What Fills up the Sociological Vacuum?
This study critically discusses the thesis on the sociological vacuum formulated by Stefan Nowak. The author’s aim is to refute the claim that the sociological vacuum is relevant for major social processes occurring in Poland. He presents the sociological vacuum in the context of the debate on micro and macro levels and discusses how the theory of fields and social network analysis is useful to reconcile the micro-macro divide. The book considers the uses of the sociological vacuum in explaining such phenomena as the Solidarność social movement, civil society, social capital, and democracy. In the empirical part, the author confronts the data on identifications with the data on relations and claims that the vacuum is not in the society but it in sociology.
Part I: The micro-macro problem in sociology: theoretical background
The aim of this chapter is to briefly present and summarize the main stances on the micro-macro problem in sociology. I discuss the differences between the micro-macro pairing with other – sometimes overlapping but still different – pairings important for sociological theory: individual-society and agency-structure. The proper understanding of the micro-macro debates and possible links between the micro and macro is an important context for the interpretation of social processes with the use of the concept of the sociological vacuum. As Jonathan H. Turner (2016: 123) noticed, “the failure of closing the micro-macro gap in sociological theory was often used by enemies of sociology that it is not a science.” Yet, Turner (2010c: 1) in his other works showed that problems with the micro-macro link are not just a domain of sociologists: in physics, subatomic physics is not very well integrated with astrophysics; in biology, genetics is still not integrated with population ecology; while in economics, the gap between micro- and macroeconomics is still not filled.
I believe that the story I present in this chapter is a story of relative success. Debating about connections between the micro and macro from the times of Karl Marx to the present, in my opinion, have brought conceptual improvement and theories which allow to understand different levels of social reality in a more refined way. The newest solutions to the micro-macro question will be presented in Chapters 2 and 3, therefore this chapter also serves as their introduction. For this reason, I do not include the network theories in this chapter and only discuss Pierre Bourdieu’s conceptions very briefly. Both social fields and social networks as links between micro and macro will be discussed in detail in separate chapters. This chapter owes very much to the works by Jeffrey Alexander and Bernhard Giesen (1987), Derek Layder (2006), and George Ritzer (1990), who made their own summaries of the micro-macro debate in sociology. I cite their remarks on the issue, but I do not fully agree with their insights. All four authors were not only reviewing the debate, but they were also taking part in it and promoting their own theoretical agendas. ← 19 | 20 →
The chapter is organized in the following manner: I start from presenting the differences between micro-macro pairing and the pairings of individual-society and agency-structure, and I clarify that micro-macro needs to be understood as an analytic dichotomy, even though some social scientists see it as an empirical dichotomy. Next, I discuss the six classical for sociology authors and try to distill their views on the micro-macro problem. Karl Marx, Emile Durkheim, Max Weber, Georg Simmel, George Herbert Mead, and Florian Znaniecki were all concerned with the micro-macro dichotomy, although they were presenting it in the terminology of individual-society dichotomy. After presenting the views of the classics, I discuss the 1980s debate on linking micro and macro in sociological theory. After a period of divide between the micro- and macro-extremism in the sociological theory in the 1980s, theorists discussed vigorously the possibilities of integrating the levels of analysis. The divide was tackled from the micro-perspective and the macro-perspective, but also from the dialectical “in the middle” approach. This debate, in my opinion, turned the theoretical focus towards the agency-structure pairing. In the subsequent section, I discuss the solutions to the micro-macro link, which are distinct to the majority of solutions from the 1980s debate, because they call for introducing intermediary levels of analysis (by some authors called meso-levels) linking micro and macro. The chapter finishes with concluding remarks.
To understand the issue of the micro-macro problem in sociological theory, it is necessary to start from looking at it in the context of other so-called pairings in sociological theory. Sociologists tackle pairs of concepts which cause tensions and often paradoxes. Next to the problem of micro and macro, the central sociological parings are individual-society and agency-structure. By a similar token, debates in sociological theory are organized also by other dualisms such as objectivism-subjectivism, dynamics-statics, materialism-idealism, and rationalism-empiricism which, in turn, lead to the formation of opposing camps and paradigms, thus revealing huge divides in the general understanding of what sociology is. I am not going to reconstruct all the possible dualisms in sociological theory, and in this section my focus is on understanding the basics of the micro-macro problem. I am sure that a better understanding of this problem is possible when it is presented in relation to the issue of individual-society and agency-structure. In the last paragraphs of this section I will also try to reconstruct how the notion of meso is understood in the theory. ← 20 | 21 →
1.2.1 Other pairings: individual-society and agency-structure
The individual-society pairing certainly has some overlap with the micro-macro pairing, as the individual is a phenomenon of a micro-scale, and society, as something large, obviously seems to be something of macro-scale. Yet, the individual is not the only entity placed on the micro-level, and there are also many other phenomena, which are to be regarded as occurring on the macro-level. Quite many theoretical approaches deal with macro-processes and macro-phenomena but reject the concept of society. The individual-society pairing is the oldest among the three and has been seriously considered in the works of social philosophers for a very long time before the establishment of sociology as a legitimate mode of theorizing about social processes. In the next section, where I am going to discuss how the classics of sociology perceived the micro-macro problem in the theory, I will demonstrate that they were rather tackling with the individual-society problem. What is now anachronically concluded about their stances on micro-macro is based, to a large extent, on their statements about individuals, society, and relations between the two. On this subject I will say more in Section 1.3 of this chapter. Here, however, I will focus only on the conception of the individual, and the way in which the individual and society mutually defining each other, as analyzed by Norbert Elias.
The reason I choose here Elias’ work is because it serves as an illustration of the mutual defining of individual and society, but also because the author provided a convincing history of the concept of the individual in Western culture. Although the word “individual,” which is used in English to denote an entity (usually bringing to mind a human individual), comes from Latin, it was not used in the classical Latin of Roman empire. The word appears for the first time in medieval theological treaties about discussing the indivisibility of the Holy Trinity. “Individual” originally denoted an entity, which could not be divided (Elias 1991) and it was only later that the term was also attached to the concept of a human individual. As Elias claims, the concept of the individual and our self-perception as individuals is a product of the Western culture and the processes that started in Medieval period and then flourished in Renaissance. Elias (1969; 1982) calls this a “civilizational process.” In Elias’ interpretation of Western culture, the conception of the individual dominated our thinking (Layder 2006: 153). This obsession with our own individuality is for him also present in sociological theories which focus excessively on social roles and the individual-society relation. His way of breaking with this, as he perceived, ill-thought dilemma was the figurational theory, which highlighted that society and individual are mutually creating, which means that it is not possible to think of one them without referring to the other. ← 21 | 22 →
The discussion of the pairing of individual-society requires having the concept of the individual rooted in one’s conceptual toolkit. The second concept – society – demands some explanation as well. It is commonly believed that sociology as a science was created when thinkers started to problematize society as a population of a modern state. Later, the concept was either rejected, or made more complex. The fact that society is made-up of individuals (but is not always understood as a simple aggregation of individuals), and that humans are at the same time social animals unable to function without their social environment, is a kind of quibble requiring explanation from sociologists. Another issue is the double meaning of the term “society” which does not only denote a population of a large size, but also a company, an association, or a club, which is yet another reminder of the fact that the existence of an isolated individual is impossible. For the sociologically trained, it is a banal statement, but as Derek Layder (2006: 3) noticed, there are still people who speak about individuals as if they were outside of social forces. For this reason, it is important to recall the individual-society pairing, even though as a theoretical problem it is no longer particularly interesting for sociologists.
I will now turn to the agency-structure pairing, which is now at the center of the sociological debate. Although it was also not directly described by the classics, its strong traces can be found in their works. According to the agency paradox, people are able to act while simultaneously being constrained by structure, and through acting, they are capable of influencing this very structure. To understand this pairing, it is necessary to understand both notions, which, of course, have many definitions in sociological theory. To simplify things, I will assume after Layder (2006: 4–5) that structure can be defined as “the social relationships which provide the social context or conditions under which people act.” The tradition of sociological studies of the social structure is a very long one, and it includes classical analyses of social classes, as well as recent works on social networks. What is common in this thinking is the understanding that a structure which contains elements and the relations between these elements is, in a way, stable or durable, and constrains social actors (individual and collective). Agency is a capability of acting, the state of being a subject not an object. As it was understood by Anthony Giddens (1984), it is the ability to make a difference in the world. Agency is connected with the key sociological notion of social action since Max Weber (1978) recognized it as a crucial subject of sociological inquiry. The tension in the agency-structure pairing is therefore connected with more classical tensions, such as voluntarism-determinism or change-stability. It is a pairing which is strictly a product of the sociological way of theorizing and has occupied the minds of sociologists in the recent decades. A good example of ← 22 | 23 → this sociological focus on agency-structure pairing is the monumental collection of readings edited by Mike O’Donnell (2010a). This collection of four volumes includes the works of classics and current theorists tackling the problem of agency and structure. The most influential contributions in late 20th century sociological theory by Bourdieu (1977), Giddens (1984), and Habermas (1984) are recognized as valuable because of their attempts to solve the agency-structure paradoxes. In the introduction to the above-mentioned collection of readings discussing the agency and structure issue, Mike O’Donnell (2010b: xxvii) stated that the proper understanding of individual-society and micro-macro pairings is necessary for dealing with the crucial for social theory pairing of agency and structure.
What is thus the difference between micro-macro pairing and agency-structure pairing? George Ritzer (1990: 363) defines it in the following way: “Agency is usually micro, but may be macro. Structure is usually macro, but may be micro. Micro usually indicates agency, but may include mindless behavior. Macro usually means structure, but may refer to culture.” Agency is the ability of a social actor, which as a point of departure denotes a human individual (which is connected to the earlier discussion of individual-society pairing). There are, however, also large collective actors, whose agency is performed on macro-level. A canonic example of this is the state, but some authors treat as collective actors (of obviously macro-scale) also nations or social classes.
Many – usually the ones closely attached to methodological individualism – would reject treatment of such large objects as actors (Hindess 1986: 115). According to Hindess, acting requires two capabilities: the capability to make a decision, and then to act on this decision. A nation or social class cannot be treated as an actor because it does not have a mechanism of decision taking, and calling either of them an actor cane be done only in an allegorical sense (Hindess 1986: 115). The state, however, with its procedure of taking collective decisions is, according to this approach, an actor. Hindess (1986) persuasively claims that the macro perspective is also a reductionist one. Conventionally, the micro-perspective is treated as a reduction of complex social processes into their “atomic” level. Yet, holistic approaches are also reductionist, because they reduce the complexity of actors’ internal construction to simple consequences of structural determinants. This reminds of the famous critique of sociology by Denis H. Wrong (1961), claiming that it employs the oversocialized conception of man which reduces human individuals to enactors of social pressures. In a similar fashion, classical economics undersocializes the concept of man, reducing the complexity of social forces influencing decisions of individual. According to Hindess (1986), social structures of macro-level influence the action in its both phases. The decisions are to be taken ← 23 | 24 → only in the framework of certain discourses which allow perception and categorization. Decisions are influenced by macro-cultural structures and acting on these decision is constrained by social and material conditions external to the actor.
The micro-macro pairing was firstly introduced in economics. The author who is credited for coining the terminology to distinguish micro- and macro-economics is Ragnar Frisch (1933), however, it is conventionally regarded that the creator of the subdiscipline of macro-economy is John Maynard Keynes (Canterbery 2011: 490). Keynes (1937) focused his studies on employment, interest, and money. His approach was labeled as “macroeconomics” because it was focused on the processes occurring on the national level of economy: aggregate national income, product, employment, and overall price level. In comparison, microeconomics is focused on choices of smaller decision units (as sociologists would say, “social actors”) such as consumers, households, and firms. Since that time, the division between micro-economics and macro-economics, institutionalized in economics teaching curricula, became commonly accepted. As economics is for sociology very often a discipline to look up to – although sociologists very rarely admit to it – it was quite easy for them to start thinking about the processes they were considering in terms of micro or macro as well.
1.2.2 What does micro and macro actually mean?
Jeffrey C. Alexander and Bernhard Giesen (1987) in the introduction to the collected volume The Micro-Macro Link (Alexander et al. 1987) state that the micro-macro dichotomy is an analytical distinction. This, according to Alexander and Giesen, is what distinguishes micro-macro from previously discussed concrete (substantial) pairings. Individual versus society, and action versus order (another label for agency-structures dichotomy) are of substantive nature, which means that is possible to indicate the real entities behind them. Yet, seen from the more constructivist perspectives, this is not so straightforward: there are quite strong arguments to claim that both individual and society, or agency and structure, are notions developed by researchers in order to better understand social reality. To support Alexander and Giesen’s view, it is necessary to state that “individual” and “society” are categories of practice, while the micro and macro distinction is only used by social scientists analyzing social phenomena. It would be also hard to point to agency and structure as categories of practice but – putting aside the huge sociological debate about agency and structure – it is possible to point to their substantial designates.
Having agreed that micro versus macro is a distinction of an analytical character, I need to point to the consequences of this fact: it is the decision of a ← 24 | 25 → researcher what to name micro and macro; it is also his or her decision how to conceptualize their relation. There are two possible general ways of conceptualizing the link between micro and macro: the first approach is to describe the mechanism mediating between the two; the second one is to introduce another analytical category of meso.
As George Ritzer (1990: 349) noticed, many sociologists use the analytical distinction between micro and macro empirically. The reason for this may be the strong institutionalization of these analytical categories in the scientific field. The consequence of such an use, however, may be their reification and production of artifacts. Thus, scholars often make an equation between the micro-level and the individual-level reality of everyday life, while the macro-level is equated with the social world or whole social reality. The artifacts of analytical framing might be easily produced if what is useful as a tool to analytically perceive reality (i.e. as social processes on different levels) might be smuggled – or in unnoticed way transformed – into the statements on the substantial entities. This is an important lesson regarding the recurrent problem of social theories in which analytical categories are often treated as empirical categories, which may lead to the production of knowledge that does not pertain to social reality but the sociological perception of this very reality.
It is a contingent outcome of the development of sociological theory that micro and macro are so close to other theoretical pairings. As an analytic categorization they are both relational and arbitrary, so something treated as micro from one perspective could be treated as macro from another (Alexander 1987: 291). Apparently, there is some overlap between the analytical micro-macro distinction and the two other key theoretical distinctions, which I have already mentioned in the previous section. My guess would be that many scholars do not sufficiently understand this overlap and the differences between the pairings, and often treat the micro-macro dichotomy as more or less equal to, for instance, individual-society dichotomy. This makes it quite easy to state something about the micro-level having in mind individuals or their interactions. Although in this chapter I often interpret claims of classical theorists regarding other pairings as claims about the micro-macro dichotomy, I keep in mind that this has to be done very cautiously.
It is also important to be very cautious in the attempts of operationalizing micro and macro as observable and unobservable. It is not true that it is possible to plainly observe micro-level entities or processes such as motives, personalities, or biographies. On the other hand, some macro-entities (like legal systems or income distribution) are quite easily to be seen by social scientists (Alexander, Giesen 1987: 21), which is an important argument for treating the micro-macro ← 25 | 26 → pairing as an analytical distinction. Another similar assumption made by some authors is that processes and entities on the micro-level are simple or uniform, whereas processes and entities on the macro-level are complex (Knorr-Cetina 1981: 20). It is not true, because the micro-transactions of everyday life might be enormously complex, while some processes (or at least their representations) on a macro-scale might be quite simple. Some approaches to macro-level treat it as an aggregation of micro-level processes or entities, which in consequence may lead to the assumption that macro-level is more complex as it is a totality of micro-level phenomena. It is a naive way of thinking, because – having in mind that micro and macro are analytical categories – it is necessary to reduce the complexity on both levels through the development of models of reality. The model of macro-process does not include all the elements of its micro-components.
Another important point about the micro-macro pairing is that it might be treated as a dichotomy or a continuum. George Ritzer (1990: 364) claims that micro and macro are just ends of a continuum. Yet, as they are analytical tools, only their usefulness for a better understanding of reality should be important here. Again, the treatment of micro-macro as a dichotomy is quite often connected to its overlap with other pairings which are “more dichotomous” in nature. In the case of individual-society, there is no place for continuum, although researchers often show that in social groupings of different sizes there are different dynamics of social processes. In the case of agency and structure there is no place for continuum as well. Thus, to picture micro and macro as a floor and a ceiling, to some extent arbitrary set by a researcher, it is possible to point to the space between them and try to conceptualize it, as if it included some other levels (i.e. meso); it is also possible to talk about something being more macro (i.e. closer to the imagined ceiling) or more micro (closer to the imagined floor).
To conclude, micro and macro do not mean anything, or they might mean quite a lot. As analytical tools they might be very handy in order to better understand the social reality, but taking them as something substantive might be very deceptive. I truly believe that the use of micro and macro in sociological analysis is helpful in creating a conceptual order in the chaos of various substantive entities. Yet, we – as scholars – need not to reify it.
As pointed out by George Ritzer (1990: 349), the classics of sociology such as Marx, Weber, Durkheim, Simmel, and Mead were very much concerned with the micro-macro linkage. The problem of the relation of individual and his or her experience to the processes of the whole society was preoccupying them because its appropriate tackling would allow to claim that they created a social science – meaning the science of the society – which would help in understanding individuals. In this section, I will briefly present how the relation between micro and macro was theorized by the authors recognized as classics of sociological thought. My arbitrary – though, in my opinion, not controversial – selection of classics includes: Karl Marx, Emile Durkheim, Max Weber, Georg Simmel, George Herbert Mead, and Florian Znaniecki. Much of the discussion of their views on the micro-macro linkage is indebted to authors who were analyzing and sometimes refreshing their ideas, because – what needs to be emphasized at the beginning of this section – the classics were not using the terms “micro” and “macro.” This goes along with a postulate, expressed by George Ritzer (1990: 366), that there is a need to rethink the work of theory of the masters in line with the new perspectives on the micro-macro problem. Their interpretations require, as mentioned in the previous section, distilling the micro-macro question from overlapping questions concerning individual-society and agency-structure.
1.3.1 Karl Marx: the structural conditions
Marx’s perspective on larger collectives’ capability of action started a long tradition of treating as social actors large entities who might be called actors only in an allegorical sense (Hindess 1986: 115). Describing a social class as an actor is a generalization which is actually very similar to describing in the same way a social system. This problem is closer to the debates on agency, and it stems from the fact that Marx’s structural approach had a problem with emergence. It was a quite persuasive theory of explaining the micro-level phenomena with their macro conditions – i.e. superstructure of ideology with material ground – but it ← 27 | 28 → lacked the conceptual tools of understanding how micro phenomena could be transformed into macro ones.
Contrary to Hechter (1983) and Ritzer (1990), Layder (2006: 63) interprets Marx as not very much interested in the issue of micro- and macro-levels of analysis. Nevertheless, he still believes that it is possible to find in some of Marx’s statements hints on how to resolve the issues of agency and structure, and – by the same token – resolve the issue of the micro and macro. One of the most often repeated quotations of Marx comes from the essay “The Eighteenth Brumaire of Louis Napoleon” (Marx 1907: 5): “Man makes his own history, but he does not make it out of the whole cloth; he does not make it out of conditions chosen by himself, but out of such as he finds close at hand. The tradition of all past generations weighs like an alp upon the brain of the living.” This quote opens an avenue for thinking about agency as something limited by social circumstances, but also of seeing these circumstances as an aggregated product of actions. Marx was also calling for revolutionary action in order to change social conditions. This element of Marxism is often presented as contradictory to his strong views on structural conditions influencing the situation of individuals. These incongruences are probably also the reason why Marx was so passionately debated among his disciples and his work led to the formation of contradictory schools. What is, for example, in contradiction with calling for revolutionary change of social conditions is another famous quote from Marx, stating that the only freedom which the market gives workers is the freedom to starve (Hechter 1983: 5). Thus, despite possible perspectives on linking micro and macro, the problem of structural explanation of agency and the focus on macro-historical processes make it possible to consider Marx as the father of structural explanations in sociology (Alexander, Giesen 1987: 7; Hechter 1983: 5).
