Hermann Cohen’s Neo-Kantian Ethical Socialism

Abstract

Hermann Cohen, the founding father of Marburg neo-Kantianism, is known for criticising capitalism from a Kantian ethical perspective. Thus far, the role of the notion of humanity in this critique has been viewed as grounding what I shall call the ‘purposive labour reading’. This reading takes Cohen’s primary interest to lie in a reorganisation of work so that our humanity, which requires us to be treated as ends, remains intact. With the aim to better understand the relevant notion of humanity, I contextualise the discussion within the overall framework of Cohens’ neo-Kantian account of ethical cognition and situate his ideas in the context of his contemporary interlocutors. Revisiting Cohen’s remarks on socialism and capitalism against this backdrop reveals that his discussion of labour serves as an exemplar, showcasing how ethical rationality manifests in the liberal socialists’ demands. I argue that his primary aim was not to prepare the ground for a prescriptive labour theory – though this is likely to follow – but to argue for a framework alternative to historical materialism, allowing us to perceive and interpret social practices in an ethical light.

1. Introduction

Hermann Cohen, the founding father of Marburg neo-Kantianism, is known for critiquing capitalism from a Kantian ethical perspective. In his criticism of capitalism, we repeatedly encounter passages drawing on Kant’s formula of humanity. In Kant’s Foundation of Ethics (hereafter, Foundation of Ethics), ¹ he states: ‘The less man is used as a mere means, the more he is free in his actions and destinies’ (Cohen 2001: 286). ² In his second major book on ethics, Ethics of the Pure Will (hereafter, Pure Will), ³ Cohen writes, ‘Man … may be a means only so far as he remains an end at the same time. Therefore, he is an end in himself’ (Cohen 1981: 320). In Kant’s Foundations of Aesthetics (hereafter, Foundations of Aesthetics), published in 1889, he claims, ‘As long as we are regarded merely as labour values, we … function as a means and … [are] utilized and consumed as such’ (Cohen 2009: 139). This idea resurfaces in his mature ethics, where he states, ‘An object possesses value, that is, a market price or exchange value. A person, on the other hand, does not possess value but dignity’ (1981: 322). We see that the notion of humanity plays a crucial role in his refutation of capitalist practices.

Thus far, the role of the notion of humanity has been viewed as grounding what I shall call the ‘purposive labour reading’ (Furner 2019; see also van der Linden 1988). ⁴ This view emphasises Cohen’s claim that capitalism roots in a reduction of workers to ‘mere means’, suggesting that Cohen primarily aimed at a reorganisation of work in a way that preserves our humanity. However, this reading appears limited when we consider the notion of humanity within the context of his unique neo-Kantian account. With the aim to better understand not only Cohen’s usage of ‘humanity’ but the overall ethical approach underpinning his account of socialism, I contextualise his remarks within Cohen’s ethical framework and situate his ideas in the context of the discussions of his contemporary interlocutors. Revisiting key passages against the backdrop of his neo-Kantian account of ethical cognition reveals that Cohen’s discussion of labour serves as an exemplar, showcasing how ethical rationality manifests in the liberal socialists’ demands. I argue that his primary aim was not to prepare the ground for a prescriptive labour theory – though this is likely to follow – but to argue for a framework alternative to historical materialism, allowing us to perceive and interpret social practices in an ethical light.

I will rely on three interpretative assumptions. First, I adhere to the standard assumption in scholarship on neo-Kantianism that Cohen is not just a commentator on Kant but a philosopher who develops his own unique philosophical system. ⁵ Consequently, I do not attempt to determine the extent to which Cohen follows or deviates from Kant. Instead, I discuss his systematic theory on its own terms and place his ideas in the context of his time. Second, I align with interpreters who argue that his main works on ethics, Foundation of Ethics and Pure Will, complement one another (see Schmid 2001: 31; 1995). Third, I follow commentators who take Religion of Reason (Cohen 1966) as a continuation and further exploration of Cohen’s ethical system rather than as marking a systematic shift in his philosophical thinking (see, e.g., Poma 1997, Wiehl 2000: 63, Novak 2000: 227). ⁶

Thus far, I have introduced the overall objective of this article. In section 2, I will first briefly discuss four features that characterise Cohen’s neo-Kantian approach to ethics and then go on to provide a more detailed discussion of each of these features. Finally, in section 3, I will argue that Cohen’s aim was to present an account alternative to historical materialism, allowing us to view socialist movements through an ethical lens.

2. Cohen’s neo-Kantian ethical programme

Cohen’s philosophical programme is best understood as a ‘response to the Materialism Controversy’ (Nahme 2019: 123). Nineteenth-century scientific materialism represented a view that (i) assumed the existence of a reality independent of the mind, (ii) rejected idealism, and (iii) asserted that objective judgements about morals were valid only if grounded in inductive reasoning (Gregory 1977: 7–8). Carl Vogt’s infamous statement that ‘thoughts stand in the same relation to the brain as bile to the liver or urine to the kidneys’ became characteristic for this movement (Vogt 2012: 5–6). Prima facie, it may seem trivial to mention that a Kant scholar, who holds the belief that moral judgements are based on deductive inferences presupposing an idea of universality, disagrees with materialism. However, Cohen would be misunderstood if his idealist critique of materialism were simply considered a refutation of a theory prevalent at the time. Rather, his deep concern with materialism is part of his philosophical programme. Cohen’s philosophy is not limited to studying the a priori conditions of knowledge. Instead, it incorporates a materialist aspect, as it is founded on the premise that rational cognition interacts with ideas materialised in history. In this context, Cohen aims to capture the psychological shifts in concepts that we undergo when we come reflectively to realisation that certain realms, previously perceived as causally determined by matter, are now more accurately understood under the free causality of rational laws.

The psychological aspect of objective concepts comes most prominently to the fore in Cohen’s early ethno-psychological (völkerpsychologischen) phase as illustrated in ‘Poetic phantasy and mechanism’ (2012a), ‘The Platonic idea, psychologically considered’ (2012b), and ‘Mythical Concepts of God and the Soul’ (2012c). In the late 1860s, the young Cohen is interested in how, from a psychological viewpoint, scientific and moral progress comes conceptually into being. Instead of evaluating existing concepts regarding their truth value, he focuses on the social and historical aspect of epistemic concepts, thereby attempting to explain conceptual shifts from a causal point of view in history (Kusch 2019: 251). To achieve this aim, Cohen relies on two assumptions. First, he posits that knowledge is inevitably conceptual. Due to our inability to process a multitude of sensory inputs, Cohen suggests that we ‘compress’ these inputs into concepts, which are then recalled when exposed to the same object (ibid.). Second, he argues that learning processes build upon and synthesise already known concepts (Cohen 2012a). Although Cohen later moves away from his völkerpsychological approach in favour of a critical foundation of experience (Widmer 2021: 255, Edgar 2020: 268–9), he maintains the thesis that rational cognition is conceptual and, as such, inevitably engages with ideas reflecting past standards of morality and truth.

