Recently, I have been reading David Lindley’s The Dream Universe. A lot of thoughts went through my mind, during reading, and I tried to put them ‘on paper’. Any reaction is welcome!

I am sometime astonished how the anthropic principle is used in modern physics. I think it is used upside-down. I find it funny how it only came up in the context of fine tuning, or as explanation of why the universe is as it is.

Science aims to understand the world as it is: therefore, from the beginning, it is clear that it can only find a universe in which we find the conditions that makes our existence possible. So independent of how deep our understanding of the world around us is, we are guaranteed to find theories that show the possibility of our existence.

That being said, I think there are (at least) 4 preconceptions that lie at the heart of the hard sciences:

1. Events are causally related.

2. Objects are defined by the causal role they can play in events.

3. The causal potentialities of objects can be explained by the causal behaviour of their parts.

4. The universe is a coherent unity, so a unifying theory of it should be possible: the universe cannot behave inconsistently.

The question is, are these principles guaranteed to work endlessly? Or do we reach certain limits?

In my opinion we are reaching these limits. Examples:

1. The electron is generally considered as a particle that has no parts. What an electron is can only be described by its possible causal relationships.

2. In quantum physics, for calculations a wave function is used, that in itself is not an observable. It only can be deduced from the statistics of repeating the same experiment over and over again. Applied on a single particle, it only gives us a probability distribution. There is no exact causal explanation for where the particle arrives.

3. The universe has an observational horizon. At the moment it is the CMBR.

4. With the known laws of physics, we can understand what could have happened after less than a second, but these laws break down at a shorter time.

Given such limits, I think it makes no sense to wonder why the universe is so perfectly tuned for our existence. The fundamental constants have arisen in our descriptions of the possible causal relationships between objects; and the other way round, the objects that we suppose to exist do so, because we require descriptions in terms of causal relationships. To make exact predictions, we need mathematics, mathematics needs regularities, and causal relationships deliver regularities.

Somehow I see a parallel with Douglas Adams' puddle: why should we be astonished that the universe fits to our existence?

And instead of ‘explaining’ our universe, saying that there are are many more universes (string landscape, eternal inflation) by using the anthropological principle, I think it means we are closing in on the limits of what science can explain.

Edited October 13Oct 13 by Eise

A worthwhile discussion and introduction to it. +1

That does not mean I agree with it all.

1 hour ago, Eise said:

I am sometime astonished how the anthropic principle is used in modern physics. I think it is used upside-down. I find it funny how it only came up in the context of fine tuning, or as explanation of why the universe is as it is.

Science aims to understand the world as it is: therefore, from the beginning, it is clear that it can only find a universe in which we find the conditions that makes our existence possible. So independent of how deep our understanding of the world around us is, we are guaranteed to find theories that show the possibility of our existence.

How does the anthropic principle play with a holographic universe where both the observers (us) and the observed are simply in tyhe mind of a computer or organism ?

1 hour ago, Eise said:

Somehow I see a parallel with Douglas Adams' puddle: why should we be astonished that the universe fits to our existence?

But does it ?

Some bits do, more or less

Even the Canadians might blanch at over minus 200^oC below. Yet Nasa thinks thre is a good possibility of life on Pluto.

There is no such thing as philosophy-free science; there is only science whose philosophical baggage is taken on board without examination.

I agree

1 hour ago, Eise said:

3. The causal potentialities of objects can be explained by the causal behaviour of their parts.

Emergence ?

Banach-Tarski paradox?

1 hour ago, Eise said:

The question is, are these principles guaranteed to work endlessly? Or do we reach certain limits?

In my opinion we are reaching these limits. Examples:

Sounds a bit like complacent 19th cent Physics and the age of the Earth etc etc.

We are always discovering and testing new things, new ideas.

eg The iridium boundary ( A triumph of hardline Physics over Geology)

So yeah, let the discussion begin.

12 minutes ago, studiot said:

Sounds a bit like complacent 19th cent Physics

Yeah.
I'm reminded of his teacher telling Max Planck not to study Physics as there was nothing new to discover in that field, and it just needed to tie up a few loose ends, before Max went on to unveil a whole new domain of science, and usher in a paradigm change in how we view reality.

Interesting topic none-the-less ...

Edited October 13Oct 13 by MigL

5 minutes ago, MigL said:

Yeah.
I'm reminded of his teacher telling Max Planck not to study Physics as there was nothing new to discover in that field, and it just needed to tie up a few loose ends, before Max went on to unveil a whole new domain of science, and usher in a paradigm change in how we view reality

Max Planck.

