The extra-territorial paradox of voting: the duty to vote in extra-territorial elections

ABSTRACT The question of why individuals vote, the so-called “paradox of voting”, has been a crucial debate within political science, conceived deductively as an interaction between costs, benefits, and, as some argue, duties. This article situates the question of why individuals vote within the context of extra-territorial elections, focusing on how and why those who acquire citizenship kin-states participate in kin-state elections following citizenship acquisition, while continuing to reside outside of the kin-state. The article uses the case of newly acquired Romanian citizens in Moldova, who have never resided in nor intend to reside in Romania, to unpack whether, how, and why individuals acquiring Romanian citizenship in Moldova vote in Romanian elections. The article uses an interpretive and inductive approach to explore from the bottom up both the experiences of and motivations for political participation of extra-territorial citizens. The article finds, unexpectedly, how those acquiring Romanian citizenship in Moldova are motivated by a duty to participate. Overall, the article argues for a relational and reciprocal understanding of citizenship and voting, which focuses on the relationship between the kin-state, facilitating citizenship as a right, and the kin-citizen, performing their duty to vote.


Introduction
The question of why individuals vote has long been an issue of interest to political science. Rational choice approaches to voting conceive of turnout as an interaction between costs, benefits, 3 and, as some argue, a duty to vote. 4 This article seeks not to answer these theoretical questions, but poses a conceptual question concerning the participation of citizens in elections in states in which they do not reside (that is, extra-territorial participation): why do individuals vote in elections in a state in which they neither reside nor have ever resided? In these settings, it would be expected that the benefits of voting are more remote, the costs of voting higher than in domestic elections, while the sense of duty to vote, in a state in which individuals do not reside, largely absent. However, this article finds the reverse: Extra-territorial citizens do conceive of a duty to participate, and relate this duty directly to the idea of new citizens' legitimate performance. This article approaches political participation not from a deductive or aggregate/statistical perspective (as is more common), but inductively from a bottom-up perspective, focusing on the experience of, and motivations for, voting in extra-territorial elections.
Typically, extra-territorial participation has been studied in terms of diaspora, as a migration-centred analysis. 5 With the proliferation of dual and multiple citizenships, post-national approaches argue that citizenship is no longer bounded by the "nation", but has become more inclusive via migration. The demos too has become post-nationalized: By the beginning of the twenty-first century, over 100 states had enfranchised the right of permanently external citizens "to keep their citizenship and voting rights". 6 This article shifts away from the migration-centred focus of citizenship acquisition and political participation towards the enfranchisement of citizens following their acquisition of citizenship from kin-states in which they have never resided (kin-citizens). This expansion, and extra-territorialization, of citizenries is especially concentrated in postcommunist examples (for example, Romania, Croatia, Bulgaria, Hungary, and Serbia) where states facilitate citizenship acquisition for external communities considered "kin" (co-ethnic). Several cases (Croatia, Romania) also facilitate enfranchisement of kin-citizens, regardless of current or previous residency, where these (kin-)states permit and facilitate kin-citizens' right to vote while remaining (permanently) non-residents, including in external parliamentary constituencies. 7 This offers kin-citizens the possibility of determining election results, in external constituencies and presidential elections. For example, in Croatia's 2005 presidential elections, extra-territorial kin-citizens forced a run-off election between the incumbent president and the more nationalist Croatian Democratic Union (HDZ) candidate, a party favoured by kin-citizens, especially ethnic Croats residing in Bosnia Herzegovina (BiH). 8 Analysing this puzzle of extra-territorial kin-citizen participation, this article uses the case study of Moldova to analyse those who have "reacquired" Romanian citizenship, and unpack whether, how, and why Romanian kin-citizens within Moldova participate in Romanian elections.
The article adopts an inductive agent-centred, bottom-up approach, to analyse the lived experiences of, and motivations for, extra-territorial political participation. This complements existing top-down, institutional approaches by offering a deeper exploration of these extra-territorial political practices. This article is concerned with more than the affective side of voting, in terms of the lived experiences of political participation. It seeks greater conceptual insight into why citizens vote in states they do not reside in nor have ever resided in, and aims to inform studies of extra-territorial voting, which are largely top-down, as well as engage in more general political science debates about political participation (debates of costs, benefits, and duties).
The article finds a reciprocal relationship between the kin-state, facilitating citizenship as a right, and the citizen, performing their citizenship duties by voting. In other words, by constructing voting as a duty attached to the practice of extra-territorial citizenship, individuals gain legitimacy in performing as voters and, hence, as more legitimate citizens. This supports, empirically and conceptually, arguments in political science 9 that stretch beyond extra-territorial examples about why people vote, signalling the need to look beyond rational choice assumptions of costs vs benefits to consider the ideas behind voting and, in particular, the notion of voting as a duty and the inherent legitimacy of this construction, which ties notions of citizenship to voting.
In the rest of the article I, first, outline Romania as a case of extra-territorial politics. Second, I review two areas of literature: political participationto consider voting in terms of rights, benefits, and dutiesand diaspora votingto consider the extra-territorialization of voting, as well as the need to consider not diaspora voters but also kincitizens. Third, I discuss the contribution of studying extra-territorial participation inductively and from the bottom up, in terms of the experiences of, and motivations for, voting in extra-territorial elections. Fourth, I review the methodology of the article to discuss how respondents were selected and interviewed. Fifth, I move to the empirical material to focus on three dimensions of extra-territorial voting that emerged inductively from interviews in Moldova (voting obligations, preferences, and intentions). The article concludes by emphasizing the importance of considering the duty of voting as an important element of voting, even for those who have never resided in the state in which they can vote.