This conceptual sensitivity – rooted in the acknowledgement that rational judgements manifest in historically evolved concepts – is a key aspect where Cohen diverges from Kant. Unlike Kant, who, according to Cohen, aimed to justify a priori principles as metaphysical truths independent of their cultural manifestation, Cohen advocates for a ‘dynamic’ understanding of rational principles (cf. Luft 2015: 54; Friedman 2001). This approach allows for inferences about subjective categories through their empirical manifestation. In Cohen’s view, the role of ethics is to liberate ourselves from concepts unjustifiably categorised according to the causal realm of nature and to render them correctly within the self-attributed realm of free causality. In what follows, I will identify four features that characterise Cohen’s neo-Kantian ethical programme.

1. (i) Ethics, as a philosophical discipline, examines not only the normative statements and their underlying concepts but also the methodological application of moral concepts aimed at promoting historical progress.

2. (ii) Historical progress is rooted in ethical judgements that evaluate the relationships between means and ends, presupposing the idea of the unity of ends.

3. (iii) The coherence of ends is grounded in the concept of humanity, which functions as a hermeneutical guide for discerning moral content.

4. (iv) Duties and rights are relational principles that represent a break from previous social life forms, which were incompatible with the demand that everyone must be acknowledged and treated according to their humanity.

Each of the following sections will explore one of these features. The hierarchy proposed here does not reflect different grades of importance, but rather different levels of abstractness, allowing us to illuminate Cohen’s ethics from various viewpoints. We will begin with the most abstract principle, concerned with the role of ethics in Cohen’s system of critical idealism. With each step, we explore this idea in further detail, ending with the most concrete guiding rule for evaluation of human practices.

2.1 The ‘transcendental method’, ethical frameworks, and historical progress

Cohen conceives of ethical activity as an ongoing cultural practice manifesting in normative judgements. Philosophy theorises about the ethical categories that underlie these judgements, providing frameworks to guide our moral deliberation. To capture this process, Cohen develops what he refers to as the ‘transcendental method’: This method aims not only to explore the logical categories that enable accurate ethical judgements but also operates under the assumption that methodically guided thinking – the correct application of ethical concepts – fosters cultural progress. The aim of this section is to demonstrate that, although Cohen’s ethics makes universal claims, these claims are also considered in their material manifestation (Widmer 2024). Similarly to the case of artefacts that are both objects resulting from method-guided activity and integral to the historical evolution of art, I will illustrate that Cohen’s ethics employs a dual approach: One perspective examines the correct way to engage in ethical deliberation, while the other assesses the effectiveness of ethical frameworks in the advancement of culture.

To understand Cohen’s view, it is helpful to explore the influences that shaped his thinking, leading him to conceptualise universal claims as deeply intertwined with our tangible experiences. Numerous scholars have emphasised the profound influence of Adolf Trendelenburg, Cohen’s teacher, on Cohen (see, e.g., Edgar 2021, Beiser 2018, Kinzel 2023). ⁷ Trendelenburg’s conservative political views did not align with Cohen’s socialism. As a ‘loyal Prussian’, Trendelenburg believed that ‘the Hohenzollers’ (the ruling aristocracy) were the epitome of enlightened rule (Beiser 2013: 71). Nevertheless, Cohen found inspiration in Trendelenburg’s historically ‘embedded’ view of ethical rationality, which serves as the foundation for the normative constraints of natural law (Widmer 2023a). According to Trendelenburg, the Kantian notion of universality is evident in ‘concrete’ normative concepts that shape our ethical deliberation and, consequently, the norms that become realised in history (Brüllmann 2019: 207).

With an origin in ideas tracing back to his Logical Treatises of 1840 (second edition 1870), Trendelenburg agrees with Hegel that rationality is a historical practice. However, in his Natural Law based on Ethics, he highlights what he considers a crucial oversight in Hegel’s dialectical logic: the absence of ‘spatial movement’ (1868: 42). The category of spatial movement would enable us to attribute two contradictory states to one and the same object. For example, Trendelenburg argues that while the sentence, ‘While the day is coming, it is already, and it is not yet’ (p. 38), would violate the law of non-contradiction, which states that A and not-A cannot both be true simultaneously, nevertheless, we would still be capable of perceiving an object in its transitional state. Our ‘receptive intuitions’ allow us to ascribe contradictory states to one and the same object (p. 56).

The ability to perceive an object in its transitional state is also a significant aspect of Trendelenburg’s practical philosophy. In Natural Law based on Ethics, Trendelenburg argues that historically developed norms can be understood in two seemingly contradictory ways: (i) in their spatial-temporal appearance, causally determined by natural laws, and (ii) concerning their underlying purpose, constituted by a free rational will. The latter perspective enables us to assess the rightness of the norms in question: ‘As the spirit opens its senses, … the a priori principle of freedom is found in its movement in physical objects and ethical norms’ (1868: 237). Freedom is not merely an abstract idea; it also manifests itself in appearances that shape our history. Just as we can intuitively ascribe both A and not-A to the same object, we can perceive historical facts as causally determined appearances while simultaneously considering them as originally caused by a free will.

Trendelenburg adopts a loosely Kantian perspective, emphasising an a priori understanding of the ideal purpose, which he considers the ‘ethical origin’ of consciousness (cf. Hartung 2008: 297). However, he not only historicises and materialises moral norms more profoundly than Kant but also abandons the systematic separation of law and morality (Trendelenburg 1868: 16). Trendelenburg argues that all norms, whether moral norms or positive laws, should be assessed with respect to one universal purpose. The aim is to identify legal norms in history established for universal ends, allowing for the distinction between moral developments and those exemplifying non-universalisable, immoral ends. The task of philosophy is to elucidate the underlying logic that permits an ethical interpretation of historical events, regardless of their material manifestation in history.

Resonating with Trendelenburg, Cohen employs the notion of concrete universality, considering ‘cultural facts’ both as part of the empirical realm subject to causal explanations and as expressions of ethical rationality orientated towards a moral purpose. As he conceives of philosophical ethics as the rational and ‘scientific’ endeavour to conceptualise this practice, Cohen encounters a novel problem: If ethical rationality inevitably manifests itself in the spatial-temporal realm, how are supposedly rational and atemporal ethical concepts such as ‘universality’, ‘humanity’, or the ‘end in itself’ to be addressed?

Cohen considers ethics not only as a discipline that promotes objective principles intended to be realised in practice. By regarding his own framework as a ‘method’ that guides moral thinking, he is also concerned with the effectiveness of ethical concepts and their potential to foster cultural progress (2001: 33). To illustrate this dual perspective, consider a scenario in which a morally charged case is presented to a materialistic monist and an idealistic dualist. The materialistic monist denies the possibility of universal judgements on a priori grounds, asserting that our sense of morality is subjective, contingent, and entirely rooted in our experiences. Consequently, objectivity in ethics may only pertain to descriptive statements regarding what people consider to be right. On the other hand, the idealist maintains that, even though the content of our moral judgements varies, we can distinguish between right and wrong based on an implicit a priori principle. These two positions can be examined in two ways: first, in terms of the validity of their theories, and second, in terms of the effectiveness of these methodological frameworks within the context of cultural progression.

