Introduction

The theme of social entrepreneurship (SE) has drawn increasing attention from researchers in recent decades, and its definition has been widely debated among researchers and practitioners alike (Granados et al., 2011). Conceptual debate occupies considerable space in the literature (Dacin et al., 2011), and different schools of thought on SE have been identified (Bacq & Janssen, 2011; Dees & Anderson, 2006; Defourny & Nyssens, 2010; Galera & Borzaga, 2009; Young & Lecy, 2014). The field has now moved beyond narrow definitions (Doherty et al., 2014), and made considerable efforts to describe the diversity of SE and to build typologies for the range of organisations within this label (Alter, 2007; Kerlin, 2013; Defourny & Nyssens, 2017; Mike Bull and Rory Ridley-Duff, 2018). The field of social entrepreneurship (henceforth SE) has been steadily growing in the last few decades (Dacin et al., 2010). However, according to some scholars, the boundaries of the field remain blurred (Santos, 2012), without reaching a consensus on what makes SE different from regular businesses. We argue that defining SE (as opposed to classical business) is difficult because “social” is a value-loaded concept (Choi & Majumdar, 2014) and its normative basis is unclear (Bruder, 2020). This is problematic because if SE scholars ambition to provide practitioners with useful knowledge to help them progress, they will need theories that are explicit regarding the type of goals SEs are supposed to achieve. Following this idea, our research explores the normative assumptions behind SE scholarship through political philosophy theories, which provide coherent logical frameworks to classify normative ideas.

This paper explores the normative basis of SE research through a deductive content analysis of the top 100 publications in the domain. Using eight contemporary schools of thought in political philosophy as a template for analysis, our goal was to identify the main normative assumptions and principles sustaining SE investigation. In our analysis, we have identified the philosophies underlying SE literature, and the important consequences of their (often ignored) underlying normative assumptions. More specifically, we identified the presence of normatively ambiguous concepts, fundamental justifications, and critiques around some central philosophies’ normative assumptions, and important central normative contradictions.

Our paper contributes to the literature on SE in three distinct ways. First, by we propose that political philosophy can help to identify what counts as ‘social’ in SE. Next, we show some of the inherent normative tensions that could dampen its impact. Finally, we propose ways in which a normative awareness would help to establish the basis upon which to evaluate and demonstrate the social, economic, and cultural impact of SE.

The Need for Normative Theories of Social Enterprise

We have found diverse answers in the literature on “which objectives can legitimately be considered social” (Nicholls & Cho, 2008, p. 101), as well as which means (Miller et al., 2012). SE scholars sometimes define the ‘social’ orientation in opposition to the profit motive. Peredo and McLean (2006) define SE by stating that these enterprises aim at increasing ‘social value’; that is, their intention is “to contribute to the welfare or wellbeing in a given human community” (Peredo & McLean, 2006, p. 59). They situate different types of SE on a continuum according to the prominence of the social objective or profit motives. The literature on hybridity similarly suggests that business or commercial objectives are in tension with social ends (Doherty et al., 2014). On the other hand, the social mission can also be defined as providing goods and services that address unmet needs (Defourny & Nyssens, 2017, p. 2487), foster economic democracy, or address particular social problems like “unemployment, poverty, underdevelopment or handicaps of all kind, among other factors, which may cause marginalisation or exclusion” (Defourny, 2009, p. xiii). For other authors, SE intervenes to create positive externalities (Santos, 2012) or responds to market failures (Austin et al., 2006).

Regarding means, some SE authors defend the idea of internal democratic participation (Domenico et al., 2010) as essential to the work of organisations, “which involves various parties affected by the activity” (Defourny & Nyssens, 2010, p. 43). Others suggest that ethics of care (Mort et al., 2003; Smith et al., 2012) are needed to sustain the empathy needed for social initiatives. Finally, studies have also explored existing shortcomings in working conditions inside SEs (Jones et al., 2008), revealing “problematic account[s] of work/life balance centred on extreme self-sacrifice” (Dempsey & Sanders, 2010, p. 439).

How might this diversity of social goal and means definitions be explained? We argue in this paper that behind any definition of ‘social’ there is an implicit normative criterion. The word ‘social’ has a positive connotation, and several authors have noted the normative implications behind the definition of SE. For Cho, defining ‘social’ is equivalent to defining what is in the interests of society, which is a political task (Cho, 2006, p. 36). Choi and Majumdar similarly highlight that the concept of SE generates value-laden debates (Choi & Majumdar, 2014, p. 365). Bull and Ridley-Duff (2018) note that diverse conceptualisations of SE imply various, and under-theorised, political foundations, while Boddice (2011) points to the ideological foundations of the SE movement. Lyon and Sepulveda (2009) also propose that mapping SE empirically requires political choices in interpreting SE. Therefore, to define and discuss SE, we must first build understanding of what these political stances are.

The Relevance of Political Philosophy to the Understanding of SE

Moral philosophy focuses on what people may or may not do, while political philosophy debates which of the obligations we have to each other should be enforced through the state (Kymlicka, 2002). Consideration of the normative dimension of SE has recently been encouraged in the field of business ethics (Chell et al., 2016), and some attempts made to apply moral theories to SE. For example, André and Pache (2016) make prescriptions about how social entrepreneurs can apply ethics of care. Using moral philosophy concepts to define SE, Tan, Williams and Tan (2005) define the ‘social’ dimensions of SE through the altruistic orientation of the social entrepreneur. Their definition of a social entrepreneur, thus, is a “legal person engaged in the process of entrepreneurship that involve a segment of society with the altruist objective that benefits accrue to that segment of society” (Tan et al., 2005, p. 360).

This kind of definition, based on the intentions of actors and moral principles (altruism, egoism, sense of justice, etc.), is interesting in its identification of different behaviours within SE. However, it is also problematic. First, it is difficult to transpose moral concepts to the organisational level. We can define what counts as an altruistic individual, but identifying an altruistic organisation is trickier due to difficulty in specifying its ‘personal interest’. Second, individual intention is insufficient to judge the rightness of an action or its consequences.

Imagine two fictional cases: in the first, a group of poor cocoa producers start a collective SE to increase their income; and in the second, a student starts an SE and works voluntarily to allow business school students to enjoy car racing. We may describe the intention as collective self-interest in the first case and altruism in the second. However, to judge which organisation has a superior normative value and thus a greater claim to the adjective ‘social’, a theory of justice is needed. For example, from an egalitarian perspective, the first organisation may be judged as more “social” because its goal is to satisfy the basic needs of poor people, while the second case seeks to satisfy a group’s leisure preferences. Furthermore, we have to determine what the person/group is entitled to, in order to distinguish whether giving it away is altruist.

