On the nature of evidence

[Skeptic Ginger]

....I take slight exception here. In the absence of other information the anecdotes are evidence of fairies. They are also evidence that the person(s) relating the anecdotes want you to believe that they observed something which they interpreted as "fairies".

It is only when the anecdotes are examined in light of other available evidence that one can conclude that the "fairy interpretation" is probably incorrect, but the "vision of fairies interpretation" might still be plausible.

I'm pretty sure this is what you were saying earlier in the post, but you seemed to almost turn it around here.

It takes a while sometimes for the statement clarifying what one is saying finally comes to one.

No, it is not what I am saying. I am saying that if the person claiming to have seen a fairy can show the evidence they relied on to someone else to examine, then you could actually examine the evidence. But without that evidence you have a person's conclusion, you do not have the evidence the person drew the conclusion from.

In this discussion, the person's conclusion is being substituted for the evidence. The conclusion is not the evidence.

 

[Skeptic Ginger]

Oy, are you ever gonna be in trouble when that invisible pink elephant decides to sit down on the same chair YOU happen to be sitting in...

Not to worry. I keep lot's of invisible pink unicorn horns around to poke elephants with. He'll not get very far into the chair.

 

[Skeptic Ginger]

I struggling to see how you can separate conclusions from evidence. As soon as you describe an observation you believe is evidence of something, you have made a conclusion that the observation is what you believe it to be.

Without linking an observation to a belief, the observation is not evidence (of anything). It is necessary to make a provisional statement of belief before it is possible to use observations to add or remove support for that belief.

And I'm struggling to see why that is such a difficult concept for people. It is so simple, so obvious, and so 'not hard' for me.

I have to wonder where our brains are different here. I mean that literally. We are organizing the data differently. (That is the collective we, BTW, not just me and you.)

Let's re-examine the nature of the word 'evidence'.

A thing has certain properties. Take a bicycle as an example. It has wheels, it has metal, it has brakes, it has color and weight, and so on.

It is also a vehicle for transportation. If I refer to the bicycle as a means of transportation, is that not a property of that bicycle? If I refer to it as a child's toy, is that not a property? In referring to those properties, does a person need to be using the bicycle as a toy or transportation device for me to conceptualize the bike in that way?

If I refer to the bicycle as evidence, I am indeed assigning a property to the bicycle. But I don't need to be using the bike as evidence, I don't need to know what it is evidence of in order to consider the bike as evidence. It could be something I see with a label that says "evidence", sitting in a police evidence room and nothing more. I might have no clue what it is evidence of. It could be evidence of a theft or a missing person or an accident or even a mistake and not evidence after all.

So why must one include a conclusion about the evidence in order to see its property of 'evidence'?

 

[Dr H]

Just consider that except for quantum mechanics, observations do not change the evidence.

OK, I'm going off on a tangent here, but you've made this statement several times in this thread and I have to say that I think it's based on a misconception.

The fact that a wave function can have multiple outcomes with equal or similar probability, and that the act of observation is believed to determine in certain circumstances which outcome becomes manifest, does not mean that 'observation has changed the evidence'. To draw such a conclusion is, in fact, to fall into the same error you so eloquently warn against in other posts in this thread. than is: blending conclusion with evidence.

The evidence in this case is not the observed outcome; the evidence is the wave function which describes the probabilistic array of outcomes for the system. The observation provides the additional information which changes the context in which the evidence is evaluated, and in that new context you reach a conclusion: the collapse of the function to a single state.

Sorry about the derail.

 

[Skeptic Ginger]

Why absurd?

Surely you've heard of not being so open minded your brain falls out?

 

[athon]

A quibble: I would say it's weak evidence because it is anecdotal.
Anecdotal evidence is inherently weak because it can be the result of a great number of things which can be difficult to control for: misperception; mistaken interpretation; dellusion; hallucination; deliberate fabrication; etc.

I fully agree.

Going on your responses above, I think I've clarified my own position a little better. Thanks. I agree that observations are weak when viewed in isolation, which is why cumulative observations strengthen a model. However, I'm still not sure that it is then no longer evidence at all. Ultimately I realise this discussion has certain parallels with the old mathematical 'is approaching zero the same as zero?' question. We both agree that an anecdotal observation concerning a phenomena which has relatively little to no additional supportive evidence is effectively useless.


