Repetition, adaptation, institutionalization—How the narratives of political communities change

At times when migration and diversity are politically salient and controversially discussed, the rhetoric of staying ‘as we are’ is widespread. But how do ‘we’ actually change and how would ‘we’ know when it happens? Based on the premise that political communities are the products of narratives of peoplehood, this paper explores how such narratives evolve over time. It conceptualizes different modes of balancing narrative continuity and change. These modes – repetition, adaptation, and institutionalization – are illustrated with reference to evolving German narratives of peoplehood centring around (not) being a country of immigration. The paper argues that all modes lead to some degree of change in narratives of peoplehood. Against the backdrop of different understandings of the core of a narrative, it further discusses when such changes fundamentally affect who ‘we’ are. Overall, the paper invites scholars, policymakers, and citizens to think critically about the essential aspects of their political communities’ narratives and to be aware of the stories that ‘we’ are told and that ‘we’ tell ourselves.


Introduction
Following the 'summer of migration' in 2015, the German public debated whether and how the arrival of hundreds and thousands of people seeking asylum was going to change Germany. In this debate, politicians across the political spectrum tried to speak to what social psychologists call the 'resistance to change' of their constituents by emphasizing that 'we' should and will stay fundamentally 'as we are' (Jost, 2015: 607-613;Za´rate et al., 2019). This has taken very different forms, for instance with the far-right AfD stating that 'we want to leave a country to our descendants that is still recognizable as our Germany' 1 (Alternative fu¨r Deutschland, 2017: 28), while the Green Party vouched in view of anti-immigrant populism to preserve fundamental rights, democracy and principles of the open society that supposedly 'define us and make us strong' (B € UNDNIS 90/DIE GR € UNEN, 2017: 146). Chancellor Angela Merkel, too, reassured citizens a year after her famous statement 'Wir schaffen das' (We can handle it) that, despite significant changes due to immigration, 'Germany will stay Germany, with all that we hold dear and care about' (Merkel, 2016a). But how exactly do political communities (avoid) change over time? What is it that 'we hold dear' and that should not change? And how can 'we' know whether 'we' have indeed changed?
The paper seeks to shed new light on these questions. It builds on the premise that political communities are constructed through narratives of peoplehood. Combining this approach with an understanding of narrative inspired by narratology, it conceptualizes different modes of (re)constructing political communities. The first mode is repetition, in which a similar story is retold using similar words, often with the intention of avoiding change. In a second mode, adaptation, central events and characters are intentionally changed, often to maintain credibility. Lastly, political leaders can choose a mode of institutionalization, cementing the narrative in a form that is more authoritative to the community. Regardless of the narrator's intentions, all three modes imply some degree of change to a political community's narrative of peoplehood and thus to the community itself. However, while a political community cannot stay completely 'as it is', it depends on the understanding of the core of a narrative of peoplehood whether a change is fundamental enough to construct a new 'we'.
The paper's argument is relevant to both scholarly and public debates. It points to the analytical potential that lies in connecting theories on political peoplehood with narrative analysis by offering a set of specific narrative components through which a political community's development can be observed. Furthermore, it can give citizens tools to attentively listen to, actively engage with, and critically challenge the narratives of peoplehood they are told and that they tell themselves.
The paper proceeds in four sections. The first section outlines the theoretical background of the narrative construction of political communities. The second section outlines three basic narrative components -story, text, and narration -which can be used to trace continuity and change in political communities. It further outlines three options of defining the core of a narrative of peoplehood -the moral message of a narrative, the concise narrative, and its main character. The third section conceptualizes three modes of balancing narrative continuity and change and illustrates them with competing narratives of German peoplehood. The first of these concerns the repetition of the narrative of no-country-of-immigration, the second involves the adaptation of the narrative of country-of-immigration; and the third entails the institutionalization of both narratives in citizenship and immigration legislation. The section further discusses how the modes interact with the cores of narratives of peoplehood. The final section concludes.

Political communities as narrative constructs
Most social scientists agree that social groups in general, and political communities in particular, are not natural entities; instead, they are social constructs or, in Benedict Anderson's famous wording, imagined communities (Anderson, 1983;Brubaker, 2002;De Cillia et al., 1999;Smith, 2003). Many scholars have differentiated between various types of these imagined communities, often building on (or criticizing) the distinction between ethnic and civic constructions (Brubaker, 1992;Jensen, 2019;Joppke, 2007). This paper has a different interest. It looks primarily at the tools with which political communities are (re)constructed rather than at the outcomes themselves.
Psychological research has shown that language in general, and narratives and storytelling in particular, are a central tool through which humans make sense of their social world and their lives (Bruner, 1991;Koschorke, 2018;McAdams, 2011;Polkinghorne, 1988). Indeed, children begin to tell stories early on, before they learn to understand or produce logical arguments (Bruner, 1990(Bruner, : 77-80, see also 1991. In a similar vein, sociologists and narratologists have pointed out how people invest and create meaning in their social environments, including the groups they imagine they belong to, by constructing narratives (Somers, 1994;Sommer, 2017).
