andrewducker: (Default)
[personal profile] andrewducker
I’m not ruling out unicorns.

It is impossible to prove that something does not exist.  Unless you have looked everywhere, you can’t be 100% sure that whatever you were looking for doesn’t exist.  And if you did look everywhere then you couldn’t be sure that it hadn’t walked around behind you, making faces behind your back (as I believe the Loch Ness monster does every time they run sonar up and down the loch).

Not only that, but there’s always the possibility of things we don’t know how to detect (like, say, neutrinos until recently), or things that look like something else, or even the sceptical equivalent of the atom bomb – we could be brains in a jar (although I believe that nowadays the brain in a jar gets to wear cool clothing and learn Kung-Fu, so it’s not all bad).

So, given that we can’t disprove the existence of unicorns, it therefore makes perfect sense that if someone declares that Black Holes could be full of unicorns that we give them the benefit of the doubt.  After all, we can’t see into a black hole, there could be _anything_ in there, couldn’t there?

But you’ll excuse me if I don’t believe it.  Because if I believe it (and I’m not even talking about certainty here – not 100% pure, positive, unshakeable belief, just your common-or-garden, “My history teacher said it so it must be true”) then by the same reasoning I have to believe there could be pixies in there or an army of Nicole Kidman clones or Great Cthulu himself.

Eventually, when you get right down to it, all we have to go on is what seems right to us.  We can’t have definitive proof, we can’t be completely sure; there will always be room for scepticism on anything before the basic Cogito Ergo Sum (and even that is arguable, for certain values of “Ergo” and “Sum”).  This means that we are left to pick and choose based on what fits into our worldview.

Which doesn’t mean that you’ve got free rein to decide that the sky is yellow, cats bark and Paris is near Tokyo, because believing things that are clearly contradictory to your perceptions will quickly make it impossible to deal with anyone not living in your personal world.  But when you discover that the US is invading Iraq (something for which we have clear evidence) then it’s your worldview that tells you that whether he’s doing it for the oil, to free Iraqis, to stabilise the Middle East, because Saddam had his dad shot, to bring about the Rapture or because Vladimir Putin bet him he wouldn’t.  The evidence is always open for interpretation.

Evidence varies in how much interpretation it allows; experiments have shown that witnesses are far less accurate in their perceptions than used to be thought.  People fill in information from a few glimpses, taking in a few cues and then adding layers of detail based on their expectations.  We use shortcuts to leap to conclusions, choose evidence selectively to prove whatever points we are attached to and then remember the stories we tell as the truth.  Once we’ve repeated a story a few times as if it were truth it becomes very hard to remember the original embellishments we applied.

Given all of this, deciding what fits your worldview is fraught with peril.  How can you tell that you’re not just playing to your own prejudices?  For instance, I have a liking for superstring theory because it plays to my appreciation for simplicity.  But I wouldn’t say I believed it any more than one of the other quantum theories because there’s no decisive evidence for it.  When it comes to deciding what to believe I am generally very happy to turn to science.  But why should science be privileged over, say, Christianity?

When you learn science at school you largely conduct experiments and then interpret them.  This leads people to think that these two stages are what make up ‘science’.  And if that’s all there is to it – do something, decide what it means – then why should the opinion of said scientist matter any more than yours when it comes to announcing what’s really going on? 

The answer is that science demands more than that – the scientific method means that when you come up with a hypothesis that you design an experiment whose results will be different if your hypothesis is true than if it is not.  If a scientist thinks that the speed of light varies depending on height above sea level, then they have to design an experiment that measures it at various heights and show the difference.  They also have to defend the design of his experiment against other scientists who think something different.  And when, 50 years later, someone designs an experiment that shows that the speed of light varies depending on air pressure, not height, they have to face up to the fact that their work was merely transitory, giving an approximation to reality that’s been superseded.