Romania: a case of extra-territorial politics
This article analyses the extra-territorial political participation of Romania's kin-citizens within Moldova (who have recently become, or are becoming, Romanian citizens). Since 1991, Romania has facilitated citizenship "reacquisition" (redobândire) for those residing in former territories of Greater Romania (1918Romania ( -1940, Bessarabia (most of Moldova), and Bukovina (a region of Ukraine), including those who can prove they descended (up to third generation) from citizens of former Greater Romania. 10 This has permitted large numbers of Moldovan residents to reacquire Romanian citizenship, while retaining Moldovan citizenship (since 2003) and residing in Moldova. 11 This reacquisition of citizenship facilitates voting rights in Romanian elections, within Moldova, on the same basis as Romanian migrant diaspora (for example, voting abroad in polling booths in the United Kingdom, France, Italy, and Spain). These diaspora voters (both migrant and kin-citizens) are institutionalized within external constituencies: Romania has four external constituencies for deputy elections (lower house), allocated geographically, and two external constituencies for senate elections (upper house), giving migrants and kin-citizens the right to decide political outcomes in these constituencies.
After 2007, when Romania acceded to the EU, the numbers reacquiring Romanian citizenship increased in Moldova (although it is unclear by how much). 12 Alongside this expansion of Romania's citizenry, the number eligible to participate in Romania elections and participating in Romanian elections increased substantially (Figure 1), even between the first and second round of the 2014 presidential elections. Despite declining domestic support, Moldovan voters have shown continued electoral loyalty to Romania's President Băsescu (2004, notably in the 2009 presidential election (Figure 2), and his affiliates, such as Eugen Tomac (2012 parliamentary elections, Figure 3) and Viorel Badea (P-DL senator for external constituency including Moldova). 13 Moldova dominates its external Romanian constituency overwhelmingly (in comparison to other diaspora voters), particularly in the lower chamber elections. In 2012, 94% of votes in the second overseas constituency (Eastern Europe and Asia), were from Moldovan polling booths, 82% of whom voted for Tomac, securing his victory ( Figure 3). 14 Romanian politicians have also begun to "pay greater attention to the voting power of the Diaspora", in particular since external constituencies were founded in 2008. 15 They have opened local offices in Chișinău, Moldova's capital, for  Romanian politicians, primarily from the Social Democratic Party (PSD) and Democratic Liberal Party (P-DL), and fielded local candidates. 16 Romania therefore provides an interesting case, theoretically and empirically, to unpack extra-territorial voting practices. Romania is neither an exception in facilitating the right of extra-territorial citizens to vote (about 100 states currently do), nor in facilitating extra-territorial constituencies (for example, Ecuador, France, Italy, Portugal, Croatia). However, it is one of the "most permissive" states for facilitating diaspora voting rights which, Burean argues, stems from a normative stance of improving "democratic quality" via more inclusiveness. 17 More cynically, Romania, alongside other kin-state cases, including now Hungary (since 2014), view diaspora as electoral capital. Alongside the continued modification of electoral rules to suit the interests of the incumbent regimes, 18 Romania's diaspora form an important battleground in this popular franchise, not least because they have been electorally decisive (like Croatia), allowing Băsescu (P-DL) to win Romania's 2009 presidential elections. 19 However, the motivations of those participating in Romania's largest, and most expanding, enfranchised extra-territorial citizenry remain under-scrutinized.

