John Fischer

John Fischer (University of California-Riverside) “Freedom, Foreknowledge, and Frankfurt: A Reply to Vihvelin,” with commentary by Kadri Vihvelin  (University of Southern California).  The paper and commentary can both be found here.

Comments

[Brad Weslake]

The Vihvelin paper to which Fischer is responding can also be read online:

http://www.usc.edu/schools/college/philosophy/private/docs/vihvelin/Freedom.pdf

[Neil]

There’s a lot to say about these two rich papers (and that’s how we should think of them: not a paper by Fischer with a reply by Vihvelin, but an exchange between Fischer and Vihvelin). I’ll try to say some of it, though I’m not always certain what’s going on in the debate.

One reading: Fischer concedes that in the coinflip case, it is (in some sense) possible that the coin comes down tails). But he says that the case is disanalogous to a FSE, because in a FSE the counterfactual intervener remains on the scene to intervene if necessary. Because the intervener remains on the scene, the agent cannot do otherwise. Vihvelin replies that this confuses counterfactual and conditional intervention. We’re supposed to be talking about counterfactual intervention, because conditional intervention does leave the agent able to do something (eg, to begin to perform the forbidden act). Putting the intervener back on the scene allows them to be a conditional intervener, but that doesn’t show that the agent lacks the relevant abilities. Taking the intervener away makes her purely counterfactual, and then the agent is relevantly like the coin: it is possible for the agent to do otherwise. The intervener would just turn out to have had a false belief.

Here’s a reply to this line of argument. We ought to concede that the intervener is a conditional intervener. But – think of case like Pereboom’s tax evasion – the condition upon which the intervener would intervene is not the agent beginning to perform the forbidden act itself, but beginning to perform a necessary condition of that action, which is not itself a free action. So the agent remains able to do something, but that thing is not a free action; it’s a mere ‘flicker of freedom’. Of course, the intervener remains counterfactual in the actual scenario, just in the sense that she doesn’t actually intervene at all.

As to the coinflip claim, and the sense in which it is possible for the coin to come down tails: well, I must admit that I don’t think I’ve got this part of the dialectic at all. First, I don’t see why the case is a good analogy at all: coins don’t have abilities, just objective probabilities (of coming down heads or tails).

Second, consider this case. Suppose Black has god-like powers, but lacks omniscience: that is, it is not the case that Black has infallible knowledge of the future. In that case, even though Black is very reliable predictor of what will happen, Black’s predictions are not necessarily true. But Black has the god-like ability to roll-back time (like the God of van Inwagen’s well-known thought experiment), and then replay it, as many times as he likes. Suppose that Black wants the coin to come down heads, and that it is undetermined how the coin comes down. Black allows time to go forward; if the coin comes down tails, he rolls time back to t, well before the coin flip. He repeats until the coin comes down heads. He then rolls back time to t once more and allows events to proceed. The probability that the coin comes down tails is zero, yet Black is, in the actual sequence, as counterfactual an intervener as you like.

[Yentz Mahogany]

This is an interesting conversation betweeny Fischer and Vihvelin. (Also, a big thank you goes out to Neil for his thoughts.) However, assuming that I have properly understood these gifted philosophers, I have problems with both. I apologize for the length. In all likelihood, if I were a better writer, I would phrase my objections only in aphorisms. As it happens, I'll just summarize each point I want to make right now, and the reader can choose whether or not they want to slog through the rest.

1. Jones is necessarily responsible only when he a) can choose, b) is in control of his actions. If not, then bad consequences.
2. 'Possibility' requires interpretation.
3. Ability entails possibility of performance.
4. Fatalism is determinism.
5. Transitivity is just fine.

-=-=-

1. 'Freedom to do otherwise' has little to do with voluntariness and everything to do with liberty. Contra Vihvelin, I come to the table predisposed to believe that liberty (freedom-to-do-otherwise) is not a necessary condition for moral responsibility. It just seems wrong to say that a person, like Jones, is not responsible for the choices they make, regardless of the strange ambient circumstances which, unbeknownst to him, limit his freedom of action by limiting his options. In other words, you may not be free to do otherwise, and yet you (the agent) still may be accountable, depending on whether or not you made the choice to do the act. For if we -- the moral punishers with perfect knowledge of the agentic factors in this situation -- are to make judgments about accountability, then we have every lisence to hold Jones responsible for the act which he chooses to do, so long as he is not aware of Black. Else, in morally relevant situations, we would be letting a rule go unheeded, which can produce ill social consequences.

