I’ve recently found myself having to deal more and more with the gnarly issue of souls in my work. My research focuses primarily on the relation between personal identity and ethics, but the metaphysical possibility of the existence of souls continually throws a monkey wrench into attempts to draw firm conclusions about the nature of selves, human beings, and/or persons with respect to ethical issues. Most contemporary theorists assume a materialist conception of these objects, but it’s very easy to undermine their conclusions by simply positing the existence of an immaterial substance at our core. For example, Singer and Kuhse have shown that if you believe that an embryo is a human being (with full moral status) from the moment of conception, the possibility of twinning gets you into serious trouble. But the trouble only comes if you assume a materialist conception of human beings. If, on the other hand, you maintain that what makes the embryo a human being from the moment of conception is that it houses a soul, then you can avoid the problems they cite. After all, upon twinning, one soul could migrate to one of the new organisms, while a new soul could pop into existence to be housed in the other new organism.
Because arguments about souls continually creep into public debates about this issue, I think it’s incumbent upon ethicists to deal seriously with it, rather than ignoring it or casting ad hominems upon its advocates. So that’s what I’ve been trying to do. Please forgive, then, the following movement into some rather straightforward metaphysics. It’s necessary, I think, to wrestle with the important ethical issues at stake in both the abortion and the stem cell research controversies.
It seems to me that the best way to deal with the problem of twinning is by maintaining an Augustinian (rather than Thomistic) conception of the soul (for reasons I won’t go into here). The Augustinian conception maintains that souls are purely immaterial substances, and they are housed (some might say “trapped”) in a material body during its life on earth. This move makes a human being consist in an ordered pair of body and soul, i.e., {B, S}. Now the most plausible way to avoid the twinning problem as a pure materialist is to go four-dimensionalist: the pre-twinning embryo is actually two human beings, worm-bodies that split upon twinning, much like two distinct roads that may share a stretch before splitting off into two different directions. So the two human beings at this point share all of their temporal parts.
But if souls are involved in making a material organism a human being, then we need to fit two of them in somewhere in this ontology, viz., by attaching them to temporal parts of bodies. What is wholly present at t1, then, pre-fission, would be two souls attached to a wholly present temporal part of two spacetime worm-bodies. The ordered pairs involved would thus be functions from temporal parts of distinct bodies to distinct souls, i.e., {B1(t1), S1} and {B2(t1), S2}, where B1 and B2 are spacetime worm-bodies whose temporal parts at t1 wholly overlap with one another. Articulating the matter in this way allows for the most plausible account of what happens to the bodies in the fission case (both spacetime worm-bodies share a temporal part pre-fission) while also allowing that (a) both body-worms could be ensouled from conception, and (b) the two post-fission body-stages would each be unified with the pre-fission body-stage as part of the same body (while not being unified with each other).
Nevertheless, while making this move renders the identity of the bodies involved utterly unproblematic, it raises a host of questions about the identity of the human beings involved. For example, do human beings (ensouled bodies) endure or perdure? On this account, the bodies involved perdure, while the souls involved — because they are simple, and thus three-dimensional substances — endure, but how are we to make sense of what happens to the combination of the two? If human beings are truly ordered pairs of bodies and souls, then what precisely is wholly present at t1? There are only (temporal) parts of two bodies wholly present, but there are two wholly present souls. Set aside for now the problem of fission and consider an embryo at t1 that won’t twin, though. Combining bodies and souls under these competing ontologies in this simple case yields a straightforward dilemma resting on a logically exhaustive disjunction: human beings either endure or perdure. On the one hand, if human beings endure, then all of their parts must be wholly present at any given time. But on this view only the t1 temporal part of our human’s body is present at t1; its remaining parts have yet to occur. Thus not all of the human being’s parts are wholly present at t1, so it cannot endure. On the other hand, if human beings perdure, then what is wholly present at any given time must be only temporal parts of human beings. But on this view the soul renders the embryo housing it a human being, and because the soul is a simple substance it can have no parts whatsoever, including temporal parts, so at t1 what is wholly present is not just the temporal part of the embryo/body but the entire soul. Since the entire soul, and not merely a temporal part of it, is wholly present at t1, the human being involved cannot perdure. Ultimately, then, the view implies that human beings neither endure nor perdure. But if these are logically exhaustive possibilities, then the soul view (in this version) is incoherent.
