A Puzzle About Twins

Suppose people start existing on the first day after conception--just suppose.  Here's an argument that seems to show that identical twins must be an exception.  Take Betty and Casey, who come into existence as my diagram depicts--

Jeff McMahan makes this argument--
[Betty and Casey] ... cannot both be identical with the original zygote for, given the transitivity of identity, that would imply that they are identical with each other, which they clearly are not. We must conclude, therefore, that when monozygotic twinning occurs, the zygote that divides thereby ceases to exist and two new zygotes begin to exist at that point ....  What this means is that, even if most adult human organisms begin life at conception, monozygotic twin organisms began to exist somewhat later, when a zygote that began life at conception divided. (Jeff McMahan, The Ethics of Killing, pp. 25-26)
Maybe I'm missing something, but it seems most natural to say that there are two entities here, Betty and Casey, with an overlapping stage--Zygote (A).   A is a stage of both Betty and Casey.  Betty isn't identical to A, and Casey isn't identical to A, because A is just a stage.  Nevertheless, A is the first stage in the life of both Betty and Casey.  They start their lives when A begins. 

It seems like lots of things have overlapping stages like this, or at least you can dream up lots of things. Take for example The Two Funerals.  Two funerals take place during the same two-hour time period in a funeral home. The scheduler messed up, and the reception for each takes place in the same room.  Mourners talk to each other about their respective deceased, sign the same guest book, drink from the same wine bottles, etc. Then the two funerals split apart, with services taking place in two separate rooms.  Paralleling McMahan's view of the twins, you'd have to say three funerals took place--the reception-funeral, which ended at the end of the reception, and then the two services that took place in the separate rooms.  But no, surely, there are two funerals, which shared a first stage.

Boy, what fun I could have dreaming up more overlapping events like this.  Two parades start off separate, but merge for 10 minutes, then separate again.  Using McMahan's reasoning, we've got to think there are actually five parades here.  At the merge, the first two come to an end.  Then there's the merged parade. After the merge, another two come into existence.  Five, count 'em, five!

If you think life generally begins on the first day of gestation, it seems to me you can think so for identical twins as well.  But is that what we should think? I would like very much to think about personal identity and gestation without playing abortion-chess--i.e. looking ahead to the implications of various views for abortion ethics.  Most of the time when we're coming to terms with when life begins, we're thinking about a life that continued (our own, our children's lives) or that will continue (a wanted pregnancy). It seems like this ought to be the context for reflection at least some of the time, instead of abortion ethics always being the elephant in the room.


Jeremy Bowman said...

Identity depends on what we count as a "property", and that is contextual. "The" wave that sunk the Edmund Fitzgerald is one and the same wave as the one that nearly sank its sister ship a few minutes before - but only if we discount what would count as properties in many other contexts.

In the case of growing persons, we have to take account of temporal position, and often do.

Alan Cooper said...

Yes, I agree that "identity" is a different issue from "life" (in which we are all branches of a continuous stream which began running billions of years ago). But it is not just in the sense that one can never step into the same stream twice, and he question of existence as individual human identity is indeed problematic for a blastula which has the potential for splitting into two or more.

However perhaps the questions of humanity, personhood, and individual identity are actually all distinct. Consider the science fiction replicator which acts by destroying the object to scan its structure and then builds two identical copies (which share the memories but not the atoms of their deceased progenitor). The "humanity" of all three is probably universally accepted for moral purposes, though the actual *number* of individuals may be less well understood.

arvizzi said...

Your argument seems to depend upon the zygote split being somehow predetermined, genetically or otherwise. However we know that the split is basically a random occurrence, though presumably with influencing factors. You want to take the split event and retroactively impose some idea of overlapping individuals in the past. But the split could very well not have happened. Unless we want to get all funky by introducing reverse causation, a zygote isn't influenced by its future state.

Jean Kazez said...

It seems I can believe the spit is determined, or not determined, and still have the same position. In either case, it's going to turn out, in the fullness of time, that the zygote splits. This is not to say there's some sort of reverse causation, with the future affecting the past. It simply will split. So it will turn out that the zygote was a stage of two different individuals.

This seems like the right thing to say, as long as there are no general problems with thinking of a zygote as a stage in the development of an individual (in a non-twin case). If there's a problem, it seems to me it's with the general picture of zygotes as stages--there's no special problem involving twins (as the funeral and parade examples show).

faust said...

McMahan's argument is as follows:

1.Identity is transitive
2.Twin zygotes (B) and (C) are not identical with each other.
3.Therefore twin zygotes (B) and (C) cannot be identical to (A)

Your response is to acknowledge that neither (B) nor (C) is identical to (A) but that (A) is clearly a “shared stage.” If we view the lives of either (B) or (C) in toto, then we will find that those lives are composed of a series of stages and that at a certain point they shared a stage.