Focussing solely on individual morality leaves key normative questions unanswered: Which segment of society deserves help? How do we know that the actions directed at others are ‘good’ for them? When is self-interested behaviour legitimate? Political philosophy can provide conceptual tools to answer these kinds of questions. Additionally, political philosophy is needed because SE is supposed to transform the world (Sen, 2007) or change social structures (Mair, & Marti., 2006). If we state that SE aims to improve society, we need to know what standard is used. That is why we choose political philosophy here, which focuses on the societal level, rather than moral philosophy.

Political philosophy may also be more able than other approaches like discourse analysis to provide solid theoretical grounds and articulate normative concerns and descriptive theories. Indeed, discourse analysis usually focus on “naturally occurring” normative discourses. These kind of discourses are not always internally coherent because of the complexity of the social life, and do not necessarily include clear justifications for the prescribed behaviours. On the other hand, political philosophy allows discussing deeply the logical coherence of different normative justifications of human institutions. As such, it could provide more solid ground to build rigorous theories of SE.

In our analysis, we use eight schools of thought of western contemporary political philosophy: utilitarianism, liberal egalitarianism, libertarianism, Marxism, communitarianism, citizenship theory, multiculturalism, and feminism, briefly described in the Online Appendix A. This typology is based on Will Kymlicka’s (2002) book Contemporary Political Philosophy: An Introduction, a widely-cited reference in the field. Kymlicka (2002) presents the philosophies underpinning established schools of thought, rather than key debates (Christiano & Christman, 2009; Matravers & Pike, 2005), subjects (Graham, 1982; Solomon & Murphy, 1999), or canonical texts or authors (Carter et al., 2007; Goodin & Pettit, 2006). Kymlicka focuses on Anglo-Saxon streams of thought which, we suggest, fit with the similar tradition that fuels SE literature (Granados et al., 2011). That said, his work is mostly silent on recent moral debates around the environment, the protection of which has increasingly been studied as an important ‘social goal’.

Methodology

In this section, we outline how we used the above philosophies as templates for a content analysis (Seuring & Gold, 2012) of the top 100 influential articles on SE. The goal of this research was not to conduct a systematic literature review and compare empirical or theoretical results, as has been done elsewhere (Phillips et al., 2015). Our aim was rather to identify the implicit normative assumptions present in the most influential SE research, i.e. our goal is to analyse the SE literature, and not its practice. This approach is in line with previous comparative methodologies such as Boltanski and Thevenot’s in De la justification (1991), in which they used political philosophies to build their model of different ‘worlds’ (inspired, domestic, fame, civic, market, and industrial). We first present the data collection process with the database used and the selection criteria, before turning to the coding process and the analysis of our sample of documents.

Database choice

Various database sources that can be used to perform bibliographic analyses: Web of Science, Scopus, Google Scholar, Microsoft Academics, Dimensions, Crossref, COCI, etc. (Harzing, 2019; Martín-Martín et al., 2020). Since our study focuses on influential articles, the most important identification criteria were the database’s scope and ability to count citations. The databases with broadest coverage are Google Scholar and Microsoft Academics; Web of Science, Scopus, and Dimension have lower coverage of citations, and are quite similar (Martín-Martín et al., 2020).

The second criteria was the accessibility of the database. Extracting data from Google Scholar is difficult (Martín-Martín et al., 2020), Microsoft Academics requires a programming interface, and Web of Science and Scopus require subscriptions, while the basic functions of Dimensions are free. We initially used the WOS database in our first analysis. However, WOS aims specifically to index “high quality” journals, and in the SE field, specialised journals may be excluded from it because of lower impact factors, while they are influential in this emergent field. We thus performed an additional search in the Dimension database.

Sample Constitution

The main keywords used in existing literature reviews are “social entrepreneurship”, “social entrepreneur”, “social enterprise”, and “social venture” (Dionisio, 2019; Gonçalves et al., 2016; Granados et al., 2011; Hill et al., 2010). We searched the databases using only the keywords “social entrepreneurship” or “social entrepreneur” to focus on the core literature. These keywords were entered in the Institute for Scientific Information Web of Science (ISI WOS) and Dimension databases, and publications were searched for all available years: 1934–2020 for Dimension, and 1975–2020 for WOS. Search characteristics are summarised in 1—Sample selection criteria (Table 1).

Table 1 Sample selection criteria

We exported the top 500 most-cited documents for each database and merged the items present in Dimension, WOS, or both databases into a single excel file. We considered citations as an indicator of a paper’s influence and likelihood of representing some foundations and inspiration for the field. To reliably identify which papers were the most cited, for each paper we added the number of citations calculated by Google Scholar, the database with highest citation coverage (Martín-Martín et al., 2020). We recorded the highest number of citations identified by Dimension, WOS or Google Scholar, and ranked the articles accordingly.

Then, we read the abstracts for the 130 most-cited documents to check their relevance to the SE literature. Both authors discussed doubtful cases, and we made sampling decisions together. We included book chapters, editorial pieces, introductions to special issues, and literature reviews, as these are likely to frame the field and its concepts. Conversely, we excluded articles whose focus was not on SE. For example, the topic was broader (CSR) or narrower (cooperatives). Although papers on teaching SE can be influential, we also considered that their central contribution was on education rather than SE and did not include them. We excluded practitioner-oriented articles because they are more likely to summarise existing ideas than develop new ones.

After excluding irrelevant items, our sample was established with the Top 100 most-cited papers. This final sample contains 93 articles and 7 book chapters (references are available in Online Appendix B: References of the sample). The most represented outlets are Entrepreneurship Theory and Practices, Journal of Business Ethics, Journal of Business Venturing, and Journal of World Business. The publication dates range from 1991 to 2019 (Figs. 1, 2).

Fig. 1
figure 1

Journals represented in reviewed publications

Fig. 2
figure 2

Number of articles per year

Data Analysis

Our data analysis had three phases. First, we performed a deductive content analysis based on keywords related to the eight philosophies informed by the literature, followed by a qualitative reading of the coded excerpts to confirm the latent coded content. Finally, we conducted an individual document analysis to understand each author’s use of the identified philosophies in their research.

Deductive Data Coding

The sample documents were listed in the reference management software Zotero and exported with their metadata into the qualitative data analysis software Atlas.Ti 8. Their software readability was verified. We then used a two-step approach to conduct the content analysis, following Seuring and Gold (2012): a deductive coding anchored in theory (here, philosophies) followed by iterative cycles of inductive refinement of the resulting categories. The first step thus aimed to build a coding template based on Kymlicka (2002). For each of the political philosophies detailed in Online Appendix A, we listed all the key ideas and concepts with which they were associated. Then, we selected the more specific ones that were less likely to overlap across different streams. This step was mainly based on our qualitative understanding of the different philosophies. Online Appendix C: Initial political philosophy keywords presents the initial keywords list.