Now it seems to me that the only way the Augustinian can rescue the view from incoherence is to somehow maintain that souls can have temporal parts, in which case we can treat them as four-dimensional objects as well, which would enable the theologian to avoid the twinning problem altogether. So this possible “out” leads to my question: can souls have temporal parts? I don’t believe they can.
While it is obvious why a soul could not have any spatial parts, it may be less obvious why a soul could not have any temporal parts. Normally, of course, temporal parts are thought to exist, if they do at all, in spatiotemporal objects, which, if true, would immediately rule out souls (non-spatial objects) as having any temporal parts. Nevertheless, it seems in principle possible for there to be objects having no spatial location that still have temporal parts. What exactly would this mean, though? The soul would have to have parts that exist only instantaneously (or perhaps over a more extended interval of instants) and be such that each one of those parts nevertheless wholly overlaps with everything that is a part of the soul at that instant (or interval). Now insofar as the soul is supposed to be a simple substance, it is also not supposed to be divisible at all, either into spatial parts or temporal parts like this. But why not? The restriction against spatial parts is obviously required by the soul’s immaterial nature. The restriction against temporal parts would have to follow, I believe, from two aspects of the temporal parts view that are problematic for immaterial substances. First, the notion of what it means for a temporal part to “overlap” at t with everything that is part of the soul at t is an idea of which it is exceedingly difficult to make sense with respect to immaterial substances. This is no problem, of course, with respect to material objects: for a temporal part of my mug to overlap at t with every part of the spacetime worm of my mug that exists at t, it just has to be the case that every particle of physical stuff making up the temporal part of the mug at t takes up the same space as everything that is part of the spacetime worm-mug at t. But if an object, like the soul, takes up no space to begin with, it is difficult to see how to specify what it would mean for its temporal part to overlap with anything. Second, if a soul were to have temporal parts, there would need to be a unifying principle connecting the various parts as one unified soul across time. But it is again difficult to imagine what that unifying principle – intrinsic to the soul-parts themselves – might consist in, if there were no material or even psychological elements constituting them. For these reasons (and there may well be others), it seems we have to read the “indivisibility” constraint on souls as meaning they have neither spatial nor temporal parts.
Now I could very well be wrong about this, but it seems that if the theologian is committed to the view that the soul is a simple substance, then it cannot be divisible into either spatial or temporal parts. But perhaps there’s a way to maintain that a soul has no spatial parts while also somehow maintaining that it could have (coherently specified) temporal parts. Anyone out there know how to pull that off?
Pardon an objection at the threshold, but it seems as much a mistake for a materialist to attempt to do ethics on a Christian ontology as it would be for a Christian to do ethics on a materialist ontology.
Even if we had a univocal Christian account of the identity conditions for a soul (which, as you point out, we don’t), we would still need to know the eschatological details. E.g., is it bad knowingly to end the life of a soul-laden blastocyst? Perhaps–if doing so annihilates the soul. Certainly not–if doing so releases the soul directly into eternal bliss.
Such details would vary unmanageably if doing ethics on a materialist ontology were deemed too parochial and we opted instead to accommodate all variant ontologies. (Why restrict ourselves to Christian varieties?)
Dave,
You say “If, on the other hand, you maintain that what makes the embryo a human being from the moment of conception is that it houses a soul, then you can avoid the problems they cite. After all, upon twinning, one soul could migrate to one of the new organisms, while a new soul could pop into existence to be housed in the other new organism.”
This seems false. If the “other new organism” gets its soul only after twinning (which happens days after conception), then it’s false that that organism had its soul from conception. Also, assuming that each of the twinned organism has a distinct soul, it doesn’t solve the problem of recombination, such that up to 14 days (if memory serves) after conception, the twinned organisms can recombine. This presents the unwanted implication that either one organism can have two souls or a soul can leave a body without that body dying.
