Regarding a putative kind X (mathematical objects, artifacts, races, theoretical entities, etc.), there are three different metaphysical questions: do Xs exist and, if so, what do their existence depend on?, why are they X?, and why do we make a difference between Xs and Not-Xs?
The first one is the issue of whether certain kind of objects exists and, if so, whether they are natural, socially constructed, fictitious, etc., the second is the question for the metaphysical ground of certain truths or facts, and the third is the issue of whether or not our concepts cut reality at its seams. The questions are different and, for the most part, independent.
Thus, for example, in the philosophy of physics, it is important to distinguish between the questions:
- Do forces exist, i.e., do the entities in the extension of the predicate “(is a) force” exist, and if so, are they part of the fundamental furniture of the world or does their existence depends on the existence of other objects or facts regarding, for example, human conventions, the constraints of cognitive architecture, cultural perspective, personal preferences, etc.?
- Why are forces forces, i.e., what makes the objects in the extension of the predicate “(is a) force” be forces? Is there a fact of the matter whether or not something is a force or not, or does it depend on our conventions, cognitive architecture, cultural perspective, personal preferences, etc.?
- Why do we make a distinction between forces and non-forces? Does the concept “force” cuts reality at its seams or does it only make sense for our social practices, cognitive architecture, preferences, etc.? Is it a natural kind or is it socially constructed, fictitious, etc.?
Notice that the third question is not why a particular entity x is an X rather than a non-X (that would be the second question), but the question of why do we make a difference between the Xs and the Ys. Thus, the second question is independent of the first in so far as it applies to all sorts of entities: fundamental, derived, fictitious, socially constructed, etc. However, it is not completely independent from the second because, even though socially constructed properties cannot be natural kinds, not all natural properties are natural kinds, and thus it makes sense to make the further question of why do we mark certain differences and not others. Whether the Xs are natural, fictitious, socially constructed, etc. it is still a further question whether the concept X is joint carving or not.
The third question is a question of joint-carving, because, presumable, if the answer to a question of this third kind necessarily and substantially appeals to our practices, cognitive architecture, preferences, etc., then the distinction between Xs and not-Xs will not be joint carving.
Social Constructivism in Metaphysics
Being a realist (or a fictionalist or a social constructivist, etc.) regarding a type of objects means different things depending if one is taking a position regarding each of these three questions. Consider the first question: If one is a Quinean, being a realist regarding the Xs does not mean much else besides believing that the Xs exist, and that must be the end of it; but if one is an Aristotelean, one may still wonder whether the X are fundamental or not (or not all, but only some) and if they are not (all) fundamental, what does their existence depend on. If one believes only things with objective existence are real, then one will reject as real those entities that exist, but whose existence depends on human conventions, the constraints of cognitive architecture, cultural perspective, personal preferences, etc.
Consider some examples. We usually make a distinction between socially constructed entities like words, passports, baseball bats and nations on the one hand, and not socially constructed entities like neutrons, lumps of coal, and clouds on the other, with entities that are difficult to classify such as domestic cats and melodies. What makes a nation socially constructed is that its existence metaphysically (not just causally) depends on certain human social actions and/or practices. Clouds, in contrast, are usually considered to exist independently of our social actions and practices. We have done nothing to make them exist and they could have existed even if society had never developed on this earth. This is a distinction at the level of question one, since it concerns the existence of entities of a certain sort.
Regarding the second question, we usually make a distinction between socially constructed facts like fanny packs being uncool and Paris being the capital of France, and not socially constructed facts like every planet with an atmosphere having clouds or Mauna Kea having an altitude of 4,205 meters. Again, we have cases that are difficult to classify like the seventh note of a musical scale being its leading tone or tomatoes tasting good. In this regards, we say that Paris is the capital of France is a socially constructed fact because part of why this is actually so is because of our social practices, institutions and actions. In contrast, the altitude of Mauna Kea is putatively independent of our social constructions, that is, nothing we have done or could have done could have changed its altitude (without causally affecting its physical reality, like for example, bombing its top off!).
