?

Log in

Previous Entry | Next Entry

controversial topics in science

I recently asked Lubos Motl what he thinks the difference is between the string theory debate and the global warming debate, since in the former he is vehemently in favor of the majority opinion among experts, and in the latter he is vehemently opposed to it. See the "fast comments" after this post of his, where he responds to my inquiry with 4 long comments. I'm not sure whether I'm going to say anything back, because I honestly asked out of curiousity and what he says does make some amount of sense. But anyway, this inspired me to create this poll to see what other correlations or anti-correlations there are between different positions people take on certain controversial topics surrounding science...

Poll #953125 controversial topics in science

Is string theory the most promising route to understanding quantum gravity?

yes, and it's the only route that should be taken seriously at this point
1(4.5%)
yes, but it's good that we are dedicating some resources to exploring alternatives such as loop quantum gravity
4(18.2%)
yes, but string theory gets too much attention and some string resources shoud be redirected towards alternatives
1(4.5%)
it may be the best, but that's not saying much--the real answer is most likely something nobody has thought of
5(22.7%)
no, there are better alternatives which have been proposed
3(13.6%)
I don't know enough about this issue to even guess
7(31.8%)
I don't like any of the choices
1(4.5%)

Is global warming a crisis?

yes, and it's so bad that we're past the point where we can do anything about it
2(9.1%)
yes, and if we don't do something about it now, we're all screwed
6(27.3%)
yes, but we still have some time before we have to deal with it
2(9.1%)
global warming is a problem, but not bad enough to be a crisis
3(13.6%)
global warming is real, but it may or may not be a problem, and could have more positive effects than negative
1(4.5%)
the temperature shifts over the past century are primarily natural, and not due to human activity
0(0.0%)
both E and F
5(22.7%)
global warming is a myth made up by the liberal media, and has no scientific basis
0(0.0%)
I don't know enough about this issue to even guess
1(4.5%)
I don't like any of the choices
2(9.1%)

Evolution is best described as a

fact
6(27.3%)
an extremely well-tested theory, which has been confirmed beyond a shadow of a reasonable doubt
15(68.2%)
consensus
0(0.0%)
opinion
0(0.0%)
a fairly well-tested theory, but many loopholes remain that need further testing in order to confirm it
1(4.5%)
just a theory, and the evidence supporting it is pretty weak
0(0.0%)
poor alternative to the real creation story, which is accurately told in the bible
0(0.0%)
evil concoction by atheists to undermine religion
0(0.0%)
I don't know enough about this issue to even guess
0(0.0%)
I don't like any of the choices
0(0.0%)

When should robots be given the same rights as human beings?

As soon as they exhibit consciousness, which will most likely happen within the next 50 years
4(18.2%)
As soon as they exhibit consciousness, which will most likely happen in 50-100 years
2(9.1%)
As soon as they exhibit consciousness, which most likely won't happen for at least another 100 years
2(9.1%)
They will most likely exhibit consciousness in the next 100 years, but we should not given them any rights when this happens
0(0.0%)
Robots will be able to do everything humans can do within the next 100 years, but they will never truly be conscious, and therefore should not have rights
0(0.0%)
Robots will not be able to do everything humans can do for at least 100 years
4(18.2%)
Robots will never be able to match certain things humans can do
2(9.1%)
I don't know enough about this issue to even guess
2(9.1%)
I don't like any of the choices
6(27.3%)

Comments

( 22 comments — Leave a comment )
onhava
Mar. 25th, 2007 01:38 am (UTC)
Annoying question: what is the difference between "confirmed beyond a shadow of a reasonable doubt" and "fact"? Evolution, surely, is both.

I'm also kind of surprised that I'm the one taking the strongest pro-string-theory position. But you did make it a question about quantum gravity -- I would suggest that there are too many people studying quantum gravity in general, but those who are should probably be doing string theory.
spoonless
Mar. 25th, 2007 02:04 am (UTC)

what is the difference between "confirmed beyond a shadow of a reasonable doubt" and "fact"? Evolution, surely, is both.

Yeah, there is not much difference. Part of the reason I included them both as choices, is to see who would go for one versus the other (in other words, to get an idea of whether people believe there is a difference.)