Also, how are we to make sense of one zygote (prior to twinning) having two spacetime worm-bodies? There are really two questions here. First, are we to understand a (say) four-cell zygote as having two of its cells making up the spatial aspect of one worm-body and the other two making the other? This seems counterintuitive, and it seems hard to see any principled way to determine, among the four cells, which two cells are assigned to one worm-body and which two cells are assigned to the other. But maybe the half-the-cells approach is mistaken; maybe I just don’t understand what a worm-body is. (I don’t find it very intuitive to think of one road as containing two roads prior to the split; I think of it as being one road, which then *changes* into two roads.) Second, regardless of how that first question is answered, how is it possible to have two worm-bodies in (what is initially) a single-cell zygote? It seems impossible to say that there is anything, at level of organism, that can be conceived of as composed of more than one individual without destroying that organism. That is, as a single-cell zygote, it seems indivisible qua zygote (let alone qua human being!). And, of course, two does not equal one.
First, briefly, to Tadlow: I’m not trying to do ethics from one ontology or another. I’m simply trying to see whether or not there’s a coherent conception of immaterial substance that can make sense of the claim that a zygote is a human being. There are a variety of conceptions of the soul, to be sure, but it seems it’s the Augustinian conception that holds sway among popular proponents of the argument, and it’s also the one that’s been most worked out in the Catholic Church. The Thomistic conception has also been carefully articulated, but for the purposes of my argument here, I’ve focused solely on the former. (This is actually part of a larger project that does seek to deal in detail with the Thomistic conception as well.)
To Josh: you’re right that the “lives of all human beings begins at conception” line would be undermined by the twinning case if the theologian maintains that a new soul pops into existence at that point to be housed in one of the “new” organisms. I explicitly recognize that point in the larger paper from which this is drawn. That’s why I focus here on the more plausible move, which is for the theologian to try and go four-dimensionalist instead and suggest that the pre-twinning embryo actually houses two overlapping souls which then separate along with the bodies upon twinning.
This move thus addresses your point about which cells would house which soul. The answer is that there would have to be two bodies whose temporal parts entirely overlap (so it’s not the case that a different part of that body houses one soul — the two worm-bodies would wholly overlap at that point in time). To take an L.A. example, the 101 and 134 freeways have parts that wholly overlap around Glendale before separating off east of there. They are two distinct roads that simply coincide for a stretch. The same would go for the two distinct worm-bodies pre-twinning: they share *temporal* parts for a few days before separating. But if these material bodies are to be human beings, they must have souls that also overlap with one another in *their* temporal parts. But I fail to see how we can make sense of souls having temporal parts. If instead they are simple substances, they can’t have temporal parts (or so it seems to me), in which case you can’t view them as part of a four-dimensionalist ontology. So the bodies must be viewed from a four-dimensionalist ontology (to best address the twinning worry), while the souls must be viewed from a three-dimensionalist ontology, which renders the question, “Do human beings endure or perdure?” impossible to answer.
The question of recombination is also quite important, but again, if you go four-dimensionalist, the question of what happens is easy enough to answer *for the bodies involved*. They would simply be two distinct worm-bodies with a long stretch of coinciding temporal parts (post fusion). But you’d have the same problem accounting for the souls that would have to be involved if you can’t go four-dimensionalist for them as well.
David,
Interesting topic – I never thought before about whether souls should be thought to endure or perdure.
It seems to me that there are ways for both the “enduring soul view” and the “perduring soul view” to avoid your objections.
1. About the “enduring soul view.” You say “the view implies that human beings neither endure nor perdure. But if these are logically exhaustive possibilities, then the soul view (in this version) is incoherent.” But doesn’t the very case that you are considering — the possibility that a person is composed of an enduring thing and a perduring thing — show that these are *not* logically exhaustive possibilities?
