Social Action in Quantum Social Science

Introduction

In its critical mode, Quantum Mind and Social Science offers a fundamental challenge to the materialist, determinist, mechanist, and individualist underpinnings of modern social science. Wendt is not alone in this endeavour. He joins a long tradition of post-classical philosophy and social theory that goes back to Whitehead and Dewey and forward to Deleuze and Guattari and Connolly. ¹ This is an important project and Wendt offers a clear, systematic, and convincing critique of the classical worldview. By discrediting the reigning view, Wendt lowers the burden of proof for his more controversial positive arguments. Although the title of the book suggests that Wendt articulates a quantum social science, what he offers are the building blocks of a quantum social theory. Social theories provide an account of agency, an account of social structure, and an account of how the two relate. For reasons of space, I focus on Wendt’s quantum theory of human agency, but I also briefly review Wendt’s account of how quantum agents are entangled with social structure.

Some of the most compelling evidence for Wendt’s quantum theory of social action comes from the rapidly developing field of quantum decision theory (QDT). ² QDT is striking because it predicts and explains most of the anomalies in rational behaviour that Kahneman and Tversky documented in the 1970s and 1980s. ³ This, for Wendt, provides prima facie evidence of the quantum consciousness hypothesis. After all, if humans did not have quantum brains, why would they behave in the exact way we expect agents with quantum brains to behave? Put more sharply, ‘it could be that reality is quantum at the sub-atomic level, classical at the brain level, and then quantum again at the behavioral level – but how likely is that?’ ⁴

Focusing on the role of QDT in Wendt’s account allows me to explore two central issues that I think get us to the heart of quantum social theory. First, to make the inference that the success of QDT implies a quantum brain, we have to believe that the mathematics of QDT actually maps humans and their social behaviours. There is one good reason not to suppose this: people working in this area are themselves reticent to say that their formalisms capture quantum processes in the brain. Wendt is upfront with the reader about this. ⁵ He asks us to bracket that problem and trust in the elegance of the claim that a quantum brain is producing mathematically legible quantum action. Or, if we aren’t willing to go that far, Wendt would at least like us to accept that the success of QDT shifts the burden of proof onto those who would argue that a classical brain could produce quantum behaviour. In either case, to go along with Wendt we need to accept that quantum mathematics faithfully maps human and social reality. It is worth interrogating this assumption. As I outline below, even without questioning the general applicability of quantification, there are some good reasons to question whether QDT maps what Wendt suggests it does.

Second, the claim that a quantum brain produces quantum behaviour presupposes a mechanism that links the brain to action. The argument here depends on an account of how social action emerges from the quantum brain. In exploring this, I show that Quantum Mind offers an image of an agent embedded in phenomenological, social, and temporal contexts. However, this image of agency already circulates widely in the social sciences, so it is worth questioning what the value-added of the quantum contextual agent is.

My main claim is that while Wendt is right that we need to challenge the classical underpinnings of social science, I am not convinced we should construct the alternatives under a quantum physical constraint. Such a constraint could limit the possibilities of post-classical social theory by leading its theorists to force their ideas into quantum forms. Quantum social science should be a key part of redefining and reshaping social theory in a post-classical age, but social scientists should not restrict their theories to naturalist categories drawn from the physical sciences. Instead, we can with Whitehead and others draw inspiration from the quantum revolution without being constrained by it. ⁶

Mapping Social Reality

In Parts I and II of Quantum Mind and Social Science, Wendt motivates the argument and lays out the quantum brain hypothesis. In Part III, Wendt moves from the quantum brain hypothesis to quantum social behaviour. Here, Wendt contends that the success of QDT provides evidence that human behaviour is quantum and that this implies that humans have quantum brains. To see the power of QDT, consider the case of the fictional experimental subject ‘Linda’:

Linda is first described as having been a philosophy major in college, bright, and concerned with discrimination and social justice. Then, the judges are asked whether it is more likely that Linda is (A) a feminist bank teller or (B) a bank teller. Classical probability theory tells us that B is more probable, since it encompasses A, but includes other possibilities as well. Yet, judges tend to pick A… ⁷