Wasn't one of those 'fundamental constants' named after him ?

+1

7 hours ago, studiot said:

Banach-Tarski paradox?

Not sure how this would be applicable to physical objects

8 hours ago, Eise said:

Given such limits, I think it makes no sense to wonder why the universe is so perfectly tuned for our existence. The fundamental constants have arisen in our descriptions of the possible causal relationships between objects; and the other way round, the objects that we suppose to exist do so, because we require descriptions in terms of causal relationships. To make exact predictions, we need mathematics, mathematics needs regularities, and causal relationships deliver regularities.

I don’t see this as having a lot of traction; you can certainly investigate what happens if the constants have different values and see how (or that) things fail to work — changing the characteristics of fusion, for example.

What we might not know is whether you can change only some of the constants without affecting others, but I’m not sure how that gets tested.

There’s a limit to any irregularities for similar reasons. If the nature of interactions were inconsistent, how do we end up where we are? How do we get data of various vintages that’s all consistent with the interactions being the same?

9 hours ago, Eise said:

I am sometime astonished how the anthropic principle is used in modern physics. I think it is used upside-down. I find it funny how it only came up in the context of fine tuning, or as explanation of why the universe is as it is.

Interesting, the AP was mentioned this weekend in another thread, in regards to the fine tuning argument and creationists.

https://scienceforums.net/topic/139476-does-it-make-sense-to-debate-ideological-fanatics/page/3/#comment-1301643

I sort of land in the puddle argument - strong anthropic principle (SAP). Sentient observers will always be observing where the constants are such that complex molecular structures may happen. And where those constants, and the causal relations they represent, have a consistent value over time. Either endless bubbleverses, or serial "bounce" universes, or a multiverse will allow all constant values to be eventually instantiated. Some will support chemistry, some won't.

1 hour ago, TheVat said:

I sort of land in the puddle argument - strong anthropic principle (SAP). Sentient observers will always be observing where the constants are such that complex molecular structures may happen. And where those constants, and the causal relations they represent, have a consistent value over time. Either endless bubbleverses, or serial "bounce" universes, or a multiverse will allow all constant values to be eventually instantiated. Some will support chemistry, some won't.

The problem with this is it assumes it is logically possible for the constants to have different values. If one adopts this view, then one has given up trying to answer the question of why a given constant has the value it does. My own personal view is that the dimensionless constants are ultimately based on mathematical constants.

17 hours ago, studiot said:

How does the anthropic principle play with a holographic universe where both the observers (us) and the observed are simply in tyhe mind of a computer or organism ?

First, the hypothesis of the holographic universe and that we are some software implementation in a 'higher order' reality are not the same.

Secondly, whatever the universe in a metaphysical sense is, the fact is that science as described by me in my OP works. Science's aim is to understand our universe, and in this 'our' the anthropic principle is already given. Unless some empirical evidence is found that we live on 'holographic surface' or in the mind/program of a 'meta-entity', I lay these hypothesis aside. And if some evidence would be found, then it becomes our universe.

17 hours ago, studiot said:

Emergence ?

Fits under preconception 3. One could speak of two directions of explanations:

• Downwards: that is reductionism. We explain the 'macro behaviour' of objects from its parts

• Upwards: different objects that are in interaction can give rise to properties that the parts do not have.

17 hours ago, studiot said:

Sounds a bit like complacent 19th cent Physics and the age of the Earth etc etc.

17 hours ago, MigL said:

I'm reminded of his teacher telling Max Planck not to study Physics as there was nothing new to discover in that field, and it just needed to tie up a few loose ends, before Max went on to unveil a whole new domain of science, and usher in a paradigm change in how we view reality.

It may remind you of that, but it is not what I mean. I am not saying that we already know everything, which seemed to be the physical sentiment at the end of the 1800s. We see that we get at some epistemological limits. Quantum physics is 100 years old now, made an incredible progress during these years, but an interpretation of QM on which everybody agrees is not in sight. My position is that we might near such empirical limits also in other areas, like the the Big Bang (therefore I wrote 'At the moment it is the CMBR').

10 hours ago, swansont said:

I don’t see this as having a lot of traction; you can certainly investigate what happens if the constants have different values and see how (or that) things fail to work — changing the characteristics of fusion, for example.