Science isn’t a doctrine of facts to be believed.  It’s a method of discovering approximations to reality, of constructing models of reality, of producing rigorous evidence for a particular view of reality.  It’s not perfect, of course, no system ever is, especially not one that involves people.  Scientists hold onto theories that are terribly outdated, attack each other in print (and occasionally in person) and occasionally behave like spoiled brats.  But when the evidence is sufficient the vast majority of scientists can be persuaded and minds can be changed.

Scientifically you don’t have much of a theory unless you have some way of differentiating it from other possibilities.  If you don’t have a way of looking into the black hole to see whether there are unicorns inside then saying that there might be is just meaningless supposition.  It may feel right to you.  You may have a gut feeling that they are there.  But unless it’s possible to look in some way, assuming their existence is taking a step out of rationality.

So, returning to the original example, when people say “But black holes could be caused by unicorns.” I look at whether this fits in with whatever I already know, I factor in whether there is any new evidence for the claim, whether people whose judgement I value (either personally or by reputation) agree with it, whether the people making the claim have any obvious ulterior motive for doing so and whether there is any way of checking to see if the claim is correct.

So, having never seen any proof of unicorns, lacking any credible people in favour of the claim, noticing that the unicorn claimants have a strong emotional attachment to the existence of unicorns and that there’s no way of telling if there are unicorns in the black hole, I don’t consider there to be any reason to add the claim to my list of beliefs.  Which doesn’t mean I’m ruling it out permanently; many claims have seemed incredible, weren’t believed at first and were uncheckable at the time later turned out to be valid.  But this doesn’t mean that I’m willing to commit to it at any point before there is a strong reason to do so.

This was, by the way, inspired by the post on the journal of [livejournal.com profile] green_amber here.

Date: 2004-05-06 11:10 am (UTC)
From: [identity profile] in-thy-bounty.livejournal.com
I'd like to be conquered by an army of Nicole Kidman clones.

Date: 2004-05-06 11:15 am (UTC)
From: [identity profile] green-amber.livejournal.com
Need lj-cut babe.

OK, as pre advertised

So, having never seen any proof of unicorns, lacking any credible people in favour of the claim, noticing that the unicorn claimants have a strong emotional attachment to the existence of unicorns and that there’s no way of telling if there are unicorns in the black hole, I don’t consider there to be any reason to add the claim to my list of beliefs

But this is EXACTLY how I fel about strong AI (the belief that machines can develop intelligence/"consciousness" like that of humans).

a) I've never seen any proof. we can both agree, I think, that as yet no one has developed anything that loks remotely like a machine with consciousness.

b)Credible people? well, who's credible in the AI camp? The entire history of the discipline Searle and Minsky and al the rest of em have been disagreeing with each other. I'm not sure that "credible witness" is a useful criterion to add here as any domain where we are looking at the evidenmce will have conflict and disciples either way.

c)noticing that the unicorn claimants have a strong emotional attachment to the existence of unicorns

We discussed this. It seems apparent that adherents of strong AI have emotional investments in it which transcend the emotional attachment people had/have to, say, believing we could isolate stem cells, or create transuranic elements or identify dark matter or other fields of scientific discovery/enquiry.

Wha are the emotional investments in it for adherents of strong AI? It's for people who feel they're not quite part of the human race - geeks at the Aspergers end again, I'm afraid. The human race either fails them - or rejects them - so they look fo a better substituite in artificial intelligence. The "Weird Science" effect, as I jokingly called it - girls don't go for me cos I'm a geek so I'll build my own robot gal. It's a sort of mishumanism and misogyny and tho subliminal, it upsets me.

And then there's transhumanism ,which translates as, "oh my god we're all going to die!" If you have strong AI and the brain is all there is - not a mind that is or has something seperate - then being able to put the mind into an eternal silicon home is obviously possible. This is very attractive for some of the current aging commentators. Basically, it's about (understandable )fear of death, plus (equally understandable) rejection of anything spiritual remaining after death.

d) there’s no way of telling if there are unicorns in the black hole

Well in this example this either doesn't apply, or I refer you to (a). There are as yet no conscious AIs and no way I can test to see if future machines might have anything like it.

