I claim that any theory of properties should aim primarily to be a useful intellectual tool; to be an abstraction that covers up the messy details of the world. That there is no one best theory about properties for all people and in all situations is one consequence of that claim. However, it remains to be shown exactly how a theory about properties can be a useful intellectual tool, and the burden of proof rests on me to demonstrate that it can serve as such a tool. Below are some of my notes about how a theory about properties may find applications outside of philosophy, and where one theory may be more useful than another.
1. Properties themselves can be attributed to our need to abstract details from specific situations in order to reason about things in general, and to communicate without each other. Perhaps in the experience of infants, before the conceptual apparatus of language is developed (which brings properties with it in the form of adjectives), objects and events are experienced as unified wholes – without parts and thus without properties. But, as mentioned, without breaking the experience of an object or event up into properties you can’t talk about it (except, perhaps, by pointing), or reason about it, because each experience is wholly unique. Thus the idea of properties itself is a useful abstraction that permeates almost our entire intellectual life. I’d say that’s a pretty useful piece of philosophy.
2. One of the rare cases where we may explicitly invoke the idea of properties outside philosophy is when trying to define something. In attempting to come up with a good definition we self-consciously think about the properties a thing has that makes it unique. Consider a general definition, one that picks out a class of things rather than a particular object. Specifically consider a definition/description of a regular monopoly board, in general. It is reasonable to include “has a space labeled ‘Boardwalk’” as part of our description of what a monopoly board is. But what about a damaged monopoly board, one that is missing Boardwalk? Does it cease to be a monopoly board? The fact that we refer to it still as a monopoly board, albeit a damaged one, strongly implies that it should still count. But yet it is still valid to describe a monopoly board in general as having Boardwalk. When we run into a case such as this having a theory about properties and what they are properties of may come in handy. If the theory you are working with asserts that properties must apply to objects you may be faced with a dilemma. Either you accept that there is some abstract “monopoly board in general” object that has these properties, which is unpalatable, or the properties of monopoly boards in general are properties of each monopoly board, which creates problems in the case of the damaged board. Now these “problems” can be resolved. You could say, for example, that the properties of a monopoly board in general are only properties that most, but not all, monopoly boards have. But then to a question such as “do monopoly boards include Boardwalk?” you must answer “the majority have Boardwalk but a minority may not” when what you want to be able to say is “yes”. Again, the theory is defective here not because we can’t make it do what we want, but because instead of being a useful tool it makes us adopt unusual and unhelpful ways of thinking to make it work. Alternately, if we still want to hold on to the idea that all properties are properties of objects, we could treat “monopoly boards” as a list of properties that individual game boards can fall under – in other words a complicated property that we define in terms of other properties. Again though this leads to the need to jump though hoops to make this approach work in certain situations. Instead of being able to say that monopoly boards are game boards (fall under a more general category), we would have to say that all the individual boards that fall under the category “monopoly board” fall under the category “game board”. There is nothing wrong with this, but it’s not a convenient or particularly useful way of thinking about the situation. Finally, we come to the theory best suited to this situation: properties simply clump together; there are no objects, just property clumps. Under this theory we can take the idea of a monopoly board in general to be its own clump of properties, which is not an object in any normal sense because it lacks properties such as a location in time and space, individual existence, and so on. The relation of particular, possibly defective, boards, to this clump of properties representing the abstraction of a monopoly board in general can de described in one word: “approximates”. Obviously in this case the last approach is far superior to the rest – it provides a consistent framework that allows us to talk about monopoly boards in general, the category’s relation to particular boards, and its relation to larger categories in exactly the way we want, all without committing us to any metaphysical extravagance. Of course that’s just what’s right in this extended example; it may not always be the best approach.
3. The properties-as-clumps versus properties-as-attached-to-objects distinction may also come up when how the properties are related to each other is at issue. The properties-as-clumps picture promotes a view where the properties in the clump are connected to each other, and possibly depend on each other. In contrast the properties-as-attached-to-objects picture promotes a view where the properties are like tags that are stuck to the object, and which can be added or removed without substantially affecting each other. I can think of situations where both perspectives are appropriate. When thinking about a case where the properties in question are closely connected to each other, such as the properties that define someone’s personality, thinking of them as linked together is helpful. It is not the case that you can simply switch off one aspect of someone’s personality or add something new in isolation. The manifestations of an individual personality are interconnected at a deep level, and any change in one aspect is going to have an effect elsewhere as well. And the properties-as-attached-to-objects theory may lead us to think, contrary to this, that we could simply swap out greed for altruism without other changes resulting from the switch. On the other hand, there are cases where properties are like tags stuck to an object – easily added and removed without affecting each other – perhaps literally. In a case like that it would be foolish dwelling on the interconnectedness of properties when they simply aren’t.
4. Another difference between theories about properties is whether different objects can have the same properties. Most theories do provide some way in which the properties of different objects can be described as literally the same. However, there are also approaches in which the properties that are found in individual objects are not the same, at least not in a normal sense; the best that can be said is that they are similar. In most cases the former sort of theories are preferable to the latter. After all the whole point of the abstraction of properties is to abstract and communicate, which would seem to necessitate thinking of properties found in different objects as the same. However, there are cases where we want to emphasize that there are subtle differences, even when the objects in question are described as having the “same” properties, in which case the latter sort of theory is superior. For example, we may be classifying people by personality but at the same time want to keep in mind that Bob’s pacifism is not exactly the same as Charlie’s pacifism.
5. Cases where we need to demonstrate that two objects do in fact have the same property (or where questions of how we know that two properties are the same come up) need a specific kind of approach to properties as well. Any theory that takes the sameness of properties in different objects to be a bare fact will not be helpful. In such cases theories that take the properties found in each object to be essentially distinct and attempt to explain how properties in different objects are similar or fall under the same category are more useful. Because, in doing so, those theories usually outline tests or methods by which the properties of different objects can be compared to see whether they are similar or whether they fall under the same category. And that is exactly what is needed.
6. But what’s wrong with flying by the seat of our pants? Obviously there is an intuitive conception of properties that most people, not having extensive training in philosophy, get by with just fine. Well, there’s nothing inherently wrong about it. I often compare philosophical theories to tools. Using an intuitive conception of properties is like using a rough and unrefined tool – it may get the job done but it isn’t optimal. Generally the rough and ready tool is unsystematic. If we were to analyze it as a philosophical theory it would look like a combination of approaches, each of which is deployed in specific situations. The problem with this is that it may lead to internal contradictions. If you hammer with your tool in one place and it leaves a certain kind of mark and you hammer in another and it leaves a different kind of mark then if you try to put the two together they may not fit. Again, this isn’t an fatal defect, we are always free to fiddle with the results of our unrefined tool to make them fit together, but it is an example of how the unrefined tool can occasionally get in the way. Another, larger, problem with sticking with an intuitive conception of properties is that it simply may not be the right tool for the job. As illustrated above, which approach is best varies from situation to situation. If we are masters of a number of different theories then we are free to pick the one that best suits the task at hand, which will yield the best results more often than always trusting whatever approach we intuitively find ourselves using.