A Simple Question: Should We Believe Scientists?

[SciameriKen]

You avoided my previous question - but I'll post again - what is your larger point here? If its that we can't know everything 100% I"ll be largely unimpressed.

I wasn't aware that knowledge, at least propositional knowledge (knowledge-that as opposed to knowledge-how) which is what we're examining here, admits of degrees.

It seems to me a person either knows a given proposition or does not know. How could you, for example, "70% know" that Paris is the capital of France, or that DNA has a double helix structure, or that the Earth is three weeks old?

Looks like you'll have to remain 90% unimpressed until we can figure out how a person could be 10% pregnant.


I think I sense what you mean to ask, though, and my answer is simply this: many people's confidence in scientific claims to knowledge is, in my opinion, incommensurate with the appropriate degree of epistemic weight we ought to assign to such claims.

Yes, I know it's very hard, when you're totally in the grip of a theory, to be skeptical of one's own research; after all, it really does seem that "the evidence is overwhelming". At times like these it's hard to be objective, stand back, and take into consideration the evidence from the historical record.

Perhaps the best remedy for this exaggerated confidence is a journey through the history and philosophy of science; books written by historians and philosophers of science rather than scientists themselves. After all, as Thomas Kuhn pointed out, "[sciences have] a persistent tendency to make [their] own history ... look linear and cumulative."


Two questions for you, Ken:

(i) How would you estimate the ratio of scientific hypotheses/theories ever proposed and once widely regarded as true only to be amended or abandoned vs those still regarded as true? 10:1? 1000:1? 100,000:1? Or what?

(ii) In light of your answer to (i), what do you feel is the appropriate degree of confidence we should assign to, say, the theory of evolution, as it stands right now, being true?

I find the world much easier to understand by not requiring knowledge to be at a 100% threshold for certainty. Here is an experiment you can try too - I suspect there is a mug on my desk - my hypothesis is that if I look at my desk I will visually confirm the existence of a mug. I just carried out my experiment and my hypothesis was proven true. Does this mean I can with 100% certainty proclaim the fact that there is a mug on my desk? Well what if as indicated by a different thread that was just posted on this site that in fact we are all existing in a computer simulation, and I am wired up to believe a mug is on my desk when in fact not only is there no mug, but there isn't even a desk - and I truly do not have eye balls! Chances of this are who knows? 0.00000000000000000000000000000000000000001%, if that? Regardless, I cannot with 100% certainty, for this and for whatever other crazy reason say that there is a mug on my desk. However, if I am comfortable with 99.999...% then I can move on and behave as if I was 100% certain - and if that crazy scenario proves to be true - will I will have to adapt - as the world does when a seemingly proven theory runs into contradicting evidence.

I would argue that this is not exaggerated confidence - it is confidence while it is working, and if the evidence suggest otherwise, then moving on.

So regarding your questions -
#i: I'll answer the first with a question - does the exact number matter? Are people wrong versus are people wrong a lot matter? Two things that would be interesting is if this number changes with time - I would suspect the ratio shifts towards stability as techniques/methodologies improve - and if given infinite time if 99.999...% shifts towards being disproven?

#ii: Ultimately what matters is the predictive power or utility of it - I'd have confidence in an idea that had only 1% chance of being correct if that was the best explanation possible - of course always being wary of it being incorrect. That being said, the theory of evolution I have at least 99% confidence in it being correct. I have personally applied the theory of evolution to my own research and it has yielded new directions for me. Creationism cannot have done this for me as God isn't available to explain what was done :). However, if God does appear - explains how all creatures were created - and the explanations yield scientific results for me - well then, the theory of evolution can take a hike!

[Positor]

Let's try an example. And let's make the stakes high to prevent flippant answers. An evil demon has taken one of us hostage, oh let's say BraininVat, and is holding a gun to his head. We're told:

"Scientific experts currently estimate the age of the Earth to be 4.5 billion years. You can choose to bet for or against this being the correct answer. I'm in a good mood so let's say I'll allow a margin of error of half a billion years on either side. Oh, by the way, I'm omniscient. If you bet wisely, this wretch goes free; otherwise he eats lead. Place your bets, please."

I'd bet against, and no, not because the experts are dumb. These things do tend to get revised a lot though.

I am interested to know:

How large would the allowed margin of error have to be for you to bet for 4.5 billion years?

[Lomax]

You avoided my previous question - but I'll post again - what is your larger point here? If its that we can't know everything 100% I"ll be largely unimpressed.

I wasn't aware that knowledge, at least propositional knowledge (knowledge-that as opposed to knowledge-how) which is what we're examining here, admits of degrees.

It seems to me a person either knows a given proposition or does not know. How could you, for example, "70% know" that Paris is the capital of France, or that DNA has a double helix structure, or that the Earth is three weeks old?

Hmm? Are you familiar with fuzzy logic?

[Eclogite]

As far as I can now recall I could find no correlation between ambient temperature and arachnid habit in this context. However, the increase in temperature would likely change the mix of bird species, through direct and indirect means. This would probably affect the success of these spiders, either increasing or decreasing their numbers and doubtless altering some behaviours, dependent upon the particular birds that became the favoured target.