(In what follows, I will speak of a person on this view as being a *fusion* of an enduring soul and a perduring body rather than as a *set* containing an enduring soul and a perduring body. I’m not sure whether sets are persisting objects at all.)
Here’s a parody argument. “Everything is either abstract or concrete. But consider the fusion of my coffee mug with the number two. This is neither abstract nor concrete. Therefore there is no such thing as the fusion of my coffee mug with the number two.” Now, maybe there are other reasons to deny the existence of such an odd object, but I don’t think this argument provides a good one. If you had independent reasons to believe in such an object, you would thereby have good reason to deny the premise that everything is either abstract or concrete.
Likewise, it seems to me, when it comes to endurance and perdurance. If we have reason to believe in fusions of enduring and perduring objects, then it seems we should just say that there are persisting things that don’t exactly endure and don’t exactly perdure. And it shouldn’t sound mysterious — we are not committed to some mysterious third way of persisting. We know exactly what is going on: the thing has a part the endures and a part the perdures, and that’s that.
2. About the “perduring soul view.”
a. The overlap objection. Your first objection is that since souls are not located in space, there is no way to define ‘overlap’ (whereas ‘overlap’ for material objects can be defined in terms of spatial occupation). But I don’t think we need space to define ‘overlap’ (either for material or immaterial objects). ‘Overlap’, I believe, is typically defined in terms of parthood:
x overlaps y =df. there is some z such that z is a part of x and z is a part of y.
(I do believe some formulations of classical mereology take ‘overlap’ as primitive and define ‘parthood’ in terms of it. But I think it makes no real difference.)
You complain that “if an object, like the soul, takes up no space to begin with, it is difficult to see how to specify what it would mean for its temporal part to overlap with anything.” I think it just means that there is some part of the soul — temporal part, of course — that is also a part of that thing.
b. The unifying principle objection. Why not go for a “psychological continuity” unifying principle for souls? You say there are no “psychological elements constituting” the souls. There are indeed no psychological parts (at a time) to the soul, but surely the soul itself instantiates psychological properties. This is consistent with the soul having no parts (at a time) and is just what is needed for psychological continuity.
3. I think I have become convinced that there are no good formal objections to a soul view. The view can be made coherent. Perhaps the only objection comes from Ockham’s razor.
Dave,
I’m still a bit confused. You say “there would have to be two bodies whose temporal parts entirely overlap (so it’s not the case that a different part of that body houses one soul — the two worm-bodies would wholly overlap at that point in time). To take an L.A. example, the 101 and 134 freeways have parts that wholly overlap around Glendale before separating off east of there.”
Forget human bodies (and souls) for a moment and focus on the freeway. I don’t at all find it intuitive to say that there are two freeways whose parts (temporal or not — I’m not seeing any real help here provided by four-dimensionalism) wholly overlap. I find it much more intuitive to say that there is one freeway during that stretch, which we think is appropriate to call either 134 or 101. When they split, then we have two freeways, one of which is called 134 and the other 101.
So, again, I’ve got two questions. Are there any cases, unrelated to the issue of fetuses, where someone with my intuitions about the freeway might find it intuitive to think that there are actually two wholly overlapping spacetime bodies in one object? And, is there any way of determining when we have two, three, four, or simply one (if not, why should we say there are multiple bodies)? Maybe this is just one side of Chris’s Ockhamist point.
Josh,
You ask: “Are there any cases, unrelated to the issue of fetuses, where someone with my intuitions about the freeway might find it intuitive to think that there are actually two wholly overlapping spacetime bodies in one object?”
Here’s a different case of partial spatial overlap. Consider a pair conjoined twins that are conjoined near the hand. We’ve got two people sharing a hand. And let’s say that the shared hand is truly shared — there are no grounds for saying it “belongs more” to one of the twins than to the other.
Now suppose you see a hand reach into the mail slot in your door and drop off a letter. It looks for all the world that just one person reached his hand in. After all, you saw just one hand. If you had pointed to the hand you would have been certain you were pointing to just one person. But in reality, you would have been pointing to two people. You just didn’t know it because you could see only a part of a person, and it happened to be a shared part.
