Naturalistic ontology as a guide to unified social science (draft)

Introduction

Ontology studies “what there is” in the world: what exists genuinely and nonredundantly (Baker, 2019). Any science strives to answer this question. Ontological commitments shape scientific explanations, implicitly or explicitly, and explanatory power of theories reflects its ontological assumptions. The choice of study methods, in its turn, reflects an intended way of explanation and demonstrates a commitment to a particular “version of the world” (Moon & Blackman, 2014)¹.

In any science, the relative coherence of its ontological picture paired with its explanatory power is connected to its “maturity”. There are predictively and explanatory powerful theories of quantum gravity, autocatalytic cycles and social network dynamics as soon as there are empirically adequate descriptions of entities invoked in these theories².

A viable ontology is both a product and a project of any science. It is a foil for future research and an ultimate goal of knowing “what there is”. Conjectures about the nature of objects of study influence the ways of approaching them: if the social world, for example, is comprised of individuals, you will probably study individuals, their properties and relationships between them. Philosophy and history of science have seen cases when mistaken ontology led to scientific mistakes³. In social science, the issue of ontology is even more complicated due to its multi-paradigm nature and lack of shared foundations: diverse objects of study, methods used for studying them and explanatory strategies⁴.

The relevance of studying ontology in social science stems from two things:

• its connection with cumulative growth of knowledge (Lauer, 2019) which is the main aim of science (bird2022?)
• its connection to policy-making and potential usage in social design (Tromp & Vial, 2023; Vink & Koskela-Huotari, 2023), which is said to be detached from social science (Eriksson, 2010).

Having a set of core concepts and a unifying principle like mechanics in physics and evolution by natural selection in biology would propel social science to higher explanatory and predictive power. Although some scholars argue that multi-paradigm nature of social science is a “feature” and not a “bug” as social reality is intrinsically complex and highly contextual (Little, 2016), it still impedes cumulative growth of knowledge (Genov, 2019). As Porpora notes, in terms of Kuhn (2009), sociology is not a mature science, but a pre-science (porpora2015?). And, as Kuhn suggested, an appropriate activity in such a state is continuing the search for a discipline-wide consensus.

Turner (2001) suggested that social science is still at a stage similar to biology before Darwin’s theory of evolution. In biology, scientists have studied flora and fauna extensively, understanding their local habitats, predicting natural phenomena like frost damage to buds and soil improvement leading to better growth. This study resulted in the creation of “field guides” documenting this detailed knowledge. While these studies provided rich insights into individual species and their environments (like niche analysis), they did not explain how species originate and change over time. The unifying principle consolidated ontological pictures by explaining the mechanisms of emergence of entities like species. Further research and subsequent evolutionary synthesis enriched ontological picture of biology.

Just as biology relied on evolutionary theory to understand biological processes more broadly and universally, social science is currently using various methodologies to study specific societies, cultures, or other phenomena, but needs a comprehensive explanatory framework to address foundational questions about how these systems emerge and change and what does exist in the social realm. There have been candidates for such a framework, from older ones like structural functionalism (Parsons, 1937) and self-referential system theory (Luhmann, 1995) to newer ones like rational choice theory (Fararo, 1989; coleman1990?), critical realism (Elder-Vass, 2010; archer1995?), assemblage theory (DeLanda, 2019) and ‘analytical’ sociology based around the notion of social mechanisms (Demeulenaere, 2010; Hedstrom, 2005; Hedström & Bearman, 2009). All of them made valuable contributions, but none have become mainstream discipline-wide and, more importantly, outside of sociology.

Along with sociologists, philosophers of social science have been offering definitions of the scope and objects of sociology and broader social science, devising analytical truths about the social domain. Some of them claim that analytic social ontology based on conceptual analysis is prior to social scientific methodology (Epstein, 2016; Lauer, 2019; J. Searle, 1995). Others criticize this approach as detached from actual science and not “naturalistic” (Elder-Vass, 2007; Kincaid, 2024; Little, 2020; Ross, 2023; Sarkia & Kaidesoja, 2023). Their focus is meta-ontological rather than ontological as they wonder what it means to ask questions about ontology of social science and what are the way to arrive at an ontology instead of proposing or criticizing particular ontological pictures.