The first perspective aims to clarify who is deploying the right concepts. As we will see in more detail in section 2.2, Cohen rejects the prevailing materialist trend of the nineteenth century, as it carries the risk of moral relativism. With his goal of preserving ethics as a discipline dealing with objective judgements, Cohen, a self-proclaimed ‘Platonist’, conceives of ‘ethical reality’ as an arrangement of rationally constructed ideal principles (2001: 36). He asserts that it is ‘not the stars in the sky that are the objects considered by the [transcendental] method as knowledge; rather, the transcendental gaze is directed towards astronomical calculations and the facts of scientific reality’ (pp. 27–8). Similarly, in the context of ethical judgements, we are not concerned with a descriptive analysis of actual moral practices, but with rationally constructed principles aimed at a unified purpose. ⁸ Thus, though mediated through historically embedded concepts in space and time, concepts such as ‘universality’, ‘humanity’, or ‘end in itself’ are perceived as revealing aspects of our moral consciousness. To signify the intermediacy of these concepts, Cohen translates them into ‘functional’ terms, as will be detailed in section 2.3.

The second perspective assesses the expected historical effects when these concepts are applied. Just as we evaluate an artist’s work based on its impact on the course of art history, moral theories are judged by their potential to promote progress. Though Cohen thinks of the materialistic monist as deploying categories that fail to reflect important aspects, these concepts can still effectively advance human history. As we later see in Cohen’s engagement with the history of monotheism and Marxism, it is this perspective that enables him to appreciate theories with which he philosophically disagrees. This teleological perspective on ethical concepts forms the basis for the first feature of Cohen’s neo-Kantian ethics:

Ethics, as a philosophical discipline, examines not only normative statements and their underlying concepts, but also the methodological application of moral concepts aimed at promoting historical progress.

Inspired by Trendelenburg’s account of embedded ethical rationality, which allows us to understand norms as substantive appearances established for an ethical end, Cohen employs an understanding of ethical rationality that is sensitive to both the evolving content of moral norms and the changing frameworks by which we evaluate them.

As we will explore in more detail, these two perspectives play an integral role in Cohen’s discussion of socialism. Socialism is regarded as a movement that constitutes a ‘fact of culture’. Marxism and Cohen’s neo-Kantianism represent two differing views for interpreting this development. Although the historical materialist framework is effective in promoting progress, it is limited by its relativist stance on ethical normativity. In contrast, the framework proposed by Cohen aims to offer more advanced concepts that facilitate rational progress through existing legal institutions and measures.

2.2 The ultimate end

We will now explore Cohen’s conceptual framework in greater depth. At this point, we are not yet concerned with the formal will that guides ethical deliberations or the principle that provides ethical content. Instead, we draw attention to Cohen’s aim to elucidate the notion of universality when we engage in moral deliberation more generally. His goal is to demonstrate that in every objective statement – whether in science or ethics – we presuppose the idea of a systematic unity. Objective judgements in ethics, according to Cohen, are possible only when we consider the social realm as regulated by laws consistent with the overarching purpose of non-contradictory ends. Throughout his works, Cohen refers to this concept interchangeably as the ‘ultimate purpose’, the ‘ideal of the idea of purpose’, the ‘idea of the unity of ends’, the ‘idea of God’, the ‘intellectus archetypus’, or – in Pure Will – ‘allness’ (Allheit):

The theological idea [of God] provides the concept of purpose … as the … archetype of the realm of ends. God, as the correlate of the intelligible contingency of all conceivable things, is the ideal of the idea of purpose, the intellectus archetypus, and the absolute purpose. (Cohen 2001: 305)

According to Cohen, objective ethical judgements become possible when we presuppose the idea of an absolute unity of ends. Although the ultimate end as an ideal rational concept has manifested in various conceptual forms throughout human history, he distinctly separates this line of inquiry from an empirical-psychological approach. This latter approach is a descriptive discipline that reduces moral values to inductive inferences. While the strict separation of deductive and inductive psychological approaches might not appear ground-breaking to those familiar with Kant, it is important for Cohen. Cohen is confronted with the reduction of normative objectivity to empirical truths through his Marburg predecessor and friend, Friedrich Albert Lange. A revival of Kantian ethics at that time could only be successful if it addressed the concerns arising from the empiricist and scientific zeitgeist of the era.

The cultural climate against which Cohen is arguing is best depicted in Lange’s influential book, The History of Materialism (2015). In this work, Lange advocates for a psychologisation of the conditions of experience, which coincides with an aestheticisation of moral values. Lange argues that the conditions of experience are inductively based on empirical assessments of our psychological preconditions. Similarly to when we look through a telescope with a stained lens, where we can infer that the stain appears whenever we do so, Lange argues that judgements of necessity concerning the conditions of the possibility of experience are based on contingent features of our nature (Lange 2015: 265; see also Widmer 2022: 260–1). In this vein, Lange argues that ethical ideals are merely products of our fictive imagination. What Kant would have regarded as the ‘general will’ simply reflects a fictive idea of a unified whole stemming from our aesthetic inclination to harmonise sensual input (Lange 2015: 509; see also Russo Krauss 2023). Just as our psychology drives us to seek symmetry in art, Lange argues that ethical deliberation arises from our aesthetic inclination to perceive the social world as a cohesive whole. The aesthetic ‘standpoint of the ideal’, as he calls it, is considered distinct from our rational pursuit of objectively true principles (2015: 509). Lange thus presents a critique of Kant’s ethical formalism that resonates with the Marxist tradition, suggesting that although Kant’s ethical theory aims to yield objective moral claims, it merely reflects the moral standards of that era. Progress, thus, can only be achieved through empirical insights into wrongs.

In contrast, Cohen argues – initially against Lange and subsequently against Marxists – that while acknowledging that moral cognition inevitably operates with concepts reflecting historically evolved moral standards, objective judgements in ethics depend on a notion of unity that is distinct in kind, requiring deductive inferences (2001: 352). Cohen acknowledges that Kant himself introduces the notion of a focus imaginarius in the Critique of Pure Reason (p. 360; CPR, A644/B672). However, he stresses that the aspect of imagination does not refer to a fictive component of our cognition but to a rationally justified notion of the intellectual archetype, under which a normative viewpoint on our ‘ethical reality’ becomes possible (ibid.). In this vein, Cohen rejects Lange’s use of the term innere Stammesbegriffe to refer to the concept of the systematic unity of ends. Even though Kant himself uses this term, Cohen notes that the notion of Stamm – perhaps even suggesting the notion of innate ‘tribal’ concepts – carries a problematic naturalistic undertone that he seeks to avoid. In any case, neither epistemology nor ethics deals with inborn categories or, to use his own term referencing Kant, an ‘innate’ account of ‘apriority’ (p. 34). By insisting on this point, he seeks to show that a priori principles are not a reflection of our psychological nature or the economic mode of production but rather rational concepts that are necessarily presupposed when we seek to make the social realm intelligible. By regarding ethics as bearing on deliberation regarding means-ends relations under the ultimate purpose of the systematic unity of ends – whatever the empirical manifestation of this idea may be – Cohen emphasises that ethics is distinct from the psychological and physiological conditions of thinking. Instead of seeking causal explanations of norms, ethics focuses on judgements subsumed under right ends.