But this logic commits us to the view that individuals are composed of series of stages none of which we are identical to. Does this make sense? It certainly seems to. In fact, because human life seems to be generally cumulative NO stage of human life (even if you view it second by second) is completely identical to any of the preceding stages of life. Is fetus identical to the zygote even in a non-twin scenario? Is the newborn baby identical to the fetus? Is the child of 5 identical to the newborn; the pubescent identical to the 5 year old; or the well educated professor to an undergraduate self?

Using McMahan’s logic it’s not clear to me why I need to think that any stage of life is IDENTICAL to any other. He uses the twin split of the zygote to clearly illustrate that the two entities are non identical to the previous stage. But I don’t see why you even need that splitting. Mere evolution over time seems sufficient to establish the discontinuity.

But what of your funeral? It certainly seems to make sense to say that it is the same funeral throughout, but that is because we are here defining funeral as “the event at which we mourn and/or remember the passing of X.” So defined we have broad leeway in thinking of a wide variety of configurations as essentially “the funeral of X”

In the case of human life we try to do something similar. Here is the LIFE of X it BEGAN here and continued until THIS TIME. Therefore all of these stages constitute THE LIFE X. It is the LIFE X that is unified phenomenon that is the unified object, even though its internal stages are not identical to each other.

On the other hand McMahan is correct if what we want to find is the UNIQUE life of (B) or (C). Arvizzi’s commentary about the teletransporters seems thoroughly appropriate here. LIFE X that goes through a splitting teletransporter and becomes LIFE Y and LIFE Z faces precisely the same identity challenges as the zygote example.

If we ask Y and Z about their lives they might very much say something like what McMahan says about the zygotes. “Before we stepped through the teletransporter we were the same person. Now we are different people. In a sense each of us began a NEW LIFE when we stepped through the machine."

arvizzi said...

"It seems I can believe the split is determined, or not determined, and still have the same position. In either case, it's going to turn out, in the fullness of time, that the zygote splits."

I guess I don't understand. If the split is not determined, then it might not turn out that the zygote splits. It seems to me that you are assuming determinism.

Pre-split, there is no test that can tell us whether or not a split is going to happen. Such a test is not possible even in principle -- the information just isn't there. A zygote is a zygote is a zygote. The idea of overlapping identities comes down to a metaphysical supposition (stemming from determinism).

I think Alan Cooper had the right idea above -- this is merely a variation of the replicator experiment, covered e.g. in Reasons and Persons.

An interesting point to think about is that whatever argument we make must be consistent with the phenomenon of a chimera -- two zygotes fused into one. I don't remember Parfit covering that! The only thing which comes to mind is The Fly.

Jean Kazez said...

(Just FYI-- I have no commitments on these issues. I'm just thinking aloud.)

It seems like stage-talk is natural for zygotes. Either a zygote is a stage in the development of one organism, or 2, or 3, or whatever-- We'll know in the fullness of time. Whatever the answer, a zygote is never more than a stage. If you go with that, it has the happy result that you can say twins start to exist at conception, just like singletons. It seems downright strange to tell different stories about when existence begins for twins and singletons.

On the other hand, if you've got a 50 year old person who's replicated/destroyed, and two doubles come out the other end, we have the intuition that the original is a full-blown person, not merely a stage. So we're pushed toward saying there are three individuals in the story, not two that share an initial stage. The first dies, the subsequent two come into existence as a result of the replicator.

In other words, zygotes seem especially amenable to being treated as stages in something larger. That's the idea...

arvizzi said...

But everything I've said about the phenomenon of twinning is antithetical to the idea of stages. It's not something in the plan of a zygote; it's not in the genetics. A zygote is a zygote. Twinning is a random accident with an interesting outcome. It happens more often for older mothers, and this fact underscores the "something unexpected happened" nature of it, as older mothers are more likely to encounter complications.

Suppose you have a car accident on the way to work, in which your car's front grille is broken in two. Can we say that from the time you left your house to the time of the accident your grille was in the stage of becoming two half grilles? After all, in the fullness of time we know that one will become two.

The first problem is that there's nothing about the grille itself which suggests a stage of development into two half-grilles. Second, even if we accept the idea of a "grille-doubling" stage, it only exists in hindsight, or in a metaphysical sense if determinism is assumed.

Jean Kazez said...

Arvizzi, That's all food for thought. I need to think more about stage-talk--when it's appropriate, what it accomplishes, etc. I may possibly come around to what you're saying, but at this point I'm going to have to plead "agnostic". Thanks for your comments.