In short, judges commit the ‘conjunctive fallacy’. QDT explains this by arguing that judges perceive Linda’s attributes as ‘incompatible’ (in a quantum sense) with the category bank teller. So judges use the label feminist to modify bank teller and thus make it more plausible as an occupation for Linda than bank teller alone. In technical terms, ‘the mind projects itself first into the feminist sub-space… then “rotates” to consider the alternative – bank teller – subspace’. ⁸ Moving from the feminist subspace and then to the bank teller subspace makes it easier to imagine Linda as a bank teller. Classical decision theory cannot explain or predict the conjunctive fallacy, so QDT’s mathematical and phenomenological explanation constitutes a major advance.

Given Wendt’s constructivist past, some readers may be surprised by the extent to which Wendt leans on a mathematical description of human behaviour here. However, Wendt’s reliance on QDT is consistent with his defence of scientific realism in his first book. ⁹ For Wendt, if a theory is empirically successful, it must capture or depict reality in a meaningful way. Thus, if QDT explains human decision-making better than the classical model, this would suggest that humans themselves are quantum decision-makers not classical ones.

Exponents of QDT themselves do not make the leap from quantum models of behaviour to the quantum brain. ¹⁰ However, they do defend the mathematical pedigree of quantum probability in order to justify applying it to human cognition. In a prominent introduction to quantum cognition for an interdisciplinary audience, Pothos and Busemeyer point out that quantum theory was the result of intensive effort by ‘some of the greatest scientists of all time’. ¹¹ Moreover, part of the appeal of applying quantum probability to cognition is the ‘robustness of its mathematics’ which is applicable to ‘any science in which there is a need to formalise uncertainty’. ¹² But they stop short of making any ontological claims about whether the success of quantum cognition implies an underlying quantum entity. In effect, Pothos and Busemeyer seek to borrow the equations and pedigree of quantum theory without taking on any metaphysical baggage.

In either case, to be minimally intelligible as explanations (an account of why or how behaviour came to be) such arguments must make one of two ontological claims. First, in a general sense, one could maintain that mathematics is built into the structure of the world and so behaviour of all kinds conforms to mathematical patterns. I’ll call this the structural ontological claim. This seems to be Pothos and Busemeyer’s implicit claim: quantum probability is a well worked out mathematical theory that applies in lots of domains, regardless of their specific characteristics. Mathematics is the language of reality and so it is just not surprising that reality falls into recurring mathematical patterns.

Second, the claim could be ontological in a more specific sense: a successful mathematical model corresponds to a real entity (e.g. a brain) or process (e.g. computation) that performs the calculations or their functional equivalent necessary to produce those mathematical patterns. I will call this the isomorphic ontological claim. A variant of this claim underlies applications of decision theory or game theory to biological evolution. There, it is the process of natural selection that produces behaviour that looks like it was the result of a complex process of cognition or calculation.

Wendt relies on a version of the isomorphic claim. As such, Wendt is braver and more consistent than the quantum decision theorists who decline to put their ontological money where their models are. In appropriating QDT, Wendt presupposes that its equations describe the operations of a quantum entity – the human brain. How might we assess this claim? In order for the mathematics of QDT to imply a quantum brain there needs to be a clear fit between symbols and reality. How can we assess such a fit? In social science, quantitative measurement is assessed according to reliability (the method regularly produces the same results) and validity (the measure corresponds to a real entity we care about). ¹³ QDT seems reliable, but is it valid?