I guess my view is one in which the label evidence is not relying on the conclusion on whether the model is true or not (as I said before, promoting circular reasoning). For instance, skeptigirl has said 'seeing a flying man, in conjunction with existing evidence, is not evidence of fairies'. I feel the only reason for this is that there is effectively too little supportive evidence in favour of fairies existing, rather than any physical universal law which would need to be contradicted in order for it to be true. So, should we later find a tiny humanoid winged corpse matching the description of the observation, that observation once again becomes evidence.

In other words, the label evidence only applies once a conclusion of sorts has been arrived at. On its own, the anecdotal evidence of a single observation when in conjunction with what we understand (the absence of collaborating evidence) seems to rely on a tentative conclusion to be reached which decides whether it can be called evidence or not. While I might be wrong, this seems to be circular and entirely provisional.

Another problem is that associating other observations with it is speculative. That in itself is unavoidable, however it doesn't automatically mean absolute relationships. For instance, associative evidence supporting the view that the observation of a flying humanoid is a hallucination might be the presence of a brain tumour in the subject. It certainly adds weight to the position, but the observation and brain tumour still only present circumstantial evidence of a hallucination. Far better than the fairy model, still...however it still troubles me that the label of evidence would be present only for the hallucination model and not the fairy model only on the grounds of speculation that more evidence exists for the hallucination model.

You are aware of future responses?

I think you meant subsequent responses.

'Future' from your relative position of the post you are responding to.

But whatever...it's your time. I just don't understand why you don't read a thread through and then respond to the most relevant posts - the last ones to respond to the point you are commenting on.

This still doesn't change what I am saying. Unless the person is lying (and I assume we are not including that possibility here) then they saw 'something'. It wouldn't matter what it was they saw. I agree that whatever they saw was 'evidence'. But here is where we differ. No matter what they conclude about what they saw, their conclusion does not change the evidence.

I'm not sure what this means. Evidence is only relative to a conclusion. If you say 'it's evidence', the natural question is 'of what?'. You can't respond, 'I don't know, but it's evidence of something'.

I think this is where we differ - to me, one observation can be evidence for a number of things, even if it is stronger evidence for some things than for others. It is not called evidence on the provision of a conclusion.

It is not weak evidence of fairies. It is not evidence of fairies at all.

Hence the core of this argument. I disagree.

First off, you do not have the evidence you are calling 'weak evidence'. You have a person's interpretation of some evidence.

This in itself is an absurd comment. I don't 'have' evidence of many, many things - I have the interpretation of somebody else's observation. So this in itself is irrelevant. What I think it is you're trying to say is we always evaluation evidence as it relates to the party presenting it. Which is indeed true. That does not mean the observation itself is no longer evidence because we personally don't make the direct observation - it means the nature of the observer is taken into account when weighing the evidence for its worth.

If you could examine the evidence the person drew the conclusion from, that would be something else altogether. Then you would have 'evidence' but in your scenario, you have a 'conclusion' that someone else formed, you do not have the actual evidence.

So are you saying you have no evidence for atoms being made of protons and neutrons? I assume you've never observed them first hand; you certainly would be relying on another's interpretation of their observations. Dismissing something as evidence because it isn't observed first hand is ridiculous, and I find it hard to believe this is what you meant.

Wrong. You can say that the individual concluded from some evidence that fairies existed. But you cannot say there is any evidence for fairies, weak or not, because you don''t have that evidence, you only have the conclusion.

Again, I've acknowledged the error of labelling the observation as a conclusion. A man saying 'I saw a winged flying humanoid' is an observation. He says 'it was a fairy', it would be a conclusion. I would say 'it is evidence for fairies, however there is greater evidence that you are a) lying or b) mistaken'. To be pedantic, indeed for me personally the evidence is not his observation itself, but rather his communicating of that observation. Yet this extends to most things I understand about the universe.