Theorizing extra-territorial citizenship and political participation
To unpack this kin-citizen puzzle of extra-territorial political participation, this section addresses two areas of literature. First, I review political participation literature to consider voting in terms of costs, benefits, and duties. This deductive literature provides a theoretical basis to understand political participation while it contrasts with the inductive approach of this article, which aims to see how costs, benefits, and duties are discussed in practice.
Second, I review citizenship literature to consider how the demos is becoming increasingly extra-territorialized. This literature offers an overly migration-centred account, overlooking the enfranchisement of kin-citizens, who have more tenuous ties to kin-states, and kin-state elections.

Political participation
Theories of voting within political science have long pondered the (ir)rationality of voting because it is "rarely of self-interest to vote". 20 Downs conceptualized voting in terms of the costs (C) vs benefits (B), offset by the probability of an individual's vote being decisive (P) 21 and argued that voting was irrational because the costs exceeded the expected returns/benefits. The "paradox" of voting is, thus, that we observe individuals, en masse, participating in elections even if, according to rational choice, this action seems "irrational". 22 Since Downs, explaining this paradox has taken on numerous deviations, including empirical testing of whether socio-economic resources might affect the costs of voting and interest in politics, 23 whether electoral systems and district magnitude might affect how far votes are considered "decisive", 24 the significance of the election, 25 as well as other social and political factors (mobilization potential of politicians, social networks). Beyond these factors, and of greatest interest to this article, is the explanation that voting incurs a sense of "duty" where individuals feel they should participate, or achieve a sense of gratification, by voting which might offset the costs of voting. 26 Blais situated this sense of gratification and dutythe idea that voting is a correct practiceoutside the rational choice model, arguing that individuals vote because they are pledging support of certain political values, notably democracy and their right to participate. 27 Testing this cross-nationally, Blais argued from an empirical perspective that, at best, rational choice (costs vs benefits) is a partial explanation of voting alongside the sense of duty. Critically, Blais finds that students who have never voted (because of ineligibility) still express their sense of duty to vote.
Though deductive, this literature provides a conceptual framework for inductively analysing why kin-citizens might vote in extra-territorial elections. For example, the benefits of participation would appear to be more remote when individuals vote in an external (kin-)state. Moreover, the sense of duty to participate would also seem less applicable to extra-territorial elections, where individuals may be less committed to uphold democracy in an external state.

Diaspora vs kin-citizen political participation
In considering political participation, it is also necessary to consider the extra-territorialization of citizenries and political participation. Second, it is necessary to move beyond migration-centred understandings of extra-territorial voting to address the gap in understanding kin-citizens' extra-territorial political practices (Table 1).
With increasing migration, citizenship has become a transnational institution as an instrument of exclusion and inclusion, determining who can access the state they reside in (as immigrants) and retain ties to the state they have emigrated from. 28 During the twentieth century, states became more tolerant of allowing, and facilitating, the holding of dual and multiple citizenships to enable individuals to form and retain ties and participate in the states to which they have affected interests. 29 From an international principle of "one nationality only", 30 where dual citizenship was conceived as a form of "polygamy", dual citizenship became de-securitized, with multiple ties of citizenship no longer perceived as a threat to the state. 31 Citizenship, via its increasing inclusiveness, demonstrates a cosmopolitan postnationalization of the state, which in turn implies a de-ethnicization of citizenship and de-territorialization of the state. 32 Here, political boundaries are no longer the regulatory mechanism of "membership"; instead states are becoming "fuzzy" entities, by granting political and social rights, via dual citizenship, to migrants and, crucially, without requiring migrants to renounce their original citizenship. 33 With citizenship (in most cases) providing the gateway to enfranchisement (that is, the right to vote), on the basis of "one citizen, one vote", 34 the demos too has become de-territorialized and extra-territorialized. Lafleur argues diaspora enfranchisement occurred through democratization and the normalization of the right to vote externally. 35 There is a wide variation in terms of these diaspora voting practices, in terms of low and high turnout, and the marginal vs greater impact of these practices. While in some cases diaspora votes can determine electoral outcomes (for example, the Senegalese presidential elections in 2000), 36 Hutcheson and Arrighi argue these cases, though infamous, remain exceptional because, predominantly, the electoral impact of diaspora voters is low because of low turnout. 37 Alongside these post-national policies of migrant inclusion, kin-states have advanced citizenship acquisition policies for external communities, allowing kin to become kin-citizens and voters. Ragazzi and Balalovska argue this type of citizenship expansion does not resemble a more inclusive "cosmopolitan" post-nationalization of the state, but a post-territorialization of the state, by advancing preferential ethno-cultural privileges to co-ethnic communities. 38 This facilitated kin-state citizenship rarely concerns "cosmopolitan norms" but, instead, advances a policy of extra-territorial nation building. 39 For example, it is often populist and right-wing regimes which expand kin-citizenship rights to win domestic and co-ethnic support via the creation of new extra-territorial/co-ethnic electorates. 40 However, so far, much like the discussion for citizenship, research has failed to disentangle migration-centred/post-national voting practices and kin-state voting practices (Table 1). It is important to consider these political dynamics together, given that states can be both reaching out to far away diaspora and nearby kin-community and often have similar strategic logics underpinning this engagement (for example, the potential "resources" to be gained from these external communities). 41 However, from the bottom up, it is important to consider the different dynamics of diaspora and kin-citizen political participation because of different socialization experiences of these communities vis-à-vis the state in which they are voting. For example, diaspora voters have been socialized within their home states, to which they retain voting rights, where their political preferences demonstrate pre-existing voting cleavages formed while residing in their home states. 42 However, kin-citizens, newly enfranchised within kin-states, have not experienced the same socialization; instead, they create political ties and form preferences, tabula rasa, vis-à-vis kin-states, reinforced by their extra-territorial residence. Normatively, how far kin-citizens have "affected interests" vis-à-vis kin-states can be disputed because they fail to demonstrate a stake in the kin-state by virtue of their continued residence outside of the kin-state. 43 However, kin-states such as Romania and Croatia offer an emotive and reparative justification for kin-citizen enfranchisement, where citizenship is both a form of "moral compensation" and reward for past contributions. 44 This logic needs to be combined with a more cynical analysis of domestic politics, where this kind of enfranchisement (via citizenship facilitation of co-ethnic communities) is advanced primarily by right-wing populist governments in states with dynamic election rules to use post-territorial nation-building as a mechanism also of electioneering by creating sympathetic new co-ethnic citizens, and hence, voters. 45 Empirically, kin-citizen voting practices remain more contentious and influential, such as in Romania. Croatian extra-territorial kin-citizens, especially in BiH, have expressed clear and consistent electoral support for the more nationalistic HDZ. 46 These co-ethnic communities came to "make or break candidates" in Croatian elections (especially 2005 and 2007) regardless of low turnout. 47 Second, Croatia's enfranchisement of external co-ethnic Croats contrasts with policies that inhibited the political participation of non-ethnic Croats within Croatia, demonstrating a tension (or preference) for kin-citizen inclusion with immigrant exclusion.