So I put it to you that by your own criteria there is no reason why I should believe that a belief in strong AI is a ratinal scientific belief as opposed to a credo, a religion.

Or to use your own summary:

It’s (scence) is a method of discovering approximations to reality, of constructing models of reality, of producing rigorous evidence for a particular view of reality.

That rigorous scientific evidence for strong AI is where?? You started with the presumption it was correct and expected me to defend my belief to the contrary. I don't think I'm required to by rules of rationalaity.

Otherwise, vg. Now we wait for the votes of the Luxemburg jury!

Date: 2004-05-06 02:06 pm (UTC)
From: [identity profile] green-amber.livejournal.com
So what we end up with is that neither of us has proof, either that our own instinct is right or the other's instinct is worong. That's fine with me. Is it fine with you? It's whether one of these positions is "presumptively" right that bothers me.

Date: 2004-05-06 02:24 pm (UTC)
From: [identity profile] green-amber.livejournal.com
I'll take that as a no then :->

Back to bed..

Date: 2004-05-06 02:26 pm (UTC)
From: [identity profile] green-amber.livejournal.com
Also..

I never demanded the existence of anything.

What I said was that you had no evidence that consciousness was *not* more than simply the sum of the physicality of the human brain.

You seem to have admitted exactly that. No?

Date: 2004-05-06 02:53 pm (UTC)
From: [identity profile] green-amber.livejournal.com
So what we end up with is that neither of us has proof, either that our own instinct is right or the other's instinct is worong. That's fine with me.

Date: 2004-05-06 02:54 pm (UTC)
From: [identity profile] green-amber.livejournal.com
oops, score this bit - either that our own instinct is right or - OK?

Date: 2004-05-07 01:30 am (UTC)
From: [identity profile] green-amber.livejournal.com
OK! I give in! Just get me some Lem Sip wil you :-)

Date: 2004-05-06 12:49 pm (UTC)
From: [identity profile] blue-condition.livejournal.com

But this is EXACTLY how I fel about strong AI (the belief that machines can develop intelligence/"consciousness" like that of humans).

a) I've never seen any proof. we can both agree, I think, that as yet no one has developed anything that loks remotely like a machine with consciousness.



I still believe that the "Strong" and "A" in "Strong AI" are just noise. It's incredibly chauvinist to assume that the only solution to intelligence is slavish mimicry of a set of biological mechanisms we don't fully understand. It's Cargo Cult science of the worst stamp - I have this vision of a bunch of grass-skirted AI researchers at the consoles of cardboard Lisp machines praying to some John Frum-esque Minsky figure ;)

Date: 2004-05-06 12:15 pm (UTC)
From: [identity profile] steer.livejournal.com
It is impossible to prove that something does not exist.

I can cite you convincing proofs that the highest prime number does not exist if you like.

Date: 2004-05-06 12:30 pm (UTC)
From: [identity profile] steer.livejournal.com
It depends what you are taking as proof. Once outside the rigorously realms of logic, philosophy and mathematics, nothing can really be "proved".

We can "prove" Formula One cars manufactured by Arrows GP post 2003 do not exist with a much greater certainty than we can "prove" that (say) Newtonian gravity is an accurate model for low speeds high mass bodies. [In the latter case we can say that no exceptions have been observed so far except by unreliable witnesses, in the former case we can cite evidence as to where the team personel were and they couldn't possible have time.]

It's hard to "prove" anything in the real world - existance or non-existance. I guess you can accept that if someone shows you something then it most likely exists - but it could all be smoke and mirrors. [After all, despite being shown the proof, I don't believe David Copperfield can make tigers vanish.]

Date: 2004-05-06 12:41 pm (UTC)
From: [identity profile] steer.livejournal.com
No - you said you can't prove something doesn't exist. An arrows formula one car manufactured post 2003 is a concrete thing which we can prove doesn't exist with the standards of proof you would accept to be fine for proving physical theories.