It is therefore fitting that scholars studying political communities have often utilized terms related to narratives. Anthony Smith, for instance, recognized that narrated myths and memories are a core feature of nations (Smith, 2008: 39-46). Postcolonial approaches have gone even further and described nations generally as narration (Bhabha, 1990). States, too, have been discussed as narrative constructs in studies of political theory (Depenheuer, 2011) and international relations (Berenskoetter, 2014;Khoury, 2018). Legal theorists, as well, have thought of constitutional communities as narratively constructed (Addis, 2018;Rosenfeld, 2010: 45). And finally, for the political theorist Rogers Smith, all political communities -be they national, religious, urban, rural, or else -rely on 'persuasive historical stories that prompt people to embrace the valorized identities, play the stirring roles, and have the fulfilling experiences that political leaders strive to evoke for them' (Smith, 2003: 45). Since Smith's concept of 'stories of peoplehood' explicitly relies on narrative terminology, 2 since it connects well to approaches in narrative theory, and since it has a broad scope going beyond accounts of nationhood or statehood, it is an appropriate starting point for this paper's interest in how political communities' narratives change.
Narratives of peoplehood share a couple of features. First, by constructing a people, they also construct who the people are not (Smith, 2003: 56). In other words, narratives of peoplehood establish substantive markers of belonging, such as citizenship, nationhood, ethnicity, religious affiliation, and so on (Martin, 1995: 10).
Second, the process of building a people through storytelling is always ongoing, since existing narratives are constantly challenged by competing versions (Smith, 2003: 53-56). Thus, a narrative of peoplehood is never set in stone.
Third, in order to marginalize competing narratives and shape social reality, narratives of peoplehood need to be widely believed (Wodak and Triandafyllidou, 2003: 214). In political psychology, this can be studied through individuals' appropriation of master narratives for their own life stories (Hammack and Pilecki, 2012: 84-88). Some aspects make it more likely for a narrative of peoplehood to become dominant in this way. To begin with, it is important that narratives of peoplehood convey -quite literally -a good story, or to possess what is called 'tellability' in narratology (Ryan, 2005;Shenhav, 2015: 75). A narrative of peoplehood's tellability may be grounded for instance in evoking emotions (Koschorke, 2018: 80-82), or in promising political power, economic success and ascribing an inherent normative value to a political community (Smith, 2003: 59-71).
Furthermore, narratives of peoplehood are more likely to be (or become) dominant if they are told by narrators with wide discursive reach or 'power over ideas' (Carstensen and Schmidt, 2016: 326-328;Koschorke, 2018: 198). Often, these tend to be societal leaders such as politicians, religious leaders, or spokespeople of social movements (Smith, 2003(Smith, : 22, 32-36, 2015. Ordinary people may, of course, choose (not) to believe their narratives, offer counter-narratives instead and understand their community in a very different way than the elite's accounts (Fox and Miller-Idriss, 2008;Koschorke, 2018: 64-68;Smith, 2003: 34). However, these narrators of everyday peoplehood tend to be less well positioned to shape the collectively imagined community. Especially political leaders have the opportunity to tell their narratives in foundational documents, such as constitutions (and their interpretations), policy documents, public school curricula, public symbols (such as national anthems), and key political speeches (e.g. in election campaigns and public commemorations) that are widely reported on and shared in (social) media (Smith, 2003: 32-33).
Additionally, narratives of peoplehood have to make sense of peoples' lived reality to remain credible. This does not mean that they have to be historically complete or accurate. On the contrary, narratives are, to a degree, independent of past and present facts (Koschorke, 2018: 7-9, 202). However, if the tenuousness of the link between narratives of peoplehood and external circumstances becomes too apparent, narratives may lose their acceptance among listeners and be marginalized by competing and more convincing narratives (Koschorke, 2018: 197-202;Smith, 2003: 32-35).
Lastly, the credibility of a narrative also hinges on the discursive space it enters. A new counter-narrative of peoplehood that does not relate in any way, not even in contradiction, to established narratives is less likely to be persuasive or understandable to listeners (Koschorke, 2018: 197-204;Risse, 2010: 30-33;Smith, 2003: 48-49).
When discussing the ongoing competition of narratives of peoplehood, Smith has argued that continuity will mostly outweigh change. Particularly in wellestablished political communities, he did not expect fundamentally different narratives of peoplehood to be credible enough to win people's allegiances (Smith, 2003: 54-55). Similarly, narrative theorist Albrecht Koschorke speaks of a 'law of inertia for collective narratives' (Koschorke, 2018: 204).
Even though students of peoplehood regularly employ narrative terminology, its implications are seldom fully exploited. For example, Smith was quite explicit about not placing much emphasis on it (Smith, 2003: 44). In more recent work, however, he has called on political scientists to learn from literary theory (Smith, 2015: 8). And indeed, as this paper argues, there is much analytical potential in the concept of narrative, including in order to understand how political communities evolve over time, and what kind of continuity outweighs what kind of change.