Next, we used semi-automatic coding based on these keywords (Atlas.Ti 8 auto-coding function). Each time the software identified a keyword in an article, it linked the sentence in which the word was present to the corresponding philosophy. A preliminary reading of the results of the automatic coding led us to refine the keywords by suppressing terms or adding synonyms. For example, the term ‘contract’ was too general, as it could refer to legal contracts as well as to the philosophical idea of a social contract, thus we kept only the term ‘social contract’.

We also split each philosophy into categories of keywords. For example, we iteratively refined the utilitarian category because the keywords ‘maximisation’ and ‘efficiency’ present in the utilitarian codes covered multiple ideas. To deepen the validity of our philosophical keywords, we engaged two experts in political philosophy to confirm the soundness of our choice of philosophical schools. Online Appendix D: Final political philosophy keywords lists the final automatic coding request entered into the software for each philosophy.

Qualitative Reading

For each automatic coding, we then conducted a qualitative analysis of the content of the coded sentences, because the philosophical ideas in the papers constitute latent content (Potter and Levine‐Donnerstein, 1999) requiring interpretation. The keyword search was useful for identifying relevant fragments of text, but a qualitative reading was necessary to confirm correspondence to a particular philosophy.

To increase the consistency of the results, both authors undertook independent coding on 10% of the coded sentences, compared findings, and discussed divergences. We read every code, referring to the full text when context was necessary, and split each category into themes using Atlas.TI. We excluded quotations unrelated to philosophical concepts (e.g. energy efficiency in the theme of efficiency), and sentences not directly concerning SE (e.g. participatory municipal budgets in the citizenship category, gender control variables for the feminist category). We also recoded some sentences into more appropriate categories; for example, the idea of social ownership initially appeared in the libertarian category ‘property rights’, but was a closer match for the democratic philosophy category, as collective ownership is a tool for democratic governance.

Document Analysis

The coding process took ideas as units of analysis and linked them to different philosophies. However, the coded fragment of text may not represent the author’s own views. For instance, fragments may be sentences presenting another scholar’s ideas, empirical results, or critiques. We thus conducted an additional step of analysis to study how different philosophies coexist within the same piece of research, using Atlas.Ti 9’s semantic network function. A network was built for each document to map the codes present and their relationship (see, for example, Fig. 3).

Fig. 3
figure 3

Excerpt of semantic network based on Santos, (2012)

Based on the semantic networks, we coded each paper depending on the presence/absence of each idea as “pro”, “cons”, “neutral”, “instrumental” or “empirical”. When the paper argument was critical, for example defending a certain vision of SE against another, we coded the opposed ideas as “pro” and “cons”. For example, the sentence “[SE] are more than just tools for achieving the most efficient and effective mode of service delivery; they are also important vehicles for creating and maintaining a strong civil society” (Eikenberry & Kluver, 2004, p. 136) led to code “cons/efficiency” and “pro/citizenship” because it criticises what is judged as and excessive focus on efficiency. Similarly, when something was presented as good, useful, or efficient it was coded as “pro”. For example, the sentence “market orientation can imply the employment of commercial activities directly linked to the social mission to ensure the most effective and efficient distribution of social services and products” (Choi & Majumdar, 2014, p. 368) led to codes “pro/libertarian” (pro market) and “neutral/efficiency”. Occurrences of case descriptions or informant quotes were coded as “empirical”. Finally, we coded ideas presented as just a mean to get resources (e.g. democratic organising being a way to gain legitimacy) as “instrumental”. In case of a priori contradictory ideas within the same text, the final coding was based on a deeper analysis of the main paper’s arguments, identified through its semantic network. This coding allowed us to build a table summarising the global presence/absence of different philosophies in each document, and to establish a global overview of the ideas present in the literature. With this table, we identified philosophical tensions within each document and across the sample in general.

Findings: Normative Debates, Contradictions, and Ambiguity

Our results enabled the identification of the presence or absence of keys ideas from different political philosophies in the reviewed articles. Most papers do not take a clear normative stance, but many use ideas that can be linked to specific philosophies. In this section, we first present the philosophies underlying SE literature and their respective normative ideas in order of representation. Next, we introduce the normatively ambiguous concepts that are transversally employed by authors from different philosophical perspectives in the literature. Then, we explore the justifications and critiques around some central philosophies identified in the reviewed papers. Finally, we underline the central normative contradictions found in the literature.

Main Philosophy-related Themes

The most represented philosophies in the SE literature are Libertarianism, Citizenship/democracy, Egalitarianism, Communitarianism, and Utilitarianism. Table 2, below, summarises the different implicit philosophies in the social enterprise literature and associated verbatim, and Fig. 4 shows the number of papers presenting at least one occurrence of a theme.

Table 2 Underlying philosophies in the social enterprise literature
Fig. 4
figure 4

Number of papers containing at least one occurrence of a given philosophical theme. Reading: 39 paper mentioned at least once democracy-related ideas, 5 did it in a positive way, 18 through empirical examples, 12 in neutral terms and 4 in an instrumental way. (A paper can discuss several ideas but each paper is counted once for each philosophy, depending the main stance it take on it)

Libertarianism: Market and Property Rights

Libertarianism is associated with a defence of free market and limited state intervention. These ideas represent the socioeconomic context of SE well. Several authors note that the emergence of social enterprises is linked to a marketisation trend in the social sector. In several states there has been a shift in funding from subsidies to procurement contracts (Bull, 2008; Defourny & Nyssens, 2010; Miller et al., 2012; Nicholls, 2009), and/or welfare state withdrawal (Eikenberry & Kluver, 2004; Lehner, 2013; Mort et al., 2003; Roper & Cheney, 2005).

Different manifestations of SE reflect variations in welfare states’ situations (Chell et al., 2010), and the diversity of national contexts have led SE scholars to study how configurations of government (in)action and institutional infrastructure influence engagement in SE (Estrin et al., 2013; Stephan et al., 2015). In government action, the enforcement of contracts and property rights is the minimal state intervention, as accepted by libertarians. Several SE scholar mention these functions (Santos, 2012; Stephan et al., 2015), that are deemed necessary for the emergence of classical enterprise as well as SE (Estrin et al., 2013).

Weak property rights enforcement can be seen as detrimental to innovation in the case of intellectual property (Bradley et al., 2012; Desa, 2012), or as aggravating the conditions of poor populations. Weak property rights can, for example, prevent people from making use of their capital—for instance, by using their home as collateral for a loan (Bradley et al., 2012). Mair and Marti, (2009) examine how these “institutional voids” prevent participation in markets, and how SEs evolve in this context. As libertarians defend the rightful acquisition of property rights, indigenous claims to ownership of colonised land (Anderson et al., 2006) may also be compatible with this philosophy.

