A long time ago, people tried to explain the redshifts without
expansion of the Universe, or Doppler effect. They proposed ordinary
Raman effect, but it blurs the images and the spectra, thus it cannot
work.
 In 1968, redshifts were observed by an interaction of ultrashort laser
pulses with matter. This observation  led to the development of the
"Impulsive Stimulated Raman Spectroscopy" (ISRS) about which there are
lots of publications.
 Making, for my students a classification of the linear light matter
interactions, I found a white box that I filled by an optical effect
which I name now "time-Incoherent Light space-Coherent Raman Scattering"
(ILCRS) This effect redshifts the hot light beams of ordinary incoherent
light (Hot being defined from Planck's law) and may be confused with a
Doppler effect: it does not blur the images, nor the spectra, and the
relative frequency shift \Delta\nu/\nu is nearly constant in the
spectra. More, the energy lost by the frequency shift is transferred to
cold beams (In particular to the thermal 2.7K radiation).
 ILCRS is an avatar of ISRS corresponding to the replacement of the
short, powerful laser pulses by the much longer and less powerful pulses
which make the ordinary incoherent light. This replacement has
quantitative effects:
- To produce ILCRS, the matter must be very low pressure gas
- This gas must have Raman transitions in the radiofrequencies region;
thus it must have hyperfine structure; it may be molecules having an odd
number of electrons, any atom or molecule with Stark or Zeeman
effect,...
- It has a qualitative effect: while ISRS is nonlinear and produces
redshifts which depend on the intensity, ILCRS is linear, it does not
distort the spectra.

A consequence of ILCRS is that molecules which are destroyed by their
first collision, such as H2+ cannot be observed !

 The two main "proofs" of the big bang may fail!

 You can see references (in good reviews of optics) in my last
manuscript: arXiv:astro-ph/0110525.

 I am a criticist of the principles of quantum mechanics (not of its
formalism which I taught 40 years); if you are interested, look for my
manuscript on arXiv.

 I wish to discuss of those problems; if you know names of interested
people, please send me references.

With best regards
Jacques Moret-Bailly