Jump to content


Photo

The Epistemology Corner: How Do We Know What We Know?


  • Please log in to reply
526 replies to this topic

#1 Mattias

Mattias

    Member

  • Veteran Member
  • PipPipPip
  • 180 posts
  • Age: 43
  • no affiliation
  • Atheist
  • Sweden

Posted 11 January 2015 - 08:21 AM

Here I will post discussions of general interest regarding epistemology: the study of how we know what we can know. I invite everybody to join in and post their own examples.


  • svigil777 likes this

#2 mike the wiz

mike the wiz

    Veteran member

  • Veteran Member
  • PipPipPipPip
  • 3,375 posts
  • Gender:Male
  • Interests:mikey mischief.
  • Age: 36
  • Christian
  • Creationist
  • England

Posted 11 January 2015 - 09:58 AM

My understanding is there is an epistemological definition of "knowledge" that could be regarded as a Justified, true, belief, also known as a J.T.B. The other is the direct translation of "science" which means, "knowledge", if I remember correctly.

 

The problem with a JTB is that it seems to be an argument that is inductively-probable, IMHO, depending upon a scale of various inductive arguments from weak-to-strong.

 

The problem with defining science as knowledge is that we can see from the past, some theories such as Steady-State-Universe and Spontaneous Generation, we didn't actually, "know" they were true, and they were proven not to be, which proves science isn't really knowledge unless the theory in question is a very incontrovertibly strong case.

 

Examples:

 

Gravity. (Inverse square law)

Exotic air. (presence of different types of air in the atmosphere)

Germ theory of disease. (That micro-forms exist)

Laws of physics, that aren't inductively broken such as the law of Biogenesis, and the various physical forces such as linear momentum, conservation of angular momentum, and so forth.

 

These are all examples of JTB-knowledges but they are also very strong both deductively and inductively, so strong the burden-of-proof is upon those that claim other facts to the contrary.

 

Perhaps this topic belongs more in the Miscellaneous-section.

 

Weak examples of science/knowledge, are:

 

Planet collissions.

Aggregation.

Biological evolution.

Abiogenesis.

Big Bang.

Uniformitarianism. 

 

This is because such theories rely HEAVILY on conjecture, and even contradictory or only circumstantial evidence. The more amenable the theory is to change, the more plastic and un-knowledgeable it is. For example, evolution has competing hypotheses for bird-evolution, therefore how can we, "KNOW" bird evolution happened? If we base such knowledge on fragmentary evidence, we can see that that line of evidence may well be abandoned. Another example is the list of some 200 vestigial parts that were told to be evolutionary-leftovers in the human body, all but a handful of these are still claimed to be vestigial. As scientific knowledge grown, it was discovered that the various organs/body parts did in fact serve some purpose to the human body, and were in fact NOT evolutionary leftovers.

 

If we, "KNEW" they were evolutionary leftovers, how can we UN-KNOW that they are? Once a knowledge is truly known, if it is later disproven, then reasonably and logically, it was never knowledge to begin with.



#3 mike the wiz

mike the wiz

    Veteran member

  • Veteran Member
  • PipPipPipPip
  • 3,375 posts
  • Gender:Male
  • Interests:mikey mischief.
  • Age: 36
  • Christian
  • Creationist
  • England

Posted 12 January 2015 - 01:11 PM

For my example of a J.T.B I have thought of an analogy that I hope you will read, Mattias.

 

Imagine I looked down at my watch and it reads, 8.10, whether am or pm, you can assume either. Now imagine if I walked into the kitchen and happened to look at the clock and it said, 8.10. I would have a justified, true belief that the time was 8.10. Would I therefore, "know" it was 8.10? I would venture to say, "no" to that question.

 

So then our argument is Probable-Induction:

 

I observe x, I observe x again, ergo x.

 

So then it is probably 8.10. But then let us imagine the following scenario. Let us pretend that in fact both the watch and clock had stopped at 8.10 and I simply hadn't noticed the second hands were still.

 

Probability-wise, what are the chances?