What we might not know is whether you can change only some of the constants without affecting others, but I’m not sure how that gets tested.

I am not sure if I understand the points you are making.

Of course one can theoretically play around with other values of universal constants, but that is what it is: playing. We want to understand the universe as it is, so here, the constants must be as we found, otherwise we would not exist. Why would one wonder about that? A wonder would be that constants have other values, that would make life impossible, but still, here we are!

And we even have no idea if the constants could have other values. How would we know that they can have other values in other bubble-universes in the inflationary multiverse, especially because they are causally disconnected? Similar for the superstring landscape.

10 hours ago, swansont said:

There’s a limit to any irregularities for similar reasons. If the nature of interactions were inconsistent, how do we end up where we are? How do we get data of various vintages that’s all consistent with the interactions being the same?

Sure. I think that is exactly what I am saying: we can only speak of causality if there are regularities. The success of physics speaks for itself.

10 hours ago, TheVat said:

Either endless bubbleverses, or serial "bounce" universes, or a multiverse will allow all constant values to be eventually instantiated.

Well, maybe not. How would you know our universal constants are not also 'multiuniversal' constants, i.e. have the same values in every bubble universe, so the physics in every bubble is the same, just with different histories. But where is the empirical evidence that other universes exist anyway? Because we wonder about something that we should not wonder about at all? As said above, why wonder that we find universal constants that make life possible? Something else would be inconsistent from the beginning.

8 hours ago, KJW said:

The problem with this is it assumes it is logically possible for the constants to have different values. If one adopts this view, then one has given up trying to answer the question of why a given constant has the value it does.

Agree.

8 hours ago, KJW said:

My own personal view is that the dimensionless constants are ultimately based on mathematical constants.

I would call that the wet dream of a physicist. I cannot exclude such idea, but I think we are far, far away from that, to say the least. It would mean we can derive the existence of our universe from mathematics alone. Do you really think that is possible?

Edited October 14Oct 14 by Eise

1 hour ago, Eise said:

First, the hypothesis of the holographic universe and that we are some software implementation in a 'higher order' reality are not the same.

This is really ducking the issue.

I used the term holographic (as invented by someone or other with nothing better to do than dream up fancy terminology) to cover all possibilities where the observer and observed are part of the same system.

1 hour ago, Eise said:

I would call that the wet dream of a physicist. I cannot exclude such idea, but I think we are far, far away from that, to say the least. It would mean we can derive the existence of our universe from mathematics alone. Do you really think that is possible?

great stuff +1

1 hour ago, Eise said:

Fits under preconception 3. One could speak of two directions of explanations:

• Downwards: that is reductionism. We explain the 'macro behaviour' of objects from its parts

• Upwards: different objects that are in interaction can give rise to properties that the parts do not have.

21 hours ago, Eise said:

3. The causal potentialities of objects can be explained by the causal behaviour of their parts.

Why only objects ?

Surely (material) objects have properties.

And much causality and emergence comes from these properties.

And most properties are non material.

13 hours ago, swansont said:

20 hours ago, studiot said:

Banach-Tarski paradox?

Not sure how this would be applicable to physical objects

I was thinking of Kuracharski's explanation of BT and how maths is adapting to accomodate this.

Proof - Professor Adam Kucharski 2025 p 48-49.

"How to make two copies of an infinite line of apples."

The point about this is that we are not sure any longer about underlying theory such as dimension.

If you can make another line of apples, you can make an infinite number of lines.

This really impinges on Eise's points 2 and 3. and emergence.

We are starting to realise that the whole may be greater than the sum of its parts as well as less than or equal to ie not as in the triangle inequaltiy (which underlies QM) .

OK. We are not reaching the limits of all there is to know.
We may be reaching the limits of all that is possible/allowed for us to know ?

That is obviously true in some ways, but not all.
Even something as simple as the HUP prevents us from knowing certain aspects of the universe.
And sure, there are no 'work-arounds', but often, these 'prohibitions' that the universe erects in our journey to understanding, tell us a lot more about how the universe works than they prevent.

Edited October 14Oct 14 by MigL

2 hours ago, Eise said:

How would you know our universal constants are not also 'multiuniversal' constants, i.e. have the same values in every bubble universe, so the physics in every bubble is the same, just with different histories

Oh, I'm not saying any of this is knowable - unless we can construct toy universes and somehow observe them, which is a wild conjecture that seems to present conceptual problems. I was just saying a SAP which rests on universes with different physics is conceivable. As @KJW points out, there could be logical grounds that physical constants arrive at the values we find, and so all universes will tend towards baryonic matter with fusion powered stars and so on. It could be that, without needing to bring in a universe that "wants" life, it's just a happy (for us) random outcome that fusing stars and complex molecules will sometimes give rise to life.

