The Meaning of Organizational Learning

The Functionalist Paradigm

The functionalist paradigm, which is dominant in organizational and OL studies, is based on the ontology of being that conceptualizes reality as external and objective and phenomena as “fundamentally ‘real’,” discrete, and static, as summarized by Chia (1995, p. 585; see also Boal, Hunt, & Jaros, 2003). The functionalist epistemology of consensus strives to obtain knowledge that mirrors reality and is value-free, valid, and reliable, and can explain and predict individual and collective behavior (Deetz, 1996; Donaldson, 2003b; Gergen & Thatchenkery, 1996; Reed, 1997).

Based on these assumptions, functionalists conceptualize organizations as rational hierarchies and systems with formalized structures, identifiable boundaries, and describable attributes (Argyris & Schön, 1978; Clark, 1985; Cyert & March, 1963/1992; Scott, 1975). For them, organizations aim to achieve specific outcomes, such as developing new products and services and improving performance and effectiveness (Daft, 2009; Duncan & Weiss, 1979; Kogut & Zander, 1992; Parsons, 1951, 1956). Functionalist studies assume the determinism of environmental influences on organizations in terms of accessing resources and imposing changes on their structures (Aldrich, 1992; Donaldson, 2003a).

Three major functionalist approaches to OL are behavioral, cognitive, and the approach based on Parsons’ (1951) social action theory. The behavioral approach considers OL an adaptation to change in the external environment through changes in organizational behaviors (Lant & Mezias, 1992; Levinthal & March, 1993). The cognitive approach views organizations as open information-processing and interpretive systems that have cognitions and memories so that OL involves developing accurate representations of the external world and changing organizational knowledge structures (Argyris & Schön, 1978, 1996; Daft & Weick, 1984; Magalhães, 1998; Walsh & Ungson, 1991). The approach based on social action theory considers OL as a system of activities that brings information into organization, interprets, disseminates, and stores it. As a result, organizations increase their adaptive capacity (Casey, 2005; Schwandt & Marquardt, 2000).

Despite differences between these three approaches, they share several important similarities. All approaches are predicated on the assumption that OL is a linear process that is the same across all levels of analysis. As a result, researchers frequently borrow models and metaphors developed for learning at the individual level to analyze collective learning processes (Dodgson, 1993; Hedberg, 1981). All three approaches view OL as purposive organized activity shaped by organization leaders (Argyris & Schön, 1978, 1996; Daft & Weick, 1984; Mackenzie, 1994). Thus, the discussion of functionalist conceptualization of OL below will draw on all three approaches.

From the functionalist standpoint, OL is a synonym of information processing within organizations that involves the dichotomy of “the exploration of new possibilities and the exploitation of old certainties” (March, 1991, p. 71). Organizations receive information from the environment in multiple ways, for example, spontaneously through economic and political news or through deliberate search (Daft & Huber, 1987; Daft & Weick, 1984; Huber, 1991). Organizations “sift” external information through their “perceptual filters,” or worldviews that restructure existing organizational cognitive linkages and diminish equivocality in their systems (Daft & Weick, 1984; Hedberg, 1981; Levinthal & March, 1993; Murray, 2002; Schwandt & Marquardt, 2000). Information is then interpreted and becomes knowledge, which is disseminated within organizations through formal processes, policies, technologies, and informal interpersonal communication influenced by perceptions of approachability of the knowledge source, his or her credibility, and trustworthiness (Andrews & Delahaye, 2000; Epple, Argote, & Devadas, 1991; Snyder, 1996; Tompkins, 1995).

Organizations store knowledge in their memory (Cyert & March, 1963/1992; DeFillippi & Ornstein, 2005; Hedberg, 1981; Huber, 1991; Levitt & March, 1988; Walsh & Ungson, 1991). Organizational memory is organized in five “bins”: individuals, culture, transformations, structures, and ecology (Walsh & Ungson, 1991). As Olivera (2000) shows, technology-based databases and social networks within organizations function similarly to these memory bins. Social networks are instrumental in retrieving knowledge from technology-based organizational memory systems, providing fast access to needed contextualized knowledge.

