Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Pre-BB Evidence

  1. Jul 30, 2015 #1


    Staff: Mentor

    Two recent PF threads were interesting IMO. Why no time before the BB? and
    A Farewell to Falsifiability

    Here on PF I can read about collapse-rebound, quantum fluctuations, multiverses, and many more speculative theories about pre-BB eras or universes beyond the observable. But it is my understanding that we have no observational evidence to support or refute these theories, and also no prospects for obtaining such evidence.

    No disrespect intended, but total lack of evidence or prospects for evidence makes it impossible to separate these theories from "turtles all the way down."

    But Professor Susskind once said, "Physicists aren't interested in what is true. They want to know what is useful." In his lectures on string theory he also said that some of the math developed by string researchers has been useful in refining the standard model, even if string theory itself is not yet falsifiable. The point being that there can be serendipitous side benefits to research into unprovable theories.

    My questions for researchers in these fields are:

    What kinds of observational evidence do you prospectively hope to see?

    In what ways might research in these theories become useful even the theories can't be proved?
  2. jcsd
  3. Jul 30, 2015 #2
    "Physicists aren't interested in what is true. They want to know what is useful."

    I agree with this. Real science should not drift from demanding theories make testable predictions and abandoning theories that do not.
  4. Jul 30, 2015 #3


    User Avatar
    Science Advisor

    I don't know. Short of engaging in a sophisticated philosophical debate about what is "truth", I think this view is a cop out. In fact, it seems a restatement of George Box's famous quote that "all models are wrong, but some are useful." The difference is that Box was a statistician. No, I think scientists should care very much about what is reflective of reality, to the extent that this is possible. We want to understand how the universe works, not simply how to calculate cross sections.
  5. Jul 30, 2015 #4
    I disagree. I think it is more of an acknowledgement of the high probability that whatever model we are working with is not completely general and perfectly accurate.

    The really interesting part of physics (for me) is probing the limits of applicability and accuracy of different models. All that would be much less exciting if I thought a theory I was working with was perfectly accurate and had no limits of applicability.

    I am happy enough to find more accurate theories with broader applicability, because these are more useful. Are they true? I don't think one can make truth statements about scientific theories without discussing the error bars and scope of applicability.
  6. Jul 30, 2015 #5


    User Avatar
    Science Advisor

    I don't disagree that we must be aware of the limits of our models. But Susskind's quote seems to suggest that we should be content with the model so long as it's useful, regardless of how close it might come to truth.
    I agree. Nothing I've written should make you think otherwise.
    No, they are more applicable. Define "useful".
    Of course one can't. I haven't written anything that would suggest that uncertainties in scientific measurements aren't vital for understanding the accuracy and applicability of a model.
  7. Jul 30, 2015 #6
    I guess I don't distinguish between applicable and useful, at least not as a practical matter. I use non-relativistic classical mechanics very often in most of my work in ballistics and blast. I use these ideas to invent new devices, investigate models, predict the outcome of experiments, and publish papers.

    However, I have too much knowledge about quantum mechanics and relativity to be confident in absolute truth statements about classical mechanics. I know two much about QED and limitations of the Correspondence Principle to be confident in absolute truth statements about quantum mechanics. And so on. This kind of thinking has served me well by empowering useful doubt when entering new fields when the limits of applicability and potential falsifiability of ideas held with confidence have not been adequetely explored. My skepticism may not always be true, but it sure is useful.
  8. Jul 30, 2015 #7


    User Avatar
    Science Advisor

    That all seems very sensible. I think we might be arguing different things. I agree that each theoretical construct is an approximation with limits, and that falsifyability is a key component of the scientific method. Those are practical concerns whereas I take Susskind's quote to be more philosophical. When we do a QED calculation, do we not care whether the lines in our Feynman graphs correspond to actual particles? Is the "electron" merely a computational device that's "useful"? Or is it something that makes up the real universe? Scientists should be attempting to *describe* how the universe works; a collection of "useful" models and abstractions does not support that pursuit.
  9. Jul 30, 2015 #8


    Staff: Mentor

    Everyone, please keep the thread on topic. The actual questions asked in the OP are not about the philosophy of physics; they are practical questions about string theory specifically:

    Please focus the discussion on these questions.
  10. Jul 30, 2015 #9


    User Avatar
    Gold Member

    All theories are unproven until you put substantial effort into going about proving them. So choosing unproven and unproveable theories and conjectures is a very important part of the process. On one hand, you have to identify "not even wrong" theories that aren't worth pursuing, and on the other hand, you have to acknowledge that behind every research agenda is an unproven theory, in order to recognize that there are risks into devoting all of your efforts on it without considering alternatives or trying to generate alternatives to your assumptions.

    Most really bad mistakes made by smart people flow from false assumptions, not from faulty logic or flawed execution of experiments.

    Probably the most useful purpose of theories that can't be proved due to the impossibility of acquiring observational evidence is that it provides researchers with intuition on what theories for which observational evidence is available are worth pursuing because of Baysean priors driven by the theory about what important problems are truly unexplained and about what further investigation is likely to reveal.

    For example, if you have some pre-BB theory that makes it sensible and natural for there to be baryon asymmetry in the universe between matter and anti-matter at t=0 of the Big Bang, an investigator's perspective on whether it is sensible to invest a lot of experimental and theoretical effort into trying to devise a theory that permits violations of baryon number (which is conserved in the Standard Model) is dramatically reduced relative to an investigator who sincerely believes that baryon number at t=0 had to be zero which makes the elusive hunt for baryon number violating processes a matter of deep importance that just has to be out there.

