Wittgenstein, Religion and Deep Epistemic Injustice

Abstract

In his article ‘Epistemic Injustice and Religion’, Ian James Kidd raises the possibility that some epistemic injustices might be deep. To spell out exactly what might be involved in deep epistemic injustices, especially those involving religious worldviews, an obvious place to look is Wittgenstein’s work on religion. Careful reflection on Wittgenstein’s remarks in the ‘Lectures on Religious Belief’ and his late work collected in On Certainty will have implications for how we are to understand the relationships between belief and evidence and for the ways in which we might enrich our hermeneutical sensitivities, and so Wittgenstein’s remarks are helpful for understanding epistemic injustices more generally. This paper will focus on epistemic injustices involving Islamophobia since Islamophobia has, so far, been given little attention in the literature on epistemic injustice.

1. Religion and Epistemic Injustice

The literature on epistemic injustice since the time of Miranda Fricker’s book Epistemic Injustice has had relatively little to say about discrimination on the basis of religion, despite the fact that religion is an arena of identity power (See Fricker 2007, p. 14). As Ian James Kidd notes, “the literature…is modest and mostly represented by the work of the eminent feminist philosopher of religion, Pamela Sue Anderson”.1 In her book, Re-visioning Gender in Philosophy of Religion, Anderson’s focus is on the ways in which the Christian religion unjustly silences women and downgrades their credibility, as well as upon academics in the philosophy of religion whose discussions of Christianity treat women unjustly. For example, she points to passages in the Bible that depict women as being at the bottom of a patriarchal hierarchy (e.g., Corinthians 11:3 “…the head of every man is Christ; and then head of the woman is the man; and the head of Christ is God” and Corinthians 11:7 “For a man indeed ought not to cover his head, forasmuch as he is the image and glory of God: but the woman is the glory of the man”) and she discusses how sexist passages in the Bible have been used to suppress the voices of Christian women (Anderson [2012] 2017, p. 14). Both Anderson and Kidd have noted that the abstract ways in which theodicies are formulated within the philosophy of religion can lead to injustices (ibid., p. 41). For example, John Hick has argued that there is a deep problem of evil which is the upshot of evils being “distributed in random and meaningless ways” (Hick 2010, p. 333) but, as Kidd observes, if we look at the way that terrible events and suffering are distributed in reality, we see that it is “socially and materially conditioned”, such that “certain groups of people—women and the poor, say—suffer more intensely and extensively from evils…its distribution is not ‘random’”(Kidd 2017, p. 391).2 The abstraction away from the reality of suffering and its connection to social identities is a source of epistemic injustice because it disregards the testimonies of those who suffer and distorts our understanding of evil.

While Anderson has focused on the ways in which religion and philosophers of religion have perpetrated injustices I want to particularly focus on the ways in which followers of a religion might be treated unjustly as followers of the religion in question. My focus will be on Islamophobia, which is arguably being promoted by leading figures in the most populous and powerful nations on the planet,3 and yet discussions of epistemic injustice have rarely mentioned it. I will start by briefly clarifying the notion of Islamophobia in Section 2 before going on to look at how it is implicated in various forms of epistemic injustice in Section 3. I will then begin to explore how we might understand deep epistemic injustice by looking at ways to understand deep disagreement (in particular, deep religious disagreement) in Section 4. This discussion of deep disagreement will suggest that the standard picture of testimonial injustice, where a speaker presents a piece of information or an item of knowledge and the hearer makes a credibility judgment concerning what is said, will not apply in any straightforward way to cases where the disagreement between speaker and hearer is deep. In Section 5, I will contrast testimonial injustice with deep epistemic injustice and provide some examples of deep epistemic injustice.

2. What Is Islamophobia?

Before moving on to look at how Islamophobia is implicated in epistemic injustices, let us first of all get a bit clearer about what Islamophobia is. To this end, it is helpful to compare it with antisemitism. Islamophobia no more refers to a fear of a religion than antisemitism refers to opposition to (‘anti’) Semites. The social philosopher Brian Klug draws upon Wittgenstein in order to drive this point home. It is not the etymology of a word that determines its meaning but its use.4 The meaning of ‘antisemitism’ is not fixed by its etymology. It is used to refer to “hostile attitudes and acts directed at Jews whether those acts and attitudes are based in biological racism or not” (Klug 2014, p. 449), and its meaning has changed since the nineteenth century when the term was introduced (when it was tied much more closely to biological racism).5 Similarly, ‘Islamophobia’ refers to hostile attitudes and acts directed towards a racialized minority, Muslims, as Muslims. Neither etymology nor provenance tells us the meaning of ‘antisemitism’ or ‘Islamophobia’. It is their use that we must attend to, and their use has changed over time. Klug notes that Wittgenstein famously said that there was no one thing in common between the various things we call ‘games’,6 and he suggests that something similar might be said about the meaning of ‘antisemitism’ or the meaning of ‘Islamophobia’ as they change over time. There are “similarities overlapping and criss-crossing” between the uses of ‘antisemitism’ and ‘Islamophobia’ as the negative qualities that are ascribed to Jewish people or to Muslims over time by racists change.

