Events, Objects, and Eliminativism
A discussion of beliefs is very important in this day and age. The analysis of belief usually starts with sense perception, evidence, justification, and truth; instead, let’s start with the status of objects themselves. If we could get a better conceptual picture of what really exists, perhaps when can have a better debate about when to believe (or have justification in believing) in anything (theories, ideas, patterns, trends, memes, medical advise, and the like). I want to avoid the frustration of traditional ontological arguments that hinge on a confusion of predicates (is ‘existence’ a predicate or property, like the property of being red vs the property of just being?) or the ambiguity of universal/existential quantifiers like ‘something,’ ‘everything’ or ‘nothing.’ Anslem’s famous ontological argument for the existence of God has those classic language game puzzles. For an argument about ‘why nothing exists,’ see Aaron James, which seems more like an exercise to spot the flaws in the argument.
I also want to put aside the discussion of the ontology of persons and focus on ordinary objects. Though the distinction between ordinary objects and persons has been a hot topic in ontology, I want to assume that the distinction is grounded, either from reasons in the debate on persistence or philosophy of mind. However, even that debate could be guided by new developments in the philosophy of biology (see Godfrey-Smith 2020). Trenton Merricks, on the other hand, argues for the ontological elimination of ordinary things but not of persons. I will argue that Merricks’ elimination of objects is unwarranted from the lack of consideration that the objects we encounter in everyday life are more strictly speaking, event-parts of four-dimensional wholes. It is my contention that since we cannot eliminate events, we cannot eliminate ordinary objects from our ontology.
It is a commonly held belief that baseballs, statues, and chairs exist. Why believe in them? A brand of “eliminativism” argues that just holding up a number of different baseballs and saying, “look, here is another baseball—so they do exist” is not sufficient to believe in them. We think we believe in baseballs because we can see and feel them. Those in favor of unrestricted composition or fusion hold that there is an object, O, composed of the set of any object or part, A1….An; that is, there is an object, say, composed of my hand and the chair I am sitting on right now. But why believe in that object? The skeptic would argue that any “external” object could be questioned. Yes, that is an important epistemological debate; but the ontological considerations are just as important, especially since the very science that we use to understand that “external” and “objective” world makes assumptions about what objects are really "out there.” On the other hand, scientific theory can also help us to consider which primary objects we should import into our everyday ontology (i.e. what objects are real objects, and not just mental constructions that are useful in everyday talk). Therefore, I generally endorse (but not dependent on) in this line of argument a “naturalized epistemology."
Merricks argues that we need philosophical reasons to believe in objects like my hand-cum-chair, and for the same reason we also need philosophical reasons to believe in baseballs. (Note - “cum” is the latin for “with” or “joined,” like in cum laude or cumulative; philosophy uses this locution to describe objects that we don’t usually “see” as objects). This is because it is commonly believed that an object is something that is over and above just the parts that compose it. But if this is so, then there is something over and above the atoms arranged baseball-wise that is causing a window to break. But what about the physical atoms—don’t they cause the window to break? It seems very counterintuitive to think that the atoms making up the alleged baseballs do no casual work. Moreover, to think that two distinct physical things equally cause the window to break is also counterintuitive—the baseball and the atoms arranged baseball-wise can’t both cause the window to break. Therefore, Merrick argues that baseballs, chairs, and all other ordinary objects do not exist. There are other objects discussed in philosophy, like the status of mental or imaginary objects (Pegasus), and mathematical objects; however, whether or not these other objects exist in some way, let’s focus on the objects that have physical and causal properties, or in some way, can be studied by the physical sciences.
Merricks uses the locution of ‘atoms arranged object-wise’ to get to the idea that there are physical things structured in a specific way that make up every alleged object:
"I have in mind the atoms of physics, not Democritus. For there is no need to build a commitment to (or, for that matter, against) simples into eliminativism. Then again there is no need to build a commitment to atoms of physics either. So consider my claims about the atoms of physics to be useful but expendable. Such claims are really placeholders for claim about whatever microscopic entities are down there" (Merricks, p.3).
