Ron Mallon aims to provide an account whereby a human kind can be both socially constructed and real. His book has much to recommend it. It is careful, scholarly, clear and tackles important issues. The book is aimed at philosophers with a special interest in social construction. The text is clear enough to be understood by advanced undergraduates, but an acquaintance with the relevant literature is assumed. Only those who are already familiar with, say Ian Hacking’s work on human kinds, or with some of the debates about the social construction of gender and race kinds, will be able to make much of this book. But for readers who are acquainted with these texts, Mallon’s work is very useful.

 

 

The book starts with a long chapter that considers whether concepts of race can fairly be said to be of comparatively recent origin. People have, of course, always been aware that humans come with different skin colors, but the idea that such superficial markers might indicate deeper, inherited differences is claimed by some to be recent. Mallon looks at evidence from evolutionary psychology and argues that essentialistic thinking about types of person is likely supported by some universal psychological mechanism. Such thinking might be curtailed in certain cultures, but the underlying tendency to think in such ways is there.  He thinks that concepts of race are unlikely to be modern. Still, some significant shift in race kinds likely occurred comparatively recently.  Mallon suggests that the shift was plausibly not conceptual but was perhaps “political, institutional, or technological” (p.47). Readers with an interest in history might want to hear more about what the political, institutional, or technological shift might be. Here and elsewhere Mallon’s book is very much a philosopher’s book. He is much more interested in constructing an account of how socially constructed kinds might possibly work than in discussing how they actually do work in the real wold. While this may be a perfectly legitimate project, it won’t be to everyone’s taste.

 

 

Chapters 2 and 3 address the issue of how it is that human kinds, like race and gender kinds, might be both socially constructed and also causally powerful. Mallon points out that much existing literature (for example, Hacking’s work) concentrates on the role of descriptions in shaping intentional actions. Mallon thinks that it is important also to consider the ways in which representations shape not only intentional, but also automatic, processes (as shown in work on implicit bias, for example). The role that representations play in shaping social roles and social institutions (which can develop over time to become entrenched and hard to change) also plays a key role in Mallon’s account of the ways in which human kinds, once socially constructed, can be expected to develop robust causal powers. Chapter 3 ends with an interesting and provocative thought experiment. Mallon considers what would happen if people suddenly forgot their beliefs about human kinds:

 

if we could turn on our Neurosemantic Eraser Ray, blasting an entire urban area so as to erase, say, sexual or racial conceptions and stereotypes from people’s brains, producing localized conceptual deficits, to what extent would the categories themselves cease to be? Where a social role is sustained almost exclusively by psychological mechanisms and behavior, the Eraser Ray may be enough to erase the category as well. In contrast, where there is a great deal of environmental construction we should expect that our collective forgetting would be followed by people reconstructing those concepts and stereotypes from stored media repositories of inherited culture, from their own experience, and from the many material facts that encode them. In such a context, concepts would seem to be needed to explain the many distinctions that would continue to co-vary. And public theorizing about such differences might, in turn, give rise to common knowledge of them. If we magically lost the idea of sex or of race, it is plausible that many of us would have to reinvent it in order to understand the social realities that we inhabit. (p.89)

 

 

 

Chapter 4 is an interesting chapter on representation and moral hazard. Social constructionists often claim to have some sort of moral superiority over those who think that race and gender kinds are natural, biological kinds. Mallon considers whether there are any grounds for thinking that it is morally or politically better to believe a kind socially constructed. He considers psychological evidence that suggests that people have a tendency to think that when a human behavior is ‘natural’ this reduces moral responsibility for the behavior. Even if it a logical error to think in such ways, the fact that people do think in such ways may create a reason for not broadcasting claims that some behavior is ‘natural’. Actors who entertain such beliefs, for example that it is natural for men to be sexually unfaithful, may try less hard to control themselves, and so behave less well. “These inferences themselves may be unsound, but the crucial thing is that, insofar as subjects make them, representations may result in a reduction of attribution of responsibility, producing moral hazard” (p.106). Mallon concludes that even when it is true that some behaviour is ‘natural’ it might sometimes not be right to assert such a claim. Mallon’s position here is interesting. To my disappointment he doesn’t give much guidance as to how epistemic and moral/ political considerations might be weighed against each other. He says,

 

This question probably has no general answer. In some cases, it can seem obvious that one or another value is overriding, but resolution plausibly involves real conflicts and tradeoffs where respecting one value means ceding satisfaction of another” (p.110)

 

While doubtless true, there is little guidance here for the politically concerned naturalist as to what they should actually do. It also strikes me that Mallon’s ‘moral hazard’ concern might in some cases go the other way. Nice actions as well as nasty actions might be thought ‘natural’. If the assertion of claims like ‘It is natural for men to be unfaithful’ might increase the chances of men being unfaithful then equally the assertion of claims like ‘It is natural for people to be compassionate” might increase the chances of people being compassionate. If Mallon’s general line of argument works here, then surely sometimes naturalist, rather than social constructionist, claims will have the consequentialist advantage.