1.3.2 Durkheim: the collective representations
For Émile Durkheim there existed five levels of scientific analysis: the physical, chemical, biological, psychological, and social (Durkheim 1951: 325; Wiley 1988: 255). Durkheim’s claim that society is a sui generis phenomenon impacted sociological thinking for the entire 20th century. Sui generis means that the society is of its own genus and it is impossible to reduce that which is societal to other levels of analysis. This holism under the label of sociologism was a powerful legitimization for sociology at its birth as a science. The macro-level of society required its own discipline, if there was no possibility of its simple reduction to the micro-level of individual’s psychology. The sociology of Durkheim mostly considers the individual-society pairing, yet obviously his conceptualization of this relation ← 28 | 29 → had important consequences for the sociological treatment of the micro-macro problem. Both levels require its own scientific disciplines – psychology and sociology respectively – but Durkheim’s vision of sociology was of an imperial kind. Sociological level of analysis was, according to him, providing powerful explanations for psychological phenomena. The most striking example was Suicide (Durkheim 1951), which was an analysis of social conditions’ influence on the most individual and most unexplainable act a human is capable of.
The connection between individual and societal in Durkheim’s theory is made through the concept of representations. In his sociology there is no place for the micro-phenomena of interaction (Wiley 1988: 257). On the one hand, collective representations are synthesis of individual representations. But on the other hand, collective representations allow individuals to think, perceive reality, and communicate. The collective representations return to individuals as master categories (Durkheim, Mauss 1963). The lack of interactional level in Durkheim’s sociology creates a quite large distance between micro and macro, which makes his conception of micro-macro fragile. There is a place mostly for structural and dispositional explanations in this theory, but the relational explanations are omitted. Another possible link between micro and macro in Durkheim’s sociology is the social organization. In case of “primitive” societies, the contact of individual and society through collective representations, strengthened by religion was immediate – mechanical. In case of modern societies having a complicated division of labor, the organic contact of the individual and society was mediated through secondary and occupational groups (Durkheim 1933). This mode of theorizing about society opened the way for groupism, which for a long time was the main sociological perception of social structure. The open question for future theoretical studies and revitalization of old concepts remains: was the issue of anomie – disconnection of an individual from society – an artifact of Durkheim’s theoretical framework, or was it rather a genuine social problem? If I am right that Durkheim’s theory had a conceptual problem with relations and connecting micro- and macro-levels, then the answer to this question would be that the conception of anomie was an artifact.
1.3.3 Weber: the methodological individualism
Max Weber, praised by so many as one of the founders of sociology, among other important contributions defining sociological debates for more than a century, introduced methodological individualism into the sociological toolkit for understanding the social reality. As Lars Udehn (2001: 97) phrased it: “By way of extreme simplification, I suggest that Weber took methodological individualism ← 29 | 30 → from the Austrian School of Economics, supported it by a neo-Kantian view of concept formation and turned it into subjectivist methodological individualism with the help of Dilthey and Simmel.” Max Weber rejected what was foundational for classical sociologists (like Durkheim), being the use of collective concepts such as Volk, Volkgeist, or “general will.” According to him, the German Historical School, influenced by German Romanticism and Hegel, relied on metaphysical concepts, and the study of society and its economy should take as a point of departure individual beliefs (Weber 1975; Udehn 2001: 98). Max Weber was very much influenced by economics (many considered him rather as a historian of economy) and modern economy formed with strong individualistic assumptions. Later in this chapter I will discuss how transferring the developments from the field of economics to the field of sociology contributed to the understanding of the micro-macro relations.
Weber in his promotion of methodological individualism was very critical to all forms of holism, thus the subject of sociology for him was not society but social action (Weber 1978). In fact, the concept of society was recognized by him as useless –it was rather a category of practice, which for scientific inquiries could be reduced to social relations and social institutions (Weber 1971: 36; Udehn 2001: 99). Social action could be studied only with the use of the meaningful understanding method (Verstehen). Weber, trained not only in economics and sociology but also in law, treated collective actors (organizations, states, and all possible legal persons) as reducible to the actions of human individuals. Thus, he was fully aware that their actorhood was a category of common practice and the use of language, and therefore that the human individuals under study were the ones to give meanings to the acts of collectives. Individuals attach the meaning to actions of other individuals and take them into consideration when acting – this is constitutional for Weber’s use of the concept of social relationship.
Larger social structures exist for Weber only in the sense that individuals attach meaning to them and thus they drive their actions. It could be said that this way of thinking preceded Thomas’ theorem (Thomas, Thomas 1928) that defining a situation as real has real consequences. Micro, for Weber, is concentrated mostly on actions of individuals and individuals relating to other individuals. He was not interested in individual-society paring as he refuted the notion of society. The perspective on the micro-macro link was taken by Weber with micro as a point of a departure. The best example are his writings on the emergence of capitalism (Weber 1930; 1978). In essence, in the view of Weber, capitalism as a social formation – understood as the organization of formally free labor in methodical, rational, and disciplined work (Birnbaum 1953: 137) – developed, because a ← 30 | 31 → growing number of merchants began to emphasize this-worldly values. The aggregation of individual actions driven by the motive to earn and then reinvest the profits transformed the whole order of economy. Of course, there were also certain conditions, such as the development of the bureaucratic state, legal citizenship, and complex of military, administrative, and religious factors, that played important role in the process (Collins 1980). Yet, the key element for the emergence of the capitalism were the values which drove the actions of individuals. The unanticipated consequence of these individuals’ actions was the reshaping of the organization of economy (Birnbaum 1953). The theory of Max Weber was praised by Alexander and Giesen (1987:15) at the first synthetic formulation of the micro-macro problem. Undoubtedly, his introduction of methodological individualism in order to understand macro-processes was revolutionary for sociological theory.
1.3.4 Simmel: sociability and money
Georg Simmel is conventionally recognized as the founder of the relational approach in sociological theory. His other idea, which later turned in the attempts to resolve sociological pairings, is his processual approach to society. Frank Lechner (1990) claimed that Simmel’s focus on individuals, small groups, and relations is consequential for macrosociology. Simmel’s study of social differentiation is a study of co-creation of modernity as a macro-phenomenon and individuality as a micro-phenomenon. As he claimed, in society, which is becoming increasingly diverse, a human individual participates in more than just one social circle and enacts numerous social roles (Simmel 1980; Lechner 1990). This process transformed the human individual into an individuality – a coherent entity which is capable of being in these different circumstances.
Simmel rejected the idea that society, understood as aggregation of individuals, is an object of sociological study. For him, the object of sociology was sociability (Vergesellschaftung) – a process of continuous interaction between individuals. Simmel was therefore looking at the micro-macro problem from the micro-position and he did not perceive that there exists any kind of sui generis macro-entity. Yet, as Lechner (1990: 172) claims, Simmel was interested in studying a larger process that was making the encounters of individuals possible: “Sociability, as specifically social form, is quite independent of either objective (functional, specific) or subjective (psychological, emotional) contents; rather, it is the form of interaction in which individuals from different specialized sphere can engage without having to give up any part of their individual autonomy.”
In The Philosophy of Money (Simmel 1978), Simmel argued that contacts between individuals require mediation and some sort of standardization. Money, ← 31 | 32 → in his view, is here a perfect tool which serves not only for economic exchanges, but also contributes to the process of individualization, as it makes the human individual more independent from specific collectivities. The work of Simmel is called impressionistic, it usually took the form of essays, and his approach rejected the possibility of creating theoretical systems. Yet, his remarks still include quite a lot of valuable, theoretical framing for the connection between individuals involved in never-ending interactions and exchanges, and larger scale processes. What is interesting for the subject of this book is that Simmel was not only interested in the individual-society pairing, but he also brought into the equation interactions and mediating entities.
1.3.5 Mead: “The self, the I, and the me” – and the society
George Herbert Mead is recognized as a classic thinker for interactionist sociology, symbolic interactionism, and interpretive thinking in the social sciences. For this reason, the references to this author will be found mostly in works of scholars interested in analysis on the micro-level. There is not much to be found in Mead’s work on macro-processes and, as Alexander and Giesen (1987: 9) put it, Mead lacked an institutional theory linking the interactions with larger social entities. Yet, as George Ritzer (1990: 349) claims, Mead was also among the founders of sociology who were very much concerned with the micro-macro link in the theory, despite the fact that he was not using this kind of terminology. Mead’s biggest concern was how the society shapes different aspects of the human individual constitution, yet in his work there are also remarks on the individuals’ influence on society. Thus, in his view it is apparently a feedback mechanism.
In his considerations on the expression of self, Mead highlights that “me” is a member of social group or community, while “I” is expressed in relation with others. “Me” is said to be more sophisticated and suppressing the “I” (Mead 1934: 207). The “self” is a product of the tension between “I” and “me.” Thus, a human individual cannot be considered without his or her relation to other individuals and a social group as an important reference. This analysis is well known and was a strong drive for the development of social psychology and micro-oriented sociology. Mead also had some remarks about the impact of an individual on larger social entities, but he did not conduct an in-depth analysis of this part of the micro-macro interdependency.
As to the aforementioned feedback mechanism, Mead (1934: 202) described it in the following words: “The individual, as we have seen, is continually reacting back against this society. Every adjustment involves some sort of change in the community to which the individual adjusts himself.” When an individual ← 32 | 33 → is an artist, inventor, or scientist, then – according to Mead – his or her impact on society is stronger: outstanding individuals are even capable of forming society. Mead was mostly interested in the individual-society pairing, or maybe rather – using his words – “individual-community” pairing. Yet, this pairing has consequences for other pairings and passages of Mead’s work on realization of self in social situations are also quoted in regards to the agency-structure pairing; thus, they are consequential for the micro-macro pairing. Mead analyzed mostly phenomena occurring on the micro-level – how an individual is shaped in interactions with other individuals and how an individual perceives himself or herself in relation to the society. Yet, he also provided statements about the change on the macro-level. This relation is not theorized by him: he simply states that an individual triggers some (usually) incremental changes of the larger grouping.
1.3.6 Znaniecki: from the attitude to norm and back
According to Norbert Wiley (1988: 260), the most interesting conceptualization of the micro-macro link in classical sociological theory was the one provided by William I. Thomas and Florian Znaniecki (1918) in their work on Polish immigrants to the USA. On the individual level, members of social groups were attributed with subjective attitudes, while on the societal level, were the values – objective elements of culture. This shows that another important sociological pairing which troubled the authors was the subjective-objective. Yet, Thomas and Znaniecki’s approach was processual and structurational: they were neither micro- or macro-imperialists. In their view, individual attitudes were in constant interplay with societal values, so “structural closure was never complete” (Wiley 1988: 260). Thomas and Znaniecki, like other classics, were also obsessed with demarcation between disciplines of social sciences. According to them, social psychology was the study of attitudes, while sociology was the study of social values.
The interplay of attitudes and values has, in Thomas and Znaniecki’s sociology, consequence for the on-going process of social disorganization and reorganization. In their processual approach there are always individuals with attitudes unfit to the social values causing perpetual social change. Another perspective on the relation between individual and society was presented in Znaniecki’s (1934) interest in positive deviance discussed in his book Ludzie teraźniejsi a cywilizacja przyszłości [Contemporary People and the Civilization of the Future]. According to this conception, some of the individuals not fitting into society – but in a positive and not negative way – are agents of societal reorganization. ← 33 | 34 →
This brief presentation of the classics’ vision of the micro-macro link in sociological theory leads to three most important conclusions. Firstly, the micro-macro link, although not defined using this particular terms of micro and macro, was a problem of concern for the classics. They were working hard on understanding how the phenomena that can be noticed on the ground level (acts of individuals and their interactions) are related to larger entities such as society, class, or historical formation. Secondly, for most of the classics, the key sociological pairing was the individual-society, and the key concern was understanding how the individual was connected to society – even if the concept of society was refuted, as in the case of Simmel and Weber. Thirdly, it is possible to find in the works of the classics seeds of all current approaches to dealing with the micro-macro problem in sociological theory. The problem obviously needed to be defined by the use of the micro and macro terminology, which was not known for the classics. The micro-macro framing of levels of analysis migrated to the field of sociology from economics, in which it was coined in the 1930s.
In the 1980s, after 50 years of theoretical extremism and divides between the macro- and micro-oriented sociologists, a vivid debate about the possible links between the camps took place. As Alexander and Giesen (1987: 3) phrased it, the conflict over reduction – either micro to macro, or macro to micro – was replaced by the search for a linkage between the two. In the 1980s, both American Sociological Association4 and International Sociological Association5 organized meetings focusing on micro, macro, and possible bridges between them. There were also several published books and articles dealing with the problem and even publications reviewing the debate. In this section I will briefly describe the theoretical developments of that period (although some of them have earlier beginnings6). I will ← 34 | 35 → also show that the debate successively moved to the focus on the agency-structure problem, which is now at the center of theoretical attention in sociology. As some would argue, macro-extremists such as Talcott Parsons, Ralf Dahrendorf, and Peter Blau, and – on the other side – micro-extremists such as Herbert Blumer, George Homans, and Harold Garfinkel, dominated the sociological theory for half of the century before the 1980s, and their inability to come to terms and link micro and macro was the reason for the theoretical crisis of sociology.
According to a comprehensive summary of the debate by George Ritzer (1990: 348), there were three approaches to tackle the issue of the micro-macro in sociological theory: (1) Attempts to integrate micro and macro theories; (2) attempts to develop a theory that deals in an integrated manner with micro and macro phenomena; and (3) attempts to do both at the same time. Another possible perspective to understand the debate is to look at the starting point that was taken in these attempts. Some authors departed from micro, some from macro, and some tried to avoid taking either of the ends of the continuum as a starting point. To order this section I will refer to the latter – quite simple – typology.
1.4.1 The micro-end perspective
The powerful theoretical attack on the micro-macro issue had strong footing in the economic tradition. The rational choice theory, applied in sociology to explain macro-phenomena using micro-level assumptions on how actors make their decisions, was well-developed by earlier explorations in the field of economics. For example, Mancur Olson (1965) showed that individual interests do not simply translate into the interests of groups or organizations, but they require institutionalized coercion, which works as a mediating mechanism. Another important inspiration came from the work of Thomas Schelling (1969), who modeled how the decisions of individuals may produce the structural effects of segregation.
As one of the important authors of the rational choice camp, Michael Hechter (1983: 9), pointed out, the assumptions about wealth-maximization might be unrealistic and not useful to explain the behaviors of individuals, but they are actually very useful in explaining the aggregated level of macro-social process due to the virtue of canceling the idiosyncratic preferences of individuals. Thus, Hechter (1987) employed rational-choice theory to explain group solidarity. It was done, ← 35 | 36 → in a way, to counter the established sociological tradition, in which norms of solidarity were used rather in order to explain behaviors of individuals. Hechter aimed at showing that this solidarity is a product of interests of individuals.
James Coleman was another prominent member of the camp of sociologists employing assumptions on the rationality of the individual actor in order to tackle the problems of the macro-level. As he phrased it, “a central intellectual problem in the discipline [of social science theory] is the movement from the individual level, where observations are made, to the systemic level, where the problem of interest lies. This has been called ‘micro-to-macro problem,’ and it is a problem that is pervasive in the social sciences generally” (Coleman 1987: 154). As this shows, for Coleman micro was equal to the individual but macro was on the level of the system and not the society. The system is something broader (but it could also be something on much smaller scale) than society. For example, in Coleman’s analyses capitalism was treated as a system but also panic was a system level phenomenon, because it was a system of behaviors. Coleman, in order to explain causal relations involving micro- and macro-levels, was using a diagram, often referred to as “Coleman’s boat” or “Coleman’s bathtub” (Raub, Voss 2017).
The line of thinking assuming the rationality of individuals was eventually the base for Coleman’s (1990) opus magnum Foundations of Social Theory, which was planned as a comprehensive theory of all social processes, similar in its ambitions to Talcott Parsons’ (1936) Theory of Social Action. I will focus on the Coleman’s (1987: 156–157) solution to the micro-macro problem by looking at his critique of Ted Gurr’s (1970) explanation of revolutions by relative deprivation. Coleman placed improved social conditions in his bathtub-diagram as macro-conditions which when deteriorating caused frustration (as micro-conditions). The frustration of individuals was causing their aggression (micro-outcomes), which – according to Ted Gurr – was supposed to produce an aggregated effect ← 36 | 37 → of revolution (macro-outcome). Coleman (1987: 157) ridiculed this causal explanation by saying that it supposed “a simple aggregation of individual aggression somehow to magically produce a social product (that is, revolution).” What Coleman pointed to as missing in this mechanism was social organization or, in cases of other failed micro-macro causal explanations, the lack of institution. The properly identified institution is therefore a mediating mechanism allowing the transition from individual to systemic level – a simple aggregation of individual behavior was not enough for him. According to Coleman, a good example of institution as a mechanism explaining the transition from the individual to systemic level was the political institution of electoral system, which may differently translate voters’ preferences into victory in the elections.
It is also worth highlighting that Coleman’s understanding of the micro and macro notions was analytical in a way that he placed on the micro-level all entities and processes that are directly observable. On the macro-level, he saw “hidden” important entities and processes which exist but are not directly observable. Here, the pairing of micro and macro has also its methodological dimension.
Raymond Boudon (1979) saw sociology as a science close to economics – because it applied individualistic methodology – and close to history – because it was interested in banal matter and individual facts. Yet, what – according to him – made sociology different from economics was its attempt to develop a general theory of action. The difference between sociology and history lied in the use of individual facts to search for general structures. In Boudon’s approach to the micro-macro link (to a great extent overlapping with the interest in the agency-structure dilemma), the point of departure was an individual actor and rationality of his or her choices. Boudon (1998), however, highlighted the limitations of the rational choice approach – especially in Coleman’s formulation. Boudon was also reinterpreting Durkheim’s work not as holism but relational realism. In his friendly understanding, the scientific program of Durkheim (also applied by Boudon) was to study the complex influence of the structure of interaction systems on actions and emotions of actors participating in these systems. This shows a feedbacking mechanism between actions of individuals (micro-level) and systems of interactions (macro-level). In this line of interpretation, anomie or division of labor are not simply societal-level mysterious forces, but unintended effects of aggregation of individual actions (Boudon 1979; 1982). Thus, in Boudon’s linking of micro and macro sociology, the center of analysis was an individual actor, whose actions were possible to be explained by relation to the structure of interaction system.
An entirely different attempt to link micro with macro from the micro-end was taken by scholars from the interactionist school of thought. Karin Knorr-Cetina ← 37 | 38 → (1981), in the introduction to the volume on integration of micro- and macro-sociologies (Knorr-Cetina, Cicourel 1981), stated that the micro-macro debate was dominated by the methodological dichotomy of methodological individualism versus methodological collectivism (holism). According to her, the solution to the problem was the methodological situationism embedded in the theoretical tradition derived from Simmel and Mead. Knorr-Cetina’s understanding of the micro-macro division was quite simple. In her view, macro-sociology was the study of society, social institutions, or socio-cultural change on an aggregated level, while micro-sociology was interested in self, routines, conversations, and situations (Knorr-Cetina 1981: 1). The momentum to the development of micro-level theories was given by the cognitive turn, thanks to which social order is not understood as a normative and given order, but as order constructed by active and knowing subjects. The process of constructing the order manifests itself through every-day encounters, interactions, or routines of individuals. To study this phenomenon, according to Knorr-Cetina (1981: 8), the most suitable is methodological situationism, in which the interaction in social situation is treated as the relevant unit of analysis. In comparison to methodological individualism (so important for other micro-end attempts to micro-macro dilemma), methodological situationism is assumed not to be reductionist. This line of micro-sociological analysis takes into account that participants of situations make references to institutional arrangements of macro-scale which cannot be reduced to the level of interactions (Knorr-Cetina 1981: 12). In this way, methodological situationism is seen as a “third way” and a possible solution to the micro-macro dilemma translated by Knorr-Cetina as methodological individualism versus methodological collectivism dichotomy: “methodological situationism promoted in micro-sociological research challenges methodological individualism for the simplifying assumption that the locus of social action is the individual human being, and it challenges methodological collectivism for the equally simplifying and presumably related assumption that interview responses, or data in the form of reports and organizational records, constitute direct, valid sources of macroscopic inferences” (Knorr-Cetina 1981: 15). Going further, Knorr-Cetina states that there are three possible ways of reconstructing the macro-sociology from the micro-perspective. The first one is to treat macro-phenomena as aggregations and repetitions of micro-episodes (Knorr-Cetina 1981: 25) and, according to her, it is the approach of Randall Collins (see Collins 1981b). The second way is to treat macro-phenomena as unintended consequences emerging from the micro-events (Knorr-Cetina 1981: 27). Knorr-Cetina believes that Giddens’ (1984) theory of structuration is an example of that approach. It is also possible to include here Boudon’s (1982) ← 38 | 39 → earlier-discussed conception of perverse effects as another example of this line of reasoning. What needs to be underlined, is that these two approaches treat the micro- and macro-levels not just analytically but also substantially. If macro is an aggregation of many micros or an unintended consequence of many micros, then the differentiation is of empirical character. Yet, Knorr-Cetina (1981: 30) favors the “third way” of reconstructing the macro-sociology from the micro-perspective, which she calls “the representation hypothesis.” According to this perspective, “macro appears no longer as particular layers of social reality on top of micro-episodes composed of their interrelations (macro-sociologies), their aggregation (aggregation hypothesis), or their unforeseen effects (hypothesis of unintended consequences). Rather, it is seen to reside within these micro-episodes where it results from the structuring practices of agents” (Knorr-Cetina 1981: 34). Thus, according to this perspective, only micro-level entities such as individuals or encounters are substantial. Macro-level entities are representations constructed by actors involved in micro-level which have a different ontological status, yet they are often reified, treated as real, and having real consequences. Thus, only because a small group of people interacting together in a room is treated as the government of the state, their actions may be represented as acts of the state and have consequences for many other similar scale meetings.