According to Cohen, what empiricists and materialists overlook is that with any objective statement, we presuppose the idea of a coherent unity. Recall the previously mentioned case of the materialistic monist and the idealistic dualist, each employing distinct frameworks of objectivity. Cohen holds the position that both materialists and idealists implicitly rely on the concept of absolute unity. The difference lies in the idealist’s ability to properly comprehend this epistemological presupposition. Cohen argues that without the idea of an absolute unity of ends, our social interactions would be incomprehensible to us on a rational level. In the theoretical sphere, the idea of unity grounds what Cohen calls the ‘axiom of cognizability’ (Axiom der Erkennbarkeit), which is the assumption that our natural world is governed by a coherent unity of causal laws (2001: 47). To study the laws of nature, we need to perceive the world as if it consisted of logically coherent laws that can be discovered by us. Consequently, our scientific endeavour to seek objectivity is necessarily preceded by the idea of a ‘unity of reason of the unconditioned’, even though it never appears in experience (p. 80). Cohen criticises anti-idealist positions, as found in historical materialism, for insufficiently reflecting this ideal component. Even in inductive judgements, where we infer general principles from individual cases, he argues that we necessarily rely on the presupposition of the idea of absolute unity. Because we can never know all individual instances, generalisations in inductive judgements presuppose this idea, which transcends our empirical knowledge. In the normative sphere, the ‘axiom of cognizability’ is the idea of an absolute purpose where all purposes merge into a coherent system of ends. According to Cohen, the question is not ‘how and by what is freedom as a cause possible?’ but rather ‘to what end does it show itself effectively within the systematic unity of cognition?’ (p. 269, emphasis added).

Although Ernst Troeltsch and other historicists of that era criticised Cohen for interpreting history through an idealised telos (cf. Klubak 1987: 217), Cohen viewed his idealist stance as a counteraction against the value-free investigation of history that was prevalent in his time. He believed that it is only through this approach that we can achieve a rational perspective on social movements, one that emphasises ethical transformations over time. An illustration of this idea is to be found in Religion of Reason, where Cohen argues that the Jewish conception of God was the first emerging concept that allowed for an objective assessment of moral norms. The Talmud spread during a time when most people believed in polytheism, a relativistic moral belief system that did not yet possess the conceptual tools to differentiate objectively between right and wrong (Cohen 1966: 399). According to Cohen’s assessment, the ancient Greeks envisioned their gods as human-like entities, each with a different moral code. While the Greeks depicted their gods making inconsistent claims – one god demanding an act while another forbade it – the Jewish conception of a monotheistic God allowed for a logically coherent moral belief system. Cohen emphasises that the adoption of this framework had significant moral effects in overcoming cruel sacrificial practices that were deeply rooted in ‘mythical’ thinking (1966: 32, see also Widmer 2023b, 2024).

In the modern period, the notion of a systematic unity of ends is regarded as a logical concept that provides us with the most accurate framework for an ‘ethical epistemology’ – as Cohen calls it at times. Although section 2.1 has shown that he acknowledges the possibility of evaluating materialist perspectives based on their potential to promote progress, we now see that, for Cohen, only a transcendental idealist position – one that considers the assumptions that render the social realm ethically intelligible to us – offers the appropriate evaluative framework. With his critical interpretation of the idea of a unified system of ends, Cohen not only aims to demonstrate, in contrast to the prevalent worldview, that Kantian deductive logic is the right method for evaluating historical movements. As we will see later, he also uses this argument to counter Marxist strands that refute the validity of idealism and ethical normativity. Thus, the second feature of Cohen’s neo-Kantian ethics can be put as follows:

Historical progress is rooted in ethical judgements that evaluate the relationships between means and ends, presupposing the idea of the unity of ends.

The systematic view of ends also enables an objective evaluation of historically evolved practices. In this vein, Cohen argues that the objective reality of moral principles ‘is not real in experience’, but one that ‘becomes real through our actions’; it creates the ‘transcendent perspective’ or the ‘standpoint of the noumenon’ from which our ‘ethical reality’ is presented to us (2001: 309).

2.3 The functional principle of humanity

We have seen that, in Cohen’s system, the systematicity of ends plays a crucial role. However, coherence on its own is not enough. If coherence were the only criterion, then the maxim ‘to lie if it serves my purpose’ could easily become a general rule if we do not simultaneously accept the contradicting principle that lying is wrong. Both Kant and Cohen would be dissatisfied with such logic. Ethical judgements are not solely concerned with a unified set of ends; the ends of practices must, at the same time, be coherent under the presupposition that everyone is viewed as a ‘member of humanity’ (2001: 362). In this section, I aim to demonstrate that the concept of humanity, specifically the principle of never treating someone solely as a means but always also as an end in itself, serves as a hermeneutical guiding principle in Cohen’s neo-Kantian system for discerning moral content.

Neo-Kantian and contemporary scholars have approached the concept of humanity in various ways. James Furner interprets Cohen’s formula of humanity as a principle that establishes a perfect duty to condemn exploitative working practices under capitalism. This, he argues, arises from the logical inconsistency between the principle of ‘never treating humanity in a person merely as a means’ and the entailments of labour practices such as ‘menial machine labour’ (Furner 2019: 17). A more accurate rendering of Cohen’s formula of humanity can be found in interpretations where the notion of the end in itself is viewed as an ‘interpretative concept’, constituting a ‘hermeneutical approach’ (Schmid 1995; see also in contemporary Kant scholarship (Wood 2017: 65). According to this approach, the concept of humanity, by itself, does not yet establish a specific duty. Instead, it enables us to interpret human behaviour in a moral light, discerning whether an action is ‘respectful or disrespectful, caring or uncaring’, etc. (ibid.). As the subsequent paragraphs show, Cohen utilises the formula of humanity similarly as a hermeneutical principle in his own unique manner.

