This post will probably be of little interest to those who don’t care about metaphysical issues in modern analytic philosophy i.e. nearly everyone. But lately I’ve been thinking a lot about modal logic and essentialism, and I wanted to draw out one of Peter Van Inwagen’s interesting arguments about objects, parts, and continuity through space and time for those that are interested.
In a famous essay, Peter Van Inwagen argues that the Doctrine of Arbitrary Undetached Parts (DAUP) is false because it commits one to accepting Mereological Near-Essentialism. What is DAUP? The general view of DAUP states that there is an object corresponding to any part of a material object, no matter how one cuts it up. In other words, this doctrine is stating that any region of space of a material object can be divided up into a part and that the part constitutes a material object.
For example, any part of the Egyptian pyramids or John Coltrane’s saxophone could be divided into a part, no matter how random or arbitrary, and the part would be a material object in and of itself. On the face of it, DAUP appears to be obvious and intuitively true. We can speak of and point out all the different possible parts of an object and consider them as objects in and of themselves.
It seems clear that a small chunk of the top of an Egyptian pyramid, or any other part of the pyramid for that matter, can be pointed out and considered a material object. Furthermore, DAUP seems to coincide well with the intuitive view of universalism. Universalism, in the metaphysical sense in which I am using it, states that trying to draw a distinction between what parts of an object are arbitrary (and not to be considered material objects of their own right) and what parts of an object are not would have no clear justification.
While it is true that some parts of an object may be more interesting than others and someone’s leg may seem more interesting as a ‘part’ than three square inches of their right thigh, this doesn’t mean that DAUP is false. If anything, to say that certain parts of the Egyptian pyramids are material objects while other parts are not would seem dubious and unwarranted. While certain portions of an Egyptian pyramid may seem more logical to consider as a part, this doesn’t preclude seemingly more random parts from being material objects.
Additionally, our intuitions about what happens when we break a material object appear to be in line with DAUP. If one randomly breaks a graham cracker into two pieces, it seems unlikely, one would argue, the two separate pieces of graham cracker just came into existence. The parts of graham cracker existed as material objects before the break took place.
Because of the intuitive power in favor of DAUP and universalism, arguments against DAUP must be very strong to be given serious consideration. Since Van Inwagen argues that DAUP is false, even though DAUP has much intuitive plausibility, Van Inwagen’s argument against DAUP can be viewed as an argument and/or a puzzle to be explained. If DAUP is false, then universalism is also false, and giving up these views would not be easy for most people.
To argue against the truth of DAUP, Van Inwagen states that a commitment to DAUP leads to Mereological Essentialism and since Mereological Essentialism is plainly false (on the face of it), DAUP must also be false. What is Mereological Essentialism? In short, the thesis of Mereological Essentialism is the view that an object has all of its parts essentially. If an object has parts, it always and necessarily has these parts according to Mereological Essentialism. In Van Inwagen’s argument against DAUP, he claims that his argument shows that DAUP leads to a weaker version of Mereological Essentialism which he calls Mereological Near-Essentialism.
The difference between Mereological Essentialism and Mereological Near-Essentialism, the view Van Inwagen argues DAUP entails, is that Mereological Essentialism contends that if a part is removed from an object it ceases to exist in all cases. Mereological Near-Essentialism allows for a part to replace a part that was removed so as to allow the object to continue to exist. The reason Van Inwagen believes that his reductio ad absurdum will be successful, even though all it does is show that DAUP leads to Mereological Near-Essentialism, is that he believes Mereological Near-Essentialism is clearly false. Furthermore, Van Inwagen must believe that the clear falsity of Mereological Near-Essentialism is more compelling than the intuitive power in favor of DAUP.
There are very persuasive reasons to think that the thesis of Mereological Near-Essentialism is in fact false. For example, if Mereological Near-Essentialism is true, virtually no objects persist through time. Organisms and objects are constantly losing atoms and parts. If an object must have all of its parts essentially and cannot survive the loss of even the tiniest or most seemingly irrelevant part, then according to Mereological Near-Essentialism, it no longer exists.
This clearly goes against our basic intuitions. If one were to lose an arm in a car accident, no one (or at least most people) would say that one is a new person after losing that appendage. It doesn’t seem to be the case that an arm is so essential that the loss of it results in a person ceasing to exist. It appears clear that they still exist, whether with their arm or not.
On a side note, it is interesting that one of the most compelling arguments in favor of DAUP and universalism is that they are loose, casual doctrines. DAUP appears intuitively compelling because one feels it would be too strict and arbitrary to declare that some parts of an object are in fact parts while others are not. But if Van Inwagen is right, and DAUP commits one to Mereological Near-Essentialism, a commitment to DAUP does not leave one with a liberal doctrine but instead forces one into an extremely strict view about objects and their persistence through time.
Van Inwagen is not saying that objects don’t have any parts at all; he is arguing that objects don’t have arbitrary parts. Van Inwagen, because he gives a reductio ad absurdum to show that DAUP is false, begins by assuming that DAUP is true and also asks us to assume that Mereological Near-Essentialism is false.
Here’s Van Inwagen’s argument, in short: Imagine there is an object O, with a part P, at time 1, and that the object O could survive the loss of part P. According to DAUP, there can be an object which is the difference between O and P at time 1. This object, which is the difference between O and P at time 1, will be referred to as O-minus. O-minus and O are clearly different objects. They occupy different regions of space, have different parts, and are numerically diverse.
Now, at time 2, imagine that O has lost part P, but because Mereological Near-Essentialism is false, O exists and survived the loss of part P. At time 2, O-minus still exists and its’ qualities are the same as they were at time 1. But now, at time 2, as a result of O losing part P, O and O-minus are material objects wholly in the same place at the same time, sharing all the same properties, and composed of all the same parts.
Van Inwagen argues O and O-minus, which were once distinct material objects at time 1, are now the same thing at time 2. Van Inwagen believes that the idea of two different objects coinciding in the exact same place at the exact same time with all the same properties makes no sense whatsoever. So DAUP, which allows for any part of an object to be an object itself, leaves us with this seeming contradiction; O and O-minus, which were once distinct objects, are now the same.
Van Inwagen concludes that O-minus doesn’t exist and not every part of an object can be divided into an object. Van Inwagen’s argument, which assumes Mereological Near-Essentialism to be false and DAUP to be true, shows that the conjunction of these views leads to a contradiction. If one accepts Mereological Near-Essentialism, one can argue that when O lost part P it went out of existence so there is no seeming contradiction. However, as pointed out before, Van Inwagen and many others find this view of Mereological Near-Essentialism untenable. Therefore DAUP entails Mereological Near-Essentialism, and since Mereological Near-Essentialism is false, DAUP must also be false.