Resurrecting organization without renouncing society: A response to Ahrne, Brunsson and Seidl

In a recent article in this journal, Ahrne, Brunsson, and Seidl (2016) suggest a definition of organization as a ‘decided social order’ composed of five elements (membership, rules, hierarchies, monitoring, and sanctions) which rest on decisions. ‘Partial organization’ uses only one or a few of these decidable elements while ‘complete organization’ uses them all. Such decided orders may also occur outside formal organizations, as the authors observe. Although we appreciate the idea of improving our understanding of organization(s) in modern society, we believe that Ahrne, Brunsson, and Seidl's suggestion jeopardizes the concept of organization by blurring its specific meaning. As the authors already draw on the work of Niklas Luhmann, we propose taking this exploration a step further and the potential of systems theory more seriously. Organizational analysis would then be able to retain a distinctive notion of formal organization on the one hand while benefiting from an encompassing theory of modern society on the other. With this extended conceptual framework, we would expect to gain a deeper understanding of how organizations implement and shape different societal realms as well as mediate between their particular logics, and, not least, how they are related to non-organizational social forms (e.g. families). © 2017 Elsevier Ltd. All rights reserved.


Introduction
In a recent article published in this journal, Ahrne, Brunsson and Seidl (2016) advanced a programmatic claim to "extend the notion of organization" (p. 93) beyond the current understanding of formal organizations by putting decisions back at the center of organization research. Their claim is complex and extends beyond a rehashing of historical debates and competition between theories of formal organization and decision-making approaches. First, their claim relates formal and decisional aspects of organizations; second, it reflects recent developments (in theory as well as in applying empirical evidence); and third, it traverses the traditional range of organization studies.
Based on recent developments in our field, which have empirically and/or theoretically tended to dissociate from organization studies in the narrow sense, the authors underscore that organization represents a highly important and very specific phenomenon in modern society. Thus, organization is "not a mere reflection of a more general social order that can be adequately understood by concepts and theories describing society in general" (Ahrne et al., 2016, p. 93). In other words, the authors suggest that the concept of organization can be maintained and strengthened by distinguishing it clearly from other concepts and phenomena in modern society. By defining organizations as decision-based social orders, the authors explicitly draw on the latest works of the sociologist Niklas Luhmann (2000bLuhmann ( , 2003, who described organizations as constantly making and reproducing decisions while also deciding their own structures. With respect to the elements of formal organizational structure, Ahrne et al. depart from Luhmann by distinguishing five basic elements of organization: membership, rules, monitoring, sanctions, and hierarchy (Ahrne & Brunsson, 2011, p. 86;Ahrne et al., 2016, p. 95).
Building on this framework, the authors go beyond the classic concept of formal organization. By making use of the five formal elements, they suggest that different degrees of "organizationality" (Ahrne et al., 2016, p. 98) can be realized on a continuum that ranges from complete to partial organization. Although a complete organization would incorporate all five elements, the authors see the possibility of so-called "partial organization" (p. 95) that only uses a few or even one of these elements. Compared with the classic ideas of organization, a new variety of organizational forms appears on the agenda, including forms that had not been previously considered in organization studies but rather were left to other specialists and addressed by the theory of society, e.g., in the case of families. However, the authors not only advocate for a specific theory of organization that will lead to a better understanding of the phenomenon, but they explicitly claim that expanding the concept of organization will provide "insights […] necessary for understanding modern society" (Ahrne et al., 2016, p. 93).
Ahrne et al. touch on a sore spot. As US scholars stress, organization studies are "facing a kind of existential crisis" (Barley, 2016, p. 3; see also similarly Davis, 2014;Gorman, 2014) because organizations seem to morph furiously into new forms, and "old theories are no longer as relevant as they once were" (p. 2). However, Ahrne et al. make clear that European organization studies have already spawned means for understanding such new forms of organization.
We concur that a criterion is required so that organizations can be understood as a specific form of social order, and we believe that this criterion can be found in the peculiarity of organizational decision-making. Further, we also agree that organizational theories already have the means to understand "organizational" phenomena, which may initially appear to be outside the scope of the traditional conceptualization of organizations.
However, we believe that the contribution of Ahrne et al. presents two major difficulties. The first difficulty concerns the current idea of partial organization. As the authors make clear, partial organization can represent (1) a formal organization that only uses some of the possible five structural elements, and (2) certain structural elements that exist outside of formal organizations, such as standards. These forms of organization are all considered "decided orders." Therefore, we argue that although Ahrne et al. want to sharpen the distinctiveness of the concept of organization, their proposal threatens to blur the lines even more and to result in the loss of a distinctive concept of formal organization.
Our second concern is related to Ahrne et al.'s claim that their proposal can offer insights for understanding modern society. As sociologists, we assert that the authors' reflections on society fall short because they fail to connect their theory of organization with a key insight of sociological theory: the concept that modern society is differentiated into distinctive realms of social reality defined by others as "societal sectors" (Scott & Meyer, 1983), "social fields" (Bourdieu, 1988), "value spheres" (Weber, 1946), "social worlds" (Guston, 2001) or "subsystems of society" (Luhmann, 1994). Ahrne et al. avoid using a similar societal theory and limit themselves to general remarks on the relevance of "decided orders" as distinctive phenomena used to understand society. Therefore, Ahrne et al. cannot provide an explanation of how organizations and their decision-based processes are related to other forms of social order. Although Ahrne et al. (2016, p. 94) criticize Neo-Institutionalism for not presenting an elaborated theory of organizations when trying to understand the relationship between organizations and society, we are concerned that an attempt to understand this relationship with a theory of organization but without a theory of modern society introduces the same flaw only in the other direction.
We believe that both problems stem from the fact that the authors do not take their underlying Luhmannian framework seriously enough. The sociological theory of Niklas Luhmann offers a far more radical perspective of how we can understand organizations as systems of decisions and an elaborated understanding of modern society. By thoroughly acknowledging Luhmann's concepts, one can more clearly show how organizations play a major role in modern society. Our response is structured as follows.
First, we introduce a radical concept of formal organization by returning to the Luhmannian definition of an organization as interconnected decision processes. In our view, it is not simply the capability to make a decision but the process of interconnected decision making that constitutes the difference between formal organizations and other social orders. Second, we suggest that the role of organization(s) in our society can be understood only if we are able to sociologically describe modes of building order, which differs from organizational order. We argue that the concept of institution is too weak and return to the Luhmannian concept of functional differentiation to describe the building of order in different societal realms. In the last step, we elaborate on how a radicalized version of Ahrne et al.'s proposal can be used for empirical research on the role of organizations in society. Thus, our contribution supports Ahrne et al.'s (2016) effort to place "organization studies at the heart of social sciences" (p. 99).

