nothing wholly new emerges out of nothing like magic when things are just arranged in the right way. — Pfhorrest

I would dispute that with three counter examples.

1. A philosophical book is just an arrangement of words. If there is nothing new in philosophical books, why write them? Why read them?

2. A steam engine is just an arrangement of steel, water and fire. And yet when it was invented, it was pretty revolutionary. And if there's nothing new in a steam engine, how come the pharaohs of antiquity didn't think of building a Memphis-Thebes railroad?

3. A living organism is just (supposedly) an arrangement of atoms of hydrogen, carbon, oxygen etc. And yet a living organism can reproduce, which an atom cannot do. To be precise, to reproduce an atom would mean very little, because what is reproduced in life is the (complex, biological) structure of the organism, the shapes the molecules make with atoms, not the atoms themselves. So the concept of reproduction has a clear meaning in biology but not in physics.

1. A philosophical book is just an arrangement of words. If there is nothing new in philosophical books, why write them? Why read them? — Olivier5

For the same reason that we collect a variety of goods and organize them on shelves in the same building called a "store": it's much nicer to go to one place and find all the things you need, than to have to wander far and wide trying to find each one who-knows-where out there. The same is true of discoveries (in any field, not just philosophy) as it is of goods: it's an improvement to have them all organized in one place for anyone to browse, than for everyone to have to go find them all over again.

2. A steam engine is just an arrangement of steel, water and fire. And yet when it was invented, it was pretty revolutionary. And if there's nothing new in a steam engine, how come the pharaohs of antiquity didn't think of building a Memphis-Thebes railroad? — Olivier5

Because they hadn't discovered that possibility yet. The possibility was always there, like all the discoveries we collect in books as above, they just didn't know about it yet.

3. A living organism is just (supposedly) an arrangement of atoms of hydrogen, carbon, oxygen etc. And yet a living organism can reproduce, which an atom cannot do. To be precise, to reproduce an atom would mean very little, because what is reproduced in life is the (complex, biological) structure of the organism, the shapes the molecules make with atoms, not the atoms themselves. So the concept of reproduction has a clear meaning in biology but not in physics. — Olivier5

This gets most to the point: reproducing a living thing is an aggregate of a whole (frickin') lot of ordinary mechanical processes that atoms already do. A multicellular organism like a human is assembled molecule-by-molecule by (?)illions of tiny nano-machines, which like all machines are just physical things that transform flows of energy through them. A human being is an absolutely insanely complicated thing, but when you analyze one sufficiently it turns out to be an aggregate of a (whole frickin') bunch of simple molecular reactions.

What you seem to be missing is the realization that the idea that human decision making is determined by natural forces is a groundless assumption. How would you ever set about testing it? — Janus

I don't think human decisions are determined by physical forces but neither are they determined by psychic forces outside of nature (as they seemingly would need to be for the sort of 'rollback' libertarianism under discussion to be cogent). I rather think that human decisions are determined by human beings. The sort of causation at issue is a sort of agent causation, which is an exemplification both of substance causation and of rational causation, on my view. Embodied and encultured rational human beings can determine things to happen on rational (and ethical) grounds.

1. A philosophical book is just an arrangement of words. If there is nothing new in philosophical books, why write them? Why read them?
— Olivier5

For the same reason that we collect a variety of goods and organize them on shelves in the same building called a "store": — Pfhorrest

You are talking of dictionaries. I am talking of philosophy books.

The possibility [of a steam engine] was always there, — Pfhorrest

The possibility of a steam engine is something very different from a real steam engine, though.

human being is an absolutely insanely complicated thing, but when you analyze one sufficiently it turns out to be an aggregate of a (whole frickin') bunch of simple molecular reactions. — Pfhorrest

The point is that this thing can reproduce, while an atom cannot. It can decide to fight or flee. It can sleep. It can eat and drink. It can observe, and it can think.

These behaviors emerged through life. They have no meaning outside if it.

The point is that this thing can reproduce, while an atom cannot. It can decide to fight or flee. It can sleep. It can eat and drink. It can observe, and it can think. — Olivier5

All of which can in principle be broken down to complex arrangements of behaviors of atoms, without requiring that anything happen besides what those atoms could already do.