Rather, the 'control condition' appears to be necessary (to borrow a phrase from Julia Driver's paper; hopefully I'm using it properly). All other things equal, if Black compels Jones to do X, then Jones cannot be responsible; Jones has not chosen. If Jones does the act himself, (or at any rate, if his conscious mind fails to prevent his own creative striving,) then he himself is responsible; he has chosen. Yet this, unfortunately, seems to counter part of Fischer's point. For Jones, the agent, really does have the power to choose (contra p.10); it's just that he lacks the liberty to do otherwise.

2. There is a bit of ambiguity in modal terms which might be worth more emphasis than V. has given. Take:

(1) "The coin can't come up tails."

The modal word "can't" may be interpreted in two ways. By default, we infer that it is used relative to the general laws of physics of the universe and the generic state of affairs which make it possible for coin-flipping to be possible. Call this 'interpretation A'. (1) is obviously false under (A).

However, in another sense -- call it 'interpretation B' -- the word "can't" can be used relative to the state of affairs specified by the thought-experiment. Take it as a given that Black has magnificent powers, and assume that he, in at least this instance, has already used them to successfully predict the outcome. In this case, under (B), (1) is true: it is necessarily the case that Black was accurate in his predictions, and thus, that the coin must not come up tails, because we stipulated it. Stipulations have the force of necessity in a thought-experiment. So for the purposes of the case, and relative to a few facts, he really is infallible. [In this sense, if we feel fatalistic, it is because we have been awestruck by the monstrous powers that we've granted Black.]

One interpretation, (B), seems more like the Frankfurt story than the other, (A). However, it seems as though accepting (B) comes at the cost of meandering from Vihvelin's intended goals for presenting the thought-experiment in the first place: for she emphasizes that Jones can still perform, but (it so happens) will not. So perhaps (A) is most consistent with Fischer's comment that "the modal facts in Vihvelin's story and the Frankfurt examples are crucially different". I'm not sure I see that that's the case with (B), though.

I am not sure what to make of V.'s counterpoint in her original paper. She claims there is no equivocation (in her example, between the logically possible and the metaphysically possible). We are told that the coin toss is a "genuine chancy event" by stipulation. But "genuine chance" (whatever sense I can make of that) is incompatible with a deterministic universe. All chance is illusion. And this is not fatalism, unless by "fatalism" she means "determinism". In that case, by all means.

3. It seems to me that the "counterfactual intervener" (as well as the very strong form of the "conditional intervener", p.10 of V.'s reply) has the power to alienate Jones from the power to choose a particular act, so Jones is not responsible for any of the affected areas over which Black has dominion in those cases where Black intervenes. It's not just a case of lost opportunities; it's a case of meddling with agency. While the risk of a loss of the ability to express a desire as a choice is a sort of "lost opportunity", I guess, it *really is* also a limit upon the scope of Jones's agency and ability, and a transfer of agency to Black (at least with respect to certain specified goals, i.e., in riding a bike). (Although all the while, Jones retains responsibility for those actions he does perform and control, regardless of Black's machinations. The important point is that responsibility exists only if Jones is in control of his behavior.)

For an ability, to be an ability, really does demand the possibility that it may be performed. Loss of the possibility of performance due to psycho-cognitive factors is identical to loss of ability. For any given period of time during which there is a loss in the possibility of future performance, there is either the ignorance of the loss of ability (as in the Jones case), or the illusion of retained ability (as with amputees who feel phantom limbs and think they can still move their arm though it is not there). That which was once a case of "knowing-how" is sucked of all which makes it 'knowledge' or 'ability', and downgraded to the status of fiction. For the sake of that given time period, it is as plausible to say that Jones has the ability to ride a bike as it is to say that he has the ability to fly, just because he nightly dreams of flying. (Of course, this is surely a controversial claim, with competing intuitions which are themselves tempting; but there you have it.)

This is different from a weaker case of the conditional intervener, where Black simply limits Jones's opportunities by thwarting his access to the bicycle, but does not interfere with his agency. Let's return to the two interpretations of possibility, because they might shed some insight on how this might be. Take:

(A') Jones can ride his bicycle relative to the laws of the universe and the generic state of affairs (there is an Earth, humanity exists, etc).
(B') Jones cannot ride his bicycle relative to the specific state of affairs involving Black.