Organizations and beyond
Ahrne et al. as well as Ahrne and Brunsson (2011) built their analysis of organization on the crucial idea that decisions are the central feature of organization, and they define organization (i.e., in the singular) as "decided social order." The idea of Ahrne et al. (2016) is that the "concept of organization as 'decided order' allows for the transfer of the term to other domains outside formal organization, while simultaneously preserving its distinctiveness" (p. 95).
Based on the authors' new conceptualization of the term, "organization" can be used to describe several decided social orders that encompass decisions on at least one of the following elements: rules, hierarchies, membership, monitoring, and sanctions. In this respect, the authors see a continuum of degrees of "organizationality" (Ahrne et al., 2016, p. 98). If a decided order uses all five structural elements, it is denoted complete organization. If the order only uses certain elements, it is called partial organization. However, Ahrne et al. also maintain a notion of formal organization and, interestingly, suggest that formal organizations can be complete or partial: "This conception of organization opens up the possibility that organization may come in parts, such that only one or a few elements of organization are actually used within or outside a formal organization." (Ahrne et al., 2016, p. 95) From our perspective, this duality leads to considerable confusion because a formal organization can be a partial organization while partial organization can also be a type of organization that is outside of a formal organization. What distinguishes formal organizations from other decided orders outside of them? Unfortunately, Ahrne et al. do not present a clear answer to this question.
To overcome this ambiguity, we propose applying the concept of organization elaborated by Niklas Luhmann, which Ahrne et al. build on but do not entirely adopt. Luhmann (2000bLuhmann ( , 2003 offers a radical understanding of formal organizations (i.e., in the plural) as systems of decisions (see also Seidl & Becker, 2006;Seidl, 2005).
Consequently, organizations are not only decided orders but are also constituted by ongoing decision-making processes: "Organization is not about single decision makers and their individual attempts but about decision making as a process, about linking decisions" (Tacke, 2014, p. 10). Thus, organizations build an interconnected network of ongoing "processes of decisions, whereby one decision calls forth ensuing decisions, resulting in a selfreproducing stream of decision" (Ahrne et al., 2016, p. 95), whereas other elements are excluded. This aspect of organizations leads to the insight that organizations gain unique decision-making capabilities and complexity that are lacking in other types of social systems.
Moreover, this concept emphasizes a dynamic view of organizations and directs the analytical gaze to the question of how current decisions link to former decisions. By acknowledging this connection, the analysis of how decisions connect to each other becomes central: in formal organizations, the link between decisions is guaranteed above all by formal structures, which are themselves a matter of decision and function as specific premises for subsequent decisions (Seidl & Becker, 2006;Seidl, 2005). For example, organizational decision-making capabilities make it possible to decide on hierarchies or rules that do not have to be constantly (re-)negotiated while maintaining the possibility of changing these rules by making new decisions. Therefore, we acknowledge that the elements described by the concept of partial organization (membership, rules, hierarchies, monitoring, sanctions) are helpful for making sense of the different structural elements that facilitate the link with subsequent decisions in a specific organization. We also acknowledge Ahrne et al.'s view that a formal organization can also decide not to make decisions regarding some of these elements. Therefore, different types of organizations might, for example, differ on how membership is defined or not defined.
This concept can be used to disband the old concept of organization, in which enterprises and public administrations are prototypical forms and that considers all five structural elements listed by Ahrne et al. as constitutive. Applying the Luhmannian concept of organization, we can, for example, also analyze organizations, such as parties or associations, that are essential in a democratic society but do not have hierarchies similar to those of enterprises and use limited sanctioning instruments. Additionally, meta-organizations, which are described by Ahrne et al. (2016, p. 96), are formal organizations that partially use decided structures. Moreover, this concept can be beneficial for analyzing new forms of organizations, such as virtual organizations, project organizations, and hacker collectives, in which aspects such as membership are contested or unclear (see Dobusch & Schoeneborn, 2015;Schrey€ ogg & Sydow, 2010).
Based on these reflections and in accordance with the tradition of organizational studies, we want to reemphasize the necessity of developing a specific understanding of formal organization as a modern phenomenon with distinctive features. By understanding formal organizations as operatively closed systems of ongoing interconnected decision processes, we assert that the distinction between "complete" and "partial" organization is only valid at the structural level and not the constitutional level of organizations. Moreover, although we agree that the five listed elements are important structural elements of formal organizations, an explanation of this selection of elements is necessary (e.g., why are formal goals or department structures not considered?) e not least to justify the idea of "completeness." Consequently, we suggest renouncing the distinction between partial and complete organization. While this distinction defines organization by listing structural features and leaves open the implications of "complete organization", we favor considering organizations as decision systems, which defines them by identifying their constitutive operating principle (Luhmann, 2003;Seidl & Becker, 2006).
Moreover, we believe that another distinction developed by Ahrne et al. could be placed at the forefront of organizational analysis: the distinction between institutional orders and decided orders. We will outline how this distinction can be placed in the forefront in the following section.