Researching kin-citizens' political practices
Thus, it is necessary to study kin-state citizenship and voting practices separately to diaspora voting practices, at least from the bottom up, given the different political dynamics driving the expansion of kin-state citizenship and enfranchisement, compared to similar policies for diaspora. Understanding the link between becoming a citizen and extra-territorial political participation is important to decipher the relationship between citizenship as a status (something to possess) and citizenship in itself as participation (incurring a sense of duty to behave as a citizen through participating, for example in elections). However these questions are often centred on states, for example why do states enfranchise external voters?, 48 rather than on agentswhy do individuals (kin-citizens) participate in extra-territorial elections?returning to the ideas raised by the theoretical literature on political participation, where voting is (or is not) a calculus between costs, benefits, and duties. Thus, it is important to study these voting practices directly, by analysing the agency-centred demand-side perspective of these extra-territorial practices, unpacking whether, how, and why kin-citizens participate in kin-state elections.
Voting too, as a political practice, is usually studied deductively, via theoretical, historical, statistical, or "institutional structural comparisons", 49 to "explain and understand" voting preferences via "quantification", 50 and to understand how far rationality, "incentives structures", or values and norms affect these preferences. 51 This article adopts a less common inductive and interpretivist approach by focusing on the meanings and experiences of political participation. 52 Advocating for interpretive approaches to voting, Bevir and Rhodes criticize using statistical models and surveys to assume there is a "correlation or deductive link" between "beliefs and practices", whereby studying practices captures beliefs. 53 Instead, Bevir and Rhodes argue that "beliefs and practices" are "constitutive of each other" where practices of voting can only be captured by understanding, and engaging with, the "intersubjective beliefs that underpin the practice". 54 Thus, this article analyses extra-territorial voting practices, inductively and from the bottom up, to understand the experiences of, and motivations for, participating in extraterritorial Romanian elections, to understand how and why individuals become, and are, engaged with Romania, via political participation. Here the purpose is not to understand only the affective side of voting, in terms of the experiences and performances of voting, but to determine why new citizens might be compelled to vote, who they might vote for (their preferences) and what might constrain their voting practices (costs), to weigh up their sense of duty to participate against mediating factors, such as the formation of political preferences and the costs of voting (for example, time at the polling booth).