I would suggest that the difference is that Jones loses his ability according to B' but not according to A', and point out that, by default, words like "possibility" and "necessity" have a (probably superstitious) commitment to appealing to the outer reaches of what is metaphysically expectable. In which case A' is a more felicitous proposition than B', though neither are prohibited, and the matter is more verbal than substantial.

So what's the upshot? I don't know. The above remarks may add another dimension of analysis, but they don't seem to prove anything against the other two authors. Hopefully, though, it is not unwelcome exposition.

4. The word "fatalism" is used quite a bit by V. in chastizing the Frankfurtians in her reply to Fischer. However, my understanding of the term (albeit perhaps a misunderstanding) is that it is identical to deterministic-incompatibilism (perhaps with a bit of providence or God-like teleology thrown in). And surely, in this sense, one cannot accuse a Frankfurtian compatibilist of being a fatalist without misrepresenting them; unless one shows that the compatibilist stance collapses into incompatibilism, which I don't think V. has.

The entire point of determinism is that it allows us to go from the real to the necessary, although to be valid it makes its argument in more than one step. Perhaps it would be fruitful to hear V.'s thoughts on Andrew Bailey's paper and the cases presented therein.

5. I'm not sure I grasp the cogency of Lewis's argument against transitivity. In the case of Otto, Anna, and Waldo, the premise "If Anna had gone, then Waldo would have gone" is false, since there is a conceivable circumstance where Anna would go and Waldo would not. And, by contrast, if the premises were stipulated to be true, then our background knowledge about Waldo trying to avoid Otto would be false.

[Neil]

Upon sober reflection, I see that my last point is daffy. I can't make the strands internally consistent - not without some ad hoc assumptions. I withdraw it! Perhaps this is better: given the godlike ability to roll back time, Black can wake up the next morning and see if the coinflip went the way he wanted the evening before. If not, he can roll back time 24 hours; repeat as necessary. In that case, he can get the result he wants and not intervene at all in the (final) actual sequence.

My main point remains: that Vihvelin's dilemma (counterfactual intervention allows for alternative possibilities, but conditional intervention allows the agent to begin to act, a beginning sufficient to get them off the hook) fails if the conditional intervention has as its cue a mere flicker of freedom: not a robust alternative possibility, but a necessary (though not sufficient) condition of making the choice that Black wants to avoid.

[Justin Fisher]

I enjoyed reading the papers, but admit having trouble seeing a substantive conflict here.

The question at issue supposedly is this: "Does free action require that it have been possible that the actor act differently?" Of course, the answer will depend on what you mean by 'possibly'.

If you mean "possibly, consistent with all the laws of nature, and a full specification of antecedent conditions" then the incompatibilist will answer yes (free action does require alternative possibilities in this sense), while the compatibilist will answer no (free action doesn't require alternative possibilities in this sense). As far as I can tell, Kadri admits that Frankfurt-style cases weigh in favor of the compatibilist here, for these are cases where Black's nomological oomph removes the nomological possibility of alternative actions, without precluding that Jones's action be free.

Kadri's main point involves introducing what she thinks is a more interesting reading of 'possibly': "possibly, consistent with the set of skills/abilities that the actor currently has, but perhaps requiring that she have different desires than she in fact has, and perhaps requiring the absence of Black-like intrusions." I see how one could debate whether 'alternative possibilities' like these are an important prerequisite for 'free action'. But, as far as I can tell, that's not what John's doing.

Kadri also argues that Frankfurt-style cases don't do anything to show that free action doesn't require alternative possibilities in the 'abilities' sense just described. So, she concludes, Frankfurt-style cases aren't really relevant to assessing the question of whether free action requires such 'abilities to do otherwise'. But, as far as I can tell, John's hasn't given any argument that Frankfurt-style cases *would* be relevant to this question.

So, I give. What's the substantive dispute here? Nobody in the present debate disagrees about the relevance of Frankfurt-style cases to questions about *nomologically* possible alternatives; nobody disagrees about the irrelevance of such cases to questions about Kadri's 'abilities to do otherwise'; and nobody here is offering vocal arguments in favor of concentrating on one of these notions of 'alternative possibilities' rather than the other. So where's the conflict?