An important yet subtle distinction must be made before proceeding: that of “sociological ontology” and social ontology. The former is the inward-oriented study of the foundations of sociology and is closely connected with empirical and theoretical research within sociology. The works of critical realists (Elder-Vass, 2010; Lawson, 2019; archer1995?), Giddens (giddens1984?) and Bourdieu (bourdieu1990?) exemplify this strand⁵. The latter, social ontology is a science-independent philosophical study of the social entities started in analytic philosophy. It uses the conceptual analysis and tools like set theory, possible worlds semantics, ontological dependence and the like for which sometimes get criticized as detached from science. Although there is no such distinction in the literature, adopting one can clarify the contribution of the current work, which is to sociological ontology.

The remainder of this chapter is structured as follows. In the second section I argue that sociological ontology is the most reliable way of doing social ontology. I do that by defending inference to the best explanation (IBE) as a methodology for doing social ontology.In the third section, I argue that multi-paradigm character of sociology and the alleged lack of widely-accepted foundations is due to the lack of ontological constraints on concepts. There, I review major attempts at unification of sociology and identify key useful notions and a “checklist” for building social ontology.

What it means to talk about ‘social ontology’

A good deal of current social ontology is detached from actual social science (Sarkia & Kaidesoja, 2023). It is mostly developed in analytic philosophy (Baker, 2019; Epstein, 2015; Gilbert, 1992; J. Searle, 1995) and social scientific strand tradition of critical realism (Elder-Vass, 2010; archer1995?). There are views that social ontology is intrinsically more complex that any “natural” ontology for being “interactive”, or dependent on beliefs of individuals and hence changeable (Epstein, 2015; Hacking, 1999) and that social reality is inherently normative which means it cannot be descriptive and hence cannot import methodologies from “natural” ontologies (Ásta, 2024; Haslanger, 2012) which render and justify the importance of an “extra-scientific” study of the social world. To me, these are two distinct projects: a set of mutually coherent and explanatory powerful descriptions of social phenomena and a normative theory of thereof.

According to Epstein (2018), the field of social ontology divides into two distinct inquiries. The first one deals with the constituents of social entities and addresses the question “what is the social world composed of?”. The second strand of research is concerned with the construction of social categories, or kinds, and with the question “how do social kinds like money, borders, marriage and others get established?”. Individual people constituting a social group exemplify the former inquiry and children playing a game where stuffed animals have a tea party exemplify the latter (as they have “set up” the conditions for something to be a tea party) (Epstein, 2015, p. 57).

The difference between strands is in metaphysical relation of individuals to social facts. In the case of individuals and groups, social facts supervene on the facts about individuals, meaning that social properties cannot change without changing the individual ones. In the case of a toy toy party, facts about individuals set up the conditions for something to count as a social fact (2015, p. 58). One of the most famous expressions of the latter is Searle’s (1995) formula “X counts as Y in context C”. For example, “bills issued by the Bureau of Engraving and Printing (X) count as dollars (Y) in the United States (C)“ (1995, p. 28). Epstein calls the former relationship”grounding” and the latter “anchoring”, where grounding is responsible for instantiation conditions for a social fact or category like money, and anchoring is responsible for the mechanism of establishing these conditions. For example, in Searle’s case, collective acceptance of the fact X is such a mechanism⁶.

Such consideration exemplifies one of the two main views on the role of metaphysics⁷ in scientific inquiry, specifically in the sociological one. The first, empiricist one, renders metaphysics irrelevant and redundant or, if taken seriously, expects its constraints on models to impede the knowledge growth. The other view presupposes special metaphysical concepts beyond the vocabulary of science itself. For example, they discern grades of possibility and necessity largely unknown to science and go beyond causal dependence widely used in science (Ross, 2023).

Ross (2023) suggests the distinction of analytic and scientific social ontology, where the former is derived from conceptual analyses and the latter is based on empirical investigations. For Ross, ‘naturalistic metaphysics’ as he calls it, does not apply transcendental concepts except those featured in theories, models or explanations of “first-order science”. Ross defends what he calls ‘radically naturalistic metaphysics’ which is aimed at promoting scientific progress and might involve unifying or synthesising scientific discoveries.

Physics provides better contrast for comparing scientific and analytic metaphysics. Ontology of physics is formulated in mathematical terms requiring some technical proficiency. Formulating ontology of physics in an analytic way with “possible worlds” and “propositions” would be counterintuitive and unnecessarily difficult. It would require translation of already existing scientific concepts into the language of analytic metaphysics.

