Article body

What Critique?

In recent years, the field of educational sciences has seen a surge in critique of critique. On the grounds of growing skepticism about the performance and functionality of critique in general (cf. Latour, 2004), a discussion of a new approach is underway that acknowledges but still tries to surpass critique as a framework for the educational sciences (cf. Hodgson et al., 2017). The main argument against critical approaches in the educational sciences seems to be the detrimental effect it purportedly has on pedagogy – both as a practice and as the reflection of practice. However, this critique often fails to engage with the fundamental validity of critique itself, instead focusing on its potential excesses and consequences. This paper takes a different approach by examining the mechanisms of critique itself. I propose that critique within the educational sciences has become a “trivial machine” that produces generic outputs and lacks the ability to generate new insights. Importantly, I do not claim to assess all forms of critique in this field; such a broad evaluation would require extensive typologies and historical context that could complicate rather than clarify the discussion.[1]

Instead, this paper aims to reconstruct a critique machine that yields a variety of recognizable critiques prevalent in the educational sciences. This exploration serves two purposes. First, the critique machine can generate critique of pedagogical concepts for educational purposes. Students in higher education can use this tool to identify a critical stance towards language use in pedagogical fields. Second, it helps to discern generic critique from generative, more specifically focused forms of critique. Generative critique – as opposed to generic critique – produces knowledge by confronting ideas with experience in a dynamic, non-trivial, non-mechanistic process. Generic critique, on the contrary, cannot learn from experience, since its internal structure is immutable, strictly determined, and thus linear. This makes it trivial in a technical sense: it cannot adapt based on feedback or context. When the critique machine performs the same task repeatedly, it fails to recognize or respond to its own patterns and their limitations. Finally, a generically and trivially operating machine is removed from experience of the world. It operates (in Kantian terms) analytically by simply unfolding complex concepts, whereas generative and dynamic machines go “beyond the initial description of the object,” since they “appeal to particular experience” (Guyer & Wood, 1998, p. 52).

-> See the list of tables

Generic Critique Reconstructed: The Construction Plan of the Critique Machine

The machine metaphor of critique was introduced by Foucault, who describes his own archaeology as an apparatus and as a “bizarre machinery” (Foucault 1972[1969], p. 132). It has been further popularized by Latour (2004) and is often used to criticize the “repetitive process of critique” (Noys, 2019, p. 31) or its degradation as a ubiquitous routine and genre (cf. Anker and Felski, 2017), as well as the tendency to adopt a “uniform posture of critical dogmatism” (Raffnsøe, 2015, p. 6). If this is true, it must be possible to reconstruct such critique in the form of a trivial machine.

A trivial machine, according to Heinz von Foerster, is entirely predictable, since its strict internal organization produces the same outputs on the basis of the same inputs (cf. von Foerster, 1993, p. 138). As said above, a trivial machine learns neither from the past nor from the material it transforms. But can we have such a machine in our heads? As the Frankfurt School famously argued, it is not only possible but rather probable that the human spirit transforms itself into such a trivial machine in modern society. Instrumental reason, adaptation to societal demands, reification of the self, and bureaucratization of thinking and (inter)acting are critical diagnoses that indicate the tendency towards self-mechanization due to societal pressure. However, two of the most prominent opponents of the Frankfurt School, namely Martin Heidegger and Ernst Cassirer, came to similar conclusions. The Heideggerian “Gestell” as the essence of modern technology is not something external to the human spirit but rather its dominant modality of being-in-the-world (cf. Heidegger, 1956, p. 60). Cassirer’s attempt to describe various symbolic forms (e.g., art, language, religion, science) as productive, generative, originary, and dynamic modalities of the human spirit is implicitly set against the privative or degraded modalities of that same spirit (cf. Cassirer, 1923, pp. 8–9). Hence, the idea that forms of generic critique occupy the reasoning in the critical educational sciences is fully in line with critical thinking and even with philosophies that do not describe themselves as critical. Examples of the output of the critique machine will give an idea of the prevalence of this problem.