Methodology
The article uses semi-structured interviews that I conducted face-to-face with respondents in Chișinău, Moldova's capital city (2012)(2013). A subset of these interviews (40/55) is used in this article, where I selected according to whether they held or had applied for Romanian citizenship (15/55 respondents expressed no interest in applying for Romanian citizenship). These interviews formed part of a larger project to analyse everyday experiences of being Romanian, and engaging with Romanian practices (for example, citizenship acquisition, voting, scholarship programmes), where the criteria of analysis is identifying (culturally, politically, ethnically) either as Romanian and/or Moldovan, leaving out those who primarily identified with minority ethnic groups (for example, Ukrainian or Russian in Moldova). Throughout the empirical section below, respondents are numbered, for example MD-1; a full list of respondents is located in the supplementary materials (Appendix Table A1).
This fieldwork took an interpretive approach, where the focus was on everyday experiences of identification and citizenship acquisition to gain "experience-near"/ emic understandings of identities, institutions, and concepts, 55 as opposed to deductive, top-down or "experience-far" concepts. That respondents discussed their political practices was something that "emerged" during initial fieldwork, alongside the observation of Romanian political activities in Moldova, such as the visit of President Băsescu in 2013 and the opening of offices of Romanian politicians and parties on both sides of the left-right Romanian political spectrum (2012-2013).

Interview guide
During the face-to-face semi-structured interviews, which were conducted in whatever language the respondent preferred (Romanian, English, or Russian), I adopted a consistent approach in asking similarly constructed questions across respondents to ensure the comparability of data collected. To capture respondents' experiences of political participation I asked those with Romanian citizenship, first, "What can you do as a Romanian citizen?" to establish how far respondents were aware they could vote and/or whether they volunteered this information. This was followed by more substantive questions about whether they could vote and, if they believed they could, whether they had voted or would vote. Lastly, respondents were asked to explain their rationale, to unpack why and how respondents chose to participate, or not, in Romanian elections.

Respondent selection
This article does not claim to be representative because the number of respondents is neither large enough nor sufficiently random. 56 In accessing respondents, I engaged with a breadth of "multiple perspectives" and "contradictory narratives" 57 concerning their in-depth experiences and practices of everyday extra-territorial citizens in Moldova, as opposed to expert or elite opinions. The everyday focus of the project guided me to seek respondents across the political spectrum (for example, youth wings of political parties), youth and student organizations not directly involved in politics, as well as those not affiliated to organizations, via university networks and contacts (Table 2, Figure 4). The number of respondents from these different categories is small (for example, from different political parties); rather, this breadth enables diversity and contrast within the sample and this explorative research, as opposed to enabling an analysis of causal inferences, where certain characteristics are the cause of these inductive insights; instead, these could be a direction for future research.
A large number of potential respondents were contacted by phone or email, either by cold-calling (by contacting without recommendations based on internet sources of organizations) or snowballing (using existing contacts and previous respondents' recommendations). I did not choose respondents as much as choose types of respondents (for example, students) from which I contacted large numbers (for example, via student organizations listed online), seeking their response before arranging interviews. Even in terms of identity, where I consider only those who identified as Romanian and/or Moldova (in this article and the larger project), I did not choose respondents based on identity but rather only analyse respondents fitting this criteria. Pragmatically, the focus came to be predominantly on the post-Soviet generation (that is, those who had spent the majority of their life in post-Soviet Moldova), since these individuals were easier to access (via the phone, internet, and snowballing) and build a rapport with.

From kin-citizens to kin-state voters
As discussed above, the right to vote in Romanian elections is acquired when individuals reacquire Romanian citizenship, via the practice of citizenship re-acquisition (redobândire). Redobândire was instrumental in transforming respondents into political actors in the Romanian state from within Moldova, via enfranchisement that was packaged within citizenship practices. This section explores how respondents used this right, and their experiences and sentiments attached to extra-territorial voting practices. The article focuses on three aspects of participation in Romanian elections which demonstrate the different dimensions of how voting was discussed by respondents: (1) obligations, (2) preferences, and (3) intentions.
In terms of obligations, the article considers whether and how individuals engaged with voting, as a duty derived from citizenship. This focus on obligations links to theorizing of turnout, as explained by a "sense of duty", and theorizing of citizenship, conceptualized as an interweaving of rights from the state and duties towards the state; both of these factors can be observed via respondents' explanations of this sense of obligation, derived from citizenship, to participate. In terms of preferences, the article unpacks how respondents expressed sentiments of who they would like to vote for and why they expressed these preferences (for example, self-interest, interest for Moldova). Finally, for intentions, the article unpacks respondents' practice of voting and what might mitigate intentions to vote (for example, availability of voting booths).
Each dimension is now considered in turnas well as the interaction between these dimensionsto unpack how a sense of obligation expressed by respondents was mediated by voting preferences (whether respondents have or have not formed voting preferences and how they formed these preferences) and constraints (how much time it requires to participate). Respondents' expression of a "duty to vote" remains an interesting, and relevant, conceptual contribution, demonstrating a reciprocity of kin-state citizenship, where citizenship as a right becomes performed through voting as a duty.