OL processes within organizations have differing levels of complexity and outcomes. At the lower (single-loop) level, OL results in developing a rudimentary association between behaviors and outcomes, adapting to the environment, and improving organizational effectiveness (Argyris & Schön, 1978, 1996; Fiol & Lyles, 1985; Friedman, 2001; Lam, 2001; Levitt & March, 1988). At the higher level (double-loop) learning, organizations restructure their knowledge, memories, and build new capabilities (Argote, 1999; Argyris & Schön, 1978, 1996; Fiol & Lyles, 1985; Huber, 1991; Mackenzie, 1994). The triple-loop learning (deutero-learning) enables organizations to learn about their learning processes (Argyris & Schön, 1978, 1996).

The functionalist paradigm considers individuals as key agents in collecting, interpreting, disseminating, storing, and retrieving information within organizations. Functionalist studies show that OL levels of analysis are connected by individuals interacting with organizational structures that provide frames of reference for individual cognitions and behaviors. As a result of this interaction, individual mental models are incorporated into organizational mental models (Duncan & Weiss, 1979; Jelinek & Litterer, 1994; Kim, 2004; Kogut & Zander, 1992; Levitt & March, 1988; Murray, 2002). At the same time, the biggest weakness of the functionalist paradigm is the anthropomorphization of OL by theoretizing it as fundamentally similar across all levels of analysis and thus not explaining the connections between individuals and the collective (Glynn, Lant, & Milliken, 1994).

Functionalist researchers assume that learning at the individual level of analysis aggregates at the collective level but do not discuss specifics of such aggregation (Cohen & Levinthal, 1990; Hedberg, 1981; Jelinek & Litterer, 1994; C. L. Wang & Ahmed, 2003). For example, Cangelosi and Dill (1965) consider learning as a tension between adaptation at the individual and collective levels of analysis but do not explain the connection between them. Similarly, March and Olsen (1976) do not clarify the interconnection of individual beliefs, individual action, organizational action, and environmental response that are parts of the OL process. Kim (2004) conceptualizes OL as a double-loop process (in terms of Argyris & Schön, 1978) that links individual mental models to organizational mental models, although the linkage itself remains unclear.

The Critical Paradigm

The critical paradigm is marginal in organization and OL studies and is based on the ontology of being, but adopts the epistemology of dissensus (Deetz, 1996). Critical theorists “de-objectify” social structures, recognize their historical and contextual situatedness, and underscore the impacts of technical rationality—or concern with improved effectiveness—on human communication (Karataş-Özkan & Murphy, 2010; Scherer, 2009; Steffy & Grimes, 1986; Willmott, 2003). The critical paradigm exposes the use of power, challenges inequality within organizations, and reveals sacrifices that organizations make in the name of efficiency, effectiveness, and profitability (Fenwick, 2004).

The critical paradigm in OL is influenced by Marxian concepts of dialectics, determinism of structure over superstructure, and oppression inherent in human condition (Agger, 1991; Crotty, 1998; Hardy & Clegg, 1997). Another key contribution to the critical paradigm is the Frankfurt School and the work of Habermas (Willmott, 2003). Habermas (1972) views humans as organizing their reality according to cognitive interests that include producing technical knowledge to improve effectiveness (technical interest), achieving mutual understanding with others (communicative action), and emancipation. Communicative action is one of the central components of Habermas’s (1984, 1987) philosophy, and he believes that it is suppressed by discourses of technical reasoning or one-directional thinking guided by “the theoretical and hypothetical, and focusing on control through the development of means-ends chains” (Alvesson & Deetz, 1996, p. 200; Steffy & Grimes, 1986).

From a critical perspective, organizations are discrete entities that develop historically and contain structures of domination reinforced by technology (Morgan & Spicer, 2009). Organizations are contested terrains of knowledge, which is shaped by technical rationality, assumed homogeneity of individual and collective identities, equating management self-interests with interests of organizations, and organizational cultures that reproduce dominant discourses (Alvesson & Deetz, 1996; Fenwick, 2004; Grant, Iedema, & Oswick, 2009; Karataş-Özkan & Murphy, 2010).

