1. Marxism and history

At the turn of the twentieth century, few philosophical ideas in Marx’s work received as much attention as his account of history – the thesis that history, at its core, reflects the development of human productive powers, with societies emerging or declining based on how effectively they foster or hinder this progress.Footnote ¹ Supporters of this view regarded the development thesis as central to Marx’s critical framework. Consider, for instance, Marx’s discussion of the Irish famine in the 23rd chapter of Capital.Footnote ² Marx uses statistical data to demonstrate that, although the financial means to import food and prevent the famine were available, the security of capital accumulation took precedence (2020 [1872]: 645–659). Marx does not aim to describe a singular incident but to illustrate a general tendency in (capitalist) history. By claiming that class struggles arise from the contradiction between the development of economic forces and existing social relations, Marx aims not only to provide a descriptive analysis but also to criticise a structural issue that calls for systemic transformation. One of the leading questions at the time was the following: Is it true that there are laws governing our social history, and if so, how can we meaningfully conceptualise them? While orthodox Marxists endeavoured to closely follow the Marxist development thesis, the Austromarxist Max Adler (1873–1937), influenced by neo-Kantianism, took more liberties in interpreting – or, perhaps more accurately, ‘reinventing’ – the law of history in ‘teleological’ terms.Footnote ³

In this article, I seek to reconstruct Adler’s neo-Kantian ‘reinvention’ of Marx’s account of history. According to Adler, the notion of ‘necessity’ that underpins critical judgements is not grounded in a presumed regularity of the productive forces in history but rather in our moral judgements about how social history should develop. Adler relies here on what I will call an ‘experiential’ account of practical judgments. This approach maintains that promoting progress requires more than merely following the constraints imposed by practical reason; to determine practical laws that lead to progress, we ought to investigate inductively the causes of social inequalities that need to be removed. More specifically, Adler claims that learning about events such as the Irish famine means going through a two-step procedure: First, we gain theoretical insights into the empirical circumstances through statistical studies, revealing whether we could have responded differently to the crop failures. If so, then, in a second step, we evaluate how, from the standpoint of moral necessity, we should have responded.

My reconstruction is divided into two parts. First, I will make explicit Adler’s implicit critique of Marx’s thesis that economic forces determine the course of human history. By interpreting human progress as a possibility that presents itself as a necessity from the standpoint of practical rationality, I shall show that, with his neo-Kantian renewal of Marx’s development thesis, Adler laid the basis for a critique that only later gained traction (see Popper Reference Popper2002; J. Cohen Reference Cohen1982; Dickman Reference Dickman1990; Elster Reference Elster1985; Ruben Reference Ruben1980).

Second, while Marxists may fear that Kantian formalism cannot sufficiently address misguided ideological beliefs, I will demonstrate that this worry does not apply to Adler. By showing that Adler’s account is rooted in nineteenth-century neo-Kantianism, a tradition capable of tracking ideology or distorted beliefs among oppressors and the oppressed, I argue that Adler’s notion of ‘progress’ is robustly anti-ideological as it takes the recognition of exploitative structures as an ideology-emancipating transformation.

My reconstructive approach is guided by the goal of accurately portraying Adler’s philosophical concerns, while also bringing his ideas into conversation with more recent debates in analytical Marxism. To achieve this goal, I combine methodological aspects of historical reconstructivism and rational reconstructivism. In my efforts to understand Adler and the philosophical problems he sought to address, I rely on the historical contextualist assumption that ideas can only be fully understood if we understand the author’s intentions within their own context (Skinner Reference Skinner2012). However, in my analysis of Adler’s ideas, I follow the lead of rational reconstructivists who hold that philosophical arguments should be brought into a contemporary context (Rorty Reference Rorty, Rorty, Schneewind and Skinner1984: 247).

The article unfolds as follows. In §2, I introduce Max Adler’s neo-Kantian account of progress, conceptualised in his idea of ‘original communism’. In §3, I introduce Marx’s development thesis through the lens of G. A. Cohen’s reading. In §4, I outline Adler’s neo-Kantian reinvention of history from the standpoint of freedom. In §5, I use Marx’s example of the Irish famine to illustrate the differences between the two thinkers’ approaches. In §6, I reconstruct Adler’s critique of the Marxist development thesis and show that, by seeking to circumvent the problems associated with this account, he worked out a notion of progress that is robustly anti-ideological. Finally, in §7, I briefly summarise the main argument.

2. Experiential Kantianism and the communist state as a ‘hypothesis’

According to Adler, the communist state is not a timeless ideal; understood as a ‘hypothesis’, it reflects one attempt to substantialise the conditions under which the unification of ends at a specific time and place becomes possible. In this section, my aim is to explore the concept of the state in Adler and trace its roots in the tradition of nineteenth-century neo-Kantianism.Footnote ⁴

Prima facie, placing Adler in the neo-Kantian tradition might appear strange. In Causality and Teleology, Adler explicitly distances himself from ‘neo-Kantian socialist accounts’ that seek to understand socialism in purely ‘ethical’ terms: an approach typically associated with Hermann Cohen (1843–1918), the founding father of the Marburg School of neo-Kantianism (Widmer 2024a).Footnote ⁵ However, Adler clarifies that it is not Hermann Cohen or anyone from the Southwestern School he seeks to distance himself from, but rather the neo-Kantian revisionist ‘Back to Kant’ movement, as it was – in Adler’s view – crudely defended outside of academia by Karl Kautsky and Eduard Bernstein (Adler Reference Adler1904b: 92–93).Footnote ⁶ By adhering to the ‘theoretical direction of Kantian epistemology’, Adler clarifies that he ‘entirely agrees’ with Hermann Cohen, who, in Adler’s view, ‘has revealed to us the full depth of the transcendental method’ (p. 92).