We will feed the critique machine only with pedagogical semantics. The idea is that critique in the educational sciences is mostly focused on pedagogy as a profession – in other words:, as the self-description of pedagogical practice. From a systems theoretical perspective, pedagogical practice depends on orientating self-description that consists of affirmative semantics (Krönig, 2019; Luhmann, 1990). These semantics (pedagogical semantics will be put in italics from now on), like participation, inclusion, reflection, student centredness, growth orientation, holistic education, resilience, empathy, strength-based approaches, active learning, self-efficacy, self-determination, and play are prevalent in published pedagogical self-descriptions such as practice-oriented journals, mission statements, educational visions, or philosophies of educational institutions. When these words occur in the educational sciences, they no longer function as affirmative semantics that orient us and tell us what is “good” and what “should be done” or should be focused on. Rather, they become research objects, either in the form of critique or in the form of empirical material. Of course, the educational sciences do not only criticize or provide research into pedagogical semantics, but this is nevertheless one major focus of educational theory and research. Hence, in this analysis, we do not claim to speak about the entire field of educational sciences, only the self-described critical educational sciences.

Illustration 1

The Critique Machine: Overview

The Critique Machine: Overview

-> See the list of figures

The First Process Stage of the Critique Machine: Introduce Contingency Universally

Regardless of the semantics in question, the critique machine operates on the assumption that the semantics have eliminated several modes of contingency that must be unveiled and reintroduced. For our purposes, it is sufficient to conceive of contingency in the logical sense as the negation of necessity and impossibility (cf. Luhmann, 1992, p. 96). We do not need to allude to metaphysical concepts of contingency that refer to the possibility of not-being, or to the effects that contingency might have for the people’s life-worldly experience of insecurity in modern society (cf. Habermas, 1973, p. 176). As the difference between impossibility and possibility plays no role whatsoever in our discussions, we can simply conceive of contingency as “not necessary.” However, this non-necessity does not just refer to being/not-being, but rather to the innumerable possibilities of being-different. What is contingent could very well take various forms simultaneously.

The critique machine introduces contingency in all cases – that is, universally. In doing so, it criticizes that the semantics in question annihilate, invisibilize, or de-thematizate their own contingency. Whatever they describe as necessary, the critique machine unveils, could be different or could not be at all. The machine’s default to generically maximize the contingency level can be interpreted as an epistemological a priori. Whereas in transcendental philosophy truth value was linked to necessity, in the critique machine it is linked to the opposite – that is, to contingency. On the basis of this epistemological default, all pedagogical semantics are suspected of universalization and reification. The machine debunks universalization and reification by introducing contingency to all semantics to the point that there remains no trace of necessity, normalcy, or truth claim.

Universalization eliminates the contingency of a semantic (this applies to all semantics) by the invisibilization of a variety of differences regarding the standpoint of its articulation. This standpoint is historically and culturally determined and shaped by the identities and interests of the speakers and the discourses they reproduce: in other times, other cultural spheres, and from the perspectives of other people with other intersectional identities, and in the framework of other discourses and in other language games, the semantic would have different meanings – that is, it would include and exclude other meanings. All these forms of otherness are possible and their possibility is invisibilized by simply not thematizing, much less discussing them. Thus, the critique machine transforms universality into pseudo-universality by simply pointing out these contingencies, rendering the semantic in question a particular semantic. Its particularity is – in subsequent operations (discussed further below) – rendered hegemonic.

Reification eliminates the contingency of constructions. Consequently, the critique machine deconstructs the essentialist assumption that the semantic in question denotes something “which can be encountered within-the-world” (Heidegger, 1962, p. 93) and which stands in a seemingly neutral and self-sufficient juxtaposition to a disjunct Cartesian subject. The critique machine debunks all pedagogical semantics as labels for pseudo-things. In an everyday mode, people use words as labels or as signposts for objects. This (so-called ontic) behaviour, which allows people to reduce complexity and to function pragmatically, arguably also characterizes the pedagogical – that is, professional language use. Semantics, like participation and inclusion, are, then, treated as quasi-things that can be organized, managed, and measured, as opposed to constructions that merely “exist” in the framework of narrations and discourses. A good example is the way in which Adorno deconstructs the reality of needs (Adorno, 1979) – a semantic that is particularly prone to reification. For him, there are no naturally or biologically necessary needs; instead, all needs are socially mediated and historically prefigured, and therefore contingent. The reason for this leads us to the thematic side of contingency elimination.

The Second Process Stage of the Critique Machine: Introduce Contingency Locally

There are two primary regions of contingency that are addressed specifically by the critique machine. The machine is particularly triggered by the contingency elimination strategies of naturalization and culturalization. Not all pedagogical semantics can be subject to these forms of critique. A limited, albeit still large, number of semantics in the thematic field of bio/life sciences trigger the machine’s naturalization critique program. This thematic limitation is even more relevant when it comes to culturalization, as we will see shortly, after we have dealt with naturalization.