This concept, ‘arranged object-wise,’ then, commits those who believe in, say, baseballs to suppose that if there is a unitary thing with a mass and center of gravity, it non-trivially supervenes on the composition and properties of the atoms arranged baseball-wise. (Supervenience means that if we could change a set of properties at a lower level, it must change the set of properties or facts at a higher level). That is, worlds that are exactly alike at the microscopic level are exactly alike with respect to the existence of baseballs (Ibid p. 4). With this in hand, we can move on to the argument of eliminativism (Ibid p. 56).
Merrick’s ‘Overdetermination argument’ runs as follows:
(2) The shattering of the window is caused by the atoms, acting in concert.
(3) The shattering of the window is not overdetermined. Therefore,
(4) If the baseball exists, it does not cause the shattering of the window.
This is enough for Merricks to conclude that we have no reason to believe in macrophysical simples (e.g. baseballs, chairs, fingernails…). This argument is prima facie valid. But is it sound? I want to focus first on premise (2). What I want to point out is that atoms acting in concert at any give time instantiate an event. So the shattering of the window is an event caused by the atoms acting in concert—but what is also an event is ‘the baseball existing.’ That there is causation going on at the atomic and subatomic level, we can say that there is something arranged baseball-wise, which is causing the existence of what we commonly call a baseball—which is an event or a process. There are good scientific reasons to consider “events” as simples in our ontology. Special and General Relativity even reject the Newtonian assumption that there is an absolute point in space and time that can be “filled” by some material object, no matter how infinitesimal. Instead, events that produce light can be taken as the foundation for the geometry of space-time, and hence where we might find the “stuff” that we experience: every event can be mapped with its own Light-cone, in which it can be said to “be-existing” within space-time. It makes no sense to say that any two objects in the universe are “existing” at the same time. (see Maudlin 2012).
The moral of the overdetermination argument is that the existence of a baseball would cause the shattering of the window, and the atoms arranged baseball-wise also cause the shattering of the window. The shattering of the window would then be overdetermined; that is, the baseball causing the shattering of the window is redundant, for it is sufficient for just the atoms arranged baseball-wise to do all the casual work. So if we accept (2), then we are conceding that each of the atoms do cause something. This is of course correct. For one, particles (atoms, or particles of the atoms) are moving and arranged in ways that instantiate certain properties. But, for an object to be “arranged baseball-wise” is not like there are stationary puzzle pieces arranged in a way to make a puzzle. At some lower level, (which ever it may be; or maybe even a “higher” level?) there is either an event or a process going on that instantiates the properties we observe scientifically or in everyday observation. The moral of this line of thought is that an object existing at some level is an event or process. But is there still at some level a physical simple? Merricks says he doesn’t assume that there are physical simples, but I am not sure he really does within his arguments. Moreover, he concedes that to be a macrophysical object “is to have some causal powers” (Ibid p. 81). Therefore, if the atoms arranged baseball-wise do all the causal work (he argues), baseballs do not exist—for we have to reason to believe that they exist if we take the overdetermination argument seriously.
It is controversial whether or not events, processes, or physical stuff are the basic agents in causation. What I think is not so controversial is that there are events—and things or other events cause them. But can we give ontological status to the event that a baseball is existing? If so, (3) would have to be defended in another way. Let’s look at it again:
(3) The shattering of the widow is not overdetermined.
An effect is overdetermined if the following are true: “that effect is caused by an object; that object is causally irrelevant to whether some other—i.e. numerically distinct—object or objects cause that effect; and the other object or objects cause that effect”(Ibid p. 58). One way to respond to this is to say that the shattering is “pseudo-overdetermined” by being caused by the object and the event in which the object participates. But the objection is blocked by analyzing it as not being a case of overdetermination; that is, object-causation and event-causation do different kinds of causal work. So what it is for a baseball or atoms arranged baseball-wise to shatter a window is for it to participate in a window-shattering event. I will concede that event-causation and object-causation are by definition different in kind, but what makes it the case that atoms arranged baseball-wise do not participate in a “baseball-existing” event? If they do, then that event should have some causal role in the shattering of the window event. Though, as Merricks argues, that there is event-causation and object-causation together does not give us overdetermination. Therefore, (3) holds up but we have a sense that (2) may be ill-conceived, since the event of the window shattering cannot be analyzed sufficiently in terms of just the object-causation of the “atoms arranged baseball-wise acting in concert.”