You’re talking about “weak emergence”, about which there’s really no debate; weak emergence is still reducible. I’m against strong emergence, as I already specified in the OP.

those atoms could already do — Pfhorrest

I repeat: atoms cannot reproduce.

And I repeat: reproduction is a product of things atoms can do.

It has no meaning at their level.

What you seem to be missing is the realization that the idea that human decision making is determined by natural forces is a groundless assumption. How would you ever set about testing it? — Janus


I don't think human decisions are determined by physical forces but neither are they determined by psychic forces outside of nature (as they seemingly would need to be for the sort of 'rollback' libertarianism under discussion to be cogent). I rather think that human decisions are determined by human beings. The sort of causation at issue is a sort of agent causation, which is an exemplification both of substance causation and of rational causation, on my view. Embodied and encultured rational human beings can determine things to happen on rational (and ethical) grounds. — Pierre-Normand

OK, but where I have used "natural" you have substituted "physical". I'm not sure whether you draw a distinction between them, but as far as I understand, determinism is the thesis that all events are fully determined by antecedent physical events. This is often expressed in the thought experiment wherein it is claimed that if the evolution of the universe were to be played out again from the Big Bang everything would unfold again exactly as it has.

So, for me the kind of determinism which incorporates reductive physicalism is logically incompatible with the kind of freedom that could rationally be understood to justify the idea of moral responsibility.

So, I agree with what you've said above, but as I read it, what you've said does not support compatibilism, but rather rejects it.

It has no meaning at their level. — Olivier5

Sure, but that's not relevant to the point at hand. One cannot meaningfully give the temperature of a single particle, only an ensemble of particles, yet nevertheless the temperature of that ensemble consists of nothing more than an aggregate of properties of the particles: it's not like there's the kinetic of energy of every single particles plus the temperature of the ensemble as a separate thing; the temperature of the ensemble just is an aggregate of the kinetic energies of the individual particles.

Likewise, a human being doesn't do some separate thing of "reproduction" that then somehow relates to a bunch of separate nanoscopic molecular actions that lead to the construction of a new human body atom by atom; the act of human reproduction just is all of those nanoscopic molecular actions in aggregate.

Like I said, you're talking about weak emergence, which is not anything I'm opposed to. I'm only against strong emergence. See for explanation: https://en.wikipedia.org/wiki/Emergence#Strong_and_weak_emergence

the temperature of the ensemble just is an aggregate of the kinetic energies of the individual particles. — Pfhorrest

Un fact it is more complicated than that, but I see what you mean.

The concept of emergence does not imply that magic things happen between elements when they arranged a certain way. This is a strawman. These concepts simply mean that certain things happen when elements are arranged a certain way, which would not happen to the elements taken in isolation. So for instance a sentence has meaning, in a way that the letters composing it lack. The meaning of the phrase Cogito ergo sum is not the meaning of C + the meaning of O + the meaning of G, etc. It is not even the meaning of "cogito" + the meaning of "ergo" + the meaning of "sum", although that'd be a bit closer to it.

The concept of emergence does not imply that magic things happen between elements when they arranged a certain way. — Olivier5

Specifically strong emergence does, and as I've repeated at least three times now, including right in the OP, that is the specific kind of emergence I'm against. Weak emergence is fine, but also trivial; nobody is opposed to that, so there are no problems (caused by the lack of it) that it would solve.

That distinction between strong and weak emergence is unclear to me and I suspect it is spurious. Either some new property or behavior emerges when elements are arranged a certain way, which would not happen to the elements taken in isolation or arranged another way, or it does not emerge.

It's a question of whether the new behavior is an aggregate of the behavior of the constituents (weak emergence) or not (strong emergence).

E.g. if you modeled or simulated a bunch of particles and had no concept of "temperature" baked into your model, just the motion of particles, you would see temperature phenomena just show up in your model or simulation automatically (for large enough ensembles of particles), because it's just a direct product of the properties of the particles. That's weak emergence: you could just pay attention to the constituents and you would get the aggregate behavior out of it anyway whether you paid any attention to it or not.