Organizations and society
Thus far, our discussion has focused on the concept of organization, but the discussion should extend beyond organizational borders. Organizations make decisions on their own structures but also make decisions that affect society more broadly (e.g., when they decide on standards). Therefore, organizations establish decided orders. Moreover, decisions also occur outside of formal organizations (e.g., in families), and these decisions can create new decided orders. Decided orders can differ regarding the structural elements they include: membership, rules, monitoring, sanctions, or hierarchy (Ahrne et al., 2016).
In this respect, we agree with Ahrne et al. that there is a lot more decided order in the world than it is usually presumed. However, we assert that to understand these phenomena, we need not only a theory of organization but also a theory of what an organization is not, which means that we need a theory of society. Concerning the analysis of society, Ahrne et al. (2016) refer to institutions and distinguish between "decided orders" and "institutional orders"; therefore, they make a connection to the framework of neoinstitutionalism. Institutional orders are understood as emergent orders that are not a result of decisions. However, we suggest that a consideration of the concept of institutions does not lead to a sufficient understanding of societal order. Indeed, neo-institutionalists have recently started to categorize institutions into coherent logics on a societal level (Friedland & Alford, 1991;Thornton & Ocasio, 2008); however, they are not able to define the constitutive features of these logics in a convincing way (Besio & Meyer, 2015). In contrast, the theory of society by Luhmann offers adequate concepts to describe modern heterogeneity.

