Cross­cultural implications of linguistic future time reference and institutional uncertainty on social entrepreneurship

Research Summary: Using a sample of 205,792 individuals in 70 countries with 39 languages, this paper presents novel empirical evidence for how a language's future time reference , defined as the requirement that speakers mark time in the future, affects a speaker's likelihood of engaging in social entrepreneurship. FTR subtly shapes a speaker's temporal orientation, such that speaking a futured language (i.e., strong FTR) favors a short-term orientation which positively affects the likelihood of being a social entrepreneur. Furthermore, institutional uncertainty arising from weakly entrenched institutions moderates this relationship. Individuals who speak futured languages in contexts characterized by regulatory institutional uncertainty (weak rule of law, weak property rights, and strong corruption) are more likely to engage in social entrepreneurship. Theoretical and practical implications and future research directions are discussed. Managerial Summary: This paper examines the influence of language and regulatory institutions on an individual's decision to engage in social entrepreneurship. We show that entrepreneurial strategies towards social value creation are more likely to be pursued in countries where language is characterized as futured (i.e., using sentences with “ will/ shall + infinitive ” ) with the linguistic

reference (FTR). Moreover, the positive impact FTR has on the strategic decision to become a social entrepreneur intensifies when there is regulatory institutional uncertainty. That is, the probability an individual is a social entrepreneur increases when there is weak rule of law, weak property rights, and strong corruption within a country. In sum, our study highlights the important interplay between language, an overlooked cognitive institution, and regulatory institutions in shaping entrepreneurial behaviors and social outcomes.

| INTRODUCTION
Social entrepreneurship is understood as any venturing effort that aims to create and leverage value for diverse stakeholders affected by persistent and unresolved social problems (Dorado & Ventresca, 2013;Hervieux & Voltan, 2018). Social entrepreneurs are important change agents who can generate substantial positive outcomes (Dacin, Dacin, & Matear, 2010;Terjesen, Bosma, & Stam, 2016). Prior work documents that social entrepreneurship varies widely across countries. While some societies benefit from many social entrepreneurs, other societies have a shortage of them, which raises questions about what key factors drive individuals to start and run social ventures (Stephan, Uhlaner, & Stride, 2015).
We suggest that temporal orientation may explain cross-cultural differences in social entrepreneurship. Prior research shows that societies experience and perceive time differently (Venaik, Zhu, & Brewer, 2013). Specifically, while some societies have a temporal orientation that sees the future as distant, other societies see the future as something temporally present, with consequences for attitudes, intentions, and behaviors (Chen, 2013a). Temporal orientation captures how individuals tend to think about subjective time (Tang, Richter, & Nadkarni, 2020). Subjective time is a qualitative notion of time that relates to how individuals experience and perceive time (Lévesque & Stephan, 2020;Shipp & Jansen, 2021;Zimbardo & Boyd, 1999). Subjective time is now garnering renewed attention in entrepreneurship (Lévesque & Stephan, 2020;Wood, Bakker, & Fisher, 2021), nearly two decades after seminal contributions highlighting its theoretical relevance (Bird & West, 1997). Yet, we still do not fully understand how subjective time affects social entrepreneurship. Examining cross-cultural differences in how individuals intersubjectively interpret subjective time within the domain of social entrepreneurship is important because an individual's understanding of subjective time governs most human thought and action (Shipp & Jansen, 2021). Given these considerations, we focus on the following question: How does temporal orientation affect the likelihood that an individual engages in social entrepreneurship?
Regrettably, the vast social entrepreneurship literature (Gupta, Chauhan, Paul, & Jaiswal, 2020;Saebi, Foss, & Linder, 2019) has overlooked subjective time perception as a potential causal mechanism. We start this conversation by drawing on weak neo-Whorfianism to highlight the role of language in shaping the temporal orientations of individuals cross-culturally. Weak neo-Whorfianism broadly claims that the grammatical structures of our native language subtly influence the way we understand reality and make sense of the world, that is, our cognitions or the way we think (Dyke, 2021;Montemayor, 2019). Using weak neo-Whorfianism, we examine how the linguistic feature of future time reference (FTR) affects engagement in social entrepreneurship cross-culturally (Chen, 2013a;Kim, Jeong, & Jun, 2021;Roberts, Winters, & Chen, 2015). FTR captures how a language grammatically marks the future and nonfuture. Specially, strong FTR languages oblige speakers to use future tense (e.g., will/shall) to distinguish between the present and the future, whereas weak FTR languages do not oblige speakers to do so (Chen, 2013a). We theorize that FTR shapes an individual's temporal orientation, favoring either a short-term orientation (i.e., "strong" FTR) or a long-term orientation (i.e., "weak" FTR) (Jäggi, Sato, Gillioz, & Gygax, 2022;Park et al., 2017). We argue that the grammatical obligation to use the future tense to describe future events (i.e., strong FTR) subtly influences individuals to temporally discount the future. In other words, strong FTR speakers perceive future rewards as temporally distant and less valuable, leading to a short-term orientation that is reflected by a preference for proximate outcomes over distant outcomes. Given that social entrepreneurs address persistent and unresolved societal challenges, they want to solve societal problems as soon as possible in the present rather than in the far future. Hence, we hypothesize that strong FTR positively affects individuals participating in social entrepreneurship because they favor achieving proximate benefits for others affected by social dilemmas while discounting the long-term costs related to venturing.
However, engagement in social entrepreneurship is a function of an individual's entrepreneurial vision within a particular institutional matrix that can facilitate or constrain action (Estrin, Mickiewicz, & Stephan, 2013). Thus, we integrate new institutional economics with weak neo-Whorfianism to also question: How does the institutional environment shape the relationship between an individual's temporal orientation and engagement in social entrepreneurship?
We attempt to answer this question by presenting a model of embedded agency where institutional uncertainty within the institutional environment moderates the relationship between how individuals subjectively understand time through language (i.e., FTR) and their engagement in social entrepreneurship. The institutional environment is comprised of formal (e.g., legal enforcement mechanisms) and informal (e.g., market norms) elements, or "rules of the game" (North, 2005), that shape incentive structures for social entrepreneurs. Institutional uncertainty stems from unstable and unreliable "rules of the game" (North, 1990) that produce widespread "market imperfections, inadequate information flows, and fragile legal … frameworks" (Ge, Carney, & Kellermanns, 2018, p. 3). Using new institutional economics (North, 1990;Williamson, 2000), we examine two competing theoretical logics for how institutional uncertainty affects social entrepreneurship. We argue that different incentive structures operate as a distal social mechanism to shape the effect between temporal orientation and the likelihood of engaging in social entrepreneurship. First, the institutional uncertainty argument (e.g., Dacin et al., 2010;Mair & Martí, 2009;Mair, Martí, & Ventresca, 2012) suggests that social entrepreneurship increases as a response to alter unpredictable "rules of the game" when institutions fail to address societal issues (e.g., poverty, discrimination, human rights violations).
Second, the institutional clarity argument (e.g., Estrin et al., 2013;Stephan et al., 2015) suggests that social entrepreneurship increases as a response to abide to well-entrenched and enforced institutions because they create more stable and predictable "rules of the game." To examine these questions, we combine data from Global Entrepreneurship Monitor (GEM), World Bank (WB), the World Atlas of Language Structures (WALS), and the Varieties of Democracies (V-DEM) databases. Using a sample of 205,792 individuals from 70 countries and 39 languages, we find that (1) strong FTR speakers are more likely to engage in social entrepreneurship and (2) institutional uncertainty strengthens the strong FTR-social entrepreneurship relationship. Our paper offers three important contributions. First, while many scholars investigate the antecedents of social entrepreneurship, there is little research on how language shapes the temporal orientation of individuals and its consequences for social entrepreneurship. This oversight is surprising because prior research highlights that subjective time perception does influence many human actions (Shipp & Jansen, 2021). By drawing on weak neo-Whorfianism, we provide evidence that FTR shapes an individual's temporal orientation, influencing an individual's strategic decision to become a social entrepreneur. Second, our study provides preliminary evidence that social entrepreneurship appears to be a present-oriented activity connected to a short-term orientation rather than a future-oriented activity connected to a long-term orientation. And that social entrepreneurship differs from commercial entrepreneurship, which appears to be a future-oriented activity connected to a long-term orientation. Third, our work complements previous research on the crucial role of institutional elements in explaining social entrepreneurship. Specifically, in line with the "institutional voids" thesis 1 and the idea that social entrepreneurship responds to unpredictable regulative institutional elements (Mair & Martí, 2009), our results suggest that FTR speakers are more likely to venture socially within institutionally uncertain contexts-that is, environments characterized by a weak rule of law and property protection, suppressed individual and political rights, and high corruption. New institutionalists are increasingly interested in "aspects of cognition that determine how humans interpret reality, which in turn shape the institutional environment they build" (Ménard & Shirley, 2005, p. 11). Our work suggests that language is an overlooked cognitive institution that merits further attention to understand strategic entrepreneurial behaviors. We also shed some light on the potential risk of path dependence that language has on entrepreneurs' social entrepreneurship-related cognitions (e.g., Grégoire, Corbett, & McMullen, 2011;Tang, Kacmar, & Busenitz, 2012) in contexts generating specific and complex institutional configurations (Zhao & Lounsbury, 2016). In sum, we offer a rich empirical exploration showcasing that the language an individual speaks affects their adoption of a particular strategic behavior for value creation (i.e., their engagement in social entrepreneurship activities).

