Blast from the Past: "The Doctrine of No Self" Cri

Newbie Joined: Oct 25, 2009 Posts: 14

(I could revise this forever and this will probably not get any better, so here it is:)

In Defense of Self
A Review of "The Doctrine of No Self"

I just recently finished listening to the IG show with Charles Goodman regarding the doctrine of no self, and I remain unconvinced that I have no self. Now, Goodman claims that the atheist cannot support the idea that he has a self, and tries to show that using just rational arguments. The problem is that the arguments don't hold water in any way. They are chock-full of fallacies, false premises, and plain ol' faulty reasoning, and as such I remain unimpressed.

(This review is going to have some math, but I'm a math-guy. You have been warned!)

In Defense of Time and Identity:

It is really the defense of self in terms of continuity across time that is the key bastion that must be defended in order to preserve the concept of self. Once you give up that you are the same self as you were five years ago, in terms of identity, it's a losing battle from there on. On the other hand, if the concept of continuity of self across time is upheld, then Goodman's further arguments become almost trivially easy to defend against. It is here that a defense of self must receive the most attention.

Goodman sets up what seems to be a damning argument. He asks us to consider what I would call the "perfect copy machine", although he presents it in terms of a Star Trek: The Next Generation episode where the role of the copy machine is played by the transporter being all goofy. He asks whether a copy of a person is the same person than the original person, after fifteen years of being separated. The answer is arguably no, as the pairs' experiences have changed them in fundamental ways. If there's such thing as identity, and the the two are otherwise interchangable, if one of the pair is not the same person as was copied, then by symmetry, the other must share that status. After all, as materialists, we do not believe in anything other than what is material and physical, and the two are identical physically just after the copy.

It's a powerful and pursuasive argument at first blush. Goodman uses it to defeat the notion that we have a self that is preserved across time. However, the argument presented above has some very deep logical flaws that reveal themselves on close examination.

Intuitionalism:

From what I read about Charles Goodman, he has all the trappings of a philosopher. Philosophers have a very annoying tendency to put their intuition on the same level as (or even above) real physical evidence or conclusions born of axioms and logic. The interview Goodman gave on the show bears out this observation, as he tries to use your intuition of how situations should be resolved in various scenarioes to force you into thinking that the notion of self has a problem.

While the usual formulation no doubt does have problems, it doesn't mean a priori that a good axiomatic treatement of self will not resolve the contradictions, albeit likely in a counterintuitive way. We should not be scared of counterintuitive notions. After all, a person's intuition is based on a very narrow range of experience. It should be of no surprise that it lets us down when pushed beyond that narrow range of experience.

In other words, when considering unfamiliar circumstances, one's intuition should not be used as a guide to the truth. If the answer seems counterintuitive, then assume that your intuition is wrong.

As it applies in Goodman's argument, we have an intuitive sense that a person should have exactly one identity that belongs only to him/her; no more, no less, and held by no other person. We have a sense that if 'is the same person' is false for a pair of people, then they do not share an identity. Goodman is (perhaps unconsciously) counting on you holding this background notion to defeat the self.

But why should personal identity work this way in such a circumstance? Because our intuition tells us? No, that's not a good reason. Our notions of 'the same person' were developed in the absence of perfect copy machines. Our experience and intuition cannot be any reliable guide to how identity should behave when considering the perfect copy machine. That is a matter of axiom and logical argument in a rigorous treatment of personal identity.

Equivocation:

We now examine the phrase 'is the same person as' a bit more closely. We want it to be in regards to a shared personal identity across time. However, the phrase 'is the same person as' doesn't always mean that, even in the absence of a perfect copy machine. It can mean that the person has some continuity of identity across time and space, or that the person preserves his character over his experiences.

It is the former meaning of this phrase when we say there is such a person as "Thomas Jefferys" that exists over a lifetime. However, it is the latter sense of the phrase 'is the same person as' that Goodman uses to argue against the self — I'm not the same fellow (in terms of character) that I was five years ago, or even moment to moment, therefore my self cannot be defined in time.

Goodman is arguing that the latter use of 'is the same person as' cannot be satisfied across time, so the former use of that same phrase cannot apply. But we don't take the phrase to mean that for the purposes of establishing an identity: it is not my intent to prove that I am of exactly the same character when I whip out my driver's licence, or say "I'm the same guy you met last week." This makes Goodman's argument based on a fallacy of equivocation.

Needless to say, I am not impressed by the use of this fallacy, even unconsiously. A doctorate should know better.

Self as a Dynamical System:

If we cannot found identity on equality, we must find another foundation for identity. As a guide, we look to the study of dynamical systems. As with the self, which does seem to change over time while retaining identity, dynamical systems too change over time while retaining identity even as their values fluctuate. If equality is necessary for identity, then this entire, well-founded mathematical field would fail spectacularly; they are systems that —by their nature— do not remain equal to themselves, yet maintian an identity across time.