Notice that socially constructed facts could involve both socially constructed entities – that is, entities whose existence is socially constructed – and not-socially constructed entities; just as socially constructed entities and not socially constructed entities could both be involved in not socially constructed facts. This is why I have insisted that the socially constructed nature of entities is independent of the social construction of facts. For example, it is a socially constructed fact that diamonds are precious stones, even though the existence of diamonds is not socially constructed. On the other hand, it is not a socially constructed fact that Paris is rainy, even if Paris is a socially constructed entity.
Now, a property is socially-constructed if having it is a socially constructed fact, and not otherwise. Thus, being a precious stone, or cool, or expensive, etc. are socially constructed properties; being rainy, having certain given mass or being located in certain spatio-temporal point are not socially-constructed properties.
Social Construction and Language
One might argue that since Paris would not exist were it not for our social practices and conventions, it could not be rainy without them either. Fair enough, so a more precise formulation of what makes a fact socially-constructed would be that a fact is socially constructed if it depends on our social practices and actions for more than just the existence of the objects involved.
Similarly, one could argue that Mauna Kea having an altitude of 4,205 meters is socially constructed since there would be no such thing as meters if not for our social practices of measurement. However, this would be a mistake that is easily avoided if one is careful to make a difference between predicates (in language) and properties (in the world), in particular, between how we use a predicate to fix a property and the property thus fixed. A good way to illustrate this example is through a joke:
Suppose there is a young girl standing just to the left of a huge rock, half buried in the ground (the rock, not the girl). She claims to be able to move the rock from her right to her left side without using any tool or machinery, and willing to bet half a thousand dollars to prove it. Furthermore, you are allowed to try moving the rock yourself so that you can verify that it is not a trick rock or something else pretending to be a rock, etc. After trying to move the rock, you confirm that it is heavy indeed and half buried to the ground. So you agree to the bet. Once the bet is set, she turns around on her place 180 degrees. “Now – she says – the rock is no longer to my right, it is to my left”. Thus she wins the bet.
The joke is funny, presumably, because of an equivocation in the expressions “to my right” and “to my left” as uttered by the cunning girl. The rock did not actually move, in so far as it did not change location. Yet, it is true that it was on the girl’s left side and now it is on her right side. This is because we usually use the expressions “to my right” and “to my left” to refer to spatial locations using ourselves as point of reference. However, we can also use them to talk about our spatial relation to such spatial locations and the objects that occupy them. When the girl claimed that she could move the rock from her left to her right, we assumed her to be using those expressions in the first way: we assumed she was using herself as a point of reference to fix a couple of spatial locations, not in the second sense. Thus, we believed she was going to change the location of the rock, not her spatial relation to the rock.
This joke illustrates the importance of making a distinction between the property we talk about and how we fix such property. When we use “to my left” to talk about the location of an object, we use ourselves as props to fix the spatial property, but we are not part of the property. Consequently, whether an object has such property does not depend on us. In contrast when we use “to my left” to talk about our spatial relation to an object, we place ourselves in the property, so to speak. Something similar happens when we use expressions like “four days”, “4,205 meters”, etc. We use social conventions to fix the properties corresponding to these predicates, but the conventions themselves are not part of the properties expressed. Thus, we can truly say that many years had passed before we developed the convention of measuring time in years; and that Mars was already million kilometres from the Earth before the development of the metric system. Furthermore, we can also say that Mars would still be that far, even if we had never developed the metric system, for the former fact is independent of the later.
Thus, we could talk about using predicates like “millions of light years from the Earth” or “three pounds” to socially fix properties that are not themselves socially constructed. These properties are not socially constructed because what makes an object being millions of light years from the Earth or weighting three pounds is not any social convention, practice or anything similar. None of our social practices put the sun at the distance it is, but our social practices of measurements allowed us to describe such fact by using the expression “149.600.000 kilometers from earth”.