Learning from narratology
Story, text, and narration Competing definitions of narrative exist in narratology (Scheffel, 2010). This paper draws on an understanding of narrative that provides useful heuristic tools for the analysis of narratives of peoplehood in the social sciences (Shenhav, 2015). It distinguishes three main components of narrative: the story, a successive set of events connected in a plot, which are expressed in a particular text, and communicated through narration. 3 These components can be illustrated using the statement Chancellor Merkel made in an interview alluded to in the introduction: 'Germany will stay Germany, with all that we hold dear and care about. But Germany has also changed time and again since the foundation of the Federal Republic [. . .]' (Merkel, 2016a).
A story is the component of a narrative that is concerned with its content. Stories consist of several elements, the most basic being an event, 'something that happens' (Shenhav, 2015: 21;Weixler, 2017: 14). In the example, the foundation of the Federal Republic, several unspecified instances of past change, and the future continuity of 'all we hold dear' are such events. Another element is the participating characters, such as individuals, groups, institutions, or conceptsin this case, Germany (which has changed) and 'we' (who hold it dear). Spatial and temporal settings are additional story elements (Weixler, 2017: 17-18). Here, the story is set in Germany, starting with the foundation of the Federal Republic in 1949 and stretching apparently into eternity. The plot moulds the raw material of setting, events, and characters into a single unit (Kukkonen, 2014;Shenhav, 2015: 32-34). A basic plot, grounded in everyday storytelling, entails the disturbance and reestablishment of an equilibrium (Todorov and Weinstein, 1969: 75). The plot of the example is more linear: both change and continuity have happened in the past and will happen in the future. Overall, studying a story reveals who and what is (not) part of a narrative of peoplehood, and its relation to time and space. For instance, Merkel narrates the foundation of the Federal Republic -rather than e.g. the establishment of the German Reich or reunification -as the event of Germany's birth.
A text gives a story its concrete expression, often in words or (moving) images. The text reveals a particular perspective on the story. In the example, Merkel speaks from the point of view of an imagined 'we' that seemingly values the same things about Germany. A text may further accord special importance to selected story elements, for example through its wording or through the presentation of events in a particular order, or frequency, or duration that may differ from the story's chronology. A text may, for instance, start with the last event in the story or exhaustively describe an event that took only a second (Shenhav, 2015: 39-45). In the example, Germany's future of continuity is told before its past changes, thus qualifying the former. When repeating the narrative a week later, Merkel presented the events in the opposite order, speaking first about past changes and ending her speech with a promise of continuity, thus emphasizing the latter (Merkel, 2016b). Studying a text requires interpretation to reveal what and whose take on the story is presented.
Finally, narration captures the social and communicative act in which narrative texts are created, told, and listened to. Some elements that can be studied under the auspices of narration include the author of the story (Merkel and her speechwriters), the speaker who lends their voice to the story (Merkel), the intended and actual audience (actual: interviewers, intended: newspaper readership and general public), and the reasons why a narrative is told at a particular place and time (the anniversary of Merkel's 'Wir schaffen das' speech) (Shenhav, 2015: 47-53). Studying narration reveals the meaning that a story and text acquire when they are told in a certain context.

Narrative cores
Not all aspects of narratives of peoplehood are equally relevant for the political communities they construct. Not every small variation in story, text, or narration is important. Instead, political communities arguably fundamentally change only when some core aspects of their narratives of peoplehood are altered. This section discusses three ways to think about such narrative cores -the moral message, the concise narrative, and the main character.
A first possible definition of the core of a narrative pertains to its moral message. Narratives are not value-neutral but provide an evaluation of the story, so that to 'tell a story is inescapably to take a moral stance' (Bruner, 1990: 51; see also Koschorke, 2018: 49-51). Merkel, for instance, listed liberalism, democracy, the rule of law, and a social market economy as characteristics that 'make Germany Germany' (Merkel, 2016a). If this message was understood as the core of her narrative of peoplehood, 'Germany' would only cease to be 'Germany with all that we care about' if its narrative no longer conveyed these values. However, the same moral message might be derived from very different narratives. The UK, for instance, can narrate democracy as a long-held tradition, while Central and Eastern European countries may rather narrate it as a recent achievement. This might not make any difference for the moral message of democracy, but it does matter for the way that these political communities understand themselves. Therefore, the moral message alone, arguably, does not capture the meaningful core of a narrative of peoplehood.
A second option for defining the core of a narrative is to look at the essence of the story. Shenhav captures this with his notion of 'concise narratives'. These are found in the parts of a narrative text that cover the entire time span of a storysentences like: We have been and will always be this way (Shenhav, 2015: 62-65). Such phrases convey short complete narratives with events and characters, and they 'usually reflect the speaker's core ideological perceptions or his political identity' (Shenhav, 2005: 321). If these concise narratives are understood as the core of a narrative of peoplehood, changing their events and characters would fundamentally alter a political community.
A third option for defining the core of a narrative pertains to its main character, or the marker of belonging indicating which people a narrative of peoplehood is (not) about. There are two ways of looking at the main character. On the one hand, it could be understood as a specific group of people -whether they are called a nation, a citizenry, a religious sect, or something else. On the other hand, the main character could hinge on a particular name invoked for any group of people -placing the importance on being called a nation, a community of citizens, a religious sect, or something else, independent of which specific group is meant. In both cases, a shift in the marker of belonging in substance (in the story component) or name (in the text component) could fundamentally change a narrative of peoplehood.