Randall Collins was another author who took part in the 1980s debate on the micro-macro problem in sociological theory. He too represented the micro-end perspective with his approach of radical micro-sociology (Collins 1981a). According to this approach, only the translation of macro-concepts into aggregates of micro-events makes them fully empirical (Collins 1981a: 984). There are only three pure macro-variables which cannot be reduced to micro-events: time, space, and number (Collins 1981b: 98). All other variables of an alleged macro-level are reducible to micro-reality: “structural variables often turn out to be sheer numbers of people in various kinds of micro situations” (Collins 1981b: 99). Thus, according to Collins, only the reality of interactions between individuals is empirically researchable – and this is the only kind of social reality. The variables or concepts of the macro-level are of different ontological status – they are not real per se, although participants of interactions make references to these macro-concepts.
Building on the propositions presented in his work Conflict Sociology (Collins 1975), Collins listed four main types of macro-references made in social situations: individual micro-histories, situational macro-views, pure macro-variables, and analysts’ macro-comparisons (Collins 1981b). The individual micro-histories are implicit macro-references – in many micro-situations the ← 39 | 40 → participants refer to some broader segments of time or space, which are external to a given situation. The pure macro-variables of time, space, and number are abstract, yet they allow to construct temporal, numerical, or spatial aggregations of micro-experiences. This kind of aggregations is constitutive, according to Collins (1981b: 99), for the macro-level of analysis. The fourth type of macro-reference is different from the first three types: analysts’ macro-comparisons refer to the frames of situations, while the first three types are within these frames (Collins 1981b: 101). Humans have the analytical ability to compare various situations, and by virtue of this ability they are also able to create macro-concepts. Sociology is a special case of this ability: sociologists produce macro-concepts by comparing micro-situations, even though some of them seem to believe that they discover some hidden variables, and not simply produce them in concrete micro-situations of doing the actual analysis.
Randall Collins’ approach to the micro-macro issue is one of the most advanced products of the interactionist tradition represented by authors such as Herbert Blumer, George Homans, Erving Goffman, Emmanuel Schlegoff, or Harald Garfinkel. This stream of theorizing about micro and macro did not have such a powerful legitimizing ally in economics as the rational choice approach. Yet, the interactionist tradition is still influential for the micro-macro debate because it was an important inspiration for authors attempting to develop the dialectical conception of linking micro with macro. But before I will turn to the scholars who wanted to link micro with macro from the middle, I will briefly discuss the theoretical positions of authors who attempted at tackling the dilemma from the macro-end: Jürgen Habermas, Niklas Luhmann, and Jeffrey Alexander.
1.4.2 The macro-end perspective
Jürgen Habermas, a social philosopher influential for the sociological imagination of the late 20th century, started with the materialist inspiration of Karl Marx in order to integrate the system theory with action theory. For Habermas (1975), the micro-level is the level of action, while the macro-level is the level of the system. The lifeword is the level, where “participants in communication come to an understanding with one another about something” (Habermas 1984: 337). Communication is the requirement of action, as emphasized in the title of Habermas’ (1984) opus magnum The Theory of Communicative Action. What is important is that the two levels of Habermas’ theory are not only analytically constructed: the difference between them is of existential character (Ritzer 1990: 354). The lifeword and the system are different modes of organization of action. In the lifeword, social actors communicate with each other by referring to what ← 40 | 41 → is common for them thanks to everyday experience. The system is more abstract as it develops because of the growing complexity of institutions of power and expert systems. Habermas noticed that the system is colonizing the lifeword. The lifeword and the system are organized around different principles of social integration: the lifeword is organized around communicative understanding, while the system is organized around markets and institutions of power. The system developed in an evolutionary way – which is a Marxist inspiration – because of the changes in the organization of production but, contrary to the Marx’s thesis that the base determines the superstructure, the lifeword has its own dynamics (Habermas 1975). Of course, lifeword in Habermas’ theory is not equal to the superstructure. In some interpretations, it may be treated as a conceptualization of the micro-level, while the system is considered to be a conceptualization of the macro-level (Knorr-Cetina 1981; Layder 2006; Ritzer 1990). Nevertheless, the most important concern of Habermas is the paring of agency and structure. And when tackling his enormous work from this perspective, both the lifeword and the system may be treated as structures for action, but organized around different principles of coordination: either communicative understandings or abstracted institutional mediations.
Niklas Luhmann is another figure listed among theorists who tackled the micro-macro problem. Being mostly interested in autopoietic systems, this author was also looking at the problem from the macro-end. Luhmann (1987) argued that the distinction between interaction and society as systems is of an empirical character, while the distinction between them as levels of analysis is of a logical character. The use of levels helps, according to Luhmann, in excluding self-references and, as a result, putting aside the tautologies and paradoxes. In that sense, Luhmann found the distinction between micro and macro useful for reducing complexity in the description of objects. Microprocesses, in Luhmann’s theory, are existential needs to reduce complexity (Alexander, Giesen 1987: 34). Yet, the crucial distinction for Luhmann was the one between systems of society (which could be somehow equalized with the macro-level) and interaction (which could be somehow equalized with the micro-level). Thus, societies encompass all operations that have qualities of communication, while interactions – as systems – take into account also the fact that some communications are from outside the system. For interactions, persons are devices used in defining the system boundaries (Luhmann 1987: 114).
Jeffrey Alexander was one of the leaders of the 1980s debate on the micro-macro link and he represented the view from the macro-end. Alexander became known as the theorist of the social system and scholar attempting to revive ← 41 | 42 → the tradition of Talcott Parsons. In his interpretation, Parsons’ approach to the micro-macro problem was a synthetic formulation comparable to Max Weber’s contribution to sociological theory. It is an obvious exaggeration and many authors who tried to tackle the micro-macro issue from the micro-perspective (i.e. Hechter 1983) either adopted the integrative approach (Ritzer 1990) or simply treated Parsons as a macro-extremist or macro-chauvinist, who was “explaining social reality in terms of actions that become institutionalized to meet fundamental system needs” (Turner 2010c: 3). Yet, according to Alexander and Giesen (1987: 22–24), Parsons managed to link micro and macro thanks to the notions of internationalization and social role. In my opinion, this of course is a link between micro and macro, yet through this link the conditions from the macro-level are simply transmitted into the micro-level phenomena, i.e. actions of individuals are explained with the fact that they internalized broader norms. Keeping in mind that micro- and macro-levels are analytical categories, it is not possible to solve such disputes.
Alexander’s perspective on the micro-macro problem starts from the macro-end because he first looks at systems as action collective environments. Regardless of all the stipulations, micro is for Alexander a level of action, whereas macro is the level of the system. In his view, the problem of the micro-macro divide was a consequence of the lack of holistic sociological theory. Previous to him, macro-theorists were trying to build theories only about one particular system and one particular action mode. Therefore, authors dealing with system of economy were assuming the action as strategic, cultural theorists for whom culture was a systemic environment assumed action as typification, and social movement theorists assumed an inventive mode of action (Alexander 1987: 31). According to Alexander, these are just some variables, which require integration into one larger theory.
Alexander attempted at building a theory bridging the micro- and macro-levels, and he stated that action (the micro-level) and collective environment (the macro-level) are mutually producing themselves: “The collective environments of action simultaneously inspire and confine it. If I have conceptualized action correctly, these environments will be seen as its products; if I can conceptualize the environments correctly, action will be seen as their result” (Alexander 1987: 303). This statement resembles Giddens’ (1984) theory of structuration, according to which actions reproduce structure, while structure influences actions. To some extent, the approach of Alexander is a bit more sophisticated, and it could be interpreted as a kind of analogy to Heisenberg’s uncertainty principle: if theory starts to be built from the conceptualization of action, then the collective environments will be inevitably seen as products of action, while if the ← 42 | 43 → theorizing begins with a description of systems or environments then actions will be seen as their consequences. The only way out of this quibble – although not pointed by Alexander – would be to start building the theory not from the micro or macro end. Yet is it possible? George Ritzer (1990) claimed that this is what Bourdieu, Giddens, and he himself at least attempted to do by starting to apply the dialectical approach and building the theory from the middle of the micro-macro continuum.
1.4.3 The dialectical perspective
There will be much more space devoted to the thought of Pierre Bourdieu in Chapter 2, in which I discuss his theory of field. Here, however, I am only going to briefly discuss how his concept of habitus is regarded as a solution for linking micro with macro. Micro, for Bourdieu, is an individual with his or her will, while macro is a structure of positions. In this sense, the macro could be the whole society but also a field, which in the next chapter I will be discussing as a social space on the meso-level of analysis.7 The distinction between micro and macro is for Bourdieu (1981: 305) similar to distinctions between events and long durée or the great man and collective forces.
One important caveat is that, according to Bourdieu himself, the theoretical tools he uses are understandable only in relations between them and only in the context of the whole theoretical system they constitute (Bourdieu, Wacquant 1992: 96). Yet, I am not subscribing myself to the chapel of Bourdiean zealots, and – like other interpreters – I allow myself to take only a part of his theory to present here. Bourdieu is often said to be trying to reconcile his two main theoretical inspirations – existentialism and structuralism. In this sense, it might be said that Bourdieu was concerned mostly with the subjectivism-objectivism dualism, but in the most prominent interpretations, this allowed him to contribute to solving the agency-structure dichotomy (O’Donnell 2010b; Ritzer 2010). Again, it is justified to discuss his work in the context of micro- and macro-levels of analysis, because Bourdieu was very much interested in individuals’ biographies (existentialist influence), shaped by and opposed to societal forces (structurationist influence). Bourdieu focused on notions of social practice and habitus as theoretical tools serving the objective of getting rid of what he believed to be, the false dichotomy of subjective and objective. ← 43 | 44 → Habitus is defined8 as a “socially constituted system of structured and structuring dispositions acquired in practice and constantly aimed at practical functions” (Bourdieu, Wacquant 1992: 121). Habitus, as an attribute of an individual, is a product of structures, but it also shapes the practices of an individual, thus feed-backing the structures – the dispositions are both structured and structuring: “there would be no game without belief in the game and without the wills, intentions and aspirations which actuate the agents; these impulses, produced by the game, depend on the agents’ positions in the game” (Bourdieu 1981: 308). The structure of the game produces aspirations of individuals, who then “really” love something or “really” desire to achieve something. By these acts aiming at fulfilling their desires and aspirations, individuals contribute to the reproduction of the structure of the game which, in turn, influences them. The habitus is, therefore, neither free will, nor simply the determined product of external societal forces.
Critics of Bourdieu claim that he was not able to get out from structural determinism. In his view, biography is an illusion shaped by social forces (Bourdieu 1998). Habitus often seems to be fixed by social structures and then to disable its bearer from the possibility of changing the position in a given field. On the other hand, habitus is a dynamic structure which changes with time and new situations faced by individuals. Bourdieu’s conception of habitus – at least in my opinion – did not provide a solution to the micro-macro issue, and the notion of habitus to a great extent resembles internalized norms in the functional theory. Nevertheless, Bourdieu to a large degree inspired the debate on agency and structure, as well as the micro-macro link, by pointing to the need of dialectical, not dichotomous, understanding of agency-structure and micro-macro. Next, I will present Giddens’ theory of structuration which has two main similarities with the theory of Bourdieu: focus on the reproduction of structures by action, and focus on social practices.
Antony Giddens (1984) created an eclectic sociological theory of structuration explained in his key work Constitution of Society. Giddens’ interpreters agree that his main theoretical objective was to resolve the agency-structure problem in sociological theory. Yet, since – as the guiding frame of this chapter assumes – the agency-structure pairing has important overlaps with the micro-macro pairing, the structuration theory brings strong insights into the micro-macro issue. Giddens is an influential theorist, and his conceptions cover the most important ← 44 | 45 → subjects of the current sociological inquiries, which makes it simply impossible to discuss all his thoughts in this one chapter. Giddens inspirations are very diverse. He seriously discussed Marxism, he was critical but knowledgeable about structural functionalism, he revisited Durkheim’s methodology and clashed it with interpretive ones (Giddens 1976). Nevertheless, Giddens’ most important influence was his teacher, Norbert Elias. The novelty in the approach of Giddens – although the influence of Elias here is quite visible – lies in the rejection of the discussion about the agency-structure, either from the action or from the structure (or order, or system) perspectives. All theories built on strong assumptions about the system-level influence on individuals end up quite weak when dealing with individuals reflexivity. At the same time, approaches focused on individuals or situations have serious problems in taking broader social structures into account. Giddens (1984) starts with an assumption that action is in a dialectical relation with structure, and that it is impossible to simply claim that structure determines action, or that action determines structure. Agency and structure are co-creating each other, so the individual is confronted with social structures, systems, and institutions which limit his or her choices. Yet, every action of the individual reproduces the structure and has a potential for its slight transformation. Giddens (1984) avoided starting either from the micro- or the macro-end by putting in the center of analysis social practices. They are not just experiences of individuals (micro-end) or social totality (macro-end): social practices are recursive activities continuously recreated by actors. In this sense, they are expanding from a single actor but also from a single situation, which was the key focus for micro-oriented authors such as Knorr-Cetina, Cicourel, or Collins. At the micro-level, Giddens is interested in individuals who are capable of agency, if they are able to make a difference in the world. At the macro-level, Giddens (1984: 17) talks about social systems, which are “reproduced relations between actors or collectives organized as regular social practices.” Structure, in Giddens’ view, is understood as rules and resources which allow practices to exist in different places and moments of time; structure binds together actors and systems. The structure is not external to individuals, because it exists only through their actions,9 thus enabling practices and linking together agentic individuals with social systems. This takes place through the process of structuration, which is an everyday reproduction of structure by action. Social structure is, therefore, both the medium and the outcome of an action. It allows individuals to act, but it also ← 45 | 46 → constrains their actions. Individuals by actions on the micro-level co-create social systems (macro-level entities), which are usually unintended consequences of individuals’ actions. These unintended consequences of previous actions become conditions for future actions. In this line of theorizing, Giddens is actually not that far from Boudon (1982) or Coleman (1990), who also recognized the aggregation of unintended consequences as mechanism binding micro with macro (see Mica 2015). The main concern for Giddens is the relation between agency and structure. The pairing of micro-macro is for him only a background, but the strong focus on the dialectics between agency and structure also helps to understand the micro-macro relation in dialectical terms.
George Ritzer did not only summarize the discussion about the micro-macro link (see Ritzer 1990; 2010), but also developed his own theory linking micro and macro. First presented in his book Toward an Integrated Sociological Paradigm (Ritzer 1981), the theory was repeated and slightly upgraded in many of his later writings, especially in the numerous editions of his handbooks of sociological theory (see Ritzer 2010). According to Ritzer’s so-called integrated sociological paradigm, the micro- and macro-level are in dialectical relation, and it is not possible to understand one without its relation to the other. Ritzer (1995) exemplified this dialectic with the idea of personal troubles being always in relation to public issues.10 Troubles of many individuals may translate into the issues of public concern and, conversely, the concerns of the public policy may have an impact on the personal experiences of individuals. In Ritzer’s integrated sociological paradigm, the dimension (Ritzer always highlights that micro-macro is rather a continuum and not a dichotomy) is crosscut with the subjective-objective continuum.11 Thanks to this maneuver, Ritzer (2010: 503) developed four major levels of analysis: I. Macro-objective (i.e. society, law, bureaucracy, architecture, technology, language); II. Macro-subjective (i.e. culture, norms, and values); III. Micro-objective (i.e. patterns of behavior, action, and interaction); and IV. Micro-subjective (i.e. perceptions, beliefs, various facets of the social construction of reality). The difference between the objective and the subjective in Ritzer’s approach actually equals the difference between the ← 46 | 47 → material and the mental. Dialectical relations, according to Ritzer (2010: 503), occur between all four levels of analysis, so in their graphical representation there are six arrows symbolizing all possible connections. Ritzer’s approach to the micro-macro issue is not particularly innovative: he only states that the dialectical relation between the micro- and macro-levels of analysis needs to be taken into consideration, and sees this magical formula as a solution of the entire problem. Also, his adding of the subjective-objective dimension does not bring much novelty to the debate. Nevertheless, it is necessary to credit Ritzer for his great work in reviewing the debate on micro-macro and clarifying the most important statements of its participants.
In this section I will briefly discuss two other approaches to the micro-macro link which are strongly rooted in the 1980s debate but, at the same time, they depart from it. I will summarize the conceptions of Anselm Strauss, David R. Maines, Jonathan H. Turner and Derek Layder.
1.5.1 Is there a meso in between micro and macro?
Having established that micro and macro are ends of a continuum, and that they are of analytical – not substantial – character, it is possible to start to speculate about what is in between them. In this section, I present some of the authors who used the term “meso,” and I attempt to evaluate whether these uses had any application for the debate on the micro-macro pairing.
Jerald Hage (1980), in his work on the sociology of organizations, distinguished three levels: micro, meso, and macro. On the micro-level, he placed social positions and groups, while on the macro-level, multi-organization and task environment. The focus of his interest – organization – was placed on the meso-level,12 because it was a set of social positions and groups (micro-level entities) operating (quite often as a part of a multi-organization) in some kind of a task environment, supplying it with input and receiving its output. What is interesting in this categorization, which with its functionalist position is now rather outdated, is that Hage’s phenomena from three levels are, on one hand, irreducible, and that – on the other hand – it is not possible to understand organizational processes without taking into account the processes from micro-, meso-, and macro-levels of analysis. In this approach, micro, meso, and macro are not only ← 47 | 48 → treated as analytical categories, but they are also labels for categories of entities of various size and dynamics.
The search for the mesostructure (Maines 1982) and negotiated orders (Strauss 1978) was another attempt from the micro-end to find a solution for the micro-macro dilemma. Anselm Strauss, a disciple of Herbert Blumer and co-founder of grounded theory, started with the empirical assumption that social actors are continuously involved in negotiations, yet the negotiations cannot be understood only in the context of given interaction. The theoretical aim is not to reconstruct the macro-concepts from the micro-perspective but to look for the context which allows to understand interactions of individuals and the influence of macro-structures. The key notion of this approach is the one of social order. Strauss did not have an ambition to reduce macro-structures to aggregations or unintended consequences of micro-events. For him it was important that macro-structures are treated by actors as given, so they are elements of social reality. Yet, actors never experience the structural context as such – it is always mediated by the negotiations context (Strauss 1978: 237), which should be treated as a mesostructure and the proper level of analysis (Maines 1982). David R. Maines (1982: 278) claimed that this is the most adequate level of analysis which should replace the antiquated divide between micro and macro. Strauss attempted at reconciling not only microscopic interactions and macroscopic structural problems, but also the tension between cooperation and conflict. According to him, all negotiations consist of elements of actual or potential conflict and cooperation (Strauss 1978: 250), and therefore social orders are always at the same time negotiated but also coerced (Strauss 1978: 262). Although the notion of negotiated order was originally developed with the aim to understand the interactions between individuals in the context of the organization of a psychiatric hospital (Strauss et al. 1963), it was later applied to study other kind of negotiations, i.e. diplomatic negotiations. Maines (1982: 275) defined the negotiated orders “as forms of labor, [which] contain dialectical activity in which the human subject constitutes and in turn is constituted by a social object.” Thus, the mesostructure is a context which exceeds the given situation of negotiations, but it is closer to social actors than the overall structural context. In this sense, the notion of mesostructure resembles the notion of social field, which I am going to discuss in Chapter 2.