Voting obligations
That new Romanian citizens in Moldova felt an obligation to vote was a motivation that arose inductively from interviews. During interviews, respondents were asked whether and why they participated. Indeed, the extent to which many respondents expressed such an obligation and their rationale for this framing was a surprise. For example, several described how they felt a "big responsibility" to vote because they "enjoy" Romanian citizenship and want to "contribute, get involved with the Romanian state" to help their "mother state" (patria mama) have a good future .
This expression of obligation allowed respondents to feel more legitimate as Romanian citizens. Being able to vote, and engaging in this practice, allowed these young respondents who had acquired Romanian citizenship to see themselves neither as "materialist" nor "as some opportunists" who gained Romanian citizenship wholly for pragmatic reasons . Rather, they were using Romanian citizenship as a "right to express themselves as citizens" of Romania, as anyone from within Romania would do .
Moreover, several respondents explained that it was not a volition, as they did not "want too much to vote" (own emphasis) . Rather it was the attachment to the sense they had a "duty" and "responsibility" to vote that overtook their volition, by the normative compulsion to feel that if the state "gave me these opportunities" then "I want to offer something back, something good" .
This emotional content of redobândire demonstrated how far this practice resulted in becoming "more involved emotionally" in Romanian politics, as a female student expressed . This emotional impact of citizenship acquisition has often been overlooked, at least by political studies. 58 Yet, respondents indicated that they developed an emotional connection to Romania, and a sense of legitimacy for this connection, even if this did not alter how they self-identified. This demonstrates too how citizenship practices created political actors because it became more "legitimate" for them as Romanian citizens "to have a voice" in Romanian politics and to express this by voting .
However, there were respondents that did not subscribe to this idea of obligation and legitimacy. For example MD-50, a male student in Romania, explained his ideological restrictions from not wanting "to decide in a country in which I don't know too much" because he "wouldn't like to get involved in another country's politics". Thus, he felt an obligation not to vote, because he did not believe he was sufficiently involved in Romanian political and social life to participate and to influence the outcome of Romanian elections.
Overall, most respondents explained their sense of duty to vote, demonstrating their sense of compulsion and legitimacy derived from this practice. They felt a responsibility to give something back to the state that had given them this right and responsibility to show that their motivations underpinning citizenship were not simply material selfinterest. However, their interests were not "affected" so much by Romania, even if they thought they were gaining the tools to express their interests vis-à-vis Romania, because their affected interests remained within Moldova, where they continued, and intended, to reside.
Theoretically, this sense of obligation to participate demonstrates the presence of a sense of "duty" to vote. 59 However, this was also a different expression of duty, expressed not in terms of upholding the values and system of democracy, as Blais argues, but rather by legitimizing their citizenship practices. This obligation to vote converted them, as a status, to active and legitimate citizens, rather than material citizens, who felt indebted to the state from which they gained rights, demonstrating their active engagement in a rights/duties relationship with the kin-state.

