That is presuming nothing. That is an entirely different matter than presuming innocence. As I read it, the comment you initially replied to basically said (in a manner not entirely clear or sympathetic) "you are assuming facts not in evidence."
I was not advocating presuming guilt in any context - that would be even more nonsensical.
If you believe that people are born in a state of innocence, meaning they have not committed a crime, as I do, then you believe they are innocent until they have been shown to be guilty.
In this case a presumption of innocence means "things are the same as I understand them to be based on previous evidence until I am shown that they are now different." You are not born guilty so you must remain in a state of innocence until you commit a crime. And because that crime must be proven you are innocent until proven guilty.
In this case neutrality is the same thing as believing in someone's innocence.
This is fundamentally different than, for example, the distinction between a lack of a belief of God and a belief that there is no God - because we know something about people, all people, from experience, whereas we know nothing about God.
If you interpret "presumption of innocence" as purely a check against the base rate fallacy, then sure - it applies anywhere humans are doing reasoning.
I think that it serves other important roles in regard to punishment in particular, and those other roles are harmful in situations where you are seeking truth as opposed to asking whether to punish.
In particular, consider the obviously fallaceous:
"Should we put more resources toward investigating rape allegations? Well, in each of those cases where we have not put enough resources to determine the veracity of the allegations, we need to assume the innocence of the accused. By that reasoning, we can conclude that 100% of uninvestigated allegations are false, so we clearly should not spend any more resources on them."
Knowing that some uninvestigated allegations must certainly be true, based on probability alone, I cannot tell you from that information which of the allegations are true, and which are not true. So from an epistemic standpoint I am no better off than I am if the allegation were not made at all.
Lets say we know from experience that about 10% of the prevously uninvestigated cases end up proving guilt. Does that mean the allegation itself means the person being accused has a 10% increased probability of being guilty, merely because they were accused? In other words, does merely accusing them make them less innocent? I would have to say no - their objective innocence or guilt has nothing to do with whether or not an allegation was made.
You said: "those other roles are harmful in situations where you are seeking truth as opposed to asking whether to punish."
To use your example about investigating rape allegations, the presumption of innocence doesn't stop the legal system from investigating (possibly true) allegations to begin with, so why would it stop our political system from being open to the possibility of devoting more resources to investigating rape allegations, or journalists from pursuing a story to determine the truth of what happened?
In other words, the possible truth of an allegation shouldn't nullify the presumption of innocence because possible truth is not evidence. We are in no better position now that this particular allegation has been made than we were before it ws made - we still have no idea what happened. And that's true whether we're a prosecutor or just a regular person wanting to know what happened.
In fact I'd say in all cases the presumption of innocence is not harmful to seeking truth and in fact enhances it as it prompts the production of further evidence.
Assuming anything less than innocence seems more harmful to seeking truth as it allows instances like the Duke Lacrosse case or (perhaps) this recent Michael Arrington ordeal to take place, which causes more skepticism when future victims make a legitimate accusation.
Of course an allegation is evidence. If it was not evidence, we would expect to do as well investigating cases where there have been no allegations as cases where there have. That's crazy.
And what you've done above isn't "unpacking things" - you're using the same terms in the same ways with all the same assumptions.
If the allegation is evidence then we have to treat the denial as evidence too. Which one do we weigh more? We're in the same position as we were before the allegation was made. We still have to assume the accused is innocent, because that's what we assumed before the allegation was made.
My use of "unpacking" was not serious and in fact meant to be over-the-top condescending, as yours was.
I agree that it is not productive to continue this conversation. But for the record if you decide to argue against a bedrock ethical principle like the presumption of innocence, you should expect some pushback.
In the future you could probably earn more respect in debate, and perhaps even friends, by being less snarky and more respectful/charitable towards your opponent
"If the allegation is evidence then we have to treat the denial as evidence too."
Unquestionably.
"Which one do we weigh more?"
The one that is stronger evidence, in the precise mathematical sense - P(B|A)/P(B).
In the case of accusation, "the odds you are accused, given that you are guilty" may be low - let's say that only 10% of women so assaulted come forward. But the number of people accused is far less than one in ten.
In the case of denial, "the odds of denial, given innocence" is very nearly the same as "the odds of denial, regardless", so the odds don't shift back much.
So it is more likely after accusation and denial than before accusation and denial. Note that this doesn't mean that it is more probable than not, it means that it is more probable than it was. That depends on the base rate that we agree needs to be incorporated. But there being a low base rate doesn't change what is or isn't evidence, or the strength of that evidence.
"We're in the same position as we were before the allegation was made. We still have to assume the accused is innocent, because that's what we assumed before the allegation was made."
If that were true, it would make no sense to look into allegations - you'd be equally likely to find guilt looking into a random person.
'My use of "unpacking" was not serious and in fact meant to be over-the-top condescending, as yours was.'
Mine was not at all intended to be condescending. It was very specifically intended to say, "We aren't getting anywhere referring to 'presumption of innocence' without talking about what we mean by presumption of innocence", which is what I then tried to do.
"I agree that it is not productive to continue this conversation."
And yet we continue to. This isn't reflecting very well on either of us, I suppose.
"But for the record if you decide to argue against a bedrock ethical principle like the presumption of innocence, you should expect some pushback."
I was not arguing against the presumption of innocence. From my POV, I was arguing against misapplication, overextension, and dilution of the presumption of innocence. We are both of the opinion that "the presumption of innocence" as we understand it needs protecting.
"In the future you could probably earn more respect in debate, and perhaps even friends, by being less snarky and more respectful/charitable towards your opponent"
Pot, possibly lighter kettle. I fully agree we should be respectful and charitable. I try. I fail. When I fail, call me on the particular violation. You responded to perceived snark with snark, where snark wasn't even intended!
I was not advocating presuming guilt in any context - that would be even more nonsensical.