The same goes for fission cases. It looks for all the world like there is just one embryo (or whatever) there. But if you were to point at it, you’d be pointing at two people (or embryos or whatever). You just wouldn’t know it because you could see only a part of a person (a temporal part, that is), and it happened to be a shared (temporal) part.
I too have an objection from the start. I am uncertain why the metaphysical possibility of souls means that souls must be taken into account when considering the ethical nature of the human person. As you point out, materialists simply ignore the idea. I fail to see a problem with this. If souls exist, is there any way to tell that they do? It seems to me that belief in souls is completely a matter of faith, for what arguments or evidence could be submitted to convince one who does not believe that souls exist to change his or her mind? It might not be satisfying to think that Christian ethicists and materialist ethicists discourse in fundamentally separate spheres, but that seems to be the conclusion one must draw once it is realized that belief in souls is a matter of faith.
Chris,
Thanks for the case. I confess that now I share the intuition that two persons could (partially) exist in one body part. But I’m still confused about three things. First, while I share the intuition that the hand is shared by each of the twins, I have no idea how to understand the possibility of this being the case. Second, are there any cases where someone with my intuitions about the freeway would be convinced that two people could wholly and not merely partially overlap? Third, I still wonder whether there’s any principled criterion for distinguishing when one body contains merely one, or, instead, two, three, or four overlapping persons. I noticed that in your last comment, you mentioned that we can’t tell that there are two persons there, because there’s only one body. But on this score, I’m not even asking for a decision procedure for how to tell when two persons occupy one body. Rather than placing such a heavy burden on this view, I’m merely wanting to know the criterion specifying that in virtue of which it is the case that one or two or three persons, rather than any other of these options, occupy a single body.
Josh,
You ask: “are there any cases where someone with my intuitions about the freeway would be convinced that two people could wholly and not merely partially overlap.”
About the freeway case, you said above, “I don’t at all find it intuitive to say that there are two freeways whose parts … overlap. I find it much more intuitive to say that there is one freeway during that stretch.” But I think what you find unintuitive is compatible with what you find intuitive. There *is* just one freeway during that stretch – there’s just one freeway-segment during that stretch. And that one freeway-segment is a part of two distinct freeways (each of which is much longer than the shared segment). When you are on the shared freeway-segment, you can say you are on two freeways at once. Just as when you point to the shared hand in the conjoined twins case, you can say you point to two people at once.
I’m not sure it makes any sense to say that two people *wholly* overlap. Two things can never wholly overlap. If “they” did, then they would share all the same parts, and would be (according to classical mereology anyway) just one thing. I don’t think in any of these cases were we being asked to imagine two things that exactly overlap.
You say: “I still wonder whether there’s any principled criterion for distinguishing when one body contains merely one, or, instead, two, three, or four overlapping persons.”
There is. It is determined by what will happen in the future. If tomorrow you will fission into two people, then (at least according to the 4D worm view), there are two people in your chair right now (just as there were two people sticking their hand through the mail slot). If tomorrow you will fission into ten people, then there are ten people in your chair right now.
If we’re ignorant of how many people are in your chair, this is because we’re ignorant of the future (or of the past, if we’re talking about a fusion case). Just as if we’re ignorant of how many people are sticking their hand through the mail slot, this is because we’re ignorant of what’s happening on the other side of our door.
Chris,
Thanks for the discussion on this – it’s helping me sort through what I think about this stuff.
As for whether there are supposed to be two persons that wholly overlap, Dave wrote in his first reply to me that “the two worm-bodies would wholly overlap at that point in time”. Granting for the sake of argument that at T+1, after fission, the two worm-bodies would not wholly overlap, if I’m following correctly, Dave’s claim is that at T they do wholly overlap. Or, put differently, that temporal segment of the two worm-bodies would wholly overlap. This seems counterintuitive to me.