[Neil]

Justin, I think you're right there's something odd going on here. I take it you read (only) Kadri's reply? As I read her, she's not responding to John at all: in between 2000, when the paper John is replying to was published and today, she's changed her view about what the most important flaw with FSEs is. She now, as you say, thinks it turns on the question of what an agent's abilities are. In her 2000, the problem she focused on was instead a problem with the set-up of FSEs, which she expressed in the dilemma I replied to above. Her new claim is very interesting; it's defended here:

http://www.usc.edu/schools/college/philosophy/private/docs/vihvelin/FreeWillDemystified.doc

As for your claim that no one thinks this question is relevant to Frankfurt style cases, well that's false. Michael Smith, for one, has used this kind of argument to rebut FSEs. I work through this debate in my paper, coming next week on this site.

[Justin Fisher]

Thanks Neil. You're right that I didn't read Kadri's 2000 paper -- instead only John's paper here and Kadri's commentary on it. So I guess you're thinking that the substantive debate here involves whether Kadri's stronger anti-FSE rhetoric in the 2000 paper was really merited?

I'll still admit a bit of puzzlement. I took the essence of Kadri's *2000* argument against the conditional versions of FSEs to be that they didn't rule out the thought that an important sort of alternative possibility -- namely something like an ability to do otherwise -- might still be a prerequisite for free action. Am I right in thinking that that's how that part of the 2000 argument went?

If so, then that seems like about enough to make the debate have the overall structure I layed out above: there are two readings of 'alternative possibilities'; Kadri and John are focusing on different readings; and hence their disagreement about the relevance of FSE's is merely apparent, not substantive.

I'll look forward to your paper. Just to be clear, my 'irrelevance' claim was that FSE's aren't relevant to the question of whether Kadri-style 'abilities to do otherwise' are a pre-requisite for free action. (For, as Kadri's bicycle analogy makes pretty clear, the fact that Black is lurking seems to take away *opportunities*, not *abilities* -- hence, for all these cases show, abilities could be a prerequisite for free action afterall.) Is this really the claim that you and/or Michael Smith question?

[Neil]

Well, I think Kadri had better speak for herself! My reading is that the 2000 paper doesn't put forward the different account of abilities that she now wants to press. But I also don't think that she takes her current view to be weaker than the earlier one (in fact, I don't think she's withdrawn the earlier objection; she downplays it only because she doesn't think she needs it: the abilities claim can do all the work).

Michael's claim is that PAP is true: we test the claim that an agent could do otherwise by asking whether, in a near-by possible world the agent would do otherwise. We hold fixed the agent's intrinsic properties to ask this question, but not the presence of the counterfactual intervener. As to what my view is: well, you'll just have to wait another few hours to find out (I'm too tired from staying up way past my bedtime to watch the FA cup final to try to sketch it).

[Paul Torek]

Justin, you stole my thunder! I just want to add, though, that in my opinion this exchange teaches us that we *ought* to be offering vocal arguments in favor of concentrating on one of these notions of 'alternative possibilities' rather than the other. At least, we ought to if we care about whether we should say that such possibilities are compatible with determinism.

[Kadri Vihvelin]

Justin complains that he finds it hard to see any substantive disagreement. His reasoning appears to be as follows: Fischer and I agree that moral responsibility is compatible with determinism and thus agree that the nomological impossibility (in Justin’s sense) of doing otherwise does not preclude an agent’s responsibility for what she does. I say that there is a sense – the ability sense -- in which it is possible for a deterministic agent to do otherwise, even in Frankfurt stories, but Fischer never denies this. So our disagreement is merely verbal.

Justin misunderstands. First, my disagreement with Fischer is not about the thesis that moral responsibility is compatible with determinism; it is about whether “Frankfurt-style examples” (FSE’s) play any part in a good argument for that conclusion. Fischer thinks they do; I disagree. Second, it is a mistake to think of the literature about FSE’s as turning on a debate about what we mean, or should mean, when we say things like ‘he could have done otherwise’ or ‘he had no alternative’. The debate that Frankfurt sought to displace was about this question, and compatibilists were losing that debate. (See Berofsky, “Ifs, Cans, and Free Will: The Issues”, the Oxford Handbook of Free Will, 2002, for a helpful summary of the literature concerning the so-called “Conditional Analysis” of “could have done otherwise”.) Frankfurt’s aim was to undercut this traditional debate by showing that it rested on the false assumption that he called “The Principle of Alternate Possibilities” (PAP): a person is morally responsible for what he has done only if he could have done otherwise. His schematic story about Jones and Black was supposed to show that, regardless of how you understand ‘could have done otherwise’, you should agree that the Principle of Alternate Possibilities is false.