George Ritzer (1990: 358) commented in his summary of the 1980s debate on the micro-macro dilemma that meso-level perspectives might not be sufficient to explain the marco-level and micro-level phenomena. According to him, authors calling for meso-level analyses still needed to demonstrate that they are capable of providing a satisfactory integration of sociological theory. The well-developed ← 48 | 49 → approaches to the micro-macro dilemma which matured after the 1980’s debate will be presented in the next two chapters. What I would like to add here as a comment is that the development of the meso-perspective could also achieve its own dynamics. This means that the meso-level might be treated not only as a direction of integrating micro with macro, but as a level of analysis valuable and sufficient by itself. It is an attractive level of analysis for authors who are not interested in the micro-realm of individuals or interactions but also do not have the ambitions of explaining macro-structures or macro-processes.
1.5.2 Turner: corporate and categoric units
In the recent years, Turner (2010a; 2010b; 2010c) has built a grand sociological theory which, according to him, “is close to doing what no other sciences has done: explain all levels of its operative universe theoretically” (Turner 2016: 146). As he explains, the levels of social realm are embedded in each other, and thus the macro-level through the meso-level constrains the micro-level encounters and emotions released during these encounters. It is important to remember that Turner declares himself as neither a “macro chauvinist” or “micro chauvinist,” but he simply focuses on top-down relations between the levels – according to him, general sociological theory needs to deliver top-down and bottom-up explanations of relations between the levels of analysis (Turner 2016: 123).13 Turner lists the following macro realms: inter-societal systems, societies, institutional domains, and stratification systems. On the meso-level, there are corporate units and categorical units. Eventually, on the micro realm there are encounters of individuals.
Thus, in each society (but also across societies, in the world of global relations) there are two macro realms: institutional domains and stratification system. The most important institutional domains are: kinship, economy, polity, law, religion, ← 49 | 50 → education, science, medicine, sport, and arts – each having its distinctive symbolic media. Stratification systems are “built up from unequal distribution of valued resources that are distributed unequally by the divisions of labor in corporate units within institutional domains” (Turner 2016: 130).
Corporate units – from the most interesting in this section meso-level of social reality – are groups, organizations, and communities, and they reveal the division of labor. The second kind of units of the meso-level are categoric units “composed by members defined by traits or characteristics” (Turner 2016: 125): gender, ethnicity, age, religion etc. All encounters (the micro realm) occur on the ground of some corporate unit – i.e. two individuals interact in their workplace, which is an organization, and most often deploy expectations and valuations based on category membership.
In Turner’s theory, meso-level realms are unequal. Corporate units are quite concrete and sometimes of a relatively small size (social groups, organizations, or local communities), while categorical units might be quite abstract and broad (for example, ethnicity). Yet, corporate units may also be of an enormous size – like international corporations – whereas categorical units may also be very small in size, like small in numbers ethnic or legal categories. Turner perceives them as meso-level realms, because they are directly constraining that which, according to his theorization, is on the micro-level: encounters between individuals, who take into account the categories that the partners they encounter – always in the frame of some corporate unit – belong to. Turner’s approach is, probably, the last attempt at building a grand theory in sociology: his overwhelming work Theoretical Principles of Sociology contains three volumes and more than 1,200 pages (Turner 2010a; 2010b; 2010c). Its trick in linking micro and macro lies in the assumption that categoric and corporate units are on the meso-level. Keeping in mind that this differentiation is only of an analytical character, Turner’s solution may be criticized as shuffling with the placing of units on various levels of analysis.
1.5.3 Layder: social domains
Derek Layder (2006), in his book Understanding Social Theory, which was first published in 1994, describes the debate on micro-macro as one of the most important attempts to understand sociological theories. His comments and interpretations of other scholars’ stances on the micro-macro dilemma were cited by me earlier in this chapter. Here, I am going to briefly discuss Layder’s own original conception of social domains. Although Layder did not use the term “meso,” I include him in the category of theorists that construct additional levels between ← 50 | 51 → micro and macro in order to link them, which is why I present his conception in this section. Layder (2006) proposes to differentiate four domains of sociological analysis. He rejects both thinking in dualist terms (such as micro-macro, individual-society, or agency-structure) and attempts to integrate these dual oppositions. According to him, social reality consists of four various domains: psychobiography, situated activity, social settings, and contextual resources. This distinction spans from the more intimate to the more impersonal and distant aspects of social reality. The domains stretch across time and space, and are connected through social relations of power. The domain of psychobiography is built on the unique experiences of a person obtained in the course of his or her life trajectory. The next domain listed by Layder is the domain of situated activity framing episodes, situations, and social interactions. Layder (2006: 279) defines situated activity as “a subtle and complex amalgam of the powers, emotions and mutual influences of multiple individuals that unfolds in the real time of the encounter.” This domain, according to Layder, is crucial (but not exclusive) for production of meaning. The third domain is the domain of social settings, which pertain to the environment for situated activity. Layder (2006: 280) as examples of social settings lists formal organizations, as well as informal networks which, according to him, are “local aggregations of reproduced social relations, positions and practices.” The fourth domain is the domain of contextual resources which have two aspects: material and cultural. According to Layder (2006: 281), the cultural aspect of contextual resources has some similarity to the concept of cultural system in Talcott Parsons’ (1936) theory, although it is less strictly coupled and includes various sorts of cultural codes. Layder’s attempt at providing a new theoretical solution to the problem of three main dualisms in sociology (individual-society, agency-structure, and micro-macro) did not attract much attention of other scholars. His book is rather praised for being a well-done analysis of other theorists’ work than for being an original contribution to the discussion. In my opinion, the problem in Layder’s theoretical proposal lies in the fact that he did not provide any useful meso-level or mechanism linking micro and macro. As a matter of fact, his proliferation of levels of analysis (into the four domains) only made things more complicated.
The aim of this chapter has been to present the classical and the most influential approaches to the problem of the micro- and macro-levels of analysis in sociological theory. I have compared the micro-macro pairing with other important for sociological theory pairings, such as individual-society and agency-structure. ← 51 | 52 → I have also discussed the inspiration coming from the field of economics to distinguish these two levels of analysis. The distinction between micro and macro (as well as the meso or more levels, if they are introduced) is a distinction of an analytical character, although many theorists saw it as something empirical – sometimes because they built on different theoretical assumptions, sometimes because they confused micro-macro with the two other pairings (individual-society and agency-structure).
I have presented the stances of classical sociologists (Marx, Durkheim, Weber, Simmel, Mead, and Znaniecki), which might be interpreted as their views on micro and macro. Although they did not use the micro-macro terminology, they can be seen as antecedents of all current approaches to the issue. I have also presented, what I called, the 1980s debate on the micro-macro link. The actual debate began before the 1980s and was continued also in the later years, nevertheless, the 1980s were a time of particular focus on solving the problem of reconciling the micro-macro dichotomy in sociological theory. The debate concerned also the agency-structure pairing, which is still a crucial problem for sociological theory.
In the last section of the chapter, I have discussed attempts to link micro with macro, not through some kind of mechanism or by reduction, but by establishing a level (or levels) of analysis between them, referred to by some theorists as the meso-level. In this section I have presented theoretical achievements which were accomplished mostly during the 1980s debate, but also a later theory of Jonathan H. Turner (2010c). In my opinion, however, the ideas of Turner are a product of this debate and might be recognized, next to Ritzer’s (2010) work, as a kind of its summary.
The reason why I have not discussed network theorists, who have already been involved in the debates on micro-macro in the discussed period, is that I do this in a separate chapter devoted to social networks in the latter part of my work (Chapter 3). It is the same in the case of social field theory – initiated by works of Pierre Bourdieu already in the 1970s and by the institutional sociology of organizations in the 1980s – which I discuss in Chapter 2. Both social network theory and social field theory were already mentioned in the 1980s debate, but they were far from its mainstream. I devote to these two approaches separate chapters, because I believe they represent the main change that took place in sociological theory since the 1980s debate: social networks and social field are now key concepts of sociological mainstream theory, and this is where it is now necessary to look for hints on how to understand the connections between the micro- and macro-level social processes. ← 52 | 53 →
In the summary of this chapter, there is one more thing that needs to be mentioned about the relations between the micro-macro pairing and the individual-society and agency-structure pairings in sociological theory: it is the dynamics of the development of sociological theorizing. The classics were preoccupied with the individual-society pairing. They wanted to understand how these two realities influence each other. Later, after economists with their division to micro-economics and macro-economics influenced the way sociologists started to categorize levels of their analysis, the center of the debate was on the micro-macro dichotomy. The 1980s were the peak of the interest in micro and macro, but already at this time many theorists focused on tackling the agency-structure pairing. I disagree with Ritzer (2010), who claimed that American sociologists were mostly interested in the micro-macro pairing while the European sociologists focused on the agency-structure pairing. Both American and European sociologists were equally active in dealing with the micro-macro issue, as well as the agency-structure issue. Similarly, as it is possible to interpret classical thought on the individual-society pairing in terms of the micro-macro pairing, it is now possible to interpret the still “hot” sociological debate on agency-structure in terms of the micro-macro pairing. The interest in the individual-society pairing would now seem outdated, which is a sign of development in the theory. I claim that shifts in focus on various pairings are a good base for a valuable periodization of the development of sociological theory: first, the theory was interested in individual-society as two realities; then, it was mostly concerned with micro-macro as two levels of analysis; now, the central issue is agency-structure. There is still a quite large theoretical production on the micro-macro link, but I believe this divide has already been reconciled. The approaches to linking micro with macro (sometimes using meso-level of analysis) are in the mainstream of the theory, and both micro-chauvinists and macro-chauvinists (to use Turner’s terminology) are on its margins.
The key objective of the present book is to understand the problem of the sociological vacuum and for this reason, it is still necessary to present and summarize discussions on levels of analysis. I believe that the proper understanding of relations between various (micro, macro and maybe meso) levels of analysis will help in a deeper discussion of the empirical findings, which are often interpreted with reference to the sociological vacuum conception. ← 53 | 54 →
The field theory is praised to be the one of the most important theoretical contributions made in sociology in the past 40 years (Kluttz, Fligstein 2016: 202).14 The social field is a meso-level social order, which means that this unit of analysis is not on a macro-level where the forces behind social processes operate independently of social actors, and it is not on a micro-level, because it does not help to understand preferences and motivations of given single actors. Since the field is a social order, it provides stability for actors (participants of a given field). Yet, it is also an arena of social conflict and rivalization over resources – there is an ongoing game (according to the rules governing a given field) over the stakes defined for a given field. The field is a relation space, where actors orient to each other and, more or less, share the same understanding of what is going on in a given field – even though important conflicts occurring in fields concern defining the legitimate understandings of what is going on and of field boundaries (who is a member and who is not) (Kluttz, Fligstein 2016; Wooten, Hoffman 2008).
The concept of field is believed to be crucial for understanding social action that always occurs in a somehow defined social arena in which actors take each other into account (Kluttz, Fligstein 2016: 186). Without the institutional context provided by social field, it is impossible to understand action. Earlier social theories were focusing either on the micro-context of action, such as situation ← 54 | 55 → or interaction, or macro-drivers behind the action of individuals, such as shared norms, attitudes, or identities. Yet, action always occurs in a specific institutional setting defined by wider social space. To quote John Levi Martin’s (2011: 309) interpretation of the famous experiment conducted by Milgram (1974), people under research were not driven in their action by the norm they had shared: “don’t torture the innocent.” Additionally, they were not simply carrying out orders (like in behavioral schema of stimulus-reaction), but conforming to the cultural frame of acting according to the authority of a scientist. Thus, one may understand their actions only in the context of a social arena defined by the field of science in modern society.
Fields are game-like arenas (Martin 2011; Kluttz, Fligstein 2016), which means that field participants share some kind of understanding on what the rules and stakes of the game are and who plays it. Thus, in relation to the famous Douglass C. North’s (1990: 3) definition of institutions as “rules of the game in a society or, more formally, […] the humanly devised constraints that shape human interaction,” it might be said that the field is a space were the game is played by the constellations of the players.
Authors who were studying the archeology of the field concept, like John Levi Martin (2011), derive its origins from the 19th century physics interested in electromagnetism and mechanics of fluids. In social sciences, first similar conceptions were developed in German Gestalt theory in psychology. According to W. Richard Scott (2014: 202), other important conceptual ancestors of the field concept are “niche space,” used by Chicago school of urban ecology, or “social world,” used in symbolic interactionism. Conventionally, it is Kurt Lewin who is mentioned as the first author to coin the notion of a field in social theory. In Lewin’s (1951) theory inspired by Gestalt psychology, the field is an important tool to understand the psychological environment of a person.
In this chapter, I focus on three approaches to the concept of field important for current sociology. My special focus is on the positioning of a field as a meso-level of analysis. First, I discuss the notion of field in the thought of Pierre Bourdieu. Then, I present the conception of an organizational field, developed in new institutional sociology. Finally, I turn to the theory of strategic action fields recently announced by Neil Fligstein and Doug McAdam. After discussing the three approaches to the concept of field, I present concluding remarks.
“Fields present themselves synchronically as structured spaces of positions (or posts) whose properties depend on their position within these spaces and which ← 55 | 56 → can be analysed independently of the characteristics of their occupants (which are partly determined by them)” (Bourdieu 1993: 72) – this is one of the broadly quoted definitions of fields made by Bourdieu. Yet, when discussing his approach to social fields, which influenced very much the way of thinking about meso-level social structures, it is important to keep in mind that Bourdieu was abstaining from formulating intentional definitions, as he believed that doing so was a characteristic of positivisms in social sciences. According to Bourdieu’s way of sociological reasoning, notions should be understood in the empirical context, rather than clearly defined.
For Bourdieu, the field is an indispensable level of analysis to mediate between the larger social structures and individuals. He explained this issue in the following passage: “the sociology of art or literature that directly relates works of art to the producers’ or clients’ position in social space (their social class) without considering their position in the field of production (a ‘reduction’ which is, strictly, only valid for ‘naïve’ artists) sweeps aside everything that the work owes to the field and its history – that is to say, precisely that which makes it a work of art, or science, or philosophy” (Bourdieu 1993: 75). The quote shows that sociology which attempts to build explanations on the basis of correlations of variables positioning individuals in the society (like social class, or level of education) is, according to Bourdieu, mistaken. In his view, specific social roles (i.e. artist, scientist) and objects with their symbolic meaning (i.e. a piece of art, or a scientific discovery) are produced by social fields. Without this mediating (meso-level) structure and its stakes, games, logics, and power relations, analysts and interpreters of social reality are lost.
In order to function, a field requires stakes and people who are willing to play the game of a given field and have a formed habitus of perceiving the achievement of stakes of a field to be in their interest. These people (and sometimes broader collectives) are unequal in power. Some of them are challengers to the field’s status quo, and some of them hold dominant positions, and it is in their interest to maintain their hegemony. Thus, the structure of a field is actually a structure of power relations of players engaged in field struggles. The power is manifested in a form of symbolic capital, specific to a given field. The so-called “triad” or “trio” (Emirbayer, Johnson 2008; Thomson 2012) of concepts, which enables Bourdieu to analyze social processes, is actually allowing to connect phenomena of a different analytical level: habitus is a characteristic of an individual; capital is a characteristic of an individual but it is visible, useful, and salient only when there are relations between at least two individuals. Finally, the field is a context, but also an overall structure, of relations between individuals. Society for Bourdieu is in fact ← 56 | 57 → a constellation of relatively autonomous fields (Bourdieu, Wacquant 1992: 17), which allows to say that in his theory they are on a meso-level: below the level of society, but above the level of an individual or the level of a relation. Yet, the field is not just a level of analysis, but it is the only proper level of analysis. Society is for him an empty notion, and an individual is meaningless without the field which provides meaning to the roles played by an individual. Thus, for Bourdieu the field is a true object of social sciences (Bourdieu, Wacquant 1992: 107).
The metaphor of the game is one way to understand the social field, but another inspiration comes from natural sciences: it is the magnetic field in which forces influence positions and trajectories of particles. The use of the metaphor of a field of force, reveals that explaining the behavior of an individual (a particle) is impossible without the broader context of field forces. The interactions between individuals – clashes of particles – are not a sufficiently broad context to understand them. Martin (2011) explains this advancement in theory also using the analogy to natural sciences and showing that field theory allows to understand actions of individuals in other way than mechanist approaches allow. In mechanist approaches, actors need to interact with each other (particles need to collide) in order to act, while in field approaches the field force (relating to other actors, but not necessarily interacting with them) triggers movement. According to Bourdieu, the forces active in the field are the ones which define the capital specific for that field (Bourdieu, Wacquant 1992: 101). Knowing what form of capital gives what kind of advantage over other members of a given field is an impulse towards actions, which – in consequence – affect other participants of the field. This is the reason why the magnetic field metaphor illustrates so well the relational character of field processes. There is no need for interaction because relating (sometimes imagined) constituted by comparisons of capitals impacts individuals.
The consequence of introducing the notion of social field to the theory is that it brings the attention to that which is “invisible.” Thanks to the notion of the field, the analysis does not have to be limited to entities visible at first sight, such as individuals, groups, or interactions. The forces of the field cannot be always simply perceived but, according to Bourdieu, they are present there (Bourdieu, Wacquant 1992: 96–97). These are the imagined relations and motivations constructed by the field structure that drive actions. Apparently, such line of theorizing creates the risk of bringing in metaphysics, yet the intention of Bourdieu was to analyze the forces obliging actors to act in a given way reflecting their position in the field.
The interplay of visibility and invisibility, or the mechanistic and electromagnetic metaphors, bring to the issue of the objectivism – subjectivism debate in sociological theory. According to one of the reconstructions of the concept of ← 57 | 58 → social field, it is “the objective network or configuration of relations (again structuring and structured) to be found in any social space or particular context” (Grenfell 2012: 47). Bourdieu was attempting to break with the objective – subjective dualism. He was mostly influenced by structuralism (the objectivist inspiration) and existentialism (the subjectivist inspiration). The concept of field is said to be the objective (structuralist inspiration) side of Bourdieu’s theory, while habitus is said to be the subjective side. Commentators stress that these two concepts cannot be separated. Yet, for many scholars inspired by Bourdieu, only the concept of field remained salient. This is especially the case of researchers studying collective actors, of whose habitus it is difficult to talk about.15
Bourdieu’s concept of field has several shortcomings, some of which it shares with his general works: murkiness and lack of clarity covered with sophisticated, ornate, and sometimes even bizarre language. He was purposively using open concepts and claiming that the proper understanding of his concepts is possible only in the context of the whole theoretical system they constitute (Bourdieu, Wacquant 1992: 96). Some of the problems regarding the conception of a field come from the fact that Bourdieu did not build a grand theory of a field, but he was rather using it to solve practical problems and pursue empirical operations (Thomson 2012: 79). Yet, it is important to remember that Bourdieu (1993: 72) claimed that there is a need for such general theory and that “general laws of fields” do exist. Most often, the critique of Bourdieu’s understanding of the concept of field regards the following aspects: blurriness of borders; relations between fields; too many fields or focus on just one field; change in fields; architecture of fields; actors’ focus on their own position; individuals as actors in fields.
The borders of a given field are rarely clearly visible. Although Bourdieu was convinced that fields are objective reality, he was pointing that one of the games which is played in all fields concerns defining the field’s borders. At least for this reason it is never certain who is to be included in the field as its participant. Some of the more powerful actors might be claiming that other less powerful actors are outside of a given field. This exactly might the point of a boundary game in a given field: that outsiders are challenging to be recognized as members of a field. The problem is of a theoretical nature as Bourdieu stressed the contention of fields and their antagonistic character (Bourdieu, Wacquant 1992: 103). It is also a very important practical problem for researchers, who need to decide which objects and ← 58 | 59 → subjects select for observation. Bourdieu’s answer would be that proper research ought to be long-lasting, preferably with an ethnographic component, so eventually the researcher will “know for sure” what the exact borders of the field are.
Bourdieu underthorized relations between fields (Fligstein, McAdam 2012: 26). He often mentioned them by pointing that the field of education is crucial for other fields because it distributes symbolic capital, or that the field of power is the field of fields, and that all other fields are subordinated to it. Yet, Bourdieu did not provide too many hints on how relations between fields should be studied, and he did not theorize about these relations, confining himself to confirming that the question of inter-field relations is a difficult one (Bourdieu, Wacquant 1992: 109). This seems to be a crucial omission in his conception. One of the hints he gave about relations between fields is to study the transfer of symbolic capital between fields (Bourdieu 1993: 73). In other approaches to the field theory (discussed below) the interdependence of fields is a very important and often addressed issue.