There are very few conceptual and empirical studies of OL from a critical perspective. For example, Coopey and Burgoyne (2000) view OL as a function of political processes within organizations that frequently deny employees their learning space and suppress communicative action. In the critical paradigm, learning is a way to foster organizational change, and researchers are agents of such change. For example, Meyerson and Kolb (2000) facilitated organizational change through action research where they worked collaboratively with participants within the organization to develop critical gender pedagogy.

Among the critical paradigm’s weaknesses is lack of empirical studies and deterministic view of social and organizational structures (Alvesson & Deetz, 1996; Hardy & Clegg, 1997). Critical theorists view power as oppressive and emanating from organizational leadership, though research shows that all organization members strive to gain power over others (as summarized by Bouchikhi, 1998). Similar to functionalist OL research, the critical paradigm emphasizes individual action and learning as opposed to the collective, and when discussing the dialectic of individual and organization in OL, it is not clear how this dialectic comes into being.

The Constructionist Paradigm

The constructionist paradigm in OL is grounded in the ontology of becoming that views reality as emergent, incomplete, and consisting of micro-practices of social interaction (Crotty, 1998). The reality is socially constructed through a process of individuals externalizing their experiences, objectifying, and internalizing them (Berger & Luckmann, 1967; Karataş-Özkan & Murphy, 2010). Constructionists adopt the epistemology of consensus; they examine how “‘the-taken-for-granted’ becomes taken for granted” (Czarniawska, 2003, p. 137), but do not problematize existing social order. They believe that the knowledge of reality is socially constructed, practice-based, and tentative, and is rooted in the value systems of people who create it (Boland & Tenkasi, 1995; Brown & Duguid, 1991; Cook & Brown, 1999; Gherardi, 1999; Gherardi & Nicolini, 2001; Nonaka, 1994; Yanow, 2004).

From the constructionist perspective, organizations are evolving heterarchical fragmented entities that are socially constructed and enacted through symbolically mediated interactions as their members negotiate, make sense of their experience, and produce and reproduce organizational structures (Dandridge, Mitroff, & Joyce, 1980; Gray, Bougon, & Donnellon, 1985; Hatch, 2006; Langley & Tsoukas, 2010; Reed, 2003; Weick, 1979, 2001). Constructionists consider the environment inseparable from cognition so that organizations react to their perceptions of reality and enact the environment they perceive and anticipate (Hatch, 2006; Weick, 1979, 2003). Organizations symbolically construct their boundaries and articulate them through rituals, symbols, and myths (Dandridge et al., 1980; Turner, 1986, 1992). Their structures and routines are emergent and prone to improvisation (Weick, 1998) that occurs in action nets (Czarniawska, 2008), activity systems (Engeström, 1999, 2001), and spontaneous individual activities (Gabriel, 1995).

Constructionists view OL as “generative social practice in the lived-in world” that produces and reproduces social structures (Gergen, 1985; Gherardi, 1999; Gherardi & Nicolini, 2001; Ghosh, 2004; Giddens, 1984; Lave & Wenger, 1991, p. 35). Thus, OL involves not only acquiring abstract formal knowledge about the practice but becoming a practitioner through participating in collective practices (Brown & Duguid, 1991; David & Victor, 2002; Elkjaer, 2005). When participating in practices, individuals acquire, sustain, and change “intersubjective meanings through the artifactual vehicles of their expression and transmission and the collective actions of the group” (Cook & Yanow, 1993, p. 384; Easterby-Smith & Araujo, 1999; Gherardi, Nicolini, & Odella, 1998; Peters, Gassenheimer, & Johnston, 2009). Participation in practices is a foundation for organizational memory, which is not equivalent to psychological memory because it does not contain an accurate picture of the past, as suggested by functionalist scholars (Rowlinson et al., 2010), but rather its interpretation.

Constructionist views of OL have been shaped by two major theoretical developments: the structuration theory of Giddens (1979, 1982, 1984) and conceptualizing organizations as cultures (Cook & Yanow, 1993). For Giddens (1982), social life is constituted through “regularized practices, produced and reproduced by social actors in the contingent context of social life” (p. 110; see also Tenkasi & Boland, 1993). Structuration as an underlying process of OL affects texts and communication patterns (genres) that people enact to “realize particular social purposes” (Chanal, 2004; Orlikowski & Yates, 1994, p. 542). Berends, Boersma, and Weggeman (2003) show that OL is instantiated in organizational practices as a result of the interaction of individuals and organizational structures. According to Y.-J. Lee and Roth (2007), individual actions reproduce these structures in a novel form so that individuals and organization learn simultaneously.