3 hours ago, Eise said:

Because we wonder about something that we should not wonder about at all? As said above, why wonder that we find universal constants that make life possible?

I suspect that the wondering is just that any life exists - which I can agree has no point beyond just philosophical awe. That we do in fact arise from a universe which allows complex stable matter and stable energy flows and chemotaxis, etc. couid come to be a very obvious thing if we further unlock the workings of abiogenesis.

Edited October 14Oct 14 by TheVat
Kfvnmjngf

1 hour ago, studiot said:

  3 hours ago, Eise said:

First, the hypothesis of the holographic universe and that we are some software implementation in a 'higher order' reality are not the same.

This is really ducking the issue.

That was 'First'. Read again and read 'Secondly'...

1 hour ago, MigL said:

And sure, there are no 'work-arounds', but often, these 'prohibitions' that the universe erects in our journey to understanding, tell us a lot more about how the universe works than they prevent.

Can you give examples of this?

21 minutes ago, TheVat said:

I was just saying a SAP which rests on universes with different physics is conceivable.

'Conceivable', yes. But find empirical evidence for it? With causally disconnected universes?

24 minutes ago, TheVat said:

I suspect that the wondering is just that any life exists - which I can agree has no point beyond just philosophical awe.

Yep, that is something different, especially if you look what conditions must all come together that life can develop, and strive for such a long time, that it can give rise to organisms that are conscious and do science.

1 hour ago, studiot said:

Why only objects ?

Surely (material) objects have properties.

But properties are exactly descriptions of how objects can partake in causal relationships! We derive the properties from an object by the ways it can causally interact with other objects. (Oh, and I did not say 'material objects').

Edited October 14Oct 14 by Eise

30 minutes ago, Eise said:

Can you give examples of this?

I thoought I had ... the HUP.

31 minutes ago, Eise said:

But properties are exactly descriptions of how objects can partake in causal relationships!

ALL descriptions we have of ANY object are due to the causal interactions, which is what we can measure.
I recently had this discussion with @KJW as to whether space-time is actually curved, when all we can measure is how it causally interacts with test objects; the underlying 'reality' stays hidden.

25 minutes ago, Eise said:

Conceivable', yes. But find empirical evidence for it? With causally disconnected universes?

Well, I had mentioned in that post that I thought hypotheticals like toy universes* to study were "a wild conjecture that seems to present conceptual problems." And by that I meant that it seemed problematic that any observation could be done of a realm with different physics. How do beings with one physics observe a different one, given that observation itself depends on interactions that are causally bound by universal constants? It's incoherent to picture light leaving a toy universe source at one velocity and then changing velocity at some interuniverse boundary. Or a particle emitted somehow altering its fundamental properties in transit before my observation collapses it. (Amusing fantasy, for sure!)

*(Henceforth I want to use the German for this, spielzeuguniversen, it just sounds better)

1 hour ago, Eise said:

3 hours ago, studiot said:

Why only objects ?

Surely (material) objects have properties.

But properties are exactly descriptions of how objects can partake in causal relationships! We derive the properties from an object by the ways it can causally interact with other objects. (Oh, and I did not say 'material objects').

What is an object ?

No it is true I thought you were referring to material objects, but if you are not:

Take a piece of graph paper.

Draw a 1x 1 square on it.

Absolutely nothing else.

That (in my opinion) is a non material object as it is a gedanken experiment so no material graph paper will actually be defaced during the course of the experiment.

Now what causal or other relationships does that square possess ?

9 hours ago, Eise said:

I am not sure if I understand the points you are making.

Of course one can theoretically play around with other values of universal constants, but that is what it is: playing. We want to understand the universe as it is, so here, the constants must be as we found, otherwise we would not exist. Why would one wonder about that? A wonder would be that constants have other values, that would make life impossible, but still, here we are!

And we even have no idea if the constants could have other values. How would we know that they can have other values in other bubble-universes in the inflationary multiverse, especially because they are causally disconnected? Similar for the superstring landscape.

It sounded like you were saying there’s no point in thinking about or investigating why the constants have the value they do. I don’t think certain physicists (or philosophers, for that matter) would be keen on being told not to do that. Or that they shouldn’t check to see if they are indeed constant.

