Two left turns to science: Gramsci and Du Bois on the emancipatory potential of the social sciences

Introduction

There is a paradoxical relationship between science and the left, including intellectuals, social movements, and, more generally, left-leaning individuals. On the one hand, there seems to be a general skepticism toward and rejection of science as a potential source of authority. On the other hand, with the current politicization of issues like climate change or public health policies, ‘Science is real’ has become a left-wing slogan. Skepticism toward science has been especially strong among Marxism-inspired theorists in the post-war era. For example, Ernesto Laclau and Chantal Mouffe claim, in their book Hegemony and Socialist Strategy, that ‘the roots of authoritarian politics lies in this interweaving of science and politics’ (Laclau and Mouffe, 2014: 49). In all fairness, they refer to a specific kind of interweaving that attributes a necessary and existential priority to some interests over others. But they nonetheless speak about science, not religion or philosophy. On the other hand, Bruno Latour in his 2004 paper ‘Why Has Critique Run Out of Steam?’, reflects on the possible danger of scientific skepticism and worries that his critique and study of the natural sciences could have contributed to conspiracy theories and climate change denial (Latour, 2004: 227). This presents two contradictory concerns: one worries about granting science too much authority, and the other about granting it too little authority. These concerns are not new but were already present among socialist thinkers of the early 20th century.

I use the works of Antonio Gramsci and W. E. B. Du Bois as representatives of these approaches toward the role of science in an emancipatory project. I take Gramsci to reflect the first set of concerns: a skepticism toward science in general, and the social sciences in particular, as an element of the ruling classes’ ideology that is embedded in the status quo. In contrast, Du Bois reflects the second approach, namely, despair towards the impotence of science, as a practice aimed towards the discovery of truth, to dispel prejudices and biased beliefs that are the product of ingrained customs and habits. Despite their differences, I argue that Gramsci and Du Bois overlap in many of their critiques and appraisals of the practices of the social sciences. Moreover, I argue that where they do diverge, their views do not form a contradiction, but instead present a productive tension that can be used to better understand the possibilities and limitations of the use of the social sciences in emancipatory projects.

The second part of the article builds on these productive tensions. Through a critical analysis of their work, I assess how the flaws in their framework can be overcome without abandoning the valuable insights in a way that can contribute to current debates in the philosophy of science. I argue that Du Bois's strong non-epistemic value-free conception of science can be modified by adopting Gramsci's conception of science as translation without relinquishing his resistance to wishful thinking. On the other hand, I argue that Gramsci overestimates the receptivity of people toward scientific findings, which, to change minds, must be accompanied by rhetoric, as Du Bois argues. Even when taken together, Gramsci and Du Bois's approaches do not offer a panacea to the tension on the Left regarding science. However, they are helpful in underscoring the way in which the potential emancipatory role of science is mediated by the social world in both the way it is produced and the way in which it is expressed.

Strangers, scholars, and activists

Gramsci and Du Bois never met each other; nor did they read each other's work. They lived miles apart and were confronted with different political contexts with different challenges. Nonetheless, both shared a similar world-historical background: the rise of imperialism, fascism, World War I, the Russian Revolution, and so forth. Moreover, despite their geographic separation, they shared a similar academic background. Both read and were influenced by the work of William James, ¹ Max Weber, ² and of course Karl Marx. Their political projects also overlap much more than it would seem at first glance. True enough, the main thread running through Du Bois's work is race, which received only marginal attention in Gramsci's Notebooks. It is also true that Gramsci was engaged with communism and Marxism much earlier and more deeply than Du Bois, who became a member of the communist party only at the end of his life. Nonetheless, both are politically committed to an emancipatory project for those whom we might term the submerged subalterns. Subaltern is a concept used by Gramsci to refer to all non-ruling classes; this thus involves groups that are not fully marginalized but have stakes in the status quo (Gramsci, 1977: 1722–3; 1992: 92). In contrast, Du Bois used the term ‘submerged tenth’ to refer to the most marginalized and oppressed group in different societies (Du Bois, 1998[1935]: 46, 206; 1999[1920]: 83–4). The submerged subalterns, then, are the group that has stronger grievances against the status quo than other groups – the co-opted subalterns – which may either support the reactionary project of the ruling classes or join the submerged in an emancipatory project.

While this is a feature that Gramsci and Du Bois have in common with many other thinkers, they nonetheless stand out in their emancipatory strategy in advocating for an alternative to strategies of revolution and reform. Both men became disillusioned with the prominent strategy for social change in their time. For Gramsci, disillusionment came after the Biennio Rosso, the two red years after World War I that saw factory occupations and peasant revolts all over Italy but, instead of toppling capitalism, ended by paving the way for fascism. For Du Bois, it came after the lynching of Sam Hose in 1899 and the Atlanta riots in 1906, when he came categorically to reject Booker T. Washington's assimilationist approach. Instead of reform and revolution, Gramsci and Du Bois came to advocate a strategy of trench warfare (Du Bois, 1986[1934]: 1258–9; 1995[1940]: 6, 194, 199–200; Gramsci, 1971: 235; 2007: 162–3). The main feature of this is the conceptualization of emancipation as a long-term struggle, which means that while progressive groups need to keep up the pressure, they also need to brace themselves in order to be able to continue and uphold this struggle. Furthermore, this aspect of trench warfare emphasizes the importance of coalitions between different classes and social groups. Finally, it sees the struggle as taking place in broader and more diverse institutional arenas, including academia. More explicitly, they reflect on whether or not science and academia can form an arena that can transform the ingrained beliefs and habits of the subalterns, both submerged and co-opted, in a way that fosters action and the formation of a coalition for social change. Hence the dilemma of scientific authority, whereby science may have too much authority, and simply justify the status quo, or too little, and be powerless to change people’s beliefs.

