Trust the Process

1 Introduction

Moral progress happens when society improves in a morally desirable direction. But does it happen at all?

For the longest time, the very idea of progress seemed thoroughly discredited by way of catastrophically definitive historical disconfirmation: “We should probably abandon the nineteenth-century expectation for a steady progress of humanity toward greater and greater overall moral achievement. The wars of the twentieth century extinguished that teleological expectation, and the twenty-first, so far, gives us no reason to revive it.” (Nussbaum 2007, 939) Moral regression, it seemed, was always just around the corner: “And with the element of self-inflicted regression, what we experienced in the twentieth century is revealed as the intrinsic break with civilisation: far from a “relapse into barbarism”, it is the entirely new, henceforth ever present possibility of the moral disintegration of an entire nation that had considered itself “civilised” by the standards of the time” (Habermas 2022, 174). Moral progress is dead, and we killed it.

But over the past ten years or so, the idea of progress has been reanimated. The world is better than it seems, was the common refrain, and if we play our cards right, it may get better still (see https://ourworldindata.org). Regression is always possible, yes; but progress is also possible, and indeed real, and habitual doomsayers have it all wrong.

At first, this renanimation attempt concerned human progress generally, rather than moral progress specifically. Health, technology, poverty, equality – everything, a large body of literature purported to show, seemed to be improving (Pinker 2011, 2018). More recently, an increasing amount of philosophical attention has been devoted to the moral dimension of progress. If society gets better, do our moral norms and values also improve? This has led to a series of contributions to what remains a lively debate (Sauer et al. 2021).

In her most recent book (Jaeggi 2025), Rahel Jaeggi tries to develop an account of this twin pair of concepts, progress and regression. In what follows, I will ask for some clarifications regarding how Jaeggi frames this project. What sets Jaeggi apart from the dominant mainstream of thinking about moral progress are, among many other things, two features: her account is proceduralist and negativist. Progress, Jaeggi holds, is about the how, not the what. It can be found in the way social change happens – its form – not in which social changes happen – its substance. For instance, judgments about moral progress would not be committed to substantive moral judgments about gay rights, but merely to judgments about how certain social changes came about. This is the proceduralist feature. And we identify progress not so much by finding examples of when things go right, but cases where things don’t go wrong, instances not of moral improvements being promoted, but of moral regression being forestalled. This is the negativist feature.

In this paper, I will ask what motivates both of these theoretical decisions and express my suspicion that they are unnecessary. I want to show that the methodological commitments of her account of progress stem from an underlying anxiety that there is something morally objectionable about making judgments about progress in the first place. The suggestion seems to be that verdicts about progress are often themselves morally objectionable: a purported license to say that some people somewhere are irredeemably backward, inferior, condemned to perpetually remain confined to history’s waiting room, unless the white or the Christian or the enlightened man saves them from their self-inflicted state of moral retardation. It is this type of – ultimately unwarranted, I will suggest – epistemic hesitation that motivates Jaeggi’s negativism and proceduralism.

2 Negativism and Proceduralism

Theories of moral progress come in various flavors. Some focus on what moral progress consists in: an expanding circle of moral status (Greene 2013; McCullough 2020; Singer 2011), improvements in well-being (Inglehart 2018), or the emancipation from harmful or unnecessary norms (Buchanan and Powell 2018). Some are about the mechanisms that bring moral progress about, from scientific discoveries, over moral reasoning (Kumar and Campbell 2022) to prestige struggles (Appiah 2011), the evolution of norm compliance and prosociality (Hare 2017), cultural (Henrich 2020) and to institutional evolution (Acemoglu and Robinson 2012; Sauer 2023). Some are about more straightforwardly philosophical questions in metaethics and moral epistemology, such as whether the idea of moral progress is committed to a strong form of moral realism or what the moral epistemological implications of the notion of progress are (Huemer 2016; Sauer 2019).

Critical theory has always had an awkward relationship with the issue of moral progress. On the one hand, describing a given social change as progress seemed to indulge an undue, even problematic, degree of optimism, perhaps even a defense of the status quo – the very thing critical theorists want to avoid. Then again, the normative concern for emancipation characteristic of the Frankfurt School is sometimes fulfilled, for instance when the social recognition of minorities improves. And when this happens, it becomes hard to deny that some sort of moral improvement has occurred.