The hermeneutical aspect is expressed in Cohen’s ‘functional’ reading of humanity (1981: 35–6, see also Widmer 2023b, 2024). While we repeatedly encounter the claim that ethical judgements should be considered with regard to their ‘function’, Cohen himself does not elaborate much on what he means by that. However, his student, Ernst Cassirer, wrote an in-depth treatise on how to deal with a priori principles in functionalist terms, arguing that while we cannot have any insights into the forms of principles without their content, we gain access to them through their symbolic appearance as they manifest in forms of life. It seems in order to interpret Cohen’s functional take on humanity through the lens of his student. As a first step, I will highlight the relevant aspects in Cassirer, which will provide us with a better understanding of how ethical concepts are considered in functional terms. Then I will demonstrate that Cohen’s notion of humanity is also to be understood as a functional concept, providing us with moral content created by the most advanced moral framework available at the time.

In 1910, two years after the publication of the second edition of Pure Will and the same year in which the second edition of Foundation of Ethics appears, Cassirer introduces a functional interpretation of the Kantian principles in Concept of Substance and Concept of Function (Cassirer 1910). To illustrate what he considers an improvement over the original Kantian theory, he uses the following example: If we see an object, we can only notice some parts simultaneously, necessarily bound to a ‘perspective’ (p. 383). But despite our limited perceptual access to things as they are, we inevitably create an idea of an object, which allows us to recognise the object as a unified entity even as we look at it from a multiplicity of angles. This gives us an indication of how our consciousness follows universal laws that structure our experience. The ‘function of the law’, grounded in our ‘consciousness’, is to ‘produce the idea of an object’ as, without it, we would not be able to recognise an object as a unified entity (ibid.).

In the same vein, Cassirer interprets the ‘pure will’ in functional terms. He states that a scientific rendering of ethical judgements would leave us with a ‘functional’ interpretation of Kant’s ‘ethical basic concepts’ (1939: 83). These ethical concepts never appear in experience but are mediated through normative judgements in time and space. In Freedom and Form, Cassirer argues that the ‘function of the pure will cannot be considered without its relation to the object’ (2001: 237). For example, the idea that sacrificial practices are morally wrong is not an immediate truth that is just given to us. Rather, we become gradually more and more aware of the pure functions of judgements through our changing moral frameworks and conceptions of rightness. What all moral judgements have in common is their ‘function’ to create morally better forms of life (ibid.). The formal requirements are thus presented in ethical judgements. They become apparent when we examine the purpose of these judgements in the course of history. Whereas the content and the conceptualisation of our ethical concepts change over time, their function to create a better society in the course of human history remains the same.

As Cassirer’s An Essay on Man shows, the empirical outlook provides the basis to track the evolution of ethical rationality. Cassirer argues that any symbolic form can function either as an instrument of oppression or a product of ethical rationality seeking progress. At the heart of evaluation of the matter lies the Kantian notion of autonomy, functionally perceived: it serves as the ideal criterion against which materialised moral frameworks are measured. Illustrative of this idea is Cassirer’s differentiation between ‘static religion’ and ‘dynamic religion’ (1944: 116). Static religion is a ‘product of social pressure.’ Dynamic religion, by contrast, breaks with ‘all the former social bonds’, seeking more autonomous forms of life and leading to cultural progress (p. 131).

Though Cohen is less explicit about his functional interpretation of Kantian concepts, his perspective on the concept of humanity is best understood in Cassirerian terms, namely as a moral concept that, even though the manifestation of the idea is subject to historical change, remains functionally the same.

The transcendental method [is interested in] … the supreme principles of an experience given in printed books and actualized in history. If these principles claim to hold a priori within the presupposed experience, they cannot be derived from a world of things – because a priori is that which we ourselves place into things. … The principles receive their origins in the forms of our thinking, in the functions of judgement. (1981: 35–6, emphasis added)

As a functional concept, the notion of humanity grounds the ‘limiting condition’ that renders the content for duties and rights.

[T]he community of autonomous beings is the absolute end, that is to say, the subject of the moral law or humanity is the ultimate end. … [H]umanity is the supreme limiting condition of individuals. (1981: 274, emphasis added)

The idea of treating individuals according to their humanity serves the function of rendering social norms in an ethical light. It sets the boundaries of an ethical framework and the hermeneutical condition under which the ethical content is determined.

Passages from Religion of Reason are helpful in illustrating how Cohen applies this idea. Although the concept of humanity was introduced in modern times by Kant, Cohen argues that a similar expression of the law already appeared in ancient times. According to Cohen, the Jewish concept of ‘mizwa’ signifies both ‘law and duty’, without relying on the mythical idea of a transcendent soul existing beyond worldly existence (1966: 401). However, Cohen notes that the Pauline conception of law reintroduced mythical elements back into our cultural consciousness, such as the belief in a ‘true self’ existing beyond death (p. 401). Similarly to Cassirer, with the Kantian notion of humanity in mind, Cohen compares different religious concepts functionally. In this vein, he seeks to show that the Pauline conception of law deviates drastically from an ethical understanding of law. Driven by the fear of punishment, it suppresses the individual’s free choice. This marks a step backward in the evolution of ethical rationality. By tying moral laws to the idea of a transcendent soul, Cohen asserts that a more advanced Jewish ethical understanding of the concept of law had been suffocated, only replaced much later in Lutheran Protestantism by ‘an epistemology of liberalism’ (Nahme 2019: 82). Though the concept of humanity emerged in its pure form much later, it had, according to Cohen, functionally already been introduced by the Jewish tradition.

Following the two-fold account of the transcendental method, Cohen thus employs the notion of humanity in two ways: (i) explicitly to gather the moral content of ongoing practices and (ii) implicitly to trace progressive movements in history. In both cases, the Kantian principle of humanity or its equivalent at a given time serves the purpose of interpreting normative practices in a moral light. This provides us with the third feature of Cohen’s neo-Kantian ethics:

The coherence of ends is grounded in the concept of humanity, which functions as a hermeneutical guide for discerning moral content.

2.4 Relational duties and rights

Having discussed the critical hermeneutical framework for regarding the social sphere, we will now look more closely at the types of duties and rights in Cohen’s work, which are understood relationally and serve as norms guiding our social interactions. The social component is apparent when Cohen introduces what he calls the ‘grounding principle of ethics’:

Every human being, as an autonomous author of the moral law and a member of the realm of ends, must be considered as an end in itself, regardless of how much this individual mortal may or may not align with the idea of humanity that they themselves carry as the prototype of their actions in their soul. No person ought to be used merely as a means. Every person must always be regarded as an end in itself in the administration of a moral world. For the human being, as a moral being, is an end in itself; that is to say, they are free. (2001: 290, emphasis added)

This passage refers to both the rational entailments of the ultimate end, as specified above in 2.2, and the moral content determined by the principle of humanity, as discussed in 2.3. However, in this section, we shall focus on the relational aspect of norms, as illustrated in Cohen’s reference to the ‘administration of the moral world’. Cohen’s ethics is not concerned with maxims guiding the action of individuals, but rather with rules that constitute our sociality: our legal frameworks.