These understandings of narrative components and cores have important implications for studying evolving narratives of peoplehood. They allow tracing concretely how story, text, and narration change over time and what core elements do (not) change. However, such an approach is rare in the literature. Social scientists who have adopted a similar understanding of narrative and who have developed quantitative and qualitative methods of narrative analysis (Elliott, 2005;Franzosi, 2010;Gadinger et al., 2014;Shanahan et al., 2018;Shenhav, 2015), have focused mostly on narratives of particular policies, historical events, individual lives, or concepts (Andrews et al., 2015;De Fina, 2018;Jones et al., 2014;Shenhav, 2004). By contrast, research on the social construction of political communities has more often employed (critical) discourse or content analysis, even if the underlying theories often refer to narrative (De Cillia et al., 1999;De Fina, 2018: 239). This paper seeks to address this gap.

Modes of balancing continuity and change
Drawing on the narrative components of story, text, and narration, this section discusses different modes in which narrative continuity and change can be balanced. It conceptualizes three such modes -repetition, adaptation, and institutionalization -and illustrates them with evolving narratives of German peoplehood from the context of migration, integration, and citizenship policies. These illustrations are based on some exploratory narrative analysis of selected policy documents and on the secondary literature. Finally, the section discusses the theoretical implications of each mode for the continuity and change of a narrative's core, whether understood as moral message, concise narrative or main character.
Migration, integration, and citizenship are not the only policy areas within which narratives of peoplehood are constructed. But they offer particularly fertile ground for the narration of peoplehood, as defining who is (not) and can(not) become a member of a political community -a function of all three of these policies areas -requires a narrative of 'who' that community is in the first place. Furthermore, Germany is a suitable case for illustration because of the remarkable development of its migration, integration, and citizenship policy. 4 In the 19th and early 20th century, Germany witnessed both emigration and immigration. After decades characterized by forced migration in the context of two bloody world wars, it became a country with continuous positive net migration in the second half of the 20th century when West Germany actively recruited guest workers to satisfy the needs of its growing economy. After the economic situation deteriorated due to the oil crisis in 1973, guest worker recruitment was halted, but many immigrants remained in Germany and were followed by spouses and children under family reunification schemes (Oltmer, 2010). In this situation, competing narratives of peoplehood were put forward to make sense of the continuing presence of former guest workers. How the country has slowly begun to renarrate itself as a country of immigration and accepted its post-migrant reality has been the topic of much research (among others Bade & Bommes, 2004;Foroutan, 2014Foroutan, , 2016Fuller, 2019;Hell, 2005;Meier-Braun, 2013;Williams, 2014). The approach presented here aims to add a new perspective to this work by pointing out how specific elements of governments' narratives of German peoplehood have (not) changed over time and what this implies for the narratives' cores.
The analysis below has a number of important limitations. As a conceptual argument, it draws on empirical evidence for illustrative purposes. The three modes therefore do not represent an exhaustive or explanatory typology based on a systematic review of all possible narratives of German peoplehood. The application to other political communities and different time spans would require additional study to further develop, refine, and test this framework. Furthermore, since the theoretical framework expects narratives of peoplehood told by political elites to be particularly powerful, it does not cover influential counter-narratives of civil society and ordinary citizens. Lastly, it is not within the scope of this paper to explain the circumstances in which political leaders prefer a certain mode over another, or when narratives can be expected to change the most or the least.
More modestly, the following section outlines three different modes within which German narratives of peoplehood were partly modified and maintained in the past decades.

Repetition: no country of immigration
A straight-forward option to maintain a political community 'as it is' seems to lie in the repetition of an existing narrative of peoplehood. In this mode, the story and text are held fairly stable in the reiterations of the narrative, which are necessary to continuously form people's perceptions of reality (Shenhav, 2015: 56-57). Crucially, while the story and text can be repeated at different points in timethe same events are told in the same or similar words -the component of narration, by definition, cannot. At least the context of storytelling changes every single time a narrative is retold. In that process, its overall meaning may change as well. This mode can be illustrated with the repetition of the German narrative of nocountry-of-immigration, as is documented in the literature and as can be illustrated with policy documents and elite discourse.
An early version of the narrative is embodied in the position paper outlining what was meant to be a comprehensive policy towards foreigners in 1977. Developed by a commission of federal and state government as well as employment agency representatives, the first basic position finally adopted by state ministers of labor and social affairs represented a powerful narrative which was shared widely among the political elite at the time (Yıldız, 2017: 84-85): The German Federal Republic is no country of immigration. It understands itself as a country of residence for foreigners, who normally decide to return to their home countries after a shorter or longer period of stay (cited in Hell, 2005: 80 andHeßler, 1993: 137).
The position paper denied that Germany is a country of immigration and supported strict limits on future immigration. While it proposed to make the legal status of long-term resident foreigners more secure, it aimed for their integration on a temporary basis, ultimately stressing the priority of maintaining immigrants' willingness to return. This would also be a policy focus in the coming years (Bade, 2004: 391-395;Heßler, 1993: 134-145;Meier-Braun, 2002;Yıldız, 2017).