3. Testimonial Injustice, Hermeneutical Injustice, and Epistemic Violence

Having clarified the meaning of ‘Islamophobia’, we can now examine how Islamophobia might be implicated in various forms of epistemic injustice. Miranda Fricker’s book Epistemic Injustice has generated a huge discussion of injustices done to people in their capacities as knowers.7 The central cases of epistemic injustice that she identified in that book were ones involving identity prejudices that track people through different aspects of their lives, i.e., prejudices like racism, sexism, homophobia, transphobia, and discrimination on the basis of religion. In these cases, people are not just suffering injustices done to them in their capacities as knowers (given deflated credibility or denied the hermeneutical resources that would help make sense of their oppression), but they are also discriminated against in the job market, in the legal system, in education, and are subject to violence and abuse from people holding the relevant prejudices. The central case of testimonial injustice is what Fricker calls systematic testimonial injustice (Fricker 2007, p. 27). We can see that epistemic injustices that are committed against people on the basis of Islamophobia will be of this nature. For example, there is evidence that Muslims in the United Kingdom face discrimination in employment,8 in the legal system and the prison system,9 and that they face far greater levels of violence and abuse in the UK than other religious communities.10

In Epistemic Injustice, Fricker distinguishes two main forms of epistemic injustice: testimonial injustice and hermeneutical injustice. In introducing these concepts at the beginning of her book, she says that testimonial injustice “occurs when prejudice causes a hearer to give a deflated level of credibility to a speaker’s word” (Fricker 2007, p. 1) and, as we have seen above, she regards the central case of testimonial injustice to be systematic testimonial injustice involving identity prejudices. In making credibility judgments of other people, Fricker thinks that we must avail ourselves of social generalizations or stereotypes in gauging how likely it is that what a person has said is true. This becomes problematic when identity prejudice is at work in the stereotype; that is, when the stereotype is unreliable and is “made or maintained without proper regard to the evidence” (ibid., pp. 31–34). If Muslims are discriminated against in the job market, the legal system, and in politics, then we can say that Muslims are victims of identity prejudice, as Fricker describes it, and a Muslim (or a group of Muslims) who suffers deflated credibility due to this form of identity prejudice is the victim of systematic testimonial injustice.

An example of testimonial injustice involving Islamophobia is the way that the testimonies of Muslims were ignored when the French government banned face veiling.11 Many French Muslim women view the hijab in positive terms, as a way of asserting their identity as both French and Muslim, as something that makes them feel safe from objectifying gazes and probing from outsiders, as an empowering assertion of control over “body, mind, and soul”, and as a symbol of unity or solidarity in the face of state repression of Muslims (see Croucher 2008, p. 201 and also Statista Research Department 2024). However, this testimony has been disregarded by French lawmakers, who have instead tended to present veiling as a “mark of submission” or as being imposed by men (see Rosman 2020). This is an epistemic injustice because many French lawmakers have chosen to remain ignorant of Muslim women’s reasons for wearing veils. They have presented the position that many Muslim women adhere to as lacking in credibility, and so they have denied themselves and others knowledge of the reasons Muslim women give for wearing the veil.12 The irony of presenting their measures against Muslim women as being in aid of tackling sexism while they try to control what Muslim women wear and impose fines on them for disobeying seems to be lost on them.13

