[Animats]

Physics used to be "hard science" - if it wasn't testable, it wasn't meaningful. Physicists used to look down on other sciences for that reason. This article shows how much worse things have become. Nobody can figure out a way to test string theory. In cosmology, you can only observe, not experiment. As the article points out, the big questions are out of reach for scaling reasons.

Trying to fix the problem through weaker definitions of "testable" is desperation. All those smart people hate to face the fact that what they're doing may be total bullshit. This has major economic consequences for physicists - why should they be funded? Physics is funded because it produced the atomic bomb and semiconductors. Those came from the testable parts of physics. Untestable physics cannot produce engineering technology.

(The article says "no one has ever seen an atom". That's just wrong. There are lots of picture of atoms. Even pictures of atoms lined up to spell "IBM".[1])

[1] http://www.nytimes.com/1990/04/05/us/2-researchers-spell-ibm...

[adekok]

> Physics used to be "hard science" -

Physics is still a hard science. However, the edges of physics are difficult to test.

> In cosmology, you can only observe, not experiment

Saying that is like saying "we don't know evolution is true, because we don't see new species evolve every day". It's a naive approach to science.

In both cases, you can use theories to make predictions about new things to look for. e.g. http://tiktaalik.uchicago.edu/

> Physics is funded because it produced the atomic bomb and semiconductors.

Physics is funded because it's cool. And it still gets less funding than major sports leagues. So by your definition, the NFL is funded because it's part of a national defence strategy?

[Avshalom]

>> In cosmology, you can only observe, not experiment

>Saying that is like saying "we don't know evolution is true, because we don't see new species evolve every day". It's a naive approach to science.

In general it's also true of almost all astronomical research, and large whacks of geology. Which is fine because experiments are just observations that scientists are the impetus for. Science works by gathering data, experiments just let us produce data to gather at our own speed instead of having to wait around for something to happen and hope we're looking in the right direction when it does. In the case of astronomy instead of relying on experiments to produce more data we mostly build better instruments.

[rsfern]

Actually, at least in the US, physics is funded because of things like the atomic bomb and the transistor, as GP points out. Observe the rhetoric of any congress member. They don't care if physics is "cool", any more than the general public probably does.

Football isn't funded by the federal government at all, so your point about national security is just silly.

[GFK_of_xmaspast]

Maybe not by the feds, but they squeeze an incredible amount of money from state and local governments.

[vacri]

> Saying that is like saying "we don't know evolution is true, because we don't see new species evolve every day".

We do see new species evolve every day. We also see microevolution so commonly that it's a standard high-school biology task.

> Physics is funded because it's cool.

The point being made is that for a long time, it was common for physicists to look down their nose at other sciences "because what we do is real". Now they're finding themselves in the same boat.

[purpled_haze]

> We do see new species evolve every day. We also see microevolution so commonly that it's a standard high-school biology task.

Can you describe?

My memory of high school genetics involved experiments with Drosophila eye color, etc. which were really about the passing of genes and, sometimes, mutation. However these experiments didn't prove that natural selection or seemingly random mutation were necessarily what drove species to change over time to create new species based on their environment. I had to accept that. I also accept that people tell me human and chimp DNA is similar.

Theoretical physicists also attempt to explain things that they cannot see or experience, and if people believe it, they believe it.

Science, in part, requires faith.

Think of everything you think you understand about science and about our world. How much of it did you really prove with your hands, and how much instead did you just read or hear and understand to be true because others wrote or said it was true?

You can put your faith in theories, and that has proved practical for many physicists and biologists, just to name a few. A god or gods could also be driving those evolutionary processes in part, or could have set things in motion. Accepting religion and scientific facts both require faith, and that's ok.

[vacri]

> Science, in part, requires faith.

Only at the bleeding edge, which you don't need to use in your everyday life. Most science-related things at human scale can be done with your own two hands; you don't require billion-dollar tools to fiddle with materials science, for example. Youtube is filled with videos of people doing human-scale science.

The whole 'science requires faith' argument is largely thrown out to give credence to religion. The problem with that argument is that the fundamental building blocks of science do not require faith, and you build from those. The SI units are convention - you can make your own. What constitutes validity is convention - you can decide your own way. You don't require faith to start doing basic science.