Date: 2004-05-06 01:00 pm (UTC)
From: [identity profile] steer.livejournal.com
So what is proved permanently?

It could well be that something you previously thought was "proved to exist" (say Edinburgh castle) turned out to be an elaborate hoax with projectors and stage sets.

I think there's something a bit wrong with the concept of "proof" here.

You don't mean proof in the mathematical sense nor in the legal sense. You seemed to be asserting that it was impossible to prove an absence but an absence can sometimes more easily be proved than a presence. [It is easier for me to show there is no unicorn in my living room than it is for me to show that my fourth jeweller's screwdriver is in my living room -- but I'm bloody sure the little bugger is somewhere.]

Date: 2004-05-06 01:18 pm (UTC)
From: [identity profile] steer.livejournal.com
Which makes your assertion that you can't prove something doesn't exist baffling.

Date: 2004-05-06 04:31 pm (UTC)
From: [identity profile] steer.livejournal.com
It is merely harder to "prove" such a thing doesn't exist. Assuming we have some minimal information about the thing - let us say that it is a green swan. We begin with a search area and examine every swan in that area to see if it is green. We separate the previously searched area with something which is not permeable to swans and, eventually, we will have searched the entire area in question. Note that if the area is the universe this may be difficult.

Of course you might hypothesise I guess that green swans could spontaneously materialise in a green swan free area. But if what we're searching for is ordinary common or garden non-mystical green swans then it is merely extremely difficult and time consuming.

Date: 2004-05-07 03:07 am (UTC)
From: [identity profile] steer.livejournal.com
These objections (or in the case of two a variant) also bear on the problem of proving a white swan exists.

1) There may be no white swans and your imperfect detection mechanism makes you think there is.

2) A researcher could be painting the green swans white to make you think there are white swans.

3) You may be a brain in a jar in a totally swan free world being fooled into thinking there are swans.

Sorry to harp on - but my point remains, the problem of proving an absence is merely a variant of the problem of proving a presence albeit a somewhat more difficult variant.

Date: 2004-05-07 12:15 pm (UTC)
From: [identity profile] steer.livejournal.com
I think we're going to have to agree to disagree. You can't prove there will *never* be a swan if that is what you mean -- in the same way you can't prove there will always be a swan.

Date: 2004-05-07 12:24 pm (UTC)
From: [identity profile] steer.livejournal.com
(grin) Indeed...

Date: 2004-05-07 01:45 pm (UTC)
From: [identity profile] green-amber.livejournal.com
There's no swans without moorhen..

Date: 2004-05-08 06:56 am (UTC)
From: [identity profile] drainboy.livejournal.com
When people say "X exists" they are in fact saying "X exists here and now", when they say "X does not exist" they are normally saying "X does not exist anywhere, now" or, in some circumstances "X does not exist anywhere, anywhen". You define "X does not exist" (quite properly as the opposite of "X exists") as "X does not exist here and now".

("Here" and "now" being some constraints not a precise point in space-time)

If I'm reading this right, the two of you aren't arguing the same point.

If you can both agree on a definition of X and that X is defined as "X as we can reasonably detect it" and the constraint of X not existing is a searchable area then "X does not exist" is proveable as trivially as X exists. I don't think you're agreeing though ;)

Date: 2004-05-06 12:45 pm (UTC)
From: [identity profile] blue-condition.livejournal.com
You describe my position perfectly. Thanks.

Date: 2004-05-06 02:09 pm (UTC)
From: [identity profile] green-amber.livejournal.com
And this is the difference between your friends and my friends :-> (Yes I'm just Old.)

Date: 2004-05-08 02:30 am (UTC)
From: [identity profile] spaj.livejournal.com
excuse me?

Date: 2004-05-07 05:48 pm (UTC)
From: [identity profile] stillcarl.livejournal.com
I looked, and got the following...