Needs, attachment behaviour, empathy, resilience, talent, ability, competencies, and learning styles are examples of pedagogical semantics that the critique machine automatically detects as naturalizations (see 2.2.1 in the above illustration). The machine is triggered by the ethological (cf. Vicedo, 2013), biological or biomedical (Clarke & Shim, 2011), developmental (cf. Burman, 2008), and anthropological (cf. Fabian, 2014) backgrounds, reference, or at least possible connotations of these semantics. In cases in which this is unclear or ambiguous, for example in the case of competencies or learning styles, the machine operates on the basis of suspicion: competencies and learning styles can be conceived of as innate or congenital. This possibility is sufficient for the critique machine to start operating. The operation is not a transformation but simply a labelling of these semantics as biologizations, psychologizations, neurologizations, and medicalizations. These labels generically discredit the semantics, as it is taken for granted that they unveil an illegitimate reframing. They imply heteronomization; that is, the imposition of external norms – a critique that can be emphasized by the notion of colonization in the Habermasian sense. The invasion of language from the (broadly speaking) bio/life sciences into the pedagogical field is then understood as a hegemonic move, since the language of these sciences is often deemed more paradigmatic (cf. Evans et al., 2016, p. 773; Kuhn, 2012), benefits from a higher aura of scientificity, and is thus more powerful (cf. Rose, 1999, pp. 135–154).

The critique machine detects culturalization whenever semantics display an essentialist and homogeneous understanding of cultures, which the (always affirmative) pedagogical semantics do either with an appreciative and respectful gesture or with an explanatory, sometimes even apologetic, purpose. Typically, some pedagogical semantics such as diversity and inclusion, and particularly cultural enrichment and welcoming, essentialize, reify, and homogenize cultures, by implying that these cultures and the people who bring them to the table have discernable properties and qualities that can explain behaviour and that call for generic respect. Arguably, the famous critique of the semantic of multiculturalism as “the enemy” (Hall, 1991, p. 55) and the work of Edward Said (1994) have programmed the critique machine with the culturalization critique, which is now updated regularly by the developments of postcolonial theories.

The critique machine detects the acknowledgment, affirmation, and appreciation of cultures and labels the targeted semantics as expressions of othering, if not of cultural racism. Likewise, semantics with possible references to biological or developmental psychological concepts are labelled as naturalizations (see 2.2.2 in the above illustration). On these theoretical grounds, naturalization and culturalization are debunked not only as contingency invisibilization strategies but also as power moves (see 2.2.3 in the above illustration). The critique machine suspects that the legitimization of inequalities through culturalization or naturalization reveals the necessity of the invisibilization of the structural, constitutive contradictions of modern societies. Whether these constitutive contradictions are those of capital and labour – that is, in essence, economic (as Marxists claim) or political in a broad sense of power-related (as Foucauldian scholars claim) – is an ongoing discussion of the academic and activist left (Honneth & Fraser, 2003). However, the critique machine operates with a power concept that is generic to such an extent that this important difference is undermined; that is, in effect, irrelevant, as will be discussed below.

The Third Process Stage of the Critique Machine: Infuse Power

Whereas the first stage, with its epistemological default, operates without the notion of power, the second, thematic stage already infuses power regionally. The label of naturalization draws its discrediting effect not only from the epistemological faults of contingency elimination and difference (especially contradiction) concealment, but also from the argument that the invisibilizations of contingency and difference depend on the particular power of pseudo-scientific knowledge. In the case of culturalization, this power is less grounded in the positivist and naturalist knowledge of the bio/life sciences and much more in everyday pseudo-knowledge about cultures, as well as the moral power of the acknowledgment and appreciation of diversity. In its third process stage, the critique machine adds that contingency and difference elimination strategies generally benefit the interests of the powerful and normalize, as well as stabilize, power structures and systems.