Nearly forty years later, the debate about whether FSE’s succeed in refuting PAP continues. Is it a substantive debate? The defenders of FSE’s certainly think so; they don’t think that they are simply stipulating a sense of ‘can’ and showing that in this sense Jones cannot do otherwise but is still responsible. They think that Frankfurt taught us something important about moral responsibility: that we may be responsible even if we cannot do otherwise in any interesting or “robust” sense. (I assume this includes the ability to decide otherwise.) And the debate about FSE’s is not a debate in which compatibilists line up neatly against incompatibilists; there are compatibilists who deny that FSE’s succeed in refuting PAP and there are incompatibilists who defend the claim that FSE’s show that PAP is false. So if the debate fails to be about anything substantive, it is not for the reasons Justin gives.

[Kadri Vihvelin]

Neil makes the following suggestion: If we put Black on the scene with the power to intervene before Jones even begins to act, then we have a Black who succeeds in making it true that Jones cannot even begin to do the forbidden act. The only alternative possibility left to Jones is something that is a mere ‘flicker’ (the earlier sign that is a reliable indicator of the forbidden action).

This was also Fischer’s response, and I thought I answered it, but let me try again.

Suppose the forbidden act is Jones raising his hand at a meeting and suppose an indeterministic Jones who always blushes before raising his hand even though the blushing neither causes nor is nomologically sufficient for the hand-raising. Black doesn’t want Jones to raise his hand, and plans to stop him as soon as he sees the blush. Jones doesn’t blush.

Could Jones have raised his hand? No, says Pereboom about a similar but more complicated case (Tax Evasion):

“in order to decide otherwise, a moral reason would have had to occur with the requisite force, and then the device would have been activated”. (p.22)

In my simpler, cruder case:

“In order to decide to raise his hand, Jones would have to have blushed, and then Black would have stopped him from even beginning to decide to raise his hand.”

This is a back-tracking argument, of the kind that I criticised in my response to Fischer. Here the back-tracking is a matter of seconds rather than hours, but the point remains the same. Given the way Jones is, if he were about to decide to raise his hand, he would have blushed first. And given the way Black is, if Jones had blushed, Black would have stopped Jones from even beginning to decide to raise his hand. But it doesn’t follow that if Jones had been about to decide to raise his hand, Black would have robbed Jones’ mental state of its usual causal efficacy. For as a matter of fact, Jones didn’t blush.

[Justin Fisher]

Kadri, thanks for replying to my comments. I may be dense, but this still looks to me like a merely terminological debate. Here's the playround caricature version:

Everyone before Frankfurt: "Free action seems to require alternative possibilities. But determinism rules out alternative possibilities. Long live the PAP! Death to compatibilism."

Frankfurt: "But there are clear cases of free action that seem not to involve the sorts of alternative possibilities that determinism threatens. Death to the PAP!"

Compatibilists: "Ding! Dong! The PAP is dead! The wicked PAP is dead! Long live compatibilism!"

Kadri: "But wait, everyone! There's another notion of alternative possibilities: 'abilities to do otherwise'. Frankfurt-style cases don't do anything to show that you can act freely without having alternative possibilities in this new sense. Forget Frankfurt and the old PAP. Long live the new compatibilist-friendly PAP!"

John: "Ah, but Frankfurt still deserves credit for saving compatibilism from the wicked old PAP."

Justin: "You two aren't really disagreeing. Why not concentrate on the merits of this new PAP?"

Of course this is just a caricature, but I'm afraid I still don't see how it significantly misrepresents the structure of the debate.

[Jussi Suikkanen]

Can't resist the temptation - about the everyone before Frankfurt bit:

'For in order to be free, there is no need for me to be capable of going in each of two directions; on the contrary, the more I incline to one direction... the freer is my choice.' (Descartes Meditations, 57-58).

[Neil]

Kadri,

My problem with that response, in the dialectic context, is that it's a little unfair: John was replying to your 2000, dilemma, whereas the reply you made turns on the on the bundle of dispositions view. As you'll see, if you read my paper, I'm somewhat sympathetic to the substance of your reply (though I claim that the question whether free will, understood in something like the manner you understand it, is compatible with determinism is not obvious; that's where you and I part). My response is, as I think, good against your 2000 view. I would need (and I think, have) quite different arguments against the bundle of dispositions view. But since that view is snuck into the exchange with proper defence, it seems inappropriate for me to reply to it here.