For scholars who view organizations as cultures, OL is similar to culture because it involves both collective thinking patterns and collective actions. Similar to the tacit nature of cultural knowledge, organizational knowledge is tacit and is captured in artifacts that are situated in organizational culture (Cook & Yanow, 1993). Language is a part of culture and is a socially constructed and contextually embedded artifact that mediates people’s construction of reality (Berger & Luckmann, 1967; Brown, Collins, & Duguid, 1989; Brown & Duguid, 1991; Gergen, 1985; Gherardi, 1999; Hatch, 2006). As such, language is central to OL because it is through language that individuals articulate their practices; develop common reference points, metaphors, and intersubjectivity; and learn to participate in communities of practitioners (Brown & Duguid, 1991; Elkjaer, 2005; Plaskoff, 2005).

Constructionists enrich our understanding of OL at the collective level of analysis by paying attention to practices and social relationships within organizations (Gherardi et al., 1998). The social relationships link learning at the individual and organizational levels of analysis through the interplay of tacit and explicit knowledge, thus building trust and facilitating dialogue that are crucial for learning (Peroune, 2007). For constructionists, learning occurs through informal interactions in collective networks that have their own power balance (Gherardi & Nicolini, 2001; Marsick, Watkins, Callahan, & Volpe, 2009; Peters et al., 2009; Smith, 2006) and communities of practice (COPs). COP is a “set of relations among persons, activity, and world, overtime and in relation with other tangential and overlapping communities of practice” (Wenger, 1998, p. 98). COPs provide interpretive support necessary for making sense of people’s experiences, capture tacit knowledge within organizations, and build shared collective practices (Anderson & Anderson, 2005; Hildreth, Wright, & Kimble, 2000; Lave & Wenger, 1991).

One weakness of constructionist research is underestimating the constraints imposed on individuals by organizational structures (Huysman, 1999). With the focus on tacit knowledge and informal social interaction in organizations, constructionists discount formal OL processes and explicit knowledge (Ben-Ari, 2004). By concentrating on social interaction as the underlying process of OL, the constructionist paradigm marginalizes the increase in explicit knowledge as an important learning outcome. In addition, as significant as the environment is for organizations and OL, constructionist researchers do not consider it an independent and influential factor. Finally, Allard-Poesi (2005) argues that while acknowledging the socially constructed nature of reality, constructionist researchers disengage from it and do not convey the emergent and fluid nature of learning as they claim to do.

The Post-Modernist Paradigm

Similar to the critical paradigm, the post-modernist OL paradigm is on the fringe of organization studies. It includes a heterogeneous group of theories that range from Foucault’s theoretizing about discourses, knowledge, and power to actor-network theory and complexity theory. These theories share the ontology of becoming and look for rupture and discontinuity (Chia, 1995, 2003a; Chiva, Grandío, & Alegre, 2010; M. M. Lee, 2014; Reed, 1997; Tsoukas & Chia, 2002). From a post-modernist perspective, there are multiple complex provisional realities that are constructed through discourses and media (Baudrillard, 1994; Chia, 2003b; Cooper & Burrell, 1988; Foucault, 1984; Karataş-Özkan & Murphy, 2010; Reed, 1997). The epistemology of post-modernism is dissensus and a belief that it is impossible to create an accurate account of truth because there is no objective reality and no universal truths (Hassard, 1996; Magalhães, 1998; Parker, 1992). Post-modernism celebrates complexity, difference, diversity, and spatial dimensions of human activities. It gives “voice and legitimacy to those tacit and oftentimes unrepresentable forms of knowledgeable” that are frequently overlooked in other paradigms (Agger, 1991; Calton & Kurland, 1996; Chia, 2003b, p. 127; Chiva et al., 2010; Grobman, 2005; Hernes, 2004; Lee, 2014; Stacey, 1995; Tsoukas & Chia, 2002).