19 hours ago, KJW said:

The problem with this is it assumes it is logically possible for the constants to have different values. If one adopts this view, then one has given up trying to answer the question of why a given constant has the value it does. My own personal view is that the dimensionless constants are ultimately based on mathematical constants.

Me too.

In particular, I've trying to get my head around the Planck Consortium's finding that the Equation of State for the observable universe is:

w = 1.028 +/- 0.032 or w ~ -1

This seems to imply that for at least the more recent part of the last 13 billion years, the universe has been following quite a narrow trajectory between Big Rip and Big Crunch despite significant density and compositional changes.

Moreover, recent work by Hooft, Susskind and others have indicated that the Bechenstein limit sets an upper bound on the entropy of any region of the OU in addition to the ever increasing lower bound set by the 2nd Law. Projecting these conditions backwards and forwards in time impose quite severe constraints on the rate of entropy production and in particular the number and different types of particles in play at any given stage in the evolution of the universe.

I get a very strong sense that we're missing a major causal factor that is driving the universe asymptotically down a 'preordained' route. If anything in fundamental physics was initially 'plastic', it's as if the fledgling universe took a look into the far, far future boundaries of spacetime and worked out what it had to do to its inner workings to reach that destination.

Far-fetched maybe. But I'm still holding a candle for absorber theory, or TI, or the advanced wave of the Schrodinger equation to supply the feedback from the Restaurant at the End of the Universe and collapse all those multiverse options down to one.

2 hours ago, sethoflagos said:

it's as if the fledgling universe took a look into the far, far future boundaries of spacetime and worked out what it had to do to its inner workings to reach that destination.

All evolutions of the universe are possible, but the one actually taken is governed by the Principle of Least Action ( a global version of it ).
Just like refraction, the double slit experiment, the path integrals of QED, and every other physical process.
No checking into the 'future' required.

2 hours ago, MigL said:

All evolutions of the universe are possible, but the one actually taken is governed by the Principle of Least Action ( a global version of it ).
Just like refraction, the double slit experiment, the path integrals of QED, and every other physical process.
No checking into the 'future' required.

How does one determine the path of least action if one does not know the destination?

On 10/13/2025 at 5:06 PM, Eise said:

Recently, I have been reading David Lindley’s The Dream Universe. A lot of thoughts went through my mind, during reading, and I tried to put them ‘on paper’. Any reaction is welcome!

I am sometime astonished how the anthropic principle is used in modern physics. I think it is used upside-down. I find it funny how it only came up in the context of fine tuning, or as explanation of why the universe is as it is.

Science aims to understand the world as it is: therefore, from the beginning, it is clear that it can only find a universe in which we find the conditions that makes our existence possible. So independent of how deep our understanding of the world around us is, we are guaranteed to find theories that show the possibility of our existence.

That being said, I think there are (at least) 4 preconceptions that lie at the heart of the hard sciences:

1. Events are causally related.

2. Objects are defined by the causal role they can play in events.

3. The causal potentialities of objects can be explained by the causal behaviour of their parts.

4. The universe is a coherent unity, so a unifying theory of it should be possible: the universe cannot behave inconsistently.

The question is, are these principles guaranteed to work endlessly? Or do we reach certain limits?

In my opinion we are reaching these limits. Examples:

1. The electron is generally considered as a particle that has no parts. What an electron is can only be described by its possible causal relationships.

2. In quantum physics, for calculations a wave function is used, that in itself is not an observable. It only can be deduced from the statistics of repeating the same experiment over and over again. Applied on a single particle, it only gives us a probability distribution. There is no exact causal explanation for where the particle arrives.

3. The universe has an observational horizon. At the moment it is the CMBR.

4. With the known laws of physics, we can understand what could have happened after less than a second, but these laws break down at a shorter time.

Given such limits, I think it makes no sense to wonder why the universe is so perfectly tuned for our existence. The fundamental constants have arisen in our descriptions of the possible causal relationships between objects; and the other way round, the objects that we suppose to exist do so, because we require descriptions in terms of causal relationships. To make exact predictions, we need mathematics, mathematics needs regularities, and causal relationships deliver regularities.

Somehow I see a parallel with Douglas Adams' puddle: why should we be astonished that the universe fits to our existence?

And instead of ‘explaining’ our universe, saying that there are are many more universes (string landscape, eternal inflation) by using the anthropological principle, I think it means we are closing in on the limits of what science can explain.

