Limits to Physics Insight

I’m a little late to the party, as this was covered in the New York Times back in June, but the insight stuns me — we may well not have enough data in our corner of the universe to learn important physics:

Our successors, whoever and wherever they are, may have no way of finding out about the Big Bang and the expanding universe, according to one of the more depressing scientific papers I have ever read.  If things keep going the way they are … in 100 billion years the only galaxies left visible in the sky will be the half-dozen or so bound together gravitationally into what is known as the Local Group, which is not expanding and in fact will probably merge into one starry ball.  Unable to see any galaxies flying away, those astronomers will not know the universe is expanding and will think instead that they are back in the static island universe of Einstein. …

It makes you wonder just how smug we should feel about our own knowledge.  "There may be fundamentally important things that determine the universe that we can’t see," Dr. Krauss said in an interview. "You can have right physics, but the evidence at hand could lead to the wrong conclusion. The same thing could be happening today."

I read about it in the latest Scientific American, and there is also a short article in a recent New Scientist.  Corey Tomsons blogs that this is not a new insight:

This is a ‘rediscovery’ of sorts. There are two other papers which cover this type of analysis, although Krauss and Scherrer provide the most comprehensive short treatment.

… Librarians: You might start thinking about how to store cosmological data for the next 100+ billion years. Acid free paper won’t cut it, so think big. … No matter the record, it would doubtless be received with skepticism. Think of the amount of trust future scientists would need to possess if they were to take seriously any surviving arXiv records about the Big Bang, and how empiricists of the time would heap scorn upon them. This is tremendously unsettling, but ‘m fascinated by the idea that in the distant future, the Big Bang theory would become akin to Intelligent Design – a theory contradicted by the best evidence.

To turn it around, there may well be physics that we can only learn via records saved by ancient aliens – which gives us another good reason to seek such aliens.  Krauss was at first scheduled to speak at that recent SETI workshop – it would have been fun to talk to him about this.      

GD Star Rating
Tagged as:
Trackback URL:
  • Ian C.

    There’s also this:

    Some of our naughty spacecraft are not doing what Einstein predicted. I guess we have a long way to go…

  • Robin Hanson wrote: … which gives us another good reason to seek such aliens.

    Is there any reason to suspect that this would be a fruitful endeavor? Our sun is still wasting order at an atrocious rate, and as far as we can tell so are the other stars in our vicinity. The likely explanations for this that occur to me are:
    1) No intelligent aliens ever existed in this universe, period.
    2) They existed, but not in our past light cone. We might meet them or find their remnants some day, but it’ll most likely be very far into the future.
    3) They existed in our PLC, and self-destructed in some way that didn’t leave behind a superintelligence interested in gobbling up any stars within reach.
    4) They existed, and found a way to acquire a lot of cputime and storage space without gobbling up all matter within reach, making the universe uninteresting from a resource standpoint.
    5) They’ve turned our solar system into a preserve, designed to be hard to detect from inside.

    In case 1 it would be pointless to go looking for them. In case 3 there’s likely to be no storage devices left from their civilization, unless they deliberately committed self-genocide. Possibility 4 requires that there’s something seriously wrong or incomplete about our understanding of physics; it’s hard to estimate how likely this is.
    In case 5, looking for them would likely be pointless; if they’re going to such lengths to hide from us, they’re unlikely to come out of the woodwork if we try looking for them a bit harder.
    Case 2 transitions into one of the other cases or a clear alien contact at some point in the future.

    Personally, I think 1) is the most likely explanation.

  • Great comment, Sebastian! I highly recommend you start a blog to collect and centralize your comments of this caliber on the writings of others, if nothing else.

  • conchis

    I’m curious about the extent to which we should really be worried about a situation where we might believe things that are “wrong”, because all the evidence that might tend to disconfirm the wrong theory is evidence we’re incapable of experiencing. If the theory actually has no observable effects, in what sense could having better knowledge matter to us? Isn’t the theory sort of locally right?

    I accept that you might think that the lessons we could learn could generalise to domains where they do matter, so I’m not suggesting that this is a totally uninteresting possibility. But are there reasons to get particularly worried about this?

  • Adam Safron

    Sebastian: I agree with your conclusion that seeking alien civilizations is probably a fruitless endeavor, but think possibility 2 may be more likely than possibility 1. If intelligent life evolved once every couple thousand or million galaxies (that’s a lot of planets…), then it may be very difficult to detect the existence of these civilizations.

    And if we ever did encounter a vastly older civilization that wasn’t in the process of forcefully turning us into computronium, their cosmological records would be one of the last things we’d want them to show us. We certainly couldn’t extract any useful models from the data that had not already been found by this older civilization. “Show us your data” would be one of the silliest requests you could present to a more advanced civilization.

  • Nick Tarleton

    4.5) They existed, and the ones in our past light cone that didn’t wipe themselves out decided for whatever reason not to acquire a lot of computronium (and there are few enough of them that all of them chose similarly).