 

Since there are 24 hours in a day and 60 minutes in an hour, then that is 60 X 24 = 1440 chances. BUT, since the clock can strike 8.10 twice in a day, am/pm, we divide the figure by 2, so 1440/2 = 720 = 1 in 720 chance of one clock stopping once at that time, but because we observed two clocks stopped, we multiply by 2. So 720 X 720 = 518, 400 so the chances of finding two clocks both stopped at the same time is 1 in 518, 400. Or just over half a million to one.

 

I hope my mathematics are correct. So then it is highly probable, you have a JTB that the time is 8.10 but I would venture to say that logically, technically speaking, you don't KNOW it is 8.10.

 

However I would venture to say, that the more absurd the improbability, the closer it is to "knowledge" until we reach a place were it becomes TOTALLY absurd to say that we, "do not know".

 

For three stopped clocks it would be 1 in 373, 248, 000. (Nearly 400 million to one)



#4 Mattias

Mattias

    Member

  • Veteran Member
  • PipPipPip
  • 180 posts
  • Age: 43
  • no affiliation
  • Atheist
  • Sweden

Posted 12 January 2015 - 01:26 PM

MIke, a quick response: Your probability estimates are off. It is only when you notice the first watch having stopped at 8.10 that you are starting to set a time to compare against. The probability calculation should thus be based only on the second clock. The chance is 1 in 720 if you have two clocks.



#5 piasan

piasan

    Veteran Member

  • Veteran Member
  • PipPipPipPip
  • 2,727 posts
  • Gender:Male
  • Location:Oklahoma
  • Age: 71
  • Christian
  • Theistic Evolutionist
  • Oklahoma

Posted 12 January 2015 - 02:03 PM

MIke, a quick response: Your probability estimates are off. It is only when you notice the first watch having stopped at 8.10 that you are starting to set a time to compare against. The probability calculation should thus be based only on the second clock. The chance is 1 in 720 if you have two clocks.

Exactly.

 

If we were to say what is the probability of two clocks stopping at the same time, it would be 1 in 720.  If you were to ask what is the chance of two clocks stopping at 8:10, then it would be 1 in 7202 or 1 in 518,400.  In the first example, the first clock is used to set the reference time... in the second example, the reference time (ie: goal) is already set.


  • Mattias likes this

#6 Mattias

Mattias

    Member

  • Veteran Member
  • PipPipPip
  • 180 posts
  • Age: 43
  • no affiliation
  • Atheist
  • Sweden

Posted 12 January 2015 - 02:49 PM

Mike, I understand your general thinking though, and that seems legit. But as you noticed, real probability calculations should preferably be done in advance, rather than post hoc. Retrospective probability calculations can be notoriously tricky and full of pitfalls.



#7 mike the wiz

mike the wiz

    Veteran member

  • Veteran Member
  • PipPipPipPip
  • 3,375 posts
  • Gender:Male
  • Interests:mikey mischief.
  • Age: 36
  • Christian
  • Creationist
  • England

Posted 12 January 2015 - 04:09 PM

No Mattias, I was aware of that, but actually I am assuming the odds for a particular time.

 

If there are a thousand armco barriers on a racetrack, the chances of hitting a particular barrier (without being too pedantic, please folks) are about 1 in 1,000, for the sake of argument/maths, but the chances of hitting any barrier are not the same, if a crash is inevitable, and it is inevitable to hit a barrier, then it would be a much better chance.

 

You see, the chances of flipping two heads in a row are, 1 in 2 x 2 = 4, because there are 4 possibilities.

 

1. H, H.

2. H, T

3. T, T

4. T, H

 

 

If you flip a coin, your chances of a heads first time are 1 in 2 because there are two possibilities, so you are focusing on ANY result, the chances of hitting tails or heads are 1 in 1. If you look at a clock, the chances of it being on a particular minute of an hour are 1 in 720. 

 

The time I chose was a particular minute, not ANY minute, it was a particular time of 8.10. The chances are 1 in 720 for a particular minute, the chances are 1 in 1 for ANY minute, as long as I have a watch. It's important to not conflate the two different probabilities.