Considerations about science in general, and the social sciences in particular, have a central place in the emancipatory thought of Gramsci and Du Bois. This is because their own academic education witnessed the development of the social sciences as a field. Du Bois encountered sociology as a field using quantitative methods at the University of Berlin, where he studied from 1892 to 1894. Germany at the time was a much more liberal country, racially and politically, than the United States. Du Bois studied under Gustav Schmoller, who was affiliated with the socialist-leaning German school of economics, which taught its students to view economics not as a set of universal laws, but as a reflection of cultural and ethical attitudes. This emphasized inductive reasoning instead of a deductive approach employing grand theories (Morris, 2015: 19–22). In contrast, at the time Gramsci entered the University of Turin in 1911, Italian social sciences were in the grip of positivism. In criminology, Cesare Lombroso and his followers claimed that criminality was an inherited trait linked to discernible physical features. This was used to cast Southern Italians as racially inferior and explain away the poverty in the south as a natural, unavoidable phenomenon (Gramsci, 1990: 444; 1992: 116, 143–4). In economics, too, Achille Loria expounded a form of vulgar deterministic materialism criticized by Engels in the third volume of Capital (Gramsci, 1992: 114–16, 390). Gramsci encountered this Lorian positivism on the Left mainly in Nikolai Bukharin's Historical Materialism: A System of Sociology, which represented an influential current in Marxist thought and politics at the time (Gramsci, 1971: 378–80, 457). Bukharin aimed to provide a sociology of Marxist historiography that followed Darwinist positivism and examined society based on the model of the natural sciences – in other words, one that formulated universal laws that would enable the prediction of historical developments (Gramsci, 1971: 425–7; 1992: 53–7). The positivism expressed by Lombroso, Loria, Bukharin, and their followers formed the central scientific current that Gramsci opposed throughout his life: in his studies, in his pre-prison writings, and throughout his Prison Notebooks.

In turn, Gramsci and Du Bois's own works would come to shape the social sciences. Paradoxically, while Gramsci is more often cited than known, Du Bois's methodological contribution is generally better understood than cited. As Aldon Morris's 2015 book The Scholar Denied argues, Du Bois introduced modern scientific sociology to America, using new quantitative and statistical research methods that influenced the practice of a new generation of sociologists. This contribution, however, went unacknowledged because of the racism and political rivalries that characterized American sociology (Morris, 2015: 15–99, 168–94). On the contrary, Gramsci is ‘an author more often cited than truly known’ (Foucault, quoted by Buttigieg in Gramsci, 1992: xix). Indeed, people were already discussing his Prison Notebooks all around the world before they were available in their entirety, let alone translated (see Saccarelli, 2008). Gramsci had an impact on the politics of political parties, as in the case of Eurocommunism (Bobbio, 1979; Buci-Glucksmann, 1980), but also on approaches to the social sciences. In particular, Ranajit Guha and other Indian intellectuals formed the discipline of subaltern studies based on Gramsci's work. Subaltern studies scholars, like Du Bois, aimed to give a voice to the submerged subalterns, but their methodology emphasized more interpretative and qualitative approaches (Guha and Spivak, 1988: 35–88). As a result, despite their common aim, Gramsci and Du Bois came to inspire two very different methodological approaches to the social sciences, representing different concerns regarding the role of science in this socialist emancipatory project. Understanding their views and concerns – how they differ, but also precisely how they may overlap – can help us come to a better understanding of the role and limitations of social scientific studies in projects of social change. To do so, I start by discussing Gramsci's conception of science.

External reality of the world

Gramsci offers two related definitions and interpretations of science. The first of these conceives of science as the study of phenomena: their regularity, coexistence, and succession or causal relation. Another approach focuses more on the way in which science orders phenomena in order to better comprehend and ultimately master them through thought and action; it defines science as ‘the most economic description of reality’. But Gramsci points out that the most important question is ‘whether science can give us, and if so in what way, the “certainty” of the objective existence of so-called external reality’ (Gramsci, 1995: 290–2; see also Gramsci, 1996: 189–90). I argue in what follows that Gramsci thinks that science cannot and should not try to answer this question. Nonetheless, he still believes that scientific research can fulfill its purpose, or at least help in the aim to understand and shape the world.

Gramsci argues that our belief of the objective reality of the external world originates in religion (Gramsci, 1971: 441, 446–8; 1995: 291–2; 1996: 189–90). In his reflection on reality and objectivity, his main target is often the conception of reality as external, ‘made and cataloged, defined once and for all’ (Gramsci, 2007: 365). Gramsci has several criticisms against this view. First, he rejects this idea as absurd, since humans know reality only from their perspective, which means that reality cannot be imagined outside humanity. This is the criticism he voices against Russell's claim that we cannot imagine London and Glasgow without humanity, but that we can imagine relations such as north and south without imagining human beings. Gramsci criticizes this, claiming that it is not possible to think of ‘thinking’ without the existence of humanity.

While this specific criticism of Russell may be somewhat disingenuous, it contains two crucial substantive points of Gramsci’s thought: the rejection of transcendence and the reality of the superstructure. ³ It is because of these two arguments that Gramsci turns towards subjectivism. He argues that despite its flaws, subjectivism is able to overcome ideas of transcendence, beliefs that rest on hidden gods that are at their origin religious and take the world to be already made and finished (Gramsci, 2007: 365). However, subjectivism is itself superseded by the philosophy of praxis and historical materialism through the concept of the superstructure, which gives a historicist interpretation of subjectivism (Gramsci, 1971: 442, 444). Gramsci grants that speculative subjectivism can take the most absurd forms; he discusses the example of Tolstoy, who as a child thought that by spinning fast enough he would catch his mind lagging and thus pause reality. Another example of absurdist subjectivism is the idea that by opening the newspaper, one creates the news (Gramsci, 1971: 443–4; 2007: 367–8). Gramsci implicitly rejects this view in notes discussing historical facts, criticizing the Ptolemaic astronomic system, and so forth, but also more explicitly when he discusses early research on quantum mechanics and writes,

If it were true that the infinitely small phenomena in question cannot be considered as existing independently of the subject who observes them, they would in fact not even be ‘observed’, but ‘created’ and would fall into the same domain as the pure imaginative intuition of the individual. (Gramsci, 1995: 289)

But Gramsci is not satisfied to rest at such easy criticism of subjectivism; instead, he inquires into its possible value and relation to Marxism. He writes that the philosophy of praxis, Gramsci's philosophical conception of Marxism, turns a subjective conception of reality upside down, explaining a subjective conception of reality as a real historical fact that leads to action and interaction with the world (Gramsci, 1995: 347–8). It is from this point of view that we need to consider Gramsci's reflection on Russell, in which Gramsci argues that east and west, despite being historical-cultural constructions, are nonetheless real, and allow people to travel and ‘foresee’ certain outcomes (Gramsci, 1971: 447–8; 2007: 175–6).

Gramsci's critique here is thus not an attempt to deny the mind and language independence of reality, but precisely the opposite. His argument is that given social relations, beliefs, concepts, and the superstructure can guide or limit behavior. Gramsci is thus not denying the property of obstance, the notion that reality resists our wishes (Williams, 2002: 136–40). Instead, I take him to be making the claim that historical-cultural concepts denoting relations such as east and west are real in ways similar to physical reality and, like nature, can be modified only through practice and action – even if those acts are purely social, as in the case of discussion or attempts to convince others, or sometimes mixed, because they require the development of new tools, concepts, and, crucially, structures.

