A troubled inheritance: Overcoming the temporality problem in cases of historical injustice

What are historical injustices? They are not merely injustices of historical significance, such as the trial of Captain Dreyfus. Instead, they are historical in the sense that they occurred in the remote past such that many, if not all, of those directly concerned can neither be brought to justice nor given justice. But historical injustices are also not irrelevant to the present. Indeed, to give a positive account of the relationship between historical injustices and the present is the point of the paper.

Some commonly agreed restrictions are worth mentioning from the outset. Historical injustices do not typically represent unsolved common law crimes or ancient wrongful convictions like the case of Captain Dreyfus. Instead, the paradigmatic cases of historical injustices refer to crimes committed by one (or more) groups against other groups, for example, the forced displacement of the Sámi by representatives of the Nordic states, the Spanish colonization of Latin America, or chattel slavery in the United States of America (Nuti, 2019). Moreover, not all groups seem to matter to us. We discuss and care about cases of injustice that concern groups that have some connection to the living (we hardly debate the crimes or suffering of the Carthaginians or the Hittites). Thus, historical injustices are not reducible to academic historical debates about the dead; they are normative debates for the living. Faced with a troubled past, we ask what we should think, feel, and most importantly do in the present.

So, when we speak of historical injustices we are referring to acts or events, between groups, that occurred in the sufficiently remote past, such that many of its participants are beyond punishment or reparations. This temporal dimension, referred to by Stark recently as the “temporality problem” (2024) of historical injustices, raises a unique challenge. Whether or not the passage of time lessens the duty to repair (see Sher, 1981; Spinner-Halev, 2007; Waldron, 1992), the fact that wrongdoers and the wronged are no longer with us certainly complicates answers to the question of who owes what to whom?—especially, if are trying to avoid visiting the sins of the parents on their children.

This paper aims to address the relationship between the past and the present in the case of historical injustice. We argue that the right account of historical injustice must explain the temporal dimension and relation between groups of the past and the present. To this end, we consider three accounts: the enduring or structural account, the institutional liability account, and the national community account. Due to their shortcomings, we present a novel account of inherited agency based on social learning.

Many historical injustices have ongoing effects on the living. Can this feature explain how the present is linked to the past? This has indeed become a popular way to approach historical injustices. We find such views coming from two directions: first from accounts of structural injustices and forward-looking responsibility inspired by the work of Iris Marion Young (Lu, 2017; Nuti, 2019; Spinner-Halev, 2007; Young, 2006, 2011) and from the legal theory on reparations that models historical injustices on class-action lawsuits (Magee, 1993; Matsuda, 1987). Differences aside, these views arrive at much of the same relation of the past to the present via the relative benefits and harms conferred from the past to the present.

Thus, both structural and enduring accounts of historical justice link the past with the present through those who suffer or benefit from said injustice. Such analyses are reminiscent of class-action lawsuits for events that occurred in the distant past. Imagine a chemical spill that occurred near a small town several decades ago. The current owners of the plant, which we can stipulate are not harmed by the spill, owe the present inhabitants of the affected town reparations. That is because the owners of the plant are profiting from its activity and the fact that they never paid to clean up the spill. Present inhabitants of the town are suffering from the ongoing effects of the spill, quite independently of their relation to the original inhabitants. This logic is easy to identify in the arguments for repairing the wrong of slavery in the United States of America (Magee, 1993; Matsuda, 1987). On the class-action view, living generations or populations should address the injustice of slavery because it continues to affect the living where remote historical events continue to have disparate effects (see also Allen Jr. & Chrisman, 2001; Coates, 2014).

Undoubtedly, all these views correctly identify one reason why we care about historical injustices. Injustices like chattel slavery, artificial famines, deportations, and colonization still impact present populations. These downstream effects (economic, political, psychological, or social) can advantage or disadvantage the living. Nevertheless, because these accounts focus on the effects and legacy of historical injustices over the original injustices themselves, they face two challenges.

First, by focusing on present suffering and advantage, we lose sight of the distinctiveness of historical injustices. Present reactions toward the historical injustices are chiefly important because of their content. Those who are aggrieved about the past are grief-stricken, angry, or heartbroken about what happened in the past. There is a difference between members of the Sámi community being indignant about a historical injustice, like being subjected to racial hygiene studies, and a contemporary injustice like the exploitation of their reindeer-herding land for mining. In the former case, the emotions are about what was done to their ancestors.