The problem of undertheorized relations between fields is, in my opinion, connected to the issue labeled by Patricia Thomson (2012: 77) as “the problem of too many fields.” Bourdieu often treated any context of social interaction or power games as a field, which made it possible to have at some point together the broad field of power, specific field under study, and a collective actor in that field also analyzed as a field (i.e. a given university as a field in the field of science related to the field of power). This problem is connected to the tension between understanding a field as a real entity which objectively exists (Bourdieu supported this view), and understanding a field as a heuristic which helps in studying the context of action. The solution usually applied by Bourdieu was to focus on one field at a time. Yet, the decision on how to frame an object of study always has its consequences. I claim that the focus on one field, although useful for explaining some important aspects of its internal dynamics, in consequence diminishes the importance of inter-field effects. If the field is considered as a key meso-level of analysis mediating between level of an actor and a level of society, then it is necessary to keep in mind that society (whatever it means) is filled with many overlapping or related fields. By “overlapping,” I mean a situation in which one actor partakes in two fields, and by “relating,” I mean a situation in which occurrences in one field have consequences for the occurrences in the other.
Although Bourdieu focused his attention on struggles in fields as arenas of conflict (his omission of cooperation in fields will be discussed below), his approach is often criticized for an undertheorization of change in the fields. The problem seems to be paradoxical as fields studied by Bourdieu are usually boiling ← 59 | 60 → with contention and some of the fields he studied were the emerging ones – for example, the field of literature (Bourdieu 1996). Yet, this contention and never-ending antagonism within a given field often leads to the reproduction of a field structure of positions. Changes in language and symbols are usually covering the lack of changes in relations of dominance. Therefore, the accusation of determinism is quite accurate. At the same time, the undertheorization of change in fields stems from the focus on intra-field processes, while – as I will demonstrate in the latter parts of the chapter on other approaches to field theory – many structural changes within fields are caused by their relations with other fields. This brings back to the problem discussed in the previous paragraph, and allows to say that the problems with grasping change in fields are, to a large extent, caused by the undertheorization of field relations in Bourdieu’s conception.
As Fligstein and McAdam (2012: 24) noticed, Bourdieu “has little to say about the architecture of fields beyond the general view that they contain positions that are structured by the relative power of actors.” Quite often the narration on fields in Bourdieu’s conceptualization is limited to the story about the conflict between orthodoxies and heterodoxies, and the bi-polar oppositions in field structures (certainly the inspiration of structural theory). Yet, there are certainly more types of positions in the field than the dominating and dominated ones. Fields contain specific niches and actors of special status, who may not be equipped with much power but, for example, their uniqueness is indispensable for the field, or their structural position allows them to bridge different kind of actors. Finally, in many fields there are special actors to whom other actors mandate coordinative prerogatives. This idea is going to be developed in the latter section on the strategic action fields theory.
Therefore, Bourdieu is mostly interested in conflict within fields (the issue of conflict and cooperation is going to be developed in Section 2.5) and the incessantly struggling actors “are generally only responsible to themselves and motivated by a desire to advance their interests” (Fligstein, McAdam 2012: 25). Thus, there is no place for coordinated action in fields. Actors fight for making their position better and even if Bourdieu talks about orthodoxies fighting against heterodoxies, it is rather an invisible hand mechanism which creates the impression that these two groups of position holders coordinate their actions against each other. Fields are fields of individual, self-oriented particles.
Another shortcoming of Bourdieu’s conception of social field is that it is tailored to study individual – and not collective – actors. Although Bourdieu was mentioning that collective actors are also members of fields (he referred to them as institutions), it is easy to see that his approach was not including collective ← 60 | 61 → actors. In fact, any collective actor (an organization or social group, and especially a social movement) was nearly automatically treated by him as another social field. There is nothing wrong in this perspective, however, in Bourdieu’s approach the notion of habitus is tangled with the notion of field, and habitus is strictly a disposition of a human individual. I cannot imagine a serious transposition of habitus from individual actors to collective actors. Bourdieu’s fields are spaces connecting people and shaping their biographies, but they are not very helpful in explaining organizational processes.
W. Richard Scott (2014: 106) distinguished six possible levels of institutional analysis of organizations. Starting from the one closest to the human individual these are: organizational subsystem, organization, organizational population, organizational field, societal level (society), and world system. Obviously, the level of organization was initially the most interesting for the study of organizations. As I have demonstrated in Chapter 1, for some sociologists, organization was a sufficient meso-level of analysis allowing to understand the relation between an individual and the whole society. Yet, for the sociology of organizations the important topic of studies are phenomena which occur not only inside a single organization but go beyond it, thus being located “somewhere between” organizations (Pawlak, Srokowski 2014). As for the old problem of sociology, another “obvious” level of analysis was the societal level, important for institutional studies of organizations because of the influence from the political institutionalism for which the study of nation-state institutions is crucial (March, Olsen 1989). Yet, this macro-level of analysis was not sufficient for understanding organizations. Society is “too far” from organizations as too far from them are other macro structures: sectoral or transnational ones.
Another quite “obvious” level of analysis was the level of population. If it is recognized that there exist organizations of certain types (i.e. schools, hospitals, car factories), it is possible that the study of populations of organizations of a given type will bring knowledge about these organizations as such. Michael T. Hannan and John Freeman (1977: 934) defined population as an aggregate of organizations that “must be alike in some respect, that is, they must have some unit character.” They stressed that populations are “abstractions useful for theoretical purposes” (Hannan, Freeman 1977: 934). It is the researcher who for his or her analytical purposes defines population using the criterion of unit character. Trade unions, political parties, or newspaper companies are apparent examples of organizational populations, but the unit character might be also of a hidden ← 61 | 62 → type, not perceived by members of organizations who are being researched. The work of Hannan and Freeman were inspired by population biology and ecology and aimed to build a model of analysis in order to better understand the competition and selection processes among organizations. Yet, what this approach lacks, is the good conceptualization of relations with organizations belonging to other species. Organizations do not only compete with ones similar to them, but they also relate to organizations which are very much different from them.
The concept of the organizational field allows to better understand the relations between different kinds of organizations – this seems to be the key to the success of this concept. The notion of organizational field was introduced by Paul J. DiMaggio and Walter W. Powell in seminal paper “The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields” (1983). The authors were trying to solve paradoxical puzzle: why organizations become increasingly similar as they try to change themselves? Common sense suggests that change should cause increasing dissimilarities and complexity among organizations. DiMaggio and Powell distinguished three mechanisms of institutional isomorphic change: coercive, mimetic, and normative. In order to grasp the processes resulting in organizational homogeneity there was a need to study organizations in a perspective allowing to see how they relate to each other. DiMaggio and Powell defined the organizational field as “those organizations that, in the aggregate, constitute a recognized area of institutional life: key suppliers, resource and product consumers, regulatory agencies, and other organizations that produce similar services or products” (DiMaggio, Powell 1983: 148). As an advantage over other attempts to grasp the meso-level unit of analysis, the authors highlighted that field directs attention to the totality of relevant actors and not only competitors. The organizational field could be regarded as a tool to understand a given organization in the context of its environment, but it quickly became an insightful level of analysis by itself (Scott 2014: 223).
The three types of institutional isomorphism introduced by DiMaggio and Powell (1983) go hand in hand with three pillars of institutions – the conceptualization of institutions developed by Scott (2014). The coercive isomorphism reflects the regulative pillar by which institutions are manifested in rules and codes regulating actions; the mimetic isomorphism reflects the cognitive pillar by which institutions shape the way people perceive the world outside; and the normative isomorphism reflects the normative pillar connected with moral valuations of actions. It is important to remember that Scott’s pillar distinction is an analytical one – it helps to understand the various modes by which social life is influenced by institutions. In a similar way, the three types of isomorphism are ideal types. ← 62 | 63 →
Here I intend to contribute to the development of a field theory by theorizing how different types of isomorphism connect organizations with entities from various levels of analysis: coercive isomorphism connects organizations with state and the macro-level; mimetic isomorphism connects organizations with other organizations and the meso-level; and normative isomorphism connects organizations with individuals and the micro-level.
Coercive isomorphism which stems from the actors’ preoccupation with legitimacy (DiMaggio, Powell 1983: 150) is the mechanism by which powerful actors put pressure on organizations to achieve their conformity. In many fields, the powerful actors who are capable of setting or executing regulations are state agencies.16 Clearly, in this branch of theorizing about social processes the state is not conceptualized as a single actor. State is rather described as a constellation of organizational actors: ministries, agencies, and other bodies which often compete for resources in a given field. Still, they are enabled with unique powers to coerce other organizations. I would, therefore, say that the focus on coercive isomorphism is a good way of exploring relations of organizations with the state or other huge, powerful, and complex organizations which operate on a macro-level. In this understanding, the macro-level equals societal level. Simultaneously, it helps to unpack the concept of state and other macro-actors, because the empirical studies of given fields show that the state with which actors interact or which is able to coerce them is not some kind of a far abstraction, but it manifests itself in concrete organizational units, i.e. taxation authority or immigration bureau.
The mimetic isomorphism, which is an effect of organizations’ search for successful models in conditions of uncertainty (DiMaggio, Powell 1983: 151), connects organization with other organizations. It is a phenomenon occurring strictly on the meso-level of analysis. Modeling sometimes occurs when organizations interact and thus become in direct contact with each other. Yet, very often modeling requires some kind of mediation. Models which are to be mimicked are distant, and in order to follow them, organizations use simplified images of models. In the process of diffusion, a model is translated to the local ← 63 | 64 → context after a travel of ideas (Czarniawska, Joerges 1996; Djelic 2008). In a way, boundaries of a field are defined by a range of a potential models to which uncertain organizations relate to. Thus, mimetism seems to be a phenomenon from the meso-level of analysis.
Normative isomorphism stems from the fact that members of professions are present in nearly all aspects of organizational life (DiMaggio, Powell 1983: 152). Furthermore, the salience of professions and professionals in the new institutional organization theory is perceived as growing. Scott (2008) announced that professionals are “lords of the dance” and that they are the most important agents of institutional change. Bromley and Meyer (2015) show that the global increase of organizations is strongly correlated with the growth of higher education and occupations – which are professions or aspire to be professions – that require diplomas. The authors particularly point to the fact that the occupation of a manager is becoming professionalized. Professionals are members of organizations whose important normative orientations are shaped by professions crosscutting various organizations and often various organizational fields. Professions are also a special kind of meso-level structures which could be analyzed as social fields. Thus, professionals as individuals put pressure from the inside of organizations to conform to their normative orientations which are shaped because of belonging to a profession. DiMaggio and Powell (1983) pointed to two most important factors contributing to forming normative pressure: professionals are, to a large extent, products of educational institutions which shape their norms; and professionals are usually members (sometimes obligatory) of professional associations and other kinds of networks. Normative isomorphism in this way connects organizations with other structures through individuals which, on the micro-level, in their everyday actions, bring in the norms shaped in other contexts.
The organizational field is a space where the state and general social pressures are present, but it also allows to understand actors. In the new institutional theory of organizations, the focus is on organizations as actors, but also other collective actors are recognized as field members, i.e. more loosely organized social movements or other social groups. Yet, organization is here a crucial unit of analysis. The assumption underlying the new institutional theory of organizations is that nearly all social process in the late modern world are happening in or through organizations, and organizations as a cultural form of collective action are globally expanding (Bromley, Meyer 2015).
The organizational field in this approach is a real entity. It is not just an analytical tool or a construct, but it exists, which allows to talk about the ontology of fields. They exist “to the extent that they are institutionally defined” (DiMaggio, ← 64 | 65 → Powell 1983: 148). Thus, the ontological status of fields is similar to the status of institutions. Fields bind organizations but are clearly an institutional phenomenon. The new institutional theory of organizations is in its assumptions based on social constructivism of Berger and Luckmann (1967), so both fields and institutions exist as externalizations of people’s cognition and actions. The field, in order to emerge, requires: an increased interaction between organizations; developed structures of domination and patterns of coalition; an overwhelming information load to be processed by organizations; a mutual awareness of organizations involved in “something” they share in common. In short, “one actor takes note of another and through this process of referencing one another, actors bring a field into existence” (Wooten, Hoffman 2008: 138). The field is a space of structuration process described by Anthony Giddens (1979).
As Wooten and Hoffman (2008) stressed, after the focus on isomorphic pressures resulting in relative homogeneity of organizations within the field, institutional scholars turned to study conflicts and other forms of field dynamics. The change on the field level became (before it was a change on the organization level triggered by the field level processes) a research subject of growing interest. To simplify, the causes of the change could be found in a field – the results of endogenous institutional processes; or the causes might be external to a field – like exogenous shocks destabilizing the relations between the actors in a field (Greenwood, Suddaby, Hinings 2002). Subsequently, another aspect of conflictual character of organizational fields started to be emphasized – namely, the aspect of power and dominance, so the situation of fields seemingly being settled and quite started to be explained with the work of maintenance done by powerful actors (Lounsbury, Glynn 2001). Consequently, more focus was set by new institutionalism on the heterogeneity within fields than on homogeneity. A growing body of research dealt with issues like conflicting institutional logics present in fields (Friedland, Alford 1991; Thornton, Ocasio, Lounsbury 2012), which resulted in production of hybrid organizations (Pache, Santos 2013) or the development of the ability called “institutional ambidexterity” (Jarzabkowski et al. 2013). My intention here is not to go deep in considerations of various understandings of organizational field concept in new institutionalism, but rather to show that there is an on-going theoretical discussion on the nature of fields, which is very thoroughly tested in research. New institutional conceptualization of fields (and a broader new institutional theory of organizations) is often presented as a study of isomorphism, conformity, and stability – yet it is an oversimplification usually caused by the sin of describing the new institutionalism only on the basis of its founding texts from the break of 1970s and 1980s. This sin of ← 65 | 66 → oversimplifying and not noticing the newer developments in new institutional theory is also committed by proponents of the strategic action field theory (Fligstein, McAdam 2012; Kluttz, Fligstein 2016) described below, in Section 2.4.
The belief that fields exist as a real thing is in some way shared with Bourdieu’s approach, although he underlines their objective ontological status. In the new institutional theory of organizations fields exist, but their ontological status is of an intersubjective nature. Yet, there are some researchers who would say that they use organizational fields rather as an analytical tool, and that they do not perceive fields as something real but rather a heuristic thanks to which it is easier to grasp social phenomena occurring on the meso-level. Thus, the field is then an analytical tool which helps to understand what is going on between actors constellating around a given issue which brings them more or less together (Hoffman 1999).
A good way to observe the transformations of new institutional theory of organizations is to check whether there are any definitional changes in W. Richard Scott’s Institutions and Organizations (Scott 1995; 2014) book, which is upgraded every few years and summerizes the most recent developments in theory. In 1995, he defined the organizational field as “a community of organizations that partakes of a common meaning system and whose participants interact more frequently and fatefully with one another than actors outside the field” (Scott 1995: 56). Yet, in 2014 the definition was changed to “a collection of diverse, interdependent organizations that participate in a common meaning system” (Scott 2014: 106). The changes of definition are significant and are a good illustration of the changes in the new institutional theory of organizations in the last 20 years.
Firstly, Scott replaces “a community of organizations” with “a collection of diverse interdependent organizations.” “Community” is a much stronger term denoting that a category called with this name has a kind of a common identity or “spirit.” “Collection” simply means that in a field there are organizations, but there is nothing that binds them. “Community” was replaced by “collection,” yet a description was added that these organizations are “interdependent,” which means that they are in some kind of relations, but they do not create one “body.” It was also stressed that the organizations in a field are “diverse,” which reminds about the differences between the notion of a field and the previously discussed notion of population. It also highlights the heterogeneity of fields and possible divides present in the fields. The previous use of the term “community” was obscuring the divides and rather stressing what they had in common. ← 66 | 67 →
Secondly, Scott deleted the line “whose participants interact more frequently and fatefully with one another than actors outside the field” (Scott 1995: 56). The reason for this – according to my interpretation – is that new institutional scholars realized that the crucial relation in fields is not interaction but orientation to each other. Actors who orient themselves to each other do not have to enter exchanges, conflicts, or cooperations, but they are somehow aware of their existence. The newer definition includes nothing about more frequent interactions with other actors from the field because, according to a plethora of studies, interactions between actors from various fields are quite common. Although usually studies focus on a concrete field, the scholars’ awareness of inter-field relations in the recent years has been growing.
What remained in Scott’s definition is the “common meaning system,” which means that although replacing “community” with “collection,” the theorist highlights that a field includes some shared cultural elements. Even if the field is a sphere of a conflict, its participants more-or-less agree on what the stakes of this conflict are.
In the new institutional theory of organizations, the field is one of the most central and vital terms (Scott 2014: 219; Wooten, Hoffman 2008: 142). This approach explains organizations with external processes, not with what occurs inside organizations. Thus, space, arena, industry, population, or the field are levels of analysis which seems to be indispensable. The new institutional thinking about fields was, to some extent, inspired by Bourdieu’s theory, but it developed footed in empirical studies of inter-organizational phenomena. As the new institutionalism in organizational theory has recently increased its interest in agency on expense on the interest in conformity, it also increased its interest in fields understood not only as spaces of isomorphism but as spaces bringing meaning for organizational action.
Neil Fligstein and Doug McAdam (2012) crafted together their integrated theory of fields under the label of strategic action field theory. Their goal was not only to develop the theoretical conceptualization of the meso-level but also to reintegrate sociology, which they perceived as too fragmented. In their approach they bring together the advancements of economic sociology and new institutionalism (Fligstein’s field of expertise), as well as social movements (McAdam’s field of expertise). According to them, a strategic action field is “a constructed meso-level social order in which actors (who can be individual or collective) are ← 67 | 68 → attuned to and interact with one another on the basis of shared (which is not to say consensual) understandings about the purposes of the fields, relationships to others in the field (including who has power and why), and the rules governing legitimate action in the field” (Fligstein, McAdam 2012: 9).
According to Fligstein and McAdam, the search for the meso-level of social order in many different sub-fields of fragmentized sociology has been visible at least since the 1990s. As the most important paradigms in which this search appeared, they enlisted various branches of new institutionalism, social movements literature, and network analysis. Still, the most influential inspiration for strategic action fields theory was Bourdieu’s conception of social fields – discussed above – with critical elements borrowed also from Fligstein and McAdam. The first distinctive element of strategic action field theory is that a whole theory is constructed around the concept of a field – it is a theory of fields. For other approaches fields are very important concepts, but they are rather coined in order to be used to achieve some other theoretical goals such as understanding forces shaping trajectories of individuals (in case of Bourdieu) or understanding the environment of an organization (in case of new institutional theory of organizations). For both of these approaches, the fields are indispensable and central notions, yet they are not theories of fields – fields are helpful to theorize about habitus or institution.
Fligstein and McAdam accuse new institutionalism of understanding social fields as static arenas of reproduction of institutions. It is also a common critique of strategic action fields theory that it presents Bourdieu as a theorist of reproduction unable to explain change, which is not entirely true (Swartz 2014). Indeed, Fligstein and McAdam focus on change and stability as outcomes of strategic action, yet they exaggerate in diminishing the salience of focus on change in other theories. In brief, the strategic action field theory is an attempt to synthesize new institutionalism (presented as a very good theory of stability, but lacking theory of change and emergence) with social movements theory (as a very good theory of change, but lacking theory of durability).
The actors, according to the assumptions of Fligstein and McAdam, are strategic. The fields deliver a structure in which the actors are embedded, but the actors are not determined by this structure. The actors play an on-going game of making their position better. This continuous activity is called jockeying and makes other actors “to interpret them, consider their options, and act in response” (Fligstein, McAdam 2012: 12). Thus, jockeying eventually creates a huge internal pressure on fields to change, but it may also result in reproduction – Anthony Giddens and the theory of structuration is also pointed by Fligstein and McAdam as an important inspiration. ← 68 | 69 →
Unlike in the rational actor theory or the Marx-inspired critical approach of Bourdieu, in the strategic action fields theory, actors are not only driven by their interests but also by the urge to find meaning and identity. It lies at the microfundations of Fligstein and McAdam’s theory that social activities also have an existential function for human beings. According to the authors, existential fear and uncertainty are elements of human nature, and it is also natural for humans to seek refuge from these fears by engaging in collectives. According to Max Weber (1978), the source of all social life is the ability of people to understand one another and, according to George H. Mead (1934), crucial social mechanisms provide people with a sense of self and the ability to take the role of the “other.” So for the strategic action fields theory the issues of meaning, identity, and self are not only characteristics of individuals, but they enable collective action and cooperation. According to Fligstein and McAdam’s approach, the homo economicus is a wrong model of a social actor as it treats the urge of belonging as something irrational and unexplainable, while it is one of the factors that social theory needs to treat seriously in order to understand phenomena on a collective level. As Fligstein and McAdam (2012: 43) claim, “any adequate theory of human strategic action must take this mix of instrumental and existential motives into account.” The focus on existential function of the social, leads authors to the notion of “social skill” as something important for understanding the co-operative aspects of field processes.