Citizenship: Democratic Goals and Means

A first view of citizenship is to conceptualise it as comprising rights (civil rights, political rights, social rights, etc.) (Kymlicka, 2002, p. 287). We will see in the Egalitarianism section that SE can aim to improve these rights. But, citizenship theory also focuses on the more active roles that citizens should take in order to maintain a functioning democracy, and there are examples of SE contributing to civic education and emancipation (Alvord et al., 2004; Parkinson & Howorth, 2008; Santos, 2012). One virtue or responsibility of citizens as defined in citizenship theories is to monitor elected officials (Kymlicka, 2002, p. 289). Non-profit organisations are considered to be important vehicles for this type of civic participation by facilitating the expression its constituencies’ voices (Eikenberry & Kluver, 2004), and the ability of SEs to take similar roles is discussed.

Citizenship theories consider that state democratic structures alone are not sufficient for an active democracy, and more deliberative forms of democracy are needed where voting is not the only way for citizens to participate (Kymlicka, 2002, p. 292). Establishing democratic structures in the economic sphere—where they are not the norm—can be an objective for a social enterprise. Indeed, some definitions of SE include this notion of participative and democratic structures (Anderson et al., 2006; Bull, 2008; Chell et al., 2010; Hockerts, 2017; Parkinson & Howorth, 2008; Perrini & Vurro, 2006; Roper & Cheney, 2005), and emphasise the inclusion of various stakeholders in SE governance or other participation channels (Defourny & Nyssens, 2010; Di Domenico et al., 2010; Hill et al., 2010). Democracy is also seen as a source of legitimacy (Smith et al., 2013).

Democratic governance can be transposed into particular legal statutes, leading to property rights structures other than investor-owned firms. Empirically, this “social ownership” (Shaw & Carter, 2007) notably takes the form of cooperatives (Corner & Ho, 2010; Defourny & Nyssens, 2010; Spear, 2006; Thompson & Doherty, 2006). The goal of building alternative democratic structures like cooperatives is thus associated with the broader aim to improve economic democracy (Defourny & Nyssens, 2010; Parkinson & Howorth, 2008).

In this context, the “monitoring” role of citizens can apply at the social enterprise level rather than the state level: democratic governance and members’ participation can be understood as a mode of control and accountability (Nicholls, 2010a). Some SE scholars show concern for deliberative democracy. Because the “social goals” of social enterprises are value-laden and potentially contested (Choi & Majumdar, 2014; Sharir & Lerner, 2006; Sud et al., 2009), they argue that their definition should remain eclectic (Dey & Steyaert, 2010) and be subject to deliberation (Cho, 2006).

Liberal Egalitarianism: Basic Rights and Needs

Egalitarian philosophies use concepts like primary goods, rights, or capabilities. Many SE scholars echo this, arguing that the goal of SE is to satisfy unmet social, basic, or human needs (Bacq & Janssen, 2011; Haugh, 2005; Lepoutre et al., 2013; Mair, & Marti, 2006; Nga & Shamuganathan 2010; Weerawardena, & Mort, 2006). Basic needs include, for example, freedom, equality, tolerance, and quality of life (Murphy & Coombes, 2009), and food, water, shelter, education, and medical services (Santos, 2012). Several researchers give examples of social enterprises supporting civil rights groups (Alvord et al., 2004), or developing human rights programmes (Desa, 2012; Mair & Marti, 2009; Meyskens et al., 2010; Sharir & Lerner, 2006). The explicitly egalitarian capability approach put forth by Amartya Sen is also directly referenced in some articles (Ansari et al., 2012; Bradley et al., 2012). Egalitarians also argue that primary goods, rights, or capabilities should be distributed equally, and similar concern for (in)equality of opportunities or wealth distribution was found (Cho, 2006; Kraus et al., 2014; Shaw & de Bruin, 2013) in our sample publications.

Communitarianism: Solidarity and Social Capital

Communitarian philosophy argues for greater focus on communities. It explores the shared identity, trust, and solidarity among people that is necessary to form a legitimate and stable “ethical community” in which collective decisions are accepted (Kymlicka, 2002, p. 257). In the SE context, communities can be included as partners (Ansari et al., 2012), target constituencies (Mair et al., 2012), or can themselves be entrepreneurial (Hall et al., 2012).

Some authors explore the collective action dynamics within and around SE to solve social problems (Dorado & Ventresca, 2013; Fowler, 2000; Waddock & Post, 1991). They explore how close networks, shared identity (Dorado & Ventresca, 2013; Seelos et al., 2011), common interests, and shared goals (Bacq & Janssen, 2011; Dorado & Ventresca, 2013; Waddock & Post, 1991) can be necessary for the emergence of SE through collective action processes within communities.

Nicholls identified among social entrepreneurs an institutional logic focussing on community, networks, and cooperative action (Nicholls, 2010a) in line with communitarianism. Mair et al., (2012) also noted that in relying on social capital, SEs tend to follow Boltanski and Thevenot’s “civic” order of worth, suggesting some links between communitarian and citizenship themes. Indeed, some authors note that non-profit organisations contribute to generating the social capital needed in democratic societies (Eikenberry & Kluver, 2004), and boost community cohesion (Di Domenico et al., 2010; Santos, 2012). Other scholars explore the role of communities from a more instrumental perspective by studying how SE uses social networks to gain advantages such as bargaining power (Perrini et al., 2010), resources (Di Domenico et al., 2010; Moore & Westley, 2011; Smith, & Stevens, 2010), capabilities (Nga & Shamuganathan, 2010), or legitimacy and reputation (Nga & Shamuganathan, 2010).

Utilitarianism: Maximisation of Human Welfare

The main idea of utilitarianism is to maximise human welfare, which it conceptualises as the aggregation of utilities (peoples' pleasure or preferences). Zahra et al. rely implicitly on utilitarianism by proposing to evaluate social ventures’ performance according to the ‘total wealth’ they create (Zahra, et al., 2009, p. 522). Santos defines social welfare by the aggregation of individual utility (Santos, 2012, p. 337). Townsend and Hart also offer a ParetianFootnote 1 definition of the ‘social goal’ of SE: the maximisation of social welfare without diminishing any individual’s utility (2008, p. 688).

Some argue that SE focuses on value creation over value capture (Mair & Marti, 2006; Santos, 2012). The ‘value’ here may imply ‘utility’. As mentioned by Zahra, classical economic approaches use profit as a proxy for utility creation, but it is difficult to use this metric to measure the utility created by non-profit social enterprises (Zahra et al., 2008). Some methodologies have, however, been proposed to evaluate the “value” (utility) created by social enterprises; these include cost–benefit analyses assessing the efficiency of social value creation (Santos, 2012), and the Social Return on Investment (SROI) (Nicholls, 2009). By summing the costs and benefits created by SE, these methods typically adopt a utilitarian approach.