In the following decades, this no-immigration-country narrative was used by parties across the political spectrum and became an important leitmotif in the German government discourse on migration (Bade & Bommes, 2004;Hell, 2005: 80-82;Williams, 2014). The following quote is from an op-ed by Interior Minister Manfred Kanther in 1996, in which he sets out his political vision for migration policy in a widely read newspaper. In essence, he repeats the narrative almost twenty years later: Germany is not a country of immigration. We have never in the past pursued an active policy of receiving foreigners with the objective of their permanent settlement.
Even the recruitment of foreign workers was linked to the idea of a limited stay. It is not intended to enable the immigration of persons for permanent residence, either today or in the future [. . .]. (Kanther, 1996, cited also in Hell, 2005. Both examples exhibit the no-country-of immigration narrative. Through the lens of the narrative components sketched above, it can be roughly summarized as follows: in terms of story, crucial events are (a) the arrival of foreigners, (b) their limited stay, and (c) the expected return to the country of origin, all supported by corresponding policies. The narrative follows a basic plot: immigration brings an original, seemingly stable situation into disequilibrium, which is then resolved with the return of migrants. The two main characters in this narrative are 'foreigners' and 'Germany'. The story's spatial setting draws a distinction between Germany and the 'home countries'. Its temporal setting covers eternity, either explicitly -Germany has never been and will never be a country of immigration -or implicitly with the adverb 'normally'. In terms of their text, both narratives convey a perspective speaking for a unitary Germany, by lining out how the country as a whole 'understands itself'. Both repeat word-for-word the 'not a country of immigration' phrase.
In the reiterations of this no-immigration-country narrative across political speeches, party programs, election manifestos and policy papers that have been discussed in the literature, the story seems to have remained fairly stable (Hell, 2005: 77-82;Meier-Braun, 2002;Williams, 2014: 57). This does not mean that entire narratives were copied and pasted -different iterations of the narrative were underpinned by different arguments of why migration leads to disequilibrium, including for legal, economic, social, political, and cultural reasons (Wengeler, 2003: 516-519). Also, there was a stronger focus on asylum seekers rather than guest workers as foreigners in the 1980s and 1990s, and some versions even include the event of the integration or assimilation of foreigners who could not return. But throughout the decades, the story of foreigners arriving and ideally leaving without becoming a part of Germany remained salient (Meier-Braun, 2002: 71-72).
In contrast, aspects of the narration changed significantly. While authors and speakers of the narratives naturally differ, it was especially the wider context that had fundamentally changed: In the early 1970s, many immigrants indeed planned to return, yet this share decreased as family reunification occurred after the recruitments stop and as ethnic Germans and asylum seekers immigrated to Germany in the late 1980s and 1990s. Overall, the average number of years of residence that immigrants spent in Germany steadily increased and new generations of immigrants' children were growing up in the country (Oltmer, 2010: 54). Fostered by pragmatic integration and welfare measures, many immigrants and their children de facto permanently settled (Bade & Bommes, 2004). In this context of narration, the story and text of the no-immigration-country narrative featuring the event of return had a different meaning in the late 1990s than it did in the 1970s. With the settlement of immigrants, it clashed evidently with the social reality on the ground, which arguably undermined the narrative's credibility.
Against this backdrop, two observations can be made. First, some politicians resorted to the repetition of a similar story in a similar text over decades. They did not challenge their audience to reconstruct their no-immigration-country narrative of peoplehood; instead, continuity was the objective. Second, as the narration took place in a context of continuous settlement and de facto integration, the narrative's meaning changed nevertheless. The de facto situation of permanent immigration in a country denying this fact led to a 'social schizoid paradox' (Bade, 2017: 27). This, arguably, was one factor contributing to the marginalization of the noimmigration-country-narrative throughout the 2000s (Bade & Bommes, 2004: 454-455;Williams, 2014). However, the narrative has not completely lost its appeal: A public opinion poll in 2015 still reported more than a fourth of respondents indicating that Germany was not a country of immigration (Schmidt, 2015), almost 40% of respondents in 2014 still thought having German ancestors was important for qualifying as German (Foroutan and Canan, 2016: 34-36), and the event of return still features importantly in government discourse especially about rejected asylum seekers (see for example Bundesministerium des Inneren, 2019).