Social science does not enjoy straightforfardness of methematical formulations of its concepts⁸. Instead, it invites “folk” concepts like “beliefs”, “groups” and “social facts” in its ontologies. Further divide comes from the attitude towards this ambiguity. Some scholars argue that such “folk” ontologies are sufficient (thomassen2003?) and others put an interplay of “folk” and social scientific ontologies to the fore by bridging what (sellars1962?) called “manifest” and “scientific” images⁹ of phenomena within their social ontological theorizing (Guala, 2016; asta2012?; haslanger2018?). The ambiguity between “folk” and scientific social ontology might indeed depend on mathematical apparatus and its underdeveloped character in social science. Ross (2023) also notes that ontology construction most probably will require its formulation with mathematical frameworks sensitive to structural modeling (like hierarchical Bayesian inference¹⁰) and not set theory and similar tools from analytic metaphysics.

With the distinction of analytic and scientific social metaphysics at hand, let us look at possible way of arriving at social ontology in detail. Hawley (2018) assesses three possibilities: conceptual analysis, ameliorative (normative) approach and IBE. In terms of Ross, the first two fall into analytic approach and the last one falls into scientific approach to social metaphysics. From Hawley’s point of view, only the first two are potent. From my point of view, only the last one. Let us unpack and assess the possibilities.

Conceptual analysis and the “Standard model of social ontology”

The method of conceptual analysis involves using logic and intuition to dissect concepts to arrive at necessary truths about the world. Practitioners of this approach start with a priori theorizing, taking concepts they view as important and breaking them into parts. For example, a prominent exampe of this in social ontology is Searle (1995) who asks whether it is possible to be epistemologically objective about ontologically subjective issues. How can we know the truths about things whose existence depends on our representations or feelings, for example, about money, property and marriage? By analysing these distinctions of ontology/epistemology and objectivity/subjectivity, Searle arrives at an idea of a missing ingredient that allows for an picture of ontologically subjective entities, which is constitutive rules of the form “X counts as Y in C” as in “bills issued by the Bureau of Engraving and Printing (X) count as dollars (Y) in the United States (C)” (1995, p. 28).

Guala (2007) calls the approach to social ontology that uses conceptual analysis the “Standard model of social ontology”. This “model” describes the loosely constrained individualistic foundations of social phenomena and has three key elements: reflexivity, performativity and collective intentionality, as described by Tuomela (2002). According to Guala, most philosophers endorsing the “Standard model“ subscribe under all three elements. Reflexivity is a property of social entities to be largely comprised of beliefs about beliefs. There are I-mode and we-mode formulations of reflexive beliefs. Some philosophers say that initial and most basic beliefs comprising “the fabric“ of the social are essentially in We-mode and are not reducible to I-mode (Gilbert, 1992; Schmid, 2023; Tuomela, 2002). However, there are also more individualistic accounts of reflexive beliefs based on game theory (Bicchieri, 2005; Guala, 2016). Performativity amounts to social entities needing to be continuously maintained, performed or recreated. Collective intentionality, in its turn, is joint directedness of multiple individuals towards a phenomenon that contributes to its constitution. Collective intentionality tends to be presented either as a derivative of common knowledge and I-beliefs of the form “everyone knows that everyone knows that P“, where P is some social fact like social norm (Bicchieri, 2005), or as a primitive notion which makes common knowledge redundant. Moreover, there are attempts to naturalize collective intentionality by showing its irreducibility to individual intentionality (Gallotti, 2012; Rakoczy & Tomasello, 2007).

Concepts can be arbitrary and not reflect the structure of the world. However, concepts like Gilbert’s “joint commitment” (1992) or Tuomela’s “we-intention” (2002) are not completely arbitrary, as Hawley notes, for they reflect intuitive notions of the social world. Having access to intuition makes conceptual analysis, according to Hawley, rather virtuous in the context of philosophy of social science than in philosophy of natural sciences, for mind-dependent social world is presumably more epistemically translucent to its inhabitants than mind-independent matter.

To illustrate this point, Hawley uses Hacking’s distinction of interactive and indifferent kinds, where the former are involved in reflexive loops with beliefs and the latter are not (Hacking, 1999). Our classifications of the social world help establish and maintain it, whereas non-social objects are indifferent to our classifications of it. Nature’s objects do not change their behaviour given these classifications of them as opposed to social objects. This idea illustrates the notions of reflexivity and performativity characteristic for the “Standard model”. If social entities are comprised of beliefs about beliefs, their nature depends on these beliefs, and if beliefs change, social entities change accordingly. If social entities depend on beliefs about them, it is needed to constantly perform those to maintain them. To do this, it is needed that individuals have collective intentionality about these beliefs. For example, for money to be itself, it is needed for a relevant community to hold a collective intention to believe that certain physical entities can be used as a medium of exchange.