Voting preferences
Alongside this normative framing of a duty to vote, respondents discussed who they would vote for. They demonstrated a cleavage between those who named Băsescu as their preferred candidate and those who were more unclear as to who they would vote for.
On the one hand, several respondents expressed an open and avid desire to vote for Băsescu. More interesting were the reasons underpinning this preference. They felt personally grateful to Băsescu, attributing the policy of redobândire and the facilitation of this policy to Băsescu, who they believed had deliberately made redobândire "easier" because he was "pro-passports" . Beyond these policies, these respondents felt that Băsescu "likes Moldova" and "gives a feeling of belonging" to them as Romanians in Moldova . Indeed Băsescu was framed as being so popular that he could in fact win the Moldovan presidency , though Moldova currently does not have a popularly elected president.
Many felt it was important to keep Băsescu in office by voting for him to ensure the Romanian President remained "more open for our, for my country [Moldova]", as a young male NGO worker remarked . Their preference for Băsescu stemmed from an obligation to support the politician who had helped them and to ensure Băsescu remained in office, to ensure the continued facilitation of redobândire and more favourable, from respondents' perspective, Romanian policies towards Moldova.
By contrast, these respondents had little trust for Romania's opposition (PSD): Ponta (the leader of PSD, and 2014 presidential PSD candidate) was "more measly [sic] in terms of the relationship with us" so that Moldova's "privileged" status vis-à-vis Romania was only "as long as Băsescu is in power" . This contradicts Muntean, Pop-Eleches, and Popescu's argument, in which PSD is seen as the legacy of the Romanian Communist Party, as Romanian diaspora elsewhere do, by demonstrating a different dynamic among Romanian voters in Moldova, premised more on continuity and self-interest rather than objecting to the PSD's links to Romania's Communist Party. 60 It demonstrates also a different socialization into Romanian politics and the effects this has on voting preferences.
It was crucial that voting in Romanian elections did not contradict their participation in, or threaten their sense of loyalty towards, Moldova. Respondents believed they were voting in Moldova's interests by supporting a Romanian presidential candidate they thought was best for Moldova and themselves. This goes towards explaining why Băsescu (and his affiliates) has been the preferred candidate within Moldova, even at a time of declining support within Romaniabecause of Băsescu's association with the facilitation of citizenship reacquisition which generated a personality cult surrounding Băsescu, who was seen as the best candidate for them and Moldova.
By contrast, other respondents found it harder to voice and/or form political preferences towards specific candidates. As several respondents explained, they were not "so involved" in Romania's political life and "wouldn't be able to tell the difference" between different candidates and parties, sufficient to express or form a preference for a candidate . MD-34, a professional in his 30s, discussed how a candidate could become preferred: "if somebody would really appeal to me, to the needs of me as a Romanian citizen living abroad", such as more "cooperation between these two, our two countries". This indicated that preferences were formed based first on what was best for individuals and second for Moldova, demonstrating the connection between extra-territorial voting practices (in the kin-state) and their concern for the state in which they resided (home state).
Several respondents were critical of those voting in Romanian elections, that they did not vote "in an informed way" but rather "in an emotional way" . MD-11, a young businessman involved in politics, believed this made respondents more likely to vote for Băsescu, because of his visibility and popularity in Moldova. Rather than caring which "color" or ideology they voted for, Moldovans associated Băsescu personally with Romania's preferential attitude towards Moldova .
Respondents' sense of distance from Romania and Romanian politics cemented this informational deficit and encouraged an emotional approach to voting. For example, MD-24, a young professional who was moving to Romania, believed that once she had lived in Romania she would be "a lot more aware of who are the main political actors" and would change from expressing her preference as a collective "family decision" to an individual practice because she would "know more what is my personal opinion about who I should vote for". This demonstrates the power of family networks in socializing voting preferences, constructing voting as a collective practice in the absence of sufficient information to contest this collective preference. It was only by exiting the extra-territorial community and residing in Romania that MD-24 believed she could vote as an informed and affected actor in Romanian elections (by no longer participating as an extra-territorial voter).
This section has discussed the extent to which Băsescu and his political affiliates were the explicit and implicit choice of a significant proportion of respondents, not only because of his visibility, but because of his personal association with a policy the respondents benefited from. Even if respondents had not explicitly formed their preferences, their explanation of how they would form their preferences indicated the self-interest shaping their preferences to the extent of wanting to choose a candidate that would be beneficial for them as extra-territorial individuals, and as an extra-territorial community and constituency as well as their sense of duty to Băsescu as the direct facilitator of their right to reacquire Romanian citizenship.