Hermann Cohen’s ‘transcendental method’ is best understood as an experiential approach to Kant’s theory of human progress. This means that in order to promote progress – that is, the enhancement of the development toward the ultimate end of history (the perfect proportion between virtue and happiness) – it is not only crucial to act in accordance with the form of practical laws; we must also gain insight into the empirical social circumstances hindering this development (Widmer 2024b). The removal of laws causing inequality leads to actual progress. For this to be successful, we must rely on inductive judgments about the phenomenal world. Though Hermann Cohen builds here on Kant, he argues that Kant’s idea of progress must be liberated from the providential line of argumentation we find in the Second Critique. Kant argues that God – the divine creator – is a necessary presupposition for conceptualising the causality of progress.Footnote ⁷ The notion of ‘necessity’ arises from the epistemic limits we face: Though we can strive for the realisation of the highest (political) good, we can never know whether our actions actually lead to progress. As a ‘practically necessary condition,’ however, God and the immortality of the soul are ‘the conditions under which alone we can, given the constitution of our (human) reason, conceive of the possibility of that effect of the lawful use of our freedom’ (CPrJ, AA 132). Whereas for Kant, progress remains from a theoretical perspective a mere possibility or ‘a mere matter of faith of pure reason’ (CPJ 5:470), Hermann Cohen contends that theoretical reason plays a central role in so far as it allows us to empirically study causal relations between legal and social phenomena (Cohen, 2001 [Reference Cohen1877]: B455). To conceptualise progress, we do not need to view history as if it was led by a divine being; instead, we can formulate concrete hypotheses about what leads to progress based on empirical studies of the causes of inequality (KBE, A313 B352, see Widmer Reference Widmer2023).

Adler agrees with Hermann Cohen’s experiential approach. He argues that only through the empirical study of our ‘social interrelations’ (soziale Zusammenhänge) do we ‘establish the theoretical grounds through which we become practically conscious’ (Adler 1974 [Reference Adler1904b]: 182). By rendering empirical knowledge about social interrelations from the standpoint of practical reason, ‘ethics becomes the vehicle for progress’ (p. 182). ^{Footnote 8,Footnote 9} Adler argues that Kant was correct in claiming that the ‘highest end’ of our cognitive faculties is guided by the regulative ‘idea of unity’ (1904b: 206). However, given that progress is advanced by gaining an understanding of the empirical obstacles hindering it, we need to replace Kant’s notion of causality with one that highlights our epistemic agency in its promotion. In this vein, Adler claims that the concept of ‘God’ must be replaced by the concept of the ‘state’ (Adler Reference Adler1904b: 206–207).Footnote ¹⁰

This correction rests on the neo-Kantian conception of rational ideas as linguistic expressions that relate to and are conceptualised within a society that precedes us – a ‘social apriority,’ so to speak. Because our substantive rational concepts are historically conditioned due to their linguistic representation, we can only study their ‘functional relationships’ (Adler Reference Adler1904b: 208). Once again, this traces back to an idea developed by Hermann Cohen. Cohen argues that while the final end of humanity is a rational idea, it is only mediately accessible to us through the products of our cognition, which have the ‘function’ of promoting more equal relations (1981 [1904]: 35–36). While the idea of God once served as a ‘hypothesis’, expressing – as Kant called it – ‘the assumption of a merely possible explanatory ground’ for the realisation of the highest good (CPJ 5:463), it has been replaced in modern times by the idea of the ‘state’ (Cohen Reference Cohen, Wiederbach and Holzhey2009 [1908]: 178). By replacing ‘God’ with the concept of the ‘state’ as a ‘hypothesis’, Adler signals that it is our responsibility to investigate inductively the causes of unequal relations, which must be countered at an institutional level (Adler Reference Adler1904b: 37).

The state, understood as the relevant institution for advancing the unity of ends, prescribes the inclusion of both the study of norms as historical-causal phenomena and their moral justification. These standpoints place us in a ‘different relation’ to the ‘whole’ (das Ganze) (1904b: 69). While theoretical reasoning enquires about causal explanations of disproportionate social interrelations, moral reasoning examines the moral justification of those relations (Adler Reference Adler, Sandkühler and de la Vega1974 [1904a]: 183). Sociology, which deals with ‘some sort of naturalism’, is not merely meant to study the appearances of normative facts; since normative facts are advanced by human beings, they must also be subjected to moral scrutiny (Adler Reference Adler2019 [1922]: 18). The first step of the sociological study of social interrelations incorporates in their historical emergence is the study of them as ‘isolated insights’ (Einzelerkenntnisse) (1904b: 19). In a second step, these facts need to be justified from a moral point of view, which gives them their ‘general form’ (Allgemeinerkenntnis) (1904b: 19). For instance, understanding a religious form as an ‘isolated insight’ means examining how a religious system emerged and gained power. However, because religious laws are human-made, they also require scrutiny from the viewpoint of moral necessity to determine whether the existing principles genuinely fulfil their intended function of creating a systematic unity of ends (p. 223). Regardless of how the ideal as a ‘hypothesis’ expressing the systematicity of ends is substantially conceptualised – be it as ‘God’ or, in modern times, the ‘state’ – its role of bringing forward coherent ends remains consistent over time.Footnote ¹¹