In contrast, the idea of a given external reality, whether a god, the spirit of history, or the economic structure of society, treats humanity and, in specific, the subalterns as mere things lacking agency. But Gramsci writes of the subaltern, ‘But was he ever mere “resistance”, mere “thing”, mere “nonresponsibility”? Certainly not. That is why the ineptitude and futility of mechanical determinism, of passive and smug fatalism must be exposed at all times, without waiting for the subaltern to become leader and take charge’ (Gramsci, 2007: 353). This is especially problematic in the social sciences, where scientific claims can be self-reinforcing. It creates what Ian Hacking calls looping, in that interactive kinds react to categorization in ways that reinforce given but contingent beliefs on them; in other words, it creates a self-fulfilling prophecy (Hacking, 1999, 59, 104–5; Haslanger, 2012: 465–7). ⁴ This is the case with sociology, which Gramsci defines as an attempt to create a method of historical and political science based on evolutionist positivism and the model of natural science. ⁵ It aims to derive experimentally general laws of the evolution of human societies. But, according to Gramsci, those causal laws sociology aims to uncover cannot have any causal significance, since they are nothing more than the duplication of the facts its practitioners have observed and can be valid only as long as nothing changes (Gramsci, 1971: 426–7, 430; 1996: 166;). Gramsci thus argues that statistical tools can predict politics only under the condition that the masses remain essentially passive; the predictive power of statistics requires that the independent variable remain stable. But, according to him, political actions are precisely what tends to move the masses out of passivity and thus destroy the law of large numbers (Gramsci, 1971: 429). Sociology can predict the future only to the extent that it creates a false fatalistic belief about the unchangeability of the world, and thus can predict the future only by manufacturing passivity.

In this argument, Gramsci’s critique of statistics and sociology is ultimately not so much about the methods as about the aim that they are deployed for, namely, the formulation of so-called eternal laws. He thus writes,

Reality is teeming with the most bizarre coincidences, and it is the theoretician's task to find in this bizarreness new evidence for his theory, to ‘translate’ the elements of historical life into theoretical language, but not vice versa, making reality conform to an abstract scheme. Reality will never conform to an abstract scheme, and therefore this conception is nothing but an expression of passivity. (Gramsci, 1996: 52)

Science can to some extent help us understand, which allows us to take concrete action and not simply follow an abstract scheme. For Gramsci, then, the role of science is not to answer broad questions on topics such as the objective existence of reality, but instead to provide the tools to assess the malleability of reality. He thus writes, ‘I believe it is a mistake to demand that science as such prove the objectivity of the real; that is a conception of the world, a philosophy, not a scientific datum’ (Gramsci, 1996: 189). Instead, he follows Vico in dividing the true and the certain: while the true is the domain of philosophy and aims to establish universal truth, the certain is the domain of philology that traces the empirical phenomenon of the world as a product of human choices (Vico, Bergin, and Fisch, 1948: 56). While Vico focused on both, Gramsci places the emphasis on the certain, arguing that the philosophy of praxis understands the idea of structure as something certain, which means the empirical social relations that are tied to the superstructure and the historical subjectivity of a social group (Gramsci, 1971: 346–8). But note that even that depends on the true – in other words, on a philosophy or worldview that yields a conception of history that is itself ingrained in a certain historical context. So even the philosophy of praxis provides contextual and historical truths that are not eternal but could be, would be, and aim to be abolished in the future, and its recognition of this differentiates it from idealism. This is also where the limits of the philosophy of praxis to represent the future come from: it depends on the existing class struggle and thus has tools to understand the world that are themselves limited to it (Gramsci, 1971: 404–7; 1996: 188–9, 194–6). This means that even when it engages in the study of the certain, on the question of what is and what has been, science encounters limits.

According to Gramsci, science should aim to find what is fundamentally present in some sensation in contrast to what is arbitrary – in other words, what all humans can see and feel in the same manner, provided that they adhere to the scientific conditions of verification (Gramsci, 1996: 189). ‘Experimental science is the terrain on which this objectivization has reached its maximum realization; it is the cultural element that has contributed the most to unifying humanity. It is the most objectivized and concretely universalized subjectivity’ (Gramsci, 2007: 337–8; see also Gramsci, 1971: 446). But even this aspect is conditional, because scientific truths are themselves never conclusive or unchallengeable but constant in development. This does not imply an inaccessibility of knowledge as a form of metaphysical ‘unknowable’, but instead presents it as a form of ‘not knowledge’ depending on the development of physical and theoretical tools (Gramsci, 1995: 292; 1996: 190). Moreover, the contextual conclusions that science can draw also depend on hypotheses and conceptual classifications that are both dependent on a certain ideology or conception of the world (Gramsci, 1971: 461–2; 1995: 283–4, 293); therefore, science always appears in the trappings of an ideology or worldview. Nonetheless, because of the process of abstraction inherent in the scientific methodology, it is easier to separate the facts from the ideological system of hypothesis, which allows social groups to appropriate the science of other groups without endorsing their ideology (Gramsci, 1995: 293).

Based on these considerations, Gramsci formulates tentative criteria for the formulation of predictions across scientific traditions. This involves having a strong grasp of the concepts and history of a certain problem, avoiding arbitrary changes in methods, acknowledging gaps in existing knowledge, acknowledging the limits of one's conclusions, and ultimately being cautious in one's assertions (Gramsci, 1971: 439; 2007: 350–1). Therefore, despite its uncertainty, science can provide limited predictions even in the social realm. Those conclusions can be drawn based on certain regularities and repetitions, but must take into account the determinate social relations and the specific conditions in which those regularities occur in order to avoid making claims about eternal truths (Gramsci, 1971: 410–14; 2007: 308–9). Social scientists can thus make predictive statements like ‘Observing that … then …’, which denotes some knowledge of the limitations of the statement, namely, knowledge of the limitations of the tools, material and theoretical, through which the observation is made. Secondly, this statement acknowledges that it is not about a static position but a dynamic process; and thirdly, it denotes not simply reflection on observations in theoretical terms but the possibility of modification. ⁶ In this way, Gramsci is able to appreciate the relevance and importance of the observation of regularities, without turning them into eternal truths unaffected by humans and their interactions with the world.