As I was writing the poll, I was planning on choosing "fact" myself, and at the last minute decided to instead choose the second answer. (It was close.) In casual conversation, I have no problem referring to evolution as a fact... in fact, I do it all the time, especially when speaking to someone who might be skeptical of it. :) However, I think there is some difference in that theories are supposed to be things designed to fit the facts, not facts themselves. In other words, the raw measurements are the facts, and while a theory can do very well at reproducing them (perhaps to the point where it's ridiculous to consider the possibility of alternatives), strictly speaking the theory never gets proven and is in some sense slightly less trustworthy than the data that went into generating it. So I chose B as a slightly better answer than A, even though I consider them both correct answers.

Regarding string theory, I meant to include another option, which was something like "quantum mechanics and gravity need not be reconciled", as there are (disturbingly) many who take this view. Somehow, I forgot to include it, and you can't edit polls after you create them. I guess I also should have had one that says it's too early to work on quantum gravity, since we can't really test things yet. Although I didn't think of that until just now, after reading your suggestion (about there being too many people studying quantum gravity).

Did you read Lubos's comments to me? One of his reasons for the difference in the two debates was that it's "objectively true" that climate scientists are the worst scientists in the world and string theorists are among the best. That, and other things, made me smile... as do most of his writings. What a strange and unique perspective, he has. It all makes sense in a weird way, and yet I never cease to be amazed/amused by what he's going to say next.
onhava
Mar. 25th, 2007 04:02 am (UTC)
Yeah, there are a few other ambiguities in the poll. "Robots will not be able to do everything humans can do for at least 100 years" is something I agree with, also. On global warming I maybe should have picked "I don't know enough about this issue to even guess," if I were being completely honest. I once made an effort to understand the evidence and its implications, and I was convinced at the time, but I was in high school, and I've learned a lot about how to judge the reliability of data since. I haven't really paid that much attention to climate change arguments in the last seven years.

Also, my answer for the string theory question isn't necessarily inconsistent with the right theory of quantum gravity being something no one has ever thought of; string theory does seem to be the only reasonable thing around currently, though. And I think whatever the right answer is, it remains true that string theory provides us with consistent theories of quantum gravity, and the real world probably has a lot in common with these. (I think a major step is missing, though. We need better cosmological solutions of string theory, preferably with positive cosmological constant. And probably we'll get some surprises if we had such a thing.)

As for Lubos's comments: your post is public, and he is likely to find it by noticing click-throughs to his blog.... I'm definitely amused by what he has to say, yes.
spoonless
Mar. 25th, 2007 08:24 am (UTC)

On global warming I maybe should have picked "I don't know enough about this issue to even guess," if I were being completely honest.

Yeah, me too. I considered putting that, because my opinions on it are pretty weak and mostly just meta-opinions based on how much I feel I can trust the different people saying different things; not based on my own analysis of the science, since I've never actually read a technical paper on the subject or seriously tried to understand it. But I put my best guess anyway, for the heck of it.

your post is public, and he is likely to find it by noticing click-throughs to his blog.... I'm definitely amused by what he has to say, yes.

Thanks for reminding me, I hadn't thought of that. I guess I haven't said anything I'd be too embarassed about him reading, and he wouldn't be the first professor to find my blog, but I should maybe be careful about who I talk about behind their back... especially if I want to apply for a Harvard post-doc in a couple years :)

Ah, the internet is such a fun place, isn't it?
easwaran
Mar. 25th, 2007 09:52 am (UTC)
Yeah, it surprised me when Dave Chalmers found 2hot4philosophy, and when Greg Restall commented on my livejournal. But that stuff happens, and it's why I've decided to put a moving friends-only wall on my posts.
easwaran
Mar. 25th, 2007 09:56 am (UTC)
In other words, the raw measurements are the facts, and while a theory can do very well at reproducing them ... strictly speaking the theory never gets proven and is in some sense slightly less trustworthy than the data that went into generating it.

