From: Anders Sandberg (asa@nada.kth.se)
Date: Thu Feb 21 2002 - 06:14:03 MST
On Thu, Feb 21, 2002 at 11:06:48AM +1030, Emlyn O'regan wrote:
> Anders wrote:
> >
> > Overall, an enjoyable little website. There should really be
> > more sites
> > like this for other questions (right now I'm reading a book about the
> > animal rights movement, and it has some interesting examples of how
> > tricky the issue of deciding what beings have rights and on what basis
> > is; it would probably make a good "battleground ethics" setup).
>
> How bizzarre... I am reading "Animal Liberation" by Peter Singer at the
> moment. Same issues. Same book?
Hardly. Per Svensson makes mincemeat out of Singer, and shows some of
the less palatable sides of his and other animal rights philosopher's
ideas. The problem isn't that they are deliberately anti-humanistic -
very few consider themselves that - but that their ideas do lead in that
direction.
Especially Singer's utilitarianism has some rather awful implications.
It is hard to avoid with utilitarianism since it does not place much
value in individual lives but rather on the aggregate level. And of
course, who gets to decide what increases utility the most?
Utilitarianism has a very strong centralist bias. While Singer is very
careful to point out possible slippery slopes when it comes to eating
crayfish (which *might* be able to experience pain, and if we start to
allow ourselves to eat them, we might start eating fish, and then meat),
he handwaves away slippery slopes when it comes to his proposals for
eutanizing malformed children (apparently there is no risk that this
practice will be extended to other undesirable humans, in his view).
Slippery slope arguments are not per se valid, but it is interesting to
see here that Singer allows himself quite a bit of freedom in
interpreting what is risky and what is not - would the same kind of
freedom to decide (with little or no ethical reasoning) apply to the
definers of the common ethics advocaded by many utilitiarians?
> I think that the ethical questions raised have are very strong in the
> context of transhumanism, moreso than other philosophies, in that we support
> morphological freedom, and thus must regard where we are now (human) as t(i)
> (where i is some positive integer) in an iterative process. This implies far
> less weight given to current qualities of humanity in determining who is
> covered by concepts such as the Principle of Equality... any criteria based
> on current human abilities must look entirely arbitrary. Thus with respect
> to beings with below human levels of intelligence, which are otherwise
> rather similar to us (eg: mammals), it is difficult to see where lines could
> rightly be drawn to grant us rights but deny the same to them. After all, in
> a context where we are talking about becoming SIs, etc, the difference
> between a human and, say, a dog, is negligable.
I agree this kind of issues is very important to us, and we need to
think through them carefully.
The fact that there is no clear biological dividing line between us and
animals has been used by Singer et al to argue that hence there is no
clear ethical dividing line. This does not follow - it presupposes that
ethical dividing lines must be based on biological dividing lines. In
fact, if one looks carefully at this kind of argument the problem
appears that since there is no clear dividing line between animals and
plants, plants should also be regarded as belong to our ethical
category.
I think Dawkins is right: the important thing isn't belonging to a
certain species, but other inherent properties. When I use the word
human, I usually refer to a being endowed with certain abilities such as
capacity for rational thought, freedom of action, the ability to change
behavior depending on learning, deductions and information received from
other beings and so on. Currently the only example we know of fulfilling
these properties are members of homo sapiens, and hence the usage isn't
erroneous, but we are aware of the possibility of encountering or
creating other beings with the same properties. Hence these aliens,
uplifted animals or AIs would also be "human" (at this point it would
make much sense to create alternative terms). A lot of philosophy and
humanism can be placed into this more generalized human framework with
little or no change.
Many of the human properties brought up by Singer et al as examples of
that we are not unique - language, use of tools, deduction etc, which
are found elsewhere in the animal kingdom - are actually not significant
from an ethical point of view. Sure, we wouldn't manage without language
or tools, but having them doesn't give us any special ethical status. On
the other hand, being able to largely change the way we act and live
depending on ethical ideas, that is a very significant property with
ethical implications. The list above of properties are essential for
being an ethical subject: an irrational being cannot act ethically, nor
can a being that cannot break its behavior patterns or learn. While
there are other species than homo sapiens that show these properties in
part, there is a significant difference in ability between humans and
even the brightest apes. Graded rights make sense, both within the human
category and outside it - as Per Svensson pointed out, acknowledging
humans as having a special ethical position that animals lack doesn't
imply that animals should be treated arbitrarily; you can be a humanist
without being cruel.
I would sum up by saying that from the perspective of an SI, the
difference between a man and a dog is not negligble. A dog cannot change
itself very far, while humans are in a sense a general form of being
that can at least aspire to vastly different behaviors and lives. When
it comes to being autotelic, the human has more in common with the SI
than the dog.
-- ----------------------------------------------------------------------- Anders Sandberg Towards Ascension! asa@nada.kth.se http://www.nada.kth.se/~asa/ GCS/M/S/O d++ -p+ c++++ !l u+ e++ m++ s+/+ n--- h+/* f+ g+ w++ t+ r+ !y
This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 13:37:40 MST