1 Moral revolutions, deep disagreements, and moral progress

In this paper I will be exploring the nature and implications of disagreements across what are called “moral revolutions.” Moral revolutions are rightly coming to be recognised as a philosophically interesting and historically important mode of moral change. The study of moral revolutions has the potential to illuminate aspects of our individual moral psychologies and the sociology of institutional moral change (Pleasants, 2018; Tam, 2019; Hermann, 2019; Eriksen, 2020) as well as bearing directly on first-order questions in normative ethics and moral epistemology (Anderson, 2014; Kitcher, 2021; Lowe, 2019).

What is less often acknowledged is that the very characteristics that make a moral change revolutionary pose a fundamental challenge to the possibility of moral progress. This is because moral revolutions, as I will argue below, are characterised by a diachronic form of deep moral disagreement: moral agents on either side of a moral revolution adopt different standards for assessing the merits of a moral argument, and according to the standards they adopt, agents on neither side of the revolution ought to accept the arguments being put forward by those on the other. This threatens to undermine the supposed progressive status of many of our most celebrated historical examples of moral revolutions: if a moral revolution only looks progressive in light of a change in evaluative standards it itself brought about, then moral progress is relative to a particular historical perspective; if prior to a moral revolution taking place the moral arguments in its favour fall short of the relevant standards, moral revolutions can never be the outcome of rational argumentation. This apparent relativism and irrationality sits uncomfortably alongside our common-sense intuition that the moral revolutions upon which our contemporary moral worldview rests count as genuine moral progress. If, as Michelle Moody-Adams puts it, commitment to the idea of moral progress ‘is a condition of the possibility of morally constructive action for beings like us’ (2017, p. 154), this poses a serious philosophical problem. This paper is an attempt to theorise this hitherto neglected problem, and to assemble some resources for resolving it.

First things first: what is a moral revolution? Popular and morally celebrated examples include:

  • The abolition of institutionalised slavery;

  • The moral acceptance of homosexuality as normal and natural;

  • Recognition of the moral parity of the sexes.Footnote 1

What makes these moral changes revolutionary? A frequently cited characteristic of moral revolutions is their ‘radicality, depth or fundamentality’ (Klenk et al., 2022, p. 355). This is typically understood to involve a widespread transformation in the moral practices and shared moral norms and standards of a society.Footnote 2 With a few exceptions (e.g., Roth, 2012), philosophers interested in the “revolutionary” character of moral revolutions model their analyses on Thomas Kuhn’s The Structure of Scientific Revolutions (2012) – hereafter just Structure – a noteworthy example being Robert Baker’s appositely named book The Structure of Moral Revolutions (2019). However, one concept is conspicuous in its absence from Baker’s (and many others’) analyses of the phenomena of moral revolution: that of progress.Footnote 3

One reason for avoidance might be that the relationship between scientific revolutions and scientific progress, as Kuhn presents it in Structure, is often perceived as acutely problematic. The comforting and familiar picture of scientific progress—the continuous accumulation of scientific facts, inherited from one generation of scientists by the next—is for Kuhn a propagandistic myth, perpetuated to enculturate scientific trainees into the current scientific “paradigm”: a scientific community’s current way of seeing and doing science. That myth renders invisible to scientists their own discontinuous and non-rational intellectual history.

Kuhn hoped to demonstrate that during periods of scientific “crisis,” science proceeds not by theoretical proof and experimental corroboration, but through evangelism, leaps of faith, and individual epiphanies, resulting in the eventual mass conversion of a scientific community—what Kuhn called a “paradigm shift.” Kuhn showed how, in an important sense, scientists in the new paradigm occupy a new “world,” one populated by entirely different theoretical objects and requiring new and incompatible ways of measuring scientific success.

Given the “incommensurability” of the new and old worlds—a term I will give more sense to below—a paradigm shift will always look like progress to scientists working in the new paradigm, but it will not have looked like progress to scientists working in the old one. The problem is that none of these scientists seem to be making any mistakes. They are acting perfectly in accordance with the standards of evidence and proof characteristic of the paradigm to which they belong. In this sense, disagreements between scientific paradigms are good candidates for being called “deep”.

The term “deep disagreement” was introduced by Robert Fogelin to demarcate disagreements that are in principle irresolvable by rational means from disagreements that are only contingently or idiosyncratically irresolvable by rational means. As Kuhn did for the activity of science, Fogelin distinguishes between normal and non-normal contexts of argument. Argument is normal when it occurs against a backdrop of shared beliefs, preferences, and commitments. Against such a backdrop, discussants may ‘disagree over simple questions of fact, but, in general, they agree on the method for resolving their disagreement’ (1985, p. 3).Footnote 4 Deep disagreements arise when this backdrop deteriorates: ‘to the extent that the argumentative context becomes less normal, argument, to that extent, become impossible’ (p. 4). This poses an obstacle to progressFootnote 5: insofar as progress consists in the rational resolution of disagreements, deep disagreement makes progress impossible. Neither disputant seems to be making any mistakes. They are acting perfectly in accordance with the standards of argument they accept. On the face of it, it would seem irrational for either side to change their mind.

I will argue below that, because of the paradigm-like structure of moral inquiry, disagreements across moral revolutions are “deep” if any moral disagreements are. Consequently, whether a moral revolution looks like progress to you also seems to depend on which side of the revolution you find yourself. Our belief that our current moral practices are moral improvements on the practices they replaced begs the question—or seems to, from this angle. The phenomenon of deep moral disagreement is therefore not only an apparent threat to future moral progress—how do we overcome the synchronic deep moral disagreements we currently experience?—it also threatens to undermine our confidence in the progressive status of our surest moral achievements: those that form the unquestioned basis of our modern moral worldview. In summary: as well as obstructing future progress, deep moral disagreement threatens to undermine our confidence in the moral progress we have already made.

The paper is structured as follows. I begin by introducing the three concepts I will be deploying to tackle the problem of deep disagreement across moral revolutions: scientific paradigms, epistemic world-pictures, and moral paradigms. I argue that these concepts bear certain family resemblances to one another, entitling us to explicate them along similar lines. With this established, I draw on these resemblances to theorise the ways in which moral revolutions give rise to deep moral disagreement, namely, through the phenomena of perceptual, conceptual, and practical incommensurability. Lastly, I develop broadly Wittgensteinian responses to three worries arising from a recognition of the incommensurability of moral paradigms: the relativisation of moral standards to a paradigm; the apparent irrationality of moral paradigm shifts; and the obscurity of how moral revolutions can be morally progressive.