Gramsci thus sees some value in the sciences, and particularly the social sciences, in the struggle for hegemony. While he conceptualizes science as providing context-dependent facts, he does not endorse a form of relativism in which the validity of such facts is purely contextual; in other words, it can be used to provide critique beyond its class origins. Gramsci allocates a role for a kind of academic struggle, arguing that the philosophy of praxis should not engage only with the popularized version of the hegemonic worldview, but instead attack its strongest theoretical exponents: the philosophical, scientific, and academic experts of the ruling worldview (Gramsci, 1971: 440). This means that despite incompatible ideological sets of hypotheses, scientific approaches are not incommensurable, and that it is both necessary and possible for the subalterns to challenge the ruling worldview even in institutions marked by that worldview. Furthermore, science is strategically valuable for the subalterns. Even though Gramsci emphasizes the malleability of the social and natural world, and criticizes some approaches for presenting it as unchangeable, he thinks science can provide useful information about the world that can help modify it. While he claims that one can ‘foresee’ only to the extent that one can act (Gramsci, 1971: 438), he acknowledges that in order to act it is crucial to understand the world. The aim, then, is to understand the conditions of the status quo; this allows us to see and understand precisely what needs to be modified in order to create the preconditions for change (ibid.: 157–8, 170–1).

Epistemological moves

While Du Bois came from a different scientific position than Gramsci, he had a critical positivist approach toward science in general, and sociology in particular, and his work shows many of the same concerns and sensibilities as Gramsci’s does. As Du Bois himself acknowledged, his attitude to science and its role in social uplift changed several times during his life (Du Bois, 1990[1944]: 44–51; 1995[1940]: 4–7). Nowhere does he put it as explicitly in terms of the philosophy of science as in his 1944 essay My Evolving Program for Negro Freedom. There he describes his move away from a positivistic approach, with the ultimate aim of emulating the natural sciences to determine general laws governing human society, toward a Jamesian pragmatism and radical empiricism (Du Bois, 1990[1944]: 47–8; Gooding-Williams, 2020; Taylor, 2021: 255–6). He also records a second change, which involved a turn toward Marxist analysis and certain aspects of Freud's psychoanalysis. Du Bois links this change to his visit to the Soviet Union in 1928 (Du Bois, 1990[1944]: 49–51), but even in a 1933 essay, he asks whether Marx and Capital’s analysis of the world still have any relevance at a time characterized by the problems of the color line and imperialism (Du Bois, 1995[1933]: 543–4). Therefore, despite Du Bois's assertions, scholars have still questioned whether and when he turned toward a Marxist analysis (Aptheker, 1971: 259–60). Nonetheless, in his approach to science, I would argue that Du Bois experienced an epistemological move rather than a break, retaining key assumptions from his previous position in his later work. ⁷

In ‘A Program for a Sociological Society’, one of Du Bois's earliest essays on the methods and role of sociology (ca.1897), he separates the direct aim of sociology and science – namely, the discovery of truth – from the mediate aim: social reform. ⁸ He states it even clearer in his essay ‘The Study of the Negro Problems’, where he writes,

Students must be careful to insist that science as such – be it physics, chemistry, psychology, or sociology – has but one simple aim: the discovery of truth.… Any attempt to give it a double aim, to make social reform the immediate instead of the mediate object of a search for truth, will inevitably tend to defeat both objects. (Du Bois, 1898: 16)

In ‘Sociology Hesitant’, one of the first methodological texts he wrote after his epistemological change, Du Bois challenges positivistic sociology. In the essay, he attacks the understanding of sociology as a science that studies an abstract entity called ‘Society’ rather than human action. If this were accurate, society would function as a mystical whole following logical laws (Du Bois, 2000[1905]: 38–9). The reason why, according to Du Bois, early sociologists such as Comte and Spencer hesitated to make human action the focus of their study is that they wanted to make sociology into a science that followed general laws. But human action, while sometimes having a certain rhythm, is fundamentally characterized by incalculability and chance, or, as Du Bois puts it, ‘self-directing Wills’ (ibid.: 40–2). This does not so much break with his previous views as it deepens the skepticism and doubts he expresses in earlier works. In these, he asks whether sociology is a science like the natural sciences, noting that it does not have general laws. He does notice regularities that suggest the presence of general laws, but stresses that these can be discerned only with great difficulty. ⁹ In contrast, his interests lie in studying those regularities and their contextualities, rather than in formulating ever-present laws. In other words, while he affirms the possibility of the existence of these general laws, he does not treat them as a fundamental feature of useful sociology. His interest, detectable even in his earlier work, is more in the interaction and particularity of context and history and their effect on the present situation of groups, in particular African Americans, than in the discovery of eternal truths. The University of Philadelphia’s aim in hiring Du Bois to study the Seventh Ward was to provide statistics to support the prejudicial belief that everything that went wrong in Philadelphia was caused by its African American community. Du Bois recalls in Dusk of Dawn: ‘Of this theory back of the plan, I neither knew nor cared’ (Du Bois, 1995[1940]: 58). While he may have known, he certainly did not care; instead of using statistics to support prejudices, Du Bois studied crime and poverty as symptoms driven by a context that went beyond the Seventh Ward. Furthermore, he did not treat the inhabitants of the ward as a homogenous group, but recorded the diversity of their income, work, and activities. His works were thus instrumental in rejecting an essentialist homogenization of Philadelphia's Black community, and at the same time provided guidelines for reform (Du Bois, 2007[1899]: 1–4, 268–75). Even though Du Bois intended his findings to be helpful for reform, his immediate aim was to uncover truths. While different findings might have been more agreeable to reformists, in that they could have presented reform as an easily implemented panacea, Du Bois argues that his findings illustrate the difficulty of change, but that only if this difficulty is acknowledged can reform efforts succeed (ibid.: 268).

Du Bois thus sees science as a way to uncover certain facts of life, but over time he came to revise his understanding of what constitutes facts. He writes, ‘Facts, in social science, I realized, were elusive things: emotions, loves, hates, were facts; and they were facts in the souls and minds of the scientific student, as well as in the persons studied’ (Du Bois, 1990[1944]: 47). This statement could be read as problematizing the idea of value-free science or as emphasizing the reality of beliefs and emotions as facts that move people. These are valid interpretations that are discussed further below. However, the main point that I take him to be making is that a purposefully non-emotional description of history would also be nonfactual. ¹⁰ Black Reconstruction's final chapter, ‘The Propaganda of History’, exemplifies this view. There, Du Bois criticizes the way the history of reconstruction was done in his time. He presents a choice: history can aim towards truth, which means that it must have standards of accuracy in its research and interpretation; or it can be used to inflate one's ego, in which case one must give up on the idea of history as a science (Du Bois, 1998[1935]: 714). Treating history as a science does not entail impartiality or neutrality in the description of moral wrongs. On the contrary, Du Bois berates his contemporaries for discussing American slavery as if it were merely the workings of a cosmic social and economic order for which no one could be blamed (ibid.). To take out of the equation the will, and thus the wrongdoing, of people is thus not a commitment to truth but rather the erasure of facts, which, no matter how unpleasant they are, science must not stoop to.