Relatedly, if the source of a grievance is truly a historical injustice, any recognition and repair of that injustice should not primarily be about attending to the present suffering. But to address a historical injustice we should chiefly address what's at the root of the suffering or reactive attitudes. Otherwise, we are merely offering a form of public cognitive behavioral therapy to cope with the aftereffects of the injustice. This means that an approach to historical injustice that reduces it to a structural or enduring injustice loses its proper focus. As Susan Stark puts it “the obligation to repair historical injustices is rooted in the occurrence of the past injustice, not solely in its expression in current structures, as structural views allege” (2024, 2). Indeed, this explains why living members of a group, at times quite well-off, may still seek official recognition of what was done to their ancestors (Pasternak, 2021, see also Kumar, 2014).

Second, structural and class-action views are also limited in scope: they can only address historical injustices if we can also identify living parties that are benefited or harmed. Yet, it is not a conceptual truth that historical injustices must produce these two kinds of effects. Imagine a colonial empire that eradicates a small nation in the Pacific. Now there is no living population that is poorer, suffering from lack of recognition or lower social status since the victim community has been eradicated. Whether the colonists stay or go, there is no heir population to suffer the lingering effects of the injustice. Or imagine that the Israelites completed the genocide commanded by God against the Amalekites (Carroll & Prickett, 1998, 1 Samuel 15:3). In such cases, there is no disadvantaged population that can be made whole. Class-action and structural views presume that there is always a plaintiff and a defendant, but what if there is only the latter?

These accounts also struggle to capture plausible cases with unexpected consequences. Consider a variation on the above case where the chief effect of eradicating a small Pacific nation is to severely impoverish their former trading partners. On such a view, instead of responding to the annihilation of a people, we need to focus on the impoverishment of their former trading partners. If what makes present generations respond to past injustices are ongoing consequences, then the poverty of former trading partners should concern the living whereas the actual genocide should not. Worse, it is not obvious who will be worse off due to a historical injustice. The present war in Ukraine might make the Russians worse off in the long run.

We neither claim that injustices rarely produce enduring effects in the present day nor that these are irrelevant. Historical injustices that have ongoing effects may be those that have the strongest claims upon us. However, do argue that we cannot reduce all historical injustices to this subset.

We note that Iris Marion Young's account of structural injustices may escape these objections because she was skeptical about relying on any links between the advantaged and disadvantaged groups in a structural injustice to ground relations of responsibility. Instead, she suggests a social connection model of responsibility where responsibility is adjudicated in terms of social or institutional roles within a structural injustice (2006, 119ff.). This leads us to examine accounts that focus on institutions to explain the responsibility for historical injustices.

Another way to link the past and the present in a historical injustice is by showing that there is a single “person” responsible for the historical injustice that can be brought to justice (Boonin, 2011; Fullinwider, 2000, 2004; Pasternak, 2021; Thompson, 2006). These views tend to focus on state-like institutions, whose legal status and liability exist are distinct from the status and liability of its leaders and members. Leaders and agents of states come and go, but the state, as a continuous legal person, is bound by past agreements. The liability of states mirrors the corporate liability of institutions like firms.

Let us examine a concrete case. During the 19th and 20th centuries, the Sámi were severely mistreated by state actors—scientists, courts, etc.—in several Nordic countries. These states sought to enforce their borders, provide land to settlers, and drive economic development. Therefore, they forcibly displaced many Sámi, whose livelihood and way of life centered on herding reindeer, who migrate between the Northern countries (Labba, 2020). In addition, the Sámi were subjected to so-called “racial hygiene studies”; they were humiliated and denigrated by researchers investigating racial mixing (Hagerman, 2015).

To link past and present, these accounts identify political institutions as the culprit, for example, the Norwegian or Swedish state. Swedish citizens may not have been alive when this occurred, but they must repair these injustices because they benefit from and partake in the Swedish state. Janna Thompson puts it as follows: “Correcting and repairing injustices is part of what it means to support morally reliable institutions and they have this duty even if they in no way contributed to the wrong” (2006, 263). Citizens bear no personal guilt or responsibility, but until the injustice is (sufficiently) repaired they must do their part to help the state discharge its duty of repair. Many will have noticed that this is very much the reasoning behind war reparations: the party responsible for damages is the state though this means that its citizens pay through taxation long after the generations that fought the war have passed.

When it must identify an aggrieved party that is continuous with the past aggrieved group, the institutional liability views no longer look for an institution. Often enough, there simply is no state or institutional actor. The injustices perpetrated against the Sámi did not occur between states since the Sámi did not have one. Nor have Black Americans had a state. For Fullinwider, an advocate of this view, it is sufficient that there is some population who have been adversely affected by the misdeeds of a legal person. We can reliably approximate who they are: “Because the effects of […] racial oppression have been dispersed so widely throughout the African-American community, it makes sense to adopt some scheme of reparations that morally approximates rather than actually effects the restoration of victims to their ‘rightful places’ […]” (2000, 6).