Forgive me, Eclogite, if I say this all sounds hopelessly vague -- and given the putative randomness of mutation postulated by Darwinian-based theory, any prediction would, by necessity, have to be hopelessly vague, as Cohen correctly points out.

Nonsense. It is general, not vague.

It is general because it is a casual recollection of a marginally more precise speculation based upon, at most, two hours of study on the character of bird eating spiders and the Colombian ecologies. This could be made ever more precise in proportion to the resources applied to the problem. If we wished to answer the question, to the degree that you appear to require to dissuade you from your current scepticism the investigation might proceed like this:

Determine the target species and the proportion that they form in the arachnid diet.
Establish how the target species number and behaviour may change in response to a temperature increase.
Assess the impact of this change in target species upon the arachnid.
Identify evolutionary changes that would help the arachnids adapt to their changed environment.

The latter would be of two types:
A numerical prediction of which existing genotype/phenotypes would be favoured by the changed environment and would therefore increase.
Identification of a suite of currently non-extant, but plausible mutations some of which could be expected to appear.

[NoShips]

Let's try an example. And let's make the stakes high to prevent flippant answers. An evil demon has taken one of us hostage, oh let's say BraininVat, and is holding a gun to his head. We're told:

"Scientific experts currently estimate the age of the Earth to be 4.5 billion years. You can choose to bet for or against this being the correct answer. I'm in a good mood so let's say I'll allow a margin of error of half a billion years on either side. Oh, by the way, I'm omniscient. If you bet wisely, this wretch goes free; otherwise he eats lead. Place your bets, please."

I'd bet against, and no, not because the experts are dumb. These things do tend to get revised a lot though.

I am interested to know:

How large would the allowed margin of error have to be for you to bet for 4.5 billion years?

I don't think I'd ever be inclined to bet for.

The best deal I could possibly secure for myself, assuming symmetry on the margin of error, would be a latitude of 4.5 billion years on either side. In other words, if the correct age of the Earth is in the range 0 - 9 billion years, and I bet for, then BraininVat lives to fight another day, woo the ladies, and delete more of my posts.

While this is going on, the punter who bets against monopolizes the entire time range from 9 billion years to infinity. Lucky git!

On purely a priori grounds, then, the chances of for coming up trumps would appear vanishingly miniscule; rather like betting on a given albatross who has just suffered a minor avian coronary somewhere over the Pacific Ocean landing on a rubber dinghy that just happens to be passing below.

As all parties concede, however, there's more to this than just a priori probabilities -- we have reasons pertinent to science for betting as we do. I would suggest, though, that the optimist would need an extraordinarily cogent reason to offset the daunting a priori odds; say, for example, an exceptionless record of breathless seabirds plopping neatly into dinghies.

The optimists among us (BiV and Forest) now appeal to the strength of the evidence, the veracity of dating techniques; our token miserable sod (yours truly), meanwhile, looks to the historical record only to find a graveyard of floating, watery albatross corpses.

What would it take, after all, in our current case, for the albatross to miss that dinghy and for BiV to eat lead for supper? Ans: an adjustment of the current estimate, in the direction of an older Earth, by a mere factor of two or more. (i.e., from 4.5 to 9 billion years or greater)

Has this kind of thing happened before? Well, minimal googling reveals that scientific estimates for the age of the Earth, from roughly a hundred years ago until present, have been modified, in total, by a factor of not only a paltry two, not just even ten, but a staggering forty-five!

Yet the optimists insist, with a complacency that makes albatross mothers mourn, that's not gonna happen again. "We're different now!"

How would you bet yourself, Positor?

[NoShips]

Hmm? Are you familiar with fuzzy logic?

Only in name, Lomax. What do you see as its relevance here?

By the way, you haven't told us how you'd bet yet.

[NoShips]

Nonsense. It is general, not vague.

Ahem.

"It's not a bug; it's a feature" - a dude anxious to sell something

[handmade]

A deceptively simple question perhaps.

Have scientists been right before? I have little doubt. And thanks!

Have scientists been wrong before? I have no doubt whatsoever: the historical list is both lengthy and incontrovertible. Even those theories considered most highly confirmed and embraced with certainty, or near certainty (Newtonian mechanics, say), have subsequently been abandoned by scientists themselves as an accurate representation of reality, if not as a useful tool ("Got us to the Moon", and all that).

This much, ladies and gentlemen, is, I assume, entirely uncontroversial. That said, what is the appropriate epistemological weight we should assign to the knowledge claims of science? When should we believe them, and when might a more circumspect attitude be appropriate?

Does smoking really cause cancer? (*cough*) Is global warming really due to our planetary mismanagement? Do quarks exist? Are tectonic plates real? (ever seen one?) Is the luminiferous aether real? (J. C. Maxwell allegedly claimed it was the most highly confirmed entity in all science). Is Dawkins justified in claiming that only the ignorant, stupid, insane, or wicked would doubt the scientific account of evolution?

How do we decide these things?