The future-oriented criterion for number-identification helps me understand what the 4-D view is trying to say here – “If tomorrow you will fission into ten people, then there are ten people in your chair right now.” But now (surprise!) I find this counterintuitive. But maybe I’m just not a good four-dimensionalist: my intuitions suggest that whether there are ten people or one person in my chair right now depends on what’s in my chair right now, not on what might happen to the thing in my chair tomorrow.
Finally, you say “one freeway-segment is a part of two distinct freeways (each of which is much longer than the shared segment). When you are on the shared freeway-segment, you can say you are on two freeways at once. Just as when you point to the shared hand in the conjoined twins case, you can say you point to two people at once.” I admit that you’re getting me further on board, now, but I still find it a bit counterintuitive. My intuitions make it tough to say that I am on two freeways at once. You and I agree that I am on only one freeway-segment. You think this is consistent with being on two freeways; I don’t – I think that if I’m on one freeway-segment, I can’t be on more than one freeway. But I’m not sure what would allow us to arbitrate between our clashing intuitions here.
Josh,
It definitely does sound bizarre to say that, for all you know, there are ten people in your chair right now. Lewis would say that this sounds bizarre because sometimes we count by person-stages, not whole (temporally extended) persons. And there is just one person-stage in your chair right now. So, says Lewis, if your intuition is that there is just one person in your chair right now, this must be because by ‘person’ you mean ‘person-stage’. Incidentally, I believe this puzzle is part of what motivates the so-called “stage view” variant of 4Dism, due to Ted Sider. On this view (still a perdurantist view), ordinary objects (like persons) are identified not with the 4D worms but with the 3D stages. The stage view may strike a nice balance between accommodating our 3Dist intuitions while enjoying the theoretical benefits of 4Dism.
About the freeways. I think it’s hard to deny that you’re on two freeways at once. Are you on the 134? Yes. Are you on the 101? Yes. Is the 134 identical to the 101? No.
I guess we’ve gotten way off the souls stuff!
First to Matthew: Surely matters of faith aren’t completely unsupported by reason. People certainly offer reasons for their belief in souls (I seem to recall Descartes offering some rather interesting ones), but the “faith” part simply constitutes a “leap” from the point at which reason may leave off. But insofar as reasons are offered, it seems perfectly legitimate to examine them as we would any other reasons.
Chris has been doing a very good job of explaining the 4D view, including citing the reason Sider abandons Lewis’s “worm view” for the “stage view.” I have my doubts about the stage view (perhaps similar to Josh’s), but this isn’t the place. Rather, I wanted to address Chris’s earlier arguments against the views I was advancing. First, the parody argument posits a fusion between mugs and the number two to undermine the seeming dichotomy between the abstract and the concrete. But this is disanalogous, it seems to me, given that my case was about a distinct substance subcategory — human being — that the soul theorist is positing as consisting of two (essential) parts. This is a clearly recognizable ontological object within our conceptual domain — the trick is to figure out what the identity conditions are for it. I don’t even know what to make of the mug/two object. Is it a substance? Something else? The presupposition of the soul theorist, on the other hand, is that “human being” is a substance concept, and as such, it will (it seems!) have to fall under the rubric of either enduring or perduring.
Second, defining “overlap” in terms of shared parts doesn’t help me with my difficulty of conceiving shared temporal parts of immaterial objects. Now perhaps this is just a difficulty with my own conceptual apparatus. I understand the formal nature of the definition (it’s pretty much what Sider says), but I simply can’t wrap my head around this kind of overlap.
Third, I can’t propose the unity principle of souls as consisting in psychological continuity, simply because immediately post-conception there just *is* no psychology to be had in the soul. I know of no advocate of the soul view who maintains, for example, that embryos *think*.
Finally, I don’t yet despair of showing the incoherence of this view — it’s just extremely difficult. In one of my earlier papers I called the soul a “slippery little sucker.” Slippery, perhaps, but not uncatchable.