| Social entrepreneurship: Phenomenon and academic discourse
The last six decades witnessed a rapid proliferation of research across disciplines to explain social entrepreneurial activities (Dacin, Dacin, & Tracey, 2011). 2 Despite that burgeoning interest, social entrepreneurship is a contested concept comprising sub-concepts such as the social entrepreneur, social value creation, social innovation, and the social organization (Choi & Majumdar, 2014). The literature includes various definitions ranging from non-profit activities and alternative funding (Dart, 2004) to socially responsible practices of commercial enterprises (Sagawa & Segal, 2000) and change agents providing innovative solutions to address social problems (Bosma, Schøtt, Terjesen, & Kew, 2016;Terjesen et al., 2016). Despite multiple understandings of the phenomena, many scholars agree that "social entrepreneurship balances both social and economic new value to the benefit of collective, rather than individual, interests. The distinctiveness of social entrepreneurship lies in using practices and processes that are unique to entrepreneurship to achieve aims that are distinctly social, regardless of the presence or absence of a profit motive" (Short, Moss, & Lumpkin, 2009, pp. 171-172). Although all entrepreneurs can pursue and generate social value, social value creation is the raison d'être of social entrepreneurs (Morris, Santos, & Kuratko, 2021;Santos, 2012). More specifically, social entrepreneurs use their vision to deploy market-based solutions to address overlooked and long-standing social problems that disrupt or damage society (e.g., poverty, human rights violations, gender inequality, etc.) (Dacin et al., 2010;Santos, 2012). 3 Aiming to understand social entrepreneurship as an essential element of a social entrepreneur's vision, our approach acknowledges that social entrepreneurs create ventures to solve current problems they see as important and not primarily to make profits or gain long-term competitive advantage (Santos, 2012). However, this is not to say that pursuing social value (i.e., the total value created for society, beneficiaries, and customers) is more important than value capture (i.e., the monetary value captured by an individual/organization) for all. 4 Evidence from the Global Entrepreneurship Monitor indicates that a significant share of social entrepreneurs do not agree with the statement that "generating value to society and the environment is more important than generating financial value for the company" (Bosma et al., 2016, p. 14). This suggests that social entrepreneurs might use their judgment to intertemporally trade-off monetization opportunities with opportunities to do good. Indeed, there is considerable evidence that, when compared with commercial entrepreneurs, social entrepreneurs must balance economic demands for starting and managing the social start-up with opportunities to create societal value (Mitzinneck & Besharov, 2019;Short et al., 2009). Thus, a social entrepreneur is more likely to explore intertemporal trade-offs with various levels of magnitude and potential benefits, costs, and risks related to the pursuit of commercial/economic goals and social goals (Townsend & Hart, 2008;Van de Ven, Sapienza, & Villanueva, 2007). The cognitive task of evaluating and judging an opportunity (Tang et al., 2012), which implies deciding which goal (herein, a social or commercial/economic goal) to pursue and prioritize (Townsend & Hart, 2008), is related to temporal consciousness (Shipp & Jansen, 2021) and time perspective (Zimbardo & Boyd, 1999). Given that both temporal consciousness and time perspective are influenced by language (Boroditsky, 2000(Boroditsky, , 2001Dyke, 2021;Montemayor, 2019), we subsequently highlight that the Sapir-Whorf hypothesis is a useful theoretical tool to link language, thought, and behavior in the context of social entrepreneurial action.

| Language, time, and cognition
Anthropologists Edward Sapir and Benjamin Lee Whorf are both credited with theorizing that language has consequences for individual patterns of thought about reality-now commonly referred to as the "Sapir-Whorf hypothesis" (Sapir, 1929;Whorf, 1956). Since its introduction, the Sapir-Whorf hypothesis has faced an unprecedented cycle of acceptance and dismissal. Although linguists, cognitive scientists, and philosophers generally perceive the Sapir-Whorf hypothesis as having limited significance, it has regained attention within economics, management, and psychology domains (Dyke, 2021;Montemayor, 2019;Pinker, 1994).
The German origins of the Sapir-Whorf hypothesis can be traced to the 1760 s and 1770 s in books by Johann Georg Hamann, Johann Gottfried Herder, Wilhelm von Humboldt, and Heymann Steinthal who connected language and thought. The Herder-Humboldt line of logic connecting language and thought is a tradition that becomes most evident in Edward Sapir's and Benjamin Lee Whorf's arguments. 5 In fact, Sapir's formulation is derived from his work in ethnolinguistics, in which he views language as a guide to social reality. According to Sapir (1929, p. 207), the "language habits of our community predispose certain choices of interpretation." In other words, Sapir proposes that the language an individual speaks and has in mind shapes her perception of the world. Sapir's position does not neglect an objective reality but asserts that individuals' linguistic habits play an essential role in how they cognitively perceive reality. Whorf's formulation is notably derived from his research on the Hopi language's apparent lack of tense. 6 Specifically, Whorf (1956, p. 213) states "that the world is presented in a kaleidoscopic flux of impressions which has to be organized by the linguistic system in our minds." In other words, Whorf proposes that reality is relative to the conceptual scheme, and the language individuals speak forcibly structures their perceptions, thereby suggesting that the objective world is totally subjective. Thus, the Sapir-Whorf hypothesis consists of two contrasting dogmas: linguistic relativism and linguistic determinism. Linguistic relativism suggests that "the worldviews imposed by different languages are incommensurable" (Dyke, 2021, p. 2). By contrast, linguistic determinism suggests that "the structure of language determines our worldview" (Dyke, 2021, p. 2). Both relativism and determinism can have weak and strong interpretations. A weak interpretation of linguistic relativism holds that "different languages provide their speakers with different conceptual schemes with which to organize their experiences, but these differences can be grasped with the aid of translation" (Dyke, 2021, p. 3), whereas a strong interpretation of linguistic relativism holds that "different conceptual schemes afforded by different languages are incommensurable with one another; our different languages lock us into a view of reality such that we are unable to see the world as speakers of other languages see it" (Dyke, 2021, p. 3). Relatedly, a weak interpretation of linguistic determinism holds that "the language we speak influences or shapes the way we perceive and experience the world" (Dyke, 2021, p. 3) whereas a strong interpretation of linguistic determinism holds that "language forces us to perceive and categorize the world according to its structure, in way that we cannot escape" (Dyke, 2021, p. 3). The strong interpretations of linguistic determinism and relativism are largely discredited in cognitive science as scholars find that conceptual schemes have existed prior to language and are not determined by language (Pinker, 1994).
However, experimental evidence in cognitive psychology supports a weak interpretation of linguistic determinism and relativism regarding space (Boroditsky, 2001) and time (Montemayor, 2019). Building upon these works, Dyke (2021) advocates for a weaker interpretation of the Sapir-Whorf hypothesis, called weak neo-Whorfianism. 7 Weak neo-Whorfianism broadly claims that language has a limited influence in shaping thought. According to Dyke (2021, p. 5): "all that it requires is that there is some relation of influence between a speaker's language and some aspect of her thought. It does not have to be a specific element of a speaker's mental life, such as perception, cognition, or experience. Any such function that is influenced by language would suffice to satisfy weak neo-Whorfianism." Specifically, weak neo-Whorfianism suggests that individuals who speak different languages will then have distinct sets of experiences, perceptions, and thoughts about the world. Consequently, weak neo-Whorfianism holds that an individual's thoughts are subtly influenced by their native language's grammatical structures.
Montemayor (2019) further suggests a specific application of weak neo-Whorfianism, arguing that there is a narrow scope for language to shape later-stage time perception and thought. By connecting extensive evidence from neuroscience, cognitive psychology, linguistics, and philosophy, Montemayor (2019) makes a compelling case that language differences result in various distinctions in speakers' temporal consciousness. This view is consistent with Casasanto (2008, p. 75): "people who talk differently about time also think about it differently, in ways that correspond to the preferred metaphors in their native languages. Language not only reflects the structure of our temporal representations, but it can also shape those representations." Bylund and Athanasopoulos (2017) provide evidence in their test of the linguistic relativity thesis among Spanish and Swedish speakers' representation of duration periods. Swedish speakers tend to refer to periods of time using spatial metaphors of length (e.g., "short" and "long").
Conversely, Spanish speakers refer to time periods with space quantity metaphors (e.g., "small" or "big"). Bylund and Athanasopoulos' experimental design tests whether linguistic differences among Swedish and Spanish speakers lead to distinct representations of time duration. The authors find that, compared with Swedish speakers, Spanish speakers' preferences for duration as amount hamper their estimates of duration periods. Swedish speakers' preferences for duration as distance hamper their ability to discuss the length of growing lines. Lastly, Swedish-Spanish bilinguals' task performance highlights a different interference effect depending upon the language used to cue the task. Taken together, these works convey that temporal consciousness is influenced and modulated by language, such that language affects an individual's grouping of events and thoughts about time (Montemayor, 2019).
Distinguishing between past, present, and future helps an individual to think in terms of prospects and counterfactuals that appear essential to her perception of subjective mental reality. Thus, subjective time, or "the experience of the past, present, and future, which occurs as individuals (intrasubjectively) and collectives (intersubjectively) mentally travel through, perceive, and interpret time" (Shipp & Jansen, 2021, p. 301) is a core element in developing an individual's subjective mental reality. Building upon a narrow application of weak neo-Whorfianism, we suggest that language's time cognition shapes an individual's temporal orientation. Thus, in this paper, we take a structuredcentered approach to apply a narrow application of weak neo-Whorfianism to language shaping subjective time perception and thought. 8 This structure-centered application of weak neo-Whorfianism focuses on the lexico-grammatical structures (Lucy, 1997) of languages and examines structural differences among languages (e.g., number, gender, and aspect markings) by examining the presence or absence of a particular linguistic marker, FTR. Indeed, one language marker that should have important implications for how individuals subjectively think about time (Shipp, Edwards, & Lambert, 2009;Zimbardo & Boyd, 1999) is FTR. Specifically, FTR encodes how a language requires its speakers to mark time in the future and non-future, thereby capturing a subjective and individual-based understanding of society's language.

