Absence of evidence IS evidence of absence

quote:

---

You can use an absence of evidence as evidence of absence when you're pretty damn sure you'd know where the evidence would be, if it existed at all

---

That's not absence of evidence, it IS evidence.If you KNOW that for A to be true you MUST find B under
condition X, and you don't that is evidence against A.That's not the same thing at all.If for A to be true B must exist somewhere, under some
unspecified condition, and you do not find B, it tells
you NOTHING (except to look elsewhere).

If I search a house and don't find anyone inside,
can I conclude that the house is empty?

That does seem to be the line of reasoning ... but
it doesn't make absence of evidence evidence of absent,
it just changes the rules.If you are looking for something (A) in a well-defined search space
and you know that you have covered 100% of that space without
finding (A) then the absence of (A) has been shown.If you have not covered 100% of the search space, you have not
shown the absence of (A).You may, as your coverage increases, become convinced that you
will never find (A) ... but the accumulation of failed searches
does not amount to evidence that (A) is not present (nor that
it IS present).I think what is beign suggested is that the longer something
remains absent of evidence, then that builds a case for
the absence.This is not true. One can never rule out that one is
looking in the wrong place or for the wrong thing except in the
extreme case where the issue is so well defined that the
entire search space can be investigated with 100% coverage and
100% certainty of success were IT there.E.g. the football field -- we cannot say the field is absent of
a football until we have searched every football sized space
on the field.Things are always in the last plave you look for them

You are quite right that (A) has to be well defined too.And yes, the absence of evidence is, in that extreme case,
evidence of absence ... but only because the nature of the
search is so well defined and bounded.Until you reach 100% coverage of the search space you have
no evidence at all.You could say that (A) is becoming unlikely (that probably equates
to Absence of evidence not being proof of absence), but as soon
as you relax the definition of (A) and/or the search space
all bets are off.If you don't know how much of the search space you have covered,
or you don't sufficiently understand the nature of (A) then no
amount of absence of evidence is indicative of actual absence.

Yes.It's about the relationship of the evidence to the object being
sought.