To spell out the ‘ideal of the state,’ i.e., the conditions under which the approximation of the highest political good becomes possible (Adler Reference Adler2019 [1922]: 34–35), Adler begins with the formal requirement of the state. Inspired by Kant, he argues that laws must have a general form that reflects the ‘united will of the people’ (1974 [1904a]: 190). Here, Adler draws on Kant’s view that a state is legitimate if and only if its laws have an ‘general‘ form (TP, 8:290ff; MM, 6:263). The general or omnilateral form requires that laws be ‘independent’, meaning they are established not based on personal preferences but on principles to which everyone could rationally consent (TP, 8:297; MM, 6:263). Unlike civil self-sufficiency, which is an empirical feature of citizens actively participating in lawgiving practices, the ‘independence’ of the omnilateral will refers to an a priori form of lawgiving judgments.Footnote ¹² This requires that we establish laws so that our wills are ‘united not contingently but a priori and therefore necessarily’ (MM, 6:263, emphasis added).Footnote ¹³ Only if lawgiving judgments take such an omnilateral form can we ensure that a citizen’s innate right – freedom and equality – is externally secured (TP, 8:290; MM, 6:237).Footnote ¹⁴

While there is nothing inherently wrong with the rationality of omnilateral lawgiving, the way Kant applies the idea of the state under certain empirical assumptions can be seen as problematically insensitive to questions of material inequality. Although some interpretative approaches find conceptual space to read his political philosophy in more progressive and left-leaning terms (e.g., Williams Reference Williams1983; Ypi Reference Ypi2014; Love Reference Love2017; Hasan Reference Hasan2018; Holtman Reference Holtman2018; Vrousalis Reference Vrousalis2022), a literal reading of his texts reveals limited concern for economic causes for social inequality. We see this, for instance, in his discussion of poverty: while Kant deems it important to worry about the level of taxation for poverty relief to ensure it ‘does not become an unjust burdening of the people by the government’ (MM, 6:325–327), he regards inheritance laws as posing no threat to justice, even if they ‘bring about a considerable inequality in wealth among the members of a commonwealth’ – a position that is indirectly supported by the private right section of the Doctrine of Right (TP, 8:291; MM, 6:243–309). By refraining from theoretically investigating the empirical causes of social inequality, Kant’s political philosophy exhibits several ideological biases. These biases manifest not only in his views on people living in poverty but also in his stance toward women and wage laborers, whom Kant deems legitimately excluded from lawgiving procedures.

While Adler adopts the rational structure of the omnilateral will, his neo-Kantian experiential account comes with two crucial amendments aimed at avoiding such biases. First, Adler argues that the general will can only generate principles with an omnilateral form if those subjected to these laws actively participate in lawgiving practices. If certain groups are unreasonably denied participation, Adler argues that they are wronged in their ‘human willing and striving, purposeful and ethical judgments’ (2019 [1922]: 8).Footnote ¹⁵

Second, unlike Kant, who states that ‘material inequality’ is separated from the legal notion of equality, Adler believes that the state only upholds the general form of lawgiving when a citizen’s ‘economic self-sufficiency’ is secured (1974 [1904a]: 190). According to Adler, it is a deep economic fact that the private ownership of the most relevant means of production is the cause for material inequalities. Thus, a state is justified only under the ‘social ownership’ of the most relevant means of production (p. 190). This is why the substantialised ideal state is, on Adler’s account, a ‘communist’ state:

Because we always deal with individuals in representative roles who claim to ‘act in the name of the whole but, in reality, have personal interests in mind in their exercise of authority’, Adler acknowledges that a state governed by the omnilateral will can never be fully achieved (p. 34). By relying on inductive judgments about the causes of social inequality, Adler seeks to avoid the generalisation of ideological content that problematically legitimizes unequal social relations. With a greater focus on the conditions under which our normative ideas are discursively shaped, he aims to specify the circumstances in which freedom and equality can be realized.

Following Hermann Cohen’s experiential account, Adler argues that progress, from a theoretical viewpoint, is neither fully determined by economic conditions, as Marxists claim, nor a matter of faith, as Kant argues. By empirically examining the social relations that constitute disproportionate material relationships, we gain the ‘critical consciousness for the [required] change and transformation of the existing order’ (Adler Reference Adler, Sandkühler and de la Vega1974 [1904a]: 182). In this vein, Adler redefines ‘Marxist sociology’ as follows:

Equipped with Hermann Cohen’s experiential neo-Kantianism, Adler redefines Marx’s notion of history in terms of a moral transformation that becomes real when we seek to overcome moral contradictions in our society. For Adler, it is not the disposition of productive forces to develop in a specific manner that provides the framework for critique; rather, it is human agency and the obligations we uphold that shape the course of human history.