| The linguistic structure of future time reference
Although all languages use spatial vocabulary to refer to time, languages vary in the strength of future tense. As Dahl and Velupillai (2013, p. 1) illustrate: "In English, the sentence (1) It is cold tomorrow, with the present tense of the copula is, sounds strange: it is more natural to say (2) It will (it'll) be cold tomorrow or It is going to be cold tomorrow, using a future tense form. In Finnish, on the other hand, one may replace the adverb tänään "today" in (1) with huomenna 'tomorrow', yielding (2) without any further changes in the sentence.

Finnish: Tänään on kylmää
Today is cold. PART "It is cold today." 2. Finnish: Huomenna on kylmää Tomorrow is cold. PART 'It will be cold tomorrow.'" Following this precedent, we categorize languages into two groups: (1) strong FTR or the futured group and (2) weak FTR or the futureless group. Strong FTR languages use inflectional future-tense and require speakers to designate verb form when discussing the future (e.g., English, French). Weak FTR languages neither utilize inflectional future tense nor mark the future (e.g., Finnish), and instead use present-tensed auxiliary verbs (e.g., most other Scandinavian languages and Dutch) (Dahl, 2000;Dahl & Velupillai, 2013). The lack of an inflectional future tense leads a weak FTR speaker to not distinguish future events from present events. Instead, a weak FTR speaker locates future events subjectively closer to the speaker's own temporal position. Accordingly, weak FTR speakers are more likely to speak of the future as similar to the present and will not decouple the present from the future. Contrastingly, strong FTR language speakers habitually decouple tomorrow from today. As a result, strong FTR speakers talks of future events as more remote or distant (Chen, 2013a).
We expect that an individual's temporal orientation is affected by FTR through the narrow application of weak neo-Whorfianism. Specifically, the presence of inflectional future tense may increase a speaker's subjective temporal distance to an event. Conversely, the absence of inflectional future tense will likely decrease a speaker's subjective temporal distance to an event. Therefore, FTR has important implications for temporal discounting. Temporal discounting makes the subjective temporal distance of an outcome to an event a salient aspect that an individual considers when making a decision. Temporal discounting broadly encompasses the reasons to care less about future consequences (Ainslie & Haslam, 1992). Taken together, we contend that strong FTR (i.e., futured) languages generate a higher temporal discount rate compared with weak FTR (i.e., futureless) languages because of the obligatory use of inflectional future tense.
Strong FTR speakers decouple the future from the present, making them more likely to discount the future and less willing to sacrifice the present. Thus, strong FTR speakers favor a short-term orientation because their temporal orientation attaches them to the present . Indeed, short-term orientation describes the preference for focusing on short-range implications of how decisions and actions play out after a brief period of time. Societies characterized by a short-term orientation emphasize efforts that bring quick results (Hofstede & Minkov, 2010). Consequently, a short-term orientation corresponds to larger time discount rates and is associated with proximate payoffs (Wang, Rieger, & Hens, 2016), as well as with spontaneous prosocial behaviors (e.g., helping strangers and donating money) (Luria, Cnaan, & Boehm, 2015).
Research confirms a link between strong FTR and a short-term orientation. For instance, strong FTR speakers spend more money on experiential consumption (i.e., traveling and dining out) , have a fairly modest propensity to save for the future (Chen, 2013a;Roberts et al., 2015), and are more likely to avoid paying taxes (Cheng, Kim, Rhee, & Zhou, 2021). As the future is perceived to be more distant from the present and can mitigate speakers' painful feelings about the future, strong FTR speakers also show less suicidal behaviors (Lien & Zhang, 2020). Moreover, companies headquartered in strong FTR speaking countries engage in more earnings management to meet short-term performance benchmarks (Fasan, Gotti, Kang, & Liu, 2016;Kim, Kim, & Zhou, 2017), are less likely to engage in corporate social responsibility (CSR) initiatives, and less likely to invest in research and development (R&D) (Liang, Marquis, Renneboog, & Sun, 2018). Of particular interest to our study, research finds that strong FTR correlates with innovative early-stage entrepreneurship across countries, such that countries speaking strong FTR languages are about two times more likely to engage in early-stage innovative entrepreneurship (Tang, Yang, Ye, & Khan, 2021). In other words, strong FTR speakers can discount the risks associated with innovative early-stage venturing, enhancing their risk-seeking behavior. Taken together, prior research suggests that languages with inflectional future tense (i.e., strong FTR) create an enduring way of thinking and valuing the present among individuals, which correlates with short-term behaviors.
In contrast, by discussing the future as if it is the present, weak FTR speakers subjectively consider future outcomes as less distant in time and are less likely to discount the future. If speakers perceive the future to be closer to the present, they see present costs and sacrifices for future benefits as less of a problem. For this reason, weak FTR languages are associated with a long-term orientation, reflected in lower time discount rates, as an individual's temporal orientation is directed towards the future (Galor, Özak, & Sarid, 2016). Long-term orientation describes the preference for focusing on long-range implications of how decisions and actions play out after an extended period of time (Lumpkin, Brigham, & Moss, 2010). Societies characterized by individuals with a long-term orientation value what works more than what is right (Hofstede & Minkov, 2010). Due to their propensity to estimate future benefits as more valuable, individuals with a long-term orientation primarily seek situations that offer future rewards and are sensitive to the attributes of such rewards.
Research also confirms a link between weak FTR and a long-term orientation. Accordingly, weak FTR speakers save more often than strong FTR speakers (resulting in more retirement wealth) (Chen, 2013a). Moreover, individuals invest more in education (Galor, Özak, & Sarid, 2020) and they care more for the future of the planet (as reflected by pro-environmental behaviors) and support pro-environmental future-focused policies (Mavisakalyan, Tarverdi, & Weber, 2022;Pérez & Tavits, 2017). Individuals are also more likely to absorb the short-term costs (e.g., time, resources) associated with religiosity in favor of the ultimate long-term payoff, that is "the afterlife" (Mavisakalyan et al., 2022). At the firm level, companies headquartered in weak FTR speaking countries are more likely to emphasize R&D investments and CSR initiatives (Liang et al., 2018), have more patent counts and citations (Kong, Wang, Wang, & Zhang, 2022), have higher precautionary cash holdings (Chen, Cronqvist, Ni, & Zhang, 2017), and are more apt to pursue environmentally responsible behaviors (Mavisakalyan, Tarverdi, & Weber, 2018). Of particular relevance to our study, Campo, Nunziata, and Rocco (2020) find that immigrants in Switzerland who speak weak FTR languages are more likely to be self-employed than strong FTR speakers. Similarly, Hu, Su, and Ye (2022) find that immigrants who speak weak FTR are more likely to be self-employed in the United States. In other words, immigrants who speak weak FTR languages can absorb the short-term costs associated with self-employment, in favor of the potential long-term rewards generated by their self-employment. Collectively, research suggests that languages without inflectional future tense (i.e., weak FTR) create an enduring way of thinking and valuing the future, which in turn is correlated with more future-oriented behaviors.