2 Scientific paradigms, world-pictures, and moral paradigms

To state the problem more clearly, and to lay out the resources available for confronting it, I draw attention to five features shared jointly by what Kuhn called paradigms, what Wittgenstein called world-pictures, and what have been called “moral paradigms” in some pockets of the moral revolutions literature. This will both illuminate the problem just posed, as well as permitting constructive resource sharing between philosophies of revolutionary scientific progress and revolutionary moral progress.

What is a world-picture? The term is notoriously vague. Michael Kober describes a world-picture as ‘[t]he total of all knowledge-claims and epistemic norms of the respective practices of a community’ (1997, p. 373). Wittgenstein himself spoke of a world-picture using the simile of a shared mythology (2008, §95) that both embodies the beliefs and commitments of a particular community and gives community life its practical structure (Kober, 1996, pp. 418–419). There is nothing necessary, inevitable, or immutable about a world-picture (Wittgenstein, 2009, §97). Rather, it is localised to a particular community, with particular aims, operating in a particular context (Kober, 1996, p. 419).

Also notoriously vague is Kuhn’s term “paradigm.” Margaret Masterman (1970, pp. 61–65) famously listed ‘not less than twenty-one different senses’ of the word deployed by Kuhn in Structure.Footnote 6 Kuhn himself eventually whittled it down to just two: a broad or “global” sense, and a narrow or “local” sense. The global sense of “paradigm” is the totality of shared commitments of a scientific group; it is ‘what the members of a scientific community, and they alone, share. Conversely, it is their possession of a common paradigm that constitutes a scientific community of a group of otherwise disparate [individuals]’ (1977, p. 294). An illustrative metaphor might be the notion of a “national identity”: a national identity, if it is anything, is a set of commitments and values that uniquely demarcates citizens of one nation from other nations and which hold them together as a nation. A national identity encompasses the factual beliefs, normative commitments, and practical priorities required to get on as a citizen in a particular nation. Likewise, a scientific paradigm encompasses the factual beliefs, normative commitments, and practical priorities required to get on within a particular scientific discipline. Kuhn later suggested the term “disciplinary matrix” as a synonym for this sense of paradigm (2012, pp. 181–186). The local sense is an important subset of those commitments, the science’s archetypal successes, by analogy with which all future problems within that science are to be solved. Returning to our metaphor, a national identity may be anchored by a handful of occasions of especially great national pride—for Britain, candidates might include winning the 1966 football world cup, the creation of the NHS after the second world war, and Britain’s withdrawal from the Atlantic slave trade in the nineteenth century. Kuhn suggested the term “exemplar” as a synonym for this sense of paradigm. These occasions of national pride are to British identity what scientific exemplars are to a scientific paradigm.

In a nutshell: a paradigm in the “global” sense is that which is shared by and consolidates a particular community of inquiry; a paradigm in the “local” sense is a constituent part of the paradigm in the global sense, the examples of successful problem-solutions by analogy with which future problems are to be solved. To keep things simple, when I refer to paradigms, I will do so in the global sense. When referring to paradigms in the local sense I will use the term “exemplars”.

It is easy to point to scientific paradigms. Some specific examples of scientific paradigms discussed by Kuhn are:

  • Aristotelean hylomorphic mechanics (2012, p. 10);

  • Newtonian mechanics (p. 30, 33);

  • Quantum mechanics (p. 48);

  • Relativistic dynamics (pp. 72–75);

  • Daltonian atomism (p. 79).

Scientific inquiry is (relative to other domains of inquiry) highly structured, formally regulated, and meticulously coordinated. Scientists form a tight-knit professional community; Kuhn wrote that scientific paradigms belong to ‘communities of perhaps a hundred members, sometimes significantly fewer’ (1977, p. 297). What counts as scientific success can with little effort be gleaned from scientific journals and textbooks. All of these features make scientific paradigms relatively easy to locate and to analyse.

On the other hand, epistemic inquiry more broadly—the practice in which every epistemic agent is engaged when they doubt and assert and argue about facts—is very loosely structured. There is no community of “professional knowers”. The epistemic community consists of everyone of whom we have epistemic expectations—that is to say, all (or very nearly all) agents.

The same is true of what I will be referring to as moral paradigms. Palmer and Schagrin (1978, pp. 266–267) emphasise that, unlike science, morality is in no way the preserve of an elite and insular professional community. Rather, ‘to see moral paradigms one needs to look beyond the narrow confines of the academic community to the open social space of ordinary moral agents and their everyday moral practice’ (Pleasants, 2018, p. 579). Therefore, ‘moral agents can only be conceived as constituting a moral society, not a moral community’ (ibid.), in Kuhn’s original sense of the term “community.” The moral society consists of everyone of whom we have moral expectations—which again is to say, all (or very nearly all) agents.

These differences obscure the paradigm-like structure (if any) of general epistemic and moral inquiry from view. Because the boundaries of moral and epistemic communities are fuzzy, it is much harder to talk clearly about what it is that binds them together as communities. Despite this, in what follows I want to make the case that scientific, epistemic, and moral inquiry exhibit a “family resemblance” that licenses us to think about them similarly and to expect them to exhibit comparable revolutionary dynamics.

2.1 Five family resemblances

In the following subsections I will outline five features that can be imagined as occupying the overlapping region of a three-way Venn diagram, the circles of which enclose the features scientific, general epistemic, and moral inquiry respectively. These circles are themselves fuzzy and the five features placed at their intersection are exhibited in different ways and to different degrees by each form of inquiry. What I intend with this list is not, therefore, to enumerate sufficient or even necessary conditions for a domain of inquiry to be structured in a paradigm-like way. Rather my aim is to draw attention to some of the overlapping fibres criss-crossing between these three domains of inquiry, which allow us to see them as (pardon the mixed metaphor) belonging to the same family. This, in turn, entitles us to further explicate the concepts of a scientific paradigm, an epistemic word-picture, and a moral paradigm along similar lines.

I do not want to suggest that any of these five features are uncontroversial or that they must necessarily be bundled together. Each of the five features I identify are subject to legitimate criticism and the plausibility of each must be assessed on its own merits. There is no reason to adopt a take-it-or-leave-it stance with regards to these Kuhnian and Wittgensteinian ideas. While the case for treating scientific, epistemic, and moral inquiry as relevantly similar will be stronger if each domain of inquiry really does exhibit each of these features—as I believe they all do, to different degrees—it does not stand or fall with any of them.

2.1.1 They ground our “way of seeing”

Kuhn wrote that ‘something like a paradigm is prerequisite to perception itself’ (2012, p. 113). A theoretician is taught to see her problems in particular ways, as resembling the paradigmatic solved problems (exemplars) of her science (1977, p. 305). Equally, how an experimenter interprets her observations is in part a function of her paradigm-infused perception (2012, p. 197). For theorists and experimentalists alike, enculturation into a paradigm saturates a scientists’ perception of the world with various degrees of likeness to known exemplars; it furnishes scientists with a theoretically loaded representation of reality.