#8 mike the wiz

mike the wiz

    Veteran member

  • Veteran Member
  • PipPipPipPip
  • 3,375 posts
  • Gender:Male
  • Interests:mikey mischief.
  • Age: 36
  • Christian
  • Creationist
  • England

Posted 12 January 2015 - 04:17 PM

 

 

 Piasan: If you were to ask what is the chance of two clocks stopping at 8:10, then it would be 1 in 7202 or 1 in 518,400

 

Which is exactly what I said, you are being pedantic because I am not on the side of atheism/evolution, and you have to keep up the pretense that my intellect is thwarted.



#9 mike the wiz

mike the wiz

    Veteran member

  • Veteran Member
  • PipPipPipPip
  • 3,375 posts
  • Gender:Male
  • Interests:mikey mischief.
  • Age: 36
  • Christian
  • Creationist
  • England

Posted 12 January 2015 - 04:24 PM

Of course, the probability was only an example of how unlikely it would be that both your watch and clock would have stopped at the same time.

 

In a world with only one hour, there would be these possibilities.

 

W:1 min C: 1 min

W: 1 min C: 2 min

W: 1 min C: 3 min

 

That's just for the watch being on 1 minute (60 times), we then go through every other minute, so it would be 60 X 60 because there would be 3,600 possibilities, but to strike both 1 minute for the watch and 1 minute for the clock would be only 1 possibility in 3,600.



#10 piasan

piasan

    Veteran Member

  • Veteran Member
  • PipPipPipPip
  • 2,727 posts
  • Gender:Male
  • Location:Oklahoma
  • Age: 71
  • Christian
  • Theistic Evolutionist
  • Oklahoma

Posted 12 January 2015 - 06:27 PM

If we were to say what is the probability of two clocks stopping at the same time, it would be 1 in 720.  If you were to ask what is the chance of two clocks stopping at 8:10, then it would be 1 in 7202 or 1 in 518,400.  In the first example, the first clock is used to set the reference time... in the second example, the reference time (ie: goal) is already set.

Which is exactly what I said, you are being pedantic because I am not on the side of atheism/evolution, and you have to keep up the pretense that my intellect is thwarted.

Yeah, after I made the post, I went back and saw I had missed that.  Since my explanation covered both the time being pre-determined and it not being pre-determined, I figured my comments were still accurate.  It was an oversight on my part that I failed to notice you had pre-determined the time and had nothing at all to do with you being a creationist.

 

You really should take that chip off your shoulder and stop looking for things with which to take offense... it's not necessary here.  In my experience, when one looks for things to be offended by, one will find them ..... whether they're real or not.



#11 Mattias

Mattias

    Member

  • Veteran Member
  • PipPipPip
  • 180 posts
  • Age: 43
  • no affiliation
  • Atheist
  • Sweden

Posted 12 January 2015 - 08:48 PM

 

Which is exactly what I said, you are being pedantic because I am not on the side of atheism/evolution, and you have to keep up the pretense that my intellect is thwarted.

 

I concur with piasan. That was an entirely unwarranted comment.



#12 mike the wiz

mike the wiz

    Veteran member

  • Veteran Member
  • PipPipPipPip
  • 3,375 posts
  • Gender:Male
  • Interests:mikey mischief.
  • Age: 36
  • Christian
  • Creationist
  • England

Posted 13 January 2015 - 04:34 AM

I'm wouldn't go so far as to say I am offended, Piasan. I took umbridge, that's all, because it seems the evolutionists cant even agree with me on a neutral subject. But I appreciate that I might have jumped to a conclusion prematurely if that was not your intention.

 

 Your comment Mattias, is extraneous. You concur so that one more vote is on the side of the evolutionists to make the creationist look bad.worship.gif (joking) burp.gif  (Mikey-mischief complete.) wink.png



#13 Schera Do

Schera Do

    Banned

  • Banned
  • PipPipPipPip
  • 1,373 posts
  • Gender:Male
  • Interests:Critical analysis and referents, Ephectic, Ultimate questions & how to answer, political philosophy, Constitutional Conservatism
  • Age: 55
  • (private)
  • Agnostic
  • Northeastern U.S. of A.