The other form of epistemic injustice that Fricker focuses upon is hermeneutical injustice, which she initially says occurs “when a gap in collective interpretive resources puts someone at an unfair disadvantage when it comes to making sense of their social experiences” (Fricker 2007, p. 1). Later in her book, she specifically ties hermeneutical injustice to “structural identity prejudice”, which, she notes, helps to convey the discriminatory nature of hermeneutical injustice, and so she gives a revised definition of the term, describing hermeneutical injustice as “the injustice of having some significant area of one’s social experience obscured from collective understanding owing to a structural identity prejudice in the collective hermeneutical resource”.14 Hermeneutical injustice, she notes, is closely tied up with matters of power and influence. Those who are influential in fields like journalism, politics, and law hold more sway when it comes to the hermeneutical resources available to a society. Looking again at the example of Islamophobia in France, we can see that there have been a series of attacks on the collective interpretive resources available to understand racism and Islamophobia in particular. In 2012, the French President, François Hollande, declared that “there is no place for race in the Republic” (Bessone 2020, p. 818). In 2013, the National Assembly voted to remove the word ‘race’ from French legislation. In 2017, the Minister of Education, Jean-Michel Blanquer, said that the word ‘racialized’ was “among the most appalling words of the political vocabulary” (Bessone 2020, p. 826). In 2018, the French magazine Le Point published a letter from eighty public figures occupying influential positions, calling for terms like ‘racialized groups’, ‘structural racism’, ‘white domination’, ‘racial power’, ‘black feminism’, and ‘decolonialism’ to be removed from academic discourse. In 2020, the French government shut down the Collectif contre l’Islamophobie en France (CCIF), which collected data on Islamophobic discrimination (Wolfreys 2023, p. 174), and in 2021, Macron pressured Imams to sign a Charter of Islamic Principles, under threat of repression if they did not sign, which included articles declaring that ‘state racism does not exist’ and that mosques cannot have any political agenda (ibid., p. 166). French government ministers and cultural figures claim an anti-racist motive for trying to eliminate discussion of race and racism; that the notion of ‘race’ is one that has been decisively discredited by science as referring to a biological category and so has no referent and any attempt to revive the term is encouraging racism. However, this argument relies on an understanding of race that is clearly a very impoverished one. It has long been understood that racism combines biology and culture to various degrees and that ‘race’ is socially constructed (ibid., p. 167; see also Rattansi 2020, pp. 80–86; Bessone 2020, pp. 817, 822–24). Moreover, the task of tackling racism is made more difficult if we are not able to use terms like ‘racialized’, ‘race’, and ‘racism’ to comprehend and challenge it. Eradicating organizations dedicated to gathering information on Islamophobia, refusing to gather data relevant to racial discrimination in the census, and trying to eradicate the language used to understand racism clearly has a deleterious effect on trying to understand racism in French society. This is a kind of epistemic injustice because it is an attack on the means that oppressed groups in French society have for understanding and interpreting racism—a hermeneutical injustice.15

Since the publication of Epistemic Injustice, a number of forms of epistemic injustice that were not discussed in that book have come into view. For example, in thinking about hermeneutical injustice we might note that it does not always take the form of oppressed people being denied the means to make sense of their experiences of oppression. Oppressed people are often in a particularly good position, relative to other people, to understand their own oppression. If we want to understand racism, we would do well not to ignore the understanding of those who have suffered it. If we want to understand sexism, we would do well to listen to what women have to say about it. Those who experience oppression thereby have a source of knowledge about that oppression and the ways in which it manifests itself that is denied to others, and they are also motivated to overcome that oppression. However, oppressed people who understand their own oppression might be treated unjustly by others dismissing, ignoring, or distorting their understanding of their oppression. Oppressors have a motivation to try to ensure that understanding of oppression do not become more widespread, and so they ignore the frameworks that oppressed people develop to understand their oppression and propose alternative frameworks that try to justify the oppressed positions of the oppressed. The powerful, at least sometimes, remain willfully ignorant of the ways in which oppressed people understand their lives, preferring the narratives of the powerful in their place. This is what Gayle Pohlhaus (2012) refers to as ‘wilful hermeneutical ignorance’. Gayatri Spivak (1998), Kristie Dotson (2011), and José Medina (2023), amongst others, have developed accounts of what they call ‘epistemic violence’ which is “the violence one suffers when one is harmed as a subject of knowledge and understanding, as it happens when people are silenced, systematically misrepresented, or rendered invisible or inaudible” (Medina 2023, p. 62).

An example that serves to illustrate both willful hermeneutical ignorance and epistemic violence is the way in which much of the Western media misrepresents jihad. There have been countless headlines in the UK press (and internationally) that have presented jihad as nothing but irrational violence driven by mistaken religious beliefs. For example, the UK newspaper The Times carried a headline on 22 August 2014 saying that ‘Hundreds more UK Muslims choose jihad than army’ (Haynes and Hamilton 2014), based on claims of a Member of Parliament that 1500 Muslims had gone to fight in Iraq and Syria and that there were only 600 Muslims in the UK Armed Forces. However, according to Ministry of Defence figures, there were 640 Muslims in the UK Armed Forces at that time,16 and according to the Foreign Office, 400 UK citizens had gone to fight in Iraq or Syria independently of the army (of which half had returned). Moreover, we have no good reason to think that all of those who went to fight in Iraq and Syria went to ‘wage jihad’ if what is meant by that is irrational violence driven by mistaken religious beliefs.

According to CAGE International, an organization set up to represent people who had been unjustly imprisoned during the ‘war on terror’, many of those who went to fight in Syria had quite reasonable political motivations. Their researchers interviewed a UK citizen fighting in Syria who said, “I don’t have any plan for waging jihad or any kind of fighting in the UK… we’ve just come for one purpose, to save the oppressed people of Syria from the evil of the terror of Bashar al-Asad” (Qureshi 2014, p. 8) and they say that this interviewee is representative of commonly held views among those fighting in Syria (ibid., p. 9).

