MemberlistThe Library Index  FAQFAQ   SearchSearch   RegisterRegister   ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 
PRESUMPTIVE LOGIC (APPLIED EPISTEMOLOGY)
Reply to topic Goto page 1, 2, 3, 4, 5, 6, 7, 8, 9  Next
View previous topic :: View next topic  
Ishmael


In: Toronto
View user's profile
Reply with quote

Applied Epistemology is "The Science of Presumptive Logic". Thus it is a methodology that consists of certain rules governing what must be presumed in the absence of evidence to the contrary. It also consists of several tests for evaluating the strength of an hypothesis or theory.

This section of the Applied Epistemological library is devoted to the formalization of the rules and tests that define the nature of our method.

------------------------
Addendum (added in response to the discussion):

Applied Epistemology also offers a critique of the psychology and sociology of knowledge acquisition, using the principles uncovered by this critique to identify theories and disciplines likely suffering paradigm error.
Send private message
Ishmael


In: Toronto
View user's profile
Reply with quote

The principle of "presumptive logic" means that every rule of Applied Epistemology is subject to the caveat, "except when it isn't."

Thus, three of our more familiar and fundamental rules should be understood as,

  • The truth is always simple, except when it isn't.
  • What is is what was, except when it wasn't.
  • Same cause, same effect, except where it differs.


These caveats are typically omitted when writing the rules, else they provide a too-convenient escape hatch for those who would not be bound by the rule. The rules list those presumptions by which our models are bound where substantial data to the contrary is lacking. Those who would violate the rules are saddled with the burden of proof.

  • If the truth is not simple, prove it!
  • If what was differs from what is, prove it!
  • If different inputs will produce the same outcome, prove it!
Send private message
Ishmael


In: Toronto
View user's profile
Reply with quote

Presumptive logic means that, even where we know nothing, we already know a lot.
Send private message
Mick Harper
Site Admin

In: London
View user's profile
Reply with quote

The other version is "The more you know, the more you are bound by what you know."
Send private message
Ishmael


In: Toronto
View user's profile
Reply with quote

In no way do these rules eliminate the possibility that the solution to a problem may be complex or that the same result might be produced by completely different actions -- no more than they require the past to be a duplicate of the present. The rules simply alert us to the default state of things we are required to presume absent substantial evidence to the contrary ("bone-chilling" or of lesser quality).

Applied Epistemology is quite clear that we have no choice in the matter; we must presume a default state as specified by the rules of Applied Epistemology. There is no such thing as a blank and empty unknown. There is only the space occupied by evidence and the space occupied by a presumptive state.
Send private message
Ishmael


In: Toronto
View user's profile
Reply with quote

Mick Harper wrote:
The other version is "The more you know, the more you are bound by what you know."

Or as I have written it, "The more you know, the less you think", or, "Knowledge is the enemy of thought."

Ignorance is like virginity. Its preservation is a virtue.
Send private message
Ishmael


In: Toronto
View user's profile
Reply with quote

You raise an important oversight in my original definition of Applied Epistemology. I omitted the critique Applied Epistemology affords to the psychology and sociology of knowledge acquisition, the most famous expression of which is the notion of the "careful ignoral".

I shall add this to the original post.

I will update all definitions for all rules and principles in light of the evolving discussion.
Send private message
Grant



View user's profile
Reply with quote

Is it possible to recast AE using Bayes' Theorem?
Send private message
Ishmael


In: Toronto
View user's profile
Reply with quote

You shall have to explain it and its implications.
Send private message
Wile E. Coyote


In: Arizona
View user's profile
Reply with quote

Bayes theory aint gonna be of interest unless you are really worried about your false positives or negatives.

You are worried about HIV.

You are not gay, injecting or another risk group.

You have a test.

It shows you have HIV.

The test is say 99% accurate.

Dont panic yet.

It is still likely to be a false positive.

This is of course counterintuitive.

If you used Bayes you would understand why and you can predict other counter intuitive results. Social scientists who dont understand Bayes often mangle their stats as they cant spot a likely false positive.
Send private message
Mick Harper
Site Admin

In: London
View user's profile
Reply with quote

Yes, this sounds like what we are up against all the time. The soft 'sciences' (eg geology, archaeology, linguistics) just don't teach statistics at the individual level and certainly do not apply it at the subject level. The kind of 99% false positive finding that you describe would have been entered as "observable fact" a hundred years ago, taught as such ever since, and never investigated further by de facto fiat.

If you have been observing, for instance, our hill fort discussion you will realise that a figure as high as 99% is out of the question! 'In the majority of cases selected for investigation' is more likely.

Needless to say, Applied Epistemologists are as ignorant of Bayes Theorem (and similar statistical techniques) as their opponents, it's just as revisionists we do not carry the weight of the world's knowledge on our shoulders so can afford to be relatively insouciant about being wrong.
Send private message
Wile E. Coyote


In: Arizona
View user's profile
Reply with quote

An expert in Bayes Theorum was called by the defence in the sucessful aquittal of Sally Clarke. She was previously found guilty of murdering her two kids.

No forensic evidence.No social services involvement.

Sir Ray Meadows obviously not a statisticial genius, quoted a figure of a 1 in 73 million chance of a single family suffering 2 cot deaths.(implication being that mum murdered them)

This is known in the trade as "the prosecutors fallacy" As you read this you are probably thinking, "Ok thats a bit high....but even it was 1 in a million she probably did it....must have been guilty"

If you understood Bayes you would realise why the judge and jury got it so terribly wrong.
Send private message
Mick Harper
Site Admin

In: London
View user's profile
Reply with quote

A somewhat similar story has just emerged, the Colin Norris case (transmitted last night)

http://www.bbc.co.uk/programmes/b009wc83

I watched it off the Digibox this afternoon and was untterly dumbfounded. Sorry, can't say whether Bayes Theorem is involved, would appreciate advice.
Send private message
Hatty
Site Admin

In: Berkshire
View user's profile
Reply with quote

Wile E. Coyote wrote:
If you understood Bayes you would realise why the judge and jury got it so terribly wrong.

I went out with a judge who'd known Sir Roy and he told me that when he was trying cases he would become an expert himself in the relevant field. It would seem that high court judges imbibe prevailing orthodox wisdom to prepare for such trials in which case would their judgement be unbiased?
Send private message
Wile E. Coyote


In: Arizona
View user's profile
Reply with quote

Sir Roy will for ever be a case study in the misapplication of statistics. He was the "expert" witness.

There were quite a few other women who ended up behind bars cos of Roy.

Wiki has a bit of gossip.

"In 2004 Meadow's ex-wife, Gillian Paterson, accused Meadow of seeing 'mothers with Munchhausen Syndrome by Proxy wherever he looked', and implied that he was a misogynist: 'I don't think he likes women... although I can't go into details, I'm sure he has a serious problem with women.' The article also revealed that Meadow had starred in an amateur production of The Crucible by Arthur Miller, playing Judge Danforth who falsely and recklessly accuses women of witchcraft and child killing and sentences them to death. Meadow confided to a friend that 'he found it an uncomfortable part because he identified with this judge more than he was happy with'.


There are a number of women behind bars for force feeding their children on salt......

Meadow wrote a 1993 paper on non accidental salt poisoning.
Send private message
Display posts from previous:   
Reply to topic Goto page 1, 2, 3, 4, 5, 6, 7, 8, 9  Next

Jump to:  
Page 1 of 9

MemberlistThe Library Index  FAQFAQ   SearchSearch   RegisterRegister   ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 


Powered by phpBB © 2001, 2005 phpBB Group