The Case for Strong Emergence by Sabine Hossenfelder

Sabine,

Yes. Knew what you meant. But you missed my question(s) (accidentally?); I think Physics IS important (I include my observational cosmology work, etc.) But are you really happy eschewing real 'advancement' to just shuffle past theories?

I escaped academia, earned more but do MORE research. Yet editors & arXiv increasingly don't like even logging my papers! Is that right when our big hopes to escape this theoretical wilderness (that's YOU Bee!) seem to have dismissed the need and given up! Not long ago ALL our eminent physicists insisted new approaches & breakthroughs were needed. Chatting with Milton Freeman recently almost NONE now came to mind. Have you now given up to?

I know there's a million weird ideas out there, but can you give advice for this scenario;; Say someone just beyond access to journals made a falsifiable theoretical discovery apparently unifying QM (a classical derivation) and Relativity, also resolving a tranche of other including astrophysical anomalous findings. When editors slam the door in fear without even looking.. WHO IS IT in the academic community that would deign to even look, and if correct, help develop it to publication. Is there anybody left who might? That's an important question!

So is there still any sense of responsibility in academia to help advance mankinds understanding rather than just careers when only academics can publish papers in accredited journals?

Do you not agree physics really could help with cures for cancer etc if we really try?

best

Peter

"Time Really Passes, Science Can't Deny That" is a recent essay by Nicolas Gisin (a household name perhaps) published in 2017 and posted on the arxiv [1602.01497]. The title is somehow misleading as he mostly deals with free will. A good deal of ingenuity appears to have been invested in avoiding to mention "Time and Free Will". This is actually the English title of once a famous book by Henri Bergson, the philosopher who dared to contradict Einstein. Of course he is largely forgotten and now people are rediscovering his ideas by their own means.

Dear Sabine,

1. Why is the argument against free will you sketch

in the abstract invalidated by strong emergence?

You just do not tell in the essay. This does not

make sense. Perhaps it would be best if you just

remove the remarks about free will (they appear anyway

only in the abstract and two very short sentence in the text),

and tell the reader what the essay is really about in the

abstract ;-).

2. Who else thought that "effective field theory is

a fool proof argument against strong emergence"?

You do not quote anybody... Is the essay in the

end characterized as:

"(only) I long believed in a mistaken argument

and here I tell you why I was wrong"?

sincerely

Maurice

Dear Sabine,

I consider your essay a piece of jewelry, both by elegance, and by having sharp arguments which cut like the diamond to the essence. You give crystal clear definitions of weak and strong emergence, and simple but clear and rigorous explanations and examples. You made the best arguments against strong emergence that I saw, yet this allows you to avoid the usual misconceptions and find a loophole. In addition, among the example you gave en passing, I should say some of them are really important and yet sometimes ignored. For example, in just a couple of words you said it well about the AdS/CFT correspondence, clearing some confusion in the literature, where too often people take it literally that the duality is a mathematical isomorphism, while being a sort of physical equivalence between some particular cases. Another part I liked was the double hierarchization of theories by weak and strong emergence, and the interplay between the two kinds of emergence that you exemplified. Also the argument from the Landau pole. So while I would like to help with some criticism, I don't have any, at least for the moment. I just have a question. You wrote "nature does not allow mathematical inconsistencies", and I strongly agree. Assuming that there is a mathematical theory which describes the physical universe exactly at all levels (which maybe we will find someday, maybe not, and doesn't have to allow exact derivation of everything, effective is enough) do you believe that this would forbid strong emergence, and in particular the possibility of free will? (for example I believe such a theory is compatible with strong emergence)

Congratulations for the essay and success in the contest!

Best wishes,

Cristi

Dear Sabine,

a very interesting essay. The connections between theories reminded me of Sebastian de Haro's contribution to this contest---I think his framework could have a valuable application to your argument.

But I do have a couple of questions, if I may. First, can we not simply 'add' theories? Think of classical Newtonian gravity, and electromagnetism. Both have the same domain of applicability, but either could exist without the other, and they're not equivalent. So you could have a mass moving in a gravitational field, a charge moving in an electromagnetic field, and a massive charge moving in a combination of both fields. In the latter case, the predictions made using Newtonian gravitation would not suffice to calculate the object's motion, and similar with electromagnetism.

So it seems to me that to get an instance of 'strong' emergence, one would simply have to add a theory that postulates a 'charge' (in the general sense) which is linked to resolution, or size, or something like that; so the theory claimed fundamental no longer suffices to describe objects at a certain level, but no logical contradictions obtain---the fundamental theory simply isn't the complete description, just as Newtonian gravity isn't the complete description of the previous case.

An argument often made against such a case of strong emergence is related to Leibniz' 'principle of sufficient reason': there's no reason for additional charges such as the one I describe to suddenly emerge. But there's actually no difference to reductionism here: the fundamental facts that fix everything in a reductionist universe similarly lack further justification.