 

 

Chapter 5 considers a potential puzzle for those social constructionist accounts that claim that gender, or other human kinds, are constructed through performance. If actors perform what it is to be a woman, for example, how come they are not generally conscious that this is what they are doing? Mallon suggests that our ability to correctly attribute the causal role that is played by our mental states is dependent on the correct psychological theory being available to us. We “construct a causal explanation of our thoughts and behaviors on the basis of our background theories about what is and is not a plausible cause. The result is that we might decide, along with others in our epistemic community, that our mental states and behaviors have completely different causes than they in fact do” (p.125). So, for example, suppose my belief that women can’t throw balls very far leads me to throw a ball weakly. I see that the ball I have thrown only went a short distance. Like others in my community I think that women are naturally weak ball throwers. This leads me to suppose that the cause of my poor throw is natural weakness, and I fail to recognize the causal role played by my expectations and beliefs. In such a situation, I might perform womanly ball-throwing without knowing that this is what I am doing.

 

 

Chapter 6 considers possible conflicts between various types of social constructionist claim and realism. The chapter is clear and helpful. Mallon shows that many of the most plausible forms of social constructionism are compatible with realism. The chapter would make an ideal reading for higher-level undergraduate courses concerned with such debates.

 

 

Chapter 7 considers whether constructed human kinds can be sufficiently stable to ground successful explanations and prediction. Famously, in his work on the ‘looping effects’ of human kinds, Ian Hacking worries that human kinds can be expected to be radically unstable, such that scientists who seek to study them will always be chasing a moving target. Mallon stresses that some social processes can be expected to stabilize human kinds. Labelling effects mean that humans will sometimes tend to conform with expectations. Social institutions can also often lead to certain ways of behaving becoming entrenched. Mallon thinks that cases will vary and that sometimes stabilizing social forces will be more powerful than destabilizing forces.  The chapter ends with an appeal to evolutionary considerations that I find less than unconvincing:

 

evolutionary considerations suggest a plausible ultimate explanation for social stability rooted in the imperatives of social life: we need to coordinate and cooperate with others, and we typically have little meaningful option to ‘opt out’ of the cultural milieu in which we find ourselves. It makes sense that we would have a range of capacities that would enable us to find a role within the social ordering in which we find ourselves, and these capacities would lend themselves towards signaling that we belong to a role and understanding the roles signaled by others instead of toward undermining the destabilizing those roles. (P.180)

 

Here it seems to me that Mallon has overlooked a key fact that has been emphasized in Ian Hacking’s work: the human sciences have a special interest in forms of human deviance. When disordered, disruptive, or criminal behavior is being studied, it’s not at all clear to me that we should expect evolutionary pressures towards coordination and cooperation to be playing much of a role. Some types of deviance plausibly arise through evolutionary dysfunction. Others might be a result of frequency-dependent evolutionary strategies that emphasize cheating rather than cooperation. I think that Mallon’s general expectation of stability is insufficiently grounded.

 

 

Chapter 8 examines how a constructionist can account for the reference of human kind terms. If social constructionism is right, say about gender, then lay people are radically mistaken about the nature of gender terms. Social constructionists want to say that ‘woman’ refers to a socially constructed kind although people believe it refers to a natural kind. However, in other cases where lay beliefs about a kind are radically wrong we conclude that there has been reference failure. Thus, ‘witch’ failed to refer because the old women that lay people labelled witches failed to have the properties ascribed to them. How then, can the social constructionist say that ‘woman’ refers to a socially constructed kind, rather than being forced to say that ‘women’ don’t exist? The chapter is clear and helpful, and shows how a modified causal account of reference might be hoped to deal with this difficulty.

 

To sum up, The Construction of Human Kinds is a valuable addition to the literature. The book will be accessible to readers with a good undergraduate background in philosophy, and should be read by philosophers who want to understand the metaphysics and semantics of socially constructed kinds.

 

 

 

© 2017 Rachel Cooper

 

 

 

Rachel Cooper, Lancaster University