Post-modernists view organizations as emergent complex entities that exist in ambiguous and uncertain environments and contain micro-practices, which have no meaning outside human activity. For Chia (2003b), organizations are “nothing more than islands of relatively stabilized relational orders in a sea of ceaseless change” (p. 131). Post-modernists believe that the world appears to us through language and discourses, which are full of contradictions and inconsistencies (Alvesson & Deetz, 1996; Brown, 1990; Foucault, 1984; Gergen & Thatchenkery, 1996; Grant et al., 2009) Thus, post-modernist studies often conceptualize organizations as texts that are produced by and in discourse so that they are “miniaturized, decentered and localized discursive or representational practice” (Chiva et al., 2010; Cooper, 1989; Cooper & Burrell, 1988; Hatch, 2006; Hernes, 2010; Reed, 1997, p. 26). Post-modernists also conceptualize organizations as social networks that are temporary products of micro-processes of social ordering and discursive patterning (Law, 1994). Post-modernist researchers are generally suspicious of structures and argue that we can ascertain human activity and discourse that constitute these structures but not the structures themselves (Deetz, 1996; Hatch, 2006; Reed, 1997).

Post-modernists view organizations as containing formal rationality and informality that break the formal rules. Overall, post-modernists focus on the “unconscious, hidden, and nomadic forces shaping rational choice and deliberate planned action” in organizations (Chia, 2003a, p. 108; Gabriel, 1995). Instead of social structures, post-modernists discuss social texture, which is a tacit “connectedness in action” that “opens out in a centrifugal way and can only be experienced as an activity of creative production, in which the agent/‘reader’ is caught up as an active element in the ongoing, unfinished movement of the text” (Cooper & Fox, 1990, p. 576, 578; Gherardi, 2006).

The post-modernist view of OL focuses on knowledge, power, language, and discourse so that individual learning acquires its collective meaning through text, intertextuality, and storytelling (Boje, 1994; Levina & Orlikowski, 2009). The post-modernist research also highlights generative learning “through intuition, attention, dialogue, and inquiry” and is in opposition to the functionalist vision of learning as an adaptive process based on logical deduction and induction (Chiva et al., 2010, p. 116). Dialogue as a generative OL process is a “stream of meaning flowing among and through us and between us” so that individual perspective is elevated to the collective level (Chiva et al., 2010, p. 124; Oswick, Anthony, Keenoy, Mangham, & Grant, 2000).

Among weaknesses of the post-modernist paradigm is lack of empirical studies. The post-modernist scholarship “is highly theoretical and generalized and remains quite esoteric” (Alvesson & Deetz, 1996, p. 212), potentially because post-modernists view themselves more as philosophers than data gatherers (Hardy & Clegg, 1997). Post-modernist scholars deconstruct structures, organizational routines, and social contexts that shape organizational realities and behaviors but ignore social action and its transformative potential (Reed, 1997). Finally, post-modernist focus on language and textuality does not account for richness of human practices where the language is situated.

The Functionalist–Constructionist Transition Zone

There are several points of convergence between the functionalist and constructionist paradigms in conceptualizing OL and connections between its levels of analysis. Hatch (1993) crosses the borders of functionalism and constructionism by approaching human cognition as inseparable from social processes and linking objectivist and subjectivist perspectives of culture. She builds on Schein’s (1985/2010) functionalist model of culture as structured around assumptions, values, and artifacts but argues that the matter of “the objectivity or subjectivity of organizational culture itself is undecidable” (p. 683). To connect objectivism and subjectivism, she adds cultural symbols to Schein’s model, although unlike Schein, she views culture as dynamic and containing multiple layers of meaning, thus increasing the explanatory power of his theory.

The functionalist–constructionist transition zone contains research that generates a more nuanced understanding of organizational cognitive and behavioral learning as situated in social interactions. Kupers (2008) contends that OL occurs simultaneously at the individual and organizational levels of analysis and includes subjective feelings, objective cognitions and behaviors, intersubjective cultural artifacts, and interobjective systems. On a similar note, Nicolini and Meznar (1995) note that research into cognitive OL processes is not sufficient and needs to account for the social interactions in OL. According to them, OL transforms tacit knowledge into explicit knowledge through organizational reflection that is controlled and facilitated by organization leadership. Elkjaer (2004) integrates the functionalist view of OL as knowledge acquisition and the constructionist view of OL as participation in social processes. For her, the former underestimates the power of social processes and contexts, and the latter ignores the learning content.