Of course, one could then turn the question around and ask what 'strong' emergence actually adds. And here, I'm afraid, there is no progress on the question of free will: whether my actions are determined by microscopic facts, or whether some macroscopic facts play into them, as well, doesn't change the deterministic nature of my actions. I would not be any more free if my behavior is determined by a law that cannot be smoothly extrapolated from a microscopic law---the question simply doesn't have any relevance to my freedom.

Indeed, the whole idea that it's physics that poses the greatest threat to free will is, in my opinion, already misguided. The problem is with the concept itself. As Schopenhauer has pointed out, "man can do what he wills, but he cannot will what he wills". What he alludes to here is that there'd be an infinite regress in arguing that the will could, in some way, be self-determined: in order to be free in my action, the cause of my action---my will, my desire to act---must be free; if not, if I am forced to desire something, and hence, act accordingly, then my action obviously wasn't free. But then, likewise, the cause of my will must be free---or else, see above. So the problem with the idea of free will really is that all my behaviors are governed by my desires; but I can't freely choose my desires (without collapsing into infinite regress).

That said, I do think there is a sensible notion of free will that can be salvaged; but that would take us entirely too far afield.

Thank you for a clear and incisive essay, Sabine

As a response to another post, you state that "In case you mean to say that we shouldn't try to actually derive all of science from physics, I totally agree. It doesn't seem very practical." This suggests that the weak emergence between theories of different resolutions you discuss does not provide practical knowledge, especially in terms of providing measurements and predictions at the lower resolution.

Considering your levels of resolution (a great image), what if we find that we really need the theories at each level of resolution - that they cannot be practically reduced to the 'most fundamental' level? If we are unable to practically reduce all lower resolution theories to the highest resolution theory, where is the fundamental character of this highest resolution theory? More importantly, why are the other theories not also 'practically fundamental'?

I think this is were the reductionist philosophy fails - on the practical applied level. And if it cannot break this practical level, what good is the theoretical reasoning about reducing all lower resolution levels to any higher one (is it then philosophy and not science)?

This situation, of the practical uselessness of attempting to define all levels of reality via the 'fundamental' reductionist level, could be a limitation of our current (mathematical) tools. This could be a limitation via mathematical theory (which you seem to introduce), but it could also be a limitation in the mathematical tools used to measure reality (eg. our inability to provide a single numeric value for a complex number as we are able for 'real' numbers). If the latter, then changing the limitations of the tools (one never eliminates such a limitation) would them impact any theories built upon those tools.

Thank you for your essay,

Don

Sabine,

I think GR is fundamental for now but I hope it will become emergent when (if) a TOE is developed. QM is also fundamental for now and it is at a higher resolution than GR (at least away from BH's).

So, does your definition require that a strongly emergent theory must make the same predictions as the higher resolution theory?

Best Regards,

Gary Simpson

You are being unfair to Sabine and physicists in general, Peter. The system is far from being perfect, and our knowledge is limited, but it is completely not true that physicists are accountants "shuffling last years books into order", "eschewing real 'advancement' to just shuffle past theories". We know very well what we know and what we don't know. I can understand your frustration, and why it would resonate to other "outsiders", but it is just not true that the doors are locked for them.

I know because it happened to me. My first PhD advisor was very happy to have me, but he got angry when I showed him my papers, and he didn't even read them because I didn't work at his problems. I understand him, he wanted me to follow the "safe" path, but I didn't care, so I left and stayed without advisor for 2 years. Without any support, without mentioning the affiliation in some cases, I continued. I put my articles online, and got favorable emails from a few experts in the field, and invitations to some great conferences. I got my first job invitation at a great institute before my first paper was accepted, even if they thought that I was unaffiliated. With publications and the thesis written, I could find a new advisor and get my PhD. Before getting my PhD I already had 10 publications in peer reviewed ISI journals, single author, and without any support. This simply shows that the system is open to the outsiders, you can publish even without PhD. And none of my articles is in mainstream approaches, if you think that this is why they accepted them. I work at my own projects, even if this means more difficulties and less financial support. I am aware that only a few experts see me as more than a minor league player, but I play by my own rules. And I am not a unique case, many articles are published by people who left academia. The system may be imperfect, but if you have viable ideas, you work very hard to formulate them well, both mathematically and physically, and obtain relevant results, then the doors are open. I hope you'll take this as good news :)

Dear Sabine

You assumed that consciousness is a low resolution phenomenon, thus it causes only down causation. But there are the model connected with pansichism, One of them is my model: quantum consciousness .

Do you accept any bet about existence of quantum consciousness? But, argumentation would be still better.

