” and “sublimates formed in the human brain,” akin to the “inversion of objects on the retina.”¹¹¹ In these phrases, ideological consciousness is described, on the one hand, as a merely epiphenomenal byproduct of material processes, as if thoughts were not real or concrete but merely arose out of that which is (they are “phantoms,” “sublimates,” “echoes’). On the other, it is couched in physiological terms, as if thoughts were caught in and exhausted by the all-encompassing deterministic web of cause and effect (“reflexes,” “efflux,” “objects on the retina,” “on the human brain”). These two sets of images stand in tension with one another in some sense, but both contribute to a picture of ideological thought as non-agential, as if all that is effectual is so merely causally or mechanistically. Taken together, these passages thus suggest that the material conditioning of consciousness occurs as a deterministic derivation from material practice.

This view, in a characterization we will return to, is what Engels himself later terms “mechanical materialism,”¹¹² and indeed the Marxian tradition has long opposed Marx and Engels’ own initial formulations of the conditioning thesis, primarily contesting the first sense there given to ideology as imaginary or epiphenomenal consciousness. As Williams argues, “the language of ‘reflexes,’ ‘echoes,’ ‘phantoms,’ and ‘sublimates’ is simplistic, and has in repetition been disastrous.”¹¹³ Thus Althusser famously renounces “the notion that [ideology] consists simply of a collection of distorting representations of reality and empirically false propositions,”¹¹⁴ and argues instead that a subject’s “ideas are his material actions inserted into material practices governed by material ‘enmeshment’ which are themselves defined by the material ideological apparatus”—such as the church, the state, or the school—“from which derive the ideas of that subject.”¹¹⁵ The subject is “constituted” by these “material ideological apparatuses” and indoctrinated into ideological “rituals.”¹¹⁶ In this way, Althusser “invert[s] the notional schema of ideology,” endorsing the following “script,” as it were, as an example of the ideological process: “Kneel down, move your lips in prayer, and you will believe.”¹¹⁷ That is, he displaces the locus of the ideology theorist’s interest from the consciousness of the individual to the social phenomena which give rise to it.

An important strategy in the ensuing tradition has been to place “the emphasis on [the] social phenomena”—primarily language (“signifying chains,” “symbol-systems”) and shared action (“the rituals and practices of social action and behavior”)—in which “ideas appear, where mental events register or are realized.”¹¹⁸ In this way, ideology is not construed as a mental event, but is situated within the “material social process of signification itself.”¹¹⁹ Similarly, Foucault moved away from the concept of the historical a priori altogether as an “examin[ation of] the forms [of thought] themselves,” as he later commented on his own life’s work, to an analysis of “their formation out of practices and the modifications undergone by the latter.”¹²⁰ Thus, although he saw his concept of the historical a priori as continuous with his later work, he never returned to the language of the a priori, putting priority instead on practice.¹²¹

Haslanger, who understands herself to be following both Althusser and Foucault,¹²² insists that “ideology [is] not simply located within individual minds,” and instead exhorts her reader to attend to “social practices,” which she calls “ideological formations” as well as “cultural software.”¹²³ Practices, she writes, are “partly constituted” by “concepts, rules, norms, stereotypes, scripts, and the like,” which supply something like their ideological content.¹²⁴ She commonly cashes out these components in terms of “schemas,”¹²⁵ which, she states, “encode knowledge and also provide scripts that frame our interaction with each other and our environment; such scripts can guide group members through collective events or even organize a life.”¹²⁶ These schemas come to exist because of “our need to organize information and coordinate action, thought, and affect”; thus, these “signals” are “learned through socialization.”¹²⁷

It has even been argued that ideology is wholly attributable to the role schemata and scripts play in the social sciences.¹²⁸ According to the work relied on to support this account of ideology, these schemas are “prescriptive sequences” that “people automatically engage in.”¹²⁹ They “are like theatrical ‘scripts’: once we know which role we are playing, we just follow the script, acting in ways appropriate to our roles, without really thinking much about it.”¹³⁰ For instance, the fact that “when people go into a restaurant, they know what is likely to happen and hence they know how to behave appropriately” is explained by appealing to “different but relatively similar restaurant scripts stored in each of us,” akin to how “the software that is installed on a computer… allows a computer to process information.”¹³¹