    If you think that baryogenesis and leptogenesis is a serious unsolved problem, for example, your Baysean prior about the likely viability of Majorana mass for neutrinos and Supersymmetry (both of which allow B-L conserving process that violate B and L number conservation separately) will be much more bullish and you will devote far more resources to that, than you will if you believe that finding a mechanism that violates B or L conservation is unimportant given the resources that have already been devoted to the problem when you have already done some pretty serious searches that have failed to find it.

    If you see a natural reason for the Higgs boson to have the mass that it does, even if it is unproveable with observational evidence, you are far less likely to make finding solutions to the "hierarchy problem" a priority. If you see a natural reason that is unproveable for the strong force not to have CP violation, then you're probably not going to devote a lot of evidence to investigating the possibility that axions exist.

    An immense amount of resource allocation in physics is based upon Baysean priors and intuition about what phenomena are likely to be out there to be discovered which is based upon unproveable theories and unproven conjectures and perspectives about what is natural. These meta-theories have immense real world budgetary, staffing, and personal research agenda setting impact, and so research into these theories in order to find meta-theories that are useful because they set research priorities that turn out to be fruitful are useful even if they themselves can't be proven, while unproveable theories that tend to set research priorities that turn out not to be fruitful can do immense harm even though no one ever claimed that the theories could be directly proven or were proven.

    Hypothesis generation is a critical part of business of scientific discovery and that part of the process is driven by intuition, conjectures, unproveable theories, Baysean priors, and the sociology of the discipline. But, since they have immense real world effects, we need to be more transparent about these assumptions so that they can be challenged and so that we can think more clearly about how to evaluate if we are generating the right kinds of hypothesizes and if we are falling into the perils of group think.

    We have more well trained scientists alive today, with more resources, than at any time in the history of science. But, I fear that we are squandering an unreasonably large share of these vast resources, because group think and bad meta-theories that cannot be proven or are mere conjectures, are diverting a huge share of these resources in directions that are not fruitful or duplicative, in a manner resembling the Middle Ages when centuries of the best talent in Europe was squandered on what ultimately turned out to be a scholastic theology and Platonism based research agenda that ended up being worthless because its foundation was inherently flawed, even though its axioms seemed reasonable and attractive enough to become the group think of academia and the intelligensia for centuries. Even Newton straddled that divide, devoting as much of his immense insight and time to Unitarian theology that has long since been forgotten and made no difference to future generations, and to alchemy based upon Platonic ideals rather than the Aristotelean ideals of real chemistry, as he did to physics and calculus where he almost single handedly created classical physics, classical graviation theory, and calculus.

    To go ahead and name names, I think that our current physics establishment has far too many of its eggs in the baskets of string theory, supersymmetry, and the particle based dark matter hypothesis, and far too few of its eggs in other research programs such as investigating the implications of QCD and GR in complex systems, in modified gravity theories to explain dark matter hypotheses, and it other fundamental physics theory programs. It isn't that string theory, supersymmetry, and particle based dark matter theories don't deserve investigation and development, but these research programs should have perhaps a third or less of the resources devoted to them that we see today, and the balance of our scientific resources should be devoted to other research programs.

    String theory and its cousins have been driven too strongly by Platonic premises about beauty and naturalness, that are philosophical in origin, which in hindsight, decades later, seem to have been asking some of the wrong questions. We have overestimated the usefulness of these unproven conjectures and metatheories to our detriment. (This isn't to say that the string theory program has left us entirely empty handed, but it has hardly been a bargain for the percentage of theoretical physics and SUSY chasing HEP resources that we have thrown at it.)

    For example, the agenda of rigorously establishing the laws of nuclear physics using first principles from the Standard Model is a difficult but doable project that has sat on the shelf for half a century because no one was been willing to devote the research resources to take it on for fear that they would miss out on the next great thing in string theory.

    Since we can't determine a priori which set of conjectures and unproveable theories should drive our research agenda, we need to be transparent about what those assumptions are and then intentionally diversify and isolate groups working on different agendas in order to avoid group think and to put our eggs in more than one basket. Lots of those baskets will come up empty. But, our odds of getting a winner somewhere on big breakthroughs would be greater if we had a more balkanized research community.
    Last edited: Jul 30, 2015
  11. Jul 30, 2015 #10


    User Avatar
    Gold Member

    The kind we haven't been looking for that some bright physicist notices while we are looking for something else. No other kind of observational evidence can lead to a truly transformational breakthrough.

    There is something to be said for the brute force approach of Edison trying to find the right filament for a light bulb, knowing everything else he needed to about the system for that invention. But, it is the Archimedes noticing that water flows out of his tub when he steps into it and putting two and two together from that observation that really produces progress.

    More recently, recall the famous statement made when the muon was discovered: "Who ordered that?"

    I hope to see more "who ordered that?" moments.
  12. Jul 30, 2015 #11


    User Avatar
    Science Advisor

    Lots of early-universe models do, however, have observable consequences. Claiming they cannot just betrays a lack of imagination (and a failure to read the relevant literature).
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Similar Discussions: Pre-BB Evidence
  1. HUP and BB (Replies: 17)

  2. Before the BB (Replies: 12)