Originally posted to sci.astro on Dec 10, 2000.
David A. Smith wrote about a question on radiation from distant galaxies and quasars:
>> The light from distant sources is not scattered >> (the images are clear, if faint), ...
Jeannette Bos then asked:
> But if light from distant sources isn't scattered, why do we see in all > directions (more or less) the same cosmic background radiation?
Both are correct; there is some confusion because the word "scattered" can correctly be applied to the CMB, and correctly be raised in discussions of light from distant galaxies. Here's how:
The CMB is thought to arise from thermal radiation in the very early universe, when all matter was (nearly) uniformly hot and dense. In the earliest times, when this temperature was over three thousand degrees, all matter was ionized. Free electrons scatter electromagnetic radiation efficiently, so the universe was "opaque" -- photons couldn't move very far without bouncing off an electron and changing direction. It would have been difficult to see distant objects, had anyone been around to try.
As the universe expanded, the temperature of gas within it dropped. When the temperature dropped below about three thousand degrees, the electrons started to combine with free protons and helium nuclei to form neutral atoms. Neutral atoms don't scatter light nearly as well as free electrons, so suddenly photons were free to fly long distances through the universe. The moment when the universe changed from mostly ionized to mostly neutral was the last time that most of the photons present at that time scattered off free electrons. The gas at that time is sometimes called the "last scattering surface". All the CMB photons we see now have been flying through the universe, largely untouched, since that time.
So, we see CMB photons which were last absorbed/emitted/scattered from the "last scattering surface", when the universe started to turn neutral.
"Scattering" appears in a different context, in some discussions of the physical basis for the redshift of light emitted by distant galaxies and quasars. These galaxies were formed long, long, long after the universe turned neutral -- probably a billion years later or so. The light they emit comes from ordinary stars, not from a fog of hot gas permeating all of space.
In _this_ context, some people claim that the observed redshift is not due to a Doppler effect (due in turn to the expansion of the universe), but instead due to some sort of interaction of the photons with matter along their long, long journey across the universe. They claim that some sort of atom/photon interaction (or "scattering") causes the wavelengths of the photons to shift.
However, if this were happening, then the photons would shift not only in wavelength, but in direction. That would cause any sharp point source to appear blurry and diffuse; moreover, this blurriness would increase with redshift, in this model. Several astronomers have searched for a redshift-dependent blurriness, and failed to find it. Hence, most astronomers do not believe that the "scattering" model explains the redshift of distance sources.