The three interconnected notions—reflexivity, performativity and collective intentionality—together with a method for arriving at them—conceptual analysis—presuppose that one cannot be mistaken about the nature of social entities, for they are posited a priori and based on intuitive concepts. Hence, if epistemic access is limited and truths are infallible, conceptual analysis cannot lead to viable social ontology only by itself. As Hawley sums up, conceptual analysis is still a useful methodology when used with caution, but it cannot securely lead to genuinely naturalistic social ontology and must be suppemented with empirical investifations.

I see the argument about relative translucency of the social world based on intuition unsatisfying and argue that social ontology cannot operate solely on the basis of conceptual analysis. I follow Kincaid (Kincaid, 2021, 2024), Ross (2023) and (churchland2019?) in that ontological claims based on conceptual analysis mistake the properties of the world for properties of the word.

Kincaid (2021) argues that current social ontology often builds on unfruitful analytic metaphysics, which ignores major 20th century philosophical results like the idea that science-independent knowledge solely by conceptual a priori analysis is problematic (quine1960?). Social ontology follows this trend, rarely using real social science research. Conceptual analysis, Kincaid maintains, can indeed clarify concepts, but cannot replace empirical knowledge of the social world. More broadly, he suggests that traditional projects of conceptual analysis testing necessary and sufficient conditions against intuitions will not tell us much about the sciences we want to understand as philosophy does not first decide what is ontologically real or essential before science begins because it is a scientific question itself (Kincaid, 2014).

(churchland2019?) goes even further and, drawing on (quine1960?), indicates that conceptual analysis as a methodology for making ontological claims is insufficient, for a concept reflects current beliefs and fallaciously mistakes their features for the features of the world. Churchland argues that a concept’s applicability to real-world phenomena depends on science and discovering facts. What is even more disturbing, she adds that often conceptual analysis yields not a reflection of a concept’s meaning but a camouflaged theory with ontological claims about some phenomenon. For instance, a theory that beliefs are semantic and require language (schooler2014?) which has emerged with help of conceptual analysis and no empirical evidence.

Ultimately, conceptual analysis and foundational metaphysical projects that try to “ground” social science in analytical truths (Epstein, 2015, 2016) might be useful to clarify concepts, but are non-starters for the task of identifying a theoretical core of sociology.

Ameliorative approach

The second methodological approach to social ontology which Hawley reviews is an ameliorative one by Haslanger (2012). Haslanger distinguishes between conceptual, descriptive and ameliorative approach to analysis of social categories like “race”, “gender” and others that can be extended to institutional categories like “marriage”, “borders” and others, as well. She makes another distinction between manifest and operative concepts, where the former is a theory individuals use to make sense of the term and the latter is a practice associated with it, what do people actually mean when they invoke a concept in the “wild”.

In Haslanger’s view, conceptual approach sheds light on “our” manifest concept of, say, “marriage” by inquiring into what do “we” take it to be. It might show how different understandings of the term “marriage” changed over time and across social structure. Descriptive approach circumscribes the operative concept of “marriage” by explicating, where possible, natural (chemical, biological, neurological) constraints on paradigm cases of the term studied. Lastly, ameliorative approach defines “our” purposes regarding the term and elucidates what concept would fulfill this purpose. An important addition is that in the last case “normative input” is required. The implication of this approach to social concepts is that if one has a descriptively accurate theory but does not like its normative outcomes, she is entitled to discourage it.

As Guala notes, Haslanger wants to put forward the idea that morally unacceptable institutions can and should be reformed (guala2016b?). However, as he notes, this position is in tension with realism, for the latter presupposes that reference of a category is determined by the world and not by “us”, which falls under Churchland’s criticism of a concept reflecting current belief and not a feature of the world. Ameliorative approach might be said to be a flavour of the “Standard model” as it shares the idea of reflexivity, or dependence of social ontology on beliefs of individuals. A shortcoming of this approach is reducing ontology to politics, which means that one should discuss and converge on “our” legitimate purposes regarding a studied term, which might not be self-evident thing to do.

Ameliorative approach does not allow fallibilism which means you cannot mistake about the structure of the social world, for communal consensus is the criterion for a concept’s applicability and ontological verification. However, if one can be mistaken about operative and manifest concepts of a social category, she cannot be mistaken about its normative input as she defines it. And since normativity is explanatory essential to ameliorative approach, it is not realist and cannot help identify a theoretical core of sociology.