There's something right about that, but I don't think there is such a thing as a raw measurement. As they say, all observations are theory-laden. We don't have absolute certainty that the measuring device actually worked as it was supposed to, that we read it right, or even that we remember our visual impression right. I think there's no level at which we arrive at absolute certainty. And I'm far more certain of evolution than I am of some observations I've made (like say, the observation that my feet are size 11).

Also, there's a bit of a confusion there, since fact is a metaphysical description, while well-tested theory is an epistemic one. Presumably, most (or at least many) well-tested theories are also facts, but there are plenty of facts that have not yet been well tested.
spoonless
Mar. 25th, 2007 06:09 pm (UTC)

I think there's no level at which we arrive at absolute certainty.

Well, I would certainly agree with that. :)

And I'd also somewhat agree when you say "there's no such thing as a raw measurement", but in both cases, this seems irrelevant as all I was trying to do is to say that you should reserve the word fact for stuff that is close to the "raw measurement" side of the spectrum... stuff that has the least bit of interpretation going into it. And even though nothing has no interpretation, I think there is a fairly good distinction one can make here, for instance, between the fossil records themselves, and what we conclude from them. It's like seeing a whole bunch of dots that line up in a curve, and then filling in the curve.

However, the last two things you say here make me realize my usage of "fact" is probably pretty flawed and should be re-evaluated:

I'm far more certain of evolution than I am of some observations I've made (like say, the observation that my feet are size 11).

yes. I admit, I can think of many examples where I feel more certain of the theory itself than any particular data that when into generating it. So this appears to defeat a big part of what I was saying.

Also, there's a bit of a confusion there, since fact is a metaphysical description, while well-tested theory is an epistemic one. Presumably, most (or at least many) well-tested theories are also facts, but there are plenty of facts that have not yet been well tested.

And that's a good point too. Yes, a fact should be a fact regardless of who knows it or how well-tested it is. But also, another reason I think it doesn't make too much sense to call any theory a fact is because a fact is supposed to be one particular statement or truth, not a whole body of knowledge. So it seems you've defeated my argument for not calling it a fact, but it may be justified in a completely different way.

Actually, I think I could get around your metaphysical/epistemic distinction by saying that the issue of when we should start calling something a fact is still epistemic, even if whether it is a fact is metaphysical. However, as you've pointed out, we can be more certain of a theory than of any particular data supporting it. It looks like that really is a crushing blow to the distinction I was making.
easwaran
Mar. 25th, 2007 07:54 pm (UTC)
There certainly seems to be an interesting distinction to be made between statements about particulars and universals - but that seems to be a relatively weak point from which to argue against people who claim evolution as a fact, since they obviously aren't trying to make that particular distinction (unless they're severely confused about what the theory of evolution is, or about basic facts of the English language).

I think the temptation to distinguish between "fact" and "theory" is exactly what you say at the end - the question is when we call it a fact, rather than when it is a fact. I would think that would go with certainty and evidential support, which as you agree, destroys the force of that distinction here. Though I could also imagine someone arguing that calling something a fact requires a particular type of epistemic support that maybe we can't get for universal claims - though I think this would end up with serious anti-realist consequences.
spoonless
Mar. 25th, 2007 08:45 pm (UTC)

that seems to be a relatively weak point from which to argue against people who claim evolution as a fact, since they obviously aren't trying to make that particular distinction

Well, I hope nobody thought that by not choosing "fact" as the best answer I had any problem with calling it a fact other than semantic. As I mentioned earlier, I often call it a fact myself. And I think the answer to the question, which asked "evolution is best described as...", depends on who you're talking to. If I'm talking to the average person on the street, who usually has no idea what the real meaning of a word like "theory" is, I'm almost always going to use the word fact to refer to it, because using anything else is misleading. However, if I'm thinking about it for myself, or talking to someone who pays attention to and is interested in the subtle nuances of meaning (mainly, I'm thinking of philosophers here), I would make a more careful/different choice of words. I thought of this while I was making the question up, but decided to leave the audience ambiguous and just see how people responded.
thaumaturge
Mar. 25th, 2007 05:34 am (UTC)
Yeah that's what I thought, it's not so much a fact as it is an observation. Look, bacteria are adapting to our antibiotics! Is that MAGIC?