Thus, institutional liability appears to be a hybrid view. On the one hand, it says that a continuous political institution such as a state is the party liable to repair historical injustices. On the other hand, when we turn to the aggrieved party, much of the same reasoning as we saw in the structural injustice or class-action views is employed. Thus, the view cannot avoid (some) objections directed at structural or enduring injustice accounts.

Yet, the worst obstacle for the view is precisely its reliance on state institutions as the source of responsibility for historical injustice. On this account, we can only establish a bridge between the past and present by identifying some continuous legal person that can be held accountable. Yet, this view quickly runs into trouble whenever there is no legal institution or a discontinuous one. Consider the establishment of the state of Israel in 1948. True, state liability views allow Palestinians to make claims reaching back several decades regardless of how well they are doing, but the conflict between Jews and Arabs did not start on the day of the foundation, May 14th, 1948. What about even older injustices? There is, for example, good evidence that before May 1948, Jewish paramilitary groups committed crimes against the Arab population. On the institutional liability view, Israelis can only be held responsible for the actions of a continuous institution—their state. This betrays an artificial misalignment between a community's history and its responsibility.¹ Analogously, few would be impressed by future Palestinians denying all responsibility for crimes committed against the Jewish people on the basis that they occurred before the establishment of a Palestinian state. In sum, the history of the responsible and aggrieved communities involved in historical injustice seems to come apart from the history of their states, and when they do, the institutional liability view struggles.

In the absence of a continuous legal person such as a state, these views stumble. Consider the American case. Whether the “Republic” began in 1776 or 1783, it did not exist for the first Thanksgiving in 1621. Thus, years during which American ancestors may have mistreated slaves or indigenous populations are left out. As Kevin Bruyneel (2021) has shown, it is dangerous to make irrelevant the politically relevant by constructing historical narratives to fit the settler memory and identities. Focusing on institutional responsibility alone while neglecting the responsibilities of groups and individuals that are not directly linked to these institutions seems likely to do so.

If our only way to hold present generations responsible is to identify some legal person such as a state, then it appears that the view is underinclusive. It fails to include real or plausible cases where there is no state or it is discontinuous. And yet it seems that we do not have good reasons to restrict the idea of historical injustices and the possibilities of repair to the existence of states or other legal political institutions.

We have been searching for some way of relating the past with the present within two groups: those owed repair and those who, in some sense, owe it. As we have seen those who think that membership in purely institutional terms or in terms of being the recipient of certain benefits or harms connected to structural or enduring injustices struggle to account for the distinctive and often noninstitutional nature of historical injustices. One upshot seems to be that we desire a sense in which membership in aggrieved or responsible communities neither depends upon the present distribution of harms and benefits nor the continuous presence of states.

This is what membership in a national community view of collective responsibility does (Abdel-Nour, 2003; Butt, 2008; Miller, 2007). On such views, we should focus on grounding the relation between the present and the past and the obligation of repair in the responsibilities of a distinct kind of intergenerational community: the nation. For this view, it is membership in a pre-political community itself that is the legitimate source of collective responsibility. Historical injustices on such views can occur between different national communities (which may or may not have statehood), but they can also occur within a community such as in the cases where one part of the population targets fellow members based on features like disability or gender (cf. Nuti, 2019). Indeed, it is especially in the absence of a state, or when seeking to break away or free from an existing one, that people appeal to their membership in pre-political forms of community (Gat, 2012).²

The challenge for such views is to show that it is truly membership in a nation or pre-political community, rather than the liability attached to membership in a formal state—that explains how collective responsibility emerges and carries over across generations. Like advocates of institutional liability views, defenders of the national community view do not hold living nationals morally culpable or guilty for the misdeeds of their ancestors. Their claim is however that the community as a whole is responsible for ensuring that the nation does what it should or must. The challenge, as they understand it, is to explain why individual members of the nation alive today are responsible for what past members did. According to Daniel Butt, this implies two main tasks: the first is to understand how the responsibility can even be collective (Butt, 2008, Ch. 4 & 5); the other is to understand the basis for assigning membership to intergenerational communities (Butt, 2008, Ch. 6).