Yet another argument that seems different at first sight but is wrong for the same reason as the example with the chain is that entanglement realizes top-down causation [12]. The argument here is that entanglement is a non-local property of a system. Hence, if you have information only about a small part of a system, you have no way of knowing whether the system will begin to show novel effects due to entanglement if you look at the full system. Again, though, it is clearly possible to derive the behavior of the whole system if you have information about its entire microscopic constituents which, of course, includes entanglement between them.

I do not understand this. "if you have information only about a small part of a system," Why it is necessary to have only information about a small part of a system?

But it is interesting that Stoica agrees with you. What Zeilinger or Brukner think?

But independently what you think about free-will, I like what you think about top-down and down-top causation. It is so simple, I hope that it is true.

Please visit the My essay .

Hello Sabine,

Thanks for the opportunity to read your mind.

Let us start with a few words about what we don't know.

Insofar as 'free will' is simply the ability to make decisions in the context of multiple choices, we have to admit that free will exists, but it comes at the high price of uncertainty.

Concerning 'strict laws' we are moving towards a collective understanding that what we have called strict laws in science in the past are better represented as 'principles'. Strict laws demand compliance whereas principles are more liberal in the sense that they accommodate deviations within limits. Nature accommodates deviations either side of its statistical norms which deviations conjure up extraordinary variety, which variety enables a few of many variables to persist in an environment undergoing constant change.

Your claim that 'Large things are made of smaller things, and if you know what the small things do, you can tell what the large things do' - requires re-thinking. There is an adage that the whole is greater than the sum of its parts. If you put the parts of an aircraft together in their intended relations, you still have no idea what an aircraft can do! Reductionism only works well when it works well - i.e. to one's advantage.

Picking up on your statement that 'no one understands how gravity works' rings a bell.

In 1916 Einstein declared gravity a misconception, an effect rather than a cause. While he elaborated to state that the cause was brought about by the uneven distribution of mass in the universe, his 'cause' is open to question as to whether the dominant constituent of the cosmos, vacuum, is the direct cause of what we call 'gravity'.

In matter we find an exhibition of defiance against vacuum, the exception that proves the rule. What is the rule? The rule is that vacuum (unfilled void or space) abhors nature, and flows to fill its absence.

Regarding the question of 'curved space', if space is unlimited in its extent, how can it be 'curved'? Where space describes a discrete volume, then yes, the space within a banana skin is curved.

On to your subject of the moment: 'Emergence'. I hope that I am alone in your readership in saying that I don't understand your concept of emergence. Perhaps my limited skills in mathematics accounts for my difficulties, but if this is so, reliance upon mathematical linguistics may be stretching the evaluation criterion stipulating that essays should be 'non-specialist'. Suffice to say, I don't know what I don't know! пЃЉ

I think that we all, upon occasion, make mistaken assumptions that each question only has a single correct answer. This error often arises due to the temptation to reduce all issues to mathematics for resolution.

Circumstances are constantly changing, which changes enable other acceptable answers to surface. Truth is thus revealed as being conditional rather than absolute.

It is important for all specialists to communicate with the public in non-specialist languages because we need the support of the public to further advance each specialty.

Thanks again Sabine for the present of your presence of mind.

Gary.

Dear Sabine,

Fantastic essay - well-written, clear, entertaining, and I think genuinely profound! You do indeed put a strong case forward for strong emergence.

Before reading your essay, the only reasonable example I knew of a theory that might require strong emergence was that of consciousness: effective field theories cannot even in principle tell me how the collection of particles in my brain makes me, for example, experience the colour red in the way that I do. This is of course no proof that strong emergence is playing a role, but it is certainly a case where weak emergence is far from providing the answer. I wonder if you have an opinion on this?

I will honour the tradition of shamelessly advertising one's own essay in

a comment. Mine is almost completely unrelated to yours, but you might enjoy it nonetheless! https://fqxi.org/community/forum/topic/3091

Best of luck,

Paul

Cristi,

I know you're right, again. Nothing's impossible. A string of rejections from editors at first glance is wearing but I can improve quality and am tenacious. I recall one great advancement followed ~40 rejections.

My 'shuffling' comment was about Bee's essay specifically. I admire Bee's blog entries and approach but was very disappointed she seemed to have abandoned that incisive 'abandon' in this essay to descend into old embedded mire.

Yes, It's "far from being perfect", but it could improve in the right direction with a little effort. I think those with established authority also have a heavy responsibility which most don't recognise. Maybe Bee feels she doesn't have such authority, but she quickly would with more effort to properly analyse and help promote replacement theories. (I recognise 'quickly' in physics is subject to relativistic adjustments dilating and quadrupling the normal passage of time)!!

I recognise the issue that there are hundreds of flawed theories out there so the easiest option is to study none!

Do argue or defend your position on advancement please Bee, I apologise if I just seemed to have suggested it's in a mud bath!