I will answer according to your title, a simple question only needs a simple answer.

Should we believe scientists? NO

Should we believe in evidence the scientist provides? Yes as long as it is correct

[NoShips]

I will answer according to your title, a simple question only needs a simple answer.

Should we believe scientists? NO

Should we believe in evidence the scientist provides? Yes as long as it is correct

Sorry, Handmade, but I'm having trouble making sense of this remark. I had this problem with another member earlier in the thread who seemed to envision "evidence" as the kind of thing one might trip over during a stroll in the woods, or the kind of goodies one might accumulate in a bank vault.

Nothing is intrinsically evidence. Evidence is a relational term; something (an observation, a phenomenon, a physical object, etc) is only evidence in relation to something else -- a theory or a hypothesis, for instance.

Much is being written right now in another thread about the phenomenon of time dilation and the theory/theories of relativity. So using time dilation as an example, when you say I should "believe in the evidence [if it's correct]", what is it that you're trying to tell me, Handmade:

1. I should believe in the phenomenon of time dilation? (But time dilation in itself is not evidence; it's simply a phenomenon)

2. I should believe that the phenomenon of time dilation constitutes evidence? Constitutes evidence for what? For relativity theory? What if another scientist tells me time dilation constitutes evidence for a different theory, a rival to relativity, are they both right?


Finally, what does it mean to say evidence is "correct"? Do you mean that the object/observation/phenomenon adduced as evidence is itself beyond doubt? (i.e. time dilation is a real phenomenon). Or that its standing in an evidential relationship to a particular theory/hypothesis is beyond doubt? (i.e. time dilation really is evidence for such-and-such a theory).

[Forest_Dump]

Out of curiosity, how much do you trust medical science? Given that medical science is now a major industry and that is would seem likely that some large percentage of practitioners enter the field promarily just to make money, when you get some disease or affliction, what do you really bet on when push comes to shove? Do you put blind faith on the contents fo some pill or fluid injected into you, trusting it is pure and will cure the affliction without causing worse harm to you? Do you "trust" that x-rays taken, etc., provide actual data and will not cause radiation sickness and death? Do you get a "second opinion" from other similar experts or do you also consult faith healers, witch doctors, etc. from truly independent professions, etc? What do you really do when it is the life of yourself or someone you love at stake?

[Eclogite]

Nonsense. It is general, not vague.

Ahem.

"It's not a bug; it's a feature" - a dude anxious to sell something

Aphorisms and Arguments are different beasts.

Implicit accusations of manipulative rhetoric are self-defeating when delivered via manipulative rhetoric.

[NoShips]

Out of curiosity, how much do you trust medical science? Given that medical science is now a major industry and that is would seem likely that some large percentage of practitioners enter the field promarily just to make money, when you get some disease or affliction, what do you really bet on when push comes to shove? Do you put blind faith on the contents fo some pill or fluid injected into you, trusting it is pure and will cure the affliction without causing worse harm to you? Do you "trust" that x-rays taken, etc., provide actual data and will not cause radiation sickness and death? Do you get a "second opinion" from other similar experts or do you also consult faith healers, witch doctors, etc. from truly independent professions, etc? What do you really do when it is the life of yourself or someone you love at stake?

By and large, I listen to my doctor and take the pills, electrotherapy, or whatever other treatment he/she recommends.

A few points though:

1. It's not clear to me that a doctor is considered a scientist. I imagine most doctors at cocktail parties would respond to "I hear you're a scientist" with something like "No, I'm a doctor".

2. In any case it's not particularly salient. What you're talking about here (again!) is the practical applications of scientific theories; not the truth of the theories themselves. (and truth has been our focus since the deGrasse Tyson fiasco)

Other members keep reminding me that airplanes fly and computers work. At no point in the thread have I denied the efficacy of good scientific theories; quite the contrary, I have positively affirmed it again and again. Science works, indeed works extremely well. By this I mean, regardless of whether theories are true or false, they often allow us to derive accurate/correct/true (choose your adjective) predictions, in virtue of which we are able to intervene and gain some measure of control over natural processes.

For example, I would have great faith in scientists' predictions about what will happen (quantitatively) if an object is released near the Earth, or about the date of Halley's comet returning. In this sense I believe what scientists tell me. Where I would be more circumspect is in believing the truth of the theories from which these predictions are derived.

Newton, for one, had a theory that allowed us to make such predictions: it involved the invocation of a theoretical entity called "gravity" which was construed, as far as I understand, as an attractive force which acts instantaneously over infinite distances, apparently with no expenditure of energy, against the backdrop of a perfectly homogenous absolute time and absolute space.

I don't think anyone believes that account any more, not even the scientists themselves.

Einstein, now, has a different story -- a logically incompatible story -- to tell which predicts/explains the very same phenomena. I doubt very much his will be the final word either. This should not be construed as an attack on Newton or Einstein (sigh!); simply a statement of, what I see as, facts.

Bet yer bottom dollar Halley's comet will be back exactly when predicted, airplanes will keep flying (as BiV and others continually point out as if it's relevant to my arguments), computers will keep working, and Viagara will continue to save the day.