Feminism: Gender Equality and Care

Feminist perspectives are diverse, but all focus on women’s interests, which have often been overlooked by mainstream political philosophies. Empirically, gender equality programmes are a well-represented example of SE action. Social enterprises can aim to address gender inequalities (Kistruck & Beamish, 2010; Murphy & Coombes, 2009; Seelos et al., 2011; Zahra et al., 2009), improve women’s rights (Mair & Marti, 2009), empower women (Datta & Gailey, 2012; Hayhurst, 2014; Saebi et al., 2019), or target them specifically (Littlewood & Holt, 2018; Mair et al., 2012; Sharir & Lerner, 2006; Short et al., 2009).

Feminist approaches also led to the formulation of ethics of care by taking inspiration from “feminine” moral reasoning. The ethics of care is concerned with what qualities people should have to act morally, rather than by defining the best moral principles. This approach echoes that of SE scholars investigating the qualities of SE founders. Authors note that social entrepreneurs should show virtues such as “love, integrity, honesty and empathy” (Mort et al., 2003, p. 83), have an ethics of care (Smith et al., 2013), or be altruist (Dempsey & Sanders, 2010; Tan et al., 2005). In the study of SE’s antecedents, scholars cite moral qualities such as compassion, emotions, and prosocial motivations (Miller et al., 2012), empathy and moral judgement (Bradley et al., 2012; Hockerts, 2017; Mair & Noboa, 2006; Saebi et al., 2019; Zahra et al., 2008), and prosocial identity and personality (Miller et al., 2012; Saebi et al., 2019).

Multiculturalism: Cultural Preservation and Norms

Multiculturalism has common ground with communitarianism, with a focus on cultural injustice. Multiculturalism questions how cultural minorities can be protected from external pressures of wider society to preserve their way of life, and asks how far a community can constrain its members to follow specific rules. These two questions are present in SE literature.

First, we find examples of SE aiming to preserve specific cultures and ways of life. For example, socially responsible tourism is described as a way to preserve cultural diversity in the world (Hall et al., 2012). Additionally, Anderson et al. (2006) describe SE by indigenous people as a way to improve their socioeconomic conditions while maintaining their traditional values.

On another hand, the tension between community rules and individual freedoms also exists in the SE context (Smith & Stevens, 2010). SE actions can disrupt existing social structures and norms; some authors see this as negative (Ansari et al., 2012) because it can disrupt the bonds of solidarity within existing communities (Dey & Steyaert, 2010), and give primacy to indigenous values (Anderson et al., 2006; Fowler, 2000). Others see the disruption of some social norms as positive; for example, when the change enables the participation of women (Alvord et al., 2004; Mair & Marti, 2009; Seelos et al., 2011). In this context, multiculturalist philosophies help to discuss which cultural norms should or should not be preserved.

Marxism: Exploitation and Self-realisation

Marxists denounce exploitation and alienation, and value self-realisation in work as a goal. One study presents an example of a social entrepreneur being explicitly informed about Marxism (Jones et al., 2008), but globally, few articles in the present sample expound Marxist ideas, and those which do mainly discuss questions of working conditions within SEs.

Dempsey and Sanders (2010) note that people working in non-profits have high life satisfaction, but question “how and when non-profit work might be oppressive” (Dempsey & Sanders, 2010, p. 440). Work in social organisations is associated with low pay (Dempsey & Sanders, 2010; Jones et al., 2008), which can either be criticised, or justified as an expression of solidarity with the beneficiaries (Dees, 2012). However, beneficiaries can also be employed themselves, and social enterprises may face tensions between improving their workers’ well-being or meeting other, conflicting goals (Smith et al., 2013). As a consequence, the literature contains examples of SE exploiting its workers (Littlewood & Holt, 2018; Smith et al., 2013) or participants in ‘Bottom of the Pyramid’ programmes expressing feeling like “guinea pigs” for big companies (Ansari et al., 2012).

Normatively Ambiguous Ideas

In our analysis, we have identified the concepts that permeate various different philosophies in the SE literature. Therefore, their use creates ambiguity since they are attributed with different, sometimes contradictory, normative perspectives. While the ambiguous use of concepts might not be clear in every paper, our work has allowed us to unravel the issue at the SE domain level, and to identify the central controverted elements of disadvantagedness, empowerment, profit, and efficiency. Figure 5 presents the occurences of these ideas in our corpus.

Fig. 5
figure 5

Ambiguous ideas occurences

Disadvantagedness

We found many definitions or empirical examples describing SE as helping disadvantaged people (Di Domenico et al., 2010; Nicholls, 2010b; Seelos et al., 2011; Smith et al., 2013; Stephan et al., 2015; Tracey et al., 2011; Waddock & Post, 1991; Weerawardena et al., 2010; Weerawardena & Mort, 2006). More specifically, this can concern disabled people or groups (Desa & Basu, 2013; Hockerts, 2006; Littlewood & Holt, 2018; Nicholls, 2010b; Sharir & Lerner, 2006; Tan et al., 2005; Weerawardena et al., 2010), or poor populations (Corner & Ho, 2010; Saebi et al., 2019; Shaw & Carter, 2007; Smith et al., 2012, 2013). Disadvantaged target populations also include minorities (Robinson, 2006; Sharir & Lerner, 2006) such as migrants (Meyskens et al., 2010; Smith et al., 2013; Townsend & Hart, 2008) and refugees (Corner & Ho, 2010).

The moral justification for why disadvantaged people deserve attention from SE, or the empirical explanation for why they get more attention, is ambiguous in the literature. The question of disadvantage originally comes from our egalitarian keywords, associated with Rawls’ principles arguing that natural and social inequalities are undeserved, and that the disadvantaged should thus be compensated (Kymlicka, 2002, p. 72). But, other justifications are possible. From an ethics of care perspective, disadvantaged people’s suffering may generate more caring responses because vulnerable populations generate empathy (Dees, 2012).

From a utilitarian perspective, Santo argues that helping the disadvantaged is not the only role of SE: “efforts to help advantaged populations may also constitute SE, as long as it involves addressing problems with positive externalities with a dominant goal of value creation” (Santos, 2012, p. 343). This is consistent with a goal to maximise aggregated utility. However, focussing on disadvantaged people can also be justified from a utilitarian perspective via the assumption of decreasing marginal utility. “People who lack resources will, in general, get more utility out of each additional resource than those who already have many resources” (Kymlicka, 2002, p. 40), so serving disadvantaged people may increase total utility more than serving more affluent customers.