3. Marx’s development thesis: a contemporary defence

To fully grasp Adler’s alternative suggestion, we must first understand the position from which he seeks to distinguish himself: the Marxist development thesis, which asserts that human history follows theoretical laws determined by productive powers (Adler Reference Adler1904b: 32). The focus of this section will be on this thesis.Footnote ¹⁶ A natural approach to introducing this position would be through the defence of the development thesis proffered by Georgi Plekhanov (Reference Plekhanov2009) – a contemporary of Adler who, at the time, provided its most influential defence. Plekhanov was an engaged publicist whose works were well known to socialists. However, his radical rejection of idealism – ‘the ideal is nothing other than the material […] transformed in the human mind’ (2009 [1896]: 10) – and his support of the ‘iron regularity of history’ made his interpretation of Marx appear ‘shallow’ and ‘overly simplified’ (Kołakowski Reference Kołakowski1978: 341). While learning about Adler’s reluctance towards the theoretical regularity of history through Plekhanov’s Marxism is historically sensible, I choose instead to focus on a philosophically more robust and currently more prominent defence of this thesis, as articulated in G. A. Cohen’s Karl Marx’s Theory of History (1978).

In G. A. Cohen’s endeavour to renew ‘old-fashioned historical materialism’, he introduces the development thesis as the idea that ‘history is, fundamentally, the growth of human productive power, and forms of society rise and fall according as they enable or impede that growth’ (Cohen 2000 [1978]: x).Footnote ¹⁷ G. A. Cohen compares the emergence of the productive powers to a child who has an autonomous disposition to grow up and yet is incapable of doing so without assistance (1988: 90; see also Vrousalis Reference Vrousalis2025: 23). In capitalism, the productive forces consist of, on the one hand, the means of production – instruments of production, raw materials, premises, and spaces – and, on the other hand, labour power – physical capabilities, technical skills, and know-how of the workforce (Cohen Reference Cohen2000 [1978]: 55). The effective control of the means of production and labour power constitute, in sum, the economic structure, as well as the (legal, political, and religious) superstructure. Depending on who has effective control of the means of production, we can predict the growth of a society, i.e., the tendencies that history takes on. Under conditions of private ownership over the means of production, the productive forces prompt the increase of capital growth. This goal is determines culture to a large part.

G. A. Cohen introduces this thought in the form of the ‘primacy thesis’, which states ‘that the nature of a set of productive relations is explained by the level of development of the productive forces embraced by it (to a far greater extent than vice versa)’ (2000 [1978]: 134). Although various forces influence the superstructure, it is to a much greater extent the productive forces within current social relations that determine the course of history. Consider following example: ‘Protestantism gained strength in early modern Europe because it promoted the development of capitalism’ (Cohen Reference Cohen2000 [1978]: 249). Let us assume that the thesis (originally going back to Weber) is correct: Protestantism, a religious institution of the superstructure, was successful because it taught morals that increased the efficiency of labour power. Although the emergence of Protestantism is a time-sensitive historical event, the explanation of its occurrence is non-arbitrary as it refers to a law insensitive to time.

To get a better grasp of the time-insensitive or non-arbitrary aspect of G. A. Cohen’s functional explanations, we shall follow his own lead as he draws examples from social biology to make his theory comprehensible. Let us ask: Why did the dodo go extinct? One way to answer this is to provide a set of individual causal explanations: ‘The dodo was exposed to new predators’, or ‘The dodo lost its natural habitat due to human deforestation’. Though such explanations are propositionally true, they do not tell the whole story. For an explanation to be comprehensive, we must identify the correct properties that allow for a law-based explanation of this particular event (Vredenburg Reference Vredenburg2023). In our case, this means understanding the relevant properties to recognise the event as a type-event corresponding to the law of natural selection, which says that those organisms with optimal traits to survive in an environment will survive. While the specific individual causes that led to the extinction are arbitrary and time-sensitive (e.g., exposure to a new predator), the evolutionary type-law abstracts from these arbitrary aspects and explains the event by its time-insensitive features that hold true irrespective of how exactly the law manifests.

A helpful way to determine whether we are dealing with time-sensitive or time-insensitive explanations is to test whether the explanation remains robust under minor ‘alterations of the event in question’ (Hitchcock Reference Hitchcock2012).Footnote ¹⁸ Imagine a world in which a small number of dodos survived in an area left untouched by humans and safe from the dodo’s new predators. Although we can easily imagine a slightly altered world in which the dodo survived, the survival of the dodo would still count as an event corresponding to the law of natural selection. In our imagined world, the dodo has the traits to survive in its environment. Though the token event differs, it still gains its explanatory force from the time-insensitive law of evolution. Only in a world that is substantially altered – a world without the law of evolution – would the mode of necessity for such explanations change. This shows that explanations grounded in a law insensitive to time are ‘modally robust’: the explanatory law is necessarily true for corresponding event-types irrespective of how event-tokens manifest under arbitrary conditions in history. These explanations are characterised by a mode of necessity that remains unchanged under slight alterations because, regardless of how the token event plays out, they still correspond to a time-insensitive type-law. I shall call them henceforth ‘general causal explanations’.Footnote ¹⁹

G. A. Cohen’s functionalist explanations are general causal explanations that gain their non-arbitrariness from the idea that history has a regularity to it. Regardless of how history factually manifests, these facts are grounded in a type-law: the specific form that the ‘development’ thesis assumes under a particular mode of production. However, there seems to be a crucial difference between evolutionary explanations of organisms and economic explanations of the social structure. Whereas explanations of evolutionary biology refer to a law that is applicable to all organisms, Cohen’s explanations of social forms apply only to some. For instance, whereas the emergence of Protestantism is explained by its function to increase the growth of productive powers in capitalism, Catholicism – although also a prevalent religious life form in early modern Europe – is not. This does not apply to events corresponding to the law of evolution because every organism is necessarily subject to this law. If that were not the case, the theory would fail.