If we look to The Qur’an, to collections of Hadith, and to legal works in the Islamic tradition for an understanding of jihad, we find that the term’s basic meaning is ‘struggle’, ‘striving’, or ‘exertion’ and that a different term, qital, is used to refer specifically to fighting (Afsaruddin 2020, p. 448). The Qur’an has quite strict rules regarding the initiation of hostilities and concerning conduct in war. Killing women and children, brigandage, highway robbery, sedition, and what we now call ‘terrorism’ are all deemed illegitimate violence in The Qur’an and Islamic juridical literature (Afsaruddin 2020, pp. 466–67). So, the concept of ‘jihad’ found in the Western press is very distant from that found in Islamic religious texts and juridical literature. The Western media largely ignores the frameworks used by Muslims themselves to understand jihad (they remain willfully ignorant), and so they do a hermeneutical injustice to Muslims. Moreover, according to Medina’s understanding of epistemic violence, the systematic misrepresentation of Muslims and of jihad amounts to epistemic violence.

What I want to focus on in the following sections is the possibility, raised by Kidd (2017), that there might be deep epistemic injustices committed against religious believers by people holding a different worldview.

4. What Is a Deep Disagreement?

In order to get to grips with deep epistemic injustice, we should first look at how we should understand deep disagreement. In a recent paper, Chris Ranalli and Thirza Lagewaard have identified two mainstream ‘theories’ of deep disagreement (Ranalli and Lagewaard 2022): the Wittgensteinian account and the Fundamental Epistemic Principle account. I will not discuss the fundamental epistemic principle account here because I take it that the hinge account can accommodate the kind of ‘fundamental epistemic principles’ that Lynch discusses, and moreover, as Ranalli and Lagewaard argue, the fundamental epistemic principles account has the disadvantage of being too narrow in scope (Lynch 2010; Ranalli and Lagewaard 2022, p. 12). My focus will be on Wittgensteinian understanding of deep disagreement.

The locus classicus for discussion of deep disagreement, and one that is influenced by Wittgenstein, is Robert Fogelin’s paper ‘The Logic of Deep Disagreements’ (Fogelin 1985), in which he notes that the kinds of arguments that we have in everyday life take place against a shared background of beliefs and preferences.

A disagreement is deep if a shared background is lacking in some way. What Fogelin has in mind in the case of deep disagreement is a disagreement in deep commitments, which Wittgenstein sometimes described as analogous to hinges. It should be noted here that although Fogelin cites passages §§341–344 of Wittgenstein’s On Certainty, where Wittgenstein talks about hinges, he does not generally talk about these deep commitments as being hinges in the manner of modern-day hinge-epistemologists. Fogelin prefers to talk of a disagreement in ‘framework propositions’ or ‘rules’.

What are these ‘hinges’ or ‘framework propositions’? In §341 of On Certainty, Wittgenstein famously says that “the questions we raise and our doubts depend on the fact that some propositions are exempt from doubt, are as it were like hinges on which those turn”. The ‘hinges’, he says, “must stay put” (§343). In other places in On Certainty, Wittgenstein uses other metaphors for these propositions that are exempt from doubt. He describes them as “fixed…removed from the traffic…shunted onto an unused siding” (§210) or as “foundation walls” (in contrast with the “whole house”) (§248), or as a “riverbed” (in contrast with “the movement of the waters” of the river) (§96–99). Interestingly, in the case of the riverbed metaphor, Wittgenstein speaks of the riverbed shifting and of there not being a sharp division between the bed and the waters that flow over it. So, in each case, there is something ‘hard’ or ‘fixed’, which “stands fast” (§116) or which “stand unshakeably fast” (§144) on the one hand, and there is the moving, fluid, shifting (door/traffic/water/parts of the riverbed) element on the other.

For a clearer understanding of what ‘hinges’ are, we should look at some examples of propositions that Wittgenstein uses and say something about why they might be considered ‘fixed’ or ‘hard’. Examples of hinges include, “Here is one hand (accompanied by a gesture of presenting one’s hand)” (§1), “I have two hands” (§§23, 125), “I am writing in English” (§§70, 158), “I am a man” (§79), “This → o color is called blue” (§§126, 522), “I have just had lunch” (§659), “The earth exists” (§§208–10), and interestingly, for our purposes here, he also includes cases of ‘religious hinges’ where different people hold to quite different hinges, such as “belief in a God, or that none exists” (§107) or the belief that “every human being has two human parents” in contrast with the Catholic belief that “Jesus only had a human mother” (§239–40), and Wittgenstein talks about missionaries converting ‘natives’ to their worldview (made up of various religious hinges and other hinges).

One feature that these (hinge) propositions have in common is that it is, at the very least, peculiar for someone to ask how one knows them. If you told someone that you have two hands and they asked you, ‘How do you know?’, you would likely be baffled as to why they were asking such a strange question. It is also worth noting that it would be strange to offer up many of these propositions in the first place, although we can imagine situations in which we might. Propositions like ‘here is a hand’ and ‘the earth exists’ are very rarely articulated except in philosophical discussions. Some of them (“This → o color is called blue”, “I have two hands”) might be used in training young children in basic color concepts, in counting, or in naming parts of their body. We can get a better sense of why it is odd to articulate them and why it is odd to ask how we know them by thinking about how we might try to investigate them, produce evidence for them, or justify them.