Once enough people have examined enough evidence for certain conclusions about the nature of the Universe, we can say, there is evidence for something. If it is 'extraordinary' or new but not enough evidence has been examined by enough people, you can still say there might be evidence or there is weak evidence for something.

This is edging on argument from popularity. What is 'enough'? Does it matter who these people are? What if I have several thousand people who think they've seen a flying humanoid - would you then call it evidence on these grounds? That seems silly. Either something is evidence for something or it isn't - it doesn't magically become evidence once enough people agree.

On the other hand, I might concede that the term is entirely subjective. This in itself presents problems, however, as we've shown.

But for things like fairies, none of the conclusions that evidence exists for fairies have been backed up by producing any of that evidence for others to examine.

But here is your circular reasoning again. It's not evidence because there is no other evidence. You seem to be suggesting that it can only be called evidence if it can be reproduced on command, though. Again, this is ridiculous. It is certainly helpful, but a great deal of science concerns itself with phenomena which cannot be reproduced spontaneously, and has to rely on other forms of observation.

I would answer that logic alone should tell you there must be things for which there is no evidence for the existence of those things.

Can you rephrase this? I'm afraid I'm not sure what you mean.

Athon

 

[arthwollipot]

I can look around my room and say definitively there is no elephant in here. Yet you want to argue some philosophical principle that is not the case. I will stick with reality, thank you. There is a point where taking that philosophical principle to an absurd extreme becomes as much woo as someone not considering an evidence based Universe.

Why absurd?

Surely you've heard of not being so open minded your brain falls out?

Of course, but in this situation, the discussion is not about common everyday usage. Everyone can agree on what evidence is if there is no rigour in the question. We're talking about epistemology here, not common sense.

I started typing out a long elaboration involving the spider that isn't on my desk, but it didn't say anything that you don't already know.

 

[athon]

Art has it in one there. Many disagreements arise when there is a misunderstanding between a common understanding of a term and an epistemological understanding. The word 'theory' is a perfect example - a person might say 'I have a theory that...', only to be met with 'that's not a theory - in which peer-reviewed journal have you published your experiment's results?'. Neither person is using the term wrongly, even though the contexts conflict.

Proving negatives is another example. Common sense tells you there is no elephant in the shoebox. You look...and there is none. However the epistemology behind it does not rely on there being a proof of negative, but rather absence of positive. Philosophically, in science, this is significant, even though in everyday practicality it seems to be an irrelevant contrast.

Athon

 

[arthwollipot]

I actually liked the phrase I used "We're talking about epistemology here, not common sense", and I almost joined the ranks of solipsistic autosycophants when I posted it.

 

[Skeptic Ginger]

....I guess my view is one in which the label evidence is not relying on the conclusion on whether the model is true or not (as I said before, promoting circular reasoning). For instance, skeptigirl has said 'seeing a flying man, in conjunction with existing evidence, is not evidence of fairies'.

That is not what I am saying. "a flying man" is still a conclusion. I said this, but you keep telling me I'm not following the conversation. Yes I am. The thing the person saw, or what ever triggered them to think they saw something is the evidence, regardless of what they call it, regardless of what their brain sees, regardless if they are correctly or incorrectly observing something.

I remain surprised this is a hard concept for people to understand. However, if you understand the nature of evidence, you will understand I am correct and I do know what I am talking about.

• There is the evidence.
• There is the observation/measurement of the evidence.
• There is the conclusion drawn about the evidence.
• There is communication of that conclusion.

To understand the principles of an evidence based world view, you do need to understand the difference between these 4 things. People deal with evidence and observation/measurement of the evidence interchangeably without realizing they are doing so. What is evidence is critical though, because that is the basis of understanding the Universe. It isn't just a matter of the best explanation for the evidence, it is also important to understand what is evidence what is not evidence but rather a belief or conclusion about the evidence.

Regardless that we don't always consciously recognize what is the evidence and what is the conclusion, they are two separate things.

I'll have to elaborate further and reply to the rest of your post tomorrow. I'm out of time again.

 

[Skeptic Ginger]

Of course, but in this situation, the discussion is not about common everyday usage. Everyone can agree on what evidence is if there is no rigour in the question. We're talking about epistemology here, not common sense.