In contrast, to model or simulate a strongly emergent phenomenon, you couldn't just model the behavior of the constituents and let it happen on its own; you would need to add some extra logic to the model or simulation that says that when a certain type of collection of constituents get together this certain way, start doing this new kind of thing.

So in that way, "nothing new" happens in the weakly emergent model, you don't have to specify anything other than the behavior of the constituents to automatically get the "new" behavior of the aggregate; while on a strongly emergent model, something "wholly new" happens in aggregate, in that if you just model the behavior the constituents, you miss out on modelling the "new" behavior of the aggregate entirely; your model wouldn't behave like the supposedly strongly-emergent reality would, if your model only modeled a bunch of constituents.

Rejecting strong emergence means rejecting that there's anything in the universe that can't in principle be fully modeled just by modeling the behavior of the constituents, because you'd always automatically get the same behavior of an aggregate of them in the model as you would in reality; nothing ever strictly needs to be added to the model to handle the behavior of aggregates.

(Of course, for practical purposes, we often only care about the aggregate, and can ignore the constituents, and use simpler models of the aggregate behavior while ignoring the behavior of the constituents. That we can and often want to do that is different from a case where we have to because modelling the constituents just wouldn't do it).

It's a question of whether the new behavior is an aggregate of the behavior of the constituents (weak emergence) or not (strong emergence). — Pfhorrest

It all boils down to what kind of aggregation one is talking about then.

Aggregation: the formation of a number of things into a cluster.
"a single dose of aspirin irreversibly inhibits the normal aggregation of platelets"

This word therefore calls to mind a naturally collected, loosely structured mass of elements.

Would one call a house "an aggregate of bricks"? The bricks do not merely aggregate into a house like platelets aggregate into a clot. A house has a very definitive structure, which may be made of certain types of bricks or other material but the material does not naturally aggregate into a house.

Likewise, a word is composed of certain sound types (phonemes). When written, the word is composed of certain letters, arranged in a specific way, otherwise it's another word. It is a structure, not a mere clot of letters.

Likewise, a sentence is composed of words, but it is not a passive aggregate of words. A sentence binds its elements into a precise structure to mean something specific.

The structure is not comprised in the elements, and it brings something new. This structure is what you hide to yourself when you use the word "aggregate". It's what you cannot account for in your system. You have no sense of structure.

OK, but where I have used "natural" you have substituted "physical". I'm not sure whether you draw a distinction between them, but as far as I understand, determinism is the thesis that all events are fully determined by antecedent physical events. This is often expressed in the thought experiment wherein it is claimed that if the evolution of the universe were to be played out again from the Big Bang everything would unfold again exactly as it has.

So, for me the kind of determinism which incorporates reductive physicalism is logically incompatible with the kind of freedom that could rationally be understood to justify the idea of moral responsibility.

So, I agree with what you've said above, but as I read it, what you've said does not support compatibilism, but rather rejects it. — Janus

Yes, I had purposefully replaced "natural" with "physical" because I meant to deny that human actions having natural causes must imply that they also have (fully determinative) physical causes. Endorsing naturalism without physicalism is enough to ward off implausible forms of super-naturalism, I would hope.

It seems to be assumed by many philosophers that (1) the thesis of micro-physical determinism conjoined with both (2) the thesis of the causal closure of the physical and (3) some doctrine of supervenience, imply (4) the thesis of universal determinism, such that human actions, in addition to everything else that takes place in the natural world, are fully determined by the past physical state of the universe (at some arbitrary moment in time). This would appear to follow from Jaegwon Kim's causal exclusion argument (according to which, roughly, anything that happens has a sufficient physical causes and higher level 'causes' therefore are epiphenomenal). But I think Kim's argument fails to apply to human actions (and to scores of other natural and social phenomena) because he gets the metaphysics of those higher-level phenomena (and of higher-level material agents) wrong.