Posted 13 January 2015 - 05:21 AM

The "nature of knowledge": let us begin, as nobody has begun.

Eyes
Ears
Physical sensation
Nose
Brain

#14 Mattias

Mattias

    Member

  • Veteran Member
  • PipPipPip
  • 180 posts
  • Age: 43
  • no affiliation
  • Atheist
  • Sweden

Posted 17 January 2015 - 02:01 AM

I'm wouldn't go so far as to say I am offended, Piasan. I took umbridge, that's all, because it seems the evolutionists cant even agree with me on a neutral subject. But I appreciate that I might have jumped to a conclusion prematurely if that was not your intention.

 

 Your comment Mattias, is extraneous. You concur so that one more vote is on the side of the evolutionists to make the creationist look bad.worship.gif (joking) burp.gif  (Mikey-mischief complete.) wink.png

 

Sorry; I accidentally quoted the wrong post; above is what I intended to quote.

 

No, Mike, i concur to cast one more vote from a neutral spectator’s perspective of your manifest behaviour, which would encourage you to take a long, good look in the mirror, do some soul-searching, and ask yourself what prompted you to utter the following statement:

 

“…you are being pedantic because I am not on the side of atheism/evolution, and you have to keep up the pretense that my intellect is thwarted.

 

Because, again, the statement was entirely unwarranted, based on the conversation that you had with piasan. The conclusion you draw in no way follows from the simple fact that the two of you had a rather trivial disagreement regarding probability calculus. Thus, I must conclude that you are adding hidden assumptions to the mix. I encourage you to perform a thorough survey of your own assumptions regarding your fellow human beings, which you are adding to your evaluation of their likely motives, and ask yourself whether this is perhaps an emanation from your own mind, which you project onto them, rather than a warranted conclusion from their behaviour?

 

Besides, from your description of the situation with the clocks, it seems evident that your assumption of the specific time point of 8.10 was a retrospective hypothesis that you came up with only after you noticed that the clocks seemed to have stopped at this time. Thus, it seems like you should use the probability of 1:720 based on a single clock only. Based on your description of the situation, a mathematically precise formulation of your hypothesis would be: “Blimey, both my clocks have stopped at 8.10. Whoa, what are the odds….?” This would make it into an a posteriori hypothesis, rather than an a priori hypothesis, which would be required to assume a probability calculation based on both clocks. But both piasan and I may have misunderstood your intended hypothesis; it remains a reasonable conclusion, however, based on your description of the situation.



#15 mike the wiz

mike the wiz

    Veteran member

  • Veteran Member
  • PipPipPipPip
  • 3,375 posts
  • Gender:Male
  • Interests:mikey mischief.
  • Age: 36
  • Christian
  • Creationist
  • England

Posted 17 January 2015 - 02:14 PM

 

 

Mattias: Because, again, the statement was entirely unwarranted

 

I said I was perhaps wrong to come to that conclusion, short of sending me to prison, what else can I say? I was wrong, I made a rash conclusion. I am not perfect. What more can I say? 

 

 

 

Mattias:Besides, from your description of the situation with the clocks, it seems evident that your assumption of the specific time point of 8.10 was a retrospective hypothesis that you came up with only after you noticed that the clocks seemed to have stopped at this time. Thus, it seems like you should use the probability of 1:720 based on a single clock only. Based on your description of the situation, a mathematically precise formulation of your hypothesis would be: “Blimey, both my clocks have stopped at 8.10. Whoa, what are the odds….?” This would make it into an a posteriori hypothesis, rather than an a priori hypothesis, which would be required to assume a probability calculation based on both clocks. But both piasan and I may have misunderstood your intended hypothesis; it remains a reasonable conclusion, however, based on your description of the situation.

 

It depends upon how each person thinks about it. I still think 1 in 720 X 2 is correct, even posteriori, because logically we must ask, "what are the chances of the same particular time being struck?"

