Edit: For context, the quoted user made a (now strangely deleted) post commenting on his (hard to say) either disapproval or genuine sense that the definition can be improved as far as the 2nd post on this topic by Baden — Outlander

It’s fine that it was deleted— it was superficial. But to elaborate: I think the definition given (sincerely) is way too technical. Maybe it makes some sense and maybe it’s worth studying (as someone who respects Heidegger’s works, I grant that possibility) — but outside academia, I don’t see it being helpful in any way in life. I like ‘s description.

I like ↪T Clark ‘s description. — Mikie

If nothing else, it has the virtue of brevity.

My understanding is that Luhmann worked on social systems, thus not a general systems theory

Thank you for your contribution - I was hoping for a concise definition backed by a general systems theory. Notions of systems are a dime a dozen

I agree with your definition, even though I have used some different words. Do you know of any theory that backs up this definition?

These "irreducible entities", is that the fundamental particle that the physicist are trying to find?

I don't think systems <=> coherencies is any definition.

I don't blame Mikie for his reaction (he didn't ask the question...). I'm reading a compilation of lectures by Luhmann at the moment, so these ideas are on my mind.

My understanding is that Luhmann worked on social systems, thus not a general systems theory — Pieter R van Wyk

He built from the mathematical work of Spencer Brown and the biological work of Humberto Maturana a kind of general system's theory that could be applied to society, but can also be applied to e.g. consciousness and other (autopoietic) systems.

But, yes, others may be more on point re what you're specifically looking for.

I don't think systems <=> coherencies is any definition. — Pieter R van Wyk

I didn't say they just were that. I said this:

Systems are coherencies of (self-recreating, in the case of autopoietic systems) differences between themselves and an environment — Baden

And I also clarified what I meant.

You asked a question. I answered with a definition and a detailed follow-up. I also gave you the theoretical context (Luhmann). You had everything you needed to make some sense of it.

If you don't want to get your brain out of first gear, don't ask in the first place.

[In general, if posters want non-academic, dumbed-down answers to their questions, why not just ask AI, why ask on a philosophy forum?]

In its most general sense, a system can be understood as an organized interconnected set of at least two or more components that collectively constitute a unified whole. The behavior of the whole is conditioned by the interactions of its parts, while the parts, in turn, derive their functions and significance from their relation to the whole. In this respect, a system is indivisible, as the whole cannot exist independently of its parts, nor can the parts operate meaningfully apart from the whole. A genuine system necessarily gives rise to emergent properties, thereby becoming more than merely the sum of its parts.

Yes, in the context of physicalism.

I agree with your definition, even though I have used some different words. Do you know of any theory that backs up this definition? — Pieter R van Wyk

It's not a principle, it's a definition, so there is no theory backing it up, just a consensus of the meaning of the word among users. What I posted is my understanding of the consensus in this particular case.

Here's my version as simple as I can make it. Do with it what you will.

A system can be seen as a "coherency of differences" as follows.

Difference A: Fundamentally, a system must establish a difference between itself and its environment.
Difference B: A system must "observe" / react to differences in that which is different to itself, i. e. its environment.
Difference C: A system must operate on the basis of internal differences. A pure homogeneity excludes operationality.

A, B, C, type differences must cohere in a system for a system to be identified singly as a system.

Therefore, a system can be seen as a coherency of differences.

(No meaningful difference can arise until all those differences arise coherently. A pure environment contains nothing to establish difference. Difference requires a coherent system that is different from its environment (A) and that is different within itself (C) for the differences described in (B) to manifest.)

My problem with this is it lapses into substance ontology which is reductionist. An ontology of stuffs rather than of processes or the holism of systems of self-stabilising interaction. . . .
If we are using physical jargon, then entropy-information is a good dichotomy but also locks us into an ontology of substance rather than process. — apokrisis

I don't understand that assessment. Energy & Entropy are Processes, not substances. Information --- or EnFormAction, as I like to spell it --- is also a process. Systems are mental concepts that categorize collections of interacting "stuffs" as-if unitary things. Which, as Organized Structures, we tend to think of as single substantial objects. So, I view Holism/Systems as an Ontology of Processes (causation ; change) instead of stable-but-malleable Matter.