How can nations be collectively responsible? For Miller (2007, Ch. 5) it comes down to two features of nations. On the one hand, members of a group share collective responsibility because their whole ability to act cohesively as one rests on their members' like-mindedness. Shared action requires shared beliefs, expectations, and values—a kind of expressive and mental ecosystem. A member who shares in this ecosystem then also shares in the responsibility for its outcomes. To be clear, appealing to like-mindedness draws on our communitarian intuitions about shared identity. On the other hand, according to Miller collective responsibility can follow from participation in a fair (enough) cooperative practice. Note that participation is distinct and independent from identity. If one has partaken in a cooperative practice and shared in its benefits, then one cannot refuse to share in its burdens. Just as members of a cooperative firm may draw benefits from their participation, regardless of their attitudes or beliefs, they also must share in wage cuts when the firm is struggling.

For Miller, we can collectivize responsibility by appealing to either the like-mindedness model or the participation model, or both. Once we have established that nationals, and not only citizens, share in collective responsibility we can argue that they inherited the burden of responsibility from the past because this is the flip side of accepting the benefits conferred by past generations.

Similarly, Abdel-Nour (2003) argues that communities of identification and solidarity determine membership in a group or nation. If people identify with their forebears and take pride in their deeds, this implies a level of corresponding national responsibility for deeds done in this collective's name. Again, this is a kind of symmetry argument; if you identify with your ancestors' glorious deeds, you must also identify with their inglorious ones.

One problem for the national community view concerns communities that do not present the kind of symmetry of benefits and burdens envisioned by its advocates. This makes it unclear whether they would indeed transfer the right kind of responsibility across generations. In 1917, the Irish did not have a sovereign state. The land—a tangible benefit identified by Miller—is not under their control. The Irish may reap some benefits from inhabiting the land, but they do not control the land like the English control England. The Irish are like tenants in their homeland. Now, if we must accept the burden of responsibility for past generations because we accept the benefits conferred by them, we are faced with an issue. The Irish do not enjoy the same tangible good as the English do—the former lack proper control enjoyed by the latter. Not only have these nations inherited less of a tangible good such as land, but they might also have inherited little capital. Because many nations seem to have inherited few goods or to have inherited them in a lesser form it becomes unclear why living generations should accept responsibility for a possibly what might be a very long list of misdeeds. If this is a symmetry argument, why accept the full weight of the past if the rewards are underwhelming or scarce?

Moreover, although the national community view is distinctive from the institutional liability view since it admits that the temporal links of past and present communities need not go via statehood or any other institution, this means it must provide its normative links in another way. But are shared individual features such as like-mindedness, identification, and solidarity truly capable of generating the right kind of normative links that form a basis for duties of repair, let alone generate commitments of group complaints? Butt adds an overlapping generations model where responsibility is transferred over generations because of the moral connections between members at any given time (2008, 184). While this may be a noninstitutional version of the corporate liability model, it is not clear why a generational overlap should provide moral continuity per se.

Thus, it is easy to be skeptical about how reliable and normatively important cross-generational overlap and ties among groups such as national communities are. We might also wonder to what extent like-mindedness or cooperative practice are parasitic on things like state systems of education or public institutions. We thus turn to the next section where we argue that what truly matters is that members of national communities inherit a valuable form of group agency through practices of social learning.

Our solution to the temporal problem lies in the distinctive inherited and collective agency that emerges within lineages of social learning. This normatively relevant link applies equally to communities that demand repair and those that owe it.

Social learning is not something that depends on the existence of states or other institutions. It occurs whenever a rich set of beliefs, behavior, and emotional repertoires are inherited. Too often, we associate learning with institutions like schools. Yet, social learning does not require institutional or explicit instruction. Instead, social learning depends upon a range of practices and activities such as imitation, play, and simply observation where certain skills, beliefs, and emotions are modeled (i.e., not taught) and subsequently adopted by a learner (Heyes, 2012). Social learning occurs in all cultures. Some argue that it is distinctive of what culture is (Richerson & Boyd, 2008).³

Social learning is a powerful force for both inheritance and intergenerational continuity and therefore also relational features like identification, loyalty, and like-mindedness that neither relies directly on biology nor on the existence of certain institutions (though it certainly interacts with them). This rich source of inheritance occurs within national communities and any pre-political communities and often does produce like-mindedness, identification, and loyalty within these communities. But it is a more fundamental feature of a community than that. That is because while social learning promotes like-mindedness and conformity more generally over generations, identification and loyalty among members can lessen over time. This is not a puzzle from a social lineage view since a lineage underpinned by social learning means two things. It implies relative similarity in traits within temporal, geographic, and cultural proximity since we tend to learn to do as others do rather than go at it on our own, that is, through trial and error (Nielsen & Tomaselli, 2010). But social learning also means that over time and space, change is likely to accumulate within a lineage and a sudden uptake of innovations, much like mutations (Sperber & Hirschfeld, 2004).⁴

Thus, lineages of social learning both underlie and explain what national and pre-political communities are in the first place. This also leads to a striking difference with national community views that rely on identification or like-mindedness as being the more fundamental property of the community. One cannot simply reject one's history of social learning without heavily investing and assimilating with another social lineage. In contrast, identification views struggle since when people simply do not identify or feel loyal to the nation, they seem entitled to reject their belonging in the community rather swiftly (i.e., through simply ceasing to identify or feel loyalty).⁵ We will also return to the normative implications of our view below. The point for now is that being linked through social learning is more suitable for normative commitments than the linkage that comes about merely via identification or like-mindedness alone—indeed it also explains the emerge of the latter features.