As I had suggested earlier, human actions can't sensibly be construed as causal impacts that some agent exerts on the material world, at a moment in time, from outside of the world, as it were. Human beings are material beings in the world. As such, they are not divorced from their own pasts, and the material (bodily) and cognitive powers that they possess at some moment in time aren't features that merely determine them to act but rather constitute them acting in the world.

Hence, when a human being reflects on the physical determinations of her own embodied actions, some of those determinations indeed constitute external constraints on what she can do. (Someone, for instance, may be unable to deadlift 300 pounds, or mentally calculate 10 decimals of Pi). But some other among those physical determinations realize instances of that person's own abilities. They are enabling causes, rather than constraints, with respect to an agent's physical abilities and her powers of rational practical deliberation.

So, let us suppose that we grant the possibility of micro-physical determinism being true, and of the thesis of the physical-closure of the physical being true as well, but nothing more. It remains true, then, that given 'the past', as it was, there was no possibility that some agent, who actually did A, would (counterfactually) have done anything else consistently with the past having been as it was. So, indeed, if the state of the universe was 'rolled back' to some earlier state, the agent would still necessarily do A and nothing else. But that doesn't imply that this agent was constrained by the past physical state of the universe to do A. That's because the agent herself figures in that past (at least recently). And so she is (and was) an active participant in the process of determination of the future. And when we say that she could have done something differently (or refrained from doing it), we don't mean that she could have done it consistently with there being nothing about herself that would have been different (including the states of her character, her motivations, her reasons, habits and inclinations, etc.) but rather that nothing outside of herself fully determined her to act in the way she did.

Some worries that remain may stem from regress arguments against ultimate responsibility (such as Strawson's 'Basic Argument' for hard incompatibilism) or consequence arguments that set up regress of antecedent determinative physical causes reaching back to a past time before the agent existed. But I think I should address those worries in a separate post.

I’m not at all denying structure. If you simulate or model a house, it is enough that you simulate or model the bricks etc that it is made of — arranged into the structure of a house, of course, but you don’t then have to model properties of the house specifically, just the properties of the bricks etc, and when they are put together like that the properties of the house show up automatically.

Your language examples are a bit beside the point, because we make up the rules of language and so can make up strong emergence in them if we want. We could also make a simulated universe where when constituent come together the right way new rules start applying, in addition to the aggregate of all the rules applying to the constituents. We can make up whatever we want for abstract things we invent. The question is whether the real world behaves in that way.

when they are put together like that the properties of the house show up automatically. — Pfhorrest

Show up = emerge.

Note that one can make several different houses with the same material, so the structure of the house is additional to the material. It is not 'contained' in the material.

Your language examples are a bit beside the point, because we make up the rules of language and so can make up strong emergence in them if we want. — Pfhorrest

Okay, so strong emergence happens, I suppose...

Show up = emerge. — Olivier5

Weakly, yes, but that’s not in question.

Okay, so strong emergence happens, I suppose... — Olivier5

If we make up something with rules where it happens, sure.

The question is whether the universe follows such rules or not.

But if strong emergence happens in language, it can happen outside of it too. Otherwise, when did strong emergence emerge? :-)

That's like saying "but if magic can happen in stories, it can happen outside of them too".

We can make up anything we want for things that exist only in our minds. Letters, words, and sentences don't intrinsically mean anything that you could discover by studying just scribbles or sounds etc; they only mean things to people, and we find out what they mean by studying the people.

It's like any social construct: something "is money" not because of any properties of the thing, but because of how people treat it; to find out what is or isn't money, you don't investigate the things themselves, you investigate what people think about the things.

People can choose to assign meaning to a word that is not a composite of the meanings of the letters, as we obviously do, but that doesn't mean that there's any actual strong emergence in the real world, only in the stories and games we make up and tell each other.

To come back to money for analogy: we could decide that 100 cents together as a unit count as more (or less) than 100 times the value of a cent, if we wanted to do things that way. That doesn't tell us anything about the properties of pennies or dollars or whatever; that only tells us about the stories and games we've made up about pennies and dollars etc.

Pardon but ...

By 'weak emergence' I understand properties reducible to some 'configuration of simples' (e.g. rainfall from convection of water vapor).