[Neil]

oops, that should be "without proper defence" (which is not to say you don't have an argument for the view, but the argument is a whole other paper).

[Anibal]

Perhaps the well known PAP and the long philosophical debate that strive us is just an outcome of the cognitive, reflexive part of our minds forming with the other most emotive and automatic part the dual system of mind view, actually defend by psychologists (e.g. Kahneman),philosophers of neuroscience ( e.g. Greene), experimental philosophers ( e. g. Knobe) and others. And because each of us internalize, prioritize, rationalize... the things according with life experiences, capacities, expetancies, academic orientation and the like... (that´s not mean relativism only superficial relativism because underlying each of us exist a universal mind grammar of heuristics and judgments for inmediate behaviour and brain circuitry responding to demands of ecology including doing philsophy), some believe in alternatives and different paths of actions to speak about metaphysical freedom and atribution of responsibility. But following the science of evitability (Dennett 2003 p. 59) what you choose is the "real outcome" and this happen in a determined world or in a undetermined world, there is no necessity for constantly presenting alternatives (that if real maybe inmobilize us), just only what we choose, but neverthless what we decide freely to choose even if we don´t have open-ended alternatives.

[Paul Torek]

But Justin, to play devil's advocate for a moment, what is the alleged difference between the "wicked old" PAP and the "new" PAP? Isn't it just the difference between understanding "abilities to do otherwise" as implying - here I'm not sure what, on the "wicked old" view - versus something like "a bundle of dispositions" (Neil's term) on the new. Merely verbal? I suspect we're in the region where immediate classification into "word" vs "world" categories is a bad idea.

I also can't get over the sneaking suspicion that backtracking and Transfer principles (akin to Transfer of Non-Responsibility) are at work here. To the extent that one finds such things tempting, I suspect, one will tend to think that Black really does deprive Jones of alternatives.

[Jeffery Coleman]

I hope this isn't too off-topic, but aren't Lewis's and Stalnaker's counterexamples to transitivity, well, wrong? In the case of Otto, statement 2(where 1,2,3 are:

1)If Otto had gone to the party, then Anna would have gone.
2)If Anna had gone, then Waldo would have gone.
3)Therefore: If Otto had gone, then Waldo would have gone.) seems terribly wrong given that Waldo never risks encountering Otto. Shouldn't it read "If Anna had gone and Otto had not gone, then Waldo would have gone."? Likewise, in Stalnaker's counterexample(

1)If J. Edgar Hoover had been born a Russian, then he would have been a communist.
2)If he had been a Communist, he would have been a traitor.
3)Therefore: If he had been born a Russian, he would have been a traitor.) shouldn't statement 2 read "If he had been a communist and not a russian, he would have been a traitor," or "If he had been a communist and been born an american, he would have been a traitor"?

To treat these examples otherwise seems to miss the fact that they are counterfactual. Did I misunderstand something?

[Yentz Mahogany]

Jeffrey,

Yeah, I'm puzzled as well.

What might be replied is that the point of the examples is not to show that any of the premises are incomplete, as you did with your appending. Rather, it was to try and show how true premises might lead to a false conclusion in some instance. So telling us about what premise two *ought* to be is no help in explaining the puzzle.

But that doesn't change the fact that premise two is, in some plausible conceivable circumstance, false. Since an "if-then" is a kind of guarantee against some plausible conceivable circumstance turning up false, this either makes the counterfactual false (and the syllogism invalid and useless), or guaranteeing the conclusion (thus defeating the point).

Quite the puzzle indeed.

[Jeffery Coleman]

Right, but what I mean is, isn't premise 2 in both cases false? In the first case, the scenario where Otto shows up is a counterexample to the truth of the premise, and in the second case the fact that some communists who are russians are not traitors is again a counterexample(given that we don't specify J.E.'s nationality), otherwise we are equivocating between J.-Edgar-if-born-American and J.-Edgar-if-born-Russian(we could also say that the first example is in danger of equivocating between Anna-if-she-went-alone and Anna-if-she-went-with-Otto).

I guess I would suggest that the initial plausibility of the argument lies in our general tendency to equivocate over a range of fundamentally different possibilities with names(such as Jeff-of-five-years-ago and Jeff-of-tommorrow). However, general human behaviour is typically not borne of solid logic. Induction is everywhere!

[Yentz Mahogany]

Jeffrey,

As far as I can tell, then, our confusion is based on the same reasons. I can't say much beyond that.