| Future time reference and social entrepreneurship
Social entrepreneurship differs from commercial entrepreneurship in terms of time orientation. Commercial entrepreneurs, as bearers of risk and uncertainty, seek to generate financial returns from their economic activities (Cantillon, 1755). They often make great sacrifices to develop, improve, and sustain their businesses over time, and are willing to accept long, lean periods and forgo immediate profits. In contrast to commercial entrepreneurs' long-term orientation (Lortie, Barreto, & Cox, 2019), social entrepreneurs are more short-term oriented. Social entrepreneurs strive to solve a social problem as quickly as possible, thereby bringing rapid change by addressing social problems in the present, even if it means losing their raison d'être. Many social entrepreneurs would like to close their businesses once the social problem they address has been solved. Social entrepreneur and Ashoka Fellow Jacek Jakubowski expresses this attitude with the words: "All in all, our role is to build a world where we are not needed anymore" (Ashoka, 2022). Similarly, Dimitris Vassiliadis, founder of Giving Street, a social venture that helps the homeless, formulates his short-term goal as distinct from commercial ventures' long-term goals as follows (Noonan, 2019): "Wanting to eventually close our business might sound strange, but it's our ultimate goal for there not to be any more homeless people on the streets in the future. So that makes us different to purely commercial businesses that are completely profits driven, and therefore want to go on forever." The manifesto of Westway Law (2022), a New York City law firm by social entrepreneurs for social entrepreneurs, further exemplifies this point: "We see the world as it could be. We resolve to make change happen. Not content to wait for better times, we strive to lay foundations now." For social entrepreneurs, profits are only a means to an end, and future rewards and payouts are not the leading priority. In fact, the positive association between social entrepreneurship and short-term orientation is consistent with previous studies demonstrating urgent and impulsive prosocial behavior among individuals with short-termism (Guo, Liu, Li, & Qiao, 2018;Luria et al., 2015). Research also shows that social entrepreneurship is more prevalent in innovation economies that are characterized by a short-term orientation (Kedmenec & Strašek, 2017). Likewise, Hechavarría and Brieger (2022) report that individuals' engagement in social entrepreneurship over commercial entrepreneurship depends on both future-oriented practices and the social category of gender. Moreover, research on social dilemmas finds that individuals sometimes prefer present outcomes over future outcomes when making ethical judgments about immediate behavior (Strathman, Gleicher, Boninger, & Edwards, 1994). In sum, evidence appears to suggest that social entrepreneurs prioritize a pragmatic short-term approach to venturing.
Resultantly, we expect that strong FTR (i.e., futured) speakers are more likely to engage in social entrepreneurship because of their ability to decouple the present and the future. Strong FTR (i.e., futured) speakers perceive urgency when addressing overlooked longstanding social issues due to their short-term orientation. On average, social entrepreneurs prioritize immediate rewards that come with the solutions they develop (e.g., recognition, satisfaction, helping others) rather than a larger reward for addressing the social challenge slowly in the far future (Smith, 1988). Indeed, previous research finds that social enterprises have significantly lower leverage ratios when compared with matched commercial enterprises (Siqueira, Guenster, Vanacker, & Crucke, 2018). This coincides with strong FTR speakers perceiving the future as far removed from the present since leverage is a long-term capital strategy. This also suggests that strong FTR speakers might be better equipped to make intertemporal trade-offs for doing good and monetizing the venture since they are less concerned about future uncertainties and risks associated with debt. Taken together, strong FTR speakers perceive the future as uncertain and may feel more urgency when it comes to engaging in the boundary-spanning activity of social entrepreneurship in the present.
Prior research also highlights a predictive relationship between hyperbolic discounting of rewards and impulsiveness (Ainslie, 1975), as well as positive relationships between individuals with a present time perspective (Zimbardo & Boyd, 1999) and impulsivity (Sircova, Sokolova, & Mitina, 2008), as well as motor impulsiveness/ restlessness (Baumann & Odum, 2012). We therefore assert that strong FTR speakers perceive longstanding social problems as urgent dilemmas to be addressed (Austin, Stevenson, & Wei-Skillern, 2006;Zahra, Rawhouser, Bhawe, Neubaum, & Hayton, 2008). Longstanding social problems may motivate individuals to become social entrepreneurs and create short-term action plans (Laverty, 1996) to address these problems quickly and effectively. Social problems can also cause social entrepreneurs to act more impulsively, spontaneously, and bravely (e.g., if a pivot is required) and render these individuals incapable of resting until these issues are resolved (Austin et al., 2006). Our reasoning is consistent with Austin et al. (2006, p. 7), who explain that: "even before growth has been considered, much less a strategy for growth laid out, social entrepreneurs and their organizations are often pulled into rapid growth by pressure from funders, demand for their products or services, and pushed by their social missions to meet those needs." Thus, we hypothesize: Hypothesis (1). Compared with individuals who speak futureless languages (weak FTR), individuals who speak futured languages (strong FTR) are more likely to engage in social entrepreneurship.
3.2 | Language time reference, institutional uncertainty, and social entrepreneurship New institutional economics examines how institutions affect economic activity. Following Davis and North (1971), we conceptualize the institutional environment as the background constraints, or "rules of the game," that guide an individual's behavior. The institutional environment includes both formal rules (e.g., constitutions, laws, property rights) and/or informal rules (e.g., norms, social conventions). 9 The institutional environment influences individuals' choices by specifying under what conditions certain activities are permitted or prohibited (North, 1990). That is, the institutional environment provides the framework for where human action takes place. North (1990North ( , 2005 further contends that institutions are humanly devised constraints to order the environment and reduce uncertainty. Institutions can reduce transaction and information costs, allowing for more predictable and reliable (but not necessarily efficient) human interactions in a foreseeable environment (North, 1990(North, , 2005Williamson, 2000). Correspondingly, uncertainty is a core element of the institutional environment within new institutional economics (North, 2005;Williamson, 2000). Individuals experience institutional certainty when they can decide with predictable outcomes to pursue their interests (Frølund, 2021). This typically requires that institutions are strongly entrenched, respected, and enforced, as reflected by a strong rule of law, the protection of civil and political freedoms, property rights, and corruption control (Koyama & Rubin, 2022;North, 1990;Rubin, 2008). Prior research elaborates on the favorable influence of institutional clarity on various socio-economic outcomes (Boettke & Fink, 2011;Koyama & Rubin, 2022;North, 1990). In contrast, institutional uncertainty is experienced when laws are not clear in meaning and unequally applied, when individuals' civil and political freedoms are suppressed, when property can be confiscated arbitrarily by the state, and when corruption is immanent to the system (Bylund & McCaffrey, 2017;Milanovic, 2019;Valev, 2006). Prior work sheds light on the adverse effects of institutional uncertainty on various economic and social outcomes (Bylund & McCaffrey, 2017;Koyama & Rubin, 2022). Societies with uncertain institutional environments tend to be poorer and more unequal (Deaton, 2013;Milanovic, 2019;Rubin, 2008), more likely to violate human rights (Kizilkaya, 2021;Welzel, 2013), and more likely to experience profound sustainability issues (Lisciandra & Migliardo, 2017;Wang, Cardon, Liu, & Madni, 2020). At the firm level, an uncertain environment is costly and unpredictable, which makes it more difficult for companies to plan, manage transaction costs, and implement strategies (Luiz, Magada, & Mukumbuzi, 2021).
Entrepreneurs are strongly influenced by the institutional environment, which determines their overall scope of action and thus the extent to which they can take advantage of opportunities. Entrepreneurs may react to the institutional environment in different ways. They "can abide by the institution, completely evade it, attempt to alter it, or if none of those options seem feasible to the entrepreneur, give up entrepreneurial action" (Frølund, 2021, p. 917).
We theorize that institutional uncertainty within the institutional environment is a critical distal mechanism shaping the relationship between an individual's temporal orientation and her willingness to engage in social entrepreneurship. Consequently, institutional uncertainty may moderate the FTR to social entrepreneurship relationship in two distinct and competing ways.
First, one line of theorizing claims that an uncertain institutional environment facilitates social entrepreneurship as it allows social entrepreneurs to respond with market-based solutions to take advantage of institutional uncertainty by addressing it (Mair & Martí, 2009;Montgomery, Dacin, & Dacin, 2012;Zahra, Gedajlovic, Neubaum, & Shulman, 2009). Under these circumstances, social entrepreneurs can alter the "rules of the game" to exploit the opportunity space provided by a particular institutional environment (Bylund & McCaffrey, 2017). When a state arbitrarily applies laws and rules, suppresses individual rights, or threatens certain social groups by confiscating their property, social entrepreneurs can step in to address such institutional uncertainty. Montgomery, Dacin, and Dacin's (2012) finding that social entrepreneurs use collective action to address institutional uncertainty by changing outdated institutional environments supports this perspective. Mair, Battilana, and Cárdenas' (2012) content analysis of 200 social enterprises also finds that social enterprises are correlated with addressing institutional uncertainty in the economic, civic, law and rights spheres. Indeed, many law firms, like the Fernando Pombo Foundation, which promotes legal initiatives geared towards strengthening the rule of law and human rights, operate as social enterprises to provide direct support for those adversely affected by institutional characteristics. Since countries characterized by weak rule of law, suppressed individual and political rights, weak property protection, and public power exercised for private gain tend to have greater social problems (e.g., poverty, discrimination, human rights violations, etc.) (Elliott, 2011;Gründler & Potrafke, 2019;Inglehart & Norris, 2003;Koyama & Rubin, 2022;Milanovic, 2019;Stagnaro, Arechar, & Rand, 2017;Welzel, 2013), there are just more opportunities for individuals to enhance societal well-being. In doing so, social entrepreneurs venture with the goal of altering the institutional environment, ultimately leading to institutional change (e.g., Lawrence 1999;Dorado 2005). Under institutional uncertainty, strong FTR speakers experience a stronger sense of urgency around social problems, and thus are more likely to engage in social entrepreneurship to address these issues. This relationship functions notably via altering entrepreneurial action (Bylund & McCaffrey, 2017), as the uncertain institutional environment influences individuals to develop institutions that can deliver goods and services to those in need (Mair, Martí, & Gangly, 2007). The institutional conditions of uncertainty drive-up transaction costs for social venturing. However, the ability to decouple the present from the future results in social entrepreneurs discounting future costs, preferring the short-term rewards created for beneficiaries. Conceived as such, social entrepreneurs are seen to meliorize in terms of short-term efficiency outcomes in order to transform the institutional environment. Taken together, institutional uncertainty reinforces instances to alter the institutional environment. Therefore, we hypothesize: Hypothesis (2a). A country's institutional environment moderates the relationship between FTR and the likelihood an individual engages in social entrepreneurship, such that the more a country is characterized by high levels of institutional uncertainty, the stronger the positive relationship between futured language (strong FTR) and social entrepreneurship.
A second line of theorizing claims individuals are also more likely to pursue the productive function of social entrepreneurship if an economy is characterized by institutional certainty (Brieger, Terjesen, Hechavarría, & Welzel, 2019;Estrin et al., 2013;Stephan et al., 2015;Tran, 2019). Under these circumstances, entrepreneurs can abide by the "rules of the game" to exploit the opportunity space provided by a particular institutional environment (Bylund & McCaffrey, 2017). Indeed, individuals may choose to become social entrepreneurs if they know that they will not have to pay bribes, that their property or the fruits of their property are protected and will not be confiscated, or that laws will not be interpreted differently in their case. Supporting this perspective, Estrin et al. (2013) find that strong property rights support engagement in social entrepreneurship. Similarly, Brieger et al. (2019) report that entrepreneurs living in countries with greater individual and political rights (e.g., freedom of speech, religion, movement, assembly, and petition) place a higher value on social value creation. Indeed, many companies, like Warby Parker (a for-profit certified B Corp that offers affordable eyeglasses direct to consumers and to those in need by using a buy-one-give-one model) leverage institutional certainty to create and grow their ventures (i.e., a valuation of $1.7 billion in 2020). Such companies tend to launch because the institutional environment provides more clarity in dealmaking, which facilitates lower transaction costs. Hence, a competing line of logic suggests that low institutional uncertainty can strengthen the positive link between strong FTR and social entrepreneurship. This relationship functions notably via abiding entrepreneurial action (Bylund & McCaffrey, 2017), as the certain institutional environment allocates individuals into productive forms of entrepreneurship (Baumol, 1990;Minniti, 2016). Lower transaction and information costs and higher predictability of outcomes can offer lower returns for rent-seeking behavior and higher returns for productive forms of entrepreneurship. Arguably, social entrepreneurship is one of the most productive types of entrepreneurship due to the ever-present concern for societal well-being. Correspondingly, the ability to decouple the present from the future results in social entrepreneurs discounting the future, and thus being better equipped to make short-term intertemporal trade-offs for doing good and monetizing the venture due to institutional clarity (Frølund, 2021). Conceived as such, social entrepreneurs are seen to maximize short-term efficiency outcomes by capitalizing on the institutional environment. Therefore, an environment characterized by low institutional uncertainty can strengthen the positive effect FTR has on social entrepreneurship. Hence, we hypothesize: Hypothesis (2b). A country's institutional environment moderates the relationship between FTR and the likelihood an individual engages in social entrepreneurship, such that the more a country is characterized by low levels of institutional uncertainty, the stronger the positive relationship between a futured language (strong FTR) and social entrepreneurship.