To illustrate Cohen’s ethical concern with the legal institutionalisation of norms, it is instructive to take a closer look at his criticism of Kant’s ‘point of view’ in the well-known murderer-at-the-door case (1981: 528–9). Cohen criticises Kant’s response for failing to treat this as a case dealing with a norm that could potentially become a guiding rule in a just society. The problem, according to Cohen, lies in the fact that Kant would argue that even a case of ‘unjust coercion’ would not justify a ‘good-natured lie’, making a lie morally reprehensible even if it prevents a murder (ibid.). The issue Cohen raises here is that Kant treats the maxim as that of an individual act and not as a relational rule, leading to problematic results that go against our moral intuition. Cohen seeks to resolve this issue by adopting a social viewpoint. Instead of describing the act in terms of the individual person answering the door, he assesses the action-guiding rule relationally. In Cohen’s view, the relevant act description in the murderer-at-the-door example is not the ‘lie’, but the ‘prevention of a murder’. If an act can be reconciled with the free community without contradiction, it is considered ethically justified. Just as the rule of ‘self-defence’ is not seen as ‘murder’ or ‘a murder in self-defence’, Cohen argues that the act description of the ‘lie’ in Kant’s example should be replaced with a description that takes into account the coercive nature of the situation and its potential to become a general rule in society (ibid.).

This shift towards the social realm leads to another consequence: the dissolution of the systematic separation between internal moral laws and external legal laws (see Gibbs 2005). In Kant’s view similarly to Cassirer, the external realm is considered as systematically distinct from the internal realm, regulated by an account of freedom that states, ‘any action is considered right if it can coexist with everyone’s freedom in accordance with a universal law’ (Kant 1996: Ak. 6: 230). Cohen dismisses this approach with the argument that it would lead to a duty of ‘sanctifying the existing legal and state order’, which undermines a critical assessment thereof (Cohen 1970: 79). Thus, for Cohen, concerned with moral progress, the pure will serves as the only rational principle imposing normative constraints on the social realm. In Cohen’s words: ‘The ethics of the pure will must become the principal doctrine of the philosophy of law and state’ (p. vii). Cohen’s social ethics intends to foster a critical attitude towards existing legal norms that regulate our interpersonal relations. Accordingly, ‘ethical logic’ is to be deployed in the ‘legal sciences’ (Rechtswissenschaften) (ibid.).

Systematically, this idea traces back to Cohen’s elaboration on the logic of ‘the Other’ in Pure Will, where he raises the question of how the ‘self-consciousness of pure will’ is possible (1981: 282). According to Cohen, the reflective self is genuinely social, constituted in confrontations with others in our social interactions. ‘The I’, argues Cohen, ‘could not be defined or generated if it were not conditioned by the Other’ (p. 212). Albeit in a somewhat modified form, Cohen utilises Fichte’s concept of the ‘not-I’ in order to justify the relational ethical logic underlying juridical deliberation. The ‘A=A’ or the ‘I am I’ represents the reflecting subject in Fichte’s philosophy of scientific thinking (Fichte 1846: 95). According to Cohen, Fichte was the first to develop the concept of the ‘not-I’ as a concept of the ‘Other’ that is taken in a purely logical sense. ⁹ As Cohen understands this principle, the ‘first A’ refers to the non-reflective and heterogeneous subject. The ‘second A’ refers to the I that ‘makes itself the object of reflection’ (1981: 208). Ethical reflection is initiated by the ‘not-I’. However, whereas the not-I in Fichte encompasses objects, Cohen’s usage refers to other persons only, thereby marking the realisation of the Other ‘in terms of the concept of humanity’ (p. 209). Ethical consciousness is only attainable through confrontation with the ‘consciousness of the Other’. This encounter fosters our deliberation as to how to treat others, acknowledging that they possess the same rational capacity as we do (p. 212).

It comes as no surprise that Cohen finds inspiration in Fichte’s embedded account of subjectivity. According to Fichte, the task of philosophy is to reflectively distance oneself from ideas shaped by history and replace them with rational ones. For two reasons, Cohen, who views philosophy as a continuous rational or ‘scientific’ endeavour to refine moral concepts, finds Fichte’s approach particularly appealing. First, Cohen’s adaptation of the recognition of ‘the Other’ acknowledges that laws regulating our interpersonal relationships evolve historically. Second, the rational aspect of ‘the Other’ – in Cohen’s interpretation, the notion that everyone should be viewed in their rational capacity to set ends – provides the logical foundation for rational renegotiation of the historically evolved concepts in question.

In legal discourse, this process most evidently reflects our ethical logic. ‘The logic of pure knowledge is founded on mathematics … In a similar manner, … ethics [is orientated] towards jurisprudence. Jurisprudence is the mathematics of the social sciences’ (1981: vii). According to Cohen, jurisprudence, where persons are seen as entirely detached from their empirical and contingent characteristics, aims at justifying just rules regulating interpersonal actions. The focus lies here solely on the ‘objective’ side of the human agent (pp. 216, 223). Just as mathematics is taken to be the ‘logical method’ of the natural sciences, Cohen designates ‘ethics’ as the ‘logical method of the human sciences’ (Humanwissenschaften) (ibid.). In this vein, the ‘state’ is taken as the institution that ought to actualise moral principles (cf. p. 242). Rules of social and cultural practices are morally just only if their end is coherently rendered under a perspective where everyone is recognised and treated in accordance with their rational nature. Socialism, as we will see, is an expression of our rational inclination to renegotiate laws so that labourers are recognised as rational beings with intrinsic ends in the legal-economic sphere: a realm that has thus far been regulated by the end of accumulation of capital. The fourth feature of Cohen’s ethics is a further specification of the type of rules in question in regard to the ultimate purpose:

Duties and rights are relational principles that represent a break from previous social life forms, which were incompatible with the demand that everyone must be acknowledged and treated according to their humanity.

3. Capitalism, Marxism, and ethical socialism

Having explored various aspects of Cohen’s ethics, thus far illustrated in his engagement with the history of religion, we now shift our focus to his political philosophy. In his commentary on capitalism, socialism, and Marx, as presented in Ethics of the Pure Will (1981: 312ff, 505ff, 606ff), Cohen does not expend much effort on recapping his background theory. Prima facie, his remarks suggest that Cohen’s prescriptive statements express solely his efforts to apply Kant’s formula of humanity, both to condemn capitalist labour practices and to justify ethical forms of labour. James Furner, an advocate of this interpretation, focuses on a passage that states, ‘If through the division of labour an isolation of action is brought about, through which the unity of a person is destroyed, then the division of labour destroys the unity of culture’ (p. 607). Furner contends that the conceptual contradiction between being treated ‘merely as a means’ and ‘menial machine labour’ justifies a perfect duty to condemn capitalism (Furner 2019: 18). Similarly, van der Linden views Cohen as a labour theorist who employs the humanity principle to advocate for a democratic organisation of work. Drawing on Cohen’s assertions – ‘The vocation of morality cannot be fulfilled except on the basis of scientific insight’ and ‘All people must become part of the scientific work’ – van der Linden suggests that Cohen argues for everyone’s involvement as ‘co-legislators of the institutions in which we function’ (Cohen 1981: 505–6; van der Lindenn 1991: 350, 1988: 222). While these interpretations are not incorrect, they do not fully capture the breadth of the conceptual shifts that Cohen aims to introduce. In this section, my aim is to show that Cohen’s remarks take on a different meaning when placed within the context of his systematic ethical framework. I will argue that Cohen’s remarks are more accurately understood as an effort to introduce an ethical reinterpretation of left-wing concepts, which were predominantly defined in Marxist terms at that time.