To understand G. A. Cohen’s intended functionalism, it is important to consider how he distinguishes between the ‘matter’ and the ‘form’ of social phenomena (2000 [1978]: 89). The ‘matter’ designates variable values (Protestantism, neoliberal morals, Catholicism, etc.), while the ‘form’ corresponds to a constant type-law. However, not every variable value qualifies as a type-event corresponding to the type-law (e.g., Catholicism persists in capitalism despite neither enabling nor hindering capital growth). This is possible, according to Cohen, because the emergence of social forms is explained by ‘consequence laws’ that are ‘universal conditional statements whose antecedent is a hypothetical causal statement’ (2000 [1978]: 259, emphasis added). This means that, unlike in evolutionary biology, G. A. Cohen acknowledges that other forces can explain the emergence of social forms. For instance, religion could be explained by its function to provide hope. However, according to Cohen’s primacy thesis, history shows that social forms that enhance economic performance are significantly more likely to thrive compared to those that do not. Though this theory cannot account for the explanation of all social forms, it is meant to illuminate clear tendencies in history.

4. Max Adler’s teleological alternative

While G. A. Cohen’s theoretical type-law explains the developments in history as they occur under a specific economic structure, Adler’s view on history differs. According to him, the non-arbitrariness of critical judgements corresponds to a practical type-law.Footnote ²⁰ He agrees with Marxists that we must focus on explaining how social forms emerge. However, following Cohen’s neo-Kantianism, he deems these investigations as relevant insofar as we seek to evaluate how historical events should evolve from the standpoint of moral necessity (the omnilateral will). As mentioned previously, Adler suggests here a two-step procedure: first, we must enquire into the empirical conditions that constitute our social reality. Second, we evaluate whether social phenomena align with the omnilateral will. Highlighting moral contradictions provides motivating reasons to change the given conditions, which – once pursued – lead to empirically measurable progress.

Recall that, in section 2, I have demonstrated that, for Adler, the omnilateral will is the formal principle prescribing to render facts from the standpoint of practical necessity. We have seen that Adler distinguishes here between ‘isolated insights’ (Einzelerkenntnisse) and normative ‘facts’ that express the ‘forms of cognition’ (1904b: 210). ‘Isolated insights’ refer to statistical data, historical explanations, etc.: all kinds of inductive judgments that refer to the explanation of time-sensitive, isolated incidents. On this account, the statement, ‘Protestantism gained strength in early modern Europe because it promoted the development of capitalism’ (if true) is considered an event that necessarily happened as it did because all empirical factors led to this state. However, given that these factors are made by humans, we can imagine a world under altered conditions in which Protestantism could have gained the same strength but did not. Under slight alterations, the mode of explanation changes. Explanations of such kind are ‘modally fragile’ (Gallow Reference Gallow, Zalta and Nodelman2022). I shall call them henceforth ‘explanations of unique causal events’.Footnote ²¹

Adler’s notion of ‘isolated insights’ (Einzelerkenntnisse) covers instances of unique causal events. Rather than seeing them as grounded in a type-law corresponding to a natural account of teleology (as Marxists do), Adler understands historical states and events as a series of isolated facts we gather to understand how a state or event came into being. Though this theoretical endeavour is value-free, it can become a matter of practical interest. By theoretically assessing an event, we seek to determine whether the ‘ought implies can’ principle applies: ‘Any “ought” only holds meaning in relation to a “can-be-otherwise”’ (Adler Reference Adler1904b: 163). If a historical event turns out to be caused by human action, then the ‘ought implies can’ principle applies. In this case, the event qualifies for a moral assessment from an omnilateral viewpoint.

According to Adler, human history is not understood as developing from an inherent disposition of our human productive nature. His proposed ‘teleological perspective’ does not refer to a natural telos we inevitably approximate, but a moral end, which we achieve if we follow the principles we define from the standpoint of practical necessity. Whereas in G. A. Cohen’s account, economic conditions provide the type form of historical explanation, it is in Adler’s case a normative judgment that serves as the type form for justifying principles promoting progress. Grounded in the form of the state – i.e., the omnilateral will – this normative perspective prescribes that laws be given an ‘independent’ shape, meaning that every citizen’s interests must be equally reflected in the laws posited.