In remark §125 of On Certainty, Wittgenstein says that “If a blind man were to ask me ‘Have you got two hands?’ I should not make sure by looking. If I were to have any doubt of it, then I don’t know why I should trust my eyes. For why shouldn’t I test my eyes by looking to find out whether I have two hands? What is to be tested by what?” His point is that one cannot provide justification for the claim ‘I have two hands’ with the claim that ‘I can see that I have two hands’ because the second claim is no more certain than the claim it is supposed to support and so it does not provide support. The claim that ‘I have two hands’, we might say, is absolutely certain. There is nothing that we could be more certain of that would provide justification for it. We do not, of course, in ordinary circumstances look to check whether we have two hands. That would strike us as ridiculous. However, we can imagine circumstances in which we might look to check (upon waking up injured in hospital from a massive car crash, for example). The ‘claim’ that ‘I have two hands’ is not something that we arrive at through investigation or through discovering evidence. It is not something that we can justify, nor is it unjustified, and the same can be said of the other examples listed above. There is no investigation that I could perform to discover that I am writing in English when I am writing in English, and I could not be said to have evidence that that is what I am doing. The same can be said for ‘God exists’. People who believe in God do not arrive at belief in God through conducting an empirical investigation. They are generally raised in households where religious language is imparted to them as they engage in religious practices, like praying and religious worship, and come to believe in God in that way.

In Wittgenstein’s picture, the ‘foundational’ propositions, the hinges, are neither justified nor unjustified, and they are not arrived at through investigation, so they cannot be said to be items of knowledge. The hinges are not supported by evidence, and neither are they self-evident (they are not “intrinsically obvious or convincing” (§144)); they are ‘swallowed down’ (§143) with our training in concepts and with the acquisition of beliefs.17 The beliefs that I hold fast to form a ‘nest’ (§225); they are not held in isolation from one another, and the beliefs that stand fast are held fast by what lies around them (a vast system of beliefs) (§144). When it comes to ordinary empirical matters, we might be said to know things and might also have doubts about things, but in the case of hinges, doubt is excluded (logically) (§88). It must be that certain propositions are not doubted if the game of doubting (and justifying, enquiring, and knowing) is to get off the ground. When Wittgenstein characterizes the distinction between hinges and other kinds of belief in less metaphorical terms, he sometimes characterizes it in terms of something that might be tested by experience (ordinary empirical beliefs) and rules of testing (hinges) (§§98, 309, 319) or in terms of empirical beliefs and norms of description (§§167, 321). This rule or norm-like nature can perhaps be seen most clearly in something like the proposition ‘This → o color is called blue’ which might be used to instruct someone in how to use the term ‘blue’, whereas an empirical proposition involves using the term ‘blue’ to attribute the color to some object (e.g., ‘the cover of my notebook is blue’).

One final thing to note about hinges is that they are diverse. Trying to doubt some of them would be incoherent because it would bring everything into doubt (and therefore, the game of doubting could not get off the ground at all). So, doubting whether ‘this → is my hand’ would be to “stand before the abyss” (§370), according to Wittgenstein. Doubting that “this person here [sat opposite me] is [my old friend] N. N.” would “drag everything with it and plunge it into chaos”18 but doubting whether “the water in the kettle on the gas flame will not freeze but boil” would not (§613). Particularly relevant to our considerations here is that it seems that coming to doubt one’s religious beliefs would not “drag everything with it”, although it may well be a highly disconcerting experience, shaking up many elements of one’s worldview.19

Does the foregoing discussion of deep disagreements and hinges mean that we should understand deep disagreements over religious issues in terms of hinges? I think the fact that ‘deep disagreement’ is a term of art20 and that it was introduced by Robert Fogelin in an influential paper only forty years ago in roughly these terms means that we should take seriously the possibility that at least some deep religious disagreements will take this form. Moreover, as I will argue a bit later, I think Duncan Pritchard provides good reasons for thinking that religion is a particularly promising area to find deep disagreements over hinges.

However, I think it is doubtful that all deep religious disagreements are disagreements over hinges, and we also have good reason to doubt that all disagreements over hinges are deep disagreements. Victoria Lavorerio has pointed out that paradigmatic examples of deep disagreements, such as disagreement over intelligent design, are not (or at least not always) disagreements over hinges. She argues that the parties in the disagreement over intelligent design, the evolutionists and the proponents of intelligent design, might have different hinge commitments (proponents of intelligent design are Christians, and so believe that ‘God exists’, for example) but that those hinge commitments need not enter into the debate (the debate is an empirical one over whether certain organisms are ‘irreducibly complex’ (Behe 1996, p. 39) and the claim that some are has no scientific merit) but the disagreement might nonetheless be considered deep if, with Lavorerio, we define deep disagreements as “persistent conflicts of opinions about matters that can be said to be worldview defining” (Lavorerio 2021b, p. 416).