Empowerment

Some scholars describe individual or collective entrepreneurship as a means of empowerment (Datta & Gailey, 2012; Sharir & Lerner, 2006). The concept of empowerment can be interpreted in different ways. The (contested) libertarian interpretation focuses on market participation. Hayhurst (2014) criticises the neoliberal empowerment narrative in which empowerment is market-based, and which sees individuals as responsible for being good wealth producers. Similarly, Ansari et al. (2012) mention that ‘Bottom of the Pyramid’ literature is focussed on employment and consumption as an empowerment solution, but argue instead for a capability-based approach to empowerment.

The empowerment process may include improving one’s economic situation (Datta & Gailey, 2012; Hayhurst, 2014), accessing resources (Datta & Gailey, 2012) or mobilise underused ones (Santos, 2012), and improving assets and/or capabilities (Mair & Marti, 2009; Santos, 2012), skills, and confidence (Estrin et al., 2013; Hayhurst, 2014). Empowerment also includes some citizenship and democracy themes such as gaining political capital (Mair et al., 2012), giving a voice to people (Mair & Marti, 2009), enabling participation in the public sphere (Cho, 2006; Datta & Gailey, 2012), and boosting the capacity to face powerful adversaries (Alvord et al., 2004) and elites (Mair & Marti, 2009). From a feminist perspective, it is also important to note that empowerment may include the ability to contribute to the family and to make decisions in the household (Datta & Gailey, 2012).

Ultimately, empowerment is also described as the ability to make choices and transform action in outcomes (Santos, 2012). This includes agency and control over one’s decision (Datta & Gailey, 2012) and/or responsibility for, and control over, one’s life (Hayhurst, 2014; Thompson et al., 2000). This question of agency is associated with entrepreneurial behaviour itself (Datta & Gailey, 2012) and the process of turning ideas into reality (Bacq & Janssen, 2011). Some authors therefore describe empowerment as the involvement of stakeholders, communities, and beneficiaries in the process of solving their own problems (Chell et al., 2010; Santos, 2012; Zahra et al., 2014).

Profit

The literature suggests a continuum in engagement in profit maximisation versus social goals (Stevens et al., 2015). Many definitions of SE mention that its goal is not to maximise profit (Bacq & Janssen, 2011; Dacin et al., 2011; Dacin et al., 2010; Dempsey & Sanders, 2010; Littlewood & Holt, 2018; Rivera-Santos et al., 2015; Shaw & Carter, 2007; Smith et al., 2013; Tan et al., 2005; Townsend & Hart, 2008; Zahra et al., 2009), as per the UK government’s definition (Bull, 2008; Chell, 2007; Nicholls, 2010b; Tracey et al., 2011). A similar but more nuanced perspective suggests that there is simply no overemphasis on shareholders’ wealth maximisation (Nga & Shamuganathan, 2010), that profit and wealth are means rather than ends (Perrini & Vurro, 2006), and that SE aims to survive in the market without maximising economic value (Smith et al., 2012; Wilson & Post, 2013).

At the other end of the spectrum, SE scholars discuss the full compatibility between “social” and profit objectives. Some suggest that business revenue generation methods can be used for social purposes (Peredo & McLean, 2006), while others discuss double/triple bottom line and ‘win–win’ arguments regarding the compatibility between social and profit objectives (Hill et al., 2010; Lepoutre et al., 2013; Santos, 2012; Townsend & Hart, 2008), and entrepreneurs trying to maximise both profit and social change (Dacin et al., 2011). The ‘Bottom of the Pyramid’ approach reflects this stance by arguing that poverty eradication is reconcilable with a profit-maximising objective within an enterprise-based market system (Ansari et al., 2012).

These divergent perspectives on profit can be explained by their different philosophical foundations. From a libertarian view, the profit-making goal is legitimate in a stakeholder-owned firm since the owners should be free to use their property as they wish (this is in essence Milton Friedman’s oft-cited argument about firms’ responsibility to make profits). From a utilitarian perspective this can be legitimate too as long as it also supports global wealth improvement (this may, for example, fit the ‘inclusive growth’ paradigm). However, perspectives advocating using profits to benefit the disadvantaged (Calic & Mosakowski, 2016) may instead fit the egalitarian perspective where profit distribution to stockholders may be seen as illegitimate. Indeed, if stockholders are already wealthier than the targeted beneficiaries, dividend distribution may increase inequality.

Efficiency

The efficiency principle, present in SE discourses, has been studied through Boltanski and Thevenot industrial order of worth (Mair et al., 2012), the “business” meta-narrative used in SE rhetorical strategies (Ruebottom, 2013), and the semantic repertoire of social entrepreneurs (Dey & Steyaert, 2010; Jones et al., 2008). Moreover, some SE scholars worry that efficiency discourses disarm radical and critical approaches (Cho, 2006; Parkinson & Howorth, 2008). In our analysis, the question of efficiency was originally included in the utilitarian category; however, utilitarianism is not the only philosophy with concern for efficiency (Sen, 2001), and this theme remains ambiguous in the SE literature in terms of normative anchorage.

SE scholars discuss the divergence of views about the most efficient means to solve social problems (Miller et al., 2012) and which organisational form is the most efficient (Kistruck & Beamish, 2010; Wilson & Post, 2013). But, we may ask: efficient to do what? In our sample, authors discuss how social enterprise can be sufficiently effective and efficient to serve basic needs (Austin et al., 2006; Seelos & Mair, 2005), solve market failures (Smith & Stevens, 2010), deliver services and products (Choi & Majumdar, 2014; Miller et al., 2012), and increase global sustainable well-being (Zahra et al., 2014).

As we have seen in the previous sections, different finalities can be related to different philosophies, and efficiency in solving market failures may differ from efficiency in catering for basic needs, or improving equality or democracy. Regardless of the philosophical anchorage, efficiency arguments should also be supported by causal connections (Sen, 2001, p. 147). In facing the efficiency question, SE scholars should be more precise about the targeted outcomes, and look for empirical evidence supporting efficiency claims.

Justifications and Critique

Through the qualitative analysis of the normative basis of the reviewed papers, we could establish how different philosophies were evaluated by the scholars, and by looking at pro/cons analysis in each paper, we identified debates in the field of SE around some philosophical normative perspectives. In this sub-section, we delineate the main arguments around each of the most criticised philosophies—libertarianism, utilitarianism, and feminism (ethics of care)—as found in the SE literature.