Functional differentiation
According to Luhmann, modern society is primarily functionally differentiated into distinctive realms: the subsystems of society (Luhmann, 1977(Luhmann, , 1987(Luhmann, , 1988(Luhmann, , 1994; for a brief introduction, see also Seidl, 2005;Seidl & Becker, 2006;Turner, 2013). These subsystems include politics, economics, justice, medicine, science, education, mass media, art, religion, sports, the sphere of intimate relationships and nuclear families (Luhmann, 2012(Luhmann, , 2013 and possibly more. Each societal subsystem has its specific logic that is defined by the orientation to a binary code that serves as a guiding difference for its operations (which for social systems are communications). For example, the economic system orients its operations along the distinction of having/not having (Luhmann, 1989, p. 52); thus, it includes all communications concerned with the distribution of scarce resources. Therefore, societal subsystems can be considered systems because they are constituted by interconnected networks of communication, which in each case are oriented to only one system-specific code. These systems are not organizations. Indeed, the subsystems gain their modern complexity from the fact that they are not reducible to just one organization or one hierarchy (Luhmann, 2013, p. 150).
To show that societal subsystems are specific forms of social order and cannot be reduced to decision-making processes, we examine the case of families as an example. On a societal level, the sphere of nuclear families and intimate relationships can be considered a societal subsystem (Luhmann, 2012(Luhmann, , 2013 because of the logic of love, which orients communication in this societal sphere (Morgner, 2014). We choose this example because Ahrne et al. provocatively consider families as organization (Ahrne et al., 2016, p. 98).
As Ahrne (2015) outlined in another paper, a greater degree of decision-making occurs in families and romantic relationships than is commonly assumed: "how to arrange for their marriage, what they can expect from each other, whether they are going to have children" (Ahrne et al., 2016, p. 98). Indeed, modern couples can decide on such issues relatively independently (Ahrne, 2015). In pre-modern families, this was not possible because families were not only families but also economic production collectives. Their structures were determined by the necessities of production and reproduction as well as by estatist norms.
Modern families are no longer burdened by these institutionalized orders. Instead, love and cohesion are crucial and allow for the development of families as distinctive social systems that can be distinguished from economic entities (Luhmann, 2012(Luhmann, , 2013. However, does family become an "organizational form" as proposed by Ahrne et al. (2016, p. 98)? Does the presence of elements of decided order in families make the family an organization? We tend to disagree. Indeed, although decisions are made within families, families are not based on or constructed by decisions. Instead, they are based on cohesion and the societal logic of love. Family members do not take formalized roles but rather communicate with the expectation that they will be fully acknowledged as a person with a distinct and complete personality. Families are constituted and stabilized based on love, not their decisions, and especially love implies that one has not made a decision (Seidl, 2005, p. 39), but just happens to love the other and vice versa. Although it might be true that membership in families is welldefined (Ahrne et al., 2016, p. 98), this membership usually can hardly be considered to be a matter of decisions: one does not decide to love the other, and a child does not decide to become a child nor can it relinquish its family membership. Instead, a family is a collective based on regular interactions through which the relationship among the family members grows stronger and undecided norms emerge. Therefore, although families must make numerous decisions in everyday life and this could be a fruitful research topic, we do not consider families to represent an entity that becomes a decision-based system.