The functionalist–constructionist transition zone articulates the weakness of the functionalist conceptualization of connections between multiple OL levels of analysis. For example, Glynn et al. (1994) expand the functionalist OL model suggested by March and Olsen (1976) by adding a constructionist element to learning. Glynn et al. suggest that OL starts with diffusion of individual learning within organizations via routines and organizational memory, leading to organizational action. The collective learning can also start with individuals who develop shared meanings through their direct and indirect interaction, thus shaping organizational context.

The OL framework by Crossan, Lane, and White (1999; see also Crossan, Maurer, & White, 2011) situated on the border of the functionalist and constructionist paradigms suggests four sequential social and psychological learning processes at the heart of OL: (a) intuiting (individual preconscious pattern recognition in the form of metaphors and developing a rich interpretive schemata), (b) interpreting (explanation of insights to oneself or to others in the form of a cognitive map and developing shared interpretations and enacting them), (c) integrating (“developing shared understanding among individuals” and taking “coordinated action through mutual adjustment”), and (d) institutionalizing (incorporating actions into organizational routines) (p. 525). These processes link OL levels of analysis in a bi-directional way so that ideas flow from individuals to the organization and feed back from the organization to individuals.

Lehesvirta (2004) uses Crossan et al.’s (1999) framework, as well as insights from the functionalist researchers such as Huber (1991), and shows that the four OL processes identified by Crossan et al. (1999) occur simultaneously rather than sequentially and are affected by dynamics within the organization. For example, individuals may not necessarily share their interpretations so that no integration and institutionalization of learning occurs and individual learning activities and organizational goals become disconnected. Mohrman, Tenkasi, and Mohrman (2003) show that learning at the individual level is connected to learning at the collective level through multiple social networks that form within organization’s units, between them, and organization-wide. OL occurs at all levels of analysis, and social interaction within the networks is crucial for people to form common schemata as a precondition for changing their behavior.

The Constructionist–Post-Modernist Transition Zone

Constructionism is based on the view of organizations as processes and not static entities (organizing instead of organization, in terms of Weick, 1979), so its ontological assumptions are close to the post-modernist view of organizations as emergent (Chia, 1995). Both paradigms acknowledge the centrality of language in organizations and OL, and their mutual complementarity is illustrated by studies that discuss organizational discourses defined as “an institutionalized use of language” (Gherardi & Nicolini, 2002, p. 429). Gherardi and Nicolini (2002) demonstrate that OL involves learning a practice through participation in a community that constructs its own discourse. It is through discursive practices that individuals compare perspectives, negotiate meanings, and coordinate action. As suggested by Barrett, Thomas, and Hocevar (1995), discourses and discursive practices are maintained through official documents and policies that shape employees’ activities and coordinate their actions.

Another point of convergence between these two paradigms is looking at organizations as networks. Spinuzzi (2008) connects Engeström’s (1999, 2001) constructionist activity theory and post-modernist actor-network theory. Spinuzzi shows that organizations are historically and socially evolved collections of activity networks, which he defines as assemblages of activities that are heterogeneous, multiply linked, transformative, and having simple interface for dealing with large complex issues. His study demonstrates that networks learn through their members, who receive formal training and contextualize it through informal workplace communication and workplace learning in the form of shadowing and mentoring.

Gherardi (2006) brings together constructionist activity theory, situated learning theory, cultural perspective on OL, and post-modernist actor-network theory to develop a practice-based framework of OL. She contends that OL includes participation in practices, defined as “our production of the world and the result of this production” and as a system of activities where “knowing is not separate from doing” (Gherardi, 2006, p. 215). For her, engaging in work practices means participating in organizational discourses, acquiring knowledge, changing or transmitting it, and creating and recreating the organization.