If you agree with Donald Hoffman's Interface Theory*1, even Matter is a conscious construct, that humans use to guide their physical interactions with the world. Another way to look at Ontology is to view the Real World as a multilevel system of acting & reacting sub-systems. We physically "see" a superficial layer of reality, like "icons" on a computer screen. But lower, more fundamental, layers are where the action is. And, what we call Systems, are mostly interactions on the lower levels of reality. Our idea of a System*2 is based on our ability to conceive of invisible-intangible extra-sensory qualia --- "more than the sum" of material parts --- that makes it a Holistic concept.

That's an Idealistic philosophical approach, but for practical purposes, common-sense (science) may be a better guide to dealing with Reality. :smile:


*1. Matter in Mind :
Donald Hoffman argues that matter is not a fundamental aspect of reality but rather a symbolic representation or "icon" constructed by consciousness, similar to icons on a computer interface. In his theory of conscious realism, he posits that consciousness is the fundamental reality, and the physical world, including matter and spacetime, emerges from a network of conscious agents. Matter, in this view, doesn't exist independently of consciousness but is a useful, though not literal, construct for interacting with the world.
https://www.google.com/search?client=firefox-b-1-d&q=don+hoffman+on+matter

*2. Systems Theory :
A system can be more than the sum of its parts if it expresses synergy or emergent behavior. Changing one part of the system usually affects other parts and the whole system, with predictable patterns of behavior. More parts, means more interrelationships, and more complex properties & activities, including mental functions.
https://blog-glossary.enformationism.info/page18.html

And of course this is the thing about systems in interaction with their environments: they attempt to achieve predictability (and thus a kind of rigidity) not just by refusing to see what doesn't fit (as the counterculture would have it) but by making their environment more predictable, by eliminating what doesn't fit. Adaptation is required for the system to persist, but it can adapt itself to its environment or its environment to itself. — Srap Tasmaner

Yes, and your examples were very interesting in that respect. A system must be less complex than its environment and it reduces complexity through a kind of code that "sees" only certain things in that environment. That becomes its reality. This is why I was saying earlier that systems establish different versions of reality. They observe and are perturbed by their environment but interpret it only according to a particular internal code that identifies them as a particular type of system. They create differences that determine their reality, and thus enable a form of "meaning" or "cognition" in a broad sense. That is, they are operationally closed (operate only according to their own internal rules or code), but they are cognitively open in that they are affected by their environment and interact with it. I'm not sure I would call this a "rigidity" but perhaps a structural limitation. But paradoxically, it is the reduction of complexity that allows systems to complexify (and adapt), and in fact reach higher orders of self-referential complexity (self-managing of complexity). The more efficiently they simplify, the more efficiently they can complexify in a sense.

it is the reduction of complexity that allows systems to complexify (and adapt), and in fact reach higher orders of self-referential complexity (self-managing of complexity). The more efficiently they simplify, the more efficiently they can complexify in a sense. — Baden

I'm so glad you came back to this, because that's an excellent point. (And, for what it's worth, close to my own thinking about the utility of simplifications like logic, mathematics, language, music theory, maps, all that jazz.)

they are operationally closed (operate only according to their own internal rules or code), but they are cognitively open in that they are affected by their environment and interact with it. — Baden

Right. One thing I didn't like about my earlier post was it ends up sounding too much like we're only talking about modeling, but we want to be able talk not just about "this is how I symbolically represent and predict the outcomes of horse races" but also "this is how I cut planks to the lengths I need," where this later phrase refers not to a verbal description of me doing it, but to me doing it. The system in operation, interacting with the environment, rubber meeting the road in a more than cognitive sense.

What's tricky is to find the natural correlate of simplification by abstraction in non-cognitive (or, at least, not only cognitive) interaction. There is an obvious path in modifying the environment to simplify it (planks as simplifications of trees, extracted from them, with the rest of the tree physically abstracted away), but otherwise I'm not sure, so maybe this is just a hard difference between mental and physical interaction, that there is this freedom in cognitive behavior that you can't quite manage when dealing with the world in the raw, however it comes. Not sure.