If instead of following an Augustinian conception of soul we instead follow a Leibnizean conception of a soul, don’t most of the problems vanish? Most particularly the problem of temporality since when Leibnizean monads are temporal, space is an emergent property out of the relations of monads. Indeed while a true relativistic ontology doesn’t fit General Relativity (despite apparently Einstein’s efforts) there still are attempts within quantum gravity to adopt a Leibnizean ontology of sorts. (Whether this can ever be successful is an other matter – I bring it up merely to point out that Leibniz’ process of materialism is still a current topic in philosophy of science)
Dave,
Excellent points.
1. I guess I’m not inclined to accept with ontological seriousness the notion of substance. As I see it, if you’ve got a perduring body and an enduring soul, then you’ve got a third entity composed of the first two. Maybe this is what persons are. Never mind whether the third entity is a “substance” or has “substantial unity.” It exists — that’s good enough. (I would even think that dualists should be friendly to this approach, since it seems difficult for them to get substantial unity out of such diverse kinds of things as a material body and an immaterial soul.)
But I realize that you’ve got quite a respected tradition in philosophy on your side.
2. I agree that the idea of shared temporal parts of immaterial objects is a little strange. I think this kind of case can occur only when the objects involved are rather gerrymandered. Here’s one such case. Consider Descartes’s entire 4D soul (I’m not sure how temporally long it is – maybe it still exists). Now consider the temporal part of his soul that lasted just for the 30th year of his life. We’ve got two immaterial objects here, the latter being a temporal part of the former. So this is a case of shared temporal parts of immaterial objects (of course, one of the objects is *entirely* a part of the other, but it’s still an example of the idea).
3. Very good point about no psychology immediately after conception. I hadn’t considered that. The dualist might have to take the identity of souls across time as primitive. That’s definitely a cost, so I think your original argument is pretty good. (Incidentally, I believe Descartes maintained that souls are essentially thinking things. So I guess he could not accept the view under consideration. He would be committed to the view that the soul comes into being only when thinking (broadly construed) begins.)
Fun stuff to think about.
Chris,
You say, “About the freeways. I think it’s hard to deny that you’re on two freeways at once. Are you on the 134? Yes. Are you on the 101? Yes. Is the 134 identical to the 101? No.” I said earlier that I didn’t know of any ways to arbitrate between our clashing intuitions on this question. Your statement doesn’t help push me in another direction, but it does tell me that maybe I need to clarify my own position a bit.
Let’s assume that freeways are not some sort of abstract entities that exist apart from the stretches of concrete that constitute them, as seems implausible. Rather, I think it’s more plausible to say that freeways just are stretches of concrete. If that’s the case, then of course it seems hard to be on two freeways at once, since (highly unusual traffic accidents aside) it’s impossible to be on two distinct (highway-sized) stretches of concrete at once. So, then, when I’m on the stretch of concrete labeled 101/134, since it’s one stretch of concrete, it also seems to be one highway, given the identity of the highway and the stretch of concrete. So you’re right, I am on both the 134 and the 101. But this is a bit misleading, since using two numbers might seem to refer to two different highways; more precisely, I’m on one highway (or, even more precisely, highway-stretch) that has two different names. So I’d deny – with a qualification to be made momentarily – that this highway-stretch of the 134 isn’t, in one sense, identical to the 101. That is, it doesn’t seem “hard to deny that I’m on two freeways at once” – in fact, it seems pretty intuitive to me to deny that, since I’m on only one stretch of concrete and I identify freeways and stretches of concrete. The qualification is this. Stretches of concrete are divisible, so it makes sense to talk about parts of a freeway. Therefore, it is possible to say that part of the 134 is identical to part of the 101, while simultaneously maintaining that (other) parts of each are distinct. So in response to the question “Is the 134 identical to the 101?” I’d say the right answer is neither an unqualified “No” (as you stated) nor an unqualified “Yes,” but, rather, “Partially.” This may seem counterintuitive. After all, two things either are or are not identical. So I don’t mean that the 134 and the 101 are “somewhere between identical and non-identical.” Rather I mean that parts of them are identical and parts of them are not.