More specifically, I shall show that Cohen implicitly engages here with Marxist ideas from two distinct angles: a perspective of progress and a prescriptive viewpoint. From the perspective of progress, Cohen values historical materialism as a framework that liberates us from capitalist ideology, and by doing so promotes cultural progress. However, from the prescriptive standpoint, Cohen criticises what he perceives as the ‘natural scientific’ interpretation of Marx, advocated by ‘orthodox Marxists’, which neglects normative theory in favour of a focus on revolutionary goals. By distinguishing these perspectives, I aim to show that Cohen’s ethical interpretation of the division of labour and his support for worker unions serve as examples that illustrate how, when socialist themes are reinterpreted through his neo-Kantian ethical lens, they lead to an endorsement of a liberal-reformist programme.

Let us begin with Cohen’s examination of Marxism from a perspective of progress. For Cohen, determinism and freedom are not metaphysical concepts, but epistemological terms defining the realm of cognition and the idea of unity under which laws are properly subsumed: either the unity of causal laws or the unity of ends. This differentiation allows us to understand objects from either a natural or an ethical standpoint. A central characteristic of his theory is the effort to achieve a precise understanding of concepts within their appropriate framework, as caused either by external natural forces or by human agency. Cohen spends considerable time criticising instances where we erroneously ascribe natural-causal origins to phenomena that are actually human-made. Despite their differing philosophical foundations, Cohen and Marx share this aspect.

One of Marx’s central aims in his analysis of capitalism is to demonstrate that, contrary to what classical economists suggest, natural events causing pauperism and food shortages have been exploited to promote capital accumulation. This is exemplified in various arguments in Capital. For instance, Marx discusses a period between 1797 and 1815, during which the increasing infertility of soil necessitated the use of products that raised production costs (2020: 488ff). Contrary to expectations that higher production costs would naturally lead to a decrease in surplus value, Marx reveals that what actually transpired was an adjustment of working conditions to ensure continued capital accumulation. By employing statistical data, Marx aims to show that surplus value rises with each additional hour of work. This is evident in the legal changes of that period. According to Marx, the history of capitalism demonstrates that an increase in production costs is accompanied by longer working hours and efforts to boost efficiency of labour.

By rewriting economic history to illustrate how our labour practices were historically adjusted to secure the accumulation of capital, Marx challenges attempts to normalise capitalist developments by naturalising them. Though, on Marx’s account, Malthus proves to be a thorough thinker who – in contrast to Ricardo and West – acknowledged the extension of the working day in response to the increase of production costs, he criticises him for naturalising pauperism (Marx 2020: 489). In An Essay on the Principle of Population (1798), Malthus claimed that food production, constrained by the fertility limits of crop fields, progresses arithmetically, while the human population increases exponentially, leading naturally to periods of food shortages and overpopulation. Marx, however, objects that Malthus, due to his ‘conservative interests’, failed to see that it was, in fact, the ‘excessive extension of the working day, together with the extraordinary development of machinery and the exploitation of women’s and children’s labour’, that caused the ‘overpopulation’ of labourers and consequent poverty (pp. 489–90). Here, we see that Marx challenges theories that explain shortages and poverty as if they were effects of irrefutable ‘laws of nature’. In reality, he argues, these issues stem from the ‘historical natural laws of capitalist production’ (p. 490, emphasis in original).

Cohen also aims to counter the problematic naturalisation that occurs when society prioritises the accumulation of capital over humanity (1981: 606ff). Utilising his functional concept of humanity, Cohen criticises the prevailing capitalist worldview, which elevates objects to the status of personhood while unjustly reducing individuals to mere objects, as though they belong solely to the realm of nature. In this vein, he re-examines fundamental legal concepts, such as inheritance laws, capital, labour contracts, and the division of work. Similar to the Pauline law that reintroduced mythical elements into the collective consciousness, Cohen views inheritance laws under capitalism as also containing mythical elements – evident in the idea that a person can ‘continue to live symbolically’ after death by passing on ownership to the next generation (p. 606). The concept of capital, too, shows mythical elements. For instance, foundations under capitalism function as legal entities with rights, effectively ‘acting like persons’. In contrast, labour contracts under a capitalist division of work allow for reducing labourers to ‘mere things’, thereby losing their status as persons (p. 607). Instead of being seen as rational beings and ends in themselves, they are naturalised and objectified, becoming mere instruments for the overarching goal of capital accumulation.

Despite originating from different philosophical backgrounds, Cohen and Marx share the goal of exposing the capitalists’ interests hidden in ideas that masquerade as laws of nature, revealing them instead as human choices that could have been otherwise. Earlier, I demonstrated that while Cohen rejects materialism, his theory still allows for the evaluation of a materialist framework in terms of its progressiveness. In this regard, Cohen places significant value on ‘social physics’, a purely statistical method of exploring the social realm, as mentioned in Pure Will (1981: 311). He recognises the potential of such statistical analyses, like those found in Capital, to foster progress by liberating us from the belief that we are merely subject to capitalist laws, showing instead that these laws are changeable. In this vein, Cohen reflects on the pragmatic use of an imperfect framework that nevertheless contributes to progress, asking, ‘[W]ould it be right to think that it is indifferent under which mask help is given to people, as long as help is given to them?’ (ibid.). Although he is not completely opposed to this idea, he ultimately rejects it.

Cohen emphasises that Capital is based on the moral judgement that prioritising the goal of capital accumulation over the rational ends of humanity is wrong. According to Cohen, this judgement also entails the ethical impetus to change the formal conditions of economic measures, or the ‘legal side … of the economy’ (1981: 309–12). He aims to demonstrate that in societies where capital accumulation is the ultimate end, ‘the law adapts itself to all the forms and deformations which the obligation of labour invents’ (p. 610). In this context, ‘[c]apital is the great inventor; the law must build the catacombs in which it buries itself’ (ibid.). However, when viewed under the end of humanity, the law is not merely a tool of oppression but a means of redressing these wrongs. Through his ethical framework, which highlights capitalist inconsistencies within an ethical idea of humanity, Cohen observes that ‘the evils of culture carry within themselves the seed of their cure. The mythology of capital brings about the redemption of the working person itself’ (ibid.). The task of ethics, as Cohen sees it, is to liberate ourselves from concepts under which the law becomes an oppressive tool and to subject ourselves instead to rationally constructed laws coherent with the idea of the unity of ends.