Among differences between the constructionist and post-modernist paradigms, the studies that cross their borders point out that the former ignores the role of power in organizations and in OL. For example, Marshall and Rollinson (2004) highlight the limitations of the constructionist paradigm because it discounts the power balance within organizations. As an alternative, they suggest that the production of power and knowledge within organizations is closely connected. . Fox (2000) analyzes COPs and relies on the post-modernist actor-network theory to show that they have their own power dynamics and contain nests of micro-practices. In his view, the actor-network theory provides important insights into OL by including non-human agents (such as technologies) into workplace practices as a factor affecting network/community interaction and collective learning.

The Definition of Organization

Organizations are multi-layered social worlds, or entities that include human and non-human actors and have shared perspectives, activities, and discourses. They have the duality of stability and change that is manifested in formal structures that are enacted, negotiated, and redefined by organization members. The view of organizations as social worlds emerged in the positivist sociology (Strauss, 1978) and was further developed by constructionist (Elkjaer & Huysman, 2008) and post-modernist authors (Clarke, 2005). According to Clarke (2005), a social world is any group of people engaged in shared activities that have shared perspectives and discourses. Organizations as social worlds have structures but at the same time are fluid in terms of their membership and interaction of their members. Organizations have their geographies, technologies, identities, patterns of collective action, sub-worlds, texture, ideologies, commitments, and boundaries (Clarke, 2005; Cooper & Fox, 1990; Elkjaer & Huysman, 2008; Strauss, 1978).

Organizations as social worlds are shaped by the interaction of their members, as well as structures that—as paradigms agree—include division of labor, rules and procedures, and mechanisms of coordinating individual actions (Alvesson & Deetz, 1996; Chia, 1995; Döös & Wilhelmson, 2011; Hatch, 2006; March, 1991; Wenger, 1998). At the same time, the paradigms differ in their view of information processing in organizations. For functionalists, information comes into organizations from external environment as a result of intentional information gathering activities. Organizations process information to reduce equivocality, make decisions, and adjust to environment (Daft & Huber, 1987). Information becomes knowledge when organizations interpret it and store it in their memory. For constructionists, the source of information is both extra- and intra-organizational, and its processing involves pre-cognitive sense-making (Weick, 1995) and the interaction of tacit and explicit knowledge of organization members (Nonaka, 1994). From the constructionist perspective, new information becomes knowledge when it is socially negotiated, interpreted, and shared within the organization. For post-modernists, new information becomes knowledge when it becomes a part of organizational discourse (Hatch, 2006; Y.-J. Lee & Roth, 2007), and for critical theorists, information becomes knowledge when it reflects power relations within organizations (Coopey & Burgoyne, 2000).

The Definition of OL

OL is a social process of individuals participating in collective situated practices and discourses that reproduce and simultaneously expand organizational knowledge. The view of OL that emerged in this article brings together its cognitive aspects, social nature, and embeddedness in language (Gherardi, 1999). All paradigms agree that OL involves the creation of new knowledge. For functionalist researchers, OL means creating new knowledge and changes in organizational cognitions and behaviors that result in adapting to the environment (Kim, 2004; Walsh & Ungson, 1991). On the constructionist side, Nonaka (1994) views OL as triggered by the enlargement of individual knowledge through knoweldge conversion and forming shared perspectives during the socialization and externalization of knowledge. In his opinion, organizations learn by amplifying individual knowledge through the spiral of knowledge creation that involves the interaction of tacit and explicit knowledge at three levels of analysis (individuals, teams, and organizations). For post-modernists, knowledge is embedded in power relations and is incorporated in collectively created discourses (Foucault, 1984; Hatch, 2006). Similar to post-modernists, critical theorists consider knowledge a part of power relations that are reflected in discourses proliferating oppression and domination (Grant et al., 2009).

OL paradigms converge on conceptualizing knowledge as produced, reproduced, and expanded in collective practices. This conceptualization goes back to the critical paradigm, specifically Marxian heritage of viewing practices as shaped by historical and social environment (as summarized by Gherardi, 2000). From a functionalist perspective, practices include explicit work processes organized around work roles and re-structuring explicit organizational knowledge. For constructionists, practices are work-related coordinated social activities of groups and individuals that are situated in specific contexts (Cook & Brown, 1999; Gherardi, 2000). In practices, knowing is not separate from doing, and constructionists bring to the forefront the tacit knowledge created in practices, viewing OL is a social rather than cognitive process (Gherardi, 1999, 2000).