Close relationship between organizations and society
The close relationship between organizations and society can be understood by considering it from a historical-evolutionary perspective. Organizations are a product of modern society, and only a functionally differentiated society requires organizations. Therefore, modern society would not exist in its current form without organizations, and organizations are only possible and necessary in modern society (Schoeneborn et al., 2014). This relationship implies a complementary thesis: organizations did not occur in older societies.
"Traditional social places" in these societies primarily included households and guilds. Even if guilds are to a certain extent the precursors of modern organizations, they are completely different in a number of respects. First, members were included "totally", that is, without distinguishing between personal and role behavior. People were born, lived, and died in the same place, and time was the "natural" time of aging, not the time of work contracts. Work did not require individual motivation, and guilds did not need membership selection criteria. Second, the internal structures of the guilds mirrored the general hierarchical societal structure; for example, officers and bishops as well as the guilds' public representatives were members of the aristocracy (see Schwinek€ oper, 1985).
Let us take a closer look at the example of politics. Historically, politics were not organized. The medieval courts communicated on the basis of certain criteria, such as virtue, friendship/hostility, rivalry, secret knowledge, etc., and interactions at the court around the king and the network of households outside the court were both necessary and sufficient. Consequently, courts were based on institutional order and not on decisions. This began to change in the 17th century (in England perhaps earlier) and politics began to be organized on the basis of decisions, such as selective behavior, which always has alternatives and must prove itself in the public sphere. Interests collate into organized parties, and administrations require officials and procedures, which are no longer based on the quality of the nobility but require training and decisional premises (Zaret, 2000). Finally, modern politics requires organizations (i.e., central governments, political parties, lobbies, etc.), but it is not limited to these organizations. Politics is only an example, and similar conclusions can be extended to the economy, science, and most other societal subsystems.

Organizational re-specification of the logics of societal subsystems
Consequently, organizations play a crucial role in the Luhmannian concept of society. The "emergence of organizations as a new type of social system was a precondition for the modern, functionally differentiated society" (Schoeneborn et al., 2014, p. 297). In modern society, one cannot rely on traditional norms or authorities that are taken for granted; rather, one must decide on almost every issue. In addition, systems such as politics, justice, or economics depend on decisions to operate, and these decisions can only be delivered in the necessary quantity and quality by organizations. Although the guiding codes of societal subsystems are necessarily diffuse, organizations are able to concretize and implement these codes; thus, organizations can re-specify the logics of the societal subsystems so that decisions can be made (Drepper, 2005, p. 178;von Groddeck, 2013, p. 193). Because of their ability to make decisions and decide on structures, i.e., the creation of decided orders, organizations facilitate the management of modern society's complexity by making subsystems' logics decidable. Consequently, organizations have an important role in applying the codes as well as in shaping the code-related norms of societal subsystems. For example, the economic system relies on organizations to distribute goods and money; the legal system relies on organizations in the form of courts to make decisions on legal or illegal acts; and the political system needs organizations to decide on party leaders and candidates for president and to produce collectively binding decisions.

Organizing for coupling in modern society
Another aspect we would like to highlight is the importance of organizations for mediating the relationships among different subsystems (see Drepper, 2005). Because the operations of societal subsystems are specific and autonomous, they cannot directly connect to each other. Consequently, the question arise: How can we understand their relationship? For example, how can we explain that political decision-making affects education and how do scientific findings influence economic development? The Luhmannian theory states that organizations play a crucial role in connecting different logics, which occurs via the relationships between organizations and within organizations.
First, organizations are capable of communicating with each other (Luhmann, 2013, p. 145), whereas societal subsystems cannot communicate with each other. For example, in the case of science, it is difficult to imagine science as an actor capable of communicating with other systems. However, a number of scientific organizations, such as universities, research institutes or scientific associations, can communicate with political or economic organizations to negotiate funding.
Second, organizations are "multi-referential" (Stichweh, 2015, p. 29), meaning that in their decision-making, they can take different logics into consideration. Moreover, organizations can manage economic, juridical, scientific, educational and other logics simultaneously (Besio & Meyer, 2015;von Groddeck, 2011;Kette & Tacke, 2015). For example, enterprises manage financial matters as well as research programs, sports teams, legal departments, etc. Churches implement religious practices but also require revenue; therefore, they make investments. Organizations are entities that orchestrate several rationalities through their activities (Andersen, 2003) by fine-tuning and channeling their mutual influence. The Luhmannian perspective acknowledges that organizations are capable of deciding how they incorporate and couple differing logics (see Luhmann, 2013, p. 151), and in recent years, fruitful discussions have taken place on this topic in organizational studies (e.g., Andersen, 2003;Besio & Meyer, 2015;von Groddeck, 2011;Kette & Tacke, 2015;Roth, 2014).