According to Cohen, Marx and his ‘orthodox’ followers failed to recognise the potential inherent in the legal aspect of society. While he acknowledges that statistical measures can lead to conceptual shifts that foster progress, Cohen argues – similarly to his critique of Lange’s materialism – that a well-developed theory must acknowledge the ideal presuppositions underlying inductive inferences. The appropriate epistemological framework for making the social realm ethically intelligible involves a theory that incorporates an understanding of the unity of ends. By Cohen’s lights, a functional account of humanity rationally defines concepts that regulate our interpersonal behaviour and manifests in the legal system. If this framework is correctly applied, it will naturally lead to a reformist agenda, similar to that advocated by the liberal wing of the social democrats of the time.

Cohen completed his ethics between 1889 and 1910, a period during which Eduard Bernstein advocated for a reform-based revision of the social democratic programme. The ‘orthodox Marxists’, including thinkers such as Nikolaj Berdjajev, Karl Kautsky, and August Bebel, viewed capitalism as operating on its own logic, necessitating a revolution of the socio-economic system. Owing to their radical stance that ideals merely reflect the ideology of the ruling class, they were highly critical of Bernstein and the liberal camp. As Kautsky put it, ‘the categorical imperative is a springboard into an unknowable world. No principle is without sensory intuition, and, like all other cultural products, it merely reflects concrete societal relations’ (Kautsky 1970: 197). Though Cohen acknowledges the socialists’ call for a ‘scientific’ approach as reflecting a more advanced cultural stage, he criticises the reductionist view of the orthodox Marxists. As they limit their critique to natural causality, the orthodox Marxists reduce legality to merely a tool for serving arbitrary interests, thereby wrongly inferring that a ‘necessity of revolution’ is the only meaningful way to transform the social realm. ‘In this consequence [that the concepts of action and legal subject become obsolete] lies the error of the materialistic view of history’ (Cohen 1981: 312). Cohen aims to move beyond a Marxist framing of the problems that lacks awareness of the ethical ends they aim to achieve. His approach seeks to reframe the problems in a liberal context. In this vein, he focuses on tangible experiences – ‘social facts’ – and seeks to interpret them through an ethical lens. This approach fosters not only a critical stance towards immoral concepts but also highlights normative pathways to address these injustices.

Keeping this in mind, the framework against which Cohen defines work and other normative concepts shifts. His examination of the division of labour and worker unions is not meant to lay the groundwork for a prescriptive theory of work; rather, a detailed examination of his wording on the backdrop of his ethical theory suggests that he aims to offer an alternative viewpoint to both capitalist and Marxist frameworks, which dominated the framing of socialist issues at the time. For example, in his discussion of the capitalist division of labour, Cohen argues that the ‘isolation of action’ in the capitalist division of labour disrupts the ‘unity of culture’ (1981: 607). Behind this statement, two judgements hover: (i) that prioritising capitalist objectives over the aim of humanity is erroneous, and (ii) that understanding society primarily in terms of their material class differences, rather than on the background of a unified end, is wrong. While the capitalist rendering of the division of labour disregards workers in their relational end-setting nature, the Marxist perspective disables the normative viewpoint from which reformist changes can be properly articulated.

In the same vein, Cohen’s discussion of worker unions emerges as an ethical re-examination of a matter of controversy at the time. While the radical Marxist wing considered the pursuit of proletarian interests through worker unions – an idea originating from the liberal party in Germany – to be futile, Cohen maintains that unions facilitate a rational collective deliberation process, in which workers act according to their inherent nature of setting ends. ‘It is precisely in this independence from the current individual aspect that they [the worker unions] differ from a majority [voting system]; it constitutes their value as a totality and legal entity’ (1981: 243). His defence of worker unions implicitly criticises both: (i) the capitalist organisation of work, which undermines the personhood status of labourers, and (ii) the orthodox Marxist stance, which dismisses unions as mere reinforcements of existing class differences, thereby prioritising subjective interests over the rational united will.

Though the ‘purposive labour’ interpretation offers valuable insights into how Cohen’s prescriptive statements could lead to a normative theory of work, Cohen’s primary goal was to re-examine concepts previously defined by Marxist or, worse, capitalist frameworks. While Marx has done a great deal to provide a framework that puts the focus on the right issues, Cohen seeks to correct the Marxist worldview by providing a philosophical foundation reflecting the ethical assumptions inherent in their critique. Similar to the Protestant worldview that reintroduced a liberal epistemology to the religious consciousness, Cohen believes that the most crucial task of his age is to replace the myths introduced and perpetuated by capitalism with an ethically corrected Marxist account that places the end-setting being at the heart of legal theorising.

It is truly not only a progress of ethical culture but also indirectly a progress of ethical science that in our century … the question of optimism has been replaced by the problem of socialism. The Job of our era no longer asks whether mankind has more sunshine than rain, but rather whether one person suffers more than their neighbour, and whether in the distributive justice of pleasure there exists a predictable connection that makes an increase in pleasure for one member in the realm of morality the logical fate of another member’s decrease. (Cohen 2001: 368)

According to Cohen, by emphasising class differences rather than the unifying ethical nature of human beings, the orthodox Marxists’ focus on revolution leads to the destruction of a ‘scientific’ ethical mindset that seeks the unity of ends. ‘Whoever accepts the thesis that man is solely a product of the economy … has committed himself to Mephisto; he has abandoned the distinction between matter and form and the morality of man’ (1981: 314). Cohen maintains that ‘from the spirit of unity, the demand for reform has emerged, forcefully and defiantly opposing revolution’ (p. 253). His claim, ‘Kant was the true and real originator of German socialism’, reflects this perspective (Cohen 1970: 71). In Cohen’s view, it is the Kantian, or rather, the neo-Kantian actualisation of transcendental idealism that provides the correct ‘methodology’ for rendering legal concepts in a moral light – a task that finds its purest expression in the demands of liberal socialists of the time.

4. Summary

I have demonstrated that Cohen’s ethical justification of socialism primarily involves applying his ‘transcendental methodology’, guiding us in how we ought to reinterpret concepts that are, under capitalism, shaped by the end of accumulating capital. Though his ethical rendering of legal concepts allows for the derivation of rights and duties, I have argued that, rather than categorising him solely as a prescriptive labour theorist – as commentators have stressed – Cohen’s remarks are better understood as aiming to cast Marxist concepts in an ethical light. Though Cohen saw in Marx’s Capital a framework that fostered cultural progress, he aimed to show that, from an ethical-teleological perspective, the liberal socialists’ programme opting for legal reforms and worker unions, rather than the radical theory of revolution emphasised by orthodox Marxists, appears as the rational antidote to capitalist wrongs.