Likewise, an epistemic world-picture grounds our perceptual knowledge—it is a way of seeing. Wittgenstein distinguishes “seeing-as” from other merely sensory aspects of perception (2009, p. 207), the difference being that when one is seeing-as, one ‘is also thinking of what he sees’ (§139). Accordingly, ‘what is at issue is determination of concepts. What forces itself on one is a concept’ (§191). When I am looking at a tree (2008, §267), or at cutlery on the table (2009, p. 205), I do not judge or infer from my perceptual experience that what I am seeing is a tree or cutlery—I just see a tree, I just see cutlery, and it generally takes no effort, no act of interpretation, to see these objects as a tree or as cutlery. We do not and need not justify seeing the objects as such; they are simply what we are seeing, and ‘[w]e shall stick to this opinion, unless our whole way of seeing nature changes’ (2008, §291). This is because our concepts—the concepts of trees and cutlery, for instance—are built into our perceptions (2009, p. 217). A world-picture, then, can be thought of as a concept-infused representation of reality. This will be useful for understanding the incommensurability of successive paradigms.

As with scientific perception within a scientific paradigm, moral paradigms infuse perception with degrees of resemblance to known moral cases and kinds. According to Pleasants, to perceive wrongdoing is to see that ‘the victims of a practice are moral agents or patients of the same kind as those that they already recognize, and that the practice embodies the wrongness exemplified by other acts and practices that they already regard as unquestionably wrong’ (Pleasants, 2008, pp. 109–110). Enculturation into a moral paradigm saturates moral perception of the world with degrees of resemblance to known moral cases and kinds; it furnishes agents with a morally loaded representation of reality.

The metaphor of “moral perception” is liable to lead to some confusion. This liability has not been helped by a tendency amongst philosophers of moral perception to draw on inapt analogies, in particular, on the analogy of moral perception with colour perception.Footnote 7 While analogies with colour are illuminating in some respects—they capture an important interplay between subjective responses and objective stimulus involves in moral perception, and the non-inferential status of moral-perceptual knowledge (De Mesel, 2016, pp. 541–542)—they are obfuscating in others. One obfuscation has to do with the possibility of trying to see something a certain way. It does not make sense to try to see an object as having one colour or another—colour perception is passive.Footnote 8 But it does make sense to try to see the different moral dimensions of the same situation or event—there is, or can be, an “active element” (pp. 544–546) to moral perception.Footnote 9 Relatedly, a second obfuscation has to do with the role of learning or being taught a particular perceptual capacity. Normal colour-perceptual capacity is not acquired by “colour education” (colour words are, and so is the capacity to discern finer nuances of colour, but the general capacity to perceive colour veridically is not). Normal moral-perceptual capacity, on the other hand, is acquired by a process of “moral education,” by which one becomes a competent user of the correct moral concepts (pp. 546–548).

Much better analogies come from what Wittgenstein called “aspect perception”—unlike colour perception, aspect perception has an active element (Wittgenstein, 2009, p. 224) and requires mastery (§168), and like colour perception it captures the interplay of subjective response and objective stimulus and the non-inferential status of the resultant knowledge (De Mesel, 2016, pp. 542–544).

2.1.2 They suggest ways of getting on, often by analogy with familiar cases

Paradigm-infused perception plays an enabling role in scientific practice. It determines some empirical facts ‘to be particularly revealing of the nature of things’ (Kuhn, 2012, p. 25) and worthy of looking into. It also shapes how scientists go about solving their problems. Paradigms assure the scientist of ‘what both the world and [her] science are like’ (p. 42) and suggest ways of making progress. For instance, once a scientist has seen a degree of likeness between an unsolved problem in a new context and a paradigmatic problem with a known solution, she sees a way forward.

World-pictures also play an enabling role in practice. They suggest a way of engaging with, getting on in, and talking about the world. Our concepts ‘lead us to make investigations. They are the expression of our interest and direct our interest’ (Wittgenstein, 2009, §570). A world-picture gestures to the bits of the world that are important and suggests lines of inquiry that might lead to important results.

Likewise, our moral paradigm directs our attention to ‘what morality fundamentally is about’ and shows us ‘who or what counts morally’ (Pleasants, 2018, p. 580). Furthermore, it suggests ways of tackling ambiguous and contentious moral problems (2018, p. 581). For instance, once a moral agent sees a degree of likeness between an ambiguous problem and a moral problem with which she is familiar, she has some idea of how to proceed.

An excellent example of how paradigm-infused moral perception directs our moral interests is provided by Ian Hacking’s discussion of the concept of child abuse. Hacking is aware of the normative significance of how we classify our world:

Classifications evaluate who is troubling or in trouble. Hence they present value-laden kinds, things to do or not to do. Kinds of people to be or not to be. … Classification can change our evaluations of our personal worth, of the kind of person that we are. (1999, p. 131)

The emergence of the categories of “child abuse”, “child abuser”, and “abused child” in the latter half of the twentieth century has had a profound influence on what we think morality is fundamentally about: ‘In our present system of values, genocide is the worst thing that one group can do to another. Abusing a child is the worst thing that one person can do to another’ (p. 132). Hacking draws attention to the way the category of child abuse has drawn our attention to certain phenomena to which previously not much attention had been paid. Our concept of child abuse is both an expression of our moral interest and directs our moral interest; it has brought the phenomena that now fall under it, or even uncomfortably close to it, to the foreground of our moral consciousness. Often, our moral judgements about the permissible treatment of children turn on the perceived similarity between a contentious case and paradigmatic instances of abuse.

Another fascinating example is provided by the historian Thomas Haskell’s discussion of the role of familiar patterns of attributing moral responsibility for harm in the British abolitionist movement (1985). Haskell argues that part of what prevented sympathetic members of society from taking active steps to alleviate the well-documented suffering of slaves was the fact that they lacked what he calls “recipe knowledge”, practical know-how for intervening in harmful extended chains of cause and effect. Without such know-how, conscientious moral agents are unable to see their own lifestyles as morally implicated in the suffering of distant others. The rise of wage labour system, and the familiarity with complex networks of contractual obligation that accompanied it, provided agents with precisely such recipe knowledge, precipitating the abolitionist movement. Such familiarity provided those troubled by slavery with a model that they could apply to intervene in the suffering of distant others.