We can see that there are two existents to be analyzed in any causation—the objects and the events. (2), then, is ill formulated in terms of giving a full account of what exists. The thrust of eliminativism is to eliminate superfluous existents within the causal network. I will concede that the folk ontologist’s view of what it is to be an object must be thrown out, but it can be replaced by another ontological definition of what it is to be a macrophysical object. But now it is the burden of my formulation to give reasons for accepting it. For one, do events really exist? And how is ‘existing’ an event? Is there a circle here? Let me deal with these issues before we can return to Merricks’ argument.
An event is anything that happens. I believe this is straight forward enough to suppose that things do happen. An event can be construed as that of something changing or that change is occurring; however, this is not necessary for something to be an event. Imagine a baseball sitting at rest on a table. Secondly, image that there is not a single person or other organism perceiving (or in any causal relation to) it. I think we have a clear case that nothing is changing about the baseball and no causal powers or properties are be instantiated at that time. On the other hand, there is a sense in which something is happening—that a baseball is resting on the table. If I all the sudden walked into the room and saw the baseball, we would say that what happened was that I saw the baseball. (Light hits my retina, and so I exist in the “same light-cone”; or in another way, it would make no real sense to say that I saw a three-dimensional object while riding on the same absolute time-line). But with that in mind, I think that we can say that something happened even if we did not walk into the room—that something existed. Take this example: there is a world W1 in which all that exists is a baseball, and that the only thing that changes is the relationship between the ball and time. At t1 the baseball pops into existence and at t2 the baseball pops out of existence. What happened in W1? Well, a baseball existed.
From these examples I think that it is clear that for something to be an event, it is not necessary (but possibly sufficient) for a change to occur. It can be pointed that in a perduantist framework, temporal parts can be identified with events. Quine has argued for this view, and has said that events and physical objects should not be thought of as belonging to distinct metaphysical kinds (Lombard, p.141). I do not want to argue for or against four-dimensionalism yet, but to show that the idea that events are legitimate ontological entities is very plausible. Moreover, it is not decisive from the findings of empirical science that there are physical simples (be it atoms, protons, quarks) that are irreducible to some kind of efficacious event, process or relationship. David Chalmers points out that
"…physical theory only characterizes its basic entities relationally, in terms of their causal and other relations to other entities. Basic particles, for instance, are largely characterized in terms of their propensity to interact with other particles…the picture of the physical world that this yields is that of a giant causal flux, but the picture tells us nothing about what all this causation relates" (Chalmers, p. 153).
It may be that events are the most fundamental entities that can be described. Though it is not settled, I think that it is justified in saying that events could have equal or more ontological reality as physical simples. Moreover, that there is not a clear understanding of the nature of causation shows our lack of understanding of the relationship between event-causation and object-causation. What overdetermination reveals, however, is that an effect cannot be redundantly caused by two sufficient doses of the same causation. If atoms arranged baseball-wise is sufficient to do all the causal work to shatter atoms arranged window-wise, there is no reason to suppose that there exists another physical object (baseball) that also enjoys in doing the causal work of breaking the window.
Let’s return to the Merricks and the overdetermination argument. An objection to premise (2) involves the move I made in claiming that there are two different types of causation going on in the shattering of the window. This would mean that the baseball is not identical with its constituent atoms, and so the shattering of the window is also not identical to the event of the many scatterings of the atoms arranged window-wise. Merricks responds to this kind of objection with the following remarks:
"The ‘non-identity’ crucial to this objection is not between the single event of the scattering of the atoms and the single event of the shattering of the window. The first event—if it existed—would be identical with the second…there is, however, a lack of identity between the shattering of the window and the many events such as this atom’s heading thataway…for each of the atoms’ formerly arranged windowwise" (Merricks, p.164).