What I've been recommending is a little more caution before nailing your colors to the truth of the theories which purport to underlie these phenomena.

[Forest_Dump]

What I've been recommending is a little more caution before nailing your colors to the truth of the theories which underlie these phenomena.

Well I think here I would agree with you in that, IMHO, a scientific theory is ultimately the best explanation we have so far (i.e., pending a better one) and that we do have some ways of measuring alternative theories for how well they conform to various types of observations, allow predictions, etc. And the best theory so far is usually the one that has survived testing, relies on testable under-lying mechanicsms, conforms to logic, allows some kinds of predictions, etc. But I don't necessarily think all these criteria are always met (e.g. how do we viably predict things that may take thousands or millions of years to occur, how troubling is it that we don't fully know what causes gravity, etc.).

it involved the invocation of a theoretical entity called "gravity" which was construed, as far as I understand, as an attractive force which acts instantaneously over infinite distances,

Pretty minor point, I am sure, but is this really true? I wouldn't have guessed that and truth be told I wonder how that could be tested. Wouldn't we have to somehow create matter and then measure if its gravitational effect happened "instantly" over large distances? I would have guessed it act at about the spped of light but what do I know?

[NoShips]

it involved the invocation of a theoretical entity called "gravity" which was construed, as far as I understand, as an attractive force which acts instantaneously over infinite distances,

Pretty minor point, I am sure, but is this really true? I wouldn't have guessed that and truth be told I wonder how that could be tested. Wouldn't we have to somehow create matter and then measure if its gravitational effect happened "instantly" over large distances? I would have guessed it act at about the spped of light but what do I know?

I'm pretty sure what I said is right, although I have no sources at hand (I'll post if I come across any).

That is to say, I think I'm right in the way gravity was construed (i.e. acting instantaneously) under the Newtonian framework; not that this is believed to be true any more.

Better check with the physicists. Don't believe all they say though :-)

[handmade]

I will answer according to your title, a simple question only needs a simple answer.

Should we believe scientists? NO

Should we believe in evidence the scientist provides? Yes as long as it is correct

Sorry, Handmade, but I'm having trouble making sense of this remark. I had this problem with another member earlier in the thread who seemed to envision "evidence" as the kind of thing one might trip over during a stroll in the woods, or the kind of goodies one might accumulate in a bank vault.

Nothing is intrinsically evidence. Evidence is a relational term; something (an observation, a phenomenon, a physical object, etc) is only evidence in relation to something else -- a theory or a hypothesis, for instance.

Much is being written right now in another thread about the phenomenon of time dilation and the theory/theories of relativity. So using time dilation as an example, when you say I should "believe in the evidence [if it's correct]", what is it that you're trying to tell me, Handmade:

1. I should believe in the phenomenon of time dilation? (But time dilation in itself is not evidence; it's simply a phenomenon)

2. I should believe that the phenomenon of time dilation constitutes evidence? Constitutes evidence for what? For relativity theory? What if another scientist tells me time dilation constitutes evidence for a different theory, a rival to relativity, are they both right?


Finally, what does it mean to say evidence is "correct"? Do you mean that the object/observation/phenomenon adduced as evidence is itself beyond doubt? (i.e. time dilation is a real phenomenon). Or that its standing in an evidential relationship to a particular theory/hypothesis is beyond doubt? (i.e. time dilation really is evidence for such-and-such a theory).

Evidence comes in many different forms, however evidence can sometimes be misleading if the evidence is interpreted incorrectly.
I consider that human logic is the main evidence we could ever hope to use. Defining possibilities or impossibilities, the human mind experience being a key rudiment of all science. If the logic of a notion is false, then the process the logic is identifying must also be false.

You asked me :

1. I should believe in the phenomenon of time dilation? (But time dilation in itself is not evidence; it's simply a phenomenon)

My truthful answer is no, you should not believe in a time dilation. However you should believe in the evidence of a relative timing dilation.

2. I should believe that the phenomenon of time dilation constitutes evidence? Constitutes evidence for what? For relativity theory? What if another scientist tells me time dilation constitutes evidence for a different theory, a rival to relativity, are they both right?

It depends on the rival theory, for example my notions involve Einstein, Planck, Newton, I consider it is just an extension to their ideas but put in a more understandable and precise logical way (axioms). Relativity is correct because the interpretation of time makes it correct, but if we interpret things differently then relativity fails. Relativity can not be correct and incorrect at the same time, so therefore in my eyes I ''see'' the correctness if we redefine the term time dilation to a timing dilation , timing dilation being the correct terminology and accuracy that removes confusion from relativity.
The rest of the notion would then be correct.

Finally, what does it mean to say evidence is "correct"?

Evidence is correct if the logic that goes with it is also correct. If the logic does not make relative sense to the evidence being provided , then it must be incorrect evidence interpretation.

This question reminds me of a picture I once viewed, A women crying holding a baby why a man was seemingly waving his arms at her in anger, the observation evidence however, told a lie, the women's husband was returning from war with open arms.
Ambiguity of evidence and interpretation can be a problem, but axioms never lie.

