From: Lee Corbin (lcorbin@tsoft.com)
Date: Wed Oct 30 2002 - 23:39:54 MST
Hal writes
> In terms of civil laws, if you take on an obligation and make a copy,
> is it bound by the obligation? If not, then you could escape your
> obligation by making a copy and then killing yourself. If so, then
> your copy would have its freedom of action circumscribed; it could
> not go off to Mars, because the original might at any time suffer a
> mishap and then the copy would be obligated to perform in your place.
> Neither of these is desirable.
Yes, that's quite clear. As you suggest, it's merely a question
of extending the range of contracts. There are a few harder things
about contracts covering duplicates, but I don't think that they're
insurmountable.
> Given the variety of situations which can occur, ideally the issue can
> be resolved on a case by case basis, at the time the contract is made.
> The minimum requirement has to be that it is possible for the original
> to *voluntarily* create a copy who is bound by some obligation. Just as
> a person can bind himself with a contract, he has to be able to bind his
> copy at the instant of its birth. This allows both solutions above to
> be explored and used, as circumstances dictate.
>
> If we don't allow for this kind of copy creation, it will be too easy
> for people to escape obligations, and either copy creation will be
> illegal, or else no one will be able to bind themselves with contracts.
> I think allowing the creation of obligated copies is the lesser of all
> these evils.
Of course. It also has a great deal to do with the memories
of entities at various moments, and even more to do with
intentions. But what puzzles me is that the clearest and
best laws always steer clear of intentions.
> But at [an] extreme, if someone makes a copy, then one of them
> decides to commit a crime, but the decision was made after the copy,
> then it seems wrong to hold the innocent one responsible for the crime.
That does seem correct, but determining the exact instant in which
such a decision is made is probably at least NP-complete ;-)
Suppose I've been carrying a grudge against Mr. X, and I know
that if I go into a certain bar, then perhaps I'll lose my
temper, get into a fight with X, and kill him. Knowing that
this might occur, I might make a copy for a number of reasons,
(this being one I'm only partly conscious of). Sure enough,
some one of me wanders into that bar sooner or later and kills
X. But that's not so bad; I live on in my duplicates.
Worse, I don't even know X, but just happen to have made a large
number of duplicates for a big programming project. Any one of
them may realize that forfeiting his life isn't such a big deal,
and that the project will continue (perhaps new duplicates will
be made) anyway. It may develop that all close duplicates must
pay for a crime, otherwise, a quick suicide of the guilty xox
is too easy an out.
> Given these considerations, here is what I see as the pragmatic and
> practical approach. Let's take the easiest case first: someone commits a
> crime; shortly afterwards, they make a copy; and shortly after that they
> are caught (both them and the copy, or only the survivor if one of them
> is dead). In that case I think both should be punished. Each arguably
> shares culpability for the crime; each remembers committing it.
Whether they remember it or not doesn't necessarily matter. In
the future one may have enough control over one's memories that
one can be restored to prior state, and it will turn out (I have
predicted) that to punish someone in this state is exactly
physically identical to punishing a duplicate that was in
no way involved.
We therefore have yet another reason to hold accountable all
close duplicates.
> In practice, if we see a
> lot of cases where criminals "just happen" to create copies shortly
> before committing their crimes, juries will tend to be more skeptical
> of whether the copy was innocent. There will be a certain amount of
> feedback in the system if it starts to be abused.
>
> We might draw an analogy to the crime of conspiracy, where it takes a
> plan to commit a crime, plus a single overt act in furtherance of the
> plan, to constitute conspiracy. The copy could be said to be guilty
> of conspiracy, if at least a single overt act in furtherance of the
> planned crime occurred before the copy was made.
Yes. This is a corrective to some other posters who wondered
what would happen if two duplicates were in the same room when
a crime was committed. We already have this case: when two
ordinary people, usually friends, are together when a crime is
committed, then it doesn't work for each to plead that he cannot
personally be found guilty beyond a reasonable doubt. If it
did, then pairs of people would commit serious crimes all the
time with impunity.
> Summing up, for voluntarily relationships we should encourage voluntary
> solutions, where people simply state what they will do in advance, when
> the contract is made. This does require having the option of binding
> a copy to follow a contracted agreement as a condition of the copy's
> creation.
Quite right.
> For criminal actions, we should punish copies if they have
> a past history (including the history of the original before the copy
> was made) of committing the crime, or of taking some explicitly step in
> furtherance of a plan to commit the crime. This is necessary because
> otherwise many people will see the creation of copies as a loophole that
> allows them to commit crimes without consequences, or with a lessened
> chance of being punished.
Yes. But I predict that it will come to be necessary to deter
all close duplicates (like a single person is deterred) because
intent and motivation are too hard to determine. Yes, this is
a little like punishing a whole village for the misdeeds of
single partisans, but I doubt that anything else will work.
Lee
This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 09:17:54 MST