Responding to Pothos and Busemeyer, Lee and Vanpaemel suggest that there is a disjuncture between QDT models and any purported psychological mechanism:

There are striking disconnects between the descriptions of the formal workings, and their psychological interpretations. The former seem devoid of psychological content and the latter seem vague and verbal. For example, the projection lines onto subspaces that models the conjunction fallacy, with its mathematical precision, is “explained” as “a kind of abstraction process, so that the projection on to the feminist subspace loses some of the details about Linda”. ¹⁴

The objection here is an interpretive one: saying that these symbols mean x about the cognitive process is questionable. For Lee and Vanpaemel, QDT offers ‘impressive fits to data’, but this does not imply that the model ‘captures deep insights about the workings of the mind’. ¹⁵ So, while QDT is mathematically successful in the sense that it is experimentally reliable, it may not validly map onto brain processes.

However, Wendt’s argument does not rest solely on a conceptual correspondence between QDT and the quantum brain. Wendt points out that QDT offers a plausible mechanism that links brain and behaviour: ‘interference arising from the mind’s internal measurements on incompatible states’. ¹⁶ In experimental practice what this means is that when we measure individuals’ conceptual subspaces or semantic pathways we find that they tend to cluster and become separate (or ‘incompatible’). These subspaces ‘interfere’ with one another such that the probability of the union of two incompatible semantic pathways is less than the sum of the probabilities of the two individual pathways alone. ¹⁷ In the case of Linda, there is interference between the philosophy major/social justice subspace and the bank teller subspace that generates the conjunctive fallacy.

This is interesting but it does not go all the way down to brain processes. The interference property of semantic subspaces is a linguistic mechanism. So, it tells us how we get from language to behaviour, but leaves a gap from the brain to language. Wendt has a whole chapter on language, but he doesn’t ground this chapter in his account of the quantum brain. ¹⁸ Instead, Wendt reviews another exciting literature on quantum linguistics. But the quantum linguists are also reluctant to suggest the application of quantum probability to their work implies an underlying quantum brain. ¹⁹ So, they don’t provide a physical basis for their claims either.

In sum, while quantum cognition and quantum linguistics provide some suggestive evidence to support the quantum brain hypothesis, there are some missing links in the account. We have to have a lot of confidence in scientific realism and the power of mathematics to map human behaviour to make the leap with Wendt from quantum models of human behaviour to quantum reality. If we are not scientific realists, we hold realism to a higher standard, or we are generally sceptical about the ability of mathematics to accurately describe human behaviour, then we are unlikely to be convinced that QDT implies a quantum social theory.

But more deeply, I do not think that scientific realism should be the default epistemology of post-classical social theory. Realism itself is bound up with the classical episteme which divides the world into mind (ideas, theory) and body (matter, reality). ²⁰ This sets up a correspondence theory of truth that has been roundly criticised (as Wendt is aware). But scientific realism’s reliance on the very dualism that Wendt seeks to displace makes it unsuitable as an epistemology for a post-classical approach.

Moreover, the naturalist ethos of Wendt’s enterprise means that he accepts the basic principle that experimental results and mathematical descriptions of reality should govern what counts as a good account of social action. On this view, it is possible that some intuitive aspects of human experience may be rendered illegitimate even under the quantum physics constraint. Wendt prefers the quantum to the classical constraint because the former has more room for agency and ideas. But in borrowing the scientific authority of quantum theory for his own purposes, Wendt concedes that scientific findings should be the arbiter of social theory. Rigorously following the physics constraint would entail displacing the more complex, if imprecise, roles of human experience, philosophical coherence, and historical investigation in assessing the validity of the folk psychologies we might otherwise place at the basis of post-classical social theory.

Agency in Quantum Social Theory

Wendt uses QDT to develop a model of the quantum agent. Wendt builds this account on the claim that the brain is a quantum entity that maintains superpositions within itself and with the world. That is, humans are walking wave functions with various potentialities for action. An expression of Will collapses the wave function as action coheres in one form from an array of possibilities.

But why does this sequence ensure that human behaviour conforms to quantum theory in the way QDT describes? For Wendt the brain is a quantum entity just like a sub-atomic particle; it operates according to the same laws. But protons do not seem to make decisions or have Wills in the way that humans do, so there is some sleight of hand here. ²¹ How does a quantum brain conducting quantum computations produce the acts of a human will that in turn generate statistical regularities akin to the behaviour of sub-atomic particles?