Very best

Peter

I don't take a reductionist view but I still have trouble accepting the argument for strong emergence. If I have some theory I can run a simulation, and given enough computer power I can observe the implications of the theory in the simulation. If I do the experiment and see something else emerge then either the theory is wrong or there is some other influence I did not account for in the simulation. If the theory has a non-deterimninstic element different things will happen but the statistical behaivior in many trials will be the same in a simulation as it is in the real world. I don't see any room for strong emergence.

As for free will, the answer depends on the exact definition of the terms in the question. If those are clearly given then people will only answer differently depending on whether or not they hold a pluralist belief for mind and matter or not.

周慮歳

Dear Sabine,

Because you are by far my favorite author on the curious state of current physics, and because I would love to have your comments on my essay, I'm a little at a loss how to respond to this beautifully-written piece. Though I don't disagree with you at any point, what was in my mind throughout was your blog-post from a couple years ago on "The Unbearable Lightness of Philosophy." Actually (as a philosopher) I'm very pleased that you take philosophical arguments seriously. But really - free will?

Surely the conscious decisions we make are the result of neurons firing in our brains, or more generally, the result of things going on beneath the level of our self-awareness. That seems like a very good thing to know about ourselves, if it makes us a little more humble. And if there were any meaningful issue about our choices being "free" - which I doubt - you'd think this is where it would arise.

But I can't see why it makes any difference to anything whether the behavior of neurons is predictable from physical principles. It's hard to see how it could be, if it's not even possible to predict when an atom will emit a photon. But what does it matter whether or not what happens in my brain is computable?

I have no problem with reductionism - except for the "ism" part. That seems to imply that this is the only kind explanation science needs, which is obviously wrong. Take biology, for example. I see no reason to doubt that everything that happens in a living cell is done by the laws of physics. On the other hand, none of these very complex molecules and patterns of interaction would exist if it weren't for the ability of cells to replicate themselves and so evolve. Nothing like this happens in physics or chemistry, so it's clearly "emergent." But that's not because it's independent of the laws of atomic physics. On the contrary, if those laws weren't so precisely uniform and reliable, self-replication would be inconceivable.

So my question is, how does distinguishing between "weak" and "strong" emergence contribute to our understanding of situations like this? I wouldn't call self-replication a "truly new fundamental law," but it is a truly fundamental functionality, on which a whole new world of possibilities is founded. It's completely new, and also completely dependent on lower-level functions.

In my current essay I try to imagine the foundations of physics in a similar light, asking about the functionality of a universe that's able to define, measure and communicate all its own information. It would mean a lot to me to know if the argument of my first three pages makes sense to you.

Conrad

Dear Sabine,

I looked for your paper because I read and enjoy your backreaction blog, so I felt your astute science writing would be insightful.

Although you acknowledge that modern theoretical physics "is almost certainly incomplete" you avoid venturing into "what" it is that is more fundamental than the well-known 25 fundamental SM particles.

The discussion of how "emergence" is defined in this context was enlightening, and the examples in condensed matter 'verrry interesting (but not fundamental)'.

Here I think that the examples of strong emergence should consider cosmology, i.e. General Relativity. This is a relevant issue for insight into the research topic. I am certain that causality is a fundamental property of particles, as Seiberg has found. So I began my essay by considering the well-founded causal formulation of particles as given by the No-Boundary Wave Function.

I also discussed the fundamental requirements to establish consistency between GR and (causal) particle theory. But in your essay you suggest (without evidence) that non-renormalizable theories are "sick". But the only reason that renormalization is used is that L'Hopital's Rule doesn't work- the mathematical singularity assumed forces an infinity/infinity situation. Of course, the singularity also compels one to arbitrarily assign quanta and scalar metrics (mass and energy).

In short, the current theory violates mathematical 'laws' which there is thus strong motivation to correct... a very good starting 'point' is to not assume a particles representation geometry is a point.

It turns out that all of these criteria can be met at once, but yes, you have to let go of renormalization. The traditional approaches that keep it and seek unification via new particles just don't work out.

That said I invite you to read and comment on my essay: https://fqxi.org/community/forum/topic/3092 and hope that its insights spark an interest.

Best,

Wayne Lundberg

p.s. as a footnote, I am sure that a person's perception of free will is best discussed in the context of particle theory by considering the scale of space-time averaging. Clearly weak-scale particle theory has little bearing since the quantum algebraic states average out at micro-condensed matter scales, long before a human scale. Consider again, if you will, to be fair, just how much free will really means when you use a space-time average of say, 2 Earth orbital diameters and 10000 years. That yields a rather different result, no?

The analogy to particle theory works pretty well when you compare a human's decision tree at, say, an intersection. Compare that to a particle interaction's "channels".