The question in the OP is much too broad to be interesting, since the word "system" has multiple meanings, and those meanings are in turn quite general without further qualification.

The meanings of "system" that have been largely left out of this discussion (except by @Srap Tasmaner) are, for example, system as a theory ("Kant's system"), method ("Dewey decimal system," "gambling system"), rules of behavior ("system of discipline").

Within the material context, some definitions that have been given are too restrictive. @Baden's is mostly about differentiating a system from its environment, but in some contexts, environment is irrelevant for our purposes and can be left out of consideration. The only internal differences required of any system are those between the whole and its parts. Stability and rigidity also do not always apply: systems can be dynamic in their composition and form, although it could be argued that some essential features of a system must be invariant within the scope of consideration for it to be recognized as one system. But that is true of any named entity.

@T Clark and @punos gave good general definitions of a material system, and it is pretty clear that not much more can be said on the subject without getting into specifics of particular kinds of systems, such living organisms or ecologies. If we stay at the most general level of a "system," then we are just doing amateur lexicography.

system must be less complex than its environment and it reduces complexity through a kind of code that "sees" only certain things in that environment. That becomes its reality. — Baden

So you wish to limit your definition of a system to an organism then? Which is fine, as code-based or semiotic systems are their own class of thing. Life and mind as opposed to mere physics.

But if talk about systems is talk about some general causal model, then it has to include the physical realm. And we do talk about weather systems, solar systems, atomic systems, ocean current systems and all the other systems that are globally coherent in being hierarchically self-organising.

And the big advance in biology and neurocognition has been to recognise the continuity that underlies the “mind-world” difference. The organism is a system with a code and the environment is also a system - lacking a code but still a system of constraints.

So reductionism is left with no where to hide. It is systems all the way down. And this is why both physics and organisms can make sense within the one larger causal model offered by dissipative structure theory.

The world was already doing something organised. The organism only had to latch onto that grand entropic enterprise as a bit of viral code.

I don't understand that assessment. Energy & Entropy are Processes, not substances. Information --- or EnFormAction, as I like to spell it --- is also a process. — Gnomon

But you had to invent your own term to turn information back into informing. So you clearly can see there is an issue to be sorted.

Entropy/information arose as a way to count bits. Put a number on distinctions - whether they were a difference that made a difference, or even when they were differences that were just noise.

So that was a valuable step. Science could count the differences that any kind of system - physical or organismic - could contain. Then came the difficult bit of adding back a distinction between information that indeed informed, and entropy that instead was work or free energy.

Eventually this has led to the current rich variety of models that put meaning and action back into any counting of bits. We now do have process accounts like dissipative structure theory in physics and the Bayesian Brain theory in neuroscience.

So first we reduce everything to the bare notion of countable differences. We reduce it so far that it completely loses its larger systematic structure - the coherent context that even allows the difference to count as a difference. And then we need to repair the damage by building back some story of a process that is organising this whole show.

Ideally, the whole of reality will then be described under this one common process. But as has been said, organisms are different as they encode their environments in terms of their selfish wants. So under the most general class of systems causality - which I say is dissipative structure theory - you have at least this one major sub-class that contains the novelty; which we can call semiosis, or very loosely, information processing.

If you agree with Donald Hoffman's Interface Theory*...That's an Idealistic philosophical approach, but for practical purposes, common-sense (science) may be a better guide to dealing with Reality. — Gnomon

As epistemology, his point is mundane. As an ontological commitment, it makes the usual idealist mistake.

Idealism is the reaction against reductionism – an attempt to reject a world of only atoms blindly banging around in a void. But idealism fails to replace reductionism with anything better. It falls back on a mind stuff to replace the matter stuff. Making the same mistake in the other direction. Or tries to sustain a Cartesian dualism which tolerable to both the church and the scientist. Each side absolves itself in the other side's "great mystery".