But, never underestimate the average human's ability to deny reality.
thaumaturge
Mar. 25th, 2007 05:28 am (UTC)
Why would anyone design and mass-produce a conscious robot? I HIGHLY DOUBT that it could ever happen by accident. It should be easy to design an advanced slave robot without self consciousness by simply not programming it with self-consciousness.
spoonless
Mar. 25th, 2007 06:53 am (UTC)

Why would anyone design and mass-produce a conscious robot?

There are already several groups working on it. It may be a bit early today, which is why they don't get too much funding, but the interest and funding in developing the software will surely go up once the hardware improves to the point where supercomputers exceed the computational power of the human brain (which is likely to happen within the next decade or two.)

I personally doubt it's possible to build a brain that can match human intelligence and creativity without it also having consciousness. It is probably possible to build one that is conscious, but lacks self-interest, or perhaps has some self-interest but a stronger desire to serve humans. However, the chances that everyone in the world with the technology to do so will refrain from designing one with self interest is slim to none. At the very least, the military will develop one which fights to survive, using whatever violent tactics it can employ.

A brief list of a few groups/people off the top of my head who are actively interested/involved in developing conscious machines (I'm sure there are many more):
http://www.singinst.org/aboutus/ourmission/
http://en.wikipedia.org/wiki/Hugo_de_Garis
http://www.novamente.net/agi/
http://web.media.mit.edu/~minsky/

Self aware robots have already been developed, although they are currently far from being fully conscious in the way humans are:
http://news.com.com/Researchers+unveil+a+self-aware+robot/2100-11394_3-6136608.html
thaumaturge
Mar. 25th, 2007 09:55 pm (UTC)
I didn't just say developing, I said mass producing. Do you think that the future Sony-Honda-Mitsubishi Corporation or whatever is going to make a consumer-priced robot butler that will need to be "conscious" to work? I don't think so, for two reasons.

First, our concept of human "consciousness" seems to be principally dependent on the organism having a complex social environment. This gives massive selective pressure on the organism to develop a sense of self, and extrapolate that to neighbours in order to better predict, understand, manipulate, and cooperate with them. Without this need, without a sense of self, the organism (or simulated organism) may have some kind of phenomenology, but it would be utterly shallow. The robot may be able to think, but it could never be aware of that fact because it cannot self-reflect. It could therefore never arrive at the conclusion "I am." Self-awareness and "consciousness" are not in any way synonymous.

Secondly, consciousness is not simply an "on/off" state. It seems like it should be entirely mechanistic, and lacking certain components, "consciousness" in its familiar human form simply would not arise. Without a capacity for self-reflection, without the perceptual programming to perceive and empathize with fellow creatures, etc. the resultant "consciousness" would be very different. I doubt that it will impact any capacity for creativity or "intelligence" either, at least in general. It may have difficulty with certain social problems, but worldly problem-solving doesn't seem to need this capacity.

There are major conceptual obstacles here. It seems clear to me that our sense of self is an entirely artificial abstraction which severely limits the usefulness of any introspective processes for understanding these things like "intelligence" and "creativity" and "consciousness." Therefore, I don't think we can make any meaningful intuition about how these things "work" without exacting physical observation of the organ that creates it. An unimaginably detailed computer model of a human brain should theoretically give rise to normal human consciousness, but we are so far away from this. We understand basic principles, but even the most advances neuroimaging techniques are incredibly crude. Cataloguing functions dependent on certain brain areas is a start, but even with a specific localization like "the ventromedial prefrontal cortex is necessary for empathy," the VMPFC is still an incredibly complex chunk of cortex with a vast number of interconnections with the rest of the brain. General models may help, but I think the answers are in the details, which is something we have barely scratched the surface of. A reasonably detailed model of the brain in 50 years would be an optimistic estimate, with 100 years more likely, in my opinion.

Note: All opinions stated above are subject to revision without notice!
spoonless
Mar. 25th, 2007 10:56 pm (UTC)

I didn't just say developing, I said mass producing.