A third and final example of how paradigm-infused perception gives shape to our moral practices can be found in Nigel Pleasants’ discussion of the role of perceptions of necessity and naturalness in the persistence of practices that are known to be harmful and exploitative, like slavery (2011). Even if one were to see the victims of a harmful practice as members of a morally significant kind, and even if one were to see oneself and one’s lifestyle as causally implicated in bringing that harm about, one might not be induced to take practical steps to dismantle the practice until and unless one sees a viable alternative to that practice:

It is this ability to point to an available, plausibly superior alternative to a harmful institutionalized practice, that lifts objections to it out of the realm of merely moralistic expression and into that of efficacious radical social criticism. (p. 156)

Our paradigms fix what David Lewis called the “relative modality” (1979, p. 354) of our moral discourse, the range of moral possibilities that we regard as live or serious. Consequently, our paradigm can deem certain interventions in harmful practices as lying beyond that relative modality, thus rendering such harmful practices beyond remediation.

To sum up: the accepted moral kinds, conventions of moral responsibility, perceptions of necessity, and myriad other features of our moral paradigms interweave to give shape and direction to our moral practice.

2.1.3 They are possessed collectively by a community/society of inquiry

According to Kuhn, scientific paradigms ‘are not individual. Rather they are the tested and shared possessions of the members of a successful group’ (2012, p. 190). Indeed, they are what constitute a scientific community as a group. Furthermore, the group, and not any of the individual scientists it contains, is the exclusive arbiter of what is to count as genuine scientific success and what is illegitimate pseudo-science (p. 167). Shared scientific paradigms are what consolidate scientists into a scientific community, and also what constitute legitimate scientific activity as such.

Likewise, a general epistemic world-picture is constituted by ‘agreement not in opinion, but in form of life’ (2009, §241). Rules of inference can be followed, reports of knowledge made, only against a backdrop of shared and collectively maintained customs, usages, and institutions (2009, §199). Shared epistemic world-pictures create the very possibility of epistemic activity; therefore, to be in possession of an epistemic world-picture, and thus to engage in epistemic activity, is a collective affair.

Like a scientific paradigm, a moral paradigm is what makes social life possible and what gives a society its ideological identity (Kober, 1997, p. 376). Furthermore, moral knowledge is ‘the collective achievement of the wider moral society, and I come to share it through membership of, and participation in, that society’ (Pleasants, 2018, p. 588). In sum: shared moral paradigms are what consolidate moral agents into a moral society, and in doing so they create the very possibility of discerning genuine moral argument from moralistic rhetoric.

The point here is not merely that we could not come to have individual knowledge—whether moral or non-moral—in the absence of social institutions of knowledge creation, preservation, and transmission—though this is of course true. It is the more radical claim that both the moral and non-moral knowledge we possess as individuals cannot aptly be described as such in the absence of an epistemic community to deem it so. As Michael Williams puts it, ‘[k]nowledge is an essentially social achievement’ (2007, p. 107).

2.1.4 They are scaffolded by a system of unshakeable commitments

Paradigms are partly constituted by a network of basic commitments that are not themselves objects of inquiry within the paradigm. Kuhn (1997, p. 297f) singles out three particularly important objects of commitment: what he calls symbolic representations, models, and exemplars. Exemplars, as noted, are the archetypal successes of a science by analogy with which all future problems within the science are to be solved. Models ‘are what provide the group with preferred analogies or, when deeply held, with an ontology’ (pp. 297–298). They are (often pictorial) representations of a system, constructed from a fixed set of basic elements, intended to be manipulable by a relatively fixed set of standard methods. Models bridge the gap between exemplars and unsolved problems. Finally, symbolic representations are those scientific expressions that a scientific community regards as neither in need of justification nor liable to empirical refutation (pp. 297–298). In physics, these might include such expressions as “F = ma,” “every action has an equal and opposite reaction,” and “overall change in entropy is always positive.”

An epistemic world-picture, like a scientific paradigm, consists partly in a system of basic commitments that are themselves beyond justification and doubt in most contexts. Some of these commitments express semantic constraints within a language-game—if someone were to doubt the truth of a particular empirical proposition, it would bring into question their grasp of the terms contained within it (Wittgenstein, 2009, e.g., §§81, 82). Some commitments don’t express semantic constraints, but standards of aptitude: if someone were to have certain doubts, we would doubt their competence as an epistemic judge (e.g., §§323, 329, 672). Yet other commitments express neither semantic constraints nor standards of aptitude, but rather capture constitutive rules of the game of epistemic justification and doubt itself: if someone were to have certain doubts, they could not be described as engaging in epistemic inquiry at all (e.g., §§446, 493, 497). This is a non-exhaustive list of the ways an empirical proposition can be beyond doubt, and the ways listed surely overlap and interact with one another; the basic point is that, like the contents of a scientific community’s disciplinary matrix, they are not subject to doubt (or justification) in normal circumstances. They are certainties of our world-picture. Furthermore, that they are not doubted is a precondition of one’s membership of and participation in one’s epistemic community.Footnote 10

Similarly, a moral paradigm is partly constituted by a system of moral certainties. Members of the moral society deploy their moral certainties, in propositional form, without felt need for special justification. Some of these certainties express semantic constraints:

The truth of moral statements, as applied to familiar paradigms [i.e., exemplars], is a matter of meaning: someone who questioned the truth of such a statement would be held not to understand the meanings of the relevant words. (Hanfling, 2003, p. 26)

Following Philippa Foot (1958), we might aptly say that some moral predicates bear an “internal relation” to their objects; within our moral paradigm, it simply does not make sense to use moral terms in certain ways. Other moral certainties express standards of aptitude: if an individual were to doubt (or even to affirm!) the truth of certain moral propositions in normal circumstance, ‘we would have reason to regard him as an incompetent or corrupt moral agent’ (Pleasants, 2009, p. 677). The fact that even affirming a moral certainty in normal contexts is a sign of moral corruption is captured in a nice line from Bernard Williams:

One does not feel easy with the man who in the course of a discussion of how to deal with a business or political rival says, “Of course, we could have them killed, but we should lay that aside right from the beginning.” It should never have come into his hands to be laid aside. (1985, p. 206)

Yet other moral certainties capture constitutive rules of the game of moral justification: ‘they constitute the very possibility of acting morally or talking about moral issues’ (Kober, 1997, p. 373). For instance, the proposition that some others besides yourself possess rights that you are obliged to respect operates as a constitutive rule of moral activity; if you deny it in your reasoning, you are simply not engaging in moral reflection. Again, this list is non-exhaustive; the basic point is that some moral propositions are not subject to doubt (or justification) in normal circumstances, and the fact that they are not doubted is a precondition of one’s membership of and participation in one’s moral society.

2.1.5 They are acquired by training and participation

The features of scientific paradigms listed above are explained by the mode of their acquisition. For Kuhn, scientific communities are ‘[b]ound together by common elements in their education and apprenticeship’ (1977, p. 296). A scientist acquires a paradigm ‘through training as a part of his preparation for group membership’ (2012, p 191).