We stress that the inheritance transmitted via social learning is not limited to concrete skills and practices, but also includes beliefs about the identity and self-understandings of a community in question, with or without an institutional framework. This also helps explain competing and even dissenting views can exist within a national community. When a Black American writer like James Baldwin (2012) argues that American history is not the history of his community, he is not voicing an idiosyncratic position, but a shared sense of alienation within the Black American community. If he and his community truly did not participate at all in social learning as Americans, if the community belonged to an utterly foreign culture, say, then his complaint would be pointless. But Black Americans do also partake in the same social lineage as other Americans and share many of the same cultural references, while at the same time feeling that their distinctive history and suffering are poorly reflected in the dominant national narrative. Baldwin's anger is precisely intelligible from the perspective of someone who feels (partially) alienated from the chain of social learning to which he belongs. We might say that Baldwin belongs to several social lineages, one as an American and a subsocial lineage as a Black American.⁶ Indeed, the possibility of being an heir to more than one social lineage is not unique to the American case and explains why many feel especially troubled by historical injustices precisely because they belong to the lineages of the offended and the offending parties.

The intergenerational transmission of social learning then explains how national communities, with or without a state, can transmit a sense of cultural identity and self-understanding—including a narrative of how they have interacted with other communities. Contrary to the national community views discussed in the last section, we contend that the links through social learning are more fundamental and our inheritance far richer than a matter of mere like-mindedness or certain goods or benefits from our ancestors. Instead, we inherit a form of collective agency. This includes collectively held claims, such as the right to self-determination or the claim to an ancestral homeland, but also other features due to social learning that become part of the collective agency, such as a set of memories, and reactive emotions. Such agency involves taking seriously what past generations did or endured and even forming hopes of the future beyond one's lifespan.

Social learning gives shape and content to our social identity as a group and through the latter, we learn to think and feel on behalf of other group members. As the example of James Baldwin above illustrates it does not mean that others necessarily agree with us or that this spokesmanship is indefeasible (but note that individual agents can change their minds or be of two minds about things). Rather it means that we do not just have self-interests and speak on behalf of ourselves or our close family members, we also have interests in virtue of and speak on behalf of the social lineages of which we are a product and the social identities we come to possess.

The inherited agency that comes with social lineages also implies a normative component attached to the agency, namely an associated ledger of moral claims and debts qua members of a social group. We inherit claims on behalf of not just of our ancestors as individuals, but of the nation and social lineage as a whole. Belonging to a social lineage, like a nation, is to partake in its collective agency and to inherit the claims to a homeland and self-determination as well as debts associated with this group agent's past. In that sense, we are not just heirs of the past but also trustees given that we are to pass on that which we received to the next generations—debts as well as claims.

One might object that though one inherits an identity, via social learning, it does not automatically follow that one inherits moral claims and debts from past generations. Identity cannot explain the normative aspects of agency. On the contrary, we believe that national and other pre-political identities are normatively laden identities precisely because those identities are not detachable from certain kinds of claims. To assume the identity of a social lineage means accepting a certain moral ledger. In the same way one cannot insist on belonging to a family and deny that anything normative follows from it. Something is amiss if one says, “I am a Palestinian but that does not establish any of my commitments or rights” but nothing is confusing about saying “I am a Gemini or a cyclist, but that does not establish my commitments or rights.” Ordinary use of such identities indicates that we cannot insist on the former kind of membership and maintain that this is normatively inconsequential whereas this seems plausible in the latter case. To fully partake in the social lineage attached to nations just is to inherit a normatively laden identity.

Admittedly, some people appear to resist that some identities, like national ones, are normatively laden. However, this is not quite true. They rarely challenge the idea of inheritance; they challenge particular claims. Americans who are reluctant to inherit a debt for the practice of slavery by their ancestors do not shy from the idea that other groups inherit debts to them, for example, Americans accepting German war reparations for generations or that the French owe them gratitude for their liberation. Nor do they object to a normatively laden identity when they find the claims desirable, for example, national self-determination. Rather these nationals object to debts that they find unpleasant or unfair, but they do not reject the concept of an inherited claim or debt as the above examples show.