But there is a risk, in insisting on the primacy of practice, of falling into the very trap one was attempting to avoid. For in one’s efforts to reject the first sense of ideology as imaginary, epiphenomenal, or unreal (suggested by the terms “echoes,” “phantoms,” and “sublimates”—terms Williams deems a “deliberately degrading vocabulary”¹³²), one risks excising consciousness altogether, and ending up reproducing ideology in the second sense of its being programmed, predetermined, or the result of a physiological process. Where, in some of Marx and Engels’s passages, consciousness is the mere “reflux,” “efflux,” inscription “on the human brain” or “on the retina” of material processes, such a position risks reducing ideology to merely “habitual responses,” “cultural software,” and “automatic,” “prescriptive sequences.” In so doing, it reproduces a figuration of the subject undergoing ideology as a mere automaton incapable of agential engagement with the world. Williams contends that this stark picture, pitting the ideal against the real, amounts to a “naive dualism,” “in which the idealist separation of ‘ideas’ and ‘material reality’ ha[s] been repeated, but with its priorities reversed”:¹³³ it risks facilely suggesting that ideas and values can simply be read off of material processes.

The ensuing tradition of ideology theory has had to contend with a way out of this seeming opposition. As Shelby claims, “ideologies cannot have their peculiar and profound social impact without being received into the consciousness of human beings”;¹³⁴ so, Hall notes, “the problem for a materialist or nonidealist theory is how to deal with ideas, which are mental events, and therefore, as Marx says, can only occur ‘in thought, in the head’ (where else?), in a nonidealist, nonvulgar materialist manner.”¹³⁵ Ideology must therefore be conceived in a non-reductionist fashion, such that agents are not demoted to “actors who can only recite preexisting scripts” or to the status of merely “receptive organ[s].”¹³⁶

We can find the schematic outlines of such an account within Marx and Engels’ own writings. Engels rejects prior conceptions of materialism, as we mentioned above, as “mechanical,” akin to the reductionist mechanical views of early modern natural science. He claims that this conception of nature as a collection of objects whose movements are entirely governed by deterministic laws, explanations of which are limited to brute causality, was extrapolated to human cognition: “What the animal was to Descartes, man was to the materialists of the eighteenth century—a machine.”¹³⁷ On such a picture, thoughts would be understood as mental objects, “activated” by external stimuli. But Marx and Engels exhort us to understand the material conditioning of consciousness not in terms of mechanical, deterministic processes, but in the organic terms of “life-processes [Lebensprozeß] ” of the “production of life,” of the “practical” and “material activity [Tätigkeit]” and of the “engagement [Verkehr]” of men.¹³⁸ Yet even though they generally refer to the determination of cognition by life processes or practical activity, their usage has been eclipsed in the literature.¹³⁹

Here, as Marx had already made explicit, the domain of “life” comprises not just practical but also conscious activity: if “labor” is “life-activity, productive life itself,” the human subject also “has conscious life-activity.”¹⁴⁰ As Williams explains, “consciousness is seen from the beginning as part of the human material social process, and its products in ‘ideas’ are then as much a part of this process as material products themselves.”¹⁴¹ Thus, a more sophisticated materialism does not grasp either the ideational or the material in the “mechanical” terms of mental objects in a remote realm on the one hand, and scripted social practices on the other. Instead, it conceives of both as of a single kind: as activity. Where the material is conceived in terms of practical activity, including, importantly, productive activity or labor, the ideational is conceived in terms of conscious activity.

Moreover, conscious and practical activities are not separate, but in some sense enmeshed. Marx and Engels write that “men, developing their material production and their material engagement, alter, along with this their real existence, their thinking and the products of their thinking”; or again, that “the production of ideas, of conceptions, of consciousness, is… directly interwoven with the material activity and the material engagement of men, the language of real life.”¹⁴² It follows, in the words of Wills, that “ideas are not epiphenomenal to matter; they are an inseparable aspect of one unified whole made up of both matter and ideas.”¹⁴³ If we can account for the “inseparability” or “interwovenness” of conscious and practical activity, we will therefore have found, in Marx and Engels, some resources for avoiding the dualism of both idealism and mechanical materialism.

Our solution appeals to the a priori activities of the mind, and it is therefore not available to framework approaches that do not parse ideology in terms of cognitive activity. Marx and Engels are naturally read as holding that the practices of human beings necessarily presuppose attendant conscious activity and vice versa. For practical engagement with the world requires, at the very least, that the part of the world the agent practically engages is at least somewhat intelligible to them: an unintelligible world, an “empirical chaos,” is not one an agent can intentionally interact with.¹⁴⁴ So a priori activity, because it structures the intelligibility of the context of action, is a necessary condition for practical activity; in this sense, it is prior to practical activity. But if we need to frame, systematize, and interpret experience in order to make action possible, this must also respond to one’s practical needs, for this hermeneutic activity does not happen in a vacuum, but is perpetually responsive to the practical problems that the agent who engages with the world encounters: it is because the subject interacts with the world that they need to produce particular interpretations of it. Again, practical problems present the need for intervention into the world, which is what prompts the specification and alteration of one’s sense of how the world is, such that the subject produces the interpretations that will make responding to their practical problems possible.