| Data
Our primary data source, Global Entrepreneurship Monitor (GEM)'s Adult Population Survey (APS), is a globally recognized, publicly available, cross-country, and annual individual-level protocol for entrepreneurship (Reynolds et al., 2005). The 2009 and 2015 GEM include a special protocol to measure social entrepreneurship. We match GEM APS data with spoken language, institutional uncertainty, and other variables from a variety of gold-standard sources (see Table A1). Japan and France are excluded from the analysis because those country samples do not include the protocol to screen for social entrepreneurs. We only retain working-age respondents (ages 18-64) in the sample and exclude all respondents lacking either individual or country-specific data. The full sample of 205,792 entrepreneurs from 70 countries who answered all questions of interest is used to test whether individuals are more likely to be social entrepreneurs under certain linguistic and institutional conditions.

| Social entrepreneurship
Our dependent variable captures an individual's involvement in early-stage social entrepreneurial activity (SEA) (yes = 1; no = 0) based on self-reported engagement (or not) in social entrepreneurship activity as nascent entrepreneurs or baby business owners (Lepoutre, Justo, Terjesen, & Bosma, 2013). Nascent entrepreneurs report tangible actions to start a new business during the past 12 months, owning all or part of the new firm, actively participating in the new firm's daily management, and not yet paying salaries for anyone for more than 3 months. Baby business owners currently manage a new firm that is not more than 3.5 years old, and personally own all or part of this new firm (Lepoutre et al., 2013;Reynolds et al., 2005). Consistent with previous social entrepreneurship research (Estrin, Mickiewicz, & Stephan, 2016;Hechavarría & Brieger, 2022;Sahasranamam & Nandakumar, 2020), we operationalize social entrepreneurs based on "yes" answers to GEM's Q6A1: "Are you alone, or with others, currently trying to start any kind of social, voluntary or community service, activity or initiative? This might include providing subsidized or free training, advice, or support to individuals or organizations; profit-making activity, but where profits are used for socially orientated purposes; or self-help groups for community action." For our robustness checks, we create three additional measures to capture the nature of the business activity.
Two measures capture socially oriented business activities across the entrepreneurial ladder (van der Zwan, Verheul, Thurik, & Grilo, 2013): nascent social entrepreneurship (social entrepreneurial activity in the earliest stages of the start-up process with no more than 3 months of profitability), early-state social entrepreneurship (a profitable baby business, i.e., older than 3 months, but younger than 3.5 years old), and established social entrepreneurship (social entrepreneurial activity of older firms, i.e., older than 3.5 years old).