[kevinandrew]

What is science? Science means knowledge - specifically knowledge that can be checked/repeated. If it cannot be checked or repeated, then its not really science. In my opinion, we should be sceptical about science and only believe it if it can repeated and checked.

[NoShips]

@ Handmade -- Either you make no sense whatsoever or my watch has stopped

What is science? Science means knowledge - specifically knowledge that can be checked/repeated. If it cannot be checked or repeated, then its not really science. In my opinion, we should be sceptical about science and only believe it if it can repeated and checked.

Kevin, I think all would agree that if an observation is verified ("checked and repeated" as you say), barring radical skepticism -- evil demons and brains in vats -- we can be pretty certain that our eyes are not deceiving us and can justifiably claim knowledge.

For now I'm ignoring possible complications arising from the oft heard claim that all our observations are "theory-laden", that is to say, not only are the theories we construct dependent on observational facts, but more worryingly, observational facts themselves are dependent, at least to some degree, on the theoretical/conceptual apparatus we bring to bear upon our observations. "Theory without observation is empty; observation without theory is blind" as the wise men say. In other words, the concern is that no theory-neutral language exists within which we can simply describe the raw, brute facts: there ARE no raw brute facts; all our so-called facts are infused with theory. Most of us share a naive intuition that while theories might come and go, facts are facts (those facts, you assure us, that have been "checked and repeated"), but if our facts are indeed theory-laden, then when our theories change, the facts change too.

Rather, I'd just point out a more immediate and less philosophical objection to your encapsulation of science. As you've described it, Kevin, science would be nothing more than a cataloguer of observable facts ("an unusually bright light has been observed in the heavens", "some animal species exhibit altruistic behavior", "this lump of ice floats on water", etc., etc) -- science would be a butterfly collector!

But surely the power -- the essence! -- of science lies in its "going behind the scenes" to construct theoretical systems which unify and perhaps (purport to) explain these observational facts of which you speak. And, unlike the simple (*twitch*) facts of observation, the truth of these theories cannot be ascertained by just looking.


That said, as Yogi Berra reminds us, you can observe a lot just by looking :-)

[kevinandrew]

In my opinion, theory is separate, although supported by, science (knowledge). Scientific endeavour is about observing how the outputs of a defined process vary as the inputs are adjusted. If we collect enough data about the inputs and corresponding outputs, we might develop a theory about the mechanism of the process. And the theory might be supported or refuted by more data (science/knowledge).

The power of science lies in its ability to prove most theories are wrong.

[TheVat]

Other members keep reminding me that airplanes fly and computers work. At no point in the thread have I denied the efficacy of good scientific theories; quite the contrary, I have positively affirmed it again and again. Science works, indeed works extremely well. By this I mean, regardless of whether theories are true or false, they often allow us to derive accurate/correct/true (choose your adjective) predictions, in virtue of which we are able to intervene and gain some measure of control over natural processes.

For example, I would have great faith in scientists' predictions about what will happen (quantitatively) if an object is released near the Earth, or about the date of Halley's comet returning. In this sense I believe what scientists tell me. Where I would be more circumspect is in believing the truth of the theories from which these predictions are derived.

- NoShips

What's weird to me is the Purity Position you seem to be taking on Truth. In the sciences, claims are pretty much based on the stuff you said you already accept - things that work, that make accurate predictions, and so on. Evidence has weight when a knowledge is derived from it that has that "efficacy" we discussed earlier. I don't think too many scientists are preoccupied with some Ultimate Truth of a theory, so I find that this chat seems to me more and more just a lot of semantic handwringing. You are taking medicine, flying on jets, looking up at passing comets and eclipses, so I'd say you are, de facto, doing a lot of believing of scientists. Why is this pragmatic form of knowledge (hell, we don't even have to use the dreaded T-word!) such an issue for you? All you need is some consistent observations, some good old inductive reasoning, and you are as far towards a scientific fact as you need to be. We can leave causality, the ultimate nature of time and being and consciousness, to the philosophers. Let them natter on about Truth, keeps them out of trouble, and the rest of us can get on with our lives and living in this world.

[NoShips]

In my opinion, theory is separate, although supported by, science (knowledge). Scientific endeavour is about observing how the outputs of a defined process vary as the inputs are adjusted. If we collect enough data about the inputs and corresponding outputs, we might develop a theory about the mechanism of the process. And the theory might be supported or refuted by more data (science/knowledge).

The power of science lies in its ability to prove most theories are wrong.

Once again, Kevin, I find this response unrealistically simplistic. Two points:

1. Are you aware of the so-called "Duhem-Quine thesis"? The thesis makes clear that naive Popperian-style falsificationist doctrine simply won't do. Theories/hypotheses are not tested in isolation, but rather as part of a package, including the theory itself and various background assumptions and auxiliary hypotheses.