Returning to the conceptual questions at the core of this paper's interest, what do these observations imply for repetition as a mode of maintaining a narrative of peoplehood? Focussing on the moral message, repetition ensures continuity only if the same message emerges unambiguously from each narration. However, different moral messages might be derived from the same story and text at different points in time. The no-immigration-country narrative, for instance, has been told both to convey the value of a stable welfare system (an economic theme in Smith's typology), and to stress the importance of maintaining ethnocultural homogeneity (an ethically constitutive theme) (Bade & Bommes, 2004: 445-447;Smith, 2003: 59-71). If a narrative's moral message is ambiguous, repetition does not guarantee its continuity. Turning to the main character, repetition can maintain the main character's name as it appears in the text -Germany and foreigners, in this case. However, it might not maintain the specific group of people belonging to the character at the story level. The Germany of the 1970s, for instance, included different people than post-reunification Germany, and the people meant by 'foreigners' were much more diverse in terms of country of origin and legal status towards the end of the century. Lastly, through repetition, a concise narrative of peoplehood is likewise repeated. In this case, the concise narrative seems to be that migrants arrive, stay, and generally leave without fundamentally changing Germany. Overall, while repetition ensures continuity of the main character (in the text) and concise narrative, the fate of the no-immigration-country narrative illustrates that such continuity may be detrimental to the credibility of a narrative of peoplehood in changing circumstances. Another mode of balancing continuity and change, adaptation, seems more up to this task.

Adaptation: A country of immigration
In contrast to repetition, one can conceptualize a mode of adaptation, within which the main intention is not to maintain, but to modify a narrative of peoplehood. This can be done through excluding or including characters and events in the story, through connecting them in a new plot or, more subtly, through conveying a different perspective by choosing differently connotated wording. This mode can be illustrated with some developments that the German country-ofimmigration narrative of peoplehood has undergone throughout the years.
Since the 1970s, narratives of Germany-as-country-of-immigration continuously contested the then dominant narrative of no-immigration-country. Since 1978, the successive commissioners of integration appointed by the federal government were a voice that continuously put forward this narrative (Borkert and Bosswick, 2011: 97-98;Geiß, 2001). The development of these narratives is evident in reports that commissioners have delivered to parliament regularly since 1993. While revealing only a small part of the whole picture, they indicate how some political leaders' narratives of German peoplehood in the context of diversity changed over time. A short version of the country-of-immigration-narrative is, for instance, told by Commissioner Schmalz-Jacobsen in her 1995 report, in which she states that: . . . an important part of the foreign population will stay permanently in the Federal Republic. Calling migrants still or yet again 'guests', who live here only temporarily, is an illusion. (Beauftragte der Bundesregierung fu¨r die Belange der Ausl€ ander, 1995: 54) Twenty years later, Commissioner € Ozo� guz proposed the following narrative on the same question: It is good that in the last years, the deficit-oriented discourse about people with 'a migration background' has changed. Today, no one seriously denies that we are a diverse and colourful country [. . .] (Beauftragte der Bundesregierung fu¨r Migration, Flu¨chtlinge und Integration, 2014: 15).
The differences between these quotes illustrate some of the adaptations that the country-of-immigration narrative has undergone since it became more prominent in the 1990s, and that are evident from both an exploratory narrative analysis of the commissioners' reports and from prior studies on the German migration discourse.
With regard to the story, a first change is that the event of Germany accepting immigration to be permanent moves from the future (Beauftragte der Bundesregierung fu¨r Ausl€ anderfragen, 2000: 126; Beauftragte der Bundesregierung fu¨r die Belange der Ausl€ ander, 1994: 48) to the past (Beauftragte der Bundesregierung fu¨r Ausl€ anderfragen, 2002: 20; Beauftragte der Bundesregierung fu¨r Migration, Flu¨chtlinge und Integration, 2016: 20, 2019. Second, the 'German' character is explicitly redefined as colourful and diverse (Beauftragte der Bundesregierung fu¨r Migration, Flu¨chtlinge und Integration, 2016: 15, 19, 2019. Additionally, an important new character enters the narrative in the mid-2000s: 'people with a migration background' take over from 'foreigners' or 'migrants'. 5 Correspondingly, 'people without a migration background' partly take over from 'Germans' (Beauftragte der Bundesregierung fu¨r Migration, Flu¨chtlinge und Integration, 2007: 16;. With regard to text, the perspective on immigration has partly changed: While permanent immigration was first portrayed as an unavoidable task that must be dealt with pragmatically (as in the quote by Schmalz-Jacobsen), in later years, immigration has also been portrayed as an economic necessity and even as a success proving Germany's high standard of living (Beauftragte der Bundesregierung fu¨r Migration, Flu¨chtlinge und Integration, 2012: 12-13, 2014: 15, 2019; see also Ulbricht, 2018). With regard to narration, different politicians -including one with a 'migration background' -told their narratives in a context where migration, citizenship and integration policies became more politicized and contentious both on the national stage and increasingly at the European level. An important change in narration is, lastly, that the countryof-immigration narrative has become more dominant over time (Williams, 2014).
From this brief exploration, it seems that the country-of-immigration narrative, as told by parts of the political elite, was adapted -new characters were introduced, existing characters redefined, and the perspective on immigration has partly changed. Of course, this does not mean that there were no aspects of continuity at all. For instance, the plot consistently follows a basic structure, promising a happy ending of successful integration after earlier missed opportunities and, similar to the not-a-country-of-immigration narrative, a perspective speaking for some united German 'us' is maintained.
