The Multiverse Falsified

The July 1 issue of the Monthly Notices of the Royal Astronomy Society includes an article evaluating the standard multiverse prediction of the cosmological constant, with result:

The predicted (median) value is 50–60 times larger than the observed value. The probability of observing a value as small as our cosmological constant Λ0 is ∼2 per cent.

If your theory only makes one prediction, and that prediction is off by a factor of 50, that’s the end of it for your theory. I’m very glad that this has now been sorted out, the multiverse hypothesis has been falsified, and theorists who have been working on this can move on to more fruitful topics.

Update: As David Appell realized, the last sentence here was sarcasm (or maybe black humor). Those promoting the multiverse are doing Fake Physics™, not Physics. This is ideology, not science, and there is no chance that they will stop referring to the “successful multiverse prediction of the CC”, no matter what analysis shows a seriously incorrect prediction.

As Blake Stacey points out, this paper was on the arXiv back in January (see here), and has just been ignored by multiverse proponents. Part of doing Fake Physics™ is ignoring any information that contradicts what you want to believe. Another commenter points to this 2014 argument from Sesh Nadathur, which similarly as far as I know has just been ignored.

After appearing on the arXiv in January, this latest work was promoted by press release from Durham University back in May, which led to lots of media stories (e.g. here). For some reason, the press release didn’t really explain that this work falsifies the usual claim that the value of the CC is evidence of a multiverse. Instead, the work was promoted as showing that the multiverse is “more hospitable to life” than thought, which sounds good I guess, but seems like a bizarre way to explain the significance of this work.

For various sensible explanations of what is really going on here, see Jim Baggott, Philip Ball, and Sabine Hossenfelder. I’ve often repeated my own version of how to see there’s a problem with trying to explain the CC this way. There is no actual multiverse theory, so proponents assume a “flat measure over the anthropically allowed region” and then calculate. This is exactly the same input as my theory of the CC, which is that I have no idea what is going on, so any value is equally likely. The bottom line from the latest work on this is that, even if for some reason you believe you can get a sensible “prediction” this way, the prediction comes out wrong.

This entry was posted in Multiverse Mania. Bookmark the permalink.

15 Responses to The Multiverse Falsified

  1. David Appell says:

    I’m guessing your last sentence is sarcasm.

  2. Anon says:

    Hi Peter

    Do you know if the situation has gotten worse for the multiverse since this 2014 blog post that also appears to have similar arguments like you quoted:

  3. Daniel says:

    There’s a new article at Aeon which I think you might be interested:

  4. Mitchell Porter says:

    It remains true that Weinberg did successfully predict the size of dark energy, to within an order of magnitude, in his paper “The cosmological constant problem”. It’s there at the top of page 8: “we would expect a vacuum energy density rho_V ~ (10-100) rho_M0”. (The actual rho_darkenergy is something like 14 times rho_baryon.)

    You can take the position that he was right just by accident, and that the size of dark energy is determined by something else, or determined by nothing at all. Of course that’s possible.

    But it also remains very possible that he was basically right! The subsequent literature on the anthropic prediction of the cosmological constant just reruns his argument in a variety of concrete scenarios. This paper tells us how the argument fares, for a particular cosmological model, under a variety of auxiliary assumptions. Sometimes it works and sometimes it doesn’t.

    If I was an anthropic theorist, the conclusion I would draw from this paper, is that I should go back and study closely the details of Weinberg’s original argument, because it worked perfectly when he did it, the first time around.

  5. Anonyrat says:

    Mitchell Porter:

    The Sesh Nadatur comment referenced above is that the Weinberg argument doesn’t work with the current earliest time of galaxy formation, i.e., current observations — the maximum redshift at which a galaxy has been found. It is not just that Weinberg’s argument fails with the current cosmological model, it is that he got lucky with what the state of observational knowledge was when he wrote that paper.

  6. Peter Woit says:

    Mitchell Porter,
    I’ll leave this to the experts, but my impression was that the claim of the new paper is that Weinberg’s calculation is wrong (doesn’t agree with what we now know). In any case, the two possibilities seem to be:

    1. Weinberg’s calculation was wrong. Multiverse proponents will just ignore this and continue to tell people to believe in a multiverse because of Weinberg’s “successful” calculation.

    2. Weinberg’s “prediction” was fine, but you can get pretty much any “prediction” you want by changing your uncheckable assumptions about how the multiverse works. Again, calling this sort of “prediction” a success would be misleading.

  7. Peter Woit says:


    That is interesting, will likely soon discuss it in a posting.

  8. Lee Smolin says:

    Dear Peter,

    I’ve been making this point in books and reviews for many years. (For example hep-th/0407213). A further point is that if you expand the list of constants to be determined by Weinberg’s anthropic argument, from just the CC to the CC and Q, which measures the size of the primordial density fluctuations, the discrepancy between the predicted most likely value and the observed values grows to many orders of magnitude (See Figure 2 of astro-ph/0401424.)



  9. Peter Woit says:

    Thanks for pointing that out Lee. Yes, that’s another serious problem with the oft-repeated “Weinberg’s successful prediction” claim.

  10. Low Math, Meekly Interacting says:

    The undead quality of the “Weinberg predicted the CC” argument ranks very high on the list of habits of multiverse proponents that puzzle me. As it was apparently quite rigorously dispatched years ago, what gives? It’s not like SUSY, where an excess of flexibility makes falsification impossible. The multiverse itself has much the same virtue for those who wish to move the goalposts forever. So why bother with a very narrowly-focused prediction that HAS been falsified? Has Weinberg himself ever said anything publicly about the current state of his prediction?

  11. Peter Woit says:

    To his credit, I don’t think Weinberg himself really does this, and his comments on the multiverse always seem to be of the order of “maybe yes, maybe no”, not ever “my successful prediction shows it’s true”. For many others, especially multiverse ideologues, being able to invoke Weinberg’s name and the words “successful prediction” is just irresistible and there is nothing that will ever get them to give this up.

  12. Uncommon Sense says:

    Does any of this relate directly or analogously to Everettesque/ Many-Worlds QM?

  13. Peter Woit says:

    Uncommon Sense,
    No, not in the slightest. They are two completely different things.

  14. Blake Stacey says:

    Just to update the argument from 2014:

    The highest-redshift galaxy so far observed, as best I can tell from the literature (it’s been a long time since I worked in astrophysics) is at z = 11.1. See Oesch et al. (2016). Following Weinberg’s argument, this gives a bound on the vacuum energy density of about 5800 times the present cosmic mass density. This is three orders of magnitude larger than the observed value, a ratio well into the regime where Weinberg himself says the cosmological constant would be “so small that even the anthropic principle could not explain its smallness”.

Comments are closed.