Unlike Miller, our symmetry argument focuses on the inherited identity at the heart of collective claims and debts rather than the symmetry of goods and costs: one cannot assert an identity to exercise a right and deny that this same identity comes with duties. Those who insist on inheriting only the claims are in denial about what exactly it is that they inherit. Their willingness to press their claims to inherited debts reveals that they too accept normatively laden identities. Simply put, we commonly use the identities we inherit through social learning as if they are normatively laden even if we do so inconsistently and, at times, self-servingly.

We now need to spell out our view of collective agency. One, it is ontologically parsimonious. We are agnostic on supra-individual entities; we reject ontological profligacy. Our account simply does not need such entities. Such parsimony in social ontology is associated with work on collective intentionality (Bratman, 1992; Gilbert, 1990; Searle, 1990) but for reasons we will return to, we instead want to emphasize the role of collectivized self-perception in collective agency that has been identified by political and social psychologists as a crucial element to large-scale mobilization (Brewer, 1991; Fisher et al., 2013; Huddy, 2013).

These psychologists have identified two basic modes of self-perception: personalized and depersonalized. When we engage with the personalized self, we assert our uniqueness; we stand out from others as qualitatively and quantitatively unique. “I love you” identifies the speaker as a distinct individual, and so does “I am the author of that work.” However, the phrase “I am an immigrant” or “Our homeland is under attack” identifies one as a numerically distinct but qualitatively indistinct member of a social group. Depersonalization or collectivized self-perception involves “a shift towards the perception of self as an interchangeable exemplar of some social category and away from the perception of self as a unique person defined by individual differences with others” (Turner et al., 1987, 50).

Thus, the collective agency arrived at through social learning is not a series of individuals who are thinking in the personalized mode of self-perception. The fundamental difference in self-perception is the first sign that may not be able to transfer the sophisticated models of joint agency designed for small-scale cooperation to cases of national identities. In the former case, known individuals can directly communicate, as in the central example of taking a walk or painting a house together (Bratman, 1992; Gilbert, 1990). Yet, the collective agency that we find when we speak of “The Chinese resisted Japanese occupation” is quite different. Not only do they rely on different modes of self-perception, but the kind of coordination and specificity of intentions that seem to apply in small-scale collective agency do not seem to easily transfer to large-scale cases. It is not clear how hundreds of thousands or millions of members could or do mesh their sub-plans. For example, the Chinese people could share a superordinate aim—to expel the invaders—and yet not agree on their plans or sub-plans. Collective agency appears quite different depending on whether it engages with the personalized or depersonalized self, with small or large groups. For this reason, we focus on the collective self-perception behind choice and action rather than on the content of the intention.

We have defended a more fundamental explanation of how national and other cultural communities are normatively tied to their past. Chains of social learning connect individuals to the past and these links are constitutive of their present communal identities: to be an heir to a historical community just is to be the product of a lineage of social learning. This also allows us to clearly distinguish our view from the other ways in which one has attempted to bridge the past with the present in the case of historical injustices and overcome several objections.

First, one might wonder whether large-scale groups like nations can truly be collective agents, since we have seen that there are grounds to be skeptical about their members realizing true collective intentionality with the coordination and meshing of different sub-plans. At least it may seem like there can be no collective agency without also supposing formal political institutions where a collective agent is an “incorporated group” (Stilz, 2011, 192). And if a group like a nation can only be a collective agent with formal political institutions, our inherited agency view collapses into a variant of state institutional liability.

More forcefully, one might reject that inherited identities constitute (collective) agency. Social learning might explain the transmission of identities, depersonalized self-perception, and even normatively laden identities, but that still leaves unanswered how collective decisions are made and executed. How does this inheritance answer the question: who is a collective agent?

The challenge is daunting. We have produced a necessary condition: one must partake in the social lineage. But is this sufficient? Can the comatose take part in the collective agent if they inherit the right depersonalized identity? We readily admit that for a collective agent to emerge there must be both a collective identity and some form of public culture in which one can actively partake. This public culture can be more or less formal, but there must be some opportunity for one to pass from purely ascriptive identity to a more active membership which partakes in shared life.

Formal structures help, but it is nearly impossible to tell the tale of peoples by appealing exclusively to states and similar institutions as Stiltz does. Rather, we think that it is important to recognize that peoples without a state—for example, previously the Irish, today the Kurds, etc.—both seem to act as a group with shared duties and commitments and expect to be treated as such. Put otherwise, one inherits a collective agency and one makes use of this inheritance by acting upon the identity. Membership is necessary but it is only once it is actively used, particularly to make claims, that one is included within the ledger-possessing-group.