The identity of two points in a dynamical system is decided by whether there is a continuous orbit through them. Given that the self also seems to be a dynamical entity, we can use this notion of continuity as a foundation for personal identity. Two people in different times share an identity if there is a continuous orbit (in relativistic lingo, a world-line) through the both of them. Under normal circumstances, this works fine with our intuition — people are not normally copied nor is infernal surgery or medical technology injected into them to mess with their very selves, nor teleported to remote planets. Continuity seems like a good starting point.

By appealing to geometry, Goodman clearly thinks that equality is necessary for identity. However, we see above an alternate definition that works fine under the domain of our experience, and works better than Goodman's definition in that domain of experience.

Misuse of the Transitive Property:

For brevity, we will say that the statement "A is the same person as B" can be written as "A ∼ B", and its negation as "A ≁ B". Let's say that myself, symbolized as W, am duplicated through a perfect copy machine into Wʹ and Wʺ. Let's say Wʺ is sent to live on Mars for the rest of his life, while Wʹ lives on Earth. Then arguably, Wʹ ≁ Wʺ. If we then assume W ∼ Wʹ, then by the 'transitivity of identity', it must mean W ≁ Wʺ, and that somewhere along the way, W must have ceased being W when it became Wʺ, right?

Wrong.

Let's be clear here: Goodman was arguing that personal identity (if it exists) has the transitive property, a point which I agree with. Also, I think an argument can be made for the reflexive property (that is, A ∼ A is always true). However, these two properties are insufficient to establish an equivalence relationship like identity. A logical implication A⇒B is only false if A is true but B is false; if B is true, the logical implication is true even if A is false. Thus, transitivity allows the case where W ∼ Wʹ, Wʹ ≁ Wʺ, and W ∼ Wʺ!

We math-heads call a relation that is reflexive and transitive a preorder . The relation is only an equivalence relation when symmetry is also thrown in (that is, A ∼ B implies B ∼ A); if the relation is instead antisymmetric (that is, A ∼ B and B ∼ A implies A = B — I don't make these terms up), then the relation is a partial order. If ~ is a partial order, then we can have W identified with both Wʹ and Wʺ and yet have Wʹ not identified with, or equal to, Wʺ!

This is not a paradox! Or rather, this isn't a bad kind of paradox, the kind that involves a genuine contradiction — like Russell's paradox or Curry's paradox. Rather, this is a surprising kind of paradox, resolved by the fact that the conditions are odd, so familiar notions are behaving in a counterintuitive, but not actually inconsistant, manner.

The second thing wrong with Goodman's use of the 'transitivity of identity' is that an equivalence relationship cannot itself establish equality. On its own, an equivalence relationship only establishes non-intersecting equivalence classes . This is a subtle but important distinction. It means that two quantitites can be equivalent by one relationship, but unequivalent by another. It depends on how we break up the universe of interest into equivalence classes. The two expressions "1" and "sin(pi/2)" belong to the same equivalence class in that they both evaluate to the same real number 1 (that is sin(pi/2) = 1), but they are not in the same equivalence class in the sense that they are strings of different length. Similarly for any other mathematical expression that evaluates to 1, or any other number.

Therefore, even if ~ was an equivalence relation and W ∼ Wʹ, Wʹ ∼ Wʺ, and W ∼ Wʺ, that does not mean that Wʹ = Wʺ. So, although Wʹ would not be identical to Wʺ, they could still belong to the same identity, established by the equivalence class [W]. Again, this is not a paradox, merely a peculiar behavior resulting from our intuition of identity being pushed beyond our experience, and the subtle difference between an equivalence class and an equality.

Alternately, we could define identity by stating that a set of persons A establishes an identity if the personal identity operator ~ (a partial order) is also a total order over that particular set A. What this means is W has two identities associated with him. One of them continues on earth to be Wʹ, and the other branches away to become Wʺ. (Note that this is also compatible with the conventional notion of multiple identities.) Note that both definitions preserve our notion that personal identity has continuity with our past selves.

Goodman would probably include all the above in the 'puzzles and paradoxes' regarding the self. However, I fail to see how any are anything more than counterintuitions that only exist because we lack any experiences with such situations, and I've stated before that intuition about unfamiliar experiences is a bad guide to truth.

Other Arguments against Self:

Once personal identity is defined using some criterion of continuity, most of Goodman's other arguments become moot. Because of continuity, Goodman's "memory ≠ personal identity" argument fails to stick, because we are not defining identity through memory. While continuity does mean that two people can share the same identity, that's not paradoxical because multiple identities are not forbidden. Someone who has been cloned will have some emotional problems and legal headaches (and it did in the case of the Rikers), but that's a problem with us and our society — the universe doesn't give a damn about our legal and emotional troubles.

For the "Your Brain on Nanobots" argument, we can already maintain identity through time and changes to self by way of continuity with our past. Whether he is of the same character as before is a separate question and concept. Reggie Finley's character has undoubtedly changed, and the way we deal with him is dictacted by that character, yet there is something real to interact with. Same with "Reggie Finley, Single White Female." Continuity (albeit only of the brain) is maintained, therefore identity is preserved.