Think, for instance, of a woman who finds herself needing to care for a newborn. When she turns to the world to organize their care, she finds that the only option is to do it herself; when she asks for help, people suggest she turn to her parents and siblings; childcare support in all its forms only ever deals with “parents”; as she becomes recognized by the state as the newborn’s “mother,” she might access tax breaks and paid leave; when she looks for housing, she finds that her only options are small privatized units; and so on. In an attempt to make sense of these experiences together, she comes to understand (though not necessarily in these terms) social reproduction as organized according to the institution of the private nuclear family. This organizes her way of thinking about everything; she might for instance come to think of care as the responsibility of individuals, and form the belief that she is responsible for her newborn’s care; she might come to cognize children as necessarily related to adults in a “child”/“parent” relation, and form the belief of Alyssa that she’s John’s “daughter,” and that together with Norah, John’s “wife,” they make up “the Wilsons,” a unit identifiable by the family relation.

At the same time, she is not coming into the situation wholly fresh to the world, with no prior interpretation that has guided her past experience, but she already has an understanding of care as organized by the private nuclear family; the fact that subjects form their understanding of social reproduction from the youngest age is what makes illustrating our claims difficult. What the example shows, despite this limitation, is that on the one hand, she needs to understand the privatization of reproductive labor in the form of the private nuclear family in order to effectively care for her newborn; and on the other hand, it is from her need to care for her newborn that the impetus comes for understanding how care is organized: her hermeneutical activities are both necessary for action, and responsive to the practical problems that present themselves in action. And as her practical problems evolve, so do her sense-making activities along with them.

Thus, a priori and practical activities are “interwoven,” or enmeshed in a kind of dynamic co-alteration, where one enables the other and vice-versa: Wills speaks of a “metabolism between human minds and the world of which they are a part.”¹⁴⁵ We see this “metabolism” as an improvement on the “loop” of Section III: it is not so much that action in accordance with ideology shapes the social world, which then shapes one’s cognition of the world, and so on, but instead, that one’s cognition of the world is always responsive to and bound up with one’s action within it, so that the neat division between the ideational and the material that we assumed in Section III, and found again in both idealism and mechanical materialism earlier in this section, dissipates. If the ideational and the material are enmeshed, the products of cognitive activity, such as beliefs, are not mere reflections of the world, but are both expressions and parts of particular social practical activities—results not of the subject reading facts off the world, but of their perpetual interaction with it. Against a conception of materialism as mechanical, we have therefore arrived at a conception of materialism consistent with positing transcendental features of the mind.

V.

Throughout this article, we have developed a view on which ideology is a conscious activity—enmeshed with practical activities—of synthesizing and systematizing experience into a general worldview. Although this activity is one that is conditioned and constrained by material social practice, it is also one over which subjects can in principle come to exert agency, thus opening up the possibility of revising the a priori, i.e., of ideology critique.

As we discussed in Section I, it is hard to see how static, coherent frameworks can contain the tensions and contradictions that would motivate the need for critique. Similar problems arise for Friedman and Foucault, whose work we drew on in Section II. Friedman’s relativized a priori, because it is composed of static and coherent constitutive principles, entails that the “dynamical” nature of the a priori must be parsed in terms of a “convergent sequence of successive” constitutive principles.¹⁴⁶ But it is unclear how the succession from one discrete set of constitutive principles to another comes about.¹⁴⁷ Friedman does once refer to “an intermediate stage in which we are still in the process of (continuously) transforming the earlier framework but have not yet clearly articulated the later one,” but it is not clear that this is consistent with his broader account, since the constitutive principles he invokes are both necessary for all theorizing and always fully articulated.¹⁴⁸ Foucault’s historical a priori improves on Friedman in that it is pervaded by internal tensions; for instance, the contemporary a priori, as ordered around the human subject, is not at all internally stable or consistent, but is, according to Foucault, “about to topple.”¹⁴⁹ However, Foucault’s concept faced the common criticism, including during his own lifetime, that it could not account for change or moments of transformation between a prioris (or epistemes) and hence for critique. For instance, Sartre famously objected to it on the grounds that it consists in a mere “succession of snapshots [succession d’immobilités],” and as such, cannot explain “how people move from one mode of thought to another.”¹⁵⁰ Ideological frameworks, constitutive principles, and epistemes therefore all face similar problems: to the extent that they are coherent, it is unclear where the impetus for critique comes from; and even when they are not, their discrete and static nature raises the question of accommodating the possibility of transformation.