On the other hand, religion's basic building blocks require faith from the outset. Person X is divine, 'trust us'. God Y likes or doesn't like that, 'trust us'. And from these basic religious building blocks, whole edifices are built. For example, most catholics are unaware of how the bible was pieced together by committee. That information doesn't affect their day-to-day lives, however the religious rules that do affect them are based on those fundamental building blocks that are unverifiable by anyone. Faith is required because the foundational blocks are not verifiable at all.

This does not happen in science - you can verify the foundational rules yourself, and build as far along the tree as your interest and tooling allow. Theoretical physics existing in the 'faith' world is largely a canard - most science is not like this, and theoretical physics is at the exploratory edge of science, not at the core foundational blocks. In short, the Planck Length is defined by the meter which we can see and confirm; the meter is not defined by the Planck Length which we theorise about.

Not to mention that in general, 'scientific facts' are constantly tested, and 'religious facts' are not. The engineering capabilities of steel are in constant use in almost every part of our lives; if science was wrong about them, the material would fail in unexpected ways. The divinity of the christ figure... well, it isn't even testable. Christ is only divine to christians, but a steel bridge needs to hold up regardless of the mental state of the humans using it.

[mahranch]

> Only at the bleeding edge, which you don't need to use in your everyday life.

He means how do you know the speed of light is 186,000 mi/sec? Did you test it yourself? Or are you taking someone's word for it? It's one or the other.

> On the other hand, religion's basic building blocks require faith from the outset.

I think you and the guy above you used the wrong word. Instead of faith, I would have used "trust". I trust that Einstein and his peers who studied his work are correct in their assessment of his theory of relativity. I don't need to test it myself, I take their word for it. I trust them.

On the other hand, people who compare something like string theory to philosophy really have it backwards. String theory might be untestable right this second due to technical limitations (string theory does make quite a few unique predictions) but that doesn't mean it will remain that way forever. Hell, it may not be that way tomorrow with as fast as science has been progressing... This is in contrast to proving the existence of a God or deity which has zero hope of ever being testable because it's "inherently" untestable.

As far as string theory goes, one of the most obvious predictions are strings themselves. We cannot build an accelerator large enough to smash particles together at the speeds required to look for strings but that doesn't mean such a thing is impossible; such an accelerator already exists - supermassive black holes. They accelerate particles at the required speeds to look for strings.

[vacri]

> He means how do you know the speed of light is 186,000 mi/sec? Did you test it yourself? Or are you taking someone's word for it? It's one or the other.

I take someone's word for it, but here's the kicker: I don't care. That particular measurement doesn't affect me one jot if it's a different speed. I don't play with optics; the measurement affects nothing I choose to do. However, if I did care, I could work from first principles and find out; it's really well documented, and I can also create my own methodology. This is not true of religious requirements. Throw money in the plate on Sunday because this guy two thousand years ago was divine? There is no way I can verify that.

> Instead of faith, I would have used "trust".

I did leave out a bit saying that the GP was using two different forms of 'faith' that were apples and oranges, but left it out as I was already waffling. :)

However, you will find that as you do your own science, it pretty much always conforms to the rules found by people who have come before you. It's really only as you near the expansion fringes of science that you start getting conflicts with mainstream thought.

Again, this does not happen with religion, where often the argument becomes "god moves in mysterious ways" or "who can know his plan, but he does have one". Core rules that don't make sense with the observed world are often distorted with these phrases. There are definitely grey areas and fuzziness in science - biology has a lot of fuzziness - but there are still common behaviours that work regardless of your cultural background. Born into an Egyptian Coptic family, Christ is divine. Born into the Muslim family next door, Christ is a prophet, but not divine. The fundamental essences of science just don't move with cultural background like that; the resistivity of copper doesn't care about your state of mind.

There is definitely trust involved in science, but trust is a thing that can be broken and reshaped and is allowed to be examined and verified, where faith is not. Besides, if trust is broken, that's considered a fault on the part of the person providing the information. However, if faith is broken, that's considered a fault on the part of the person receiving the information.

But word choice is important - by saying "science relies on faith", the underlying agenda is that other things that rely on faith should be given the same level of trust, even though they don't expose themselves to the same level of inspection.

[danieltillett]

There are two types of science that are fundamentally very different to each other; observation science and experimental science. The base thinking of the two types of science are very different and even the people drawn to each type are quite different.

Physics is interesting in that both types of science are successful, where in say molecular biology or geology only one type dominates.

[choosername]

of course the image that the nfl puts forth of strong men fighting battles with what might seem like intelligent strategies plays into the hands of military types. Players get points for shooting. Is this a trap question? Casting shows OTOH are all about mating and increasing the troop strength.