The same is true of epistemic world-pictures: ‘The massive background of beliefs, recognitional capacities, investigational procedures, etc., which is required for playing the game of justification must be acquired through education and training’ (Williams, 2007, p. 107).

Likewise, moral paradigms are acquired through training and participation.Footnote 11 We learn how to engage in critical moral reflection.

By example (“exemplar”) and application, and by modelling the accredited members of the moral society, who or what counts as a moral patient or agent, and what moral rights, entitlements, duties, responsibilities and obligations pertain to them. (Pleasants, 2018, p. 580)

I think Peter Strawson put it very well when he wrote that.

We should do well to scale down our pretensions to freedom by remembering, if nothing else, the importance of the training we receive and the limited choice we exercise of the moral communities to which we belong. (1961, p. 12)

In this section, I have been working to establish a handful of family resemblances between the concepts of a scientific paradigm, an epistemic world-picture, and a moral paradigm. Having articulated a few such resemblances, I regard myself as entitled to further explicate these concepts along similar lines. In the following section, I will draw on the resemblances between moral and scientific paradigms to articulate precisely the notion of incommensurability that gives rise to the phenomenon of deep disagreement across moral revolutions. In the final section, I will draw on the resemblances between moral paradigms and epistemic world-pictures to respond to some perceived theoretical difficulties caused by the incommensurability of successive moral paradigms.

3 Paradigm shifts, incommensurability, and rational resolvability

Kuhn defines scientific revolutions as ‘those non-cumulative developmental episodes in which an older paradigm is replaced in whole or in part by an incompatible new one’ (2012, p. 92). What exactly is it that is incompatible about successive scientific paradigms? Kuhn identifies three levels of discontinuity: perceptual, conceptual, and practical. If moral inquiry is structured in a paradigm-like way, as I have just suggested, we would expect moral paradigms on either side of a moral revolution to exhibit incompatibility at the same perceptual, conceptual, and practical levels.

3.1 Perceptual incommensurability

Kuhn is perhaps most infamous for writing that ‘the historian of science may be tempted to exclaim that when paradigms change, the world itself changes with them’ (2012, p. 111). After a revolution, scientists ‘are responding to a different world’, one which ‘will seem, here and there, incommensurable with the one [they] inhabited before’ (2012, p. 112); ‘in some areas they see different things, and they see them in different relations one to the other’ (2012, p. 149). Scientists working within a particular paradigm do not experience their own perceptions as theoretically loaded; rather, they see the objects of their scientific attention as simply there, as what the world is really like. This is why it is tempting to say that when scientific paradigms change, the world itself changes with them.

Likewise, when moral paradigms change, moral agents undergo a change of moral perception. They see different things, and they see them in different relations one to the other. Similarly, moral agents occupying a particular moral paradigm do not experience their moral perceptions as organised according to a particular system of moral categories and values; rather, they see the objects of their moral attention as simply mattering, as what is objectively important. As Uriah Kriegel puts it,

the objective purport of moral commitments is phenomenologically manifest, in that what it is like for a subject to have (a conscious manifestation of) a moral commitment often involves a feeling as of homing in on an objective matter of fact. (2015, p. 160)

This explains metaphors of sensory impairment so often drawn upon in the moral revolutions literature: reticent holdouts seem “blind,” “deaf,” or even “numb” to a moral reality that we find as self-evident as other items of perceptual knowledge.Footnote 12

3.2 Conceptual (or semantic) incommensurability

The characterisation of a world-picture as a concept-infused representation of reality is helpful for understanding the connection between perceptual and conceptual incommensurability. The scientist’s world appears different because the conceptual lens through which her world is “seen” is replaced. Kuhn describes scientific revolution as ‘a displacement of the conceptual network through which scientists view the world’ (2012, p. 102). For Kuhn, the meaning of a concept is not specifiable in isolation from the conceptual network of which it is a part. On the Einsteinian revolution, he writes: ‘the whole conceptual web whose strands are space, time, matter, force, and so on, had to be shifted and laid down again on nature whole’ (p. 148).

Some philosophers of moral progress, particularly Michelle Moody-Adams, believe that moral progress can involve only the “deepening” of existing moral concepts ‘such as justice, compassion, or righteousness’ (1999, p. 173), and never the introduction of new concepts. We make progress when we come to more fully appreciate ‘the richness and the range of application of a particular moral concept (or linked set of concepts)’ (p. 169), what Moody-Adams calls its (or their) “semantic depth.” She makes her case on the basis of the psychological claim that new moral insights can be assimilated into a community’s moral practice only if they are continuous with and expressed in terms of moral concepts familiar to that community (1999, pp. 171–175). This psychological claim, in combination with Kant’s dictum that “ought implies can,” entails that the introduction of fundamentally new moral concepts is not a possible mode of moral progress. This suggests a conceptual continuity across moral revolutions: in the post-revolutionary world, we understand the same concept or concepts, only better.Footnote 13

However, if moral inquiry is structured in a paradigm-like way, as I have suggested, the meaning of a moral concept is never fully specifiable in isolation from the conceptual network of which it is a part. We must recognise the possibility that “the whole conceptual web” whose strands are justice, compassion, righteousness, and so on, can “be shifted and laid down again on nature whole.” Thus, conceptual change across moral revolutions is not what we might call vertical, from a shallow to a deep understanding, as it is during periods of “normal” moral inquiry, but lateral, a change in the meaning of the moral concepts themselves. Whether we call these changed concepts “new” or not is a somewhat arbitrary choice. The point of the vertical/lateral distinction is to emphasise that the concepts of the new paradigm, whether we call them new or not, are not the same as those of the old paradigm; the change is discontinuous.

3.3 Practical incommensurability

The connection between the meaning of a moral term and the practice in which it is embedded helpfully connects conceptual incommensurability to practical incommensurability. Kuhn describes successive paradigms as ‘incompatible modes of community life’ (2012, p. 94). He writes that, because paradigms ‘are directed not only to nature but also back upon the science that produced them’, scientific revolutions often necessitate ‘a redefinition of the corresponding science’ (p. 103). Scientific revolutions bring with them changes in the criteria by which scientists differentiate real science from pseudo-science, changes that lead scientists on either side of a revolution making different judgements of scientific success. In short, paradigm shifts introduce new scientific practices supported by new scientific rationales.