History is replete with groups acting without a state. The Great Revolt of the Egyptians lasted a generation (206–186 BC) without ever establishing a state capable of collecting taxes (Clarysse, 2004). Moreover, such requirements are vulnerable to charges of being overly legalistic and Eurocentric. The Inca never possessed a written constitution because they lacked a written language. Yet, they were capable of significant cultural transmission and social learning tantamount to generating collectivized self. If a constitution is an essential feature of an incorporated group or true collective agent, then the Incan Empire is inexplicable.

By properly tying social identities, such as nationality, to an unbroken chain of social learning we can answer the challenge skeptics who believe that state institutions are essential. Through social learning, communities perpetuate themselves without a state, despite state interference or persecution, like the Ukrainians, the Kurds, and First Nations the world over. A very substantial amount of social learning is in fact required for the survival of all intergenerational communities. New members, either born into the group or welcomed from the outside, are socialized: they learn a language or a dialect, learn manners and expressions, become familiar with artifacts and techniques, stories, and so on. There simply is no such thing as an intergenerational community, like a nation or people, without an ongoing chain of social learning that transmits an identity and culture to its members.

We do not deny the importance of institutions. Things like public institutions or writing are helpful to scaffold social learning and the transmission of social identities and their culture, but they are neither historically nor conceptually necessary. Nor is institutional learning sufficient within a majority context. Much, if not most, of what we learn comes from friends and family. Furthermore, many failed attempts at assimilation are reminders that official institutions can struggle to transmit an identity to recalcitrant populations. While institutions are useful, they are not essential.

Second, on the view of collective agency which we espouse we might also wonder why individuals should commit to such a depersonalized self and inherited agency in the first place. After all, it is not arrived at voluntarily but rather through a process of social learning—a large part of which occurs already in childhood without voluntary consent. Janna Thompson's group agency view picks up on this point and argues that our commitment to the agency stems from the importance of groups having a “temporal trustworthiness” comparable to the trustworthiness of an individual agent who typically honors her past commitments and obligations: “An intergenerational agent is a group whose temporal trustworthiness persists through generational changes of membership” (2022, 13). She argues that this value comes from the intergenerational extension of groups where there is a genuine concern for the future and the past of the group agent.

Like Butt's proposal of generational overlaps, Thompson's intergenerational extension does not however identify the right sense in which this community or agency is valuable to us because both their accounts neglect the role of social learning in establishing this commitment. If the collective agency is arrived at through the participation in social learning and lineages it not only allows us to care about the generic past and future beyond our life span. It allows us to care about specific parts of the past and the future indexed to the social learning lineage of the culture and nation in question. This is probably not only because we place value in the collective agency and collectivized self itself (its intrinsic value, as it were) but also because the content of social learning (the skills and beliefs transmitted) is valuable as are those that are our models—our cultural ancestors—and those that we transmit our culture to—our cultural progeniture. It is typically very emotionally painful, as well as morally difficult, to try to distance oneself from one's membership in a social lineage even if this membership was involuntary. In short, the notion of exit is heavy with cost: there are many things, like beliefs, relationships, and personal history that one must renounce at the same time in addition to the valuable agency itself.

Third, the account allows us to argue that the reason why we accept responsibility for the past is deeper than the fact that we may, or may not, inherit substantial material and immaterial goods from past generations. Unlike Miller, our view does not depend upon a favorable cost–benefit analysis. Rather we accept collective responsibility because we accept a social identity and its attendant agency, which are intrinsically valuable to us (as well as valuable due to the people and the process of social learning itself). Some of this inherited valuable agency will inevitably be shaped by the cultural community's role in cases of historical injustices whether as a perpetrating or suffering party. Our view of inherited agency relies on noncontingent features. Unlike structural and class action views where the downstream effect of the historical or enduring injustice are simply conceived of as benefits or harms quite independent of their role in our identities, our view would track past deeds whether they have downstream effects. Nor does it depend on whether the injured party recognizes what occurred and produces an accurate account of the downstream effect. On an inherited agency view, all of these issues are secondary to one's normatively laden identity.

Of course, the national community views we have reviewed differ from the class-action accounts as they derive duties of repair not directly from benefitting from a structural or enduring injustice. Instead, they claim that responsibility for historical injustice follows from our willingness to inherit certain goods from past generations. For them, because we claim the inherited benefits from past generations as our collective possessions, we must also accept inherited costs. We agree that in practice, members of nations accept a fair amount of inheritance and that to be consistent they must take the good with the bad. However, we disagree that the fundamental role of cultural inheritance is about accepting the burdens along with the benefits. Rather, we think that what we inherit is a form of agency involved in being a member of a genuine intergenerational social lineage with its associated ledger of moral claims and debts.

