>
>> The claim that the absence of evidence is in fact evidence of absence is
in
>> fact a logical fallacy called the appeal from ignorance (a particular
>> proposition is false because it has not been proved true). To claim, or
>> even suggest, that the Cambrian lifeforms could not have evolved because
no
>> has yet found fossilized ancestors is to engage in this fallacy.
>
>While I agree that we have to be very careful about discounting a
>theory, just because there is a portion of evidence not yet found, I
>don't think I can agree with your logical falacy point.
>
>Absence of evidence may or _may_not_ be evidence of absence.
>
That's just the point though. In science absence of evidence can never be
evidence of absence because in science concepts are verified or refuted by
the presence of evidence, not by a lack of evidence. Even when you prove a
concept is wrong, you do so by presenting evidence that demonstrates the
concept cannot be right, not by using a lack of evidence in favor of the
concept. Remember how the fallacy goes: a concept is false because it has
not been proven true. In arguing that a concept is false because no one has
been able to find the evidence to prove it true, you (editorial not
personal) are simply arguing a different version of the same fallacy.
>
>Suppose
>you have a million holes in the ground, and you want to see if at
>least one of them contains a marble. You examine 200,000 holes at
>random and find no marble. This is good evidence that, if there are
>any marbles at all, their number is quite small.
>
Assuming that the marbles were themselves distributed randomly, which you
did not specify. Otherwise there could be 500,000 of them, but placed in
specific arrangements meant to defeat a random search. As such, your lack
of evidence at this point really tells you nothing about how many there are.
If, however, you can propose good theoretical reasons for believing that the
marble distribution is random, or for believing that it is not random, then
you could make a prediction about the expected number of marbles to be
found.
>
>It is not however
>enough evidence to conclude that there are no marbles at all. If
>however you examined 950,000 of the holes and still there were no
>marbles found, it would be strong evidence that are no marbles.
>
Not really. Your hypothesis is that there is at least one marble present,
but nothing is said about the probability of there being more than one. As
such, a proper hypothesis would be that the one million holes contain one
marble. That makes the null hypothesis that there are no marbles. As such,
to verify or refute either hypothesis you would have to examine every single
hole; even 950,000 holes would be insufficient to verify or refute either
hypothesis. Even 999,999 holes would be insufficient because the last hole
still has a 50/50 chance of containing a marble.
If, however, your hypothesis was that 100,000 marbles would be hidden in a
random distribution of holes, then your null hypothesis would be that there
are not 100,000 marbles randomly distributed among the holes. As such, if
after randomly searching 200,000 holes you find no marbles, then you have
good evidence that the null hypothesis is correct, because statistically
speaking you should have found 20,000 marbles by now. Of course, at this
point you cannot say why the null hypothesis is correct: there could be
fewer than 100,000 marbles or they may not be randomly distributed. Only a
complete search of all the holes, plus a plot of the distribution of any
marbles found will be able to answer that question. But as far as your
original hypothesis is concerned, a search of 200,000 holes would be enough
to verify or refute it. Despite appearances, however, this is not a case
where a lack of evidence verifies one hypothesis and refutes another,
because in this kind of study the lack of marbles is not a lack of evidence;
you would expect to find some holes empty in any event. Rather the evidence
you did find did not match the prediction made by your hypothesis. A lack
of holes would constitute a lack of evidence, because your hypothesis
assumes a million holes would be present, but in that case the lack of holes
would be unable to verify or refute either your hypothesis or your null
hypothesis. And if someone tried to claim that a lack of holes proved that
there were not 100,000 marbles randomly distributed among the holes, that
person would be engaging in the fallacy of the appeal from ignorance.
>
>My
>point is that it is possible, at least in principle, to use absence of
>evidence as evidence for absence.
>
Not in principle, since the very nature of science is opposed to it. In
practise it often happens that scientists do jump to that kind of
conclusion; I've done it myself, but that does not make it right.
>
>I don't believe the "lack of pre-Cambrain fossils argument" correctly
>matches the "appeal from ignorance" logical falacy. If no one had
>bothered to look for fossilized Cambrian ancestors, and then somebody
>claimed that because there were no such specimins they hadn't evolved,
>_then_ this would be an appeal from ignorance falacy. But the fact
>that people have been looking without success, is admissible evidence.
>
The fallacy doesn't depend upon whether you look for the evidence or not; it
only depends upon whether you use a lack of evidence to try to prove or
disprove a concept. The fallacy doesn't say, claiming a concept is false
because no one has tried to prove it true; it says, claiming a concept is
false because it hasn't been proven true. The implication is that people
could have been trying to prove it true but had not yet succeeded. The
ignorance the fallacy refers to is not a lack of knowledge of whether the
evidence exists, it is the assumption that because we don't have, or haven't
yet found, a certain kind of evidence that it does not exist at all.
And all the inability to find such evidence tells us is that we have been
unlucky, unless someone can provide good theoretical reasons why the
evidence does not exist.
>
>Don't get me wrong, it may well be that this evidence is ineffective
>in discounting evolution. Perhaps ancestors were soft bodied, making
>fossilization near impossible --- for example. But my point is that
>the reason for the ineffectiveness has nothing to do with "appeal from
>ignorance" logical falacy.
>
I hope I have been able to explain that in fact it does.
Kevin L. O'Brien