Social skill is defined by the Fligstein and McAdam (2012: 46) as “the ability to induce cooperation by appealing to and helping to create shared meanings and collective identities.”17 It enables actors to see their field from the perspective of other actors and thanks to that empathy it is a great tool for building coalitions of cooperation which are capable of transforming social fields. Yet, social skill is also indispensable for sustaining fields. Fligstein and McAdam highlight that the theory of social skill is in strong relation with the theory of social field. As they argue, it is capable of solving the tension between determinism (actors driven by their position in social structure) and overly agentic approach in which actors are capable of creating and re-creating society by their wish: both the skills and position influence the actors’ ability to cooperate and compete. Social skill is a concept merging insights from the social movements theory (i.e. socially skilled actor is able to use opportunities and to frame situations) and from neoinstitutional theory of institutional entrepreneurs (Battilana, Leca, Boxenbaum 2009; DiMaggio 1988). However, in my opinion, this element of theory remains rather ← 69 | 70 → blurry and underdeveloped. Social skill seems to be another attempt to grasp what Weber (1978) called “charisma.”
Another novelty in field theory brought by Fligstein and McAdam (2012: 77) is the concept of internal governance unit, which they define as “organizations or associations within the field whose sole job it is to ensure the routine stability and order to the strategic action field.” Internal governance units are also usually dealing with relations between fields (i.e. associations of certain occupational categories represent them in relations with the state or other fields). Internal governance units may serve field members in relations with administration, information, regulation, enforcement, or certification. Yet, usually they are controlled, or at least very much influenced, by field incumbents and serve them as a tool of domination towards challengers.
Fligstein and McAdam (2012: 58) accuse former conceptualizations of social fields of field-centric bias and theorize about the embeddedness of strategic action fields, which are often connected in a manner similar to Russian dolls. The fields remain in different kinds of relations, which of course change over time: they might be unconnected, dependent, or interdependent. The factors shaping these links are as follows: resource dependence, mutual beneficial interactions, sharing of power, information flows, or legitimacy. State is also perceived by strategic action field theory as a field, and not as one unified actor. The authors also highlight that “formal organizations are often the central players in strategic action fields” (Fligstein, McAdam 2012: 64), but they do not limit membership in fields to organizations and show that it is possible to look at organizations also as fields (then the fields are nested in each other). This is the case of the state, whose agencies are present in virtually all social fields.
According to the strategic action field theory, an increasing focus should be put on inter-field relations, as recently it has been possible to observe three related macro-social processes: proliferation of fields, growth of higher education, and the development of professions. New fields usually require new professions (or occupational categories of ambition to become professions) and these are produced by higher education entities. The supply and demand for professionals are in feedback relation.18
To sum up, the strategic action fields theory is a synthesis of insights from the thought of Pierre Bourdieu, new institutional sociology of organizations, and social movement theory. It puts the concept of field in the center of theoretical attention, which allows to grasp meso-level social phenomena. ← 70 | 71 → It develops the understanding of relations between various fields, theorizes the relation between actor and field, and claims to contribute to better understanding of change in fields. The strategic action field theory attempts to solve the biases of the two main traditions of the notion of field uses: bias toward conflict (attributed to Bourdieu) and bias toward conformity (attributed to new institutionalism).
In this chapter I have presented the most influential approaches in current sociological theory to the field concept. In discussing Pierre Bourdieu’s blurry – but inspiring – use of social field notion, the heterogeneous and vivid understanding of organizational field conception in new institutional studies of organizations, and the synthetizing theory of Neil Fligstein and Doug McAdam I was focusing mostly on how the concept of field is useful in tying the micro- and macro-levels of analysis. Indeed, I agree with Kluttz and Fligstein (2016: 202) that the field concept is the most important contribution to the sociological theory made in the recent decades. It is so because fields link actions of single actors (individual humans or collectives) and their interactions – the phenomena on the micro-level – with broader societal processes from the macro-level. Still, the field as a meso-level social order creates its own effects and constrains the conditions for social action. The fields are recreated in continuous processes of structuration by social actors who shape fields’ boundaries in their daily practices and never-ending jockeying for a better position. But the relation of fields with actors is of a mutual character and it is the field which provides a social definition of an actor: it is impossible to be an artist without a field of art, or a scientist without a field of science, or it is impossible to run a hotel without and organizational field of hospitality services.
Social field may be a key concept in the studies focused on different phenomena – as in the case of Bourdieu’s interest in power and habitus, or neoinstitutionalism’s interest in institutions and organizations – or the center of sociological theory, as in the case of strategic action fields theory. It brings attention to that which is in between – not only in between the micro- and macro-levels of analysis, but also in between actors.
For a long time, many sociologists understood their discipline simply as a study of society. It is the case of Polish sociology preoccupied with explaining Polish society to itself as a whole (Bucholc 2016) and dealing with macro-processes such as transformation after the communist regime (Pawlak 2013). More attention to the field level brings a better understanding of action and ← 71 | 72 → actors. As scholars working with the notion of field proved, understanding social action only with the knowledge on social categories and macro-processes is always fallacious. Yet, the program of explaining a given society could be saved for national sociologies, thanks to the field theory. A synthesis of case studies of various social fields present in a given country is, in my opinion, a better strategy for drawing a picture of this country’s society than the attempts to provide a grand-sociology of this society from the macro-perspective. Many macro-processes are visible in various fields, yet they occur in their own specific ways, and this meso-perspective may provide a better understanding of what the transformation from the communist regime or Europeanization actually are.
The growing knowledge on various fields, acquired thanks to empirical studies conducted either on the field level or using the concept of the field to understand phenomena from other levels of analysis, allows to build hypotheses (or research questions) for future studies of other fields. I would restrain myself from talking about laws of fields, but theorizing about results of various studies allows researchers to ask questions about what is likely to be at stake in a field which has previously been not studied. In this sense, it is possible to talk about a theory of fields because it provides research questions for studies of phenomena – even those which are empirically very distant from the ones already described in the theoretical language of fields.
Finally, social fields are to some extent invisible. Although many social actors are very aware of their relevant social environment, the theory of fields is quite abstract in comparison to many theoretical tools of sociology. It is quite far from the common sense or folk wisdom about society. It appeared quite late in the history of sociological thought. Without this theoretical tool, sociological understanding of what takes place in between the micro-level and the macro-level was obscured. Thus, I claim that the social field theory is one of the best avenues for searching for what could fill the sociological vacuum.
The objective of this chapter is to present and discuss another paradigm of social sciences which delivers promising insights on the micro-macro link: social network analysis (SNA). During the 1980s debate on the micro-macro problem, this approach was not recognized as one of the solutions, although the early work of Ronald S. Burt (1982) was mentioned as one of the attempts of integrating micro and macro sociologies (Ritzer 1990). SNA in the 1980s was already increasing ← 72 | 73 → its influence in social sciences and its most important statements were already formulated, yet it was still not a part of the sociological mainstream. The 1980s discussion was still a debate between scholars interested either in systems or in individuals. Currently, SNA, which is the manifestation of relationist sociology, thrives at the mainstream of sociology. There are probably many reasons for which the metaphor of a network started occupying the minds of people and social theorists. The key reason is, most probably, the expansion of the internet. Theorizing in relational terms is important for the concept of social field discussed in Chapter 2. The SNA is another manifestation of theory grounded in relational assumptions: it focuses more on relations than on substances (Emirbayer 1997). By this virtue, social networks – structures tying individuals with each other – bring a new perspective on the micro-macro problem.
Theorists who were leading the 1980s debate on the micro-macro problem were categorizing the network scholars (such as Harrison White or Ronald Burt) as representatives of macro theories of social structure. I am convinced this was a mistake, and much of this chapter is an argument showing that the SNA is a promising voice in the micro-macro debate, because its interest in social structure starts from the micro-level of relations between actors. Using its quite simple building blocks, SNA contributed to the development of concepts such as “social capital” or “embeddedness.” They are both promising avenues of thinking about the linkage of the micro- and macro-levels. In this chapter I am not going to focus on the notion of social capital: it has also other antecedents than network theories and it is going to be discussed thoroughly in Chapter 7. In this chapter I am leading the narrative to the conclusion that the conception of social embeddedness is a way of linking micro with macro.
The order of this chapter is the following. I start with discussing four network approaches competing with SNA: network exchange theory, Castells’ theory of network society, actor-network theory, and the new science of networks. I present them briefly, with a special focus on their potential for solving the micro-macro problem. Then, I present the development of the SNA and its central assumptions and notions. On this basis, I am able to discuss how theorizing in terms of networks is useful for conceptualizing problems of agency-structure, individual-society and – most importantly – the problem of micro-macro. In the concluding remarks of this chapter, I claim that the conception of embeddedness is an interesting way of linking micro with macro and that SNA is a perspective which allows to notice the significance of the seemingly invisible or unimportant structures which have a great impact on social processes. ← 73 | 74 →
There is no single paradigm of studying networks in social sciences. For this reason, although this chapter is mostly focusing on SNA, in this section I will discuss also other network approaches, such as: network exchange theory, the Manuel Castells’ theory of network society, actor-network theory, and new science of networks (Pietrowicz 2016).
3.2.1 Network exchange theory
The network exchange theory is an approach which has its origins in micro-economy and behavioral psychology, although it was informed also by sociological theories of exchange and anthropology (Molm 2000). In the context of the micro-macro debate it is not a relevant approach because it concentrates on how the structure of network influences the transactions between actors. Most of the studies in this perspective take into account individual actors, although collective actors are also studied – and in this case it is perceived as an analysis on the macro-level (Molm 2000: 261). Thus, although it is not a theory of exchange between two actors, because it adds social networks as a crucial context for exchanges (Emerson 1972), it is still micro-social theory. The most important topics of study of network exchange theory are power relations and dependency in exchanges. In contrast to the SNA, the network exchange theory focuses on networks of negatively connected relations in which actors are linked because they compete over something (Molm 2000: 270). To sum up, it is a theory of micro-structures’ influence on relations between actors. It does not have the ambitions to say something on the macro-level processes, so it does not provide conceptual tools for understanding the relation between micro and macro, however, it provides evidence, well supported in experimental studies, for the importance of network context in exchanges between actors. It is a relational approach according to which structure (consisting of relations) has a primacy over actors. Yet, it needs to be stipulated that attaching a different meaning to macro and perceiving any social structure as a macro-level phenomenon would make the network exchange theory a theory linking micro and macro as it links together individuals and structures. To illustrate, the network exchange theory was recognized by Jonathan H. Turner (2010c: 8) as an example of a theory bridging micro- and macro-levels of analysis, because it focuses on the forms of exchange – not on the actors – involved in exchange. Yet, exchanges are examples of encounters traditionally qualified by the scholars to the micro-level of analysis. ← 74 | 75 →
3.2.2 Manuel Castells’ theory of network society
The theory of network society is an attempt to explain and understand contemporary world which was developed by Manuel Castells, most famous for his trilogy on the information age (Castells 1996; 1997; 1998). The network is a key notion used by the author, who received enormous attention in the late 1990s and at the beginning of the 21st century. His eclectic and essayistic books praising new technologies of communication made Castells an extremely popular public intellectual, and brought the notion of network into the mainstream of sociology (Ritzer 2010). Much of this success is – in my opinion – connected with the demand of that time for a grand theory of possible changes of social life caused by the introduction of the internet. Yet, network is used by Castells as a buzz-word: he sees networks everywhere, but does not analyze them with the methodological discipline specific for other network approaches. If every social structure is a network, it becomes a dubitable notion with no explanatory power (Urry 2003: 11). Networks together with other buzz-words and catchy concepts, such as globalization, social movements, the search for identity, or internet, create altogether a very persuasive read about the contemporary world. Nevertheless, it is very hard to look for solutions to the key theoretical problems of sociology in the work of Castells. Certainly, the concept of a network allows to understand how people are connected in global world. Castells provides some descriptions of condensation of relations in important nodes of global network, such as financial cities of global metropolises. However, a synthesis of his works allows only to say that individuals are somehow connected globally through the networks of communication.
3.2.3 Actor-network theory
Actor-network theory is a perspective developed mostly by French authors interested in science and technology studies. The best known of them is Bruno Latour, but the theory’s foundational works were written also by Michel Callon and others. The theory received much attention because it was recognized as a novel, and even revolutionary, paradigm in social sciences. Network is a key notion in the approach, which is why it is worth to mention actor-network theory in this chapter. What is different, and novel, in this approach is that networks are composed of elements of various ontological status – people, animals, technological objects etc. – and therefore this perspective, at least at the declarative level, goes beyond the study of humans and includes non-humans as well. This makes actor-network theory different from other network approaches, because the latter usually study networks of homogenous nodes, or at least nodes which are individual humans or collectives of humans (groups, organizations etc.). Here the network may consist, ← 75 | 76 → as in the famous paper by Callon (1986), of fishermen of St. Brieuc, scientific colleagues, and scallops of St. Brieuc. In fact, focusing on networks consisting only of social actors has been described by Latour (1993) as a never fulfilled project of modernity in which people are obsessed with making divides between social and natural objects. The other important element of actor-network theory is that it treats networks as actual actors. In other approaches, networks are constructed by the relations between actors. In actor-network theory agency is only achieved by associating various heterogeneous elements in a network, which then becomes capable of agency. The associating is achieved through acts of translation, which transform the objects. Actor-network theory is a strictly relational approach and since it rejects essentialism, there is no room in it for dispositional or systemic explanations. Some would say that in actor-network theory there is no room for any explanations at all (including the relational ones) because it is focused only on description. The most fundamental critique of this approach, which is still very inspiring and mind opening, is that despite the declarations that actor-network theory focuses on what is material, it is simply dealing with human narrations on translating, associating, and disassociating of various elements. Still, as a perspective in discourse analysis it is a quite powerful research approach, as demonstrated by its application to organization studies (Czarniawska, Joerges 1996).
For the sake of this chapter, I will focus more deeply only on one article written by Michel Callon and Bruno Latour (1981), in which they directly address the relations between micro- and macro-actors. It is an interesting work, because it was published before actor-network theory received a huge attention and popularity. Callon and Latour claim that the construction of macro-actors requires power and material objects connected in networks with humans. According to them, assumptions that some actors are on the micro-level and others are on the macro-level are by sociologists too often taken for granted. According to Hobbes’ paradox, “no actor is bigger than another except by means of a transaction (a translation) which must be examined” (Callon, Latour 1981: 280–281). The translation connects different objects and subjects into networks which might be considered of micro or macro size. Thus, the size of the actors depends on the kind of elements the network consists of. If the network consists of a person calling himself a prince, a palace, an army, and administrative staff of scribes, then altogether they might be translated into a macro-actor of state. Without material objects like palace and tools like armor, this network would be most probably considered as a micro-level. What is more, according to Callon and Latour, macro-actors are not more complicated than micro-actors, because they consist of black boxes, simplifications enabling them to act as macro-actors. The size of ← 76 | 77 → the actors and the existence of Leviathan-like actors is therefore a consequence of networks associating and dissociating, and the study of these associations and dissociations is, according to Callon and Latour (1981: 300), a definitional task of sociology. Yet, sociologists contribute to structuring of macro-actors – as Callon and Latour perversely attacked the mainstream sociological method in which certain macro-structures are projected and then the confirmation of their existence is sought in answers of respondents. The link between micro and macro, which is not expressed by the authors directly, is the ability to associate elements, which make something of micro-level and entities of macro-level. Thus, the key message of Callon and Latour’s (1981) paper is that sociologists help in constructing macro-structures.
If Callon and Latour had known the thesis on the sociological vacuum,19 their interpretation in line with their definition of sociology would have been as follows: Stefan Nowak managed to create associations between individual humans, certain imagined groupings, and elements of physical reality. But at the same time, Nowak managed to dissociate certain subjects. Polish society is then simultaneously associated (as a nation being a federation of primary groups) and dissociated (as not consisting of intermediary group bonds). This parallel association and dissociation in interpretations of the sociologists using Nowak’s thesis started a life of its own: the vacuum is sometimes described as agentic, as capable of influencing other objects, stopping or facilitating other social processes. This excursion into the actor-network theory way of speculating is an interesting exercise because it puts the attention to the tool-kit of the sociological creation of objects: the vacuum is constructed out of the lack of something and out of the relation of this lack of something to surrounding objects and subjects.
Actor-network theory does not solve the problems of sociology – it claims that they are irrelevant, and sometimes also ridicules them. The problem of the micro-macro link is – as Callon and Latour (1981) argue in their paper – relevant only for conventional sociologists, who seem to be ridiculous in their attempts to solve the tension between micro and macro (or solve other tensions, as the one between agency and structure). Remaining at my conventional position, I still feel enriched with the perspective brought by the actor-network theory, which obliges us to remember how we, together with subjects/objects of our studies, associate new networks capable of action. ← 77 | 78 →
3.2.4 New science of networks
In the recent years, simultaneously with the huge expansion of the internet, there has been a growing interest and popularity of studies of large networks labeled as the new science of networks. The leading authors of this approach locate it in mathematics, in the tradition of graph theory started in 18th century by Leonhard Euler (Newman, Barabási, Watts 2006). The expansion of this approach to the field of social human behavior research is a result of the development of electronic devices and databases tracking the behaviors of humans using these devices. It allows scientists trained in computational analysis techniques to study the relations and behaviors of people, which are of network character. The key here are the data and analytical techniques which allow to analyze networks of human individuals in the same way as any other networks of nodes which are not human: it does not make any difference if the studied nodes are humans using mobile phones, interconnected routers, molecules linked by biochemical reactions, or nerve cells connected by axons (Barabási 2002: 16). Especially the creation of the so-called “big data” – “data sets so large and complex that they become awkward to work with using standard statistical software” (Snijders, Matzat, Reips 2012: 1) –allowed researchers coming from the fields of applied mathematics or hard sciences to study social phenomena which, mostly due to the nature of internet, leave electronic traces. From the point of view of sociological tradition, this approach has very limited assumptions about nodes: it is simply not interested in human actors’ rationality or emotionality, and treats them as any other nodes. Nevertheless, the outcomes of such studies are very powerful and tell a lot about social reality. Some even predict that analysts of big data will contribute to a change in the paradigm of humanism by proving that humans are reducible to algorithms (Harari 2016).
Duncan Watts and Steven Strogatz (1998) proved that large networks are clustered, but it is enough to have a small number of random connections between them in order to reduce the distances between the nodes. This model of network is often used as a model of global society: people are connected densely with their neighbors, but there are some nodes that connect also distant fragments of the network. According to Barabási (2002), it was a mathematical formalization of Mark Granovetter’s (1973) conception of the strength of week ties.
The studies of the emerging world wide web allowed scholars trained in mathematics to realize that networks are not randomly constructed (as it was assumed for long time) but are ordered, and hubs – nodes with large number of links – are crucial for their structure (Barabási 2002). Ironically, what was known for social studies of elites since Vilfredo Pareto (1897) required mathematicians turn to ← 78 | 79 → computational analysis of large networks. The presence of hubs (nodes of high degree) allows communication in large networks and makes the connection between distant nodes much shorter. Put differently – it makes the world smaller.
Another important contribution of the new science of networks is a modeling of dynamic networks. Traditional network approaches are usually interested either in small-scale networks or in static networks. In contrast, in the new science of networks, scholars taking their inspiration from various branches of physics are capable of modeling the evolution of networks. Applying various assumptions about the fitness of nodes, or the mechanics of acquiring new nodes, they are able to explain the way in which networks with powerful hubs emerge.