Social Kinds and Distinctions
Finally, we usually care about whether a kind is socially constructed or not, meaning, whether it makes a socially constructed distinction or not. As aforementioned, we say that a distinction is socially constructed if it makes sense only in function of certain human socia practices, actions or institutions. For example, we usually say that electrons are a natural kind because the distinction between what is an electron and what it is not is there in nature, independently of our social practices, institutions, etc. In contrast, the distinction between the owner of a property and others is a socially constructed one because its central function is to help us regulate our social practices. Thus, owning something is a socially constructed kind, not a natural kind.
As I had mentioned before, the question of whether a kind is natural, socially constructed, subjective, etc. is not completely independent from the questions of whether the corresponding property is natural, socially constructed, subjective, etc. However they are different questions, because even though socially constructed properties cannot be natural kinds, not all natural properties are natural kinds, i.e., we can make distinctions in nature that nature itself does not make. Medicine is full of such examples. Whether a condition is endodental of periodontal, for example, does not depend on our social conventions at all, but on the physiological and physical conditions of our mouth; yet, the distinction between endodontics and periodontics is socially constructed. Nature makes no such difference, it is our practices of how we approach conditions of each kind that makes them different to us (here, today).
An example in the Philosophy of Disability
In a large body of work, Shelley Tremain has sustained that the concept of impairment is socially constructed and, presumably, this is what she means: that the distinction between impaired and not impaired bodies is not a distinction that would make sense except for our social practices, institutions, values, etc. We can express this by saying that whether a body is impaired or not depends on our social practices, institutions, values, etc. However, doing so would be very unhelpful and confounding since it would be ambiguous between substantially different claims: (1) a claim regarding the existence of impaired bodies as entities, (2) a claim about the metaphysical status of impairment as a property and (3) a claim concerning the status of impairment as a concept. I take it that the correct reading is as a claim about why we make a distinction between impaired and not impaired bodies, instead of a question about facts or entities. In other words, if impairment is socially constructed in this sense, we could change our social practices and values in such a way that bodies that are currently considered as impaired could no longer be so. However, this change would be a change similar to the one performed by the cunning girl in our joke above: the bodies would not change their intrinsic properties, but our relation to them would change. But this would not be a less important change, on the contrary. Changing our social practices would not (directly) make people who currently cannot see, see, for example, but it would make their bodies no longer impaired, and this would be a signifiant political achievement.
In contrast, Michael Oliver and the so-called Brittish Model of Disability, make a distinction between disability, which they take to be socially constructed and oppressive, and impairment, which they consider not socially constructed and thus neutral regarding social oppression. I take it that Tremain’s point is that the British Model of Disability misses the difference between the second and third questions above, i.e., between the socially constructed nature of properties and facts on the one hand, and kinds and distinctions on the other.
Understanding the social construction of impairment this way has the advantage of not giving “far too much significance to language and representation” (Tremain 2015, 10). What is socially constructed is not merely the way we fix the extension of the term “impaired” among bodies, but the way we make distinctions among bodies. And making distinctions, of course, is not something merely linguistic, but a social practice and, in the case of concepts like impairment, a social practice with enormous significance on the experiences and identities of human persons. Tremain herself states this very clearly when she writes:
“Concepts, classifications, and descriptions are never “merely” words and representations that precede what they come to represent, but rather are imbricated in (among other things) institutional practices, social policy, intersubjective relations, and medical instruments in ways that structure, that is, limit, the field of possible action for humans, including what possible self‐perceptions, behavior, and habits are made available to them in any given historical moment.” (2015, 19)
Thus, one can be a naturalist regarding both the impaired bodies as entities and the material facts behind our judgments of impairment, while also recognising that impairment as a concept, that is, as a way of making a distinction among bodies, is socially constructed and, therefore, not something given but something we should be responsible of.
In general, I gather that social-constructivism is a safer bet as an answer to the third question than it is as an answer to the first two. After all, making distinctions is something we do and, as such, it is not surprising that many times the reason why we make the distinctions we do has a lot to do with our social interactions. This would explain why people can go as far as claiming that everything is socially constructed, without saying something absurd. What they mean is that the way we parse the world is always dependant on the social context in which such parsing occurs.