Impact on society
Organizations are shaped by society and in turn shape society. Organizations reproduce the operations of societal subsystems, and in doing so, they constantly alter the social orders of societal subsystems through their decisions. Every simple and ongoing strategy for re-specifying external requirements can have consequences beyond organizational borders. Once an organizational solution is applied, it may be copied by other organizations and become the object of public debate, thereby unleashing innovation processes, among other effects.
Moreover, organizations are powerful actors that can also impact society with active strategies, such as lobbying. Organizations also affect society by making decisions on rules for others. In the words of Ahrne et al., organizations create and alter decided orders as well as replace institutional orders with decided orders. Therefore, "standards" as described by Ahrne et al. (2016, p. 96) represent an important example of organizational power and are a consequence of organizational decision-making processes. Once enacted, these standards can function as decided orders for several organizations and actors in different fields. Therefore, standards are decision premises as well as outputs of organizational decision making; however, they are not themselves organizations.
Examples of organizations developing standards are easily identifiable in several fields, such as national or international politics, education, economy, social work, etc. For instance, the issue of publication ethics in the societal subsystem of science is an interesting example because what was once an institutional social order has now become mainly shaped by an organization. The Committee on Publication Ethics (COPE), a formal organization founded in 1997, has developed codes of conduct and best-practice guidelines on publication ethics that are currently applied by many scientific publishers and journals. The implications are substantial because this development means that ethical standards for scientific publication no longer just happen to be there. Those are now decided and can be decided otherwise in the future.

Re-visiting the research agenda
A more comprehensive integration of Luhmann's theory of modern society with the ideas of Ahrne et al. leads to the following questions concerning the relationship between organizations and society: 4.1. How do organizations re-specify the codes of the societal subsystems?
One can observe that different subsystems have different dependencies on organizations. Economics and politics strongly rely on organizations, whereas families and love-based relationships as a societal sphere are not based on organizations. Moreover, different types of organizations develop within different subsystems. For example, although public administrations are typically framed by formal rules, universities are more similar to anarchies (see Cohen & March, 1974). Additionally, different types of organizations may be present within one subsystem. For example, the logic of the societal subsystem of mass media is concretized in different ways; thus, in addition to traditional editorial offices, new organizational forms are also observed, such as partially structured "latent organizations" (Starkey, Barnatt, & Tempest, 2000). In these organizations, individuals and teams coordinate in a temporary project-based manner. In this respect, Ahrne et al.'s heuristic of the partially structured formal organization may provide an opportunity to analyze different forms of re-specification of societal subsystems and their consequences.

How do organizations mediate between the different logics of societal subsystems?
Recent insights show that organizations can creatively combine different logics. For example, Besio and Meyer (2015) distinguish three forms of combining differing logics within formal organizations: loose coupling, translating and interfacing. All these strategies rely on structural elements, which are decided on by organizations. In such an analysis, the five elements listed by Ahrne et al. are of relevance. For example, loose coupling may occur via the establishment of separate departments that operate according to different logics and the simultaneous coupling of these departments at selected points, such as through specific monitoring devices. Furthermore, organizations can translate one societal logic into another and then decide on rules that regulate such translation. For example, for certain "social entrepreneurs," humanitarian aid coincides with helping a local economy, and these organizations set rules that prescribe the continuation of humanitarian aid projects that are compatible with economic purposes. Moreover, organizations connect different societal subsystems by building interfaces. In particular, organizations can make decisions that establish specific "boundary objects" (Star & Griesemer, 1989), which are structures perceived as relevant and applicable by different subsystems. These boundary objects can be departments, positions, specific programs or single members. For example, the position of "professor" at a university is oriented towards educational as well as scientific logic.
An open question in this respect is related to the roles played by different decided orders with respect to mediating different logics. Can formal organizations use specific structural elements for the mediation of specific logics? How can mediation by decided orders outside of formal organization succeed?

