Re: Solidity is degree of justification (was: Mirror neurons at autopsy)

On 28/12/2012, at 5:38, "Elliot Temple" <[email protected]> wrote:
> On Thu, Dec 27, 2012 at 4:04 AM, David Deutsch <david.deutsch-LR/[email protected]> wrote:
> > On 25 Dec 2012, at 23:58, Elliot Temple <[email protected]> wrote:
> >
> >> On Sun, Dec 23, 2012 at 11:13 AM, David Deutsch <david.deutsch-LR/[email protected]> wrote:
> >>
> >>> currently called 'justificationism', namely that knowledge is that which can be justified as true, or likely to be true, etc.
> >>
> >> That is not what justificationism means. Justificationism is the
> >> concept that ideas have an amount/degree/weight of justification
> >> (epistemic status, aka confirmation, support, authority, probability,
> >> etc), by which we should judge them.
> >
> > for truth
> 
> For epistemology.
> 
> >> Ideas with more
> >> justification/weight/evidence/etc are deemed better than ideas with less.
> >
> > Deemed to be truer, more likely to be true, more likely to contain more truth, etc. That is the mistake.
> >
> > But that does not mean that ideas cannot have any non-binary attributes. For example they can be longer or shorter to express, depend more or less on mathematics, be more or less relevant to a given problem, be more or less original, more or less useful -- and (as I have said) easier or harder to vary as explanations (relative to a given state of background knowledge).
> >
> 
> All those other non-epistemological attributes matter when there is an
> explanation of why they matter. This resolves the issue of *how* they
> matter: they matter in the way that explanation says (not by the
> generic method of increasing or decreasing any sort of epistemic
> status). Sometimes uncontroversial background knowledge covers such
> explaining for us, but sometimes it does not.
> 
> > easier or harder to vary as explanations (relative to a given state of background knowledge)
> 
> This is ambiguous. Varying things is easy without constraints, e.g.
> anything can be varied to the word "cow". The constraints intended are
> unstated. Varied under what constraints?
> 
The "as explanations" is the constraint, isn't it? If you varied an explanation to the word "cow", it can no longer perform its task as an explanation. So, the constraint is that the explanation must still do its job of explaining the thing in question.

An explanation is harder or easier to vary than another explanation of the same phenomena only if it is still able to do the job of actually explaining that phenomena. Otherwise it is not explaining that same phenomena and so cannot be regarded as being harder or easier to vary and so not preferable or not, by that measure alone (it might be preferable by those other things listed when David wrote:

> ... they can be longer or shorter to express, depend more or less on mathematics, be more or less relevant to a given problem, be more or less original, more or less useful...

When judging which of rival explanations is preferable, the harder to vary criterion can only work if the explanations are able to do the same task of explaining whatever it is and the only difference is that one is easier to vary. But does this situation crop up often in science? Usually we have just one explanation, not many rivals that are all roughly equal in their ability to explain whatever it is.

There might be many reasons why we would prefer the axis-tilt theory of seasons to the Greek Persephone myth of seasons. There is not just one. But one important difference between the two theories and one important reason to prefer the former is that the Greek Persephone season theory is so easy to vary. You can change the gods, you can change the reasons for periodic visits...one is tempted to say that everything about the story (almost) is arbitrary because gods can be assigned any old magical powers you like and which would do the job for you. The axis tilt theory is not like this. It's hard to vary in comparison. For example you cannot swap the sun for the moon in the theory because the moon can't do the job of the sun. Right?

Brett.




[Non-text portions of this message have been removed]