… of Libertarianism

As has been discussed, libertarianism is in favour of free markets. The SE literature is controversial regarding market solutions. Being commercial, trading, contracting or market-oriented is in some cases part of the very definition of SE (Choi & Majumdar, 2014; Haugh, 2005), or among the empirical examples (Defourny & Nyssens, 2010; Di Domenico et al., 2010). We found some justifications of commercial solutions, as making people pay may “empower them to complain” (Dees, 2012), decentralised commercial action may be an efficient solution to social problems (Estrin et al., 2013; Santos, 2012), and the presence of social enterprises in the market may pressure classical businesses to change practices (Wilson & Post, 2013).

As a result, some argue it is possible to achieve ‘positive social impact’ and profit at the same time in a market system (Wilson & Post, 2013, p. 729). The general idea is well summarised by Kofi Annan, quoted at the end of Wilson and Post’s article:

“Let us choose to unite the power of markets with the strength of universal ideals. Let us choose to reconcile the creative forces of private entrepreneurship with the needs of the disadvantaged and the requirements of future generations.” (Kofi Annan, cited by Wilson & Post, 2013, p. 730)

Within classical firm theory, utilitarian philosophy usually fits well with market solutions (and thus libertarian ideas) because classical economics predicts that perfect market competition and its ‘invisible hand’ will lead to efficient outcomes and maximal social welfare (Nga & Shamuganathan, 2010; Santos, 2012). However, market failures can occur in cases of public goods provision (Nga & Shamuganathan, 2010), externalities, information asymmetries, principal-agent problems, etc. For some authors, solving market failures (Austin et al., 2006, p. 2; Desa & Basu, 2013, p. 27) or producing positive externalities (Santos, 2012) is thus the distinctive domain of SE.

However, the market orientation seems in tension with the “social” goals of SE (Smith et al., 2013), and several papers are overtly critical of market-based solutions. Political philosophy can help in understanding the standpoints of these critiques. First, in the democracy critique, marketisation is described as threatening civil society and democracy (Eikenberry & Kluver, 2004; Fowler, 2000). Second, from the utilitarian perspective, it is paradoxical to “solve market failures” with market-based solutions. As Santos (2012) notes, if SE “are indeed tackling areas of neglected externalities, market-based mechanisms may not be the most efficient form of economic organisation to address them.” (p. 345). Finally, from the egalitarian standpoint, market competition does not solve inequalities (Hall et al., 2012; Santos, 2012), and neoliberal approaches focussing on individual responsibility do not fix the structure of society, thus creating marginalisation (Hayhurst, 2014).

… of Utilitarianism

While we showed in the previous sections that some authors endorse a utilitarian approach to SE, others define social enterprises in opposition to it. For example, Cho (2006) argues that the social dimension of SE cannot fit a utilitarian framework because SE is opposed to the for-profit model, thus implying a definition of the social good “irreducible to and greater than the sum total of individual welfare functions” (Cho, 2006, p. 37). Stevens et al. (2015) also oppose the “utilitarian” organisational identity of SE—including economic rationality, maximisation of profits, and self-interest—with the “normative” identity associated with social mission, traditions and symbols, ideology, and altruism (Stevens et al., 2015).

Others also criticise the discursive shift “from democratic structures to a focus on social purpose” (Parkinson & Howorth, 2008, p. 291). This shift pushes policy-makers to assess social enterprises on their results rather than on their operating methods (Eikenberry & Kluver, 2004). Furthermore, this shift could be interpreted as a transition from democratic concerns towards more utilitarian judgements.

… of Feminism (Ethics of Care)

In the sample papers, discussions around caring behaviours mainly concern efficiency issues. SE scholars worry that compassion and empathy may prevent entrepreneurs from identifying the right social problems (J. Dees, 2012) or its solutions (Miller et al., 2012), and that affective commitment may be detrimental to the start-up process (Renko, 2013). Others wonder what would happen if the social entrepreneur is not virtuous enough (Zahra et al., 2009). Dees (2012) also mentions that, since charities are led by compassion, they may worry less about being efficient. Austin et al. (2006) declare that SE can survive without being efficient in the absence of market forces sanctioning or rewarding performance.

In response, we find an emerging composite argument associating ideas of efficient markets (utilitarian-libertarian), and philanthropy (care) through ‘philanthropic markets’ (Austin et al., 2006) and ‘social return on investment’. Financial investors might trade some profit for social value creation (Nga & Shamuganathan, 2010) or respect for social norms (Townsend & Hart, 2008). Social impact measurement and reporting may thus help to build an “information rich” market (Nicholls, 2009, p. 766) in which the “invisible hand of distributed other-regarding action” can generate efficient outcomes (Santos, 2012). This emerging argument making the case for philanthropic efficiency is, however, incomplete without clarity on which social outcomes these markets are aiming for, and stronger theoretical arguments explain how this could work.

Tensions and Contradictions

Our in-depth qualitative analysis also focussed on the ways in which these philosophies were applied in the SE literature. Consequently, we could pinpoint inconsistencies not only in their use in the overall domain, but also in specific studies. Below, we delineate contradictions between macro discourses and micro practices as well as theoretical issues and empirical arguments which could dampen the impact of the studies.

Macro–micro

Notably, although libertarian and utilitarian ideas mainly concern macro-level discourses and institutions such as state policies, market systems and funding institutions, few examples of social enterprises defending such ideas were found. Empirical examples focus on egalitarian, democracy, or other goals (helping the disadvantaged, fulfilling basic needs, improving civil rights, preserving cultures, etc.). Thus, our analysis agree that “discursive shifts, driven by policy-makers, funders, the sector and academics alike, do not necessarily infiltrate ideology at the level where the action is located” (Parkinson & Howorth, 2008, p. 305).

Starting from this point, political philosophy can help us build more precise research questions regarding the paradoxes within SE, which may be the product of the tension between SEs and their environment. For example, rather than asking how SEs manage “social-business tensions” (Smith et al., 2013), we can ask “how SEs manage the pursuit of egalitarian goals in a market-based environment”. Such investigations may lead to identifying more precise SE strategic responses, e.g. how fair trade develops based on the cooperative organisation of a value chain.

Empirical-theoretical

Some of the clearest theoretical propositions around social enterprises are rooted in economic theories (Santos, 2012; Zahra et al., 2009) and aspire to positivism (Nicholls, 2009; Santos, 2012). However while economic theories tend to appear “objective”, they also tend to be rooted in utilitarianism (Myrdal, 1969, p. 44; 1990). We find this tendency reflected in that Stevens et al. (2015) oppose “utilitarian” and “normative” organisational identity, as though utilitarianism had no normative content.