One response you might make would be to say that freeways are not – or not merely – identical to stretches of concrete. So, someone might say “But the 101 heads south at Studio City, when the 134 heads east, so they can’t be the same.” Or someone might say “But given that there are parts of the 134 that are not identical to the 101, surely the 134 and the 101 (considered whole) are not the same thing.” I sympathize with these intuitions. But the only way to say that I’m on two highways at once would be to say that a highway is something else than just the concrete that constitutes it (since I’m only on one stretch of concrete and one does not equal two). I find that to be a counterintuitive conception of what a highway is. So in a sense what I want to say, to make sense of these various intuitions, is that there are actually three freeways: the 101, the 134, and the 134/101 combination, which is different from the 101 simpliciter and the 134 simpliciter. That seems more plausible than saying I’m on two freeways at once and denying that freeways just are stretches of concrete. One final point: I realize that there are reasons to deny that there are three things in cases like this (cases of persons in particular). But to me this is only to point to a philosophical problem (especially given my materialist intuitions about freeways [and, for that matter, persons]), not to solve one.
You’re right that we’re pretty far away from souls at this point, so you can have the last word on freeways.
Josh,
I’m beginning to think that maybe our disagreement is merely verbal.
I definitely agree that freeways just are stretches of concrete. You say, “it’s impossible to be on two distinct (highway-sized) stretches of concrete at once.” If by ‘distinct’ you mean ‘wholly distinct’ (as in not sharing any parts), then I definitely agree. But if by ‘distinct’ you merely mean ‘not identical’ (leaving open the possibility of shared parts), then I disagree.
I am also in complete agreement with just about everything you say in the following passage:
“Stretches of concrete are divisible, so it makes sense to talk about parts of a freeway. Therefore, it is possible to say that part of the 134 is identical to part of the 101, while simultaneously maintaining that (other) parts of each are distinct. So in response to the question “Is the 134 identical to the 101?” I’d say the right answer is neither an unqualified “No” (as you stated) nor an unqualified “Yes,” but, rather, “Partially.” This may seem counterintuitive. After all, two things either are or are not identical. So I don’t mean that the 134 and the 101 are “somewhere between identical and non-identical.” Rather I mean that parts of them are identical and parts of them are not.”
Although I will add that as I use the word ‘identical’, the answer to the question “Is the 134 identical to the 101?” is an unqualified “No.” This is consistent with the claim that they are partially identical – that they have a part in common. Partial identity is not identity (as I use the latter term). x = y only if every part of x is a part of y, and vice versa. Maybe this verbal point explains some of our apparent disagreement.
Here’s another verbal point. Consider this sentence, uttered when you are on the place where the 101 and the 134 overlap:
(1) “I am on just one freeway right now.”
You are inclined to accept it; I’m inclined to deny it. But the sentence is ambiguous, so maybe we do not disagree. It is ambiguous between:
(1a) “I am on just one freeway-segment right now.”
(1b) “I am on just one complete, whole freeway right now.”
My view is that (1a) is true, and (1b) is false. If you agree, then our apparent clash of intuitions is only apparent.
I’ll finish with my picture of the freeway case. The name ‘101’ names a long stretch of concrete. The name ‘134’ also names a long stretch of concrete. These two stretches partially overlap. They share a part. We can say you are “on” the 101 even when your car touches just a tiny part of the 101. (We surely can’t require that one be touching *every* part of the 101 in order to be able to say truly that one is “on” the 101; otherwise no one would ever say something true when he says, “I am on the 101.”) All this goes for the 134 as well. When you happen to be on the shared part of the two freeways, it is therefore true to say that you are on the 101 and true to say that you are on the 134. But in this sentence the word ‘101’ still refers to the very long stretch of concrete that is the whole 101. And the word ‘134’ still refers to the very long stretch of concrete that is the whole 134. Although these long stretches overlap in one place, they are not identical: they have parts not in common. So they are two. So there is a freeway x and a freeway y such that you are on x at t, and you are on y at t, and x is not identical to y. So you are on two freeways at t.
Or that’s how I see it, anyway.
Anyhow, this is fun topic. I’ve enjoyed discussing it.