 

To explain my point I would like to make an analogy. Imagine we are at a race-track, where there are 1,000 armco barriers. Now imagine that we number them all, 1 to 1,000

 

So then now let us imagine if there was a race at this track and there were 10 crashes, that hit armco barriers number, 2,15, 94,127, 135, 381, 456, 512, 593, 995. In each instance the relevance of the odds are only 1 in 1,000 for each barrier, BECAUSE each barrier is a DIFFERENT particular place

 

So if that happened, we would not say, "wow, what are the odds of that?" Nobody would say that, in fact all they would say is this: "I can't believe how many crashes took place in this race". So probability is not so relevant in this scenario, would you agree?

 

But now let us pretend that the race took place, and 10 crashes happened only this time the armco barriers that were hit were, barriers number, 105, 105, 105, 105, 105, 105, 105, 105, 105, 105.

 

Now what are the odds? I don't know a person that would not find this improbable. That is because in THIS example, the SAME PARTICULAR PLACE was crashed into. NOT different particular places.

 

It seems to me, that we would now calculate the odds based on the number of possibilites/options. For example, the first crash could have hit barrier 1, then the following 9 crashes hit barrier number 105. But the first could have hit barrier 2, and the following 9 hit barrier 105, or instead the first crash could have hit barrier 3, and the following 9 hit barrier 105.

 

Can you see the problem, yet? The number of possibilities is tremendous. You would have to say it was 1:1,00010

 

Isn't it the same with a coin-toss? If you said to me, "hey look at this, I have just got three flips, three heads in a row, what are the chances?" I would calculate, even though it was posteriori, that the chances are based on how many possible options can be thrown in three attempts: 1:23

 

T,T,T

T,H,H

T,T,H

T,H,T

H,T,T

H,H,T

H,H,H

H,T,H

 

But I also understand your point about an apriori probability, because those random numbers in the first crash-analogy, that were DIFFERENT particular places, would be relevant to probability, if I had predicted beforehand, that those particular barriers would be hit, which would be the same as choosing lottery numbers, apriori.

 

You see, there is no reason for the same particular barrier to be hit, or the same time to be stopped at, which is why we use probability, I would argue, because it's incredible. It tends to NOT happen much more than it DOES happen.



#16 Mattias

Mattias

    Member

  • Veteran Member
  • PipPipPip
  • 180 posts
  • Age: 43
  • no affiliation
  • Atheist
  • Sweden

Posted 18 January 2015 - 02:26 PM

Now I would like to present some case studies of general epistemological interest, concerning the consequences of hypothesis testing. Many people probably intuitively assume that the outcome of a hypothesis test is always symmetrical with regards to observations confirming or refuting the hypothesis. That is not always the case, however. There are many potential and actual scenarios where a specific observation can mainly refute or confirm the hypothesis, but not both.

 

An interesting case can be derived from the very real example of the presumed remains of Richard III – the last Plantagenet king of England, who is infamous for usurping the throne of his young nephew and probably also having him murdered (see picture below). If you want to know more about Richard himself you can check Wikipedia or listen to the fascinating, hilarious, and witty Rex Factor podcast, which presents the history of all the kings and queens of England from Alfred the Great until Elizabeth II.

 

Something that the podcast does not tell you (because this happened after the podcast was recorded), is that the presumed remains of Richard III have recently been found in an excavation of the former Greyfriars Church under a car park in Leicester. Richard was supposedly buried there after being killed during a valiant charge at the Battle of Bosworth, but knowledge of the location of Richard’s burial site had been lost since the reformation. Apart from the emotionally captivating background story, the case is also interesting because it comprises both historical and forensic details providing a very strong scientific case regarding the identity of the remains.

 

Specific deformities and battle wounds on the exhumed skeleton agreed fairly well with what the chronicles had to say about Richard’s appearance and how he was supposed to have been killed in his last battle. In addition, however, confirmation beyond reasonable doubt was obtained by means of genetic comparisons. The mitochondrial DNA haplotype from two current matrilineal relatives of Richard matched that of the mitochondrial DNA of the skeleton.