Relying on utility-oriented (economic) theories may lead to discrepancies between the philosophy behind the theoretical framing and the philosophy adopted by on-the-ground actors. For example, Santos (2012) frame SE in utilitarian terms but cite the example of Unis-cité, which aims to foster civic education and social cohesion—citizenship concerns. Philosophical tensions can also emerge in the choice of key variables. For example, in their study of the organisation BRAC, Mair and Marti (2009) choose to study market participation as the main outcome even though BRAC frames its objectives as poverty reduction. Given the value-laden nature of SE, researchers should be conscious of whether they adopt the organisational perspective (emic) or judge which outcomes are valuable (etic).

Discussion

Social entrepreneurship research is a growing field containing divergences and debates. Previous researchers have identified different schools of thought on SE (Bacq & Janssen, 2011; Dees & Anderson, 2006; Defourny & Nyssens, 2010; Galera & Borzaga, 2009; Young & Lecy, 2014). Starting from the assumption that SE research contains implicit normative concerns, our paper presents a complementary lens based on political philosophy, shows how different normative anchorages are linked to different theoretical approaches, and identifies points of divergence in the literature. In this section, we elaborate on the implications for SE research in developing more reflexive research and building normative theories.

Identifying Contradictory and Ambiguous Contours

Our paper has traced contours of the SE field based on the dominant political philosophies, namely Libertarianism, Citizenship/democracy, Feminism, Egalitarianism, Communitarianism, and Utilitarianism, with a smaller presence of both Multiculturalism and Marxism. Our results indicate prior researchers’ main implicit normative assumptions (maximisation, rights, democracy, care, solidarity, etc.) that represent the original debates on SE. We also identify less well-represented ideas (cultural preservation, exploitation, etc.) that address more specific normative concerns.

Our paper contributes to the SE literature by showing how this multiplicity of normative assumptions can lead to contradictory and ambiguous arguments within a single piece of research, as well as across the SE field. We add to the discussion by Heath et al. (2010) who, from a similar political philosophy perspective, identify inconsistency in the use of distinct normative theories at different levels of analysis in business ethics research. While we identify a similar issue regarding macro and micro levels, our analysis also finds contradictions between theoretical and empirical elements within studies which undermine their robustness, and a conceptual ambiguity in the SE literature as a whole which calls into question the clarity and consistency of its research object.

Acknowledging Normativity

Given the diversity of possible philosophical stances regarding SE, we believe it is essential that SE researchers are reflexive regarding normativity in their research. Although some theories aspire to be ‘positive’ (Nicholls, 2009; Santos, 2012), we agree with Myrdal (1969) that “the only way in which we can strive for ‘objectivity’ in theoretical analysis is to expose the valuations to full light, make them conscious, specific, and explicit, and permit them to determine the theoretical research” (p. 56). SE research, like business research in general (Taylor et al., 2015) may therefore benefit from more explicit “value premises” (Myrdal, 1969), whether these stem from current public policies, social enterprises, or the researcher herself. Normative philosophies can thus act as meta-theories, along with other philosophical assumptions regarding ontology and epistemology (Pittaway, 2005).

As different normative anchorages will lead to different approaches to defining and evaluating the ‘success’ or ‘impact’ of SEs, political philosophy can aid coherent choices about tools and methods for evaluating SE. It can help in defining value-laden concepts (e.g. if an externality is positive or negative, or what we mean by empowerment), and purposefully decide what should be measured and how. To take one example, Social Return On Investment (SROI), inspired by cost–benefit analysis (Millar & Hall, 2013), may be coherent with utilitarian reasoning, but will likely be a poor match for egalitarian philosophies since cost–benefit analysis usually neglect the fulfilment or violation of rights, improvements in liberty (Sen, 2000, p. 944), and the distribution of the value created (inequalities). For this reason, some measurement tools can conflict with an organisation’s values (Millar & Hall, 2013), and choosing a universal or standardised tool to measure SE ‘performance’ can be problematic.

Building normative social entrepreneurship theories

Because the adjective “social” in SE is a value-laden concept (Choi & Majumdar, 2014), conceptual definitions of SE may also contain normative implications, and deciding which organisations to include under the “SE” umbrella therefore becomes a political decision (Lyon & Sepulveda, 2009). Various fields of study like corporate social responsibility or stakeholder theory encounter a similar issue and thus develop both normative theories and descriptive studies. SE, however, lacks a well-defined normative logic (Nicholls, 2010a, p. 617), and future research should develop more explicit normative theories. This would correspond to what Dey and Steyaert (2012) call a normative critique: being “explicit about the kind of trajectory social entrepreneurship must endorse” (p.97). Normative theories can inspire practitioners and aid policy-makers in clarifying their political choice when adopting a definition of SE.

The results we present here constitute a basis for such a task, but the issue remain complex. Coherent normative theories at the organisational level are lacking, and this problem is not limited to social enterprises. Organisations are not “citizens”, nor “mini-states” (Heath et al., 2010), so “classical political theory and individual moral theory are inadequate for dealing with the moral problems that arise in the context of the modern corporation” (Phillips, 2003, p. 41). As a response to this problem, Heath et al. (2010) call for a ‘unified theory’ bridging business ethics and political philosophy, which would justify various institutions like markets and their regulation, corporate governance, and business norms with consistent normative concepts.

The question is relevant for SE since “what seem really at stake beyond conceptual debates are the place and the role of social enterprise within the overall economy and its interaction with the market, the civil society and public policies” (Defourny & Nyssens, 2010, p. 33). We have also seen that the normative debates around SE not only concern social enterprises, but question the articulation of different levels in society: individual and collective behaviour, forms of organisations, inter-organisational relationships, markets, state policy, etc. We thus believe that the construction of ‘unified theories’, and exploration of the causal mechanisms supporting the efficiency claims they may contain, are important avenues of future research for SE and business ethics in general.

Conclusion

The past research has shown that defining social enterprises or their social goal is a political task involving value-laden debates (Boddice, 2011; Cho, 2006; Choi & Majumdar, 2014; Lyon & Sepulveda, 2009). We have extended this work by exploring the political ideas at stake behind the question of social entrepreneurship. Our paper shows the diversity of perspectives in SE, and clarifies the goals and means SE can adopt within society according to different philosophical lenses. Political philosophy provides a coherent framework for conceptualising SE definitions, goals, and impacts, both in normative theories and in describing the political projects that social enterprises themselves can defend.

Furthermore, our analysis is also relevant from an institutional perspective of SE research. First, it can help scholars in understanding their own normative ideas about social enterprise, and making them explicit. This process will provide a solid basis for critical approaches, helping scholars to define what is criticised and from which standpoint. Then, political philosophy can inform the choice of theories, conceptualisations, and measurement tools to use. Different measurement tools and different conceptualisations reflect different types of political impacts (creating more utility, fulfilling rights, fostering democracy, etc.). Thus, scholars and practitioners should pay attention to the alignment between their values and the measurement tools or theories they choose to use.