Wendt, in effect, argues that the form of quantum measurement is replicated in human behaviour. Specifically, he contends that quantum processes are evident in the way language use and reasoning exercises are described in quantum linguistics and quantum cognition:

In quantum mechanics measurement is what brings about a wave function’s collapse, which is an inherently contextual process that involves first deciding what particular question to ask of nature and then preparing the experiment in such a way that it can be answered; if these steps are done differently, then a different result will be obtained. Similarly, in language what brings about a concept’s collapse from potential meanings into an actual one is a speech act, which may be seen as a measurement that puts it into a context, with both other words and particular listeners. ²²

Here, Wendt is focused on explaining how language use takes on a quantum form. He also points out that the Kahneman and Tversky anomalies exhibit a similar quantum-like pattern insofar as the order of operations can change the behaviour of experimental subjects. So Wendt relies on isomorphism between quantum measurement, his reading of quantum language processes, and quantum decision-making to backstop his theory. In short, what makes the quantum brain argument so elegant is that it assimilates so many phenomena into the same conceptual apparatus.

The image of quantum agency that Wendt provides here is striking. Wendt, using gendered language, calls this image the ‘quantum model of man’ and juxtaposes it to ‘classical man’: ²³

Quantum Man is physical but not wholly material, conscious in superposed rather than well-defined states, subject to and also a source of non-local causation, free, purposeful, and very much alive. In short, she is a subject rather than an object, and less an agent than an agency, someone always in a state of Becoming. Moreover, this agency is a process in and through which she is sovereign. She decides her present by how she collapses her wave function; she decides her future by projecting herself forward in time and enforcing correlations backwards, and to some extent she even decides her past, by adding to or replacing it in her practices. ²⁴

Here, processual agents with free will engage in creative action that alters both the future and the past through meaningful acts. These are the kind of agents I see revealed in, say, Talleyrand’s letters to Louis XVIII from Vienna or in Rachel Carson’s letters and papers. As a starting point for social theory, I find this image of agency appealing.

Wendt emphasises that his quantum account of agency is contextual. In doing so, he joins others in social theory and International Relations who embed agents in social, discursive, practical, and material contexts. ²⁵ Wendt outlines three contexts. First, there is the immediate phenomenological or interpretive context. ²⁶ This includes the stimuli, frames, and sequencing of elements that are excavated and manipulated in experimental psychology. Second, there is the broader social context of language, norms, and institutions. ²⁷ Wendt conjectures that human agents are entangled with these contexts through language. Third, there is the temporal context of the past (which is an active present) and the future (into and from which action is directed). ²⁸ We could add artefactual-technological contexts, biological-ecological contexts, and so on. This is all consistent with Wendt’s account. The key point is that the quantum agent is not a separable, individual, atomistic unit. Rather, the quantum agent is entangled with reality in manifold ways.

To get from this starting point to social science, we still need to translate this account of agency into a set of mechanisms that link agents to social structures in concrete ways. That is, what motivates agents and how do actions concatenate up into social and political outcomes? We might ask, how does the discursive constitution of agents and organisations work differently in quantum social theory? Or, under what conditions would we expect macro-structures like the international system to be stable or unstable in a quantum social world? These kinds of middle-range questions are beyond the scope of Quantum Mind and Social Science. Nonetheless, as Wendt moves from agent to structure in the book, he provides some suggestive hints.