As soon as the critique machine injects power globally, all educational operations, including their omission, are deemed to originate from power and to produce power: power is, therefore, both cause and effect. Every pedagogical practice, action or inaction, or communication or lack thereof, as well as every pedagogical self-description – that is, semantic – produces differences in the medium of power, as soon as the critique machine saturates the pedagogical world with power as an all-pervading aether. In this world, infused with aethereal power (cf. Krönig, 2022), all forms of contingency and difference elimination are power moves – for example, in the form of subjugation, discrimination, hegemonic practice, and exploitation that invisibilize, legitimize, and normalize power differentials and inequalities of all kinds. The power that the machine injects cannot be questioned, much less criticized; hence the machine is immune to self-critique. There are several reasons for this (cf. Krönig, 2022), but its circular, self-implicating construction is a sufficient one: When every operation takes place in the medium of power, is enabled by power, and produces power effects, the questioning of power is also a powerful operation that is automatically unveiled as an interested move by powerful actors to invisibilize exactly this. Hence, the critique machine, at this stage, not only debunks power-related pedagogical semantics such as participation, dialogue, eye level, self-determination, and empowerment as invisibilization, normalization, and legitimization attempts of hierarchies, power asymmetries, and institutional as well as societal power structures, but also applies to those semantics that have (for the mind without a critique machine in operation) no obvious relation to power. This might be the case with inclusion, reflection, and play. Here, the critique machine operates ex negativo. The omission of any direct or implicit reference to power is interpreted as a hypothesis of a world in which people can become included, can reflect rationally, and are free to play. This would be a world that is not infused and saturated by power. The critique machine has labelled the de-thematizing of power as affirmative – that is, uncritical – in the previous stage, and unveils it as an individualizing and responsibilizing strategy in the next and final stage.

The Fourth Process Stage of the Critique Machine: Monopolize Causality

Until now, it seems as if the critique machine debunks all forms of contingency and difference elimination. Universalization, reification, naturalization, and culturalization are labels for strategies that powerfully “explain” and thereby normalize inequalities of all kinds, and the pedagogical semantics are criticized for blindly operating on this basis. However, one more “explanatory principle” (Bateson, 1972, p. 38) belongs in this list. The critique machine also criticizes individualization as a means to invisibilize contingency (stages one and two) and, furthermore, as a means to deny the power of power (stage three). In a world that reproduces inequality through ubiquitous power structures, the fact that this inequality is not necessary can also be normalized and legitimized by attributing observable differences to the individual person, thereby invisibilizing the de facto causations on the societal level. For the critique machine, individualization amounts to responsibilization, in all cases (see 2.4.1 in the above illustration). When the de facto cause of inequalities is located on the societal level, individualization not only hides the underlying power dynamics but also places the blame for failure and exclusion of any kind on the powerless individual person.

This brings us to the conclusion that the critique machine debunks all possible types of causation except one. Universalization, reification, naturalization, culturalization, and individualization are powerful strategies used by the powerful to invisibilize their power and to legitimize and normalize societal power structures. Whenever the critique machine detects a pedagogical semantic that implies that a phenomenon is always and everywhere valid or effective – that is, that it is not constructed but “real” – grounded in nature or culture, or a matter of the individual person, it is triggered to perform the operations sketched out above. None of these explanatory principles can withstand the critique machine and are invalid in every case. However, this does not lead to the persistence of contingency, hence the abstention from causal explanations. On the contrary, the critique machine arrives at a conclusion that was set up as a default from the start: exactly one explanation is valid, namely the causation by societal power structures. As a theoretically grounded causal explanation, sociologization is itself a contingency and difference elimination strategy, and would, as such, have to be detected, labelled, and criticized by the critique machine by its own standards. However, the critique machine does not even use the “no alternatives argument” (Dawid et al., 2015), which claims the sociologization hypothesis to be the only one left after having debunked all others (namely naturalization, culturalization, individualization, universalization, and reification). Rather, this sociological explanation – that is, contingency elimination – seems to be part of its pre-established configuration. Due to this programming, the critique machine adheres to the principle of equifinality (see 2.4.2 in the above illustration). The output is not contingent upon the input but fully determined by the internal operations of the machine. The introduction of contingency and power, and the labelling of the semantics as naturalizations, culturalizations, and individualizations, override the specificity and idiosyncrasy of the pedagogical semantics. As a result, the semantics in question are shaped, and hence assimilated, by these procedures rather than by their original character.

In order to understand the critique machine, we have to analyze the structural determination of the machine with regard to the difference: operational versus thematic. The acceptance of sociologization does not seem to follow from its operationality. Rather, as we have seen, sociologization contradicts the processes of contingency injection. This means that there are thematic pre-configurations and operational programs that are independent of each other and that lead to mutually incompatible results.