When observation clashes with theory, falsificationist doctrine tells us the theory has been falsified and must be abandoned. In practice, as the D-Q thesis predicts and history bears witness, this almost never happens, particularly in the case of a well entrenched theory. Logic can only tell us something is wrong somewhere in the package; it cannot tell us where the problem lies (i.e. in the theory under test, or somewhere else)

A stock example (circa 1600): the Copernican model demands stellar parallax. No stellar parallax is observed. Observation disagrees with theory. Falsificationist methodology tells us the theory is false and must be dumped. Is the theory dumped? Ans: no. The blame is put somewhere else ("the stars must be much further away than we thought")


2. In many cases when two theories are vying for attention, an old and a new, talk of proof (of the new theory) or disproof (of the old theory) is wholly inappropriate. What we see happening, rather, is either that proponents of the old theory simply die off (What's their problem then, Kevin? Too dumb to understand the "disproof" of which you speak?), or else the word "conversion" seems more apposite to describe the switch of allegiance from old to new. Despite its religious overtones, the word is commonly used by scientists themselves in their writings.

This is not mathematics. In empirical science, evidence and logic seldom, if ever, lead ineluctably to a single conclusion. Talk of proof and disproof are misplaced, in my opinion.

I quote the following from the chapter on "Conversion as a Feature of Scientific Revolutions" from J. Bernard Cohen's "Revolution in Science":

"[...] That phenomenon is conversion. Max Planck is often quoted to the effect that 'new scientific truth does not triumph by convincing its opponents and making them see the light, but rather because its opponents eventually die, and a new generation grows up that is familiar with it'. A similar sentiment was expressed a half-century earlier Harvard's Professor Joseph Lovering, when he told his students that there are two theories of light, the wave and the corpuscular. Today, he is said to have remarked, everyone believes in the wave theory; the reason is that all those who believed in the corpuscular theory are dead. There is a measure of truth in such statements, as we all know, and yet a new scientific idea does win adherents, and even convinces some opponents, as has been seen in many examples throughout this book."

[NoShips]

What's weird to me is the Purity Position you seem to be taking on Truth. In the sciences, claims are pretty much based on the stuff you said you already accept - things that work, that make accurate predictions, and so on. Evidence has weight when a knowledge is derived from it that has that "efficacy" we discussed earlier. I don't think too many scientists are preoccupied with some Ultimate Truth of a theory, so I find that this chat seems to me more and more just a lot of semantic handwringing. You are taking medicine, flying on jets, looking up at passing comets and eclipses, so I'd say you are, de facto, doing a lot of believing of scientists. Why is this pragmatic form of knowledge (hell, we don't even have to use the dreaded T-word!) such an issue for you? All you need is some consistent observations, some good old inductive reasoning, and you are as far towards a scientific fact as you need to be. We can leave causality, the ultimate nature of time and being and consciousness, to the philosophers. Let them [philosophers] natter on about Truth, keeps them out of trouble, and the rest of us can get on with our lives and living in this world.

BiV, if you think scientists don't speak of truth, you're simply mistaken, and can easily be shown to be so.

Yes, no denying, there are scientists who avoid the "dreaded T-word" like the plague. But there are plenty of others who scream it louder than a sexually frustrated Tarzan. Try Steven Weinberg for one. You'll find that T-word staring at you from almost every page. (Wish I'd been keeping notes now, sigh!)

And you do remember what started off my latest fit of distemper, don't you? Ans: Neil deGrasse Tyson's proselytizing about the truth of science, yes that's right, TTTRRRRRRUUUUUTTTTTHHHHHHH!

Want me to start flooding the room with scientists' truth quotes? :-)

[NoShips]

P.S.

And to those who would deny any role to the dreaded T-word (truth) in science, what do you reply when your teenage daughter asks you, "Daddy, is it true that smoking causes cancer?"

(i) "Yes, it's true. Please don't smoke, my dear. I love you."

(ii) "Pfft and meh. Science is all about constructing models. Scientific claims have nothing to do with truth. Here's a carton of Marlboros to get you started. darling."

[Forest_Dump]

(i) "Yes, it's true. Please don't smoke, my dear. I love you."

(ii) "Pfft and meh. Science is all about constructing models. Scientific claims have nothing to do with truth. Here's a carton of Marlboros to get you started. darling."

How about (iii) Neither. There have been many studies indicating a statistical correlation between smoking and some kinds of cancer. There has been a suggestion that the causal mechanism has been identified but I am not sure how well that has been tested and independently confirmed. I won't buy you any but... need a light?

[NoShips]

Let them [philosophers] natter on about Truth, keeps them out of trouble, and the rest of us can get on with our lives and living in this world.

As promised, the BBC will intermittently be posting quotes, as we come across them, to demonstrate that it's not just philosophers with no lives who natter on about truth.

"All this [i.e. Kuhn's ideas] is wormwood to scientists like myself, who think the task of science is to bring us closer and closer to objective truth."

-- Steven Weinberg (from "Facing Up", essay 17, "The Non-Revolution of Thomas Kuhn")

[NoShips]

Ladies and gentlemen, a warm welcome please for another loser with no life who natters on a lot about truth: professor Richard Dawkins...

"There are people for whom truth doesn’t matter"

"If I am asked for a single phrase to characterize my role as Professor of the Public Understanding of Science, I think I would claim Advocate for Disinterested Truth."