But the holism of the systems science approach is perhaps much crueller than the Idealists ever had in mind. Systems thinking simply incorporates reductionism under its greater causal generality. It takes it in and then sits it firmly in its corner. :grin:

So reductionism models reality as a great complication built up from its fundamental degrees of freedom. Information/entropy are all about counting those. Instead of atoms or even fundamental particles, we can count the quantum numbers that are the discrete states being shuffled about the cosmic board by their wavefunctions. Reductionism goes as small as it could possibly go.

But then the systems view comes in and points out that all differences are differences in terms of larger context. And indeed, the context shapes those degrees of freedom to be the "simple as possible" things that they are. Quantum physics makes this plain too. A maths of symmetry and symmetry-breaking tells us why – in Platonic strength fashion – these basic quantum numbers emerge. The greater context defines its own smallest thing.

So idealism just doubles down on the mystery that reductionism creates. And systems science is the proper philosophical antidote. It shows that reductionism is one pole of the greater whole. The world can be simplified to a large degree. But in the end, it is revealed to be an irreducibly complex whole. And even physics says that directly these days. Idealism is still stuck in the 18th Century so far as metaphysical debate goes.

I think what’s most interesting about this discussion is that, once again, it’s crucial to distinguish between “common sense” notions and technical notions. In the same way as “work” can mean all kinds of things in everyday life, what it means in physics is entirely different.

My problem with what’s called philosophy and philosophers, is that much of the technical jargon often reeks of posturing, of self importance. It’s an attempt to turn philosophy into physics— looks important, and helps to justify academic funding, but one wonders how close this resembles what was done in Greek times.

That being said, why is it important to have a technical notion of “system,” and more importantly (echoing Chomsky): what explanatory theory does it belong to? Is there one?

My problem with what’s called philosophy and philosophers, is that much of the technical jargon often reeks of posturing, of self importance. — Mikie

That being said, why is it important to have a technical notion of “system,” — Mikie

Beyond language, there is the maths. Is that what perturbs you?

Beyond language, there is the maths. Is that what perturbs you? — apokrisis

No.

An account that is extremely general, simple but I believe mathematically rigorous, well-defined.

https://scholar.google.co.uk/scholar?cluster=7909771384315425233&hl=en&as_sdt=0,5&as_vis=1

https://scholar.google.co.uk/scholar?cluster=10954599080507512058&hl=en&as_sdt=0,5&as_vis=1

(quotes from second paper in abstract and introduction part)

"This monograph attempts a theory of every ‘thing’ that can be distinguished from other ‘things’ in a statistical sense. The ensuing statistical independencies, mediated by Markov blankets, speak to a recursive composition of ensembles (of things) at increasingly higher spatiotemporal scales. This decomposition provides a description of small things; e.g., quantum mechanics – via the Schrödinger equation, ensembles of small things – via statistical mechanics and related fluctuation theorems, through to big things – via classical mechanics. These descriptions are complemented with a Bayesian mechanics for autonomous or active things. Although this work provides a formulation of every ‘thing’, its main contribution is to examine the implications of Markov blankets for self- organisation to nonequilibrium steady-state. In brief, we recover an information geometry and accompanying free energy principle that allows one to interpret the internal states of something as representing or making inferences about its external states. The ensuing Bayesian mechanics is compatible with quantum, statistical and classical mechanics and may offer a formal description of lifelike particles."

"To address the nature of things, we start by asking how something can be distinguished from everything else. In pursuing a formulation of self organisation, we will call on the notion of conditional independence as the basis of this separation. More specifically, we assume that for something to exist it must possess (internal or intrinsic) states that can be separated statistically from (external or extrinsic) states that do not constitute the thing. This separation implies the existence of a Markov blanket; namely, a set of states that render the internal and external states conditionally independent. The existence of things (i.e., internal states and their blanket) further implies a partition of the Markov blanket into active and sensory states – that are not influenced by external and internal states, respectively. This may sound a bit arbitrary; however, this is the minimal set of conditional independencies" – and implicit partition of states – that licenses talk about things (that possess states). Specifically, it provides a partition that constitutes the ‘self’ in self-organisation. The subsequent sections tackle the next obvious question: what are things? At this point, we deploy the Langevin formulation of random dynamical systems as an ansatz that is recursively self-verifying, when considered in the light of Markov blankets. In brief, the formulation on offer says that the states of things (i.e., particles) comprise mixtures of blanket states, where the Markov blanket surrounds things at a smaller scale. Effectively, this eludes the question “what is a thing?” by composing things from the Markov blanket of smaller things. By induction, we have Markov blankets all the way down, which means one never has to specify the nature of things."

