Are Our Scientific Models Direct Representations of Reality?

[newb]

When we see structures like DNA for example is it to be understood that if we had a microscope powerful enough to see detail at the level of a DNA molecule we would see the double helix structure? Or are our structures simply models to help explain experimental results?

Either way, at what level do our models NOT represent things we would actually see and are just models at best? For example we know that an atom is not a point like object with an electron orbiting it like a moon orbits a planet, my question is at what point can we confirm that our models are actual physical representations of our world?

 

[vanesch]

Ha, interesting question. I would say that ultimately the representation we have of the world is always somehow a mathematical structure, even though we don't always realize that.

Now, one can consider your question on two levels. The first one is: are atomic models things we would *SEE* if we just could see such fine details ? There, the answer is simple: no. Because what we call "seeing" is an approximate intensity distribution of light on a surface (retina, photographic plate, CCD, ...), and such an intensity distribution doesn't make much sense on a resolution below the wavelength of light, which is of the scale of several hundreds of atom diameters. So there's no physical way in which we could actually see atoms with light. Ok, you might say, but what if we use "light" of a shorter wavelength (X-rays). Well, the funny thing is that you now get peculiar interactions between the atoms and the X-rays, which will give you "images" which are not faithful representations of the "geometrical picture" but things such as diffraction images. If you go to still shorter wavelengths, you will have "light" that will be so energetic that it will actually alter (damage) your atoms, like wanting to look at a painting using an automatic gun or something. So we have no optical direct means to "see" atoms.
There are other tricks which DO work better. Electron microscopy for instance. One can more or less "feel" an atom with a scanning electron microscope, but what this apparatus actually does is to measure electrical equipotential surfaces, not so much "visible pictures". In that respect however, one does "see" structures which ressemble the laboratory models of molecules somewhat (bubbles attached to one another).

On a deeper level, one should see these models as geometrical and visual aids to the description of a mathematical structure that is used in certain theories about atomic structure, which succeed in making a lot of good predictions of experimental results. So then, they are nothing but a visual mnemonic technique to help you deal with the mathematics.

 

[f95toli]

I guess it depends on what you mean by "real" and "see".
DNA is a fairly big structure; the diameter is something like a 2 nm and it is very long. This means that it is much bigger than some of the artificial structures we routinely create nowadays; e.g carbon nanotubes (diameter about 1 n). There are several types of microscopes that can be used to image DNA (although it don't think think you can actually tell that it is a double helix).

Also, the current limits of lithography for electrical circuits is only slightly bigger than DNA, somewhere around 10nm (32 nm for commercial processes). Hence, you can't really "see" the components on a microprocessor either since they are smaller than the wavelength of a visible light.

Personally, I'd say that the structure of DNA is real, but that is ultimately a philosophical question.

 

[atyy]

At the level of a piano note. It is a model of a piano that you hear, not see.

 

[Naty1]

Either way, at what level do our models NOT represent things we would actually see and are just models at best?

I tend to see this as Vanesch...math and theory trump observation. Most experimental confirmations of models involve somewhat indirect "observations" inferred from our theoretical knolwedge..
I can't think of any theoretical knowledge that is based entirely on observables...you can't "see" any of the forces, none of the elementary particles, and so while you can see planets, stars, galaxies, sunlight, most of their interactions are based on unobservable components...forces and time for example...How does a battery work, or electricity...

Your mind is your greatest tool, not your vision...

 

[f95toli]

I tend to see this as Vanesch...math and theory trump observation. Most experimental confirmations of models involve somewhat indirect "observations" inferred from our theoretical knolwedge

Does that mean that you also consider images of the components in a microprocessor to be "indirect observation"?
Remember that we use the same tools to image them (SEM, TEM, various types of SPM) and DNA molecules and they are not that different in size.
Note that I am not saying that this point of view would be "incorrect", it is just that I think it adds an unnecessary layer of abstraction.

Many of the components I work with are of the order of a hundred or so nm in size and sometimes even smaller, but I do consider images of them that we obtain using an AFM to be as "real" as images from an optical microscope, partly since I designed the masks etc that determined the shapes that appear on the AFM image.

 

[Crazy Tosser]

http://www.nocaptionneeded.com/?p=561

second from top

and this
http://scienceblogs.com/bioephemera/2008/04/juxtaposition_5.php

 

[Naty1]

Does that mean that you also consider images of the components in a microprocessor to be "indirect observation"?

well i don't want to get into a semantic discussion because I was making a rather vague general comment, but I'd agree images of the type you mention are pretty direct...say an x-ray of a broken bone...

I was thinking more of things like scattering experiments, cloud chamber condensation, and indirect observational evidence for black holes...where the interpretation of effects is not so direct...

 

[Vanadium 50]

I don't think there is a whole lot to be gained by trying to distinguish between a direct and and indirect measurement. In the examples given, scattering with light is direct, but scattering with x-rays is not - but absorption with x-rays is direct. I think it's pretty clear that drawing this distinction tells us more about what the person drawing it finds comfortable than anything physical.