It seems, then, that at least according to one definition of ‘deep disagreement’, there are cases of deep disagreement that are not cases of disagreement in hinges or disagreement in fundamental epistemic principles. However, that does not mean that there are not at least some deep disagreements that are disagreements over hinges, and Lavorerio accepts this. Pritchard agrees with Lavoreiro that not all deep disagreements are disagreements over hinges, and he also argues that not all disagreements over hinges are deep disagreements, though at least some are.21

We can see that not all deep disagreements are disagreements over hinges if we think about ‘quotidian hinges’ such as ‘I have two hands’ and ‘My name is R. V.’. It is difficult, if not impossible, to imagine a genuine dispute arising over these hinges. Pritchard argues that the Wittgensteinian position is that we could make no sense of someone who doubts these commonsense hinge commitments—that we would treat someone as lacking sanity if they doubted them. Pritchard also argues that purely philosophical disputes are not good candidates for being deep disagreements because even though philosophers do present arguments in opposition to claims like ‘there are physical objects’ and ‘the future resembles the past’ their actions “reveal that the underlying certainty in these [the relevant] hinges is unaffected” (Pritchard 2023, p. 311).22 Here, I will agree with Pritchard that disagreements that concern issues of existential importance are the most plausible candidates for cases of deep disagreement, and these disagreements, at least sometimes, are disagreements over hinges.23 In addition to being a good account of deep disagreement I also understand Pritchard’s position to be very much in tune with the way that Wittgenstein treated philosophical problems and religious problems. Though Wittgenstein spoke of philosophical problems as having “the character of depth” (Wittgenstein 2009, §111), he very often criticized past philosophical positions as being nonsensical (Wittgenstein 2009, §§119, 246, 252, PPF§328) and in the Investigations said that he wanted to teach people “to pass from unobvious nonsense to obvious nonsense” (§464) in philosophy. However, in the case of disagreements between people with different worldviews (perhaps with differing religious hinges), he did not accuse religious believers of speaking nonsense (though, of course, language can “go on holiday” (§38) in the case of religious believers just as it can for anyone else). There might be difficulties in understanding when people with different worldviews meet, but this does not mean that either party is speaking nonsense. Here, there really is potential for people to sincerely disagree over issues of existential importance.

Religion, unlike philosophy, then, seems to be a field particularly suited to deep disagreements. It is a field in which people genuinely disagree in some sense but also a field in which people might not share a common background. As was already noted above, in giving examples of hinge propositions, Wittgenstein sometimes gives examples where people with different hinges come into conflict, and several of those examples are religious ones. There is the disagreement between the person who believes in God and the one who does not (§107), there is the disagreement between the person who believes that every human being has two human parents and the person who believes that Jesus only had a human mother (§239), there is the disagreement between Moore, who says that “I know that this is wine and not blood” and the Catholic who contradicts him (§239), and there is the disagreement between (Christian) missionaries and the people they set out to convert (§612).24

I say that the people disagree in some sense because disagreements over religion are not like cases where someone asserts an empirical proposition and another person asserts the opposite, contradicting them (e.g., someone says ‘there is water on Mars’ and another person says ‘there is no water on Mars’). In his ‘Lectures on Religious Belief’, Wittgenstein gives the example of a believer who says, “I believe in a Last Judgement”, and they are met with the response: “Well, I am not so sure. Possibly”. In this case, Wittgenstein says, “[y]ou would say that there is an enormous gulf between us”. On the other hand, if someone said, “There is a German airplane overhead”, and another responded, “Possibly, I’m not so sure”, then you would not say that there is an enormous gulf between the two disputants (Wittgenstein 1966, p. 53). In the German airplane disagreement, the two people have a shared background of empirical beliefs about planes and about Germany and agree about the means for settling the dispute. One might offer the other binoculars and say, “Look—you can see that it says Deutsche Aircraft on the side of it, and that is a German make of plane”, and that may well settle the argument. However, in the case of the disagreement over a Last Judgement, the dispute cannot be settled in a similar manner. The two parties to the dispute cannot do something like pull out binoculars or cite some piece of evidence. For the believer in the Last Judgement it is not a matter of evidence or of possibility. Wittgenstein calls the belief in a Last Judgement an ’unshakeable belief’ (54) and says that “it will show…by regulating for all in his life”—the disagreement is a disagreement in form of life25 or way of life, we might say. The fact that Wittgenstein describes the belief in the Last Judgement as ‘unshakeable’ suggests that he is thinking of it as being like the fixed beliefs, beyond all doubt, that he describes as being like hinges or a riverbed in On Certainty. So, the conflict between the two disputants is a conflict between one who has a religious hinge (or something hinge-like) and one who is thinking about a Last Judgement as being something possible, as an empirical belief might be. The believer’s belief in the Last Judgement is part of a worldview (it is one element in a nest of hinges and religious beliefs) that their interlocutor does not share, and that worldview is associated with a form of life that the interlocutor does not participate in. So, although the non-believer understands the component words in what is said to him (‘God’, ‘Last Judgement’ etc.), they do not, as Genia Schönbaumsfeld says, “understand how a life with such a concept is possible, or how one would go about acquiring the qualities of heart that would make it so” (Schönbaumsfeld 2023, p. 35).26