In this line of research, the network – which is usually large (large in a sense that it cannot be comprehended without large computational forces) – is the macro-level of analysis. The nodes and edges between the nodes are the micro-level phenomena. The macro-level effects, which are surprisingly similar in networks created by humans and other kinds of networks – i.e. the fact that so many networks follow the power law and are scale-free networks (Barabási, Bonabeau 2003)20 – are aggregations of large number behaviors of nodes. The transformation from the micro-level to the macro-level is fascinating, but it is not theorized. According to Chris Snijders, Uwe Matzat, and Ulf-Dietrich Reips (2012), still only conventional social sciences are capable of explaining the micro-level of network node and tie-formation, which would be a micro-foundation for big data network analysis.
In his popular manifesto of the new science of networks approach entitled Linked, Laszlo Barabási (2002) made some statements, which allow to interpret his view on the micro and macro issue. As he states in the opening pages of the book: “The construction and structure of graphs or networks is the key to understanding the complex world around us. Small changes in the topology, affecting only a few of the nodes or links, can open up hidden doors, allowing new possibilities to emerge” (Barabási 2002: 12). This means that there is a primacy of structure over the action or individual node. The new science of networks is a study of the complexity of a structure – nodes are relevant only if they play an important role in connecting the structure. Nodes’ characteristics, such as fitness, are measurable and applicable in mathematical models, but this approach assumes it only in order to understand the structure. ← 79 | 80 →
Networks can be found everywhere, as a form of structure and organization of society. They are real, and other human heuristics, thanks to which it becomes possible to understand the surrounding reality, are just discursive tools. For example, in case of economic behaviors, Barabási goes as far as to refute the notion of market, and states: “Companies and corporations were seen as interacting not with each other but rather with ‘the market,’ a mythical entity that mediates all economic interactions. In reality, the market is nothing but a directed network. Companies, firms, corporations, financial institutions, governments, and all potential economic players are the nodes” (Barabási 2002: 208). From the perspective of the micro-macro debate, the problem remains. Although nodes and individual edges (relations) are treated as micro-level units of analysis, and the whole network is treated as a macro-level of analysis, ironically, there is still a problem with linking the two. Macro effects and conditions for single nodes emerge from massive aggregation of behaviors of the nodes and the topology of their connections. Yet, this phenomenon is not very much theorized.
The mechanism of assembling the web actually reminds of the classic invisible hand explanation of Adam Smith (1975; Mica 2015). The invisible hand, however, is substituted by the absent spider: “In the absence of a spider, there is no meticulous design behind these networks either. Real networks are self-organized. They offer a vivid example of how the independent actions of millions of nodes links lead to spectacular emergent behavior” (Barabási 2002: 221).
The new science of networks is still new. It provided some powerful explanations of large networks dynamics. It also provides analytical tools useful for the more conventional SNA. The new science of networks will be probably rapidly developing in the nearest future, so it is possible that it will have to face theoretical problems such as linking micro with macro or solving the agency paradox – at least in its applications to the networks of humans.
3.2.5 Brief history of SNA
Last but not least, I am going to briefly present the history of the SNA, the approach that is the most interesting for the topic of this book. Conventionally, the representatives of SNA paradigm refer to Georg Simmel as their classic inspiration. The history of SNA could be also derived from the work of Jacob Moreno (1934) who applied the tool of sociogram in his study on intra-group preferences, attempting to label psychological geography. Early network applications were also performed by social anthropologists such as John A. Barnes (1954), who is credited as the first to use a term “network” in his paper on social classes in a small Norwegian commune of Bremnes. Yet, the real momentum for the SNA ← 80 | 81 → came in the 1960s and the studies of Stanley Milgram and Harrison White, both employed at that time in the Department of Social Relations at Harvard University and working on social networks phenomena. Although Harrison White has a strong recognition among sociologists and social network analyst and is credited for triggering the so-called Harvard Revolution, Milgram’s experiment on small world phenomena received a much wider attention.
Stanley Milgram (1967) in his famous experiment asked participants to send letters, using chains of their personal contacts – from Omaha, Nebraska (a provincial town indeed) to Boston, Massachusetts. One hundred sixty people taking part in the experiment sent 160 letters, out of which only 44 reached their addressees. The chains varied from two to ten intermediary acquaintances, but the median length of the chain was five persons. This finding was reminded and later popularized with a dictum that each person is separated from any other person in the world by six degrees. The experiment is not only a trivia to amuse people at parties: it has important consequences for the network understanding of social structure and inspired more mathematically driven analyses. Assuming that each person maintains relations with approximately 150 other persons – the so-called Dunbar’s number (2010) – and that half of these relations are not redundant, at the fifth step it should be possible to reach more than two billion people (75 raised to the fifth power). Of course, the condition is that those people are not separated.
The seminal study of Nancy H. Lee (1969) about pregnant women searching for abortionists in times when abortion was penalized in the US, brought problems of agency and trust into studies of networks. Connection to the network eventually leading to the abortionist was enabling action – in this sense, networks constrain and allow agency. In case of the search for an abortionist, trust encapsulated in network was of a key role, because only the trusted ties were opening the possibility of acting.
I believe, however, that a real “game changer” for the SNA was the article published by Mark Granovetter (1973) entitled “Strength of Weak Ties,” which is an important inspiration for this book. It is the most often cited sociological paper up-to-date. It transformed thinking about social cohesion, labor market, and overall structure of human relations. On the empirical level, the study of Granovetter explores the role of personal connections in the search for a job. It was already known from some earlier studies (i.e. Reynolds 1951) that personal contacts are the most often used mode of getting a job. What Granovetter (1973: 1361) was able to prove was that the majority of jobs were found through weak ties (meaning relations perceived as not emotionally intense, characterized ← 81 | 82 → by a small amount of time spent together and limited reciprocal services). This finding seems counterintuitive, because many people think that the information about jobs is passed through strong relations. Yet, this is in fact impossible due to structural reasons: networks of strong ties are usually closed – a good friend of John is usually a friend of John’s other good friends. Thus, an information spreading through the network of good friends with strong ties usually cannot travel far; traveling through the network of weak ties, the same information can travel to very distant social circles.
The more general contribution brought by the conception of the strength of weak ties into sociological thinking concerns social cohesion. People, and many sociologists as well, tend to think that strong attachments are important for social cohesion. The consequences of the strength of weak ties conception inform us that for social mobilization and cohesion the weak ties are crucial. The community connected only by strong ties might be too fragmented to trigger a collective action. On a more general theoretical level, Granovetter’s reasoning challenges the thinking that large phenomena have large causes. On the contrary – important consequences are products of seemingly unimportant weak ties. Next to the studies of Milgram (1967) on the phenomenon of the small world, the strength of weak ties conception provided first network explanations of the network-like construction of societies, which later inspired also analysts of complex networks discussed in the previous section.
Harrison White’s works are recognized as having an important impact on SNA. His influence started already in 1960s with the so-called Harvard Revolution, while some of his seminal papers were published in 1970s, and his opus magnum Identity and Control (White 2008) was released in the 2000s (Azarian 2003). From the concepts formatting the thinking about network patterns of social structure White had introduced, I am going to discuss structural equivalence, blockmodelling, and cat-domains.
Fracois Lorrain together with Harrison White (1971) introduced the notion of structural equivalence, which describes network nodes which have the same type of ties to other nodes. Thanks to structural equivalence, it is possible to discuss social roles abstracted from the characteristics of concrete persons and base their definitions on their relations with other roles, and not just their attributes. Lorrain and White (1971) called this approach “categorical-functional” because it allows to discover functions played by roles in networks and categories of actors. Categories and functions are defined by the position in the structure of relations, not by the common attribute or functional role for the system. ← 82 | 83 →
Blockmodelling is an analytical procedure in which researchers aggregate existent actors in blocks. Nodes grouped in one block have a similar pattern of ties connecting them inside of the block and with actors outside the block. Actors are set into blocks, which represent one position, linked (or not linked) by bonds (the aggregated patterns of ties between positions) into various blocks. Blocks are “sets of persons structurally equivalent with respect to other such sets across several types of relation” (White, Boorman, Breiger 1976: 772). Although initially the blockmodelling was applied to very small samples of data on network relations between individual actors, the idea of aggregating structurally equivalent actors into blocks to grasp the image of the social structure behind the relations of individuals is generalizable for larger networks. By the virtue of blockmodelling and structural equivalence, it is possible to perceive the structure, which is not always apparent from the data gathered on the micro-level. Even simple aggregations – like those in the tradition of survey studies of opinions and attitudes – do not allow to track the intermediations of structures as blocks based on the equivalence. Blocks might (but, of course, also might not) be a structural base for collective action and forming identities. The crucial issue for the problem of this book is that blockmodelling allows to grasp patterns that are not always perceived by social actors themselves.
In his monumental book Identity and Control, White (2008) synthesized his previous works and aimed at explaining how social formations emerge. In this book much attention is put to the middle-range orders and the way they are created by context (White 2008: 335). Among levels analyzed by White there are netdoms and realms. In this book, White also turned from his strong conviction that the structure has primacy over the individual. Although he highlighted that the “overall context or structure directly constrains and shapes action,” he also admitted that “it is individual identities that generate the unpredictability at the core of history” (White 2008: 364). Clearly, White was very much concerned with the problem of agency-structure and eventually his claims could be categorized as similar to those of Giddens (1984) and other authors underlining the co-constitution of agency and structure. This is how the statement that “corporate actors shape and are shaped by events” (White 2008: 366) could be interpreted.
The works of White had a big influence on other authors in the field of SNA, however, his writings are less known outside this field. It is most probably due to the hermetic language and intensive use of mathematized formulations of his arguments. The most important inference from White’s work is the need for including the position in the net of relations into the analysis of social role and potential for action. Another important conclusion is that a structure can be ← 83 | 84 → better understood when the lack of relations between certain positions, and not the presence of relations are being looked at. This allows to turn to the conception of structural holes.
Ronald S. Burt (1992) in his conception of structural holes explored another paradox of the network structure. In his perspective, the structural hole argument was an expansion of the strength of weak ties argument in the sense that the weakness of the tie was rather a correlate, and not a cause, of their connecting character (Burt 1992: 27). Quite often the ties that connect two networks are weak, which might lead to thinking that they are important because of their low strength; yet, they are important because they are bridging two components of a network. Structural hole is “a relationship of nonredundancy between two contacts” (Burt 1992: 18). Thus, the actor who is capable of connecting through structural holes has an advantage over other actors. This inference is important and, at the same time, paradoxical because it is against common sense reasoning according to which it is important to have many connections. Yet, what is really advantageous is having connections which are crucial for the overall structure of the network – especially the ones no other actors have. Burt (1992: 8) highlights that his relationist sociology “escapes the debilitating social science practice of using player attributes for explanation.” The competitiveness of an actor is not so much a function of his or her attributes, but of a function of position in the structure of relations with other actors. Burt was interested in competitiveness of entrepreneurs and that is why he was showing that for entrepreneurs having some special skills is not as important as having a structural ability to connect through structural holes. Yet, from this line of theorizing develops a perspective that any virtue or vice of an actor is rather an effect of her position in the network than of her individual attributes. This is why SNA focuses on relations and network influence on individuals. Individuals create relations and thus build networks, but their position in the networks shapes actors’ qualities.
Another development in the SNA was the conception of social resources, which are defined as “the wealth, status, power as well as social ties of those persons who are directly or indirectly linked to the individual” (Lin, Ensel, Vaughn 1981: 395). According to Nan Lin and his colleagues, social resources encompass two constitutive elements: social relations and resources which are accessible through these relations. In this sense, social resources are not simply possessed by individuals – they are accessed by them. The conception of social resources later transformed into the network theory of social capital (Lin 2001). I am not going to pay much attention to the theory of social capital in this chapter, because it is going to be discussed in details in Chapter 7. Here, I just want to ← 84 | 85 → mention that the most vivid approaches to social capital are built thanks to SNA’s focus on accessing resources and services through social ties.
The internet, which is a network itself, was not only a trigger for the development of the new science of networks presented in the previous section, but also gave a new momentum for SNA. Again, the empirical research brought counterintuitive findings. The most important one was that on-line connections are a new medium for enhancing the already existing ties (Hampton, Wellman 2003; Quan-Haase, Wellman 2004). The internet, despite being a revolutionary technology, did not revolutionize the structure of social relations. Only a small share of people use the internet to create new ties – it is rather a tool of maintaining the ties that had already been created off-line (Chambers 2013; Ellison et al. 2011). The internet undoubtedly provides a lot of research material for the analysis of social networks. It has also raised the social awareness of the network character of society. Thanks to internet, and social media especially, the elements of network approaches entered the mainstream language. Currently this is how people and companies talk and perceive their place in society.
3.2.6 SNA – basic concepts and assumptions
If the SNA is a theory, what are its main concepts, assumptions, and research questions? Apparently, a key concept is a network, which from the formal perspective is a “collection of nodes […] and the linkages […] among them” (Giuffre 2013: 214). But the fact that a network is a social network adds some more qualities to it. In a social network, nodes are humans or collectives of humans (organizations or groups), and for that reason a given network receives additional qualities and becomes “a set of relations, associated meanings, and expectations that connect actors” (Castilla, Lan, Rissing 2013: 1000; cfr. Portes 1998; Burt 2005; Small 2009). This way of thinking about the social network is already connecting it with network conceptions of social capital, which will be discussed in more details in Chapter 7. Networks allow to transfer resources and they also provide signals to audiences (Castilla, Lan, Rissing 2013: 1001). Resources include information, learning, influence, and support, but networks are also built thanks to the exchange of physical commodities. Thanks to signals transmitted through networks, actors and broader audiences may draw inferences about abilities, legitimacy, status, and relationship meaning (Castilla, Lan, Rissing 2013: 1004–1005).
Network consists of two elements: nodes and edges. SNA differs from the new science of networks because it is only interested in nodes who are social actors. Thus, nodes are either human individuals or collectives, like groups or organizations ← 85 | 86 → (Giuffre 2013: 213). Edges are the ties, linkages, or relations connecting the nodes. This element opens possibility for researchers to be creative. While nodes are in a self-explanatory way either individuals or recognized social constructions, ties are operationalized in various ways. A tie may mean that two actors interact with each other or that they know each other personally, which required at least one interaction during which they were introduced to each other. Relations may have a so-called natural character, just like family ties – in this case there might be no interaction between the actors, or they might even not know each other. Crossley (2016: 172) defines a social tie as “a sedimented interaction history embellished by the anticipated likelihood of future interaction.” Ties, similarly to interactions, are not directly observable. Thus, quite often when researchers have knowledge that nodes participate in the same social setting, i.e. they are members of the same club or directorate, they assume that there is a relation between them. The relation represented as a tie may have various strengths (Granovetter 1973) measured in various ways (Marsden, Campbell 1984). They may have physical manifestations, like in case of exchanges or interactions, but they may also have important conscious or emotional components, like in case of ties of friendship.
The research on networks is inevitable for social sciences, because networks are one of the five principal ways by which people gain access to resources: market exchanges, institutional distributions, community exchanges, coercive appropriations, and self-provisioning (Wellman 1999). This distinction is of an analytical character, and social networks are in fact engaged in all modes of gaining access to resources – in case of market exchanges, it is persuasively presented in the conception of embeddedness, which will be discussed below.
Concluding this section, it needs to be reiterated that networks are metaphors thanks to which it is possible to grasp patterns of relations between actors. The relations and the regularities of these patterns are at the center of the focus of SNA. The relations between individuals – not the attributes of individuals – are key to understanding social processes (Giuffre 2013: 2–3). The dominating approach among network scholars is the structuralist approach – namely, the conviction that social structure is a causal factor for the actions and attributes of individuals. Yet, it can be noticed that there is a growing insight into the mutual co-creation of network structures and individuals. Networks are created by individuals but then, in turn, networks shape individuals. As I have demonstrated, the agency-structure problem is at the center of attention of network scholars. In the next section I will discuss how the SNA is useful for tackling the micro-macro problem in sociological theory. ← 86 | 87 →
In this section I will depart from the focus on SNA and discuss how the inferences drawn from this approach can be useful in linking micro and macro. I see two ways in which SNA contributes to the micro-macro debate. The first is the debate on the embeddedness of social action in social relations; the second is the social network theory of social capital. In this chapter, I am focusing on the conception of embeddedness. I will pay much attention to the theory of social capital as a possible mediation between the micro- and macro-levels of analysis in Chapter 7.
Nick Crossley (2016), in his recent summary of relational sociology and SNA, placed this paradigm in the middle between holism (excluding actors and agency) and individualism (excluding structure). According to him, the focus on relations and their structured patterns – networks – allows solving the holism-individualism divide, because “structure is not ‘above’ or ‘behind’ actors, from this perspective, however. It lies between them” (Crossley 2016: 182). Actors in a relational account are created by relations in which they participate – it does not make sense to analyze them in separation, even if actors are more directly observable. Crossley (2016) also rejected thinking about the micro-macro divide. He preferred to think about it as a continuum between micro and macro and placing a given phenomenon somewhere in this continuum is mostly a matter of scale. He argued that events of a large scale often start as interactions between small groups of individuals (i.e. a cabinet taking an important political decision), which later affects masses of other actors, because of the mobilization of large-scale networks. Another argument regards large-scale social divisions, such as social classes, which “manifest in patterns of connection (and lack of connection) within a population” (Crossley 2016: 181). They are possible to be tracked by search of structural equivalents (Lorrain, White 1971). The structurally equivalent positions in small (micro-scale) networks in the scale of the whole population (macro-scale) are components of social stratification.
In the “Manifesto for a Relational Sociology,” Mustafa Emirbayer (1997) stated that the crucial dualism of sociological theory is not agency-structure or individual-society, but substantialism-relationism. Yet, he considered how the relational perspective is useful for tackling the research problems on the macro-, meso-, and micro-level. He called to reconceptualize these levels of inquiry. On the macro-level, relational perspective is useful in the reconceptualization of the concept of society as “an autonomous, internally organized, self-sustaining ‘system’” (Emirbayer 1997: 294). On the meso-level, the reconceptualization should be based on locating regularities, recurrent mechanisms, patterns, and ← 87 | 88 → sequences of relations in what he calls “occasions” (Emirbayer 1997: 296). On the micro-level, relational approach should reconceptualize the individual as not unproblematic and preconstituted, but as constructed in relations with other individuals. Both Crossley and Emirbayer point to the embeddedness of actors in relations as a key to understanding social processes and structures.
Mark Granovetter (1985) in his essay “Economic Action and Social Structure: The Problem of Embeddedness” took up the issues of agency and structure, and the micro- and macro-levels of analysis by applying the network perspective. It is a very general work regarding the relations of economics and sociology with a theoretical impact on all social sciences. Granovetter applied polemical strategy by refuting the famous work by Olivier Williamson (1975) on markets and hierarchies.21 In order to achieve his theoretical goal, Granovetter critically applied the concept of “embeddedness” coined by Karl Polanyi (1944) and Denis Wrong’s (1961) theory of undesrocialized and oversocialized conceptions of man in social sciences.
The simplest explanation of the embeddedness argument is as follows: “the behavior and institutions to be analyzed are so constrained by ongoing social relations that to construe them as independent is a grievous misunderstanding” (Granovetter 1985: 482). Every action – in his essay Granovetter focuses on economic action but his inferences are generalizable for every kind of social action – is driven not only by the actor’s rational decisions (undersocialized conception of actor in classical economics) or his or her internalized general morality and norms (oversocialized conception of actor in functionalist sociology). According to Granovetter, paradoxically both conceptions of an actor are atomistic. In case of the undersocialized account applied in economics, in which the actor takes up rational decisions, and where the social position of actor is not theorized or recognized as frictional, it is quite obvious. In case of the oversocialized account, the behavioral patterns of the actor are internalized. The source of the norms driving actions is in the generalized morality, so “ongoing social relations thus have only peripheral effects on behavior” (Granovetter 1985: 485). In both accounts, social actors are atomized from their immediate social context, which has a network character of patterns of relations with other actors.
In his embeddedness argument, Granovetter claims that actors situated in social networks do not act mechanically according to their profiles of preferences (classical account) or internalized norms (functionalist account): they are influenced ← 88 | 89 → by the experience of past relations, concrete present relations, and expectations about future relations. The conditions for taking decisions by actors are continuously reconstructed because of the participation in relations and obligations to other actors. Referring to the discussion in Chapter 1, it is apparent that Granovetter is pointing to all three pairings crucial of sociological theory. Network of social relations is for him mediating in the individual-society pairing. Position in the structure of relations constrains and enables the agency of actors – that is, the perspective on the agency-structure pairing. Last, but not least, Granovetter shows that the analysis of networks is crucial in understanding how broad macrostructural circumstances are linked with proximate causes (as he calls micro-level phenomena). Thus, the embeddedness in a network of relations seems to be the link tying individuals with society. Without taking the networks of relations into account, both undersocialized and oversocialized perspectives analyze the atomized model of society, similar to the Hobbesian state of nature.