In her book Progress and Regression, Rahel Jaeggi aims to navigate this tension, and she does an impressive and extremely illuminating job at thinking through how a critical theorist should want to talk about social change for the better. The result is rich, thought-provoking and a much-needed addition to the burgeoning literature on moral change. Jaeggi’s account of progress and regression has something interesting to say about all of the issues mentioned above – does progress happen? How? Why? How much? – but I want to focus on the methodological side of things. Jaeggi holds that the best way to conceptualize moral progress is proceduralist and negativist. What does this mean?

(i) Proceduralism. The task of a non-procedural theory of progress would be to look at society, identify where it changed, how much it changed, identify the subset of social changes that are morally welcome, identify the subset of social change that are morally unwelcome, then tally which subset is bigger. If it’s the former, then there has been, as it were, net moral progress.

Jaeggi feels the temptation of such approaches – everyone does – even if she ultimately rejects them. It is difficult, she admits, to avoid classifying the spread of democracy and the rule of law, the abolition of slavery, the accomplishment of the women’s rights movement, the increasing social inclusion of homosexual people, the stigmatization of violence against children, the eradication of many infectious diseases, the expansion of general education, technological innovation that connects people across the globe and increases economic productivity by several orders of magnitude, as morally desirable social improvements, that is: moral progress.

But she wants to bypass such simplistic verdicts and holds that progress does not happen when people’s lives improve, but that “progress is a self-enriching experiential learning process for finding solutions to problems that are systemically blocked under conditions of regression”. It is “a form of change, or more precisely, a certain way of responding to crises and solving problems” (xii). Progress is not merely an “an improvement, [but] becomes a reflexive enlargement of experience involving an increase in complexity, whereas regression would entail a loss of complexity and falling below a set level of reflexivity” (132). And: “[P]rogress amounts to an increase in experience, a crisis-induced growth in reflexivity” (133). Note that this is not an incidental feature of Jaeggi’s account of progress: she calls this “the core of [her] conception” (132).

(ii) Negativism. The second methodological feature of her account is its negativism. Progress is not about good things happening, but about bad things not happening. It is “the absence of regression” (117). And: “If we want to do justice to the dialectic of progress, the meaning of progress can really only be grasped from the concept of regression or relapse” (117). This means, again, that we should not point to positive improvements in people’s values, practices or behavior to illustrate what progress consist in, but to instances in which regression has been avoided.

In what follows, I want to delve into more detail as to why I believe Jaeggi feels compelled to proceed procedurally and negatively, and I will argue that neither of these desiderata for a theory of moral progress are compelling.

3 Against Negativism and Proceduralism

Methodological negativism is of course a leitmotif in the critical theory of the Frankfurt School that goes back to (at least) Adorno. Adorno already claimed that any positive articulation of a good life and a just society is bound to fail. Human society after the Holocaust and under conditions of capitalism and modern bureaucracy is so fundamentally twisted and damaged that everything would come out all wrong, and any attempt to spell out how life should be would end up wonky, jarring, and perverse. All we can aspire to is the negativistic ‘denunciation of the inhumane.’

I always found that there is a fair amount of coquetry in such proclamations, the intellectual equivalent of a young girl telling her group of friends that she knows a big secret but isn’t going to tell anyone about it to excite their interest. In a somewhat similar vein, Habermas had little understanding for what he referred to as Foucault’s ‘crypto normativism,’ which he considered mere melodramatic pretense (Huffer 2013). Of course Foucault’s examination of power structures was normatively charged, and it is unhelpful and dishonest to cash in on the applause for ones tacitly avowed moral ideals whilst publicly disavowing them. And of course Adorno had more to say about right and wrong than his ‘No more Auschwitz’-minimalism. He was just being coy about it. I will argue that crypto normativism and negativism, while different, spring from the same source: an – albeit unwarranted – epistemic anxiety about revealing, and sticking to, one’s normative commitments.