I started typing out a long elaboration involving the spider that isn't on my desk, but it didn't say anything that you don't already know.

We were talking about you trying to tell me I could not say with certainty there was no elephant in my room. Or have you forgotten what absurd claim it was you think I should keep an open mind about?

 

[athon]

That is not what I am saying. "a flying man" is still a conclusion. I said this, but you keep telling me I'm not following the conversation. Yes I am. The thing the person saw, or what ever triggered them to think they saw something is the evidence, regardless of what they call it, regardless of what their brain sees, regardless if they are correctly or incorrectly observing something.

How the hell is 'he saw a flying man' a conclusion? How could you describe the event in some way that wouldn't, according to you, be a conclusion? Consider 'I saw a flying humanoid' is not the same as 'it was a flying humanoid I saw', which would better match your claim.

I remain surprised this is a hard concept for people to understand. However, if you understand the nature of evidence, you will understand I am correct and I do know what I am talking about.

Whoa. Well, maybe that's why you find yourself surprised - holding the arrogant stance of 'I am correct and do know what I am talking about' will only serve to confuse you enough to not consider alternative stances.

The issue is not a pragmatic one, but rather perception of the parameters of a term. The reason I started this discussion was so I could gain a better grasp of the subtly different ways people perceive the term, and it has served its purpose well. However nobody else has claimed to hold an objective, definitive 'correctness' of the matter. I actually respect your views, skeptigirl, and feel you often articulate your position well. But this 'I am correct' stance is little rich.

• There is the evidence.
• There is the observation/measurement of the evidence.
• There is the conclusion drawn about the evidence.
• There is communication of that conclusion.

I can't see where this list could possibly fit in a discussion based on defining precisely what constitutes 'evidence'. It's like Claus discussing 'sell' without defining how he's using the term.

I think a better way of putting it is

• There is the observation.
• There is the relating of the observation to a speculative cause, defining it thus as evidence.
• There is the weighing of the evidence in relation to other observations.
• Confidence in the speculation is gained or lost on the weight of the evidence.

To understand the principles of an evidence based world view, you do need to understand the difference between these 4 things. People deal with evidence and observation/measurement of the evidence interchangeably without realizing they are doing so. What is evidence is critical though, because that is the basis of understanding the Universe. It isn't just a matter of the best explanation for the evidence, it is also important to understand what is evidence what is not evidence but rather a belief or conclusion about the evidence.

Yet you use the word 'evidence' as if it is self explanatory. The very point of this thread is that the term doesn't have a simple, clear definition understood the same by everybody.

Athon

 

[arthwollipot]

We were talking about you trying to tell me I could not say with certainty there was no elephant in my room. Or have you forgotten what absurd claim it was you think I should keep an open mind about?

And I was saying that claim was not absurd when addressed from an epistemological point of view. In fact, it is a perfectly logical statement, epistemologically.

Let me be clear. Of course there is no elephant in your room. Only an absolute idiot (or an anally retentive epistemologist) would doubt your claim that there was no elephant in your room. But that's not what we're talking about. We're talking about what can be considered evidence of a negative statement.

Say that you're claiming that there is an elephant in your room. Evidence of its existence is easy to come by. I could drop over to your place and take a look at it myself if necessary. Evidence of the elephant's absence is a lot harder to justify from a logical, epistemological standpoint.

For all practical purposes, taking the logical, epistemological standpoint is just pointless wankery. And anyway, it makes my brain hurt. So how about we just use common sense and say that there's no elephant?

 

[Dr H]

It takes a while sometimes for the statement clarifying what one is saying finally comes to one.

No, it is not what I am saying. I am saying that if the person claiming to have seen a fairy can show the evidence they relied on to someone else to examine, then you could actually examine the evidence. But without that evidence you have a person's conclusion, you do not have the evidence the person drew the conclusion from.

In this discussion, the person's conclusion is being substituted for the evidence. The conclusion is not the evidence.

The conclusion is not the evidence, but the claim is evidence. It is not first hand evidence, but it is still evidence in the form of testimony, also called "anecdotal evidence," and which you gave several examples of earlier.