Another measure captures whether the business formation activity is social versus commercial entrepreneurship.
This binary variable captures an early-stage entrepreneur's venture as socially (yes = 1) or commercially (no = 0) oriented. To classify as an early-stage social entrepreneur, a respondent self-identifies as a nascent or baby business owner by answering "yes" to Q6A1. To classify as a commercial entrepreneur, a respondent self-identifies as a nascent ("yes" to Q1A, Q2A, Q2B, Q2D) or a baby business owner ("yes" to Q1C and Q3A, with <3.5 years indicated on Q2E or Q3C) and indicates no involvement in socially driven entrepreneurship ("no" to Q6A1).

| Future time reference (FTR)
FTR is a variable is operationalized using WALS. We follow Chen (2013a) and Roberts et al. (2015) who differentiate weak future tense (futureless = 0) from strong future tense (futured = 1) using the procedure developed in Dahl and Velupillai (2013) and Dahl (2000) within the EUROTYP research program. 10 Futured languages oblige speakers to inflectionally mark future tense; futureless languages do not mark future tense. We use Chen's (2013b) replication data to identify the official language in the respondent's home country. For countries not included in Chen (2013b), we determine the official language of each GEM respondent's country using WALS. For countries with more than one official language, we follow Tang et al. (2021) in identifying the country's dominant language based on the largest part of the population speaking a particular language.
As a robustness check, we use Chen's alternative FTR measure, FTR verb ratio. The variable is based on how frequently weather forecasts in different languages grammatically mark time. The variable counts the total number of verbs that are grammatically in the future tense divided by the total number of verbs that refer to the future tense, yielding a interval variable on a 0-100 scale.

| Institutional uncertainty
Following Williamson's (2000) model, we conceptualize the institutional environment in terms of the following regulatory institutions: property rights, rule of law, and corruption. Formal rules around property rights can decrease institutional uncertainty because it provides enforcement mechanisms around property ownership. Rule of law can also decrease institutional uncertainty because it guarantees the independence of the judiciary, freedom from state despotism, equal treatment of individuals, and provides mechanisms around contract enforcement for organizations.
Corruption can increase institutional uncertainty because transactions are less transparent and susceptible to opportunism. We measure institutional uncertainty with country-level measures from the Varieties of Democracy (V-Dem) database from 2009 and 2015. V-Dem's democracy and institutional measures are considered the "gold standard" due to high quality data (Coppedge, Gerring, et al., 2021). We use three items from V-Dem: rule of law (which concerns whether laws are transparently, independently, predictably, impartially, and equally enforced, and government compliance with the law), property rights (which concerns the right to private property, not actual ownership of property), and corruption (which covers the pervasive is political corruption). Institutional uncertainty is a formative index (α = .88) averaging three key formal institutional dimensions: weak rule of law (v2x_rule), weak property rights (v2xcl_prpty), and corruption (v2x_corr), where higher scores denote stronger institutional uncertainty.
To complement our main analysis, we incorporate 15 additional measures that directly or indirectly capture institutional uncertainty from the Quality of Governance project (QoG) database: the World Bank's (1) quality of governance index (WGI), (2) income inequality index, and (3) poverty gap (at $1.90/day); Heritage Foundation's (4) economic freedom index and (5) property rights; Freedom House's (6) civil liberties index, (7) rule of law index; and (8) level of democracy index; the Fund for Peace's (9) fragile state index; the Quality of Governance's Political Terror Scale project (10) societal violence index; and V-Dem's (11) public sector corruption index, (12) slavery and forced labor index, (13) egalitarian institutions index, (14) educational inequality, and (15) healthy inequality.

| Control variables
We control for 13 individual and country-level characteristics linked to social entrepreneurship. At the individual level, we control for gender (female = 1, male = 0), age (continuous and squared), education (none, some secondary, secondary degree, post-secondary, and graduate), household income (lower, middle, and upper 33%), start-up skills ("Do you have the knowledge, skill and experience required to start a new business?" 1 = yes, 0 = no), fear of failure ("Fear of failure would prevent you from starting a new business" 1 = yes, 0 = no), and know an entrepreneur ("Do you know someone who started a business in the past 2 years?" 1 = yes, 0 = no). These individual-level characteristics were previously identified as relevant to time and social entrepreneurship. For instance, investigating temporal orientation and variables such as age and gender, Park et al. (2017, p. 276) highlight that future (present) temporal orientation increases (decreases) with age and women are more future oriented. Moreover, Hechavarria and colleagues (2017) also show that women entrepreneurs are more likely to prioritize social value creation. There is a U-shaped relationship between entrepreneurs' age and social value creation, indicating entrepreneurs' greater willingness to contribute to society at younger and older ages, and a focus on personal economic benefits in middle age (Brieger, Bäro, Criaco, & Terjesen, 2021). Education, start-up skills, and income are positively, and fear is negatively associated with social entrepreneurship (Estrin et al., 2016;Sahasranamam & Nandakumar, 2020).
At the societal level, we control for language family in eleven categories: Afro-Asiatic, Austroasiatic, Austronesian, Indo-Aryan, Indo-European, Koreanic, Kra-Tai, Niger-Congo, Sino-Tibetan, Turkic, and Uralic. This approach considers historical relations and similarities in language relatedness identified in large-scale cross-linguistic studies (Roberts et al., 2015). Consistent with previous research (Estrin et al., 2013(Estrin et al., , 2016Hechavarría & Brieger, 2022), we control for gross domestic product (GDP) per capita and unemployment (in %) and the population's fluency in the country's dominant national language (in %).

| Analysis
Our sample has a hierarchical structure since individual data (level 1) are nested in countries (level 2). Ignoring the two levels' interdependency can lead to biased standard errors that reveal artificial significant effects. Therefore, we employ multilevel logistic modeling to account for a nested data structure and simultaneously estimate the dependent variable's variability within and between countries (Snijders & Bosker, 2011). We nest our data by countries and allow random intercepts by countries with an unstructured variance component covariance and robust standard errors. We calculate the null model's intraclass coefficient (ICC). For instance, an ICC value of .196 indicates that approximately 20% of the variance in social entrepreneurship occurs between countries. As ICC values above .15 represent middle to large sizes, multilevel modeling is reasonable (Hox, 2010). As our dependent variables are binary, we used the "melogit" command in Stata. and institutional uncertainty (r = .054, p < .01), indicating that individuals are more likely to start-up and run a social venture in countries marked by a strong FTR language and institutional uncertainty.

| Robustness checks
We run several robustness checks, beginning with the FTR alternative: FTR verb ratio. Second, we rerun our focal analysis with alternative institutional measures.
index, the Heritage Foundation's economic freedom index, rule of law index, and property rights index, Freedom House's civil liberties index and level of democracy index, along with V-Dem's egalitarian institutions index decrease the likelihood that individuals in strong FTR societies engage in social entrepreneurship. In contrast, the fragile state index, societal violence index, slavery and forced labor, income inequality, educational inequality, health inequality, and poverty strengthen the positive effect a strong FTR has on social entrepreneurship. In line with our theorizing, the results indicate that individuals in countries with strong FTR languages are more likely to start-up and run a social venture in institutionally uncertain or adverse environments. 12 Third, given our main analysis' focus on early-stage entrepreneurs (i.e., nascent entrepreneurs and baby business owners), we rerun our analysis with restricted samples of nascent and established entrepreneurs. Tables S2 and S3 results support our earlier findings: Strong FTR positively affects nascent social entrepreneurship and established social entrepreneurship and an increase in institutional uncertainty strengthens these relationships.
Fourth, we replicate the empirical approach for early-stage commercial entrepreneurship to examine individuals' likelihood of engaging in total entrepreneurial activity (TEA) (i.e., commercial, nascent, or baby business entrepreneurship) (Reynolds et al., 2005). Table S4 indicates that FTR does not affect early-stage commercial entrepreneurial activity. We also do not find any significant interaction effects between FTR and the institutional uncertainty measure. This finding confirms that a country's language using inflectional future tense impacts the engagement in social entrepreneurship, but not commercial entrepreneurship. Finally, Table 6 and Tables S5 and S6 report results for an only-entrepreneur sample to examine under which linguistic and institutional conditions entrepreneurs are more likely to run a social or commercial venture. Additional support for Hypothesis (1) is provided which shows FTR's positive influence: entrepreneurs are more likely to be social entrepreneurs than commercial entrepreneurs in countries speaking strong FTR languages. Moreover, the interaction of FTR and institutional uncertainty strengthens social entrepreneurship when compared with commercial entrepreneurship (see Figure 2). Thus, entrepreneurs who live in societies that speak strong FTR languages and face institutional uncertainty are even more willing to start and run a social than a commercial venture. Therefore, this evidence further supports Hypothesis (2a) and rejects Hypothesis (2b). Table S5 also uses the alternative measure to capture FTR. Again, we find a positive relationship between FTR verb ratio and social entrepreneurship, as entrepreneurs are more likely to set up and run a social venture than a commercial venture in countries with a higher FTR verb ratio. Table S6 shows the robustness of the finding that institutional uncertainty strengthens the positive effect of FTR on social entrepreneurship. The results are fully in line with the results for the main sample, with the main difference that now also Heritage's Economic Freedom index shows a significant moderating effect.