This conception of guilt in history still affirms a mind- and language-independent conception of causality, in that the actions and choices of certain men are the cause of the suffering of many others (Aptheker, 1971: 256–8). Du Bois thus sees causality as a sign of human agency instead of a deterministic hindrance. Like Gramsci, Du Bois does not reject causal claims, but understands them as an example of the malleability of the world. In a coauthored paper, Du Bois discusses and criticizes the work of Pitirim Sorokin in ways that exemplify this point. ¹¹ The article begins with a discussion of the crisis link between mind and matter, which Sorokin claims to solve by prioritizing mind over matter in an idealist manner. The criticism Du Bois and his coauthor make against Sorokin is, first, that his theory is static and does not account for change, and can therefore account only for continuity. This is the result of sociology’s attempt to be a natural science, namely, that it aims to hold fast to general laws in a way that does not allow for change, especially change caused by human will (Coulborn and Du Bois, 1942: 511–12). However, the authors recognize that Sorokin does try to offer an explanation for change, which takes the form of a transcendental deity called charisma, manifest in a few exceptional individuals (ibid.: 505–6, 517). But Du Bois and Coulborn reject this theory, arguing that what transforms societies is not abstract universal laws but human beings (ibid.: 511–13, 515). They argue that the contributions of those few great names we recall from history depended on the work of a vast number of people who participated in and enabled the transformation of their ideas. They write,

To Colbert and Adam Smith could be added a score of well-known intermediary persons, and to that score could be added hundreds of thousands of persons whose names were not recorded in history but whose thoughts and acts accumulated to make the thoughts and acts of Colbert and Adam Smith possible; and all together they were the creators of systems, pure and empirical. (ibid.: 506)

That emotions are facts reflects a broader recognition by Du Bois of the factuality of social construct, or, in Gramscian terms, the reality of the superstructure. But while Gramsci discusses concepts such as east and west, Du Bois targets the concept of race. Already in The Conservation of Races, he defines race not in biological but in historical and cultural terms, in the same way as he understands nations or civilizations. He develops there the creed ‘We believe that the Negro people, as a race, have a contribution to make to civilization and humanity, which no other race can make’ (Du Bois, 1897: 825). He would reflect on this creed many years later in a chapter of Dusk of Dawn titled ‘The Concept of Race’, where he asks, ‘What is it between us that constitutes a tie which I can feel better than I can explain?’ (Du Bois, 1995[1940]: 116). The chapter starts with a chronology of the concepts of race Du Bois has hitherto encountered in his life, constantly debunked, updated, and moved around, leading him to write, ‘The first thing which brought me to my senses in all this racial discussion was the continuous change in the proofs and arguments advanced’ (ibid.: 99). While there is thus no scientific definition of race, and the impossibility of such a definition is easily proved, he nonetheless emphasizes the reality of race and the bond it creates, not as a natural fact, but as a social one. He writes, ‘The badge of color is relatively unimportant save as a badge’ (ibid.: 117). This badge is reified in legislation:

Thus it is easy to see that scientific definition of race is impossible.… All this has nothing to do with the plain fact that throughout the world today organized groups of men by monopoly and physical power, legal enactment and intellectual training are limiting with determination and unflagging zeal the development of other groups; and that the concentration particularly of economic power today puts the majority of mankind into a slavery to the rest. (ibid.: 137–8)

Translation of reality

So far we have seen that despite representing different approaches to the authority of scientific results, Gramsci and Du Bois share similar concerns about and hopes for the social sciences. Both reject the use of sociology as an attempt to discover eternal societal truths. They warn against this approach, citing its potential to overlook human agency; to portray history and societal change as a mechanistic process, the actions of great men or a hidden god; and to forget the real actions of a multitude of people that concretely enact and enable change. Besides criticizing this approach as a mischaracterization of history and the workings of society, they also argue that it creates self-fulfilling prophecies by fostering passivity. On the other hand, they also see value in social science research as a way to warn against and resist wishful thinking. Such thinking can arise among both progressives and reactionaries. With the former, wishful thinking risks underestimating the difficulty of reform and the resilience of the status quo. With the latter, science becomes subordinated to a process of inflating the national ego and rationalizing prejudices.

In the remainder of this article, I place the works of Gramsci and Du Bois in dialogue with each other to conceptualize the uses and limitations of the social sciences for movements aimed at social change. In particular, I address the question of whether progressive movements can understand and use science against wishful thinking while keeping in mind the fallibility and normative grounding of scientific research. Secondly, I discuss how scientific research can be used to overcome reactionary wishful thinking and pierce through rationalizations and ingrained beliefs. On the first point, I argue that Gramsci can help overcome the limitations of Du Bois's value-free conception of science. On the second question, Du Bois can help us overcome the optimism of Gramsci that the widespread and persistent distribution of scientific findings is sufficient to change people's ingrained beliefs.

Du Bois, in his earlier work on the Talented Tenth, assigns a leadership role to scholars. In Dusk of Dawn, after his epistemic transformation, he still sees a role for academic research in social movements, but one that is advisory (Du Bois, 1995[1940]: 320). Du Bois no longer conceptualizes the Talented Tenth as ‘leaders of thought and missionaries of culture among their people’ (Du Bois, 1903: 861). Instead, he recognizes in Dusk of Dawn that a new philosophy, the basis of a new worldview, must come from the masses – from their needs and grievances, but most importantly from their abilities (Du Bois, 1995[1940]: 217–19). As a result, technical, scientific, and academic expertise now have a supporting rather than a leading role. But even in this auxiliary role, according to Du Bois, it is important for the scientist to aim for and endorse a value-free ideal of science. This is because, as Liam Kofi Bright has argued, scientists should not put themselves in the shoes of the voters by withholding information the latter may find useful in making a choice (Bright, 2018: 2239–41).