"We must favor verifiable evidence over private feeling. Otherwise we leave ourselves vulnerable to those who would obscure the truth"

"Gravity is not a version of the truth. It is the truth. Anybody who doubts it is invited to jump out of a tenth-floor window." [It's almost painful, though, to have to listen to this advanced as an argument - me]

"It is the plain truth that we are cousins of chimpanzees ..."

"Everybody likes a good story. Myths are fun, as long as you don't confuse them with the truth. The real truth has a magic of its own. The truth is more magical, in the best and most exciting sense of the word, than any myth or made-up mystery or miracle. Science has its own magic - the magic of reality."


The "real truth"? Hmm. As opposed to what? The fake truth? Sigh!

[SciameriKen]

Ladies and gentlemen, a warm welcome please for another loser with no life who natters on a lot about truth: professor Richard Dawkins...

"There are people for whom truth doesn’t matter"

"If I am asked for a single phrase to characterize my role as Professor of the Public Understanding of Science, I think I would claim Advocate for Disinterested Truth."

"We must favor verifiable evidence over private feeling. Otherwise we leave ourselves vulnerable to those who would obscure the truth"

"Gravity is not a version of the truth. It is the truth. Anybody who doubts it is invited to jump out of a tenth-floor window." [It's almost painful, though, to have to listen to this advanced as an argument - me]

"It is the plain truth that we are cousins of chimpanzees ..."

"Everybody likes a good story. Myths are fun, as long as you don't confuse them with the truth. The real truth has a magic of its own. The truth is more magical, in the best and most exciting sense of the word, than any myth or made-up mystery or miracle. Science has its own magic - the magic of reality."


The "real truth"? Hmm. As opposed to what? The fake truth? Sigh!

NoShips,

You've been on your soapbox for a few months now and I remain unconvinced of whatever point you are trying to make here. The topic of your thread is whether we should believe scientist. My answer is yes. An answer from a scientist will provide a greater likelihood of being accurate to nature, therefore increasing the probability of attaining a desired outcome.

[NoShips]

You've been on your soapbox for a few months now and I remain unconvinced of whatever point you are trying to make here. The topic of your thread is whether we should believe scientist. My answer is yes. An answer from a scientist will provide a greater likelihood of being accurate to nature, therefore increasing the probability of attaining a desired outcome.

This answer, even supposing it's true, won't do, I'm afraid, Ken, as already pointed out earlier in the thread.

On page 4 Eclogite said the following, redolent of your own latest asseveration:

"So, despite the ignorance of the expert, we are more likely to gain a knowledge of reality by accepting the conclusions of the expert than by taking wild assed guesses, believing in a snake oil salesman, or any of the other alternatives open to us."

to which I responded

""More likely to gain a knowledge of reality..." [my emphasis], you say? What if you have a 1% likelihood of gaining knowledge by listening to me, and a 2% likelihood by listening to the experts: should you listen to them?

("I think she has a 2% chance of being right" = "I think she's wrong", n'est-ce pas? Should you choose to believe her, it would be out of desperation, and not because you think she possesses knowledge -- "It's a long shot, Jim, but we have to try it". Is this how we should think of scientific claims?)"



If the point is still unclear, suppose the question is "What is the capital of France?" Our friendly omniscient evil demon -- a maths whiz to boot! -- informs you that my answer (a non-expert) has a 5% probability of being correct. The expert's answer, meanwhile, has a 10% probability of being right.

By believing the expert, your own criterion of achieving "a greater likelihood" is satisfied (10% > 5%). But a 10% likelihood of being right means a 90% likelihood of being wrong. It is irrational to believe that which you know is probably untrue. The rational response is to not believe the expert, or at the very least, withhold judgement.

As I said in my response to Eclogite, if you choose to believe the expert under such circumstances, it's because you're desperate; not because you think he possesses knowledge. It would be an act of faith, not rationality

Your criterion doesn't work, as far as I can see. "Greater likelihood" constitutes insufficient justification for rational commitment to belief.

[NoShips]

On reflection, Ken, I feel my answer to you directly above is somewhat hasty, unsatisfactory, and deserves some elaboration. It's important here, I think, to appeal to the traditional distinction between theoretical (or epistemic) rationality -- what it's rational to believe, and practical rationality -- what it's rational to do.

I want to reproduce again your most recent comment with sanguinary emphasis:

"The topic of your thread is whether we should believe scientist. My answer is yes. An answer from a scientist will provide a greater likelihood of being accurate to nature, therefore increasing the probability of attaining a desired outcome."

Now, unlike Eclogite (who said knowledge), you weren't explicit on what that "desired outcome" is, but if the desired outcome is knowledge then I stand by what I said above (I love Paris when it sizzles, and all that): to commit oneself to a belief that is probably untrue is to violate the canons of epistemic rationality.

On the other hand, if you're sick, and drinking my witch's nostrum gives you a 5% chance of recovery, while taking the doctor's medicine yields a 10% chance, and the "desired outcome" is good health, then, assuming there are no other options, your course of action is clear: practical rationality demands you follow the doc's orders.