In terms of producing, reproducing, and expanding knowledge in practices, functionalists view this process as the duality of exploitation of existing knowledge and maintaining existing knowledge structures, and exploration that includes gaining new knowledge (March, 1991). Constructionists focus on the social processes of “practising the practice” and engaging in practices as a way for individuals to acquire, produce, reproduce, and expand organizational knowledge (Gherardi, 2008; Antonacopoulou, 2008). In their view of language as embedded in practices, constructionists are close to post-modernists, and consider engaging in practices as “taking part in a professional language games” and professional discursive practices and trying the practice out (Antonacopoulou, 2008; Gherardi, 2000, p. 216). For critical theorists, participation in practices means not only acquiring knowledge but also contributing to social change (Meyerson & Kolb, 2000).

Ontology of OL

OL is ontologically grounded in the dialectics of individuals and organizations that links its multiple levels of analysis. All paradigms acknowledge that the core of OL is the interaction of its multiple levels of analysis, though disagree on what aspect of this interaction they emphasize. For example, functionalists accentuate organizational structures and pay close attention to collective cognitive processes and collective memory that represents “supraindividual collectivity” (Walsh & Ungson, 1991, p. 68). However, functionalist researchers acknowledge the key role of individuals in collecting, interpreting, sharing, storing, and retrieving information (Fiol & Lyles, 1985; Hedberg, 1981; Walsh & Ungson, 1991). Overall, functionalists view organizations as shaping the learning of their members but at the same time depending on their learning capacities (Hedberg, 1981).

Functionalists and constructionists agree that when interacting with each other, individuals share their cognitive maps to develop mutual understanding and collective cognitive maps (Kim, 2004; Walsh & Ungson, 1991; Weick, 1979). Close to post-modernists, constructionists emphasize social interactions as a foundation of OL (Brown & Duguid, 1991; Elkjaer, 2005). However, post-modernists view this interaction as textual and expressed in discourses. Similar to the critical paradigm, post-modernists conceptualize OL as dialectics of individual and organization, and it is through this dialectic that multiple levels of analysis of OL are connected (Y.-J. Lee & Roth, 2007). Post-modernist research based on complexity theory contends that OL “emerges at the possible connections and inter-dependencies explored across multiple levels of analysis” (Antonacopoulou & Chiva, 2007, p. 289).

Epistemology of OL

OL is epistemologically situated in specific intra- and inter-organizational contexts. All paradigms recognize the importance of context in OL, though attribute different meaning to it. Functionalists emphasize external environment that shapes OL and leads to the adaptation to this environment (Daft & Weick, 1984; Hatch, 2006; Walsh & Ungson, 1991). Other contextual factors noted by functionalist scholarship include culture that influences cognitions and behaviors of organization members, and strategy that delimits organizational decision-making, structure, and processes (Fiol & Lyles, 1985; Nevis, DiBella, & Gould, 1995).

Constructionists turn their lens inside organizations and examine specific activities, situations, and cultures where OL is situated (Lave & Wenger, 1991). The concept of situated learning was first expressed in a systematic way by Brown et al. (1989), who suggest a learning epistemology where knowledge is coded by and connected to the activity and environment that produce it. Thus, conceptual representations and theoretical knowledge are secondary to activity and perception. Brown and Duguid (1991) argue that knowledge is informal, socially constructed, orally transmitted, and situated in specific work contexts and communities of practitioners. Therefore, learning is a process of enculturation, or adopting a set of behaviors dominant within a specific community so that learners not only learn about the practice but also become practitioners. Post-modernists look down on the general and universal, so OL includes continuously emergent micro-practices that have no meaning outside the context of human activity (Cooper & Fox, 1990; Gabriel, 1995). Although critical theorists pay attention to large social structures of domination, they too reflect on micro-practices of communicative action as a context where individuals develop shared meanings and understanding (Habermas, 1984, 1987).