In making the case for quantum game theory, Wendt explains what it means for agents to be entangled: ‘entanglement in quantum games corresponds to what sociologists would call the shared normative order that constitutes us as members of a society rather than as animals in a state of nature’. ²⁹ Wendt develops this by looking at a quantum game theoretic analysis of the 2008 financial crisis. In the analysis, Hawks ‘exhibit low entanglement with others and thus behave classically, whereas Doves exhibit higher, leading to more quantum behavior’. ³⁰ The analysis suggests that Hawks could be induced to reduce risky, destructive trading practices by raising their entanglement. We might increase Hawks’ entanglement through ‘moral standards, shared experiences in training, legal reforms, and above all education… [and] reducing the material incentives for classical trading’. ³¹ The claim here seems to be that classical behaviour is strategic and quantum behaviour is pro-social. This is, more or less, the same schema Habermas offers us with his distinction between system (the domain of strategic, means-ends action) and lifeworld (the domain of communicative action). ³²

The problem with this, in social theoretic terms, is that it is too restrictive and violates Wendt’s own account of the quantum agent. Wendt tells us that humans in dialogue are entangled with one another through language and context. ³³ What puzzles me is how the strategic trader can be more or less entangled with reality than any other agent engaging in a complex task. The trader, no less than the humanitarian activist, must be entangled in social, practical, temporal, and infrastructural contexts in order to act. So the division of the world into strategic and normative realms based on degrees of entanglement seems ad hoc and untenable.

Another potential problem is that Wendt uses his quantum social theory to replicate things that social theorists might already know (whether they agree or disagree). Wendt might respond that the value-added of quantum social theory is less to produce innovative theory than to bring social science into conformity with the causal closure of physics. ³⁴ This is a rather abstract value removed from the day-to-day work of social science, so some may not be motivated by it. However, Wendt makes the case that social scientists should care about it. He suggests that on the classical view, agents are individual entities, separate from other entities. If we remain within the strictures of classical social science, we’ll either be materialist individualists (denying the reality of subjectivity and ideas) or dualists (trying to manage material and ideational in an incoherent way). We might find dualism acceptable, but Wendt cautions us that the slope of arguments will continue to push in materialist, atomistic directions and idealist, holist understandings of the world will continue to be marginalised in the social sciences. ³⁵ So quantum naturalism puts sociality on a firm physical basis from which it can exercise lasting influence on the social sciences.

There is another possible value of the quantum approach that Wendt hints at, but does not develop. Wendt says that by providing a naturalist basis for consciousness and free will, quantum social theory creates ‘the possibility of giving the human experience a home in the universe’. ³⁶ For Wendt, the classical worldview ignores consciousness and denies free will. There is no place for our phenomenological experience in a determinist, materialist, mechanist universe. The quantum view, Wendt suggests, rescues subjectivity and places it at the centre of the cosmos. Here Wendt is gesturing towards the cosmological implications of quantum social theory. Since the 17th century, Western cosmology, and more specifically, the cosmology underlying international order, has been dominated by scientific ideas. ³⁷ For the most part, this has meant that concepts from classical physics and materialist biology define the fundamental units of the universe and the place of humanity in the cosmos.

Quantum mechanics challenges the conceptual and scientific underpinnings of classical physics and materialist biology. However, the cosmological implications of this challenge have not yet been translated into lasting change in social thought. Classical Newtonianism was easily transposed into political economy and Darwinism into anthropology. However, quantum mechanics has not yet been used to redefine our concept of humanity and society. As such, quantum mechanical ideas have had limited effects on political discourses. ³⁸ From this vantage point, Wendt is doing the hard work of translating quantum mechanics into social and political terms. This could help reshape cosmological discourses in the long-run if quantum social science takes off, informing policy-making and entering general public discourse.

In short, there are some good reasons to take up quantum contextualism: it might advance theory, give social science a naturalist basis, or change cosmology. However, at the moment the most concrete contribution to social science – advancing theory via mechanisms linking action to social or political outcomes – is unclear and ‘to be determined’. ³⁹ The other two – naturalism and cosmology – are interesting and important, but removed from the day-to-day work of most social scientists. In evaluating Wendt’s wager, some may opt to borrow his insights into the operations of the contextual agent without adopting its underlying quantum physical basis.