Conclusion: Assessing the Consistency and Performance of the Critique Machine

In the end, we can conclude that the critique machine is, in fact, composed of two machines. That machines, generally, consist of two more relatively independent parts is, however, the rule rather than the exception. We call attention to this fact merely for the pragmatic reason that this differentiation might help the analysis and assessment of the machine. The first component could be labelled as a deconstruction machine. On the basis of the epistemological axiom that universality claims and reifications are always (that is, generically) wrong, the deconstruction machine reintroduces the contingencies that are eliminated by universalization and reification. The epistemological axiom is, one could say, recursively legitimized by this very operation: when contingency can be introduced into the semantics, it is at once proven that contingency had not been represented sufficiently before. Hence, the deconstruction of universality as pseudo-universality, and objectivity or “thingness” as reification, does not rest on preliminary thematic decisions – that is, machine programming. This makes the deconstruction machine the operationally determined component that, in principle, applies to all semantics in the same way. Thus, the deconstruction machine is both consistent and unbiased. Also, it performs as a reflection mechanism that provides pedagogy with a reliable tool with which to distance itself from its own affirmative semantics and thereby helps pedagogical reasoning to connect to scientific discourses.

However, this is still a machine. This means it operates without any regard for the adequacy of the deconstruction it facilitates, and substitutes generative critical thinking with a trivialized generic program. For example, the deconstruction of the child or of pedagogy itself renders the pedagogical real-world relationship to children and to one’s own practice a pure (fictional) construction. Post-critical pedagogy convincingly argues that pedagogy as an originary modality of being-in-the-world cannot be substantiated and sustained on this basis. Without “maintaining an affirmative attitude towards the world” (Vlieghe & Zamojski, 2019, p. 36), towards the child, towards things, and towards one’s own practice and role as a pedagogue, pedagogy is non-existent – and what supplants it, hell. This attitude is, evidently, not consistent with deconstruction. This post-critical position is not brought into play as an irrefutable argument against deconstruction as an integral part of pedagogical reflection, but rather as a counter-position to my admittedly premature claim above that deconstruction is a rather neutral and mere analytic compound of the critique machine. Maybe it is better to say that there are things that cannot undergo analysis or deconstruction without losing their integrity and that are not resilient, in the sense that they can revert to their pre-analytic or pre-deconstructed version. This might be true of pedagogical semantics that lose their affirmative sound and orientating function for pedagogical practice. This is all said without taking any normative position with respect to the question of whether pedagogical semantics in general or any in particular should lose their affirmative sound and function. However, it seems to be sufficiently plausible that even the deconstruction compound of the machine is transformative.

We could be prone to contrast the analytic operation of the deconstruction compound with a synthetic operation of the subsequent critique compound. However, the generic critique as reconstructed in process stages two, three, and four is completely determined by the machine’s pre-configuration. Hence, the information the machine generates is, strictly speaking, not information about the pedagogical semantics in question and therefore not synthetic. In other words, the critique adds no information about a semantic but rather displays itself by way of the semantic. That a semantic with affiliations to medicine or biology naturalizes and biologizes, and that a semantic with connotations to culture culturalizes, offers no information about the particular semantic. On the contrary, it is information about the category the machine offers. This also goes for the individualization critique. When pedagogical semantics like competence, resilience, autonomy, individuality, subjectivity, or self-determination are detected as instances of individualization – that is, the invisibilization of social power structures and processes – this is, again, a mere categorization of the semantics, without information about the particular semantics in question. The mere appearance of information expansion (and in this sense, synthesis) can be explained by the fact that the critique machine not only consists of operational structures but also provides an inventory of thematic structures as a surrogate of experience. As discussed, the machine’s preoccupation with inequalities, its preconfigured monopolization of sociological explanations and causes on the societal level, and – maybe most importantly – its reliance on the aethereal power concept are all highly contingent and invisibilized programming decisions on the level of the application software. The operating system, however, could very well run innumerable other applications. Whereas the deconstruction compound introduces contingency and labels, thereby destroying contingency elimination strategies, the critique compound reduces contingency by establishing a single contingency elimination strategy; namely, the monopolized explanation due to societal power.

Maybe it is exactly this capacity for self-deception, which derives from the critique machine’s bifurcation into two compounds, that explains much of its success in terms of long runtime and significant proliferation.