Before moving on to look at the possibility of deep epistemic injustice, let us first look at one more example of deep disagreement involving religious hinges. If we think about what the fundamental beliefs of Islam are, we may well conclude that the oneness of Allah (or God) is absolutely central.27 In sura (chapter) 112 of The Qur’an, it says, “Say, ‘He is God the One, God the eternal. He begot no one nor was He begotten. No one is comparable to Him’” and the Prophet Muhammed, commenting on this sura said that, although it was short it was equal to one-third of The Qur’an owing to its importance (The Qur’an 2015, p. 444 (sura 112, 1–4)). The centrality of this belief in Islam and the importance given to it already suggest that it might have a hinge-like nature for many Muslims. Muslims do not think that the oneness of God is a merely contingent feature of God, they take it to be of God’s essence that ‘God is the One and begot no one’ and as Wittgenstein said, “Essence is expressed in grammar” (Wittgenstein 2009, §371). It seems that Wittgenstein, when he was working on Philosophical Investigations, thought that the central religious beliefs of religions like Christianity and Islam are quite unlike ordinary empirical beliefs. Instead, he thought of them as being grammatical (Wittgenstein 2009, §373) and, therefore, rule-like, in the manner of hinges. Now, this central belief of Islam places Muslims in disagreement not only with non-believers but also with polytheists, as well as Trinitarian Christians. In this latter case, it seems plausible that what we are talking about is a disagreement in hinges. Whereas for Muslims, it is a hinge that “Allah is God the One” and that “He begot no one”, for many Christians, it is a hinge that “God is three persons: the Father (begets), the Son (is begotten), and the Holy Ghost (proceeds)” and that “Jesus is the son of God” (see, for example, Luke, 4:41). Where Muslims and Christians sincerely disagree over these deeply held beliefs we may well have a case of deep disagreement.

5. Deep Epistemic Justice

Let us now consider whether there might be cases where deep disagreement exists, and one of the disputants is done an injustice in the course of the dispute—a deep epistemic injustice. The first thing that we might reflect on is that these cases are going to look very different from typical cases of testimonial injustice, where somebody tries to convey a piece of information or knowledge, and another person tries to weigh up how likely it is that what has been said is true. In cases of deep disagreement over religious belief, where the belief in question is a hinge or hinge-like, we should notice that:

• A religious hinge is not a piece of information or knowledge (as a hinge, it is neither justified nor unjustified, it does not rest on evidence, it is beyond doubt for the person committed to it).
• There are no shared criteria by which a person might evaluate what the other person has said as true or false. The normal means of determining whether something is true—by gathering evidence, having a look—are not available to us in this case (we can gather evidence or have a closer look to determine whether a plane is a German one flying overhead, but we cannot gather evidence or have a closer look when what is under dispute is whether there will be a Last Judgement).

So, these cases do not fit Fricker’s model for testimonial injustice. However, we surely want to say that people might be treated unjustly in the uptake of what they say about their fundamental religious beliefs, given that these beliefs are deeply tied up with a person’s identity.

Thinking about the kind of cases discussed above, of newspapers printing misleading headlines about Muslims and governments failing to listen to the testimony of Muslims, I think we should first of all note that cases of Islamophobia are not necessarily deep disagreements. Muslims might disagree with non-Muslims in some of their hinges, but they are bound to agree in many others, and the religious hinges of Muslims are likely to have considerable overlap with those of other people who believe in an Abrahamic religion (most will agree, for instance, that ‘God exists’ and that this is a vital component of their religious outlook).

In thinking about possible examples of deep epistemic injustice in the area of religion, one way in which we might think about it is in terms of prejudices (such as Islamophobia or antisemitism) being hinges, or at least hinge-like.28 In that case, a disagreement between an Islamophobe, who presupposes that ‘Islam is evil’ and a non-prejudiced person who says that ‘Islam is not evil’ might be characterized as a deep one (because it involves a disagreement in something hinge-like) and the Muslim who is not believed when characterizing their religion as a religion of peace and love due to this prejudice might be characterized as suffering a deep epistemic injustice. I use the term ‘epistemic’ here because, although I think that prejudices are hinge-like in the ways that Anna Boncompagni has recently suggested (they are resistant to counterevidence (Boncompagni 2024, p. 289), often implicit (Boncompagni 2024, p. 290), swallowed down in childhood (Boncompagni 2024, p. 295)), I do not think that they can be described as neither justified nor unjustified (as Wittgenstein describes hinges). Prejudices are unwarranted or unjustified prejudgements (See (Allport 1979, p. 8) and (Fricker 2007, p. 34)). I follow Fogelin in thinking that an accusation of prejudice, like an accusation of pig-headedness or an accusation of bias, “only makes sense in a normal (or near normal) argumentative context (Fogelin 1985, p. 7). Perhaps some disagreements between Islamophobes and Muslims could be characterized as deep since prejudices are sometimes deep-seated and hinge-like (and we have already seen that not all deep disagreements are disagreements over hinges).