What is the difference? Instead of considering that we inherit benefits or costs, things that are scalar and quantitative, we should think that members of a social lineage inherit a ledger of moral claims and debts. The difference is most noticeable for occupied or exiled people who do not benefit from the land inherited from their ancestors, but still inherit a collective claim to it. Similarly, a nation might claim artifacts crafted by their ancestors, and yet this claim is compatible with them being presently in foreign hands. Inheriting a benefit is ambiguous as it does not distinguish between a “successful” inheritance which combines having the good with the rightful claim to it and the claim without the good. Our view requires only that a community inherits the claim itself.

Moreover, if we consider that we inherit collective agency, then we do not need to compare the benefits and costs of membership. The claims of a homeland or the right to self-determination are moral claims attached to group agents and they hold equally for the exiled, the occupied, and the free. We can provide an account of why all nations must accept their inheritance—as in Renan's famous inheritance of glory and regrets (Sand & Renan, 2010)—without resorting to the claim the only reason one must take responsibility for the past wrongs of ancestors is simply that there are also benefits attached to it. Intergenerational communities that form robust lineages of social learning typically make claims to certain land and self-determination, regardless of their level of freedom or prosperity.

Thus, the reason that we should accept responsibility for the past is not that the past has already provided us with a basket of goods or benefits, but because we already value the collective agency we inherited from past generations. Consider the expression “unceded land” used by First Nations or their supporters. The idea is that they have inherited a claim to the land from their ancestors even if they do not control it. Yet, the expression also betrays a very strong commitment to inheriting the consequences of past collective decisions. If the land had been legitimately ceded, it is less certain that the living could make this protest.⁷

We note one final difference with the national community view. Our view has the potential to explain more cases. While nations are often the relevant communities, we are not committed to the claim that only nations are constituted by the kind of lineages we identify. There are at least two good reasons for this. First, there is real debate about the antiquity of nations. If we wish to include ancient claims, we may wish to avoid what some would call anachronistic claims. Two, in the present there are still groups that are not clearly nations. For instance, it is not clear that a tribe is a nation. On our view, intergenerational communities formed through social learning are a genus and we are agnostic on the exact number of species included. Nations may or may not be the only species that belong to it. But our view makes room for the possibility that intergenerational communities that are not nations can be parties to historical injustices.

Let us recapitulate. We examined three kinds of answers to bridging the past with the present in historical injustices. Structural or Enduring injustice views dissolve the problem by claiming that the past still impacts the present and that we must act because present distributions of benefits and burdens are undeserved. Institutional liability views seek to bridge generations by identifying a single continuous institutional or corporate agent like a state. The dead and the living belong to the same state, and whatever the state owes its members must pay. The national community views start from our ordinary practices of claiming to inherit from past generations as part of belonging to a nation and extend this to the case of responsibility. They then urge us to be coherent—to inherit the benefits and burdens from the past.

These views, in a way, present a similar vulnerability as they seem to rely on features that are merely contingent to historical injustices. The first view does not offer a justification of why a remote injustice must continue to deliver benefits and harms to the descendants of those who were aggrieved and guilty of that injustice. This may or may not be the case. The institutional view falters because it limits its scope to historical injustices where we can identify a continuous institutional agent over time such as a state. Why should that be the case? Simply stated, a community like a nation may have a state or it may not, and yet it remains an intergenerational community. Finally, the national community view seems committed to such communities inheriting a sufficiently large share of goods such that members will accept the burden of repair.

We are also doubtful that these or any other features suggested by advocates can provide a normative basis for cross-generational commitments that repairs for historical injustices entail. To this end, we have developed a collective agency view based on the ties and value of communities which constitute cross-generational lineages of social learning. It is such agency and its value to individuals that belong to its temporal extension that grounds land claims, self-determination as well as debts such as duties to repair. In this view, all intergenerational communities are tied to past generations through the process of social learning that is constitutive of their social identities. This is a noncontingent fact about intergenerational communities such as nations. It also provides a way of picking out communities without having to identify a state, state-like institutions, or downstream benefits or costs of historical injustices. For remote injustices, this is how we propose to connect the communities, aggrieved and guilty, in the present with the past.

Funding for the work was provided by the Independent Research Fund Denmark (Sagsnummer: 9062-00049B).

The authors declare no conflicts of interest.