In this supportive role, science is not to try and confirm what a particular movement wishes were the case. I argue that Du Bois bases his support for non-epistemic value-free science on a strong prudential concern against wishful thinking. His concern is legitimate and should be taken seriously; therefore, it is important to be precise about what he does or does not do. Nonetheless, I argue that his ideal of a value-free science underestimates the link between findings and value-laden theory, in that this emphasis on value-free science does not adequately combat the problem of wishful thinking. However, Gramsci's conception of the translation of elements of historical life to theoretical language may offer an alternative that takes seriously the problem of wishful thinking.

Du Bois thinks that scientists guided by values other than a commitment to truth will provide less accurate and reliable findings that are more susceptible to wishful thinking. His position is thus that the ideal of truth-seeking will lead to the acquisition of true beliefs. There are two criticisms of this viewpoint, which differ slightly but are often taken together. One concerns the influence of prior beliefs in reporting facts; the other relates to the very possibility of differentiating theoretical and value-laden aspects from objective aspects of science. I argue that Du Bois avoids the first pitfall by falling into the second; in other words, he is able to claim that value-free science can spread truths despite the possible existence of biases because he thinks that, at least in theory, it is possible to separate prior beliefs, theories, and values relating to a subject from the discovery of evidence on that subject. In a 2017 article, Bright constructs a game-theoretical model to illustrate how pure truth-seeking scientists, scientists motivated solely by a desire to transmit true beliefs to an audience, may in fact be more likely to commit fraud by modifying or withholding information acquired through their research. Bright provides a concrete example from botany in which scientists feared that they would misinform their readers by drawing certain plants realistically, and therefore omitted features of those plants that were in fact typical, but which they had mistaken for isolated aberrations (Bright, 2017: 12). While this example is telling, it fails to capture Du Bois's ideal of value-free science because it assigns prior beliefs to the scientist about their object of study and their community. In other words, it concerns not a purely truth-seeking ideal but a truth-spreading ideal, two things Du Bois kept distinct.

In ‘A Program for a Sociological Society’, Du Bois describes an unfortunate encounter he had while gathering data. He recalls asking a woman for some information, whereupon the woman asked his reason for seeking that knowledge, and whether it would be used for the purposes of reform or change. When Du Bois insisted that his aim was solely to collect facts for their own sake, the lady reacted disapprovingly and refused to give him the information. Du Bois does not use this anecdote to offer a justification for subordinating the quest for truth to the goals of reform, or for the pragmatic use of a noble lie in order to gather information. Instead, he deplores that ‘men are so filled with ideas of reform that they haven’t time to study the abuses which they seek to abolish and baldly assume that they know all about them’. ¹² The issue for Du Bois, to use Bright’s terms, was thus not men’s desire to seek the truth, but their prior assumptions about the truth.

This opens Du Bois up to another line of criticism, namely, the idea that it is impossible to differentiate the gathering of facts from theoretical assumptions because of what Helen Longino calls underdetermination: the fact that ‘no amount of empirical data can uniquely determine theory choice’ (Longino, 1996: 39). Therefore, the attribution of meaning to empirical data must necessarily be based on criteria external to that data, criteria that will inevitably reflect contextual values and assumptions (Anderson, 1995: 29; Longino, 2001: 99). Furthermore, this does not only pertain to the interpretation of results, as latent values are already present at the data gathering and classification stage (Anderson, 1995: 49–52). This reflects Gramsci's concern that

an enquiry into a series of facts to discover the relations between them presupposes a ‘concept’ that permits one to distinguish that series from other possible series of facts. How can there take place a choice of facts to be adduced as proof of the truth of one's own assumption if one does not have a pre-existing criterion of choice? (Gramsci, 1971: 461–2)

Underdetermination does not deny the importance of the empirical adequacy of the hypothesis. One aspect that Longino emphasizes to warrant the quality of research is community feedback. While this relies on the assumption that being a member of a community does not create a shared bias, it reflects certain aspects of Gramsci's understanding of objectivity as intersubjectively verifiable. More specifically, it requires that, given similar theoretical and practical tools, different individuals can observe the same phenomenon (Gramsci, 1971: 446; 1995: 286–90; 2007: 337–8; Longino, 2001: 99–100). Nevertheless, it might be objected that it is problematic to choose one theory from two that fit the data equally well simply because it is politically preferable. However, this objection misses the point of Longino's argument. As Elizabeth Anderson argues, the key aspect is that theoretical inquiries aim to achieve more than the mere accumulation of trivial truths, in that the accumulation of all possible facts, if such were even possible, would bury any significant truths under a mass of irrelevant and trivial details. This means that theory necessarily selects and organizes facts in a way that is conducive to answering a question, which means that even though a theory may be empirically adequate, it can still yield biased conclusions. In other words, not all true statements about a phenomenon constitute an acceptable theory of it (Anderson, 1995: 37–9, 41).

Similarly, in the case of classifications, we cannot simply rely on a value-neutral standard for the existence of a genuine causal relationship between all members of a group, as there are many ways in which genuine causal relations might manifest that could translate into different classificatory systems. But this does not mean that every classificatory system is equally valid. On the contrary, a valid system must meet certain epistemic standards – for example, the inclusion of clear empirical criteria that can determine whether a phenomenon falls under a particular category (Anderson, 1995: 46–7). Finally, while contextual values such as the theoretical virtues of heterogeneity and complexity can help us choose between two equally empirically adequate theories, the theories need to remain empirically adequate, and should thus be rejected if they appear to yield false conclusions. For example, Anderson argues that Chomsky's cognitive psychology displays an advantage over Skinner's behaviorism from the point of view of heterogeneity because it allows for the possibility of humans as free and creative beings, but the preference for such a theory depends on whether humans do indeed have something that can meaningfully be called agency (ibid.: 50–1). Furthermore, in some cases a classificatory system, even though it is imperfect with regard to these contextual values, may be the most adequate for acquiring certain truths about the world; for example, quantitative research may justifiably sacrifice some complexity to address questions that cannot be answered otherwise (ibid.: 52). To endorse the role of contextual values in science is not to endorse arbitrariness; it does not mean that all theories should be considered valid irrespective of their empirical soundness or based solely on normative criteria or what is politically preferable. On the contrary, it means that choices of classificatory systems and theories need to be accounted for and defended together with their empirical adequacy.