Inference to the best explanation

The final strategy discussed by Hawley is inferring social ontology from the best social science. It takes mature and predictively successful theories to be an approximation to what there really is. This position is known as scientific realism (Psillos, 1999). Semantically, it takes scientific theories to be truth-conditioned descriptions of the structure of the world, meaning that they might be either true or false, and metaphysically it views the world as mind-independent and clustered into natural kinds¹¹. The semantic position of scientific realism implies fallibilism, for one’s theory might be mistaken about the world’s structure.

Hawley criticises this strategy, for, as she points out, there is still massive divergence in social scientific paradigms, from rational choice [#insertSource] to hermeneutics and Verstehen [#insertSource]. According to Hawley, social science is not mature enough and does not have an agreed theoretical core which can generate plausible predictions. The predictive capacity, along with maturity of science, is a prerequisite for inference from scientific theory to ontology, as argued by Psillos (1999), to whom the author refers. In Hawley’s words, “there are no social analogues of the gravitational waves or Higgs boson which bring physics to the headlines“ (2018, p. 191). For the reasons of immaturity and the lack of shared foundations in social science, she discards inference to the best explanation (IBE) as a methodology for arriving at social ontology.

The problem with Hawley’s position is that there would be no gravitational waves or Higgs bosons discovered without IBE in the first place. Of course, these discoveries did not bear massive ontological commitments and, hence, did not make any metaphysical claims. But it is further unification that allowed them to be put into a more coherent picture that philosophers can use to infer ontology from. In other words, IBE requires induction and empirical investigations in the first place to arrive at predictively successful and “mature” science. This goes in hand with Ross’s suggestions of ‘radically naturalist metaphysics’ which consists solely of first-order science and uses induction (Ross, 2023).

Saunders (2020) counters Hawley’s arguments with a case study from social epidemiology, namely income inequality hypothesis which enjoys predictive success and hence could inform ontology. Saunders critiques Hawley’s reliance on Kuhnian paradigms, suggesting that the sociological mechanisms influencing scientific consensus may not necessarily track truth. This critique highlights the complexity of establishing what constitutes a mature theory in the social sciences.

Scholz (2018) also critiques Hawley’s position and advocates for an inductive approach to metaphysics, emphasizing the importance of a posteriori sources of justification and inductive methods. This approach is rooted in the idea that metaphysical theories should evolve through empirical testing and reflection, rather than solely through abstract reasoning. Scholz highlights the necessity of making empirical assumptions explicit and testing them through inductive and abductive methods. While metaphysicians rely on concepts derived from intuitions, like social ontologies of J. Searle (1995), Gilbert (1992) and Tuomela (2002) which Scholz mentions, they still may depend on hidden empirical assumptions that could be fallible. Hence, it is more safe to explicate and empirically check them before postulating any ‘pre-scientific’, or analytical, ontology.

Guala (2016) emphasizes the close connection between ontology and explanation in case of using IBE:

Our best guide to ontology is provided by our best scientific theories. According to the widely accepted method of inference to the best explanation, we can infer what exists from the theories that best explain our observations. In light of this, we believe that those doing ontology cannot avoid being concerned with explanation. (Guala & Hindriks, 2015, p. 478)

How is IBE applicable to social ontology? The inductive strategy displays a directed relation between social ontology and social science—from the latter to the former. However, there is a controversy over which is primary. Drawing on Epstein (2015), Lohse (2017) and J. R. Searle (2008), Lauer (2019) reviews and critically assesses the argument that social ontology is prior to social scientific methodology, for it can promote success in prediction and explanation by providing a correct view of the social reality. For example, Searle notes that social ontology is prior to methodology and theory: “It is prior in the sense that unless you have a clear conception of the nature of the phenomena you are investigating, you are unlikely to develop the right methodology and the right theoretical apparatus for conducting the investigation” (2008, p. 443). Epstein expresses a similar concern: “Ontology has ramifications, and ontological mistakes lead to scientific mistakes. Commitments about the nature of the entities in a science—how they are composed, the entities on which they ontologically depend—are woven into the models of the science” (2015, p. 127). Ontology affects methodology in the sense of defining relevant aspects of a phenomenon under consideration which influences the way to approach it. However, it is hard to define these aspects a priori. Social ontology is not prior to social methodology, for relevant aspects of phenomena are subject to empirical testing, and this applies to social science, as well.