By 'strong emergence' I understand properties reducible to complexes of interrelated 'configurations of simples' and, therefore, irreducible to discrete 'configurations of simples' themselves (e.g. climate systems from chaotic interractions between local geography, ground temperature, prevailing winds, barometric pressure-gradients, average humidity, etc).

Oversimplifications notwithstanding, tell me what I'm misunderstanding here.

It's possible that the senses of the terms I'm using aren't the only once, but in those senses as I learned them, your second example would still be weak emergence, because in principle you can still get chaotic weather behavior from just modelling a whole frickin' lot of air molecules as such (it's just that, since it's chaotic behavior, a tiny chance in some of the molecules in your model will produce a big change in the aggregate behavior, which makes building a model that predicts reality really hard). You don't have to specifically add some extra property of "weather" to the model, you can (in principle) get by with just modelling a bunch of molecules.

I don't know that rainfall would count as any kind of emergence, but maybe you can explain further why it would.

Basically, the distinction as I understand it is that weakly emergent behavior is still reducible in principle to the behavior of the 'simples' as you say, while strongly emergent behavior is not reducible.

That's like saying "but if magic can happen in stories, it can happen outside of them too". — Pfhorrest

No. It's like saying: if consciousness happens in humans, it must have a precursor in the animal kingdom. It cannot stem from nothing.

People can choose to assign meaning to a word that is not a composite of the meanings of the letters, as we obviously do, but that doesn't mean that there's any actual strong emergence in the real world, — Pfhorrest

But language is real. If strong emergence is a fundamental characteristic of all human language, where does it come from? How did strong emergence emerge?

Personally, I don't buy the distinction between weak and strong emergence. I see the latter as a sum of many small (weak) emergence events. It's a quantitative difference, not a qualitative one. I use the term "strong emergence" to be better understood by @Pfhorrest only.

It's like saying: if consciousness happens in humans, it must have a precursor in the animal kingdom. It cannot stem from nothing. — Olivier5

That is, ironically, a very anti-emergentist line of argument; it's basically the line of argument that underlies my panpsychist position on phenomenal consciousness. Whatever besides ordinary physical behavior (if anything) is involved in human consciousness, some precursor of it must exist in everything, because otherwise there would be some point in the construction of a human from more fundamental things where suddenly a property that wasn't built out of (and so reducible to) more fundamental properties just appeared from nowhere. That can't happen -- that's strong emergence -- so whatever property we're talking about either doesn't exist at all even in humans or else some precursor of it exists in everything.

But language is real. If strong emergence is a fundamental characteristic of all human language, where does it come from? How did strong emergence emerge? — Olivier5

Stories are real, as in, people really to tell stories. Things happen in stories that can't happen in real life. If they can't happen in real life, how can they happen in stories? zomg big philosophical mystery? No.

We make up things all the time that doesn't mirror reality. Language is a thing we made up. It's real inasmuch as we really do make up and use languages, like stories are real. But like I said, you can't learn about a language just from studying sounds and scribbles -- you have to study how people use (and so implicitly think about) those scribbles and sounds. The facts of the language are actually facts about human thoughts (with regards to those scribbles and sounds), and those thoughts can be about things with structures that don't actually exist in reality.

Personally, I don't buy the distinction between weak and strong emergence. I see the latter as a sum of many small (weak) emergence events. — Olivier5

Then what you're calling "strong emergence" is not the same thing I'm talking about. You're still just talking about weak emergence. It's fine if you think strong emergence is a useless idea -- I do to, that's why I'm against it. But don't take me saying I'm against that useless idea to mean I'm against something normal and mundane that so far as I know nobody is against.

The whole point of bringing up strong vs weak emergence is that you take me to be arguing against something I'm not arguing against. I'm trying to say what the thing I am arguing against actually is. If you just deny that there's any difference between them, then you're taking my words to mean other than I mean them, and counter-arguing against a straw opponent whose position is not mine.

Stories are real, as in, people really to tell stories. Things happen in stories that can't happen in real life. If they can't happen in real life, how can they happen in stories? zomg big philosophical mystery? No. — Pfhorrest

But the argument is NOT about what language SAYS but about how it WORKS.