Likewise moral paradigm shifts introduce new moral practices supported by new moral rationales. Baker writes that after a moral revolution, a moral community will adopt ‘new customs, rules, and laws that obsolesce or force changes in incompatible customs, rules, and laws associated with the disestablished paradigm’ (2019, p. 43). Elizabeth Anderson, in her edifying discussion of the process by which the British slave trade came to be abolished, makes vividly clear how the standards for judging the morality of a practice were different on either side of the abolitionist revolution as the result of an expansion of the category of morally relevant beings (2014, pp. 17–23). She observes that in the course of the revolution there underwent ‘a revision of the ideal of civilization that incorporated the interests of workers, considered as proper beneficiaries of its fruits and not merely instruments for creating it’ (pp. 23–24). Thus, the standards of success in terms of which the abolitionist “experiment” was framed themselves changed as a result of the experiences of those participating in it. Similarly, Haskell emphasises that the abolition of the slave trade was accompanied by a shift in the conventionally accepted boundaries of moral responsibility. Over the course of the revolution ‘the conventional limits of moral responsibility observed by an influential minority in society expanded to encompass evils that previously had fallen outside anyone's operative sphere of responsibility’ (Haskell, 1985, p. 359). Prior to this shift the claim that members of British society had any remedial responsibility for the suffering of slaves in the colonies would have fallen—so to speak—on deaf ears (and blind eyes). A consequence of the lack of shared evaluative standards across moral revolutions is that arguments for revolution only make sense after the revolution has been initiated (cf. Rorty, 2001, p. 193).

Thinking about moral revolutions by analogy with Kuhn’s picture of scientific revolutions makes it possible to locate the sources of the deep disagreement that afflicts the occupants of successive moral paradigms. They are seeing things differently; they are understanding things differently; and they are doing things differently. The Kuhnian model of deep disagreement across moral revolutions that I am proposing therefore offers pluralistic answers to what Ranalli calls the “constitution question” and the “attitude question” (2021, p. 986) for theories of deep disagreement. The constitution question asks what the objects of deep disagreement are; the attitude question asks what disputants’ attitudes are to these objects. With regards to the former, the Kuhnian model suggests that the objects of deep disagreement are perceptions, conceptions, and practices. With regard to the latter, the model suggests that the disputants’ attitudes to the objects of their deep disagreement are, respectively, stances of certainty, states of grasping, and stances of practical readiness. To share in such attitudes with your community is precisely to occupy a paradigm. It follows that when paradigms change, the result is diachronic deep disagreement.

4 Paradigms, penetrability, persuasion, and progress

This leaves us with three related but separable problems, inherited from both the philosophy of deep disagreements and Kuhn’s philosophy of science: that of the relativism of moral standards; that of the rationality of moral revolutions; and that of the progressive status of moral revolutions. I will now proceed to discuss each in turn, drawing constructively on the established resemblance between scientific paradigms, epistemic world-pictures, and moral paradigms.

4.1 Relativism

Whether deep disagreements motivate epistemological (Hales 2014, 2014; Kusch, 2021; Lavorerio, 2018; Pritchard, 2011) or scientific (Boghossian 2006; Kusch, 2017) relativism has been a topic of some debate in the existing literature on deep disagreements. In what immediately follows, I will draw on the Kuhnian and Wittgensteinian resources made available by the established family resemblance between scientific paradigms, epistemic world-pictures, and moral paradigms to develop a novel response to worries about relativism between successive moral paradigms.

The conceptual, practical, and perceptual changes wrought by scientific revolution seem to rob successive paradigms of the critical friction necessary to assess their relative merits. What we can articulate in support of a paradigm (or world-picture) from within that paradigm (or world-picture) is fixed by that paradigm (or world-picture). Consequently, no reasons for the paradigm can be articulated or understood within the paradigm in a way that doesn’t already presuppose it.

Is this relativism? In one important sense, yes—what is articulable within a paradigm is relative to it, so we have a genuine relativism of articulability. But in another sense, no. Kuhn emphasised that scientists are human beings with personal and professional needs and interests, and science has evolved to serve those needs and interests (e.g., 2012, p. 168). What is considered good scientific practice will inevitably end up being, in part, good practice for scientists. Furthermore, those needs and interests constrain the sorts of scientific revolution that are likely to catch on. Because scientists are human beings with needs and interests, they can be won over if it can be shown that things go better for them if they adopt a new scientific practice. What they cannot do is express a scientific rationale for their change of allegiance.

This is true of our practices of justification in any domain. Even if our articulable justifications are circular, we can still experience life in one circle as going better than life in another. As Wittgenstein puts it:

Language is an instrument. Its concepts are instruments. Now perhaps one thinks that it can make no great difference which concepts we employ. As, after all, it is possible to do physics in feet and inches as well as in metres and centimetres; the difference is merely one of convenience. But even this is not true if, for instance, calculations in some system of measurement demand more time and trouble than we can afford. (2009, §569)

While this passage might give the impression that we ought to think of our concepts instrumentally, this is not the case in general. Some concepts are instrumentally useful only if we do not think of them that way—a practice might serve our purposes only ‘insofar as it is sustained by motives that are autonomous, i.e. not conditional on the practice’s functionality in any given case’ (Queloz, 2021, p. 56). Thus, we ought to steer clear of naïve functionalism or shallow pragmatism about morality (and science, and epistemic inquiry in general (Williams, 2004)). Rather, the point is that our conceptual practices.

must be understood in the light of facts which, while conceptually articulated, are not facts about our conceptual practices, because how we go on and what concepts have a point for us is contingent upon certain root facts about us and our needs in the kind of world we live in. (Queloz & Cueni, 2021, p. 765)

The impression that deep moral disagreement implies moral relativism is partly a symptom of a flawed picture of moral agents as disembodied thinkers without material needs and interests. What is considered good moral practice will inevitably end up being, in part, good practice for moral agents. As Bernard Williams emphasised,

[p]recisely because we are not unencumbered intelligences selecting in principle among all possible outlooks, we can accept that this outlook is ours just because of the history that has made it ours; or, more precisely, has both made us, and made the outlook as something that is ours.’ (2006, p. 193)

Deep disagreement between disembodied thinkers implies relativism—for disembodied thinkers, cogent arguments exhaust the available resources for persuasion. Deep disagreement between real, embodied thinkers, with material needs and interests, does not.

It is also partly a symptom of a flawed picture of our form of life as a functionally unified whole, in which every language-game within it serves a specific purpose and no language-game is at cross-purposes with any other. If this were the case, endogenous reasons for moral revolution would make no sense—pulling at even one thread will cause the whole garment to unravel, and you’ll lose everything. In reality, our form of life is a patchwork of practices, sewn together haphazardly, each patch exerting its own tensions on its neighbours (Queloz & Cueni, 2020, pp. 761–762). We possess numerous paradigms and pictures, moral and non-moral, all at once, all doing different things in different ways, and they may not all have good things to say about one another. For scientific paradigms, this creates the possibility of what Larry Laudan called “worldview difficulties” (1977, pp. 61–64), which arise when a scientific paradigm (a “research tradition” in Laudan’s idiolect) is in tension with scientists’ extra-scientific view(s) of the world. Moral paradigms, I am claiming, suffer from worldview difficulties too. Our patchwork of practices will work better for us—our material needs and interests will be met more fully—when these difficulties are meliorated.