| DISCUSSION
What does our study tell us about individuals engaging in social entrepreneurship cross-culturally? First, building upon a narrow application of weak neo-Whorfianism, this study provides evidence that a language's grammatical marking of time appears to shape an individual's temporal orientation and subsequently their social entrepreneurial action. We find that strong FTR positively affects an individual's engagement in social entrepreneurship and entrepreneurs' engagement in socially, but not commercially, oriented ventures. Our evidence suggests that the ability to decouple the present from the future among strong FTR speakers facilitates a short-term temporal orientation, Note: Dependent variable is social entrepreneurship. Coefficients are reported. Robust standard errors in parentheses. All analyses include the initial set of covariates. We report only the estimates for the interaction variables. † p < .1; *p < .05; **p < .01; ***p < .005.
which increases the likelihood of engaging in social entrepreneurial action to create short-term impact while discounting the future (Bearden, Money, & Nevins, 2006). Furthermore, we integrate new institutional economics to better understand how institutions shape the relationship between an individual's understanding of subjective time and engagement in social entrepreneurship. We highlight that institutional uncertainty moderates this relationship.
Specifically, an uncertain institutional environment (i.e., undefined and arbitrarily applied laws, suppressed individual and political rights, high levels of corruption, and the risk of losing property) appears to increase the demand and opportunity space for social entrepreneurship, thereby reinforcing the positive relationship between strong FTR and the likelihood of being a social entrepreneur. In sum, our findings reveal that institutional uncertainty plays an important moderating role in contextualizing the relationship between strong FTR and social entrepreneurship.

| Contributions
We contribute to the extant literature by providing a better understanding of the social entrepreneurship phenomenon in three important ways. First, there is a paucity of research examining subjective view of time in social entrepreneurship. This is surprising given that subjective time influences many human actions (Shipp & Jansen, 2021). By drawing on weak neo-Whorfianism, we provide a compelling explanation for how FTR acts as a mechanism that drives an individual's understanding of subjective time, ultimately affecting their propensity to engage in social entrepreneurship. We highlight that language is a powerful and culturally embedded force that shapes socially oriented entrepreneurial actions. Resultantly, we respond to calls for better incorporating language (using a weak neo-Whorfian lens) in entrepreneurship research (Hechavarría, Terjesen, Stenholm, Brännback, & Lång, 2018;Song, 2021;Tang et al., 2021). We also build upon prior research that examines how an individual's conceptualization of time (Bluedorn & Denhardt, 1988;Reilly, Souder, & Ranucci, 2016;Shipp et al., 2009;Zimbardo & Boyd, 1999) shapes her behaviors, such as allocating resources to opportunities that have different (i.e., immediate versus more delayed) payoffs . We also extend recent entrepreneurship research on self-efficacy, personal agency, and alertness (Boudreaux, Nikolaev, & Klein, 2019) because our results signal that strong FTR speakers might be more temporally conscious of the present.
Second, our work illustrates the importance of the institutional environment on social entrepreneurial activity.
Notably, our results convey that an uncertain institutional environment is a distal mechanism that amplifies the positive relationship between strong FTR and social entrepreneurship. This relationship is partly explained by more opportunities to alter the unpredictable institutions, as well as an individual's ability to discount costs associated with institutional uncertainty in favor of immediate outcomes for beneficiaries. That is, increased levels of institutional uncertainty influence social entrepreneurs who speak strong FTR to focus on immediate rather than future outcomes. This indicates that in more uncertain environments, strong FTR speakers are more likely to temporally discount future costs to ensure immediate benefits for beneficiaries. In part, social entrepreneurship is an idiosyncratic activity in which time horizon-resource allocation arguments that claim "uncertainty increases with horizon, managers are […] reluctant about the greater variability of expected returns on longer term projects" (Reilly et al., , p. 1175, and language-strategy arguments such as "language highlighting environmental uncertainty lowers the likelihood of market entry […]" (Nadkarni, Pan, & Chen, 2019, p. 134) are not well received. Instead, our findings support recent research that finds "consumers are more oriented to the 'here and now' when facing uncertainty, a message frame highlighting immediate benefits of […] products fits such a present orientation" (van der Wal, van Horen, & Grinstein, 2018, p. 384). Furthermore, our results imply that social entrepreneurs who speak strong FTR might be more concerned with the immediacy of entering markets characterized by institutional uncertainty and less concerned with balancing long-term economic and social value goals (Morris et al., 2021;Short et al., 2009;Slawinski & Bansal, 2015;Townsend & Hart, 2008). We respond to recent calls urging researchers to examine how institutional factors influence cross-cultural differences in social entrepreneurship (Sahasranamam & Nandakumar, 2020). Moreover, our research has important implications for new institutionalists who have become increasingly interested in "aspects of cognition that determine how humans interpret reality, which in turn shape the institutional environment they build" (Ménard & Shirley, 2005, p. 11). Our findings suggest that language is an overlooked cognitive institution that merits further attention.
Third, our work conveys that social entrepreneurship may differ from commercial entrepreneurship in that it relates not to long-term oriented behaviors, but instead to short-term oriented behaviors. Our study supports Guo, Liu, Li, & Qiao, (Guo et al., 2018) research showing that societies characterized by individuals with a short-term orientation are more likely to engage in prosocial behaviors. Thus, we provide additional evidence that positions social entrepreneurship as a short-term oriented action that helps solve current problems by creating proximate benefits.
However, our results also contrast earlier work suggesting that speakers of weak FTR languages appear to be more prosocial since they tend to behave more environmentally responsible  and that corporations that predominantly use weak FTR languages are more engaged in CSR activities (Liang et al., 2018). We believe that, in contrast to the short-term perspective of social entrepreneurs who want to solve a social problem as quickly as possible, both CSR activities and environmental strategies of larger multi-national organizations tend to have a long-term perspective. Large multi-national companies' CSR activities tend to be much more instrumental, aiming at securing the company's reputation, profitability and survival in the long term. In this sense, Graafland andNoorderhaven (2020, p. 1036) also argue that the longer a society's or firm's time horizon, the more likely the "long-term benefits of voluntary CSR initiatives, e.g., in terms of quality of stakeholder relations," will be taken into account.

| Implications for theory
Seminal (Das, 1987) and contemporary (Nadkarni, Chen, & Chen, 2016;Schultz & Hernes, 2020)  Although our work suggests that social entrepreneurship is a short-term oriented activity, in line with recent work highlighting its prioritization of the present over the future (Guo et al., 2018;Kedmenec & Strašek, 2017), we need to know more about how social entrepreneurs' characteristics such as how positive and negative past and present experiences (and perceptions) impact goals and activities oriented towards social value creation in the present and future (Short et al., 2009). Given that recent empirical research highlights that "the benefits of horizons do not continue without limit […]" (Souder, Reilly, Bromiley, & Mitchell, 2016, p. 1213) and that "time horizons are positively associated with asset durability […]" (DesJardine & Bansal, 2019, p. 774), we encourage entrepreneurship scholars to investigate these questions more carefully. Specifically, we suggest that scholars collect quantitative and qualitative data from social entrepreneurs across countries that score differently on the cultural dimension of subjective time, like long-term orientation. In particular, as "[…] sustainability requires meeting short-term business pressures while also serving long-term business and societal needs" (Slawinski & Bansal, 2015, p. 545), future research should investigate the relationship between social entrepreneurship and sustainable entrepreneurship, since there may be a temporal tension in terms of entrepreneurs' perceptions and actions to enhance social issues versus sustainability issues (e.g., Bansal & DesJardine, 2014). This may be done with either longitudinal data or experimental data.
Echoing Kaplan and Orlikowski (2013, p. 966), who explain that "[…] strategy cannot be understood as the product of more or less accurate forecasting without considering the multiple interpretations of present concerns and historical trajectories that help to constitute those forecasts. Individuals' projections of the future are always entangled with their views of the past and the present, and temporal work is the means by which they construct and reconstruct the strategic accounts that link them together", we invite entrepreneurship scholars to account for a comprehensive conceptualization that encompasses positive and negative aspects of both objective and subjective temporal dimensions.