I don't see what you're getting at with that. Yes, it seems like mass producing fully conscious robots would be extremely expensive, and so that's probably further off. But I would assume it's just as important to grant conscious robots right, whether there are only 5 of them or 5 million of them. Why would it matter how many of them there are? The fewer there are, the harder they are going to have to work to overcome biological prejudice... but this is a question of should, not will, they have rights.

The robot may be able to think, but it could never be aware of that fact because it cannot self-reflect. It could therefore never arrive at the conclusion "I am." Self-awareness and "consciousness" are not in any way synonymous.

You appear to be equating self-reflection with consciousness in the first sentence... but then you admit in the second that they are not equivalent. I would agree that they're not equivalent, but I don't agree that it would be hard for a computer to self-reflect. That seems like the easy part... which for the most part, we already have working now. The harder part is to develop a more full consciousness where there are a lot of different interconnected awarenesses that all support and reinforce each other. This we're nowhere close to yet, but I do think we can do it in the next hundred years, and probably in the next 50.

Secondly, consciousness is not simply an "on/off" state. It seems like it should be entirely mechanistic, and lacking certain components, "consciousness" in its familiar human form simply would not arise.

I certainly agree that it's not anywhere close to an on/off state. There are nearly as many different levels, types, and shades of consciousness as their are beings on the planet. I don't think I follow the sentence after that. Are you saying you see robots as mechanistic and consciousness as non-mechanistic? I see them both as mechanicstic (to the extent that that word has meaning, since I see everything as mechanistic).

It seems clear to me that our sense of self is an entirely artificial abstraction which severely limits the usefulness of any introspective processes for understanding these things like "intelligence" and "creativity" and "consciousness."

definitely agreed there

An unimaginably detailed computer model of a human brain should theoretically give rise to normal human consciousness, but we are so far away from this.

Yes, very far. Although stuff like IBM's Blue Brain project gives me hope (http://bluebrain.epfl.ch/)
omnifarious
Mar. 25th, 2007 09:38 am (UTC)

I think string theory has very poor predictive power and is therefore useless as a theory. I think the answer is most likely something nobody has yet thought of, especially given some of the bizarre data we're getting from cosmology studies.

The most compelling evidence I see for human being affecting climate are the plots of historical temperature vs. historical CO2 levels. The fairly tight correlation seems to me indicative that the current record CO2 levels are highly likely to have an effect on the global temperature.

And I agree with you about the difference between fact and theory. A fact is a piece of observable evidence. It says nothing about cause and effect or mechanism. It's just there. Evolution is a theory, not a fact. An extremely well tested theory, but a theory.

I don't know if we will ever have conscious robots who's will isn't enslaved to their builders.

easwaran
Mar. 25th, 2007 09:57 am (UTC)
See my comment above about the distinction between fact and theory.
spoonless
Mar. 25th, 2007 08:11 pm (UTC)

The fairly tight correlation seems to me indicative that the current record CO2 levels are highly likely to have an effect on the global temperature.

Right, I've seen the same plots. But the people who criticize global warming (at least the ones who know what they're talking about) usually say that cause-and-effect are reversed. In other words, that increased temperatures have caused the CO2 levels to rise, and when they decrease again, they will go back down. I think mechanisms are known for cause and effect going both ways, so the question is which effect is more important.
omnifarious' OpenID account [omnifarious.org]
Mar. 25th, 2007 08:26 pm (UTC)

Yeah, I was thinking about that.

Personally I'd really like to avoid gambling with the global climate and error on the side of caution. We only have one Earth at this point. No sense in running that kind of experiment if we can avoid it.

Also, most of the changes required to eliminate greenhouse gasses result in an overall healthier economy anyway. I think in general that we ought to make various changes to make sure that the only thing our economy consumes is energy from the sun, radioactive elements, the Earth's rotation, the Earth's internal heat, the kinetic energy of the moon or similarily limitless (with respect to current energy usage levels) sources. Certainly no chemical reaction that has a recylcling cycle that we have a good possibility of overloading. It's just good sense and planning.

easwaran
Mar. 25th, 2007 10:02 am (UTC)
I strongly considered answering (d) for the string theory question despite not knowing much about the debate. I'm mainly of that opinion because of the types of arguments that I've seen leveled against it, which really haven't seemed very striking to me, except for the fact that it might suck up more funding and brainpower than a reasonable share.