    Though I also think (1) is most likely.

  • George Weinberg

    , there may well be physics that we can only learn via records saved by ancient aliens

    Well, no there can’t, because alleged records from bug-eyed monsters billions of years in the past are not the same as actual observations. Anything that could be learned in this way would be mythology, not physics.

  • I’m very skeptical of this sort of claim — you have no idea what future civilizations can’t know. There could be all sorts of ways to deduce a big bang given more knowledge. What about just the second law of thermodynamics alone?

  • Science, as we have known and practiced it, is only a few thousand years old. Doesn’t it seem a little premature to speculate about how “they” will practice it millions of years hence, if at all?

    As for the problem of long-term archiving, I defer to Linus Torvalds:
    “Only wimps use tape backup: _real_ men just upload their important stuff
    on ftp, and let the rest of the world mirror it ;)”

  • Nick Tarleton: It’s a possibility, but it strikes me as very unlikely. Note that:
    1) A single civilization that decides to do otherwise would break this.
    2) This doesn’t need to be a decision of the whole civilization; if an arbitrary subset with access to sufficient technology decides to do this, they could just head out some way into space, and start converting matter once they’re far enough away that their growth curve will hand them a tactical advantage over the main civilization before the authorities notice.
    To prevent this, the civlization would either have to be a lot more homogenous than present-day humanity, or feature very effective controls against this sort of thing on their citizens.
    3) Similarly, in the absence of free storage+computation (case 4 in my previous post) a significant subset of UFAI designs (paperclip optimizers, wingnut optimizers, red stapler optimizers, …) is likely to turn into matter-gobblers.
    4) Natural selection has a tendency to produce organisms that try to maximize the resources available to them/other organisms with a similar genome (like their offspring).
    5) For very many possible goals, how well you can implement them will depend on how much storage space and cputime you can get your hands on. In our universe, cputime sems to be limited largely by the laws of thermodynamics.
    In many ways, the conservative thing to do is to turn off any accessible stars as early as possible, and take your time to think about what to do with the resources afterwards. If you don’t do this, they’ll continue to burn through order you could have put to much better use.

    Greg Egan’s Dispora features a multiverse with a number of alien civilizations, all of which deliberately (and at least in the case of human-derived civilizations, by individual choice) avoid gobbling up lots of resources. I consider that one of the least realistic aspects of the book.

  • A lesson in humility

    In a little while, our universe will look different.

  • The idea that creatures descended from us even a thousand years hence, never mind billions of years, bear any relation to us whatsoever strikes me as fanciful imagining. Technological change accelerates – there is simply nothing to contemplate about those entities that has the remotest chance of being true.

    It smells like a search-for-god-anywhere-else exercise. And to think we disdain medievals for debating how many angels can dance on the head of a pin! We at least know what a pin is, and a dance is, and have some general idea what an angel is thought to be.

    There is no epistemology in this, just stoner science-fiction.

  • Meaux

    100 Billion years, eh?

    Makes me feel, well, suicidal.


  • Pseudonymous

    Our successors, whoever and wherever they are, may have no way of finding out about the Big Bang and the expanding universe, according to one of the more depressing scientific papers I have ever read.

    Surely this is a matter of history rather than physics. And we have always known that historical knowledge can be lost.

  • My guess is also that no aliens is the most likely scenario, but the value of finding them seems so huge as to justify looking even if the chances are low. I’m not endorsing any strong claim of the form we can prove that our descendants couldn’t know – just that this is a nice clear plausible example of something they may well not know.

    Assistant, we bear a great deal of relation to any creature that exists in the same physical universe as ours.

  • michael hilton

    I dont believe in the big bang idea .
    That said
    I do believe that human knowledge of physics and cosmology will always increase , no matter what the arrangement movement and detectability of galaxies .
    I believe there are more ways to discovery than our present cosmology .
    There will be more new and counter intuitive models proposed ,
    some of which will reveal astounding new understandings .
    A look at past science shows certain such points of innovation around the examples of the works of
    Copernicus , Newton , Einstein , past and current quantum physicists and cosmologists .
    My personal hunch is that the elctromagnetic spectrum will be
    understood as a loop rather than an open continuum ,
    and that there are , will be , and also there were , intelligent beings in many and varied stellar habitats .
    The wonder of life on this planet , the awareness of it and the inherent wish to explore and understand ,
    is to me a signature of the intrinsic nature of the universe .

  • Sheila Daweila

    I had a thought about mathematical probabilities this morning. If we counted every grain of sand on the sea shores of every ocean on earth and calculated the odds of another planet with all the life forces of earth forming from just one of those grains of sand, would it ever happen? Could it ever happen? What if we gave it a lot of time? Just because we can develop calculations from such a large base doesn’t mean it is ever going to happen.

  • Pingback: Physics limits | Sugarmunkee()