A rich source of disagreements involving religion that might be characterized as deep epistemic injustices is Elizabeth Anscombe’s paper ‘Paganism, Superstition and Philosophy’ (Anscombe 2008a). There, she discusses the way in which the (pagan) Tacitus describes Jewish people in extremely unflattering terms as “hateful to the gods”, “contrary to the rest of mankind”, as allowing ‘incest’, with ‘sinister’ customs “validated by their filthiness” (2008a, pp. 49–50). Here, you have a case of people with very different worldviews (pagan and Jewish) coming into conflict where what one regards as sacred, the other regards as profane. Tacitus is undoubtedly unjust to the Jewish people. He presents a distorted and inaccurate picture of the followers of Moses, but some of the disagreement is simply a disagreement in form of life. Anscombe also discusses the way in which followers of one religion have viewed themselves as superior to the followers of other religions by not accepting that the religion in question is a religion at all: “Christians at a much later date [than Tacitus] called Islam superstition and paganism superstition and rabbinical (i.e., post-Christian) Judaism superstition. Protestants called Catholic Christianity superstition, and also Russian Orthodoxy…” (Anscombe 2008a, p. 57). In fact, Anscombe herself might be accused of doing the same thing, since she characterizes Christianity as “the true religion” (2008a, p. 58) and portrays Hinduism as a kind of paganism (Anscombe 2008a, p. 55). In another paper, she makes clear her disapproval of Hinduism on the basis that it violates the Christian prohibition on idolatry. The prohibition on idolatry, she says, “is becoming rather more relevant now: with present fashions, some of our young people are being drawn into spiritual cults which include the worship of Shiva and Krishna and other deities of the Hindu pantheon…but in any case it is clear that these are (false) gods” (Anscombe 2008b, p. 231).29

A further potential site for deep epistemic injustice is in disagreements between naturalistic worldviews and religious ones (Kidd 2017, p. 393). A particularly crude form of scientistic naturalism can be found in the work of Richard Dawkins, who claims that ‘God exists’ is a scientific hypothesis about the universe (Dawkins 2016, p. 24) and that the question of whether Jesus had a human father is a “strictly scientific question” (Dawkins 2016, p. 83). However, as Wittgenstein has shown, it is plausible that ‘God exists’ and ‘Jesus only had a human mother’ might take on a hinge-like (or grammatical) role in the lives of Christian believers, whose practices are very unlike those of scientists. Moreover, Dawkins is clearly not alone in subscribing to such scientistic views. His work has been hugely influential (The God Delusion has sold millions of copies, and Dawkins has a TV series, ‘The Root of All Evil?’ based on it). This kind of scientism, portraying religion as a crude form of empirical belief in competition with science, has a long history. Wittgenstein criticized James George Frazer for presenting ritual behaviors as “pieces of stupidity” (Wittgenstein 1993, p. 119), and Frazer’s The Golden Bough was first published in 1890. This kind of scientism, which is often accompanied by describing religious beliefs in highly unflattering terms, clearly has the potential to undergird discrimination on the basis of religion. Indeed, in Dawkins’ case, it has been combined with a particular animosity toward Islam and Muslims.30

6. Conclusions

In this paper, my aim was to give an account of what deep epistemic injustice might look like. In order to do that, I first looked at standard kinds of cases of epistemic injustice so that we might contrast the ‘deep’ case with more ordinary cases. I then proceeded to present a Wittgensteinian account of deep disagreement and argued that deep disagreements between believers and non-believers (or between adherents of different religions) might take the form of disagreements over hinges but that this was not inevitably the case. Lavorerio (2021b) and Pritchard (2021, 2023, 2024) have made a convincing case that deep disagreements might involve a clash of worldviews and yet not involve a dispute over hinges (and that clashes over hinges need not count as deep disagreements). Then, in Section 5, I explained how cases of deep epistemic injustice differ from ordinary cases. Where a dispute is over a hinge or something hinge-like, we might question whether the person offering testimony is saying something true or offering up a piece of knowledge, and the depth of the disagreement will produce corresponding difficulties in assessing the truth of what the person offering testimony says, as well as difficulties in assessing the credibility of the person offering testimony. I suggested that a disagreement between an Islamophobe and a Muslim might involve a deep epistemic injustice because prejudices are hinge-like and also suggested that Anscombe’s work provides us with many examples that are plausible examples of deep injustices. Furthermore, disagreements between scientistic and naturalistic worldviews provide a further potential site of injustice.

Throughout the paper, I used examples of Islamophobia since it has received little attention in the literature on epistemic injustice and because discrimination on the basis of religion (as well as racist discrimination) seems like a plausible area in which we might find deep epistemic injustices of the sort Kidd has described.