HELLO NON-DONATING BANDWIDTH LEECHERS.
THIS SITE IS NOT EASY TO KEEP ALIVE WITH OVER 20,000 VISITORS A DAY
AND NO DONATIONS.
YOU LOVE THE "BANANAPHONE"...
THEREFORE, BUY SOME CRAP.
IT WILL KEEP THE SITE ALIVE, AND YOU HAPPY WITH YOUR "BANANAPHONE" MERCHANDISE.

Thank you,
W.T. Snacks


So, still don't believe. :)

Date: 2004-05-08 02:30 am (UTC)
From: [identity profile] spaj.livejournal.com
Well, that's like the bible? It's an indication that there was a god, but you just have to believe? ;)

And it still works for me. :D

Adam

Date: 2004-05-06 06:25 pm (UTC)
darkoshi: (Default)
From: [personal profile] darkoshi
Even if a unicorn were found somewhere on this planet, it would probably seem to me like not so much a real unicorn as just a horse with a horn on its head.

And if I weren't so tired, I might have had a more intellectual comment for you, but as it is, some thought/memory just briefly flickered and.... too tired to... probably not related to your topic anyway. eh.

Your views and GreenAmber's don't seem contradictory to me. But then again, maybe I'm too tired to see it.

Date: 2004-05-07 04:29 am (UTC)
From: [identity profile] stillcarl.livejournal.com
the scientific method means that when you come up with a hypothesis that you design an experiment whose results will be different if your hypothesis is true than if it is not.

So, is a cat conscious or not? How would you design an experiment to find out?

Date: 2004-05-07 05:44 pm (UTC)
From: [identity profile] stillcarl.livejournal.com
Simple - you ask [livejournal.com profile] green_amber for that bit. ;-)

Date: 2004-05-07 07:24 pm (UTC)
From: [identity profile] stillcarl.livejournal.com
Less simple answer...

Most people would claim they know what consciousness is and they themselves are conscious. That they can't come up with a definition of consciousness to test for consciousness in others gives rise to three possibilities (at a quick count)...

1) Consciousness actually doesn't exist.

Given I find my own consciousness as real as this keyboard I'm typing on, (in fact I find my consciousness more real, given a keyboard may appear in one of my dreams), I reject this.

2) Consciousness does exist, as could a test for it in others, but we've just not figured out what the test is yet.

This I'm unsure of, in that I don't know if anyone claims to have a valid test for consciousness. Does anyone know of any? (Okay, it must be over a day since this thread started so anyone's probably left the building...)

3) Consciousness does exist, but only the conscious can detect it, and only in themselves. ie, a test for consciousness in others is impossible.

Instinct suggests to me that this is probably the truth of the matter. Which doesn't mean it's outside of nature - just that the test for consciousness is "Are you conscious?" addressed to yourself.

Date: 2004-05-08 04:42 am (UTC)
From: [identity profile] stillcarl.livejournal.com
2. Though I can imagine consciousness without sensation.

Date: 2004-05-08 06:43 am (UTC)
From: [identity profile] drainboy.livejournal.com
In the same way you would design an experiment to decide if a cat was hungry or not. First work out the behavioural expression of hunger in a cat based on your own experience of hunger and your projection of the associated behaviour onto cats. Then decide on an evidence driven delineation between "hunger" and "not hunger" in the subject, then observe the cat. I don't think there's any fundamental difference between behaviour due to hunger and behaviour due to consciousness (except the expression of the behaviour due to the phenomenon in question) it's all subjective definition of phenomena by experience.

Perhaps there is a difference due to the validity of projecting hunger behaviour onto non-you things and the validity of projecting consciousness behaviour onto non-you things, but that's all hugely subjective and presumes some specialness about consciousness, which I think is a fallacy.

September 2025

S M T W T F S
  12 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21222324252627
282930    

Most Popular Tags

Style Credit

Expand Cut Tags

No cut tags
Page generated Sep. 20th, 2025 01:43 pm
Powered by Dreamwidth Studios