Human language uses elements such as phonemes or letters to forge words. The meaning of words is not contained in their elements (letters).

In turn, language uses these words as elements of structured sentences, and sentences are elements of paragraphs and book. A word can have many meanings, and the meaning of a word has to be interpreted within its context (sentence or group of), and thus the relationship between elemental word meaning and sentence meaning is a complex one, a two-way relationship. It is NOT additive:

Meaning of word <> meaning of letter 1 + meaning of letter 2 + meaning of letter 3 + ...

Meaning of sentence <> meaning of words 1 + meaning of word 2 + meaning of word 3 + ...

It is in this sense that one can say: the whole is more than the sum of its parts. There is something in the whole that is not present in the parts taken in isolation: the connexions, the interrelations, the synergies.

If language is not additive, if complex structures play a central role in it, why can't the same type of non-additive structures exist elsewhere?

If such non-additive, emergent structures exist in our mind, if they are fundamental to the way we think, doesn't it stand to reason that they can exist outside of our mind? Ontologically, that is.

Even if one contends that structures are views of the mind but do not exist in and by themselves in the outside world, we still think in terms of structures, so where does this seemingly new human ability come from? Did it emerge from nothing? That would contradict your stance that emergence does not emerge...

But the argument is NOT about what language SAYS but about how it WORKS. — Olivier5

And the way that it works is something that we made up. We decided to make languages where words mean more than the sum of meanings of their phonemes etc. That structure cannot be found in the sounds or scribbles themselves; it exists only in the stories we tell ourselves about what those sounds and scribbles mean, in the rules of the games we play with them.

But indeterminism is not a threat to the metaphysical sense, since that sense just is freedom from determinism, which just is indeterminism. And then we circle back around to indeterminism not being a useful kind of freedom... which just goes to show that the metaphysical sense of the term "free will" is not a useful sense of the term. — Pfhorrest

Speaking of circling back......hope you don’t mind.

Does your statement “indeterminism is no threat to the metaphysical sense” meant to indicate a metaphysical sense of free will? In which case, the statement then becomes.....indeterminism is no threat to the metaphysical sense of free will. If so, does it follow that indeterminism is no threat to the metaphysical sense of free will because (or, iff) the will is taken to be free to make determinations of its own kind, by its own right, in a metaphysical sense? But it doesn’t follow from that, that your “that sense just is freedom from determinism”, which if indeterminism is no threat because determinism is the case, contradicts itself. Indeterminism is freedom from determinism, but the metaphysical sense of free will makes determinism necessary, so indeterminism IS a threat to the metaphysical sense of free will.

Ok, so.....indeterminism is no threat because....or, iff....the will is free as a determining functionality. How, then, does it follow that the metaphysical sense of free will is not a useful sense of the term? How is it that the metaphysical sense is not the only possible sense of free will there can be, without getting involved in that damnable “....wretched subterfuge of petty word-jugglery...” (CpR, B1,C3, Para 45, 1788)?

I submit, and that rhetorically, the term “free will” is useless because the concept of “free” does not belong to the will as much as does the concept of autonomy belong to it necessarily. But this goes further afield than your thesis admits, as far as I can tell. You know.....because you never mention the word.
————-

And to the extent that determinism is not true, indeterminism is true, which then makes the argument for hard incompatibilism: one way or another free will is impossible. — Pfhorrest

Isn’t there a need to distinguish kinds of determinism? If physical determinism is not true with respect to the metaphysical sense of free will, I don’t agree indeterminism is therefore true, under the same conditions. Given the metaphysical sense of free will, it is logically consistent that the sense of determinism should itself be metaphysical, in which case, determinism must be true if it be the case that the metaphysical sense of free will abides exclusively in its law-giving functionality. I don’t think it is reasonable to suppose that because a metaphysical sense of determinism is not susceptible to inductive support in the same way as physical determinism, that the conception is therefore inherently flawed.

Clarification, not counter-argument.

And the way that it works is something that we made up. — Pfhorrest

Correct, like all philosophy, like all art and all science... Like cars or computers, or zillions of other things.