 

[Dr H]

I fully agree.
Going on your responses above, I think I've clarified my own position a little better. Thanks. I agree that observations are weak when viewed in isolation, which is why cumulative observations strengthen a model. However, I'm still not sure that it is then no longer evidence at all.

As I said, it still is evidence --strength not withstanding -- it just may not be evidence of what you originally thought it was evidence of.

Ultimately I realise this discussion has certain parallels with the old mathematical 'is approaching zero the same as zero?' question. We both agree that an anecdotal observation concerning a phenomena which has relatively little to no additional supportive evidence is effectively useless.

Effectively useless as a explanation, yes. It may be useful for prodding the curious to go out and seek additional evidence in refutation or corroboration.

I guess my view is one in which the label evidence is not relying on the conclusion on whether the model is true or not (as I said before, promoting circular reasoning). For instance, skeptigirl has said 'seeing a flying man, in conjunction with existing evidence, is not evidence of fairies'. I feel the only reason for this is that there is effectively too little supportive evidence in favour of fairies existing, rather than any physical universal law which would need to be contradicted in order for it to be true. So, should we later find a tiny humanoid winged corpse matching the description of the observation, that observation once again becomes evidence.

In other words, the label evidence only applies once a conclusion of sorts has been arrived at. On its own, the anecdotal evidence of a single observation when in conjunction with what we understand (the absence of collaborating evidence) seems to rely on a tentative conclusion to be reached which decides whether it can be called evidence or not. While I might be wrong, this seems to be circular and entirely provisional.

I don't entirelly agree. All this does is shift the issue of defining "evidence" to that of what is meant by "conclusion." It is probably impossible for any information relayed as evidence to be completely free of conclusions of some sort on the part of the person doing the relaying. Human beings are not video recorders: all of our observations are "theory laden," to some extent. Even relating your sighting as a "winged human being" or a "flying man" involves certain conclusions as to what constitutes a "wing", a "human being," a "man", or the act of "flying".

I don't see how you can get away from this, and if you can't, then placing the requirement on information that it be absolutely free of interpretation/conclusion before it can be considered "evidence," pretty much defines "evidence" out of practical existence.

Another problem is that associating other observations with it is speculative. That in itself is unavoidable, however it doesn't automatically mean absolute relationships. For instance, associative evidence supporting the view that the observation of a flying humanoid is a hallucination might be the presence of a brain tumour in the subject. It certainly adds weight to the position, but the observation and brain tumour still only present circumstantial evidence of a hallucination. Far better than the fairy model, still...however it still troubles me that the label of evidence would be present only for the hallucination model and not the fairy model only on the grounds of speculation that more evidence exists for the hallucination model.

Well, that would be an argument to have with other posters -- I agree with you there.

 

[Dr H]

We're talking about epistemology here, not common sense.

ROTFL! Now that is a line worth quoting, if only I can figure out where.

 

[Dr H]

I remain surprised this is a hard concept for people to understand. However, if you understand the nature of evidence, you will understand I am correct and I do know what I am talking about.

• There is the evidence.
• There is the observation/measurement of the evidence.
• There is the conclusion drawn about the evidence.
• There is communication of that conclusion.

What you say makes sense, but is incomplete. I submit the following list is more reflective of the actual progression:

• There is information.
• There is observation/measurement of the information.
• There is the provisional classification of the information as evidence as regards some postulate.
• There is a conclusion drawn about the evidence.
• There is communication of that conclusion.

The postulate in step three is most often going to reflect some other, pre-existing evidence in the posession of the observer/measurer (though it could, in theory, reflect mere speculation).

To begin with "there is evidence," is incorrect, however, because "evidence" must be evidence of something or evidence for smoething. To begin the list with "there is evidence" is to begin with a conclusion that the information obtained is, in fact, evidence.

 

[Dr H]

• There is the observation.
• There is the relating of the observation to a speculative cause, defining it thus as evidence.
• There is the weighing of the evidence in relation to other observations.
• Confidence in the speculation is gained or lost on the weight of the evidence.