Wasn't it Douglas that also said something like "if anyone were to discover the truth of the universe, it would instantly change" and "that's already happened".

9 hours ago, MigL said:

No checking into the 'future' required.

No checking into the future allowed.

7 hours ago, sethoflagos said:

How does one determine the path of least action if one does not know the destination?

A system's physical path in time minimizes S, the action, which is the integral of the system's Lagrangian.
See here https://en.wikipedia.org/wiki/Action_principles

The hard part is identifying the Lagrangian for the whole universe.
( if we could we would have a Theory of Everything )

On 10/14/2025 at 3:48 PM, MigL said:

And sure, there are no 'work-arounds', but often, these 'prohibitions' that the universe erects in our journey to understanding, tell us a lot more about how the universe works than they prevent.

OK, with the HUP as example: it is a limit of what we can know. But here the different interpretations already part:

• is it a limit because of our epistemology?

• is it an attribute of nature, i.e. what we are looking for (e.g exact momentum and position at the same time) does not exist below a certain limit?

On 10/14/2025 at 5:51 PM, MigL said:

I recently had this discussion with @KJW as to whether space-time is actually curved, when all we can measure is how it causally interacts with test objects; the underlying 'reality' stays hidden.

I remember, I was there too.

On 10/14/2025 at 7:03 PM, studiot said:

That (in my opinion) is a non material object as it is a gedanken experiment so no material graph paper will actually be defaced during the course of the experiment.

No, that is just an intensional object. I hope you won't try to open that can of worms. Maybe you can make your point in a direct argument?

On 10/14/2025 at 10:37 PM, swansont said:

It sounded like you were saying there’s no point in thinking about or investigating why the constants have the value they do. I don’t think certain physicists (or philosophers, for that matter) would be keen on being told not to do that. Or that they shouldn’t check to see if they are indeed constant.

Surely that was not my intent. One can never know if some new theoretical relationships are found, something like the interdependence of e₀, u₀ and c. But I simply think that wondering why the natural constants have the values they have is funny. Whatever constants and their values one finds, one should not wonder that they are consistent with our existence.

On 10/15/2025 at 2:09 PM, MigL said:

A system's physical path in time minimizes S, the action, which is the integral of the system's Lagrangian.

In my understanding of the principle of least action, you still need the boundary conditions, i.e. initial and end values of your system.

21 hours ago, Eise said:

OK, with the HUP as example: it is a limit of what we can know. But here the different interpretations already part:

• is it a limit because of our epistemology?

• is it an attribute of nature, i.e. what we are looking for (e.g exact momentum and position at the same time) does not exist below a certain limit?

I think it's a limit of what's observable/understandable, for instance, we can understand how a dog can smell, but not what it looks like for the dog.

On 10/23/2025 at 11:23 AM, Eise said:

1. is it a limit because of our epistemology?

2. is it an attribute of nature,

I'm not sure.
It could be both.

On 10/23/2025 at 11:23 AM, Eise said:

In my understanding of the principle of least action, you still need the boundary conditions

Boundary conditions yes, but the path does determine where the system ends up.
Consider a simple Lagrangian involving kinetic and potential energy.
If we simplify it to be absent of any kinetic term, you have a test mass on a potential gradient; i.e. a falling mass.
The path of the test mass down the gradient clearly indicates where it will end up; at the bottom of the potential well.
And all of Newton's laws can be extracted from the Principle and this Lagrangian.

A few months back, it may have been @joigus who posted a sample ( not simple ) Lagrangian for all the various interactions that would need to be 'considered' for a Theory of Everything.

Edited October 25Oct 25 by MigL

11 hours ago, MigL said:

I'm not sure.
It could be both.

Boundary conditions yes, but the path does determine where the system ends up.
Consider a simple Lagrangian involving kinetic and potential energy.
If we simplify it to be absent of any kinetic term, you have a test mass on a potential gradient; i.e. a falling mass.
The path of the test mass down the gradient clearly indicates where it will end up; at the bottom of the potential well.
And all of Newton's laws can be extracted from the Principle and this Lagrangian.

A few months back, it may have been @joigus who posted a sample ( not simple ) Lagrangian for all the various interactions that would need to be 'considered' for a Theory of Everything.

Well, yes. I copied it and pasted it, but it lacked gravitation.

I'm still getting up to speed with most of the arguments here, BTW.