5. The Irish famine

To illustrate the differences between G. A. Cohen’s theoretical law-based explanations of historical events and Adler’s teleological justifications of the motivational reasons constituting or hindering historical progress, let us return to the example of the Irish famine. In Capital, Marx argues that capitalism, rather than the crop failures, was the cause of the Irish famine (2020 [1872]: 645–659).Footnote ²² Roughly put, the argument runs as follows:

If we approach this case through G. A. Cohen’s theoretical explanations, which derive their necessary form from the thesis that history follows law-based regularities, the upshot is that, under private control of the means of production, capital growth is – and will always be – prioritised, at least to a much greater extent than any other aspect. The capitalist system does not yet make use of the productive powers in the most optimal way because it comes with a great loss of labour power due to starvation. Societies in which the means of production are privately controlled are less efficient. This is a ‘fettering’ system, that is, a system in which ‘the amount of productive power it harnesses at given future times is less than what some alternative feasible system would harness’ (G. A. Cohen Reference Cohen1988: 117).Footnote ²³ Under publicly controlled means of production, we would see a more efficient system in which we are not forced to endure such a great loss of productive labour power.

By contrast, if we deploy Adler’s teleological justifications, this argument takes on a different shape. In the first step, we use statistics to gain insights about ‘isolated social insights (sozialwissenschaftliche Einzelerkenntnisse), which takes events in their ‘subjective and contingent’ shape (Adler Reference Adler1904b: 169). (2) shows that there was the option to deduct from the surplus value to redistribute wealth, which means that the ‘ought implies can’ principle applies. Once we know that, from an empirical standpoint, the natural catastrophe ‘could’ have been handled otherwise, we enter the second step, in which we evaluate whether it ‘should’ have been handled otherwise. As shown earlier, this ‘should’ traces back to the omnilateral will, which provides the form of the state: a principle that, if correctly applied, promotes progress. Private property laws that legitimize the ownership of the most relevant means of production stand in conflict with the omnilateral will. Since they legitimize conditions in which one wealthy group is significantly better off than others, we come to realize that these legal conditions must be abolished in order to achieve moral progress.

One might object that the contrast between G. A. Cohen’s explanations and Adler’s justification is not relevant, as they approach the same issue from different standpoints. Unlike G. A. Cohen’s explanations, which correspond to a notion of history governed by theoretical laws and a sense of ‘necessity,’ the form of ‘necessity’ in Adler’s evaluative judgment is grounded in practical rationality. Could we not simply say that Cohen operates from a theoretical standpoint, while Adler is operating from a practical one? Although I see the appeal of reading them as complementary, there is an insurmountable difference in their ways of conceptualising historical progress. While G. A. Cohen’s interpretation of Marx’s development thesis treats progress as a value-neutral concept driven by material productive forces, analysable independently of human agency, Adler’s view of progress depends on agents acting for (im)moral reasons. This nuance cannot be easily reconciled with the two-standpoint theory. As we will see in the following section, this distinction significantly impacts our understanding of critical judgments.

6. Experiential Kantianism: an anti-ideological formalism

One reason social theorists might be more inclined to return to a Hegelian interpretation of Marx, rather than Kant, is that the dialectical approach – which seeks to highlight contradictions within a society – seems, prima facie, better equipped to account for a critique of ideology. A potential worry Marxists have is that Kantian formalism cannot sufficiently track injustices arising from false beliefs. If the formal principle in question is incapable of identifying empirically incorrect beliefs – or even legitimizes forms of exploitation by leaving ideologically entrenched concepts that support structural power relations unquestioned – Kantian methodology risks failing to tackle the issue at its roots, or so the concern goes.Footnote ²⁴ Yet, in this section, I argue that Adler’s experiential approach is robustly anti-ideological. I seek to show that his account is not only capable of addressing the Marxist concern; it is also capable of dealing with a problem historical materialists face, namely, that their ideology critique is undertaken from a ‘totalizing’ viewpoint of history that insufficiently reflects upon the standpoint from which their critique is conducted.

Jürgen Habermas famously raised the concern of an unjustified totalizing perspective against Horkheimer and Adorno’s ideology critique, arguing that their dialectical approach was caught in a ‘performative contradiction’ (Habermas Reference Habermas and Frederick1987: 119). While they claim that everything is ideological, they contradict this very claim by engaging in a critique of ideology (Freyenhagen Reference Freyenhagen2023). In a similar vein, this problem already appears in Marx’s conception of history. While all our beliefs are understood to be shaped by economic conditions, the position from which we conduct the critique of social relations lies outside this realm. Critique becomes possible only due to a history-transcending perspective, which allows us to perceive the logical regularity initiated by the productive forces. In what follows, I shall show that, whereas this problem has also found its way into G. A. Cohen’s rehabilitation of Marx’s development thesis, Adler’s neo-Kantian critique of ideology circumvents this issue.

Despite G. A. Cohen’s otherwise impressive revival of Marx’s development thesis, its problematic anchoring of critique in an absolute standpoint of history makes it one of the ‘most contested ideas’ in Marxism (see Vrousalis Reference Vrousalis2025: 29). The problem is that the assumption of a theoretical regularity of history ignores the complexity of the subject matter. Consider G. A. Cohen’s example: ‘George was sleepless because he drank four cups of coffee’, even though ‘not everyone who drinks four cups of coffee is sleepless afterwards’ (G. A. Cohen Reference Cohen2000 [1978]: 262, 259). According to G. A. Cohen, it is sufficient to know the generalised law – ‘In a large number of people, coffee causes sleeplessness’ – to recognise the individual case as a type-event corresponding to the type-law. Intuitively, it seems that we need to know more beyond plausible speculation. Does George usually drink four cups of coffee, and does he typically suffer from insomnia afterwards? How much time has passed since his last cup of coffee? Are there any other factors that may contribute to George’s insomnia? It seems that, at best, we can only conclude that George’s coffee consumption is very likely causing his sleeplessness.