| Limitations and future directions
Although we provide a novel approach to better understand some less explored macro-level antecedents of social entrepreneurship, our work is not without limitations which can also be seen as a springboard for future research.
First, we built upon Chen's works to explore the relevance of FTR in social entrepreneurship. Chen's hypothesis and the behavior-language relationship are validated in many subsequent studies (e.g., Falk et al., 2018;Herz, Huber, Maillard-Bjedov, & Tyahlo, 2021;Sutter, Angerer, Glätzle-Rützler, & Lergetporer, 2018). For instance, Sutter and colleagues (2018, p. 29) find that "German-speaking children are significantly more likely to save than Italian-speaking children." Yet, others reveal inconclusive results. For instance, Chen, He, and Riyanto (2019, p. 7) report an "absence of evidence that the repetitive use of 'will' leads subjects to behave less patiently […]." Therefore, we invite scholars to better identify all cross-cultural semantic and syntactic distinctions that play an important role in shaping entrepreneurs' behaviors, including the propensity to engage in social and commercial activities.
Second, although we detect differences between weak and strong FTR languages in terms of social entrepreneurship, we do not investigate how these distinctions manifest in terms of normative ethics (Jones & Wicks, 1999).
Thus, future works could focus on how language structures may shape virtue and consequentialist and deontological approaches (and vice versa). Specifically, scholars could investigate the language structures of social entrepreneurs who alter their environments by focusing on founders doing what is morally right (e.g., engaging in fair trade agreements with farmers and other stakeholders) in their achievements of a higher purpose (e.g., providing food for all human beings).
Third, although we capture the percentage of each country's population speaking the dominant language, we do not account for the fact that entrepreneurs may speak multiple languages. Prior research highlights that transnational entrepreneurs leverage multiple repertoires including language (Drori, Honig, & Wright, 2009). Therefore, future research could explore how social entrepreneurs who speak multiple languages (e.g., individuals who simultaneously use strong and weak FTR) code-switch between FTR repertoires and its implication for inter-temporal venturing decisions. We contend that social entrepreneurship scholars have a fantastic opportunity to leverage experimental designs and probe intertemporal trade-off decisions such as those related to monetizing the start-up versus producing social value to beneficiaries, customers, and society.
Fourth, although our study categorizes social entrepreneurship as binary, prior works argue for a "blended values" perspective of entrepreneurship (Hechavarría et al., 2017). Therefore, social entrepreneurship scholars could include a "blended values" continuum (encompassing purely social and purely economic goals) to better understand how futured languages shape value creation and capture. Likewise, there is considerable variance in terms of social entrepreneurial activity. Some social entrepreneurs might want to grow, while others want to stay small and manageable. Some may be highly innovative, and others replicative. As a result, researchers should further probe FTR effects among different kinds of social ventures to better understand if language has differing effects on certain kinds of social venturing.
Future works could notably examine how an individual's propensity to engage in social entrepreneurship evolves over time, distinguishing between futured and futureless speakers of various age groups (Brieger et al., 2021).
Fifth, our work only focuses on regulatory institutional uncertainty and excludes normative and cognitive institutions. As FTR can shape the aggregate normative and cognitive institutions that lead to individual action, social entrepreneurship scholars could also consider how FTR links to cognitive and normative institutions impacting social entrepreneurial opportunities. Furthermore, research in (social) entrepreneurship increasingly advocates for a more nuanced understanding of institutional elements and their interrelationship. A useful lens to expand on the concept of institutional uncertainty is institutional interfaces . Institutional interfaces are built of "bits and pieces" from market and non-market institutions. Understanding how these "bits and pieces" create fragmentation will undoubtedly extend our understating of institutional uncertainty and social entrepreneurial action. Especially since our research provides evidence that language structure appears to be an overlooked cognitive institutions among institutionalists.
In all, potentially interesting future directions derived from our limitations include: What are entrepreneurs' dominant language structures? Do social entrepreneurs emphasize any particular temporal dimensions? How does a social entrepreneur's language reflect and is reflected in her (ought-to-be) actions? Are there any "lexico-grammatical structures" correlated with social entrepreneurship across culture and social categories (e.g., gender identity, race, ethnicity, sexual orientation, age, etc).? How and why (not)?

| CONCLUSION
This study draws on a narrow application of weak neo-Whorfianism to explore how individuals who rely on shared symbols through language may construct the realities of their organizations. We suggest that language is a universal, taken-for-granted attribute that affects how individuals subjectively perceive time in the mind, which ultimately leads them to be engaged in social entrepreneurship. Our research connects new institutional economics with a narrow application of weak neo-Whorfianism and presents a contextualized view of the social entrepreneurial phenomenon. We find that individuals living in societies where the dominant language is futured are significantly more likely to engage in social entrepreneurial activities, and this relationship is moderated by institutional uncertainty. In sum, we argue that we cannot fully understand economic outcomes such as social entrepreneurship without considering the institutional context within which that outcome emerged.

ACKNOWLEDGEMENTS
We would like to thank Robert Nason, Jeremy Short, Trenton Williams, and Marcus Wolfe for coordinating this special issue opportunity in SEJ. We also want to thank Katie Brownell for her comments on prior versions of this paper.
Finally, we would like to sincerely thank Robert Nason and the two anonymous reviewers for their excellent feedback and developmental comments on prior versions of this paper.

ETHICS STATEMENT
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent was obtained from all individual participants included in the study, from the respective datasets. We refer to institutional uncertainty rather than institutional voids because scholars assess (see Bothello, Nason, & Schnyder, 2019) that the institutional void concept is narrow and has its flaws as it adopts an ethno-centric Western perspective. The term "institutional uncertainty" captures how individuals feel more uncertainty in contexts where laws are applied arbitrarily, their civil and political rights are suppressed, their property or the fruits of their property are not protected, and the executive and public administration can abuse their power for personal gain. 2 Banks' (1972) first coined the term social entrepreneur when describing social movements in sociology. Drucker (2012) introduced the term social enterprise in management while discussing firms' ethical responsibilities.

ORCID
3 Despite often having an objective basis, social problems tend to be subjective, socially constructed, and based on conditions claimed as "troublesome" by groups of people (Schneider, 1985). In other words, social problems are collectively defined and can be understood as an entrepreneur's interpretation of the need to ameliorate a situation regardless of whether the affected group agrees with the entrepreneur's perception (Hervieux & Voltan, 2018). 4 Our study views entrepreneurship as a phenomenon at the nexus of the individual and the initialization of agency towards entrepreneurial action to pursue both social and economic value creation and capture.
5 For a complete historical review of the origins of the Sapir-Whorf hypothesis, see Koerner (1992). 6 Pinker (1994) challenges Whorf's assessment of the Hopi language and the way in which the Sapir-Whorf hypothesis was initially popularized. Despite such challenges, there is empirical evidence that the Sapir-Whorf hypothesis holds regarding how people understand time (Bylund & Athanasopoulos, 2017;Casasanto, 2008). 7 To distinguish her position from Whorf's original strong deterministic position, Dyke (2021) uses the term weak neo-Whorfianism and uses arguments grounded in A-theory and B-theory of time in philosophy. 8 Lucy (1997) summarizes empirical research on the Sapir-Whorf hypothesis in three main approaches: language (structure-centered), thought (domain-centered), and reality (behavior-centered). 9 According to Williamson (2000), institutions operate hierarchically at different levels, including higher-level constitutional features and lower-level regulatory and policy features. Higher-level constitutional features comprise the rule of law, civil and political rights, protection of property, and freedom from corruption. Lower-level regulations and policies include "the scale and scope of the government covering an interrelated cluster of collective choice features that include the size of government spending, the scope of the welfare system, the level of taxation" (Estrin et al., 2013, p. 483

SUPPORTING INFORMATION
Additional supporting information can be found online in the Supporting Information section at the end of this article.