Conclusion: Contextual Agency without a Physics Constraint

Read as a theory of social action, Quantum Mind and Social Science is a natural and worthy successor to Wendt’s 1999 book, Social Theory of International Politics. Social Theory brought the central questions in the philosophy of action into IR. But, for Wendt, Social Theory relied on mind-body dualism in an unsatisfying way. ⁴⁰ So Quantum Mind and Social Science can be read as way of trying to overcome that dualism in an intellectually coherent, naturalist way. Nonetheless, Quantum Mind and Social Science maintains continuities with his earlier work. Wendt remains a committed scientific realist, holist, and constructivist. That said, he has moved towards a contextual view of agency that is distinct from the normatively constituted agents Id and Ego that featured in Social Theory. I find the image of contextual agency compelling and think we should rightfully place it at the centre of our social theory. But throughout my analysis here I have raised questions about whether the quantum view of contextualism is convincing or necessary.

Quantum Mind and Social Science is a gambit. Even though Wendt cannot prove that he is right, he bets that he is. Although he would love to convince us of that, Wendt would settle for all of us treating quantum social theory as a live option among available social theoretic approaches. That is, he presents a systematic treatment of the idea that humans are walking wave functions that he hopes we will give serious consideration. Wendt’s task here is made easier by the obvious problems with the classical worldview. I agree strongly with Wendt on this point: there is no reason to privilege a classical model of the social. Many of us have been working without it anyhow.

I am convinced that quantum social theory should be part of the movement to replace classical social theory. But I don’t think that post-classical social theory should be equated with quantum social theory. By Wendt’s own admission we do not need to accept the quantum brain hypothesis to get a contextual theory of agency. We can, as many have done, ground contextualism in some mix of folk psychology, philosophy, anthropology, experimental psychology and biology. Folk psychology does not sound all that great as a foundation for social theory, but it has served social theory well. In some cases, folk psychology has been ahead of empirical research. For example, in the 1920s, Dewey articulated a view of human agency as structured by stable heuristics designed to reduce uncertainty. ⁴¹ This has been borne out by experimental psychology. ⁴² If Wendt is right about quantum contextualism, then contextualist social theory has been ahead of the experimental research too.

This is emphatically not an argument for excluding the natural sciences from informing our understandings of human agency. Theories of biology, physics, and mind can and should inspire new visions of human action, society, and nature. Indeed, we need insights from the natural sciences to help break the hold of classical thinking on folk psychology. But there is a difference between articulating social theory in conjunction with the natural sciences and articulating that theory under a naturalist, physicalist constraint. The danger in accepting QDT, quantum linguistics, and quantum game theory as the core of post-classical social theory is that we will always be working from mathematical symbol sets to human action, instead of the other way around. The temptation will be to shoe-horn our experiences into conformity with symbolic logics or trendy experimental findings. But as I pointed out above, this might be a fraught interpretive exercise, especially for those of us not trained in the mathematics of quantum probability. We can adopt a contextual view of human agency without having to constantly work our account of the social into the terms and mathematics of physics.

Wendt might say that by declining to adopt a quantum physical basis, we are de facto embracing a classical worldview. If we define classical as materialist, determinist, mechanist, and individualist, that is clearly not true. There is a long tradition of social theory outside these strictures. ⁴³ This tradition declines to operate under physics ontologically or epistemologically. This does not mean that an alternative ontology would be divorced from physics or built on fantastical ideas. Whitehead, for example, was motivated by the quantum revolution to articulate a post-classical cosmology to displace Newtonian cosmology. ⁴⁴ But when Whitehead turned to create, he did not constrain himself to that or any other scientific vocabulary. Nor did he work ad hoc, freely combining and recombining elements. Instead, he sought to build a metaphysical account of the universe that was internally consistent and consistent with other things we know from physics, biology, philosophy, anthropology, and human experience. With Whitehead and others, we should strive to create an elegant, post-classical account of human action in this way. In doing so, we must draw from the natural sciences. But to build social science under the physical constraints of quantum mechanics limits our perspective and ties our enterprise to an uncertain, although successful, body of knowledge that is notoriously hard to interpret.