For these reasons, inarticulability of moral reasons is not sufficient for moral relativism. As Michael Williams puts it,

It is one thing to admit that, in any given situation, there may be limits to our powers of argument or persuasion. It is something else entirely to assert that all ways of viewing the world are equally valid, or that all epistemic procedures are equally reliable. No route leads from the first claim to the second. (2007, p. 108)

But this is not to guarantee resolvability. A final Kuhnian thought on relativism: in general, when deep moral disagreements arise, as when a scientist cannot solve an outstanding problem, ‘[o]nly the practitioner is blamed, not his tools’ (Kuhn 1977, p. 273). If you cannot bring someone to see—if your deep disagreement is insuperable—it does not imply that you are on an equal epistemic footing, and so the choice between paradigms is arbitrary—it just means that you have failed to win over your interlocutor. To get someone to see things our way, writes Williams, ‘they would have to be willing to learn. They might not be, in which case we would be stuck’ (2007, p. 109).Footnote 14

4.2 Rationality

The rational resolvability of deep disagreements is one the central issues, if not the central issue, of the deep disagreement literature (e.g., Feldman, 2005; Kappel, 2012; Lynch, 2016; Coliva & Palmera 2020; Ranalli, 2020; Melchior, 2023). Once again drawing on philosophical resources made available by the resemblance between moral paradigm and epistemic world-pictures, I develop an original Wittgensteinian response to worries about the rationality of moral revolutions.

Does the “winning over” of a moral agent to an incommensurable moral perspective preserve their rational integrity? Given the character of paradigm shifts as akin to a change of aspect (Wittgenstein, 2009, p. 206), a scientist’s transfer of allegiance from one paradigm to another appears more like a spiritual conversion than a cognitive choice (Kuhn, 2012, p. 150). Imre Lakatos famously proclaimed (somewhat melodramatically) that ‘[f]or Kuhn, scientific change – from one “paradigm” to another – is a mystical conversion which is not and cannot be governed by rules of reason’ (1978, p. 9).

Is it the case that revolutionary moral change – from one “moral paradigm” to another – is not and cannot be governed by rules of reason? Baker gives a limited role to rational thought in creating space for moral revolutions, but sees it as largely impotent in bringing them about:

when a societally accepted morality is unable to resolve vexing challenges, philosophical critiques can undermine accepted morality, creating conceptual space for alternative moral paradigms. By themselves, these critiques and proposed alternatives seldom suffice to ignite a moral revolution; however, when they are married to emotively charged moments, they are essential to catalyzing a paradigm shift. … Such moments, I argue, enable opinion leaders to peer beyond accepted paradigms and glimpse the world through the lens of an alternative new paradigm. (2019, p. 20).

Something similar is recognised by Pleasants:

moral argument is not enough by itself to get large numbers of people to turn against a harm-inflicting institutionalized practice hosted by their society. This does not mean that moral argument is an inert or epiphenomenal form of activity, but it does indicate that it requires certain social, material, and practical conditions to become effective. (2011, p. 157).

Before a moral revolution, critique of the existing paradigm can exist but lays dormant:

Just as the utterance of a new “word” is merely a meaningless noise or string of letters unless or until it catches on among a body of speakers, so a primordial act of “criticism” directed at a hitherto uncriticized institutionalized practice would be seen as eccentricity or foolishness, not moral criticism. (Pleasants, 2008, p. 100)

What is underscored by the quoted passages is that whether or not a “primordial act of criticism” catches on depends on factors – the conditions, the moment – that are beyond the control of an individual agent. This poses a problem for those who view the integrity of rational agents as dependent on the autonomy of their judgements. It appears that autonomy in judgement, and thus rational integrity, is not preserved across moral revolutions.

A core feature of deep disagreement is that when you find yourself embroiled in one, it is impossible to ‘persuade through epistemically cogent arguments’ (Smith & Lynch, 2021, p. 6). This is a problem only if epistemically cogent arguments are the sole grounds for changing one’s mind while preserving one’s rational integrity. But if there exist modes of persuasion – ways of changing our minds – that do not depend on cogent arguments, but which also do not threaten our integrity as rational agents, then the problem of rationality goes away.

Kuhn likens practitioners of different scientific paradigms to native speakers of different languages between whom communication goes on by unavoidably imperfect translation. He describes paradigm conversion as something like achieving fluency: ‘at some point in the language-learning process they have ceased to translate and begun instead to speak the language like a native. No process quite like choice has occurred, but they are practicing the new theory nonetheless’ (1977, p. 339). Language acquisition can be something that we attempt by choice, but it can also be something that just happens to us. The analogy between acquiring a paradigm and acquiring a language is important for understanding the extent to which moral revolutions threaten our rational integrity.

It is an important theme from On Certainty that just because we hold a commitment without justification does not mean that it is unjustified. Some commitments are in no need of justification in normal contexts. So where do such commitments come from? We aren’t born with them. They can’t have originated from a rational process – otherwise they would be rationally founded beliefs, after all. As Wittgenstein originally put it,

I did not get my picture of the world by satisfying myself of its correctness; nor do I have it because I am satisfied with its correctness. (2008, §94)

In life, as in science, these unshakeable commitments are things that we are simply taught.

If there is sometimes nothing wrong with holding a commitment without articulable justification, then as a matter of logic it must be the case that sometimes there is nothing wrong with forming a commitment without articulable justification; the former demands the latter. This is a diachronic analogue to Wittgenstein’s synchronic third way between being justified and being unjustified. Being taught something needn’t be an irrational way of coming to know something.

I want to suggest that genuine moral revolutions are arational in the sense just described, not irrational. Moral revolutions are not brought about by satisfying ourselves and others of their correctness. When successful, they involve bringing people to seeing, not talking them into believing. One would not know what to say if asked to articulate a justification for seeing something; it is not clear what such a demand amounts to. Moral perceptions ‘are not based on, or amenable to, evidence, reasons, argument, deliberation or interpretation’ (Pleasants, 2018, p. 571). Similarly, in the diachronic case, one would not know what to say if asked to articulate a justification for coming to see something.

My suggestion, then, is: genuine moral revolutions involve forming a commitment without articulable justification, not forming an unjustified commitment. The solution, in other words, is to reject the Kantian ideal of enlightenment as “liberation from tutelage” (Kant, 1996, p. 17). The fact that we must sometimes be taught to see something, and not talked into believing it is not always a threat to rational integrity. Sometimes we must be willing to be taught.

