Are structure and agency real?

---

<<<HOMEPAGE ||

---

Introduction

In my publication in the Journal for the Theory of Social Behaviour, I engage with two influential approaches to the structure-agency issue: constructivist institutionalism (Hay, 2016) and the morphogenetic approach (Archer, 1995). My paper deals with numerous points of agreement and disagreement between the two approaches, but in this blog, I will focus in on one particular question that arose in the paper: are structure and agency real? Archer and her morphogenetic approach say “yes”; Hay and his constructivist institutionalism say “no”.

---

Are structure and agency real?

The “structure-agency debate” is a long running and multi-facetted disagreement in the social sciences about the relation between society (as a whole) and the individuals who live within it. A host of controversies are grouped together under the “structure-agency” label. Sometimes, the debate is about freewill and socialisation as rival explanations of people and their actions. In other instances, the debate is focussed on a particular event or process, questioning the extent to which it has been caused by individuals and the extent to which it has been caused by social context. The structure-agency debate also entails methodological controversies (primarily, should researchers focus on individuals or should they focus on institutions, areas, trends etc.?) and epistemological controversies (can research focussed on individuals produce generalisable knowledge? does “generalisable knowledge” really tell us anything about individuals?). The remainder of this blog focusses in on another strand of the structure-agency issue, the ontological strand, and specifically the most basic question… are structure and agency real entities or merely invented analytical tools?

The reality of agency is often called into question by structuralist and determinist positions, which deny the creative powers and freewill of the individual subject, arguing that choice is merely an illusion. This denial of agency is based on the assumption that individual behaviour, social events and social change are the product of structural machinations (structuralism) and/or pre-determined chains of causality (determinism). In contrast, the reality of structure is often called into question by methodological individualists, who argue that ‘society’, ‘culture’, ‘institutions’ etc. are merely other people (see King 1999). From this perspective, we may be able to identify the reality of individuals and the things they have produced, but we are unable to identify anything resembling social structure; any of the usual identifications of social structure can be reduced to individual people, living and dead (King 1999).

The disagreement between structuralism and methodological individualism is interesting in many respects, but it is unsurprising that the former denies the existence of agency and that the latter denies the existence of structure. With the disagreement between Margaret Archer and Colin Hay, we have a more interesting situation. Both authors think that any serious social explanation must include both structure and agency, and both try to develop sophisticated models with which to explain their interaction over time. And yet, Archer claims that both are real and Hay claims that neither are real. From a critical realist perspective, Archer argues that structure and agency are both real layers of existence, which we analytically separate so we can explain their interaction and use them in social explanation. In contrast, Hay argues from a social constructivist perspective, postulating structure and agency as invented concepts that have analytical value but do not refer to any real entities or phenomenon. We shall look at each position in turn.

A critical realist view of structure and agency sees the two as emergent layers of social reality. Agency is the unique creative power of individuals that emerges from the lower-level parts of which the individual is composed (organs, tissue, cells etc.). Agency is said to be ’emergent’ (and therefore real) because it has causal powers that do not belong to the parts of which it is composed. Structure can be said to be emergent (and therefore real) in the same way, because it is based on lower-level parts (including individuals and their agency) and it has unique causal powers, which cannot be explained with reference to those parts alone (see Elder-Vass, 2010). Archer accepts these positions at an ontological level, but argues that we must impose an analytical dualism between structure and agency if we are to understand their interaction. This involves treating the two as if they were separate entities. Therefore, Archer accepts the critical realist understanding of structure and agency as emergent layers that are intertwined and overlapping, but imposes a further separation between them that is purely analytical.

In contrast, from a social constructivist perspective, structure and agency are not two different layers of reality but are instead two different socially constructed concepts for understanding the same single referent (i.e. ‘society’ or ‘social change’). Hay (2016) takes influence from Berger and Luckmann (1966) and Searle (1995), arguing that social structures are not independent of human thought, and therefore only exist as long as humans think in a particular way. The thought-dependence of social structures means that we cannot draw any obvious dividing line between the theory of social structures and social structures themselves. Having rejected this core critical realist distinction, it is a small step to argue that agency (both the experience and exercise of it) is inherently tied up with the same inseparable whole. Therefore, this would perhaps make the idea of structure and agency real (just as the idea of unicorns is real), but it would seem to exclude any existence for structure and agency beyond our conceptualisation of them.

While this blog post is already too long, and while it is up to the reader to work out their own stance in this debate, I will conclude with some of the points I put against Hay in my recent paper. I think that there are two interpretations of Hay’s position, and that the ambiguity between them is a result of Hay’s shift in perspective from critical realism to social constructivism.

(1) In his earlier (and more critical realist) work, Hay (2002 and 2005) theorises the existence of “structure-agency complexes”, and he analogises structure and agency as two alloys in a metal coin. These descriptions both imply the reality of structure and agency as real entities/phenomenon/causal-forces, and yet Hay consistently denies their reality. McAnulla (2005) and Callinicos (2004) point out that Hay therefore seems to deny the reality of any entities that are intertwined, and, as nothing exists in isolation, this position would seem to problematically deny the reality of any entity.

(2) In his more recent writings, Hay (2011 and 2016) seems to have turned to a more strongly social constructivist position that does not even accept the reality of ‘structure-agency complexes’. Therefore, Hay now denies the reality of agency (he uses the term ‘volition’) and of structure, not only because they are inseparable from one another, but also because both are inseparable from human thought. The point made by McAnulla and Callinicos could perhaps again be applied: are we denying the existence of entities merely because they are intertwined with other entities? However, a more significant problem arises… if structure and agency are merely analytical constructs that do not refer to real entities or even to a real entity, it is unclear why they are analytically useful at all.

I personally find the critical realist ontology compelling and therefore accept the reality of structure and agency. I therefore think it follows that the concepts are important and analytically useful. For those who deny the reality of structure and agency, there is some work to be done to explain why the structure-agency debate still matters, and why the concepts of ‘structure’ and ‘agency’ hold any analytical utility at all.

---

References

Archer, M. (1995). Realist social theory: The morphogenetic approach. Cambridge, UK: Cambridge University Press.

Berger, P. L. and Luckmann, T. (1966). The Social Construction of Reality: A Treatise in the Sociology of Knowledge. Garden City: Doubleday.

Callinicos, A. (2004). Making history: Agency, structure and change in social theory. (2nd edition). Leiden, Netherlands: Brill.

Elder-Vass, D. (2010). The Causal Power of Social Structures. Cambridge, UK: Cambridge University Press.

Hay, C. (2002). Political analysis. Basingstoke, UK: Palgrave.

Hay, C. (2005). Making hay … or clutching at ontological straws? Notes on realism, ‘as‐if‐realism’ and actualism.
Politics, 25, 39–45.

Hay, C. (2011). Ideas and the construction of interests. In D. Béland, & R. H. Cox (Eds.), Ideas and politics in social science research. Oxford, UK: Oxford University Press.

Hay, C. (2016). Good in a crisis: The ontological institutionalism of social constructivism. New Political Economy, 21, 520–535.

King, A. (1999). Against structure: A critique of morphogenetic social theory. The Sociological Review, 47, 199–227.

McAnulla, S. (2005). Making hay with actualism? The need for a realist concept of structure. Politics, 25, 31–38.

Searle, J. R. (1995). The Construction of Social Reality. London: Penguin.

---

<<<HOMEPAGE ||

---