Returning again to conceptual questions, how does a mode of adaptation affect a narrative's core? For the narrative's moral message, adaptation of text and story can lead to both continuity and change. In this case, the moral message has arguably changed in certain aspects, away from pragmatically accepting towards rhetorically welcoming diversity. Looking at the main characters, 'Germans' and 'foreigners' were sidelined by 'people with' and 'people without a migration background', shifting the marker of belonging from legal status toward ethnicity . While in this instance, there was a change in the name of the main character (in the text), it is not as clear that there was change in terms of the group of people actually meant (in the story). In fact, many people 'with a migration background' would have been 'foreigners' before 1999. 6 Lastly, adaptation is likely to change the concise narrative through changing events and/or characters. In some early version of the concise country-of-immigration narrative, for instance, immigration was understood as the event that causes a disequilibrium and that was followed by the long denial of Germany as a country of immigration before more appropriate policies reestablished the equilibrium. In some more recent versions of the narrative, immigration is the event that reestablishes equilibrium by ensuring Germany's economic success (Ulbricht, 2018: 314).
In their everyday narrations of peoplehood (Fox and Miller-Idriss, 2008;Miller-Idriss, 2006), ordinary citizens may adapt or repeat their narratives, as well. But there is an additional, important mode of balancing narrative continuity and change that is primarily accessible to political elites: the institutionalization of narratives.
Institutionalization: Narratives of (no)-country-of-immigration in the law Narratives of peoplehood inform institutions that govern political communities, in particular the rules that define membership in a political people (Smith, 2003: 49). This process of institutionalization is a third mode of balancing continuity and change. Relative to the components of a narrative, it entails the transformation of the narration component by putting a certain story and text in a new context, namely one that is authoritative to the political community. Laws are the obvious way to do this, but national anthems, loyalty oaths, or pledges of allegiance follow a similar logic.
In the German context, institutionalization could be illustrated with many legal developments from the past decades. The focus in the following will lie on a few examples that illustrate the competing narratives of German peoplehood discussed earlier: The 1983 Assistance Act for Returning Foreigners offered financial incentives for unemployed foreigners to return to their country of origin; the 2000 Citizenship Act liberalized access to German citizenship at birth and through naturalization, the 2005 Immigration Act reformed immigration governance, and the 2020 High Skilled Immigration Act incentivized qualified workers to immigrate to Germany. The narratives that these legislative acts are meant to institutionalize can, inter alia, be found in the respective official reasoning of the government proposals. 7 In the proposal for the 1983 Assistance Act, for instance, the government narrates a threatening alternative future, stating that 'in case of a further increase of the number of foreigners, it will not be possible to implement foreigner policies targeting integration [. . .]. A consolidation effect can be reached inter alia by the stronger return of foreigners' (Bundesregierung, 1983: 7). It thus aims to translate the nocountry-of-immigration narrative into law, positing the return of immigrants as a kind of happy ending. In contrast, the 2000 Citizenship Act narrates a different version of the future. The final event is also a happy ending, but it no longer entails return. Instead, it is the 'inclusion of the foreign population permanently residing in Germany by awarding German citizenship' (Deutscher Bundestag, 1999: 11), making 'Germans' out of 'foreigners'. Finally, the justifications for the Immigration Acts entering into force in 2005 and 2020 do not depict the arrival of foreign workers only as a de-stabilizing event that is eventually resolved either through immigrants' return or their inclusion in the citizenry. Reflecting the adapted versions of the country-of-immigration narrative, immigration itself is also portrayed as a resolving event that will help to reestablish the equilibrium disrupted by demographic change and globalization (Bundesregierung, 2003(Bundesregierung, : 59, 2019. Notwithstanding their differences in the story and text of the underlying narratives of peoplehood, these pieces of legislation share the crucial similarity of transforming the narration into a legal one, giving them a different meaning through making them more binding. In this way, institutionalization can give underlying narratives of peoplehood 'resilience in the face of challenge and critique' (Hayward, 2011: 659). However, it also makes them more explicit and exposes them to public scrutiny, which is not always a recipe for success. For instance, the 1983 institutionalization of the no-immigration-country narrative was criticized for being merely symbolic, thus arguably not enhancing its credibility in the long term (Meier-Braun, 2002: 61; but see Yıldız, 2017: 220-230). Similarly, the future event of naturalization of all permanent residents as institutionalized in the 2000 Citizenship Act seems questionable considering that naturalization rates have been low for years and were at only 1.14% in 2018 (Beauftragte der Bundesregierung fu¨r Migration, Flu¨chtlinge und Integration, 2019: 290-291). And finally, Germany still attracts less skilled immigration than its economy needs (Hunger and Krannich, 2018: 230).
But, even if it fails to make the narrated events reality, can the mode of institutionalization ensure continuity of narrative cores? Arguably, this mode is most fitting to ensure the continuity of the moral messages of narratives of peoplehood. This kind of institutionalization is familiar from constitutional law: for instance, the eternity clause of the German Basic Law -article 79, section 3 -guarantees that the articles enshrining liberal fundamental rights, democracy, the rule of law, and a social market economy -values that Merkel enumerates as defining Germany -cannot be amended. But this kind of insurance, as powerful as it is, primarily guarantees the moral message of a narrative on paper; it does not necessarily maintain the narrative in political discourse or in people's minds. Institutionalizing a character does not necessarily guarantee its position as a main character in the narrative of peoplehood in people's minds, either. While it may legally define who is meant by 'German' and 'foreigner', it might not prevent the invention of new characters -such as people 'with' and 'without a migration background' -who may eventually become the main characters in narratives of peoplehood. Finally, institutionalization could directly enshrine a concise narrative, as often is the case in a preamble (see Addis, 2018), an oath or an anthem. However, in this form it confronts the same challenge as a narrative in the mode of repetition: it may not age well in changing contexts of narration.