[gamesbrainiac]

The NFL is funded because it provides entertainment, and we all know how well that sells.

[malka]

and how well that prevents people from critical thinking.

[wolfgke]

The German word for entertaiment is "Unterhaltung", which is composed of "unten" and "halten" ("down" and "hold") - and that is what entertainment is: Holding the masses down (such that they don't ascend or revolt).

Guess who is thus interested in funding entertainment.

[chki]

I think that is not only incorrect considering the content but probably also the semantics involved. There are many German words that have "Unter" in them ("Unternehmen" --> Company; "Untersuchung" --> study) where it does not relate to the word "unten" in any way.

[kgwgk]

In fact it seems to be related to the "among" meaning of "unter". The etymology is similar for other languages:

http://www.etymonline.com/index.php?term=entertain

[justinsingh]

Because having freedom in how you spend your leisure time obviously means that taking away that optionality would lead to.. revolt!!! I just don't see your logic here. It seems outdated.

[mannykannot]

Boltzmann's theories were apparently rejected by some of his influential contemporaries (the philosopher Ernst Mach, for one) partly on the grounds that there was no direct evidence for the existence of atoms and molecules [1]. It has been suggested that this may have been a contributing factor in his suicide. If so, then it is sadly ironic that his suicide occurred between Einstein's publication of his famous paper that cast Brownian motion as evidence for the existence of atoms and molecules, and the experimental verification of his statistical model of the process by Perrin in 1908 [2], which quickly put an end to the doubting of the reality of atoms.

[1] http://jfi.uchicago.edu/~leop/AboutPapers/Boltzman.pdf [2] https://en.wikipedia.org/wiki/Brownian_motion

[yk]

Most physicists are experimentalists, who work on testable predictions. And most theorists work on easily testable systems, like solid state physics or nuclear physics. And even cosmologists and astrophysicists in are very aware of the problems of ascribing astrophysical phenomena to new physics, it just took 70 years after Zwicky's invention of dark matter before anybody took it seriously as a sign of new physics.

By contrast, the small bit work done on theories that have purely theoretical appeal is getting all the press, and in a way that needs to improve a lot before one could call it grossly misleading. Take for example the sentence,

    “The imprimatur of science should be awarded only to a theory that is
     testable,” Ellis and Silk wrote, thereby disqualifying most of the 
     leading theories of the past 40 years. 

Well, what is a leading theory supposed to be, 40 years ago the standard model was just that, a model that needed confirmation. The W and Z bosons were only discovered in 1983, the Top quark in 1994 and the Higgs two years ago. General relativity only had strong support in the high field regime with the discovery of the Hull-Taylor pulsar a year earlier. And as I said, most physicists did not work on fundamental physics in the last 40 years, but instead on stuff like the quantum Hall effect ( discovered by von Klitzing in the early 80ies) or the Giant Magneto-Resonance ( discovered in the late 80ies), both solid state effects with technological applications and Nobels in the last thirty years.

[kijin]

There are different kinds of "testable".

Some theories are so easy to test, a 4th grader in her science class can test it conclusively in half an hour. We don't even question these theories anymore, because they've been set in stone for centuries.

Some theories take millions of dollars and several years to test, but they're manageable by large research universities and governments. At the very edge of this range, we have things like the Large Hadron Collider testing the existence of the Higgs boson. That one actually cost a few billion dollars and several decades, but we could live with that.

Some theories will take trillions of dollars and several centuries to test. Perhaps it will be quadrillions of dollars and several millennia instead, or even several million years. Right now, we don't even know what kind of technology we will need to test string theory. Perhaps we'll need to evolve into something else before our brains can even imagine what it would take to test it. But eventually we'll get there, if we learn to accept that science is bigger than all of us and stop being so impatient.

The difference between these three kinds of "testable" are quantitative, not qualitative. It just looks qualitative because the quantities involved are so huge. But if someone cannot grasp the idea of a theory that takes aeons to develop and test, I don't think they're qualified to judge the merits of theoretical physics.

[elorant]

Nobody can figure out a way to test string theory.

Sure they can. Build a particle accelerator with one million times the power of LHC. Not knowing and not having the technology are two different things.