While Du Bois's concern about wishful thinking is warranted, we do not need to endorse a value-free scientific ideal to avoid falling into this kind of thinking. Instead, Gramsci's view of science as the translation of empirical facts and historical elements into theory seems more useful. While this acknowledges that reality will not bend to abstract theoretical schemes, and thus that wishful thinking is pointless at best, this view nonetheless acknowledges that reality needs to be translated into a theoretical language that is open to different interpretations and requires additional work to defend or reject those interpretations (Gramsci, 1996: 48–52). Gramsci's specific use of the word translation in his note on science is telling, because for him translation is not simply the extraction of meaning from one context to another, but entails the critical assessment of a concept in relation to other elements in their cultural context. In a letter to his wife, Gramsci writes, ‘A qualified translator should be able not only to translate literally but also to translate the conceptual terms of a specific national culture into the terms of another national culture, that is, such a translator should have critical knowledge of the two civilizations’ (Gramsci, 1994b: 207).

This concept of translation does impose limitations on the translator; the task is not to write a new version of the original text according to the interests of the translator. Nonetheless, it recognizes that both the source and the target language are embedded in a cultural context that is relevant to the meaning of the translation. Translation, then, requires a critical approach that does not stop at the dictionary definition of a term but inquires further into the implicit and emotional relevance of the term in different contexts (Gramsci, 1996: 69–70; 2012: 268–9). The difference in context, and how it affects the connotation of a term, is therefore productive of meaning, and allows us to better understand both the source and the target language by providing insights into their sociohistorical contexts (see Ives, 2004; Thomas, 2020). This understanding of science as translation, in contrast to the value-free conception, does not simply aim to abstract an observed phenomenon from normative assessment. Instead, it emphasizes how the phenomenon, be it a historical period or a metal in solid form, exists in relation to other elements, and that these relations are expanded or limited by their observation, by the research question being pursued, and by the categorization scheme employed. The language in which the phenomenon is explained, similarly, is embedded in a broader theoretical and social context that allows the world to be understandable. Therefore, a good scientist, like a good translator, should not try to abstract from these contexts but, like a good translator, should assess them critically in a way that involves a theoretical and normative justification.

Truthful propaganda

The use of science to achieve social change cannot rest, as Du Bois correctly points out, simply on a desire to enact reform, but requires the gathering of facts and information. However, it also requires serious theoretical and philosophical work to interpret those facts and the concept in which they are categorized. Therefore, for the subalterns, scientific research, particularly the social sciences and historical research, can help counteract wishful thinking without necessarily reinforcing the status quo. But the hope expressed in Du Bois’s early writings about the potential of scientific research to promote social reform goes further than this; namely, he argues that racism is the product of ignorance, which can be dispelled by truth and scientific facts (Du Bois, 1995[1940]: 4–5). Later in life, he became disillusioned by this approach to the problem of racism. His later work discusses how the campaign by the National Association for the Advancement of Colored People against racial discrimination was fatally limited by its own assumptions, namely, ‘on the one hand that most race prejudice is a matter of ignorance to be cured by information; and on the other hand that much discrimination is a matter of deliberate deviltry and unwillingness to be just’ (ibid.: 194). While Du Bois acknowledges that ignorance and malevolence certainly exist and are problems, he believes that inherited customs and subconscious habits are at the core of racist attitudes. These cannot be changed by sudden assault but require, according to Du Bois, a persistent campaign of well-planned and scientific propaganda (ibid.: 6, 172, 194).

There are two crucial elements I want to investigate here: first, the ‘persistent’ aspect, and second, the notion of ‘scientific propaganda’. These two elements are also reflected, to a certain extent, in Gramsci’s work. Like Du Bois, Gramsci rejects what he calls the rationalistic viewpoint: ‘the conviction that certain phenomena are destroyed as soon as they are “realistically” explained as if they were popular superstitions (which anyway are not destroyed either merely by being explained)’ (Gramsci, 1971: 235). However, Gramsci focuses more on persistence than on propaganda, even in the peculiar sense that Du Bois gives the latter, as discussed below. Gramsci sees culture not as a means of control but more as the expression of a lived experience. He argues, for example, that Italians were not reading novels by Italian authors not because they were uninterested in literature, but because those authors were failing to capture the lived experiences of the Italian masses (Gramsci, 2012: 206–12). He thus prioritizes a certain political and economic unity over cultural unity, instead of the other way around (see Lo Piparo, 1979). The potential for the propagandist use of science in Gramsci is therefore limited. While he does not think that people will give up their cherished beliefs when confronted with the facts, he nevertheless maintains that through repetition, and the widespread dissemination of new discoveries and information, people will over time change their beliefs. Already in his earlier writings and political activities, Gramsci is concerned with the question of how to bring information that requires prior knowledge to the masses, many of whom are illiterate. Gramsci argues for the publication of journals that would provide original scientific research but also basic knowledge about the concepts on which that research depends, in order to make such knowledge accessible without having to modify it (Gramsci, 1992: 173, 1996: 68; 2012: 422–3). Besides accessibility, repetition is crucial, according to Gramsci. He endorses a certain Jamesian psychology of habits, in which habits, through repetition, become second nature. The way to combat biased beliefs about the world is therefore to repeat the objections to those beliefs (Gramsci, 1971: 339–40; 1992: 128–9; James, 2010: 75–82).

Du Bois also endorses this element of repetition and continuity, but he confronts an additional problem, namely, popular unreceptiveness to facts. This means that the presentation of evidence, instead of setting in motion a gradual process whereby biased beliefs begin to disintegrate, can reinforce those beliefs through a process of rationalization. This brings us to our second element, that of propaganda, on which Dusk of Dawn seems almost to present a split personality. Already in the first pages of the book, Du Bois writes that the Black world must fight for freedom with truth and the organization of propaganda. In the very next sentence, however, he points out how this fight, and the progress achieved, are halted by propaganda, murder, and the horrors of war. In Chapter 6, ‘The White World’, he describes the train of thought of a hypothetical white friend, who reasons that ‘deliberate propaganda is necessary and permissible; propaganda assists the truth and hurries it on; it may at times exaggerate and distort but all this is for a defensible end’ (Du Bois, 1995[1940]: 158). While Du Bois points out the inconsistency of this position, he closes the chapter by emphasizing the need for well-planned and scientific propaganda to combat racist habits. Nonetheless, I argue that Du Bois does not give up on his non-epistemic value-free ideal of science; his argument instead relates to the presentation of scientific evidence that would, if shared solely as academic findings, be rejected or rationalized away. He argues that such information must be offered in a way that allows it to change the ingrained beliefs of the co-opted subalterns, the non-ruling groups that nonetheless have a stake in the status quo. To grasp this, we need to discuss what exactly Du Bois means by propaganda.