Let me be clear that Jaeggi laudably tries to avoid these traps. But this is because in actual fact, Jaeggi’s discussion of progressive social change isn’t really negativistic at all, as her analysis of female emancipation shows (141ff.). There is a positive vision here, which is that progress, as far as women’s rights are concerned, happens when women become politically, socially and economically more free and equal. This process, as she demonstrates, has been slow, imperfect, interrupted by setbacks, and unfinished, but it is not clear how this has any bearing on whether or not progress has indeed occurred. It has occurred, as Jaeggi’s own discussion shows, even if the transition from medieval/feudal gender roles to a bourgeois gender regime she describes hasn’t been an unequivocal improvement in every possible way, which is true but neither here nor there.

What about the formality requirement? First of all, many of the paradigm cases of moral progress just are not plausibly characterized as formal at all. It isn’t a formal improvement when infant mortality falls from 40 to close to zero percent or when we no longer accept raping someone else’s daughter as appropriate compensation for the rape of my daughter or when gay people can start to live freely and without fear of discrimination or violence. And our criteria for identifying such things as progress aren’t formal ones, either, but based on substantive, first-order normative judgment such as “it is bad when children die prematurely” and “consensual sexual acts between adults are not objectionable”. We judge these examples to be substantive instances of moral improvement, not “self-enriching learning processes” or mere “increases in reflexivity”.

Moreover, better problem-solving due to enriched learning seems neither necessary nor sufficient for moral progress. Not sufficient, because the Wannsee-Conference clearly satisfies the description of experiential learning being put to use for solving the problem of, in this case, how best to murder millions of people; and not necessary, because when Thalydomid is banned or gay marriage is legally recognized, no change or enrichment in our problem-solving capacities that have previously been blocked has occurred. Things have simply become better because people have made them better, or an existing solution (marriage) has been made accessible to a larger number of people who were previously barred from it for no good reason.

4 Epistemic Anxiety

What explains this desideratum that an account of moral progress be formal? The answer, I believe, lies in a certain kind of epistemic anxiety that is fueled by the sneaking suspicion that there is something morally objectionable about identifying instances of moral progress. For when we say that this or that development or this or that instance of social change has been progress, are we not always implying that other places and peoples who haven’t made said development (yet) are therefore less progressed, indeed less morally developed?

This anxiety has a point, and it is to a considerable extent psychologically understandable. Historically speaking, judgments about progress have all too often been wielded to condescend to people or even to provide a pretext for colonization, invasion, violence, and eradication.

Then again, judgments about progress and the degree to which an individual or group still has some progress ahead of them do not logically require such abuse. When we consider how well our children can read in 2nd grade, we correctly judge their reading abilities to be inferior to ours. But we do not violently invade their schools to accelerate their development, nor would this be a defensible idea, and we do not, in expressing such a comparative assessment of their reading abilities, behave in any way disrespectfully. We are merely saying that their reading abilities are not as good as those of us adults, which is in fact true and perfectly fine.

Some authors feel less apprehensive and take the opposite route, showing up – guns blazing – to reveal their moral convictions in the plain light of day. Buchanan and Powell, for instance, state that their account for the progressive nature of demoralization simply assumes the correctness of some form of liberal modern morality: “We will take it for granted, because we are assuming a broadly liberal moral perspective, that these are all cases of proper de-moralization – that, at least from a secular liberal point of view, beliefs that these behaviors are morally wrong per se or that they warrant institutionalized punishment were unjustified, and that coming to realize the falsity of these beliefs is an instance of moral progress.” (Buchanan and Powell 2018, 111) Getting rid of what they call “surplus moral constraints”, then, unsurprisingly comes out as morally progressive. I happen to agree, but I also appreciate the sense of dissatisfaction that may be felt by the part of the audience that isn’t already convinced of this basic normative outlook of secular liberalism.