G. A. Cohen acknowledges that the truth about what makes precedence statements explanatory is complex (p. 259). However, he maintains that we can nonetheless disregard this complexity captured by probabilistic statements and subsume individual cases under a generalised law of history because it holds true regardless of its applicability to singular cases. The tendency to seek simplicity over complexity is typical of scientific models. Climate scientists create simplified models of our reality to predict consequences under specific conditions; economists develop simplified psychological models to predict consumer behaviour; etc. However, when used to understand the nature of ‘human history’ by basing it on the assumption that we are beings who primarily seek to enhance our material conditions, the appropriateness of such simplification becomes questionable. While it is common in the social sciences to base a model on an assumption of human nature, the difference is that they seek to predict particular developments that can be verified or falsified: e.g. ‘saving money is a central interest of a customer, thus lowering the price of a product increases its number of sales’, or ‘expensive day-care costs motivate parents to carry out the work themselves, thus free day-care centres will increase the number of parents who place their child in day-care’.

On a deeper philosophical level, we see that generalisations with regard to ‘history’ are possible only if we assume a history-transcendent viewpoint and the possibility of the highest political good. Though the regularity of history is initially hidden from us, we can gain insights into historical developments once we understand the interplay between economic forces and the social relations. In contrast to the simplified scientific models of climate scientists and economists, G. A. Cohen deals with statements about what ‘human history’ essentially is: the social forms generated by the growth of productive forces under specific social relations. Even if particular events do not conform to the concept of historical development, the stated primacy of its influence only works if we assume that there are certain human dispositions, which on a large scale, will materialise in a regularity showing that history working towards a final end.

Karl Popper was amongst the first to critically point out that historical materialism is based on the misleading assumption that we are capable of understanding ‘the totality of all the properties or aspects of a thing, and especially of all the relations holding between its constituent parts’, while, in fact, we are ‘always bound to a perspective, and thus can study a thing only with regard to one aspect’ (Popper Reference Popper2002 [1957]: 70). Studying human history means necessarily to ‘select certain aspects’ – an aspect he sees missing in historical materialism (p. 71).

In a similar vein, Joshua Cohen points out that G. A. Cohen’s development thesis disregards contradicting facts. Referring to a rich body of literature in Marxism, Joshua Cohen shows that the picture drawn by G. A. Cohen struggles with the fact that periods of regression remain ignored and unexplained on his (or any historical materialist) model. For instance, from 1500 to 1800, Europe experienced a notable economic regression, indicating that lords prioritised control over the growth of the productive powers in the agricultural sector (J. Cohen Reference Cohen1982: 267–268). Regardless of whether the image of history is the unfolding of the world spirit, the struggle for power, or the image of a story of decline: The subject matter of ‘social history’ is so complex that it necessarily requires a selection of data, and by simplifying history in this manner, we ignore facts that contradict the presupposed image:

Because progress of human history cannot be measured as such but only in relation to an empirically measurable object, we end up with a picture of human history that claims to be holistic while, in reality, we simply select those events that underpin our presumed idea of it, which provides the foundation for social critique.

G. A. Cohen is aware of this problem. He acknowledges that the formulation of historical laws of human nature is not yet as advanced as in social biology (p. xxiv), believing that we are in a state of history where we know about the existence of historical laws, even if we currently lack the exact scientific tools needed to describe them accurately. Cohen compares the stage of historical materialism with the Lamarckian stage of Darwinism. Lamarck lived in times when ‘men […] had no idea how to elaborate’ the ‘belief that species had the useful characters they did because they were useful’ (Cohen Reference Cohen1980: 133–134). Even though Lamarck defended ‘an unworkable idea of how to elaborate it’, Cohen contends that Lamarck ‘knew that the utility of features explained their presence. It is […] not implausible to suggest that historical materialism may be in its Lamarckian stage’ (p. 133–134).Footnote ²⁵ It is for this reason that Cohen calls historical laws ‘law sketches’, that is, ‘law-like generalisation[s] in which some (but not all) of the antecedent properties are specified only by reference to an object which has them’ (2000 [1978]: 262).

While it speaks for G. A. Cohen’s integrity that he admits a level of vagueness, it does not help in assuaging the worry that we end up with an image of history that, rather than providing a rational framework to interpret historical facts, functions more like a stencil placed over an image, covering large parts to make other parts more visibly stand out. This is also why commentators have raised the concern that, without a transcendent perspective and certain assumptions about human behaviour, G. A. Cohen’s primacy thesis becomes ‘viciously circular,’ as no additional evidence is provided to suggest that history is actually developing as proposed (J. Cohen Reference Cohen1982: 264–265; see also Vrousalis Reference Vrousalis2025: 29).