Yet you use the word 'evidence' as if it is self explanatory. The very point of this thread is that the term doesn't have a simple, clear definition understood the same by everybody.

Oops. I see I should have scrolled just that -tiny- bit further down the page before my last post, in which I essentially repeated your list in paraphrase. Didn't see this post before I sent mine into the ether.

Ah well, it would seem that our respective positions are indeed rather similar, anyway.

 

[Skeptic Ginger]

What you say makes sense, but is incomplete. I submit the following list is more reflective of the actual progression:

• There is information.
• There is observation/measurement of the information.
• There is the provisional classification of the information as evidence as regards some postulate.
• There is a conclusion drawn about the evidence.
• There is communication of that conclusion.

The postulate in step three is most often going to reflect some other, pre-existing evidence in the posession of the observer/measurer (though it could, in theory, reflect mere speculation).

To begin with "there is evidence," is incorrect, however, because "evidence" must be evidence of something or evidence for smoething. To begin the list with "there is evidence" is to begin with a conclusion that the information obtained is, in fact, evidence.

I'm having a hard time separating "provisional classification of the information as evidence as regards some postulate" from "conclusion". It would seem all you are saying is first you draw a temporary conclusion then a final one. I do not consider "conclusion" to have some exclusive finality.

 

[Skeptic Ginger]

How the hell is 'he saw a flying man' a conclusion? How could you describe the event in some way that wouldn't, according to you, be a conclusion? Consider 'I saw a flying humanoid' is not the same as 'it was a flying humanoid I saw', which would better match your claim.

In one case you have a person telling you about the evidence. In doing so, the person is drawing some conclusions, be they valid or invalid. The evidence is what the person saw, not what they tell you they saw. And in response to your earlier concerns about this, yes, you are considering other people's observations of evidence all the time. No, we do not observe all evidence directly. That doesn't change what is evidence and what is a conclusion about the evidence. It merely allows me to consider some people credible and others less so.

There is no prohibition on accepting evidence someone else observed that you find to be a credible report of that evidence. Just because you are not observing evidence directly does not mean the evidence is not useful to you. From experience, we conclude that lots of evidence we have not directly observed is nonetheless credible.

The problem here is the evidence, whatever it was, a misinterpretation of some external or internal stimuli, does not become evidence OF something just because someone concludes it is evidence of that thing.

Whoa. Well, maybe that's why you find yourself surprised - holding the arrogant stance of 'I am correct and do know what I am talking about' will only serve to confuse you enough to not consider alternative stances.
The issue is not a pragmatic one, but rather perception of the parameters of a term. The reason I started this discussion was so I could gain a better grasp of the subtly different ways people perceive the term, and it has served its purpose well. However nobody else has claimed to hold an objective, definitive 'correctness' of the matter. I actually respect your views, skeptigirl, and feel you often articulate your position well. But this 'I am correct' stance is little rich.

I recognize the arrogance problem. But when one knows they are correct, it is hard to have a different stance. There are many occasions I am not so certain as in this case.

I can't see where this list could possibly fit in a discussion based on defining precisely what constitutes 'evidence'. It's like Claus discussing 'sell' without defining how he's using the term.

I think a better way of putting it is

• There is the observation.
• There is the relating of the observation to a speculative cause, defining it thus as evidence.
• There is the weighing of the evidence in relation to other observations.
• Confidence in the speculation is gained or lost on the weight of the evidence.

Yet you use the word 'evidence' as if it is self explanatory. The very point of this thread is that the term doesn't have a simple, clear definition understood the same by everybody.

Athon

I find the problem not that the term, 'evidence', is in reality, unclear, but rather, that it is misused so often its actual definition is unclear to many people.

Here are the problems that misusing the term incurs and why it matters as well as why what I am saying is correct.

Is there really actual evidence of fairies if you do not consider mistaken conclusions about things people have observed or experienced?

If you cannot separate your definition of the term, 'evidence', then let me ask you another question. Is it possible for people to believe in something for which there is no evidence for?

There is evidence people have claimed they believe is evidence of fairies. The evidence, whatever it is, exists. Does a false conclusion about that evidence change that evidence and make it something it is not?