I have recently read Eric Marcus' Rational Causation, a book about which on the whole I wondered whether anyone who had not read the Rödl and Hornsby (and to a lesser extent Steward) to which he makes frequent reference would get a lot out of, and whether anyone who had would get much more than he or she had already gotten out of the other authors: and in general whether Marcus will find a receptive audience for his goal of showing that one can "resist the naturalistic approach [without] accepting dualism, epiphenominalism, or eliminativism" among anyone who wasn't already inclined to believe just that. But rather than saying anything interesting about the book on the whole I am in typical fashion going to niggle about something very minor and unimportant.
So! In the fifth chapter, when talking about object identity, Marcus somehow manages to avoid statements of the form "'object' and 'state' are formal categories", but much of what he does say points in that direction. A state is a thing with a principle of instantiation; an object is a thing with a principle of identity. Objects persist (and exist); states obtain and are exemplified. But saying something like "an object has a principle of identity" is explicitly not saying that anything that's an object has the same principle of identity that anything else that's an object has; rather, "objects instantiate sortals and … to instantiate a sortal is at least in part for there to be a principle of identity that determines the conditions under which the object persists. Objects, however, do not instantiate principles of identity as such, but rather only insofar as they are particular sorts of objects" (p 187). Ok: Goats and desks do not have the same principles of identity. Thus also the emptiness (but not total emptiness) of simply identifying something as an "object": this is a "dummy sortal" because it just identifies the formal category to which the thing being talked about belongs—we're talking about something that persists, has a principle of identity, has parts, etc., and not something that is, like a state, negatable, capable of instantiation, etc. That's not, by any means, not to say anything, but it is not to say anything about the particular principle of identity of what we're talking about (if we are talking about something with a particular principle of identity).
That all seems agreeable. But it makes the following argument, from only 15 pages later, utterly baffling:
Can the particularity of states be established by assimilating them to objects? Few would argue that states are objects. First, the concepts of part-hood and composition have no univocal application to objects and states. Second, objects and states do not have the same relation to space. Objects compete for space with other objects. (p 202)
But the concept of part-hood—like principles of identity!—doesn't have univocal application to objects and objects. The sense of "is the same as" is ineluctably informed by the identities of its relata; what it takes to be the same goat over time isn't the same as what it takes to be the same baseball game over time. The same, I would have thought, is the case with relations of part-hood: what it is for a goat to have a part is not the same as what it is for a desk to have a part, so why can't there be a different again way for being blue to have a part? It wouldn't do to object that we can't find a sense of part that attaches to states the way we do for one that attaches to goats because many particular states, which will (or at least in principle could) all have their own different ways of determining part relations, are ranged under the still formal category "state", whereas the many animals that fall under the category "goat" have the same way of determining part relations. First of all, we could just as well have taken not "goat" but "animal" as a leading example, which seems to get us close to a desired parallel. Second, to say that states are objects is just to say that any state has all the formal features that any (other!) object has, and that leaves it open that being a state is still further determinable. The relevant determinates may have further formal state-specific features in common, or there may be common ways that all states fulfill the formal requirements of objecthood. (Marcus at one point says that "both objects and states can be said to persist", but they persist in different ways, which seems to block this move for him, anyway.)
Here is an extended analogy. One might deny that complex numbers can be classed as numbers because there is no univocal sense of addition that applies to complex numbers and to reals, rationals, integers, and natural numbers. (For simplicity's sake let us deal specifically with addends, not numbers, and assume that there is a univocal sense of addition that applies to reals, etc.) It is true that there is no univocal sense of addition that applies to complex numbers and real numbers—to add complex numbers one performs the pairwise addition of the real and imaginary parts, and real numbers don't have real or imaginary parts, so one can't add them the way one adds complex numbers—but it is, at least, not clear why this means that complex numbers can't be addends. Whatever things we think an operation must exhibit to count as addition—let's say as a first stab that it be binary, commutative, associative, and symmetric, and have an identical left and right identity (which we might give the name $\emptyset$)—we can find for complex-number-addition:
- $(a + bi) +_{complex} (c + di) \equiv (a +_{real} c) + (b +_{real} d)i$
- $\emptyset \equiv 0 + 0i$
This is notationally confusing, since the $+$ in $a + bi$ doesn't really represent addition at all, so we might prefer to represent complex numbers as ordered pairs, in which case we will immediately see that the members of the pair don't have to be—as they were with complex numbers—real numbers, but anything that is itself an addend:
- $(a_1,b_1) +_{a,b} (a_2,b_2) \equiv (a_1 +_a a_2) + (b_1 +_b b_2)$
- $\emptyset_{a,b} \equiv (\emptyset_a, \emptyset_b)$
So we can say that ordered pairs of addends are themselves addends, even though "ordered pair of addend" is a determinable; we could get away with doing this at a stroke because all the determinate ordered pairs of addends are addends in structurally similar ways. There is absolutely no univocal sense of addition that applies to both real numbers and ordered pairs of complex numbers and strings, but we can add pairs of the former and pairs of the latter up just the same.
The problem with states and part-hood isn't that any sense of "having a part" that could apply to states wouldn't also apply to (other) objects, it's that there's no appropriate sense of having a part that applies to states in the first place. That's the challenge to someone who wants to claim that states are objects: objects generically have features $x$, $y$, and $z$, which relate to each other in such-and-such fashion and exhibit the following further properties—so now you must show me how states do that. Of course one could respond to this demand by disputing the formal characterization of objects—something to which Marcus' main argument may be vulnerable. As mentioned above, he says that both states and objects persist, but whereas objects exist, states obtain, and those are distinct things:
[B]ecause existing and obtaining are distinct, the sorts of presence and persistence that characterize objects and states are also distinct. Here, on my view, is the difference. In the case of a state, e.g., Jones's believing that Nixon was a genius, the presence of the belief is Jones's exemplifying a universal; and its persistence is his continuing to do so. In the case of an object, e.g., Jones himself, presence is just the continued existence of a particular at a time; and its persistence is its continued existence over time. (p 203)
One could put the thought like this, I think: objects and states are both—let's say—things, the salient formal feature of which is that things persist. But objects persist by existing, and states persist by obtaining. (There seems to be a further thought that the existing of an object is primitive (presence is existence at a time; persistence is existence over time) while the obtaining of a state is a matter of the exemplification of the state by something else (by an object? or just anything?), but that doesn't seem to the point to me.) Now one could resist this by attempting to establish that actually it's objects which persist and some sub-class of object which exist (and which is disjoint from states). If all that amounted to were a shuffling of labels around, there wouldn't be much point in it; but there might be some point if the thing now called "object" (which persists) retained some of the features of what was formerly called "object" (which exists).
Recent Comments