 

So, that’s all well and good for the history buffs. The mitochondrial DNA is the kind of evidence that could have both verified and falsified the hypothesis that the skeleton belonged to Richard III with reasonable certainty. But what if the scientists had instead tried to use the male Y-chromosome of the skeleton, and matched it to a single presumed patrilinieal relative of Richard III? Here we would get more of an asymmetrical relationship. A match between the skeleton and the presumed relative would still constitute very good evidence of the identity of the skeleton. However, a mismatch might just as well be a case of mistaken paternity at some point in male line of the genealogical records. This kind of evidence is thus only able to confirm, but hardly to falsify, the hypothesis.

 

We have the opposite relationship in many cases when, in the good old days before DNA fingerprinting, they could mainly use blood group analysis to determine paternity. In some cases it was possible to exclude paternity based on a mismatch in the blood group between a child and the presumed father. In most cases, however, blood groups would not be rare enough to confirm paternity with any degree of certainty, even in case of a match. So here we have evidence that can mainly falsify (in case of specific outcomes) but hardly ever confirm, a hypothesis about paternity.

 

 

Richard_III_earliest_surviving_portrait.


  • svigil777 likes this

#17 Mattias

Mattias

    Member

  • Veteran Member
  • PipPipPip
  • 180 posts
  • Age: 43
  • no affiliation
  • Atheist
  • Sweden

Posted 19 January 2015 - 01:01 AM

BWT: I see now that they have actually published their findings about the skeleton in Nature Communications, and I think that the paper is open access:

 

http://www.nature.co...ncomms6631.html



#18 mike the wiz

mike the wiz

    Veteran member

  • Veteran Member
  • PipPipPipPip
  • 3,375 posts
  • Gender:Male
  • Interests:mikey mischief.
  • Age: 36
  • Christian
  • Creationist
  • England

Posted 19 January 2015 - 03:32 AM

 

 

 Seems you are stating some obvious points about confirmation and falsification evidence.

 

They have also found the mitochondrial ancestor of the first woman, dates back to about 6,000 years from what I hear.

 

Is mitochondrial DNA, "supernatural" in your opinion? because I am confused as to why in another thread you mentioned the "supernatural" and the scientific method as though I confused the two. 

 

As for evolution, they don't have any strong forensic evidence, so perhaps you meant that evolution is supernatural? Although that is the wrong word I would use for evolution, I would say a better word is, "fiction". Notice in your Richard example, both the forensics are intensely specific ruling out many possibilities, and the history is written.(Recorded history)

acigar.gif

Those two things make this very "different" science than the fairytale. The fairytale is much more tenuous, for example whale-evolution would not be based on specific DNA, it wouldn't be based on something as accurate as matched deformites. This is what I wrote in that blog-entry Mattias, your "strong" case would be "strong" because of the match, like our case for animal-kinds is strong because of a match. When we find an "old" 200 million years or so, Coelecanth fish, and we find a living one that basically looks the same, we have a match, because deductively, a lineage MUST have lead from the historical version to the extant version, because they are identical, just like Richard was identical to his description.

 

You should read the blog carefully, as animal-kinds are a far stronger scientific case, because you can't alter their lineages. As coloured in the pink-zone, but evolutionary lineages are only 100% conjecture, and CAN be changed. But you CAN'T change the lineage for animal kinds because they MATCH.

 

http://creationworld...nimals-for.html



#19 Mattias

Mattias

    Member

  • Veteran Member
  • PipPipPip
  • 180 posts
  • Age: 43
  • no affiliation
  • Atheist
  • Sweden

Posted 19 January 2015 - 04:30 AM

Mi

 

 Seems you are stating some obvious points about confirmation and falsification evidence.

 

 

 

Yes, and that was all that my post was intended to do. Present some neutral examples of asymmetric hypothesis testing. As for the rest of your post, it is a mess of incoherent statements that I cannot even respond to. You should stop writing these broad responses where you just throw in random stuff, and instead focus on specifics.



#20 mike the wiz

mike the wiz

    Veteran member

  • Veteran Member
  • PipPipPipPip
  • 3,375 posts
  • Gender:Male
  • Interests:mikey mischief.
  • Age: 36
  • Christian
  • Creationist
  • England

Posted 19 January 2015 - 05:38 AM

:acigar:






0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users