Buchanan and Powell take no moral prisoners, as it were, while Jaeggi’s account wants to avoid such brazenness. But she overcorrects into her own brand of cryptonormativism: by refusing to articulate, other than ‘formally’ or ‘procedurally’ or ‘negativistically’, what her normative convictions consist in, she makes it hard for any part of the audience to agree with her because we are left in the dark as to what there is to be agreed or disagreed with. But such attempts to please everyone by pleasing no one at all must fail, because as moral and political philosophers, we ain’t in the prisoner-taking-business; we are in the judging-people-business.

The hope that we can avoid judging the bad guys for fear of being mean to them is futile, because whenever we deploy any kind of normative standard whatsoever, some people will end up falling short of it. Jaeggi says that many models of progress entail an “untenable hierarchy” of different stages of social change. (The German original even says “unbearable”, unerträglich): “Colonial relations of violence and exploitation are thus paternalistically justified, and “underdeveloped” or “developing” nations are goaded toward their own happiness through domination and oppression” (21).

But this is not what it means to make normative judgments. It is one thing to say that X is better than Y, and a separate thing to justify colonial violence inflicted by X on Y. But more importantly, there is no reason why in principle, negativist/proceduralist judgments about progress could not be used to justify colonial violence as well.

Even if we go formal, and seemingly avoid saying unkind things such as individual, institution or culture X is better than Y therefore Y is worse than X, instead saying things such as “X has undergone an experiential learning process for finding solutions to problems that were previously blocked”, what we are committed to is saying that any Y that hasn’t undergone a similar learning process has therefore progressed less. And even if we go negativistic, and only identify ways in which societies function the way they ought not to function, some societies will attract a greater number of such negativistic condemnations than others, and will thus count as worse, because they fall short of this standard of preventing regression. If we are negatively entitled to at least “denounce the inhumane”, to use Adorno’s expression again, how is this different, in terms of justifying colonial relations of violence and exploitation, than using positive standards of moral evaluation?

The aforementioned anxiety is thus at least partly fitting because it is indeed true that normative judgments entail that some people or places or eras are worse than others. If we substantively think that beating children and raping spouses and enslaving workers and whipping peasants and torturing prisoners and banning free speech and lynching gay people are morally worse than doing the opposite, then societies that do such things more often or condemn such things less harshly are morally worse societies. But we ought to think this, because they are.

It is, of course, a rather good idea both epistemically and morally to block the inference from “group X is in a morally worse state of development” to “we need to violently intervene in this situation to help group X upgrade to our own morally more enlightened state”. But the reason for this is not that the original normative verdict is unjustified. It is that pragmatically, moral interventions tend to start with polite words and good intentions and tend to end with people looking down the barrel of a gun. And secondly, that a group of people or region of the world is in a morally worse state does not entail that the people belonging to said group or living in said region are inherently morally worse people. Far from it, because it is almost never the case that the moral state of development at the social level is determined by the moral state of development reached by the people in that society at the individual level. It is – as any good Marxist knows – almost always the other way around, and people who live under morally regressive conditions such as oppressive theocracies or apartheid regimes are usually trapped under those circumstances by collective action problems or the fact that the processes of supraindividual cumulative cultural evolution that determine the moral infrastructure of societies everywhere simply hasn’t reached a more agreeable – more just, more equal, more prosperous – state yet.

Epistemic anxiety regarding our judgments about progress is often expressed by saying that we should not ‘project’ our own norms and values and preferences onto the past, lest we engage in some illicit form of doxastic imperialism (6). We were born in this time and place, so the values that are attached to this time and place seem best to us; but, the thought goes, had we been born in some other time and place, the values that were attached to that spatiotemporal arrangement would have seemed best to us, just as our values do to us now: “[H]ow can it be explained that corporal punishment for children, which until fairly recently was regarded as an unobjectionable and even salutary disciplinary measure (‘spare the rod and spoil the child’), is now frowned upon and has itself become a punishable offense?” (38) So how do we know that we are not the backward place, rather than them over there? What breaks the tie?