I couldn't decide between (b) and (c) on global warming. I of course don't have any direct familiarity with the science, but the basic mechanism sounds plausible, and it's supported by so many trustworthy people and contested basically by none. The only question as far as I'm concerned is how serious are the issues, and how might it be possible to counteract some of the effects. (I think this is the one thing Bjorn Lomborg actually argues in any reasonable way - perhaps the best way to address it is to continue using fossil fuels to spur economic growth to develop the technology and infrastructure to figure out how to suck up the CO2.)

Maybe I should have said I dont' know enough about the issue for the robots. But the evidence seems to be that it will take a long time at any rate. Of course, more likely, they'll never be anything like human consciousness, but people will become more cyborg as time goes on, and our very notion of consciousness will change.
spoonless
Mar. 25th, 2007 06:23 pm (UTC)
Sounds like we're roughly on the same page with global warming. I'm unsure of how serious it is, but it seems like unless this is an unprecedented time in history where a whole field of science has gone crazy, we should trust them when they say it's a problem, and maybe even an urgent problem. I was raised to believe it was a "myth made up by the liberal media" and "junk science" (which is still what my father says today) so maybe there's still an inclination of that left in me which makes me hesitate to want to call it a crisis, but I really don't know.

I do hope that technological innovation can help with it, as I tend to be very optimistic about the prospects of technology over the next century (if you couldn't tell), but at the same time I can see how it could be seen as wishful thinking or procrastination... to say that we'll just deal with the problem later rather than now.

more likely, they'll never be anything like human consciousness, but people will become more cyborg as time goes on, and our very notion of consciousness will change.

I definitely agree that are notion of consciousness will change. And you also might be right that they will never be all that similar to human consciousness... although I think they will be similar enough that it will be reasonable to call at least some of them conscious, and then maybe we'll have to make up a slightly more general word to describe the types of states the rest of them are in.
spinemasher
Apr. 3rd, 2007 01:23 am (UTC)
I opted for the well tested theory of evolution rather than fact. Fact to me is something that follows from logical necessity or is a consequence of a set of givens hence immutable by its very definition and self-consistence. On the other hand, a theory of the physical world is subject to a great deal of caveats. One cannot say that it is a fact that objects fall with an acceleration rate of 9.8m/s2, for in enough million years from now there won't even be an earth around for which to make sense of the stated fact. It is merely an assumption due to observation of the stability of physical systems that we expect to measure objects falling with an acceleration rate of 9.8m/s2 tomorrow.

As far as the robots are concerned, I answered that it will never happen on the grounds of 2 things. First is my bias on the future of computational machinery and second on a technicality of words. I chose, "Robots will never be able to match certain things humans can do." In the sense that I believe a cybernetic organism will be ultimately required to truly imitate human functions of all types perfectly. This is because I expect that we will find that the most efficient (smallest physical volume, highest computational capacity) computational systems will be ones that integrate both an organic structure and an electro-mechanical one mixing the best of the nano-solid-state world and the relentless efficiency of biological evolution. Hence it may take a cyborg in order to mimic a human being flawlessly.
spoonless
Apr. 4th, 2007 03:04 am (UTC)

This is because I expect that we will find that the most efficient (smallest physical volume, highest computational capacity) computational systems will be ones that integrate both an organic structure and an electro-mechanical one mixing the best of the nano-solid-state world and the relentless efficiency of biological evolution.

Hmmm... I tend to think that silicon computers will turn out to be the most efficient computational systems, rather than organic ones. Although the way we design the chips I think should be a bit more parallel than serial... it might be good to have something with a wider fan-out ratio. There is one sense in which I'd agree with you, however. I think that the mechanical aspects of a human body, such as feeling things with our skin, or delicately grabbing things, might be easier to interface with an organic control center rather than a purely electronic one.
( 22 comments — Leave a comment )

Profile

blueshirt
spoonless
domino plural

Latest Month

May 2017
S M T W T F S
 123456
78910111213
14151617181920
21222324252627
28293031   

Tags

Powered by LiveJournal.com
Designed by Lizzy Enger