Voting intentions
While respondents expressed their sense of obligation to vote and their preferences as to whom they would vote for, these were offset by the costs of participating. First, respondents described the number of polling stations as "too small" to accommodate the numbers that wanted to vote . As MD-33, an NGO worker in his 30s, described, he did not want to "spend time on election day waiting for the entire day in line".
Even if the number of polling stations has increased in Moldova over time (Figure 1), this increase was not sufficient to meet the increasing demand to participate, based on the increasing number of eligible voters. This provided a barrier to those actually participating regularly in Romanian elections by taking the costs of participation beyond what many respondents were willing to bear. This framed the normative sense of responsibility to vote in theoretical terms when in practice fewer respondents would vote than portrayed it as a duty to vote unless it was made easier and less time-consuming, as MD-37, a young female professional, remarked.
Beyond issues of voting supply, other respondents were more uncertain about whether they were eligible to vote. Again, MD-50 did not think he should be able to participate in a state in which he did not feel politically involved. This was accompanied by the uncertainty as to whether he could actually vote with "only a passport" or whether he needed to have an identification card [also . 61 This contrasted to the majority of respondents with Romanian citizenship who were more aware of their simple eligibility to vote, that is, that Romanians did not require an "internal" passport (which does not exist), ID card, or evidence of Romanian residency to vote, since reacquisition of Romanian citizenship and evidence of holding a Romanian passport were sufficient to register to vote. That there were several respondents who were uncertain of their eligibility to vote demonstrated a further informational deficit among the body citizenry of Romanian/Moldovan dual citizens residing in Moldova.
As much as respondents expressed, normatively, their sense of obligation to vote and to give back to Băsescu by voting for him, there were barriers that had and might inhibit their participation in Romanian elections arising from the number of polling booths and, for some respondents, a lack of clarity concerning their eligibility to vote. The increased presence of Romanian politiciansof the left and right of the Romanian party spectrumvia local visits and offices, would indicate a desire to shore up this informational uncertainty regarding eligibility and candidate preference. However, the key constraint remained the available supply of voting opportunities, an issue of concern not just for Romanian extra-territorial politics in Moldova but for the enfranchisement of Romania's diaspora more widely.
This article discussed the extra-territorialization of Romanian politics via the creation of an extra-territorial citizenry, and demos, in Moldova. Using a rich inductive and agent-centred approach, the empirical section avoided making causal claims which linked who the respondents were and how this affected their experiences and motivations. Rather, the article addressed three dimensions that arose from across the sample (regardless of gender, age, profession, or engagement in Moldovan politics): voting obligations, preferences, and intentions.
In terms of obligations, the article argued that political participation via voting was framed as a necessary obligation stemming from citizenship. In terms of preferences, respondents wanted to give back to those who directly attributed to facilitating citizenship/redobândire (Băsescu). Băsescu could construct himself as a charismatic figure of extra-territorial Romanian politics within Moldova, against a distrust of other figures, such as Victor Ponta (PSD), who respondents did not believe would advocate enough for Moldova. This charisma honed by Băsescu was cultivated while suffering from a lack of popular support at home, where Băsescu faced two impeachment referenda. However, in terms of intentions, these practices were constrained by the opportunities and costs of voting (for example, time), resulting from the insufficient provision of polling stations in the Moldovan constituency. The number of polling booths remains a key constraint of the actualization of intentions, that is, the transformation of an individual who wants and feels obligated to vote into an individual who actually participates in this political practice.

Conclusion
This article argues for understanding the duty to vote expressed by kin-citizens as well as how this can be mediated by (a) a lack of supply of voting opportunities (polling stations) and (b) a lack of information, because political socialization occurs outside of the kin-state. This article provides useful insights also for analysing extra-territorial citizenship by demonstrating the performance of citizenship not (only) as a material practice, as citizenship acquisition from kin-states has often been conceived, but more as exemplifying the sense of rights and duties. 62 This duty to vote, even if it is offset by voting supply and information deficits, demonstrates a reciprocal relationship between the kin-state facilitating citizenship as a right and the citizen performing their citizenship duties by voting: kin-state:duty of citizenship ↔ kin-citizen:duty of participation These findings speak unexpectedly to the deductive hypotheses concerning voting practicesconceived as benefits, costs, and dutiesdiscussed since Downs, and Riker and Ordeshook, and agree too with Blais' analysis of voting as a duty. However, this sense of duty was expressed differently from Blais' concept, who argued that it concerned a desire to uphold values of democracy through participation. Rather, this article has shown the sense of obligation of citizenship, expressed as reciprocal rights and duties, rather than as political values.
The key difference between this article and previous analyses is the extra-territorial dimension, where those voting acquired citizenship from a state in which they did not reside. This differs in important ways from diaspora analyses of citizenship. Rather than demonstrating existing "domestic voting cleavages"which is argued to not change the polity 63kin-citizens do redefine the existing polity as actors who are creating political preferences following citizenship acquisition. This creates kin-citizens with an obligation to participate but with different concerns and interests (for example, interests linked to their home state) to resident citizens of the kin-state, and demonstrates a redefinition of existing polities and a willingness of politicians to engage in this process of redefinition because of the votes that can be won by doing so. Thus, it is important to go further in exploring the political practices of these co-ethnic communities who stand apart from diaspora voting practices explored elsewhere, and demonstrate that with the extra-territorialization of citizenship 64 comes the extra-territorialization of political participation, and thus the demos and site of electoral politics. Moreover, enfranchisement of kin-citizens is expanding, for example with the upgrading of Hungary's Status Law to citizenship status for external communities of co-ethnic Hungarians (for example, in Romania and Slovakia).
This article demonstrates the importance of a sense of duty to participate for kin-citizens and serves as a starting point for future research beyond the cases of Romania and from an agent-centred perspective; how and why those who have acquired kin-state citizenship participate in kin-state elections. While this finding is empirically restricted to the context of this extra-territorial puzzle, this bottom-up approach provides insights that could be tested deductively and comparatively over time and across different cases, by trying to systematize the sense of duty both in terms of domestic and extraterritorial political practices.