Thursday, April 15, 2010

Physics Quote Of The Day.

The opening statement to Physical Foundations of Cosmology by Mukhanov reads:
The most important feature of our universe is its large scale homogeneity and isotropy. This feature ensures that observations made from our single vantage point are representative of the universe as a whole and can therefore be legitimately used to test cosmological models.
What an important observation.  The only reason we have any reasonable chance of understanding the universe we live in, given we can only observe it from one small spec called earth, is that the universe is so darn homogeneous and isotropic.  (See here if you don't know what that means.  Basically it means the universe looks the same everywhere inside.)

 This ensures that whatever we see here on our little "Pale Blue Dot" (See video below), we can assume applies to everywhere else.  This allows us to create physical models that account for the whole observable universe.

This means, though we live on only a tiny spec, what we see here applies everywhere.  This is vital for us to be able to say anything at all about the entire structure.


  1. It's more of an assumption than an "observation," wouldn't you say? What we observe is that our *past light cone* is homogeneous and isotropic. But as you point out, our past light cone comprises only a tiny portion of the Universe. Why do we believe the same properties hold everywhere else? The only good reason, as far as I can see, is that otherwise we simply can't do any interesting cosmology.

  2. I'm nearly finished with Sean Carroll's "From Eternity to Here" and I just read about the role inflation plays on this homogeneity. Fantastic book, though a good deal of it just goes whoosh right over my head. I wonder how different the Universe would be had Schrodinger been a dog person. =:)

  3. Bryan,

    It's true that homogeneity and isotropy started out as assumptions. However, many observations confirm that the universe is homogeneous and isotropic to one part in 10^5 above 100Mpc scales. Examples of experiments that demonstrate this are:

    1. Studying the CMB itself with WMAP.
    2. Studying large scale structure with things like Sloan Digital Sky Survey.

    See Dodelson or any other modern text book treatment of cosmology for more details on experimental evidence for this.

    However, Mukhanov (the book I quoted) points out for most inflationary theories, homogeneity and isotropy break down on "super-large" scales.

    Many of our readers may roll their eyes as this being philosophy, but again, I tend to agree with predictions of theories that have a consistant track recored of being correct.

    People are free to suggest any theory *at all* that is as successful as inflation in making verified predictions but that, however, does not *also* predict things like a larger structure (call it a multiverse) with the possibility that on the largest scales homogeneity and isotropy break down.

    I have a feeling I will be waiting until the day I die for people to find something to replace inflation. It's pretty darn solid so I suggest people getting used to the idea that our universe is only a "relatively small" inflated patch of some much larger structure.

  4. Stan,

    Yes, inflation is the best understood mechanism for why the universe should be homogeneous and isotropic. (Otherwise you have to just add it in by hand "ad hoc" as they say).

    But again, if we live on only a smal patch that inflated, the rest of the "multiverse" may or may not have these nice homogeneous and isotropic attributes. (Again, unless someone cares to explain why the universe/multiverse "just happens" to be this way with some data to back it up.)

  5. The simplest way to think about why inflation forces this is: take a crumpled up ballon in any configuration. Now inflate the ballon. The final product is nice and homogeneous and isotropic.

    This is only a crude way to think about it, but for most people, it should be good enough.

  6. Joseph: The observable evidence you cite is still inevitably about our past light cone. Which is a just a tiny, tiny fraction of the universe. Of course *that's* confirmed -- but the claim that the rest of the Universe is this way is not observable. Not confirmed. And our reasons for believing it must be much more subtle.

  7. A thought on your comment: the way you put the balloon example, it isn't really clear why people posit inflation (an early period of rapid expansion) in *addition* to the standard expanding universe cosmologies.

    I'd put it this way. There are large patches of the blown-up balloon that display similar homogeneous and isotropic matter distributions. This similarity suggests (to some) that they should have come from some common interaction -- an intersection of their past light cones -- sometime after the big-bang singularity. But if you assume that the balloon has always been blowing up at a constant rate (as the FLRW models do), then there exist similar patches that have *never* interacted, because their past light-cones don't ever intersect. More fancifully: if you wind things backward, these regions "hit the singularity before they touch each other." So, people *force* them to intersect, by positing a rapid-expansion period, in which the balloon was blowing up much more quickly than it is in the current epoch.

  8. Bryan,

    Exactly! I'm sorry if my words conveyed something different.

    1. The whole "observable" universe is homogeneous and isotropic. (Or that portion in our past light cone.) We don't know about the rest.

    2. Except, we have every reason to believe inflation created much more than our observable universe. That should be homogeneous and isotropic too.

    3. Which leads to the idea that homogeneity and isotropy may break down on the largest scales, since, like you say, we only know that a patch of the balloon has inflated. The rest may not an therefore, far, far away things may nit be homogeneous and isotropic. (In other words, on very large scales.)


To add a link to text:
<a href="URL">Text</a>