Similar to Habermas, who saw in Kant a safeguard against this problem, Adler – while closely observing a period that lacked the predicted revolution and monitoring the emerging Bolshevik communist regime – recognises in Hermann Cohen’s Kant a solution to the problem that ideology critique must be undertaken from a specific standpoint.Footnote ²⁶ Adler criticises ‘orthodox Marxists’ who, despite being confronted with a more sophisticated account of transcendental philosophy, fall back into a position of a ‘mechanistic conception of the world, in which all functions dependent on thought specific to humans are excluded’ (1904b: 32). His distinction between isolated insights and generalised facts builds on a notion of necessity of practical judgments that does not trace back to a notion of natural teleology, as this would require a thesis about the development of history; rather, it is replaced by a practical teleology – a critical judgment of what needs to be changed to promote progress. Adler’s approach emphasises that all empirical statements, including statements about history following a regularity, must be empirically measurable. However, because statements that are meant to vindicate a regularity of history are too complex to be measurable, they are impossible. By conceptualising historical necessity in experiential-practical terms – as an idea that only becomes real when we are motivated by reasons to overcome revealed contradictions – Adler seeks a notion of social critique that avoids the issues faced by Marxists defending a theory of history.

Yet, although Adler anticipated a problem associated with the development thesis, this does not necessarily mean that his Kantian counterproposal is convincing. From a Marxist standpoint, there are good reasons not to follow a moral understanding of progress, given that normative claims may be entrenched in ideology, that is, incorrect beliefs held by members of a particular historical era, geographical region, society, social strata, or social group, which serve to maintain power relations.Footnote ²⁷ It is a notable strength of Marx’s approach that he counteracts incorrect beliefs rather than providing moral reasons to change the course of history. To come back to the Irish famine, Marx seeks to refute the widely accepted but incorrect belief at the time that famines were inevitable. This belief was reinforced through Thomas Malthus’s ‘law of population’ (Reference Malthus1998), which states that while food production increases arithmetically, human population grows exponentially, inevitably leading to periods of hunger and poverty. Even if it is true that Marx went too far in claiming that all social relations can be explained by the economic conditions, his objection is so compelling because he argues on a factual basis that Malthus is wrong, asserting that capitalism is the true cause of poverty (Marx Reference Marx2020 [1872]: 489–490).

Though Adler does not explicitly discuss ideology, I believe he operates with an implicit notion of it when describing the enlightening moment of gaining insight into our structural positioning in society. As I understand him, Adler does not view normative thinking as genuinely ideological. He aligns more with contemporary theorists, who emphasize that ‘ideology is not, for the most part, thoroughly hegemonic; it does not exhaustively manage all thought and action’ (Haslanger Reference Haslanger2017: 11). According to Adler, economic forces do not determine normative beliefs all the way down; rather, we can overcome ideological beliefs by acquiring empirical knowledge about the given factual conditions. If we render these facts from the standpoint of necessity and follow the practical principles that follow from this rational assessment, we gradually promote progress, i.e., a more equitable society.

Moreover, Adler’s experiential neo-Kantianism is capable of addressing the susceptibility of normative thinking to ideological biases. Following Hermann Cohen’s non-providential interpretation of the approximation of the highest good, which replaces the idea of God with the idea of the state, Adler is not simply a blind follower of Kant. By acknowledging that even rational concepts are embedded within a specific time and place, he argues that an ideal can function as an expression of practical reason at one point in time, while at another, it may be insufficient to grasp other forms of oppression that have since been recognised (Adler Reference Adler1904b: 37). This became evident in his critique of Kant, who, by making progress a matter of faith, undermines the role of science in the promotion of progress. Adler’s Kantian-Cohenian understanding of the state as a changing ‘hypothesis’ is intended to denote the conditions for an equitable society an open-ended project that remains subject to revision. Through better understanding the causes of social phenomena, we gradually learn what the substantive state ideally is by uncovering the moral contradictions that emerge within a society.Footnote ²⁸

In this vein, Adler’s experiential Kantianism involves an ideology-emancipatory moment that ‘purifies’ our outlook on the social realm. Adler claims that individuals not yet enlightened through sociological studies are blinded by ideologies that ‘isolate’ them from others (Adler Reference Adler, Sandkühler and de la Vega1974 [1904a]: 164). Individualism is an unreflective starting point that needs to be overcome. The Irish labourer who blames the natural catastrophe for her hunger, rather than the economic conditions, or the glass collector who attributes his struggles to a lack of talent, are examples of individuals who have not yet gained insight into the structural conditions designed to support the powerful. Understanding ourselves as individuals within a society through sociological studies sets in motion a process of rational self-realisation through which we become aware of our exploitative entanglements. The enlightened person becomes a political agent or, in Adler’s terms, a ‘sociated person’ (vergesellschaftete Person): a person who, once motivated by rationality, ‘thereby becomes an acting, creating subject, a transformative causal factor of the materialist historical conception’ (Adler Reference Adler2019 [1922]: 15). By acting upon the motivating reasons to change the structures of power into which we have gained insight, we break free from ideological beliefs that previously veiled structural injustices, thereby becoming an active force in the promotion of historical progress.

7. Summary

In this article, I have defended two claims. First, while Marxists often rely on a theoretical account of historical development, I have demonstrated that Adler, with his practical account of historical progress, sought to circumvent the problems associated with this view. For Adler, factual events and states do not gain a general form because of an assumed regularity in history, but because we believe that history must develop in a specific manner judged from the standpoint of moral necessity. Second, I have argued that Adler’s notion of critique addresses the problem of ideology, which is a central reason why Marxists may be sceptical of Kantian formalism.