4.3 Progress

Can such a revolutionary moral change aptly be called moral progress? The issue – thus far largely neglected in the literature on moral revolutions – seems compounded by drawing analogies with Kuhn’s philosophy of science. According to Kuhn, that science makes progress is self-evident from the perspective of working scientists. Because scientists within a normal scientific tradition agree on what their problems are and what it takes to solve them, the accumulation of solved problems just is progress for them: ‘How could it possibly be anything else? … the scientific community could view the fruits of its work in no other way’ (p. 161). In this regard, normal scientific progress partly ‘lies simply in the eye of the beholder’ (pp. 161–162). Furthermore, claims Kuhn, scientific revolutions close with ‘a total victory’ for one of the new paradigms: ‘Will that group ever say that the result of its victory has been something less than progress?’ (p. 165). Because scientists within a normal scientific tradition inevitably see their activity as progressive, and because that activity is made possible by the paradigm they inhabit, the emergence of that paradigm is itself inevitably seen, from their perspective, as an instance of progress.

Do we want more than this from the idea of moral progress? Do we need more? These questions require us to reflect on what we want to do with our concept of progress (cf. Haslanger, 2002). We can identify at least two uses to which the concept of moral progress can be put. The first is constructing a forward-facing, action-guiding theory of progress from our past moral successes – to reverse engineer moral progress to enable us to advance forward from where we are now. Philip Kitcher, for example, seeks ‘an account of moral progress capable of helping individuals and societies fashion methods to make such progress more surely and more systematically’ (2021, p. xii). For philosophers of moral progress whose aims align with Kitcher’s, being able to explicitly articulate their reasons for believing that one practice constitutes a moral improvement over another will be very important to them – essential, even.

But this does not exhaust the possible uses of the concept of moral progress. A second is to protect and preserve what is valuable about previous progressive changes – what might aptly be called “memorialising” our past successes, by analogy with the memorialisation of past tragedies.Footnote 15 We might deploy the concept of moral progress to express appreciation of and draw attention to the value of where we are now, in comparison with where we’ve come from. Indeed, coming to appreciate the most stunning examples of moral progress with appropriate reverence contributes to future moral progress as well, by supplementing our action-guiding theory of moral progress with a sense of historical forward momentum, and by instilling in contemporary moral agents a resolute determination to pick up the baton of moral progress.

This necessarily involves some revisionist history, the anachronistic projection of current categories and conventions onto past societies and their practices. The practice of writing history backwards was well-known to Kuhn in the scientific domain:

Partly by selection and partly by distortion, the scientists of earlier ages are implicitly represented as having worked upon the same set of fixed problems and in accordance with the same set of fixed canons that the most recent revolution in scientific theory and method has made seem scientific. (2012, p. 137)

Kuhn was also aware that ‘[s]cientists are not, of course, the only group that tends to see its discipline’s past developing linearly toward its present vantage. The temptation to write history backward is both omnipresent and perennial’ (ibid.). Baker describes how participants in the American abortion debate have succumbed to this temptation (2019, pp. 85–92). Hacking, reflecting on the creation of the conceptual category of child abuse and its projection back onto history, writes of how in the wake of the creation of a new category.

[e]vents in a life can … be seen as events of a new kind, a kind that may not have been conceptualized when the event was experienced or the act performed. What we experienced becomes recollected anew, and thought in terms of that could not have been thought at the time. (1999, p. 130)

‘Every generation writes history afresh. It is to be expected that a new conceptual scheme, like that of child abuse, will be used in rewriting some history’ (p. 155). What once conformed to conventional and legal standards of moral acceptability – Hacking deploys the examples of the explorer Alexander Mackenzie, who at the age of 48 married a 14-year-old girl, and author Carroll, who created and collected nude photographs of young girls (pp. 155–156) – is now reconceptualised as child molestation and paedophilia, despite these categories not being available for Mackenzie and Carroll in their own times.

It is important for us in the present that we understand our moral efforts as continuous with the efforts of those who came before us and as having at least some hope of success. Without this, it would be almost impossible to see the moral dimension of our lives as our participation in something bigger than ourselves – and to reject this way of seeing ourselves is psychologically and practically beyond us. This is how I understand Moody-Adams' claim that the idea of moral progress ‘is a condition of the possibility of morally constructive action for beings like us’ (2017, p. 154). For this reason, things are perfectly in order when we project our current moral concepts – child abuse, for example – onto past agents and say to ourselves “never again.” It does not matter that these concepts came up for grabs at a particular moment in time, long after the relevant actions had taken place. It does not matter that past agents lacked the categories to understand our condemnation of them.Footnote 16

It seems fairly obvious that the concept of moral progress is appropriately put to use in both of these ways. There are different contexts in which its different uses are more appropriate. Articulability of moral reasons plays less of a role when our aim is to retrospectively appreciate what is important about a moral change that is the foundation of our contemporary moral practices. It plays more of a role when our aim is to proactively transform our current moral practices for the better. The significance of articulability will wax and wane depending on the aspect from which we look at moral progress. It therefore seems wise to be flexible in how we think about the connection between articulability of moral reasons and the possibility moral progress.

5 Conclusion

This paper set out to achieve three things. The first of these was to establish some family resemblances between the concepts of a scientific paradigm, an epistemic world-picture, and a moral paradigm. Having argued that these concepts form a family, I took myself to be justified in explicating them along similar lines. Second, I drew on the resemblances of moral paradigms with scientific paradigms to locate the sources of incommensurability between successive moral paradigms and to articulate more fully the problem of deep disagreement across moral revolutions. Third, I drew on the resemblances of moral paradigms with epistemic world-pictures to develop broadly Wittgensteinian responses to three philosophical worries having to do with deep disagreement across moral revolutions, namely, the relativism of moral standards, the irrationality of moral revolutions, the impossibility of moral progress across moral revolutions. I argued that the perception of relativism arises from an erroneous characterisation of moral agents as disembodied thinkers. Recognising that our paradigms need to be compatible with our material needs and interests as embodied beings soothes the relativistic anxiety. I argued that the impression that moral revolutions were irrational resulted from an unrealistic insistence on epistemic autonomy. Recognising that being taught something is not an irrational way of coming to know something makes the worry about irrationality go away. Finally, I argued that the view that moral progress across moral revolutions is impossible is the consequence of too narrow an understanding of how the concept of moral progress can legitimately be put to use. Once we recognise that the concept of moral progress can be deployed to memorialise those historical moral changes that are fundamental to our contemporary moral worldview, or to shore up the moral gains that are of special significance to us in the present, the worry that moral revolutions do not qualify as progress appears misguided.