[ExI] Universal timeless principles

[Anders Sandberg]

On 2015-10-05 20:02, Dan TheBookMan wrote:
> On Oct 5, 2558 BE, at 9:09 AM, William Flynn Wallace 
> <foozler83 at gmail.com <mailto:foozler83 at gmail.com>> wrote:
>> Philosophers could not even agree on the value of philosophy, I think.
>
> That's the argument from disagreement. If you take the argument from 
> disagreement seriously, then shouldn't you apply it to all endeavors? 
> Do scientists all agree on X? No, well, then X should be chucked out. 
> X might be extended to all of science. In any field -- science, math, 
> history, music, what and whether to have for lunch today -- you'll 
> find a dissenter.
>
> (To pre-empt the argument that science or other fields all have means 
> of settling such disagreements: No, there are dissenting voices there 
> too. Not all scientists agree on how to settle science questions. Not 
> all historians agree... And so forth.)

Disagreement does show that one should not be overconfident about a 
conclusion, though. As I argued in my uploading ethics papers, the fact 
that there is disagreement between reasonable people on whether software 
can be conscious should be enough to make even consciousness-sceptics 
agree that there is some chance that they are wrong, and hence we better 
treat potentially sentient software nicely (the opposite case, treating 
software nicely when it is actually non-sentient, is not as bad as 
treating sentient software badly). This approach works when some 
possibilities have a big and different weights, but there are still 
nontrivial issues of what counts as reasonable disagreement.

Philosophy might be the field that criticises itself most vigorously.

>> And just why do they know any more about anything than the rest of us 
>> do?
>
> I wouldn't presume they know more, though one can't be sure until one 
> familiarizes oneself with their work.

Exactly. People who discount philosophy frequently reinvent 
philosophical wheels that have been in use for millennia.

Philosophers at least know what ideas have been tried before and how 
they worked out. Many are also very good at critical reasoning about 
abstract areas: when they make claims, they typically have documented 
reasons for these claims.

We know a fair bit about what tasks humans can become expert in (see 
http://www.researchgate.net/profile/James_Shanteau/publication/4815960_Competence_in_experts_The_role_of_task_characteristics/links/00b495373e3fe7630f000000.pdf 
). Philosophy does pretty well on decomposing problems and using formal 
decision aids. Depending on branch, one may look at static things with 
agreed on properties (good for expertise building) or messy human 
domains where there is little agreement on what is what (bad). Some 
philosophers keep on applying the same method to a lot of cases (good 
for becoming expert at the method), others jump between methods (bad for 
method expertise). Feedback is typically in the form of other 
philosophers criticising - useful for becoming an expert on making good 
formal arguments, but real-world feedback is clearly worth far more and 
less common. All in all, we should expect philosophers to be really good 
at doing academic philosophy, pretty good at analysing problems, decent 
in talking about real-world situations where we get feedback on what 
works and doesn't work, and hopeless in areas where there is no simple 
feedback. This is why I prefer practical ethics and epistemology (real 
world data and consequences!) to metaethics and metaphysics.


-- 
Anders Sandberg
Future of Humanity Institute
Oxford Martin School
Oxford University

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20151006/423ccfed/attachment.html>