Fifty years ago nobody knew how to test the Higgs boson theory either. That didn’t stop them from moving on and when the technology permitted it run the tests and solidify the hypothesis. String theory didn’t just popped out of the heads of physicists who had nothing better to do with their time. Parts of string theory are already tested, the discovery of the new particle that was semi announced last week at LHC was predicted by a model of string theory. That’s how science progresses, with little steps. So don’t get pessimistic, scientists have done huge progress in the last hundred years or so.

Lastly, let’s not forget how little funding science gets these days and how many different areas they have to cover. It’s not just physics, you have cosmology, space exploration, biology, energy, just to name a few.

It’s not science’s job to answer philosophical questions. If for example philosophers feel powerless because of the multiverse theory we should stop theorizing it? If it turns out that life and everything we know is just the result of a mere stroke of luck, one universe with the proper conditions popping out from a variety of trillions, what should science do about it?

All those smart people hate to face the fact that what they're doing may be total bullshit.

It’s funny saying something like that in Hacker News. Everything each and every one of us does has the potential to turn out as total bullshit. Every single day we write code that turns to be total garbage. And yet we’re doing it. You should watch the “Particle Fever” documentary. There you’ll see the reactions of physicists who favor supersymmetry in the possibility that it proves wrong. Everyone is devastated but they accept it. It’s part of the job. You make it sound like these guys are disillusioned. Scientists know perfectly well the consequences of following a wrong path in their career but they do it either way and we should be grateful to them, not despise them for it.

[SonicSoul]

Physics used to be "hard science"

I'm not a physicist but pretty sure there's not one kind of physics. Applied physics is still hard science while theoretical physics does not lend itself as well to full scientific method. Both are highly valuable. When Einstein came up with special relativity he had no way to verify it in a lab but that turned out pretty well.

[scotty79]

> Nobody can figure out a way to test string theory.

That's why some consider it physics inspired branch of math. For now.

> why should they be funded?

Until we have room temperature superconductors they are note done.

[qb45]

> Physicists used to look down on other sciences for that reason.

Rumors say that the more practical physicists have always been looking down on cosmologists for the same reason.

[dinkumthinkum]

I don't think that is very charitable. I don't think physics looked down on things per se.

There are things that arise out of the mathematics and it seems like you think the answer is just ignore it since we do not have perfect tests right now. I agree that some claims are very grand without sufficient experimentation but I don't think we should throw the baby out with the bath water.

[letitgo12345]

I don't understand -- we've spend trillions on wars, lost trillions because of poor policies that line the pockets of pharma and wall street, hundreds of billions on sports and entertainment (which is not a waste but is not contributing to technology either...), and you have an objection to spending a few billion on producing knowledge?

[unit91]

You've got an equivocation [1], three red herrings [2], and one begged question [3].

Equivocation:

You used "we" to mean government ("trillions on wars"), then private citizens ("hundreds of billions on sports..."), then back to government ("a few billion on producing knowledge"). "Our" war spending is certainly not related to "our" watching ESPN.

Red herrings:

War spending, lining corporate pockets, and private entertainment are unrelated to whether -- and to what extent -- fundamental physics should be government-funded. In fact, one may be a peace-loving socialist and still not believe that this research is the best investment for a nation's finite resources.

Begging the question:

You've assumed that research into string theory, multiverse theory, etc. produce knowledge, which is the very question being debated here. Knowledge is generally regarded as justified true belief [4] (albeit with qualifiers [5]). At issue here is whether one's scientific beliefs are adequately justified without empirical support -- i.e., whether we're producing any knowledge at all regarding this issue.

As a disclaimer, I was a researcher at a government lab for 5 years. In many cases, I believe scientific research is worth funding. That said, we should take care to present sound arguments regardless of the topic.

1. https://en.wikipedia.org/wiki/Equivocation 2. https://en.wikipedia.org/wiki/Red_herring 3. https://en.wikipedia.org/wiki/Begging_the_question 4. https://en.wikipedia.org/wiki/Belief#Justified_true_belief 5. http://plato.stanford.edu/entries/knowledge-analysis/

[ohsnap]

bayesian reasoning is a fairly accepted form of epistemology these days, and it is a form of knowledge long prominent in non-western cultures. What is wrong with physcists/philosophers thinking about how this can be applied in science?

Seems a better critically minded approach is to consider their arguments - rather than brush them off by calling them desperate

[justifier]

yeah, that's a picture of the implied surface of the electron cloud surrounding an atom's nucleus

[justifier]

    An ordinary microscope, which employs optical lenses, could view objects
    smaller than the wavelength of light. An electron microscope could view 
    smaller things with greater clarity than an optical microscope, but still
    could not clearly view individual atoms.
    
