John Richard said:
... which means that the CMB will have to have been around us from the start.
...
Yes that is right! Caveat, we are talking about the mainstream picture here. there is a lot of evidence that it is right as far as it goes.
To get used to visualizing standard cosmology it can help enormously to play around with a cosmology calculatory. Several people here at PF have made their own----Jorrie, hellfire and I don't know who else. there are some threads where they discuss features.
but most of us use online java calculators by Ned Wright or Siobhan Morgan.
Go here
http://hubblesite.org/newscenter/archive/releases/2008/08/full/
http://scienceblogs.com/catdynamics/2008/02/liveblogging_the_high_redshift_1.php
It talks about a young galaxy at redshift z = 7.6 (if confirmed) observed as it was 12.8 billion years ago during the first wave of star formation
this is conference news, it hasn't been published yet AFAIK
So suppose you want to know the recession speed of a galaxy at z = 7.6
First try Ned Wright's
http://www.astro.ucla.edu/~wright/CosmoCalc.html
just plug in 7.6 for z, over on the left
what I get from Wright's is that the light travel time was 12.97 billion years which is a little off from the 12.8 they say in the press release but small discrepancies don't matter here. and I get that the age of the universe when the light was emitted was 0.7 billion years. Is that what you get?
the important thing to notice is over on the left Ned Wright puts in the default values of H=71, and Omega_matter = 0.27 and Lambda = 0.73. they are accepted parameter values, widely used. So when you go on to Morgan's calculator you should take them along with you and plug them in too. Morgan wants her students to work more and consciously put them in.
So then go here
http://www.uni.edu/morgans/ajjar/Cosmology/cosmos.html
and over on the left, put in 0.27 for matter, 0.73 for Lambda, and 71 for Hubble.
Then type 7.6 into the z box and see what you get. You should get some recession speeds!
Cosmology is largely non verbal. It is applying a mathematical model to the geometry of the universe, and fitting it to many different kinds of data.
That mathematical model
is built into Wright's and Morgan's calculators.
When you play with those calculators you are playing with the best model of the universe we have today.
Truly enormous amounts of data of many different kinds are used to challenge the model. it must fit to the data. that is how the numbers 0.27, 0.73 and 71 are determined. The more data, the more accurate these estimates get, and the better the fit.
What do you get for the speeds? I get about 3.3c and 2.3c. 3.3c is the recession speed of the galaxy THEN when it emitted the light (some 12.8 billion years ago) and 2.3c is the recession speed NOW as the light is arriving to us and coming down the telescope.
The distance now should be greater than the distance then by a factor of 1+z which is 8.6.
How about checking to see if that is about right? 1+z is the factor by which the light wavelengths are stretched, so it must be the factor by which distances expanded in the universe while the light was in transit to us. If the ratio doesn't work out right, please tell me!
==========================
I gathered a bunch of useful links here:
https://www.physicsforums.com/showthread.php?p=1610331#post1610331