By contrast, our appeal to a materialized a priori gives us the resources to understand why critique occurs and how it is possible. In Section I, we insisted that the a priori should be understood as an activity: as the activity of synthesizing and systematizing disparate experiences into an integrated whole; of turning an “empirical chaos” into a unified and meaningful perspective on the world. We explained that to identify ideology with this a priori activity entails that it can be riven with cracks and tensions, and that it is also subject to a dynamic attempt at integration. This is important in two respects when it comes to understanding how critique is possible. Firstly, because our account can accommodate the contradictions of ideology, it can explain why subjects would want to engage in critique in the first place: it is because ideology is riven with tensions that the need for a critique is posed. Secondly, our account already contains the seeds of how critique can proceed: in the face of recalcitrant particulars, it is perpetually necessary for the mind to find new ways of unifying experience beyond those that have proved unable to reconcile disparate elements.

We draw once again on Kant to explain how these new ways of unifying experience can be found. Kant argues that reason inevitably gives rise to a priori “illusions,” or cognitive distortions; in the face of these illusions, Kant claims that we can reorient our reason in order to make sense of experience in a new way. In order to do this, Kant claims that we should employ its ideas regulatively, where they are treated as concepts that can never be fully given rather than as determinate universals.¹⁵¹ As we have discussed, the regulative idea of systematicity is an aim that, while unattainable, perpetually motivates us to reach it. The ideas point to something more than what can be contained within the limitations or boundaries of our cognition, as what exceeds or outstrips what our finite cognitive powers can represent; they therefore call on us to stretch “our imagination in all its boundlessness.”¹⁵² In order to attempt to grasp these ideas, cognition exhibits a self-endowed “autonomy” which, in the absence of a determinate direction in which to proceed, reflexively “prescribes” cognitive principles “to itself.”¹⁵³ The arrangement of experience into a system is therefore an activity in which cognition proceeds “not merely mechanically, like an instrument, but creatively.”¹⁵⁴

We suggest, by way of concluding, that this Kantian account can be extended to a fuller account of critique, understood in terms of epistemic agency. Kant took the reorientation of reason to be limited to the prescription to employ the ideas regulatively. But given our materialization of the a priori, we suggest broadening Kant’s call for a regulative employment from the individual ideas of reason to the entire a priori. By first relativizing the a priori, then conceiving it as not just a static framework but as an ongoing activity of the mind, we can make sense of a revisable a priori, one whose ordering of disparate parts of experience into a whole can be reoriented. The a priori, or the “idea” of the “form of the whole of cognition,”¹⁵⁵ can thus be seen to be perpetually subject to reorientation: if a salient particular fails to conform to the conception we have of our experience as a whole, we can either reject the recalcitrant particular, or alter our general grid of intelligibility, as generated by the activity of the a priori, in order to accommodate it. As such, the now revisable a priori can be reoriented, in open-ended fashion, to produce a new idea of the whole or idea of the system, making the world show up as intelligible or salient in new ways. With it, possibility is opened for agency over the a priori, and therefore, for agential intervention into the “loop” between mind and world or the “metabolism” of conscious and practical activity. Subjects can therefore employ their epistemic autonomy to open up new possibilities for thought and practice instead of reinforcing the old ones, forming the world anew, in light of a vision of what it might be. It is in this way that the activity of bestowing intelligibility on experience can be employed in the service of critique.

ACKNOWLEDGMENTS

This article sprung from a Research Master’s seminar we co-taught with Titus Stahl in 2021 at the University of Groningen, and we are very grateful to Titus and to the seminar participants for rich discussions that helped us find the impetus for this article. We are deeply grateful to Liam Kofi Bright, Robin Celikates, Matthew Congdon, Nicolas Côté, Dusty Dallman, Lidal Dror, Daniele Lorenzini, John Rufo, Daniel Singer, Amia Srinivasan, Titus Stahl, Tuomo Tiisala, and Ege Yumuşak for very helpful comments on previous drafts of this article. We are especially thankful to two anonymous reviewers for insightful and probing comments which pushed us to rethink entire sections of the paper, and for helping us to rewrite and improve many passages throughout it. We are also grateful to audiences at Columbia University, the Pacific APA, the University of California, Berkeley, and the University of Pennsylvania, including Sanford Diehl, Daniela Dover, Catarina Dutilh Novaes, Verena Erlenbusch-Anderson, Axel Honneth, Patricia Kitcher, Colin Koopman, Errol Lord, Jennifer Morton, Jim Porter, and Susanna Siegel, among many others, for their probing questions and comments. S.V.B.’s work on this article was generously supported in part by funding from the Humboldt Foundation and a sabbatical leave from the University of Pennsylvania.

COMPETING INTERESTS

The authors declare that they have no competing interests.