    So Binnig and Rohrer decided to build their own instrument – something new 
    that would be capable of seeing and manipulating atoms at the nanoscale level.
    To do that, they began experimenting with tunneling, a quantum phenomenon in 
    which atoms escape the surface of a solid to form a kind of cloud that hovers 
    above the surface; when another surface approaches, its atomic cloud overlaps 
    and an atomic exchange occurs.
    
    By maneuvering a sharp metal conducting tip over the surface of a sample at an
    extremely small distance, Binnig and Rohrer found that the amount of electrical
    current flowing between the tip and the surface could be measured. Variations in
    this current could provide information about the inner structure and the
    height-relief of the surface. And from this information, one could build a
    three-dimensional atomic-scale map of the sample’s surface.

http://www-03.ibm.com/ibm/history/ibm100/us/en/icons/microsc...

[manigandham]

There's no "easy science"

Also, knowing what's testable requires testing itself. Science has always been about incremental fraying at the edges of knowledge to answer 1 little question and open up 100 more. Things might not be perfectly testable today but that doesn't mean they won't be in the future or that we can't someone make progress with theory and models.

[weinzierl]

I fully agree with your point, but I think the "picture of atoms" is not a good supporting argument. The IBM image[1] is nothing more than a visualization of measurements including a good measure of artistic license. It has not much to do with seeing in the usual sense.

[1] http://researcher.watson.ibm.com/researcher/files/us-flinte/...

[tbabb]

Every picture is a visualization of measurements.

[noobermin]

Worse, by his definition of "see", one might as well say no one has ever seen the wind.

[weinzierl]

I guess my original comment was just lost in translation.

In my first language I think most people would agree that seeing involves visible light and eyes. We say the blind can't see, because their eyes don't work. We say that we can't see in the dark, because there is not light.

Sometimes we use see in a metaphorical sense, like "the blind see with their fingers". No one would confuse the fact that blind people gather some information about their surroundings with the help of their fingers with the fact that blind people just can't see. It's not the same thing.

I could agree that one can say "seeing" in regard to an atom in a purely metaphorical sense or poetical sense but that was not my interpretation of the original statement.

Maybe in English it's just different?

[choosername]

I didn't ever think i'd seen the wind. It's not an idiom I'm familiar with, either.

[weinzierl]

Agreed, but looking at an artistic interpretation of an object is not seeing the original object.

You can take a picture of a car and you can see a car with your eyes. You can take measurements of an atom and visualise that measurements but you will never be able to see an atom.

To put it another way: Imagine you and me took a picture of a green car. In your picture the car is green, in mine the car turned out blue. Would you agree that your picture is more realistic than mine?

The color of the atoms in the IBM visualisation is purely artistic. You will never be able to say green atoms are more realistic than blue ones. I’d even go so far to say the same applies for radius and shape in the IBM visualisation.

“no one has ever seen an atom" and no one ever will, because atoms don't exist in the visual realm.

[gamesbrainiac]

This should be a quote.

[c22]

"Every picture is a visualization of measurements."

[thisrod]

People do take photographs of single atoms. It requires a very colorful atom and a very bright flash, but both those things exist. They're not particularly interesting photographs: atoms are smaller than a light wavelength and they look like dots. I've seen them before, although I couldn't find one in a minute of Googling. There are even movies, as described in this paper:

http://dx.doi.org/10.1063/1.36788

[weinzierl]

Abstract:

   We demonstrated interruptions of macroscopic duration in 
   a single trapped and cooled Ba+ ions’s 493‐nm 
   fluorescence. They are caused by transitions of the ion 
   into the ‘‘dark’’ 2D5/2 state.—Multiple simultaneous 
   jumps of three ions indicate cooperative interaction 
   with the light.

I haven't read the paper but I'd be interested in a quote from the part that supports the statement: "People do take photographs of single atoms. It requires a very colorful atom and a very bright flash,..."

I'm not aware of a definition of atomic color but maybe that's a real thing in analytical chemistry, spectroscopy or some other field. I wouldn't be surprised.

Probably it's just the wavelength of the emitted light.

[karmakaze]

I would say it's somewhere between these points. The IBM image[1] is a picture, constructed from tunnelling currents rather than light, much like a blind person can see a face with their hands. Since even in the usual sense we don't see the object but the interaction with it, I would classify them more similarly than not.