This is essentially a normal skeptical argument from the subjective indistinguishability between knowledge and error to the suspension of judgment – ignoramus et ignorabimus. Even if we were a brain in a vat, or stuck in some other type of deceptive contraption, we would not know and could not rule it out. So, it turns out that we do not know anything at all. Likewise, if the trajectory of history had washed us up on the shore of moral error rather than knowledge, we would not know and, again, could not rule it out. If we had been born centuries ago, we would believe in human sacrifice just as we now believe in human rights; if we had been born in Imperial China, we would see the point in foot binding just as much as we – now and here – consider it a bizarre and fatuous torture. But either way, the strength of our conviction would be the same. So, it turns out that we have no moral knowledge at all by which to judge other times and places.

This argument is as common as it is mistaken (Sauer 2023). There are two main reasons for this: firstly, the alleged arbitrariness in the direction of change between different historical episodes that is supposed to make the switch between their and our time and place symmetrical doesn’t exist. And secondly, the parties to the disagreement – that other culture or historical era and what it believed – aren’t nearly as monolithic and homogenous as we are inclined to assume.

Why are we not similarly perturbed by scientific disagreement, which is at least as prevalent and radical as moral diversity? Sure, from the perspective of 21st century science, witchcraft and the evil eye seem like silly superstitions; but from the perspective of those who believe in the former, our body of scientific beliefs seems just as silly, with all its invisible forces that violate the laws of common sense. The reason, however, why we do not consider such non-moral disagreements epistemically damaging to modern physics is that in reality, the latter grew and developed out of the former. The direction of change is not actually arbitrary but reflects a learning process. People who want to sedate society back into its dogmatic slumber sometimes suggest that, when it comes to evolutionary biology or ‘intelligent design theory’, we should ‘teach the controversy’. But this argument pretends that this controversy hasn’t already happened. Evolutionary biology wasn’t foisted upon society. It was arrived at, nolens volens, via empirical evidence and theoretical reflection, driven by the internal contradictions and disconfirmed predictions of the theistic framework. We don’t teach the controversy because the debate is over, and one side won. It’s just that the losing side is frustrated with this outcome and is thus asking for a rematch.

Moral change is – not always, but often – analogous. We believe that slavery is wrong, but back in the day, they believed it was fine, and if we had been born then, we would have also believed that it was fine. So, any attempt to privilege one side of this disagreement can only illegitimately privilege one point of view. But this is not what happened. In reality, the disagreement is not symmetrical because, as in the case of scientific development, progressive gains are made from within a social context, and reflect struggles, learning processes, or trajectories of institutional evolution undergone by those contexts. We can subject these kinds of disagreements to a kind of Millian test: of those who tried both, who prefers which? I am not aware of any society that used to practice slavery and, after a while, wants to reintroduce it, and the same holds for human sacrifice, witch burning, or spousal rape. Many intercultural moral disagreements do not proceed from arbitrary differences in perspectives: they are more like one person who has already tried crystal meth telling another person who is about to try it that it’s a bad idea. Now Jaeggi is of course right that this process involves experiential learning. But what makes it a case of progress is not in virtue of this fact: it is in virtue of the fact that one has moved from one state of affairs – taking crystal meth – to another – not taking crystal meth – with the latter being substantively better than the former; and it would have been substantively better even if it had not been brought about by experiential learning, but by force, say through coercion or involuntary confinement.

Secondly, it is rarely if ever true that the alleged disagreement between ‘us now’, who believe X, and ‘them then’, who believed non-X, even exists to begin with. It is often said that we think slavery is wrong, but 200 years ago in the American South, they thought it was fine. But who are they? Surely, the extension of this ‘they’ doesn’t include the many, many people from the American South who didn’t think slavery was fine and wanted to see it abolished; nor does it, for very obvious reasons, include the slaves themselves. When it comes to the kinds of radical moral disagreements that are used to vindicate a kind of decolonial suspension of judgment, a similar pattern almost always emerges: slavery, female genital mutilation, child labor, foot binding, human sacrifice, racial segregation, lewd remarks against women – who thought those were fine? Almost without fail, we find that it is not a monolithic foreign culture as a whole that approved of these, but rather small elite groups that permitted themselves to hold slaves or roast children over the fire to placate the weather god (Edgerton 1992). There is a, usually rather substantive, part of societies in other places and times with which no disquieting, progress-undermining moral disagreements exist at all.