Granovetter analyzed the role of trust and conditions for malfeasance in economic relations. He saw trust as an effect of relations with concrete actors, not as an effect of generalized morality. Networks, in his approach, at the same time might be blocking the possibility of malfeasance and enabling it – both building confidence and breaking confidence requires mobilization of other social actors. Thus, Granovetter pointed to the salience of trust for interactions without exaggerating its role, which occurred in the 1990s after the publication of works by Putnam (1993).22 As Granovetter (1985: 492) phrased it, “both enormous trust and enormous malfeasance, then, may follow from personal relations.” It is a good summary of the embeddedness argument, according to which all kinds of action are enabled and constrained by the context of social relations (structured in networks). Contrary to Williamson’s (1975) argument, there are other social structures enabling economic action than markets and hierarchies (organizations or firms) – these are networks connecting people both within the organizations and between the organizations.
Granovetter’s (1985) statements on the embeddedness of action had two major interlinked sources: first, his theoretical perspective on getting a job; second, network conceptions of social structure. In his earlier work on getting a job, Granovetter (1995: 3)23 presented the research question – “How the information that facilitates mobility is secured and disseminated?” – as a potential link ← 89 | 90 → to integrate the micro- and macro-levels of analysis. As macro-level analyses, Granovetter pointed the works on social and labor mobility in economics and sociology, and as micro-level analyses he pointed to studies on the motives of individuals wishing to change their jobs. Granovetter (1995: 4) proposed that the study of ties connecting employees with employers and transmitting the information is the link between simple aggregation of demand (there is work to be done) and supply (people will be employed to do this work).
Another way of seeing what is on the micro- and macro-level proposed by Granovetter (1995) was to equalize micro with nearby causes and macro with distant causes. Although the micro-level in his work is understood clearly as actions of individuals and their relations with other individuals (level of ego-network), the macro-level is blurred, as it consists of both distant and large-scale objects. In this sense, weak ties are important because they are capable of linking with distant social settings (they are capable of making the world small). Granovetter (1995: 100) was also interested in how rational choices of individuals regarding their employment (micro-level) may lead to institutionalizing social inequality on the macro-level.
The in-depth study presenting the network conception of social structure and its connection with agency, which influenced the conception of embeddedness, was the work by Ronald S. Burt (1982) entitled Toward a Structural Theory of Action: Network Models of Social Structure, Perception, and Action.24 Burt assumed as key for the analysis of action, the analysis of how actors evaluate the action’s utility before performing it. The approaches to the evaluation of actions’ utility are consequential for the microanalytical models of action and define ideal types of actors (Burt 1982: 331). In the atomistic perspective, typical for classical and neoclassical economics,25 it is assumed that “alternative actions are evaluated independently by separate actors so that evaluations are made without reference to other actors” (Burt 1982: 331). Thus, in this model, the actor has exogenous interests. In the normative perspective, typical for the functionalist sociology, ← 90 | 91 → it is assumed that actors evaluate their actions interdependently but “as a function of socializing processes that integrate them within a system of actors” (Burt 1982: 332). This model assumes the endogenous interests of an actor. The interdependence is created during socialization, so it is prior to the evaluation. In the assumptions made in this model, interdependence achieved by socialization sustains the social system. Both models lack the link between the micro-level of an actor and macro-level of a system. In the atomistic action model, the macro-level is just a consequence of aggregation of many actors’ actions. In the normative model, actors are formed by the systemic level in the process of socialization and then only enact the once imprinted norms.
Burt (1982) proposed the model of action, in which evaluations are made in the interdependence with other actors founded on their position in social structure (of network relations). The interests here are not exogenous or endogenous, but of a structural character. The actor (in the structural theory of action) takes into consideration the information on the network relations of other actors, whereas the actor in the atomistic theory does not. The actors (in the structural theory of action) are interdependent (like in normative theory of action), but not because the common norms shared, but because of taking into account relations with other actors. These detailed considerations of Burt are the structural basis for the conception of embeddedness. The network is a bridge between the atom of social actor and the totality of a social system.
Another perspective on the role of networks in the critical discussion on the market-hierarchy approach of Williamson (1975) came from Walter W. Powell (1990), who attempted at merging organizational theory with network analysis. Powell was not interested in the micro-macro debate, but rather in the debate on the coordination of action. His argument was that Williamson’s (1975) assumptions did not allow him to see other forms of coordinating action than market or hierarchy. Economics focuses on well-defined actors, such as physical or legal persons, and well-defined interactions, such as transactions. This framing of actors is sustained even in informal settings where actors are not legal persons and transactions are not framed by legal contracts. Under such assumptions, it is difficult to grasp long-lasting relations which exceed simple iteration of transactions, collective actors of vaguely defined boundaries, and exchange of resources which are not easily tangible or measured – for example, know-how. Markets and bureaucratic hierarchies are forms of coordination well suited for exchanges of well-defined resources or mass production of standardized goods. Powell (1990) noticed that this market-hierarchy continuum does not cover situations in which there is a need of quick exchanges (hierarchies and markets tend to select the ← 91 | 92 → information and slower its exchange) or there is an exchange of resources based on tacit knowledge. Without taking into account the network forms of organization, it is not possible to understand many forms of economic activities. Powell was especially interested in economic action involving inventions and large volume of knowledge. Powell (1990: 308) highlighted that network patterns “tend to be invisible to most observers. Instead of long-term rates of reproduction most participants observe individual acts of ranking, favors, and contacts.” This is a very consequential remark, because it shows that certain taken-for-granted frames do not allow to perceive social structures which are crucial for collective action. This taken-for-grantedness is both a problem of certain theories in social sciences (in economics and sociology as well) and a problem of perception in every-day life.
Assuming that social actors are either legal persons or physical persons misses the network context of emergence of actors. In case of new organizations, the formal founding is not the birth of an organization, but its creation (or emergence) is a process of crystallization on the base of networks of its future founders (Powell, Oberg 2017: 450). In this sense, it is possible to treat the network form as the early stage of collective action, which later – when individuals decide to take a form of organization to pursue a collective action – becomes formalized. Similar is the case of emergence of markets (as a forms of coordination of collective action) and social fields (Padgett, Powell 2012).26 Yet, both networks and institutions have co-constitutive aspects: norms shape collaborations and then collaborations form as networks, and eventually transform into organizations spreading the norms of collective action (Powell, Oberg 2017: 458). Thus, networks are not only means of transmission of information, resources, or reputation, but they are also able to transform social settings (Padgett, Powell 2012: 9). Networks are analyzed thanks to micro-level data (on actors and their relations), and this allows to see the macro-level phenomena – which, for Powel and Oberg (2017) are institutions – because “individuals are deeply embedded in multiple networks through their connections to friends, collaborators and mentors. These webs of affiliations create various demands and expectations, and identities are forged out of these divergent expectations” (Powell, Oberg 2017: 460). In this sense, embeddedness and transformative capacities of networks are means of linking the micro-level with the macro-level. ← 92 | 93 →
Mario Luis Small (2009) coined the notion of “organizational embeddedness” in order to look at the relation between organizations and networks from the other end. He was particularly interested in how organizations influence the creation of social ties and quality of networks. The organizational embeddedness perspective suggests that a “person’s social capital depends substantially on the institutional practices of the organizations in which the person routinely participates. If embedded in the right organizations, a person can acquire significant advantages, through the workings of these networks, which yield palpable effects on their well-being” (Small 2009: 177). The focus of Small (2009: 178) is less on the structure (of networks) and rather on the (organizational) context of creating ties. Personal ties are created not only thanks do everyday routines of organizations, but also because of the interconnections of organizations. The main interest of Small’s research are social inequality and the circumstances under which social ties form. These problems are connected, because the inequality in access to social networks reproduces the more general inequalities. Thus, the organizational embeddedness perspective links the micro-level processes of tie formation and maintenance with macro-level structures of state policies and inter-organizational networks. Small (2009: 190–191) labels this perspective “a meso-level approach to social inequality, one that seeks regularities in how people interact, obtain information, trust others, respond to obligations, acquire supportive services, and secure everyday material goods – and probes the organizational mechanisms by which these actions can be tied to state authority, the collaboration between the public and private sectors, and the large organizational networks central to structural inequality.” In organizations in which people participate routinely – Small studied kindergartens – actors nonpurposely form ties as by-products of interactions. This is more likely in organizational environments which influence them to interact frequently, with a focus on some activity and when they have a reason to cooperate rather than to compete (Small 2009: 15).
The attempt to merge relational sociology focused on embeddedness and social networks with institutional analysis was also made by more institutionally oriented scholars, such as Victor Nee and Paul Ingram (1998). In their view, institutions are macro-level phenomena and social networks are located closer to the micro-level. Because of the interdependence of networks and organizations, the networks finally influence the institutions, because “organizations in turn affect formal norms through political action, while their performance determines performance at the macro-level” (Nee, Ingram 1998: 32). Nee and Ingram’s perspective brought attention as another attempt to integrate institutional analysis with SNA. ← 93 | 94 →
Knowing from Granovetter (1985) that economic action is embedded in networks of social relations, and from Small (2009) that networks are organizationally embedded, Steve McDonald coined the notion of “dual embeddedness” merging the two levels. The economic action unfolds in “networks of social relations that are themselves set in specific national, social and economic institutional arrangements” (McDonald, Benton, Warner 2012: 76). McDonald with his colleagues (2012) explored an interesting paradox: in the coordinated German labor market there are more jobs obtained through informal personal contacts27 than in the loosely coordinated American labor market. Intuitively, on the loosely coordinated labor market there should be more informal matching and the coordinated labor market should be effecting in formal matching. Yet, there is convincing evidence that various organizational settings of coordinated German labor market (i.e. trade unions, work councils, occupational education establishments) produce possibilities of forming social ties, which are later helpful in informal job matching. In the US, the coordination is weaker, so people often need to use formal (coordinated) modes to get jobs (McDonald et al. 2012).
I am convinced that the conception of embeddedness is the most promising network approach to linking micro- and macro-levels of analysis. The embeddedness has then a double meaning (not only when the concept of dual embeddedness is considered): it draws attention to the fact that actions are embedded in social relations, but it also means that various levels of analysis are embedded in each other. It is impossible to understand the actions of an individual without considering the relations that form and constrain the individual. The relations have patterns of a network character, which are shaped by institutional and organizational settings. These settings are shaped by macro-powers of state and economy from the top-down approach, but from the bottom-up approach, organizations, markets, and fields emerge from the stabilization of networks of relations.
The aim of this chapter was to present the insights from the SNA to the debate on the micro-macro problem. I presented a variety of network approaches (namely, network exchange theory, Manuel Castells’ theory of network society, actor-network theory, and new science of networks) and briefly discussed their contribution to the micro-macro debate. I focused on the SNA and I argued that it is not another form of structural extremism in its radical version, or an attempt to integrate the theory from the macro theoretical end in its “soft” version, as some commenters believe (see Ritzer 1990). Thinking in network terms currently dominates in the realm of social sciences. The social structure is currently perceived as a complex combination of networks, and not as a combination of groups, as sociologists tended to do for a very long time. Now, even if the notions of group, community, or society are used by sociologists, they cannot be fully described without mentioning their network component. There is also persuasive evidence and there are strong theoretical frames of the mutual co-constitution of networks and organizations, and markets.
After an overview of the most important insights from the SNA, such as the small world problem, issue of trust in network relations, the conception of the strength of weak ties, structural equivalence, structural holes, social resources, and the studies on internet, I presented the basic assumptions and notions of SNA. I needed this base to discuss how theorizing in network terms allows to integrate the micro-macro divide. Network scholars are very much preoccupied with the agency-structure pairing. The relationist thinking, which lays at the foundation of SNA, allows to tackle the great pairings of sociological theory differently than from the substantialist perspective. In this theoretical framework, actors (individual or collective) are always shaped by the structure of the relations in which they are ← 95 | 96 → embedded, so the individual-society pairing is also considered. In case of micro-macro pairing, in my opinion, the most important contribution of SNA is brought by the two theories: the theory of embeddedness and the theory of social capital. The latter is going to be discussed separately in Chapter 7. The former was presented above with much attention to the way it relates to the micro-macro debate.
The discussion about the embeddedness – a concept coined by Polanyi (1944) and reinterpreted in network terms by Granovetter (1985) – is a good example of SNA’s capability of bringing together various levels of analysis. Actions (and therefore actors) are embedded in social relations. Patterns of relations take a network form, but these networks are created in the context of organizations, markets, and social fields. The state and other large macro-level phenomena shape the conditions of network emergence. In this way, the notion of dual embeddedness is reached. Yet, the link is not unidirectional, because networks after tipping a certain point of equilibrium trigger the emergence of new organizations and markets (Padgett, Powell 2012). The conception of embeddedness and the debate about it teaches about the complex linkages between actors, relations, structures at various levels of analysis.
The power of SNA (recently made even stronger, thanks to the tools of the new science of networks) is based on relatively simple assumptions and terminology. Nodes, ties, and networks are elements which open the possibility of building quite complicated statements and models illustrating social processes. What is a key issue for the subject of this book is that relationist assumptions and tools of SNA bring attention to phenomena which are often unnoticed. A good and still inspiring example is the conception of the strength of weak ties. The role that weak ties have for the cohesion of communities, mobilization of resources, and transfer of information is unnoticeable both for the common sense and numerous sociological theories. Social relations are often difficult to be tracked empirically, but it does not mean that they do not have an impact on social processes. The role of actors capable of bridging the structural holes and not having any structural equivalence is not always easily perceivable from the participant’s perspective and might be noticed only after the analysis of a larger network. The knowledge about the small world phenomenon informs how close (in the social sense) people are to each other and, at the same time, what the social barriers between people are. Weak ties linking individuals are often not treated as something important for them, but on the structural level they allow the exchanges of resources and information. Finally, SNA teaches that large and important events or processes do not always have large and important causes. Thanks to this knowledge, it is now recognized that looking always for some large causal forces might be misleading.
4 American Sociological Association, together with German Sociological Association, organized in 1984 a conference in Giessen. The presentations from this conference were published in the collected volume Micro-Macro Link edited by Jeffrey C. Alexander, Bernhard Giesen, Richard Münch and Neil J. Smelser (1987).
5 International Sociological Association during its congress in Mexico, in 1982, organized the symposium the aftermath was a two-volume collection edited by Shmuel N. Eisenstadt, Horst J. Helle (Eisenstadt, Helle 1985; Helle, Eisenstadt 1985).
6 The European antecedents of Coleman’s approach to the micro-macro problem were recently discussed by Werner Raub and Thomas Voss (2017). Other attempts to tackle the micro-macro problem are usually embedded in theoretical traditions, which before the 1980s were present in the mainstream sociology in their macro- or micro-extremist form.
7 It should not be forgotten that these distinctions are of an analytical level, so from one perspective one level of analysis might be meso, while from the other it might be macro.
8 It is important to remember that Bourdieu was against explicit definitions of concepts, which for him was a positivists fallacy in sociology.
9 It may be seen this way from the perspective of an analyst, but social actors treat social constructions as external and real (Berger, Luckmann 1967).
10 Another dichotomy of personal troubles and public issues was initially formulated by Charles W. Mills (1959).
11 To treat micro-macro as a continuum is quite persuasive, but to treat the subjective-objective as continuum is highly controversial from the ontological point of view. From a deeper reading of Ritzer’s works, it seems that he simply does not use the terminology consequently and does not pay attention whether the subjective-objective is a dichotomy or continuum.
12 Also O’Donnell (2010b) saw organization as located on the meso-level of analysis.
13 In this respect, Turner’s theoretical strategy is similar to the one deployed by Ralf Dahrendorf (1959), who believed that society is Janus-headed, with one face of conflict better explained by coercion theory, and the other face better explained by integration theory. Dahrendorf focused on conflict and developed the coercion theory of society, yet he admitted it is only a part of the Janus-head. Jonathan H. Turner similarly admits that in order to understand social phenomena we need to have top-down and bottom-up theories, but he focuses on developing the top-down approach. The full and detailed presentation of Turner’s (2010a; 2010b; 2010c) grand theory integrating micro-, meso-, and macro-levels is presented in his monumental three volume work entitled Theoretical Principles of Sociology. Its first volume focuses on macrodynamics, the second one on microdynamics, and third on mesodynamics.
14 It is ironic that often when authors are close to finalizing their work, someone else publishes a much better and more comprehensive piece on the same topic. This is also the case of this chapter, which I seriously reworked after reading Daniel N. Kluttz and Neil Fligstein’s (2016) summary and discussion of the field theory in sociology for the Handbook of Contemporary Sociological Theory (Abrutyn 2016). Their work focuses on the social field as a tool for analysis at the meso-level. I believe, however, that I am able to add something from my perspective on the field theory, which was omitted by Kluttz and Fligstein. Firstly, their treatment of the notion of organizational field, as it is used in recent new institutional theory of organization, has its limitations – they focus mostly on the classical understanding of the notion, as it was introduced by Paul J. DiMaggio and Walter W. Powell (1983). Yet, there is a vivid discussion and conceptual development in this theory, and organizational scholars now have very much reoriented their understanding of field processes (Wooten, Hoffman 2008). Secondly, Kluttz and Fligstein (2016) advertise strategic action fields theory of Neil Fligstein and Doug McAdam (2012) as solving shortcomings of other perspectives in field theory. Although I agree that Fligstein and McAdam added much content to our understanding of social fields, I provide here a critical approach to the strategic action fields theory.
15 Even Mustafa Emirbayer and Victoria Johnson (2008), who called for bringing the habitus concept into the organizational analysis, did not offer the idea of habitus on the collective level. They highlighted that the study of organizations will benefit from the study of their individual members’ habitus.
16 Of course, one cannot limit coercive isomorphism to relations with a state. There are other kinds of powerful actors which are capable of exerting coercive pressures. On the market, these are large companies able to set standards to which smaller companies must conform. There are supra-state level actors capable of setting coercive pressures: European Union is a good example. There are also soft-law regulators which do not have state capacities of coercion but are still able to use the organizations’ hunger for legitimacy in order to influence them: the examples here are creators of rankings, or all sorts of fair trade certifiers.
17 The notion was earlier introduced by Fligstein (2001) in his famous paper.
18 Very similar observations have been made by Bromley and Meyer (2015).
19 I love to fantasize about such encounters and I am pretty sure that Bruno Latour would have been more than thrilled by the idea to use the hard science term “vacuum” to describe social structures.
20 Power law in case of networks means that the degree distribution in a given network is highly unequal, as in Pareto distribution. Some minority of nodes have a large number of connections (high degree), while the majority of nodes have a small number of connections (low degree).
21 Olivier Williamson for his work on markets and hierarchies, and transaction costs received in 2009 the Nobel Memorial Prize in Economic Sciences.
22 The problem of trust and its incorporation into various conceptions of social capital will be discussed in Chapter 7.
23 In this book I refer to the second, extended edition of Granovetter’s PhD thesis Getting a Job: A Study of Contacts and Careers, which was first published in 1974.
24 The work of Burt (1982) was categorized by Ritzer (1990) as an example of an attempt to integrate theory from the macro theoretical end. In my opinion, a careful reading of Burt’s (1982) book shows that his approach was conducted without prior commitment to the micro or macro end. Early attempts to categorize network analysis by authors interested mostly in general theory were placing it together with other macro theories of social structure. It is not quite right, because the social network analysis is a different approach to social structure, departing from the micro-level of nodes and edges.
25 The same approach was also typical for the rational choice theory in sociology, which in the 1980s was attempting to solve the micro-macro dilemma in theory – see Chapter 1.
26 Padgett and Powell (2012), in their monumental book on the emergence of organizations and markets from earlier social networks, use the parallel of the chemical process of autocatalysis. The authors highlight that “in the short run, actors make relations, but in the long run, relations make actors” (Padgett, Powell 2012: 3).
27 The authors carefully calculated as informal also the situations in which the employee was not searching for the job. According to their analysis, unsolicited job offers are more common in Germany, thanks to the organizational settings allowing routine exchanges of information through social networks (McDonald et al. 2012). This phenomenon is called the “invisible hand of social capital” and it is discussed in Chapter 7.