Having said all that, to commit yourself to a belief in the proposition "the doctor's medicine will make me better" would remain a paradigm of epistemic irrationality. Nailing your rational colors to the medicine is the proper thing to do; not the proper thing to believe.

Now, if what you're advocating here is a kind of instrumentalist approach to scientific claims ("who cares about truth and knowledge as long as the planes fly, the computers don't break down, the Viagara works, and science can get us to the Moon and back safely") then your position is very close to my own.

But you did use the B-word (believe) in your last post, after all. What is a man to think?

[kevinandrew]

In my opinion, theory is separate, although supported by, science (knowledge). Scientific endeavour is about observing how the outputs of a defined process vary as the inputs are adjusted. If we collect enough data about the inputs and corresponding outputs, we might develop a theory about the mechanism of the process. And the theory might be supported or refuted by more data (science/knowledge).

The power of science lies in its ability to prove most theories are wrong.

Once again, Kevin, I find this response unrealistically simplistic. Two points:

1. Are you aware of the so-called "Duhem-Quine thesis"? The thesis makes clear that naive Popperian-style falsificationist doctrine simply won't do. Theories/hypotheses are not tested in isolation, but rather as part of a package, including the theory itself and various background assumptions and auxiliary hypotheses.

When observation clashes with theory, falsificationist doctrine tells us the theory has been falsified and must be abandoned. In practice, as the D-Q thesis predicts and history bears witness, this almost never happens, particularly in the case of a well entrenched theory. Logic can only tell us something is wrong somewhere in the package; it cannot tell us where the problem lies (i.e. in the theory under test, or somewhere else)

A stock example (circa 1600): the Copernican model demands stellar parallax. No stellar parallax is observed. Observation disagrees with theory. Falsificationist methodology tells us the theory is false and must be dumped. Is the theory dumped? Ans: no. The blame is put somewhere else ("the stars must be much further away than we thought")


2. In many cases when two theories are vying for attention, an old and a new, talk of proof (of the new theory) or disproof (of the old theory) is wholly inappropriate. What we see happening, rather, is either that proponents of the old theory simply die off (What's their problem then, Kevin? Too dumb to understand the "disproof" of which you speak?), or else the word "conversion" seems more apposite to describe the switch of allegiance from old to new. Despite its religious overtones, the word is commonly used by scientists themselves in their writings.

This is not mathematics. In empirical science, evidence and logic seldom, if ever, lead ineluctably to a single conclusion. Talk of proof and disproof are misplaced, in my opinion.

I quote the following from the chapter on "Conversion as a Feature of Scientific Revolutions" from J. Bernard Cohen's "Revolution in Science":

"[...] That phenomenon is conversion. Max Planck is often quoted to the effect that 'new scientific truth does not triumph by convincing its opponents and making them see the light, but rather because its opponents eventually die, and a new generation grows up that is familiar with it'. A similar sentiment was expressed a half-century earlier Harvard's Professor Joseph Lovering, when he told his students that there are two theories of light, the wave and the corpuscular. Today, he is said to have remarked, everyone believes in the wave theory; the reason is that all those who believed in the corpuscular theory are dead. There is a measure of truth in such statements, as we all know, and yet a new scientific idea does win adherents, and even convinces some opponents, as has been seen in many examples throughout this book."

Well, you ask a simple question, maybe you should expect simple answers!

Anyway, perhaps it is too idealistic to think of theories being proved or refuted by science. It maybe more accurate (and no less simplistic!) to say that a person either likes a particular theory or they don't. And what they think they know, is a key (but not the only) factor in shaping their view.

I maintain the view that science and theory are too very distinct things. Science is what we (think we) know, and theory is our attempt to explain it.

[NoShips]

Anyway, perhaps it is too idealistic to think of theories being proved or refuted by science. It maybe more accurate (and no less simplistic!) to say that a person either likes a particular theory or they don't. And what they think they know, is a key (but not the only) factor in shaping their view.

I agree. Not just idealistic, but any claim that such-and-such a theory has been proven or disproven/refuted is demonstrably false, if the word "proven" is understood to mean "logically demonstrated" such as to preclude any possibility of error. This is why I regard comments such as the following (from a thread on Intelligent Design)...


"I make a distinction between Intelligent Design (the poorly concealed creationist cover-up) and intelligent design (an acceptance of the possibility that the universe, or some components, or some of its "direction" were determined by an active intelligence).

The former, with its presumptuous capitals, has been refuted. The latter receives little or no serious attention. Why? At least in part this is out of a fear that paying it attention will give comfort to the creationists and in part because the static of ID drowns out any id signal that may be there."


... as pernicious propaganda, inasmuch as a lay audience, less familiar with the ins and outs of the philosophy of science, is likely to take the scientist at his word, and thus be led down the garden path without knowing it.

"refuted to my satisfaction", or "refuted, in my opinion" I can live with, but "refuted" simpliciter stands as an insult to intellectual integrity.

(I'm not religious in case you're wondering, not that it should matter in cases such as these)