5 The Arc of History

Before we think that there is some kind of irreconcilable moral disagreement between cultures, we should first subject the origins of such disagreement to genealogical ideology critique – a point which, by the way, should be appealing to those working in the tradition of Marx.

Again: epistemic anxiety surrounding judgments about moral progress is largely unjustified. It may feel somewhat mean to judge people and places in terms of who’s better and worse, but if we want to talk about moral progress at all, this is inevitably what we will be doing.

Jaeggi wants to prevent the normative standards we use to make judgments of progress to be deployed in the service of judging others:

Progress, as I will conceptualize it throughout this study, has nothing to do with a smugly self-congratulatory Whig history of Western imperialist societies, nor is regression the paternalistic verdict on those deemed to have been left behind by Western modernity. Rather, the progress/regression binary is primarily the conceptual vehicle for critique and self-critique of the very societies that pride themselves on their supposed progressiveness (6).

Normative standards are a vehicle of self-critique.

But why would this be so? What prevents those normative standards from spilling over into foreign territory? If I complain about my backhand stroke because of some flaw in my technique, anyone else whose backhand suffers from the same flaw is subject to the same kind of evaluation. And if my backhand stroke is better than yours, then saying so is not necessarily ‘smug’ of ‘self-congratulatory’. Moreover, the admirable wish not to engage in paternalistic judgments about other cultures carries its own risk, because if it is not supposed to be possible for ‘Western modernity’ to morally criticize non-Western non-modernity, then by the same token it is not possible for the latter to criticize the former. If ‘we’ don’t get to criticize them, they don’t get to criticize us, either. Moral relativism cuts both ways.

Behind many of these qualms, I believe, is the fundamental aversion to any kind of teleological indulgence, which only a negativist and proceduralist account of progress can supposedly immunize us against. According to Jaeggi, traditionally there have been four ways of talking about moral progress that we should avoid, which are that progress consists of an unbreakable chain, that it is irresistible, developmental, and cumulative (15). The first is the claim that progress is entangled: medical, technological, social, and moral progress ‘unbreakably’ go together. The second is that moral progress is bound to happen. The third is that it follows a developmental logic with preset stages, such that some cultures or places can be ‘behind’. The fourth is that progress is cumulative, i.e. linear and loss-free. (There are actually a few more possible aspects of teleological progress, such as that it results in exactly one possible end state of moral perfection. (see Sauer 2023))

But a negativistic and proceduralist account of progress is not guaranteed to help us avoid these mistakes, and there are lots of non-negativistic and non-proceduralist accounts of progress with none of the features above (Buchanan and Powell 2018; Kumar and Campbell 2022; Sauer 2023).

Conversely, negativistic and/or proceduralist accounts of progress do not help us avoid those features. Remember that according to Jaeggi, “progress is a self-enriching experiential learning process for finding solutions to problems that are systemically blocked under conditions of regression”. This type of social change could clearly be unbreakably entangled along various dimensions of progress, it could be guaranteed by irresistible forces, it could be cumulative and it could be linear. Just imagine a process of self-enriching learning moving away from regression that suffers no setbacks, is guided by a process that reliably leads to improvements, enhances other forms of progress and is enhanced by other forms of progress in return, and which follows a sequence of identifiable stages.

Morever, non-negativistic and non-proceduralist accounts of progress need not have any of those features either. Suppose that progress consists in society functioning more freely and more equally for a greater number of people. This is a substantive idea of progress. But this idea is easily compatible with saying that there are no developmental stages leading in this direction, that while society is changing in this direction, many disappointments and setbacks happen, gains are accompanied by losses (greater individual freedom can lead to the erosion of community), that this progress happens without any unbreakable link between it and other types of progress (sometimes technology or medicine advances in ways that are bad for society) and that it is not irresistible (we could be struck by an asteroid or a fascist takeover that reverses these gains).

6 Conclusion

To me, these are good news. The epistemic anxiety that we may wish to placate by opting for negativism and proceduralism is unjustified to begin with. As always when it comes to ethics and politics, epistemic caution is recommended; but there is a level of caution that is incompatible with pursuing the project of ethics at all.