Little (2020) rejects the a priori stance and argues that social ontology cannot be justified on non-empirical grounds. Contra Lauer (2019), Epstein (2015) and J. Searle (2010), he holds that social ontology is the most abstract edge of social scientific theorizing, and that it is iteratively connected with empirical investigation and theory formation. This stance presupposes similar inductive process as in the natural sciences and in principle supports inference from social science to social ontology. However, Little explicitly rejects the “unity of scientific ontology” implicit in Hawley’s reasoning, who presupposes inference from a “theoretical core” of science. The former implies pluralism and case-by-case investigation of social ontology similar to Kincaid’s contextualism (Kincaid, 2021), and the latter implies monism regarding the focal set of entities and relations: there are “main” parts applicable across the board. Little denies the possibility of such a complete social ontology that might serve a foundation for social scientific reasoning and that “cuts the nature at its joints”. As he says, there are no such “joints” in social reality, meaning that there are no “atoms” or fundamental elements of it. Little also denies the idea that there might be a final list of entities from which the social world is composed and in terms of which all familiar social scientific concepts should be reformulated. In other words, Little’s realism is much weaker than scientific realism implied by Hawley, and this is why he sees no problem in accommodating realism for social ontology: these are just two different notions of the same title.

Weak scientific realism like Little’s cannot be a robust answer to the “what there is” question regarding the social world, but it can be a starting point if it uses IBE.

Is it possible to be a stronger scientific realist about social world? A helpful distinction here is that of natural and social kinds. Kinds are parts of the world naturally grouped together as opposed to classes which are grouped contingently. (mill1843?) connected kinds to scientific inferences and properties. Boyd (1991) redefined them as “homeostatic property clusters”, for properties are correlated and “packed” together. “Homeostatic” refers to the stability of correlations which have underlying causal mechanisms explaining them. Kinds with such homeostatic properties are real kinds. They support inductive inference and generalisation, and are studied by the sciences. Real refers to either independence from representations or being genuine.

Hacking (1999) introduced interactive and indifferent kinds, where former depend on our classifications of them. Guala (2016) notes that although interactive kinds are not real in the Mill-Boyd sense due to the lack of stability and historical contingency, they rather lack necessity than stability, for if described as patterns of strategic interactions in game-theoretic sense, as Guala does, behavioral regularities and equilibria are indeed stable.

The problem of social ontology might be reformulated as “are (some of) social kinds natural ones?” If yes and they indeed possess homeostatic properties independent from individuals’ representations of them, they might be candidates for “basic” ontological units of the social world. It helps to identify a condition for successful IBE, namely, that an ontological unit in question should be a real rather than interactive kind so that it supports inductive inference. But the next question is how to identify a real kind?

Individualism, holism and “levels”: debates within scientific social ontology

In philosophy of social science and theoretical sociology alike, there are debates on the ultimately “correct” level of description of social reality. Paradigms and research programs in sociology depart from different levels of description. One of the consequences of this is the well-known problem of micro-macro integration (Mouzelis, 2005): how micro-accounts of face-to-face social interactions relate to social institutions, states and world-systems. The problem of the “correct” level of description has been one of the key impediments for the development theoretical core of sociology and inference to ontological claims from it [#insertSource]. Both sociologists and philosophers tried to solve this problem throughout the 20th century, from “Grand theories” as Merton (merton1949?) has dubbed them (Parsons, 1937, #insertSource) to recent advancements (elder-vassa?). In philosophy of social science, this problem has been addressed as the debate between individualism and holism and consists of the question over the primary ontological level of description of the social (Zahle & Collin, 2014).

Notably, the notion of “emergence” has been employed as a remedy for the individualism/holism (or micro/macro) struggle in social ontology and sociology respectively. Although scholars mean different things by ‘emergence’, the idea of new levels and entities coming into being as a result of interactions of entities on “lower levels” has been used quite widely, from implicit into Durkheim & Fields (1995) claims about society as sui generis reality to explicit postulation of levels, emergence and even downward causation in critical realism (archer?) and so-called speculative realism (DeLanda, 2019).

We will not assess every possible option employing “emergence” as a strategy for reconciliation of the micro- and macro-levels. Instead, we will assess the very notion of emergence and conditions in which this term can help circumscribe a viable social ontology. It is a meta-ontological task which is supposed to help identify whether “emergence” can contribute to realist ontology of sociology as a set of core focal entities and mechanisms.

The notion of real patterns allows for realist special science ontology including sociology (Ladyman, Ross, Spurrett, & Collier, 2007; Ross, 2008).

ОСТАНОВИЛСЯ ЗДЕСЬ

References