Merricks concludes that atoms arranged baseball-wise have multiple effects, namely, the multiple scatterings in addition to the scattering of the window. However, that there are multiple events occurring just on the side of the baseball (that there is a baseball moving in a certain direction and a baseball existing) disrupts the certainty of what exactly causes what. With the distinction between event-causation and object causation, there is some sense in which the event of ‘the baseball-existing’ causes the event of the shattering of the window; while at the same time, there is the effect of atoms formerly arranged window-wise scattering that is caused by the atoms arranged baseball-wise. Merricks even seems to agree that event-causation may be more basic that object causation (Ibid p. 65). What he fails to consider is the fact that our ontology could include both the thing arranged baseball-wise and the event of an existing baseball.
From these considerations, we need to conclude that (2) and (4) are false under the interpretation that the event of ‘a baseball existing’ supervenes on the atoms (or, the event of the atoms existing) arranged baseball-wise. Though folk ontology can be considered as believing in the existence of macrophysical simples, we should not wish to eliminate the existence of ontologically real events of existing parts of temporal macro-objects. Whether or not an event can be a physical object in the endurantist framework I will leave open, but it does not seem that intuitive that an object we normally think as enduring through time is an event that endures through time. What needs to be dealt with in my ontology is whether these concrete events are temporal parts and hence make up a four-dimensional whole.
Mark Heller mentions that “the confusion over the nature of temporal parts is increased by the fact that phrases as ‘temporal part’…have been used in ways that suggest such varied purported objects as process, events, ways things are…” (Heller, p.312). He argues that all parts of four-dimensional objects are themselves physical hunks of matter—be it a temporal part or a spatial part. Moreover, the perdurantist can avoid any charge of co-location of two wholly and numerically distinct objects, hence giving a seemingly neat framework to explain how objects change over time; i.e. succession of different temporal parts. But being that the parts of a temporal whole are also physical hunks of matter (in Heller’s view), there is the charge, elucidated by Gibbard, that there could be two distinct temporal wholes entirely co-located. Imagine two lumps of clay fashioned into two statue halves and then stuck together to create a statue and also a large lump. Now imagine that the resultant statue and lump are then simultaneously annihilated or ripped apart (Gibbard, 100). The perdurantist could respond to that case by saying that the statue and the lump are identical. But the lump and the statue have different modal properties—namely, that the lump could survive being squashed but the statue cannot. I think this charge is fatal to Heller’s view of four-dimensionalism. On the other hand, if we take temporal parts as being events, this charge does not hold up.
Though eliminativism seems easier to swallow along side these co-location objections to endurantist and perdurantist views of macrophysical objects, the perdurantist could insist that the temporal parts are events instead of physical hunks of matter. This is because events and the atoms arranged object-wise can be co-located. This premise follows from the considerations dealt with earlier about event and object-causation—which are both part of the same larger event. The baseball, for example, is an event of both event-causation and object-causation (assuming that there is some spatially extended simple at some reductive level). What needs to be fleshed out is whether events have essential properties—and what they are. The point is, though, that a four-dimensional statue can share all of its parts with a four-dimensional lump of clay because a lump-part-event can be co-located with a statue-part-event and hence have different modal properties.
I think it is warranted to keep tables, chairs, and baseballs in our ontology. I have argued that Merricks ‘Overdetermination argument’ in favor of eliminativism can be blocked by the rejection of premise (2) and (4). Though there can be no overdetermination, that baseballs (their temporal parts) exist as events only entails that the causal structure of larger events contains a combination of non-overdetermined event-causation and object-causation.
References:
Chalmers, David (1996), The Conscious Mind (New York, Oxford University Press).
Gibbard, Alan (1975), “Contingent Identity” repr. Metaphysics: An Anthology Ed. Kim
and Sosa (Oxford: Blackwell Publishing).
Godfrey-Smith, Peter (2020). Metazoa: Animal Life and the Birth of the Mind (New York,
Farrar, Straus and Giroux).
Heller, Mark (1990) “Temporal Parts of Four-Dimensional Objects” repr. Metaphysics:
An Anthology (1999) Ed. Kim and Sosa (Oxford: Blackwell Publishing).
Lombard, L. B. “Event Theory” in A Companion to Metaphysics (1999) Ed. Kim and
Sosa (Oxford: Blackwell Publishing).
Maudlin, Tim (2012) Philosophy of Physics: Space-Time (Princeton, Princeton
University Press).
Merricks, Trenton (2001), Objects and Persons (New York: Oxford University Press).
Comments
Post a Comment