How do organizations impact society?
Organizations reproduce and alter the code-related norms of societal subsystems by providing decisions that are binding both for themselves and for others. Organizations can co-shape societal structures. The forms by which organizations affect society are manifold. Organizations can make a new solution available, define standards or engage in public debates; however, the success of their interventions is never guaranteed. The implementation of their decisions beyond their borders can be complicated by misleading interpretations, insufficient resources, and incompatibility with existing routines. In this process, Ahrne et al. indicate a crucial difficulty concerning decided orders: because they are the results of choices between alternatives, they can always be questioned. Therefore, these orders are more fragile than institutional orders. This idea opens the way for research concerning the conditions in which organizational decision-making succeeds beyond the borders of the organization, i.e., how the decided orders are accepted and legitimized outside of formal organizations.

How are decisions made outside of organizations?
The described lines of research allow us to explain how formal organizations manage societal complexity while also transforming society with their decisions. Additionally, the approach of Ahrne et al. allows us to consider decided orders that are not a result of organizational decision-making. Decisions are also made outside of organizations, e.g., in small groups, families, epistemic communities, networks, etc., in contexts that are not constituted by decisions. Additionally, in the realm of societal subsystems, such as economy or science, decisions do not always originate in organizations. Using the example of families, which are not formal organizations but still make decisions, we argue that the heuristic instrument developed by Ahrne et al. can help us understand how families borrow decision and coordination forms from formal organizations, which calls into question the circumstances that lead families to make certain types of decisions as well as the consequences of these decisions.

Concluding remarks
We agree with Ahrne et al. that organization theory is under strain, which can partially be explained by the inability of organization theory to successfully analyze the relationship between organizations and society. Indeed, neo-institutionalism has put this issue at the heart of its analyses; however, by stressing that organizations primarily reflect external societal orders, a definition of an organization as a specific phenomenon becomes superfluous, which negates the usefulness of the analytical instruments of organization theory. Ahrne et al. attempted to counteract this trend by reminding us that society consists of institutionalized orders as well as decided orders.
However, in their attempt to highlight the role of decided orders, Ahrne et al. tended to overstretch the concept of organization. Therefore, the concept of formal organization was reduced to decided orders, while society became weakly defined by the distinction between decided and institutional orders. Therefore, we suggest that the sociological theory of Niklas Luhmann, which is already present in Ahrne et al.'s approach, should be more thoroughly considered. Luhmann offers a specific definition of formal organizations as complex decision systems and provides an encompassing description of modern society as functionally differentiated. Combining the Luhmannian concept of formal organization with Ahrne et al.'s approach would clarify the concept of organization. As a result, we suggest distinguishing between formal organizations as systems of interconnected decision-making and decided order as the structural elements decided on.
Furthermore, providing a sharp distinction between organizations and society makes it possible to describe a variety of relationships. In particular, one can analyze how formal organizations implement, mediate, and at the same time change different societal logics. In this context, the five structural elements described by Ahrne et al. can function as heuristics for observing the type of structural decisions organizations make in different fields, with different objectives and consequences. Moreover, we see the possibility to inquire how decisions and decided order come about outside formal organization, and how they affect society and its subsystems. Furthermore, the combination of the two perspectives may be utilized to trace replacements and transformations of structural elements, such as membership, rules, etc., in different social systems, which allows for an analysis of their historical developments. Ahrne et al. (2016) aim to resurrect organization as a central object of research to gain "fundamental insights into the workings of our world" (p. 99). Although we highlighted several major difficulties in their approach, we generally embrace their claims. Our purpose is to add to their proposal by offering a more specific concept of organization and a comprehensive and fully compatible concept of society. By doing so, we hope to resurrect organization as a distinctive concept but seek to avoid renouncing an elaborated notion of modern society. We believe that such a combination has the potential to establish "organization studies at the heart of social sciences" (Ahrne et al., 2016, p. 99).

Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.