Continuity also covers cases of brain damage — even if I were a vegitable, I would still be identified as the same person as before, just... less of my pre-accident character.

There is also a continuity maintained with teleporting across space, being constructed out of new matter: the configuration of your old self has been reconstituted at the new point because the configuration has been extracted from the old self and transmitted across space to construct your new self. Also, "new matter" is itself a very suspect phrase, because all particles of matter are identical. All electrons are identical, as are all protons. They literally cannot tell themselves apart. Therefore, it doesn't make a damn difference whether you use the old matter or new matter in the teleportation, the results are identical — you retain identity, provided continuity of your configuration is all that is required.

This is not to say that I have the complete, rigorous definition of the self that solves all the issues. Far from it. However, the self is in much better shape than Goodman portrays them.

Precisely Fuzzy Quantities:

Despite Goodman's claims to the contrary, we can define 'baldness' in a completely precise way that produces no paradoxes. The paradoxes that result in his argument occur because of faulty treatement, not because the concept of baldness is fuzzy. While in ordinary language we treat 'bald' as if it is a crisp definition, anyone can see that baldness comes in degrees. Treating baldness as if it comes in degrees immediately dispells the paradox.

Contrary to Goodman's claim otherwise, we know that fuzzy scientific concepts do not break down in science, because science rarely has the luxury of using completely crisp concepts: just about every measurement you care to make in science is vaguely-defined, mushy, biased, uncertain, and —on some level— wrong. We have mathematical tools that allow us to handle concepts and measurements that are vaguely-defined, mushy, biased, uncertain, and —on some level— wrong, if only we would use them and give the concepts and measurement the proper treatment and representation.

What annoyed me most about his "Baldness + 1 hair" discussion is that Reggie was about to steer clear of the trap Goodman had set up for him, because Reggie immediagely recognized baldness was a matter of degree as soon as the spear-bald man was juxtaposed with the spear-bald man plus one hair. But that didn't suit Goodman's argument, so he shoehorned Reggie, the poor fellow, into regarding baldness as a binary property which lead to the contradiction... which he purposefully set up himself.

No, the problem with baldness in that case was because it was given the wrong treatment. The proper treatment of a concept depends on its structure and behavior; to say that no treatement will work simply because one treatement you tried failed is a fallacy of hasty generalization, especially when the treatment is clearly wrong from the outset.

Buddhism:

Goodman states that the Buddhist tradition claims that if one were to believe that the self is a social construct of the same kind as the McDonald's corporation, then it would make one less selfish. He drives the point home with the terrorist example, where you are more concerned about horrible things happening to you and yours than some stranger. Buddhist doctrine claims that this is because you falsely regard yourself as a real entity. I do not buy this for the following reasons:

First, Goodman has failed to present a convincing case for the doctrine of no self, even after thinking about this with the tools of western philosophy for a long time. His argument has too many errors and fallacies to take seriously.

Second, the entire approach of peeling away (seemingly) unnecessary components of your person to (not) arrive at the self suffers from the fallacy of division. It does not follow that if an entity has a property (in this case "contains a self"), that necessarily all or any of its componets have that property. We have never observed a self outside ordinary matter — being material seems to have something to do with having a self. Yet we know that not all material things have a self, so something else is required — a particular kind of organization. Each separately is as much self as a rock or a sheet of equations. Does that mean that the self must not exist because neither component has a self? No, because it is otherwise lifeless matter being arranged in an otherwise abstract configuration that has selfness. The organization of matter may have properties not shared by either the bare matter or the particular organization itself.

Third, the Buddhist notion that divesting oneself of their sense of self will necessarily make you a kinder person suffers from three problems. The first problem with this notion is that possessing a self is not a necessary condition for producing behavior one would regard as selfish. Organisms without nervous systems —freaking plants — do all of the things we would regard as selfish behavior: behave greedily, do everything they can to stave off death (as if afraid), and show preferential treatment for nearby threats, and yet they lack complex nervous systems necessary for societies and selves.

The second problem is that neurobiology seems to point to exactly the opposite conclusion. Even by 2006, the time of the broadcast, neuroscience had already discovered and confirmed that higher animals have "mirror neurons." These neurons fire when the animal acts or when it observes another animal act. Empathy is therefore likely to be the result of being able to map another's emotion onto your own self. Thus, one's ability to feel for another's pain as your own is because you have a self, rather than in spite of it.

The third problem is that the notion that there can be suffering without selves depends on suffering being a substance separate from the beings enduring them. We have never seen an instance of suffering without a sufferer. While this is not absolute proof that suffering cannot exist without sufferers, Buddhism claims that this applies under ordinary circumstances — where our intuition and conventional definitions should still work. Conventionally, suffering requires sufferers.

The doctirine of no self remains as it began, a notion believed by followers of a prescientific system of beliefs.


Login/Register New account	Home · FAQ · News · Shows · Videos · Forums · Arcade · Listen & Chat · Book Store · Contact Us · Go Mobile