In Criteria of Negro Art, Du Bois states, ‘Thus all Art is propaganda and ever must be, despite the wailing of the purists.… I do not care a damn for any art that is not used for propaganda’ (Du Bois, 1986[1926]: 1000). This is not because art should be used for political aims alone, but because art is a method for gaining sympathy and human interest that is an expression of truth, the truth of the humanity of those treated inhumanely. In other words, ‘The apostle of Beauty thus becomes the apostle of Truth and Right not by choice but by inner and outer compulsion’ (ibid.). Melvin Rogers reads The Souls of Black Folk in this light, as a form of propaganda that through rhetoric opens the mind of the white reader to an experience, a truth, to which they would otherwise be closed (see Du Bois, 1994[1903]). To be specific, Rogers argues that Souls starts by evoking emotions of sympathy, only later in the book evoking feelings of shame that reveal a truth about the wrongs inflicted by whites against African Americans. However, if Souls had asserted these truths of shame and sympathy descriptively or separately, they would have led to the closing of the reader's mind (Rogers, 2012: 200–1). Rogers argues that this is not a form of manipulation because the latter involves a disconnect between a person’s endorsement of a belief and the reason why they endorse that belief. Someone who is manipulated would not recognize themselves in the belief that they have come to hold (ibid.: 196). However, this account involves counterfactuals that are difficult to ascertain; it is, after all, precisely the mark of a successful form of manipulation that it can make someone believe that they are the source of their belief. Instead, I take the difference between a manipulative noble lie and Du Bois's conception of truthful propaganda to be that the former fears the truth while Du Bois does not. Or, as he puts it in Criteria, ‘We can afford the Truth. White folk today cannot’ (Du Bois, 1986[1926]: 1001).

As Gooding-Williams argues, Rogers's argument focuses on rhetoric instead of the particular concept of beauty used in Criteria and developed earlier in Darkwater (Gooding-Williams, 2021: 20–3). Darkwater's concept of beauty dovetails with Du Bois's changing views on white supremacy, which he attributes to ill will rather than ignorance, and consequently is also in line with his view on the transformative potential of science (Du Bois, 1999[1920]: 17–29, 41; Gooding-Williams, 2021: 14–21). Beauty would then fill the vacuum of transformative potential left by science. However, Du Bois's affinity with rhetoric predates this epistemological change. That Du Bois already uses rhetoric purposefully to deliver a truth in Souls hints that, for him, the ideal of a value-free science may be compatible with a rhetorical form akin to a particular conception of propaganda. As Rogers points out, rhetoric, as the art of persuasion, was a central element of Du Bois's academic education (Rogers, 2012: 189, 194). Du Bois recalls in his last autobiography that his first thesis at Harvard failed because of poor grammar and lack of literary style. This made him realize that style and rhetoric were necessary for carrying his message further. Black Reconstruction is a telling case of how this realization continued to influence his academic writing throughout his life. This book was a major intervention in the field of history designed to counter the falsehoods propagated about the Reconstruction era. And yet, Du Bois spent over a year working on the literary style of Black Reconstruction, informing an increasingly impatient editor that the book was ‘not yet a piece of literature. It still resembled … a Ph.D. thesis’ (Du Bois, 1998[1935]: x–xi; see also Aptheker, 1971: 251). That Du Bois emphasized style did not hinder him from criticizing orthodox Civil War historians as propagandists in the pejorative sense. The difference is that whereas these historians used rhetoric to hide and omit truths, Du Bois left the facts of the Reconstruction era untouched and used style to make his audience more receptive to them (Du Bois, 1998[1935]: 714). Science thus never entirely lost its transformative potential for Du Bois. However, he also always knew that attention to the presentation of scientific findings, to their artistic and emotional form, was also necessary to foster his audience’s receptivity to the truth of their content.

Du Bois ends with a distinction between form and content that allows him to retain his conception of science as a value-free ideal. Style, rhetoric, and art are subordinated to politics, but politics itself is subordinated to truth. While Du Bois draws a thin line between his value-free ideal of science and the tendency of propaganda to undermine this value to other ends, there is nonetheless a line. Therefore, even if we reject the idea that science can be conceived as a non-epistemic value-free practice, this line can still be upheld between science and propaganda in the pejorative sense and help guard against the lure of wishful thinking.

Conclusion

This article opened with two contradictory concerns relating to science. On the one hand, there is the concern about granting science too much authority in a way that risks creating a deterministic and essentialist world that reinforces the status quo. On the other hand, there is the concern that if science has too little authority, arbitrary political interests, prejudices, and stereotypes cannot be opposed on any external grounds except for equally arbitrary political interests. While Gramsci and Du Bois, at first glance, might be considered to embody the two horns of this paradox, I argue that they offer valuable insights that allow us to address it. I argue that despite rejecting the notion of science of a non-epistemic value-free ideal, we can avoid wishful thinking and formulate an area of commensurability to criticize and evaluate different theoretical systems that fit the empirical evidence. Secondly, even though we can acknowledge that science should not and does not have enough authority alone to reject biased beliefs, it can be presented in ways that are affectively more effective, without resorting to a limitation of the spread of new information or the distortion of scientific research.

This, however, should not be seen as a blueprint offering a guaranteed path to navigating the tension between social aims and scientific practices. Kathryn Paige Harden’s recent work The Genetic Lottery: Why DNA Matters for Social Equality could, initially, be viewed as following the approach of Gramsci and Du Bois. Harden argues that we need to understand genetic differences in learning affinity in order to transform the education system and labor market to the inclusion of everyone, regardless of the morally arbitrary distribution of genetic variants (Harden, 2021: 20). In doing so, Harden, self-consciously, attempts to grasp what is not changeable to determine what can be changed, and to promote egalitarian values and social justice in language designed to have an intuitive emotional appeal across the political divide (Harden, 2021, 2022). However, she is not entirely able to escape the social language and history that shapes the tools she uses to study humans and how her claims are expressed. As a result, her conclusions may still rely on tools shaped by problematic conceptions of genetics and intellect that reiterate the idea of biology as an immutable force shaping individual social worth (see Coop and Przeworski, 2022a, 2022b; Roberts, 2011, 2015). Gramsci and Du Bois thus do not present a blueprint for overcoming the tension between the attribution of too much and too little authority to science. Instead, they show how the practice of science – the development of its tools but also the way it is expressed – is imbued in a social language and context that shapes its ability to capture relevant facts about the world and its potential for social progress. Ultimately, Gramsci and Du Bois do not give us a way out of the constant struggle to develop better tools, reassess the premises of seemingly neutral questions, and push, through social movements, for a more egalitarian society.