It is increasingly common to hear social categories like gender or race described as “socially constructed,” or as “social constructs.” What people mean when they say that gender, for instance, is a “social construct” varies considerably. There is a sense in which “social construct” is used to describe a phenomenon with a diminished ontological status relative to other phenomena. That is, the idea that gender is a social construct means that gender is less “real” than some other category like biological sex, or “brown-eyed,” or “Tay-Sachs carrier.” There is a notion first that social categories are less ontologically stable than so-called “natural” categories, and second, a notion that certain social categories are less ontologically stable than others. The sentence “gender is a social construct,” in one sense that it is often used, is meant to imply that gender is not an objective status, that it does not exist without a social consensus that it exists and of what it consists. In the broad literature that employs social construction in this, and many other senses, we also encounter resistance to the notion that a phenomenon’s dependence on a social context indicates that it is any less “real” than other things we might intuitively think of as “objective.” Catharine MacKinnon, for example, writes,

“Epistemologically speaking, women know the male world is out there because it hits them in the face. No matter how they think about it, try to think it out of existence or into a different shape, it remains independently real, keeps forcing them into certain molds. No matter what they think or do, they cannot get out of it. It has all the indeterminacy of a bridge abutment hit at sixty miles per hour.” (MacKinnon 1989, 123).

MacKinnon’s point is that gender is independently real in the same sense as a bridge abutment because no “individual or community of individuals can simply think them out of existence” (Haslanger 1995, 96). Despite this seeming resistance to revision or elimination, for a category to be “socially constructed” in any sense that term is employed, then that category must be dependent on a social context in some substantial way. That is, without members of a community for which a social category designates a grouping of persons, such a category does not exist. It is easier to think about this in terms of less controversial categories. In a context with no institution of marriage, for example, no “wives” exist. In a context without a higher education system, no “professors” exist. To meet the qualifications that define a person as a “wife” or a “professor,” certain agreed-upon rules have to obtain in a particular social setting. Those categories depend upon the institutions to which they refer. The shape of that agreement, what it means for social rules, expectations, assumptions, constraints, privileges, and obligations to hang together despite objection to them by individual persons or groups of persons, is a large part of what we mean when we refer to “social construction.” As MacKinnon deftly indicates, this seemingly independent “hanging together” is akin to a kind of objectivity. The existence of social categories or my being sorted into one or more category does not depend on my consent, my mental states, or my beliefs about them.

In this chapter, I have two aims. First, I develop a working concept of objectivity with regard to social categories. Second, I show how that concept of objectivity illuminates the relationship between social categories and processes of social domination and political coercion. Taken together, these form the basis of an argument for a general political model of social classification. Regarding the first of these objectives, I offer three premises:

P₁ : Social categories are socially constructed

P₂ : Certain social categories designate uneven power relationships

P₃ : Social categories track political, as opposed to natural, facts

I proffer arguments for the philosophical veridicality of each of these premises so that they may perform as given as the scope of the political theory of social classification expands. Without substantiating these premises, the arguments supporting a philosophical model of social classification struggle to get off the ground, and taking these premises as simply stipulated would render such arguments unconvincing to all but the most devout social constructionists. By developing these premises, I intend to show that the objectivity of social categories depends on social consensus or, more often, social coercion. Though social categories cannot be revised or eliminated by mere “desire, thought, or assertion,” they can be, and frequently are revised and eliminated through social and political processes. These changes are often slow and subtle, driven unconsciously by historical and material circumstances, such that persons who are embedded in the contexts in which the relevant categories operate are not aware of the changes. However, by recognizing the shape and direction of social category revision, it is plausible that we might coordinate political action to deliberately reach more desirable and equitable outcomes than the ontological status quo.

P₁ : Social Categories are Socially Constructed

I intend, in using the phrase “social construction,” the generic sense that “something is a social construction if it is an intended or unintended product of a social practice” (Haslanger, 2012, 86). As I mentioned, the category “wife,” for instance, derives its meaning from the social practice of marriage, and the institution which is an artifact of that practice.^[1] That is to say, the conditions for being a member of the category ‘wife’ include social properties and relations. The category ‘wife’ does not operate in the absence of a social network or context. Making the claim that certain other categories, such as gender categories, are socially constructed, may appear less obvious to certain philosophical dispositions. Accepting such a claim will usually turn on whether or not one understands gender categories to be social as opposed to natural categories. And this distinction in turn depends on whether we evaluate the claim that certain categories are constructed in a causal or constitutive sense. Something is “causally constructed iff social factors play a causal role in bringing it into existence, or, to some substantial extent, being the way it is” (Haslanger 2012, 87). The idea that e.g., gender categories are causally constructed is a stronger claim than a notion of constitutively constructed gender categories. A category is “constitutively constructed iff in defining it we must make reference to social factors.” Classification schema play a causal role in establishing social categories. That is, classificatory practices may do more than “map preexisting groupings of individuals.” Collectively, social contexts produce “attributions” which “have the power to both establish and reinforce groupings which may eventually come to ‘fit’ the classifications” (Haslanger 2012, 88).

P₁ on its own is self-referential, for something to be a social category, it is a category which is socially constructed, i.e., a category which is an “intended or unintended product of a social practice.” Embedded in P₁ are further specifications—I take gender and race categories to be categories which depend on social properties and relations, and not, as would be the case with so-called ‘natural’ categories, socially-independent facts.^[2] For this premise to do any work, I must clarify which categories are ‘social.’ I am opposed to the view that, the category ‘man,’ for instance, corresponds solely to a physiological or biological designation, i.e., that the gender category ‘man’ merely tracks features pertinent to the sex category ‘male.’ (I similarly reject the notion that the race category ‘Black,’ for instance, simply tracks a biological feature such as melanin content of the skin or region of ancestral origin.) There is a wealth of literature, both empirical and theoretical that develops the severability of gender and sex categories. Generally speaking, the prevailing view in social philosophy is no longer Beauvoir’s adage that “gender is the social meaning of sex,” but that gender and sex designations are frequently connected but not connected by necessity. Speaking purely anthropologically, there are multifarious variations across cultures in terms of the number of gender categories that operate, the kinds of gender categories that operate, and the extent to which they correspond to a binary sex classification scheme. There is no single biological feature which unifies all ‘women’ or all ‘men.’ That is, we cannot identify one biological attribute that is shared by all the persons we would, given our social context and any number of other features that they possess, describe regularly as a ‘woman’ or a ‘man.’ However often we may accurately describe ‘men’ according to secondary male sex characteristics, chromosomes, or natal genitalia (that is – however often the sex designation ‘male’ may match the gender designation ‘man’), there are enough exceptions to rule out the possibility that any of these features irreducibly define ‘men.’ The designations for biological sex categories and the social features which define genders can, and often do, come apart. For the majority of social theorists, social scientists, and medical professionals, this is no longer a controversial claim.

Read more