So a system is a thing with a Markov blanket that separates the kind of thing it is from its environment. This can be applied to virtually anything complicated enough, from a rock to a brain to a planetary system to... virtually anything. The internal states of the system can then be descibed as modelling the states if its environment.

This can be applied to virtually anything complicated enough, from a rock to a brain to a planetary system to... virtually anything. — Apustimelogist

You may misunderstand. A rock doesn’t actually have beliefs about its environment.

So Bayesian mechanics is based on thermo maths - the minimisation of free energy. And that formalism is a way to model the beliefs of organisms. An organism models its world with the intent to minimise its surprisal - an index of its prediction error that can be written down in thermo maths.

But a rock has no world model. Put a hot rock in a cold place and it will indeed minimise its free energy as the equations describe. It will go cold in a way that says its internal state could be regarded as a model of its external environment. But the rock never had any say in the matter.

Whereas put an organism in a place it doesn’t like and it will keep moving until it finds some place it does.

But I have mentioned the Bayesian Brain as exactly this kind of exercise of marrying the science of organisms to the science of thermodynamical systems. So same mathematical framework. But with the essential twist that an organism is in a modelling relation with its world.

The organism is at root just another thermodynamic system. However it is also this special kind of thermodynamic system.

A rock doesn’t actually have beliefs about its environment. — apokrisis

It does. Bayesian mechanics and Free energy principle can apply to anything sufficiently complicated. As said in the first paper I linked before: if something persists over time, it must be encoding a model of the relevant environment we have used in constructing or picking out this system. Doesn't matter if its a rock or a person or an ecosystem, a society. Obviously, a rock may not be very interesting though as a kind of dynamical system.

thermo maths. — apokrisis

Information theory. Rather than calling that thermo math, we should be talking about thermodynamics in terms of a statistical mech.. oh wait!

But the rock never had any say in the matter. — apokrisis

Well the notion of model being used is far more minimalist and general than what you're implying. The FEP does apply to any thing, and so provides a generic characterization of any system complicated enough to have anything interesting to say about it, including stones and even smaller, simpler systems for that matter than stones, I should think.

It does. — Apustimelogist

No, it really doesn’t. The information that the rock contains bears no resemblance to a system of belief.

You can present your evidence to the contrary if you wish of course.

Obviously, a rock may not be very interesting though as a kind of dynamical system. — Apustimelogist

I was reacting to the first paper. The second by Friston is far more challenging and I’m glad you flagged it.

A quick point is that the kind of dynamics that could even be coupled would have to be in a state of criticality. So a whole landscape of moving and eroding rock could be viewed as a hierarchical system in the way being suggested. It is a tectonic flow. A balance of geological and chemical forces over many scales of being. In some sense its own model as at some particular distance or horizon, the landscape’s smallest fluctuations become a lower bound blur, and its largest fluctuations become so large the system now appears to live inside a fixed background, captured by its laws.

So Friston is walking familiar ground. But then you can see how the humble rock lacks that kind of dynamics which brings this systems perspective into things. The rock has congealed and merely erodes. If the tectonics could be considered lively, the rock is as unlively as it gets.

I’ll have to read Friston’s monograph more closely. But on a skim, I would say he is trying too hard to explain everything by the self-organising dynamics and being too glib about the self-information or measurement aspect of a hierarchical system. But a fun read so far.

So you wish to limit your definition of a system to an organism then? — apokrisis

Just a quick point for now. Definitely not (at least not literally). This is part of Luhmann's project, actually---to extend Maturana's concepts from biology to e.g. society (and also to find a mathematical basis for systems using Spencer-Brown's work). As for the connections to biosemiotics, I'm very interested in that, but my knowledge of biosemiotics is undeveloped. I'll try to come back to your post later anyhow.