Conclusion
This paper understands political communities as constructs of narratives of peoplehood that consist of a story, text, and narration. It conceptualizes three modes for balancing continuity and change among these components and illustrates them with examples from Germany: (a) the repetition of the narrative of no-immigrationcountry; (b) the adaptation of the country-of-immigration narrative; and (c) the institutionalization of both narratives. Certainly, there will have been also instances in which the no-immigration-country narrative was adapted, or in which the country-of-immigration narrative was repeated. Narrators are not restricted to a specific mode at one point in time, and there may be additional modes for balancing narrative continuity and change. But what this paper shows is that, already conceptually, none of the three modes discussed here leaves a narrative of peoplehood completely unchanged, notwithstanding peoples' resistance to change and promises that 'we' will remain 'as we are'. As narratives have to be told again and again to continuously shape social reality, and as the exact same narrative can never be told twice due to the ever-changing contexts of narration, narratives of peoplehood in fact have to change in order to be maintained.
The paper further argues that how significant such changes are to a political community depends on the definition of the meaningful core of a narrative of peoplehood. It discusses the moral message of a narrative, the concise narrative, and the main character as options for conceptualizing this core and explores how the modes of balancing continuity and change interact with these. Repetition might be the mode most appropriate for maintaining the concise narrative and the name of main character in the text of a narrative of peoplehood. At the same time, this mode may put narratives of peoplehood at risk of decreasing credibility in changing contexts. Adaptation can maintain the main character of a narrative, by changing the name but maintaining the specific group of people meant, and it may be most suitable mode to maintain the credibility of a narrative of peoplehood in changing circumstances. Finally, institutionalization can directly enshrine a concise narrative and legally define its main characters, thus cementing it, but also facing risks of long-term credibility. However, it seems to be the most appropriate mode for enshrining the moral message of a narrative of peoplehood.
There are a number of questions that the paper raises but that lie beyond its scope. Under what conditions do political leaders opt for one mode rather than another? Is one of the modes empirically more likely to fundamentally change political communities? What are the modes that ordinary citizens use in their evolving narratives of peoplehood? Additionally, there are important normative issues. For instance, should (the cores of) narratives of peoplehood be maintained at all? Is one understanding of a narrative core more morally defensible than others? And when is change of a narrative of peoplehood (un)desirable?
Such questions continue to inform the ongoing contentious debates about how to narrate German peoplehood. On the one hand, recent changes to the German citizenship legislation tie naturalization closer to fitting into a 'German way of life' (Farahat, 2019), thus institutionalizing a cultural marker of belonging to the 'German' character. On the other hand, civil society actors, activists, and scholars have explicitly been calling for a new German narrative, arguing that such a narrative should go beyond stressing people's 'migration background' and suggesting to use 'new Germans' as the main character, and to more prominently include events such as migrants' contribution to the post-war economic recovery, post-war forced migration, or ongoing differences between East and West Germans (Foroutan, 2010(Foroutan, , 2014. Whichever mode will be adopted in the future, Germany's narrative of peoplehood will not remain the same. on earlier versions, and to Jelena Cupa� c, Jo Shaw, and Shaul Shenhav for helpful exchanges. Thanks also to Roisin Cronin for her excellent editing.

Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (grant agreement No 716350).

ORCID iD
Johanna Hase https://orcid.org/0000-0001-9103-8466 Notes 1. All translations by the author. 2. While Smith uses 'narrative' and 'story' interchangeably, a story is understood as one among several components of a narrative in this paper. His concept is thus referred to as narratives of peoplehood to avoid confusion. 3. Smith is concerned about defining narratives too narrowly and about excluding sources such as census data (Smith, 2015: 38-39). While numerical data alone indeed do not satisfy the definition of narrative used in this paper, political leaders usually present and interpret such data in press releases, reports or interviews, which again mostly fit the narrower definition of narrative. 4. Any analysis of the period before reunification in 1990 pertains to the Federal Republic of Germany and not to the German Democratic Republic, where immigration took place on a much smaller scale (Bade and Oltmer, 2011: 77). 5. According to the German Statistical Office, a person with a migration background is someone who did not have German citizenship at birth or who has at least one parent who did not have German citizenship at birth (Statistisches Bundesamt, 2019). 6. The important exception comprises ethnic Germans ([Sp€ at]aussiedler), who have a 'migration background', but receive German citizenship more easily. 7. Some scholars look for story, text, and narration within legal texts themselves (von Arnauld, 2009: 23-31). However, such research is often based on a broader understanding of narrative, since laws seldom spell out concrete events on a timeline.