Dear Luca,

I don't know if I understood your question, but I think Bauman's work might help you. In the liquid life, as he explains it, there is an emergency of. His book on "Identity" has the concept. There are other books of his you may also find about this concept.

Hope to have helped!

Eveline

My understanding is that the term emergent properties is used in a variety of ways - I have certainly used the term in different ways.  In my early work, some published in American Psychologist and some with Irving Kirsch, we focused on the relationship between physiological and psychological theories.  In my more recent work, I have used ideas from parallel distributed processing  or network theory in relation to emergent properties of the body.  I think your general description in your question is correct, but the way that applies varies considerably.

Best wishes, Michael

Thank you for your answers!

Eveline, I will check out your suggestion. My interest is more within what is usually called analytical philosophy, and the specific type of "emergence" relation I am interested in is that which is historically associated with Broad and the british emergentists. Typically it is discussed in relation to biological properties, or in relation to physiological and psychological properties. I will take a look at Bauman's work anyways, since it is always nice to have a broader outlook, at least as far as understanding what is "going on" out in the theoretical jungle.

Hi Michael, thank you for your answer. That was helpful - I will check out your work. One of the reasons I was asking is that I had been expecting to find more attempts at spelling out emergence as something that describes the relation between types of "things", not properties. Especially because a lot of the problems associated with the apparant physicalist choice between reduction and epiphenomenalism, or non-reduction and downward causality depend on examples of problematic causation under a physicalist interpretation that attributes properties at different levels to the same entity.

But since potential emergent properties are always properties of objects at a relatively higher "level", if one provided an object oriented analysis, and some theory of properties (for instance as sets of capacities or something), then the emergent status and unpredictability of properties could be an artefact of the circumstance that they are simply properties of emergent object types - and these of course don't figure in lower order theories.(I am not sure if this will work, but that was my first gut-reaction).

This would get rid of mysterious property emergence, while at the same time allowing for a materialistic theory of emergent objects, if this term makes sense in this context.  The thought is rather vague, and that is why I was a bit disappointed when I took a first glance at the litterature, since it would be really nice if there was some established approaches or sketches taking this route. Perhaps there are, or perhaps I have overlooked something crucial that explains why this approach has not been tested seriously (if I haven't just missed it!).

Anyways, I will take a look at your work and see if it helps me get a clearer overview over what has already been said and done within the topic.

Thanks again for the replies, much appreciated!

I like to define emergence as the unexpected which can be either positive or negative depending on one's point of view. If something that emerges is expected, I believe it is not appropriate to describe it as an emergent event, phenomenon, or property. Those things that emerge as unexpected events that are also surprises that cannot be satisfactorily explained even after observing them after the fact are the most fascinating. I authored a conference paper and a book chapter expanding on this definition/nuansce in 2007, and 2009 or so (?), respectively, if you're interested.

Thanks alot B.E White, that was very useful. I actually need it for an approach to group ontology that analyses Groups as social artefacts, which fits nicely with a complex systems Engineering understanding of organisations. So very useful input.

The problem with the unexpected criterion is that some unexpected things are not emergent, and some expected things are.  The oscillation of a control loop is an emergent property of the loop - and the rate of oscillation set by a combination of lag and gain.  However, it isn't really surprising that control loops oscillate if you think about the time delay round the loop.  The common sense example I give to students is that of adjusting the temperature in a shower.  If you adjust too fast, you get oscillation.

Thank you very much, Luca: If you would like a few of my publications on the subject, I would be glad to forward them to you. To make that easier, please contact me via E-mail at [email protected]. Thank you again for your feedback! Brian

1-

Simple global property that cannot be reduced to property of elements and interaction, like a gaz temperature. If you take three molecules, you cannot calculate temperature. Temperature is a mean. Weak emergence. 

2-

You have non linear equations, you know the parameters, but for certain values of the control parameters, a global property appears that cannot be analytically predicted by equations (because infinite derivative, or infinite variance) . Like in all phase transitions. For instance paramagnetism or crystalisation.

For me, in this case, even if you know all parameters, what the system is depends on what the system does. Its functioning adds some new properties to its initial description. It's also only possible when it is an open system.

3- 

Emergence in biology is much more stronger, because entanglements of levels, and not only interaction between elements. 

best wishes

PA

Hello,

You can look a the work of Jaegwon Kim, for exemple : "Making sense of emergence",1999, Philosophical studies.

His idea is that a strong concept of emergence need a causal asymetry, i.e. a causality from higher level on lower, not predictable on the lower level's properties. Other types of emergence are, in his view, just a kind of supervenience, which have no ontological novelties.

I fully disagree with Kim and its simplistic ideas

you can look for instance:

Acta Biotheor. 2008 Dec;56(4):257-74. doi: 10.1007/s10441-008-9052-y. Epub 2008 Jun 10.

On physicalism and downward causation in developmental and cancer biology.

Miquel PA, Soto AM, Sonnenschein C, 

Thank you Lucas. I have read Kim's account, and that is one of the reasons that I am looking for other accounts. I think he starts out promising, but his underlying causal principles are in my view suspect and perhaps the root cause of the perceived problems with emergentism. I don't think there are any plausible exampels of properties instantiated in the right manner that would serve to support the argument for downward causation in a problematic sense. Unfortunately I haven't found any versions of the view that corresponds to what I expected to find, and before I try to develop that view, I want to make sure that I am not overlooking any obvious contributions or problems.

Thanks Miguel!

It seems like a lot of the examples mentioned in this thread point towards emergent properties of processes, and not of objects, which I hadn't really thought about at all. It complicates the picture a bit - I was just finding my zen in the more "clean" philosophical treatments. Well, well, the World doesn't care about headaches!

I think a part of the emergentist problem comes from the fact that Kim would define as much as destroy this concept. Would that lead to a biased definition?

I have met a biologist (Nadia Haubin Horth, http://www.ibis.ulaval.ca/?pg=nadiaAubinHorth), who works on the evolutionary history of fishes' social behavior. She has shown that an identical ability (for example, inter or intra specific recognition, rigorously quantifiable), appears as a result of many different physiological pathways and gene assemblages, even in closely related species.

I.e., there is no unique molecule, pathway or genes assemblage leading to a corresponding behaviour in related fishes, but a lot of different physical properties causing the same behavioural ability.

I think we have here a great example of demonstrated emergent properties. 

That would seem to be a MR property, and I think strong emergence is sometimes taken to be nonsupervenient, in the sense that MR is due to the sui generis, ontological emergent status of the property. Behavior is a prime example of functional phenomena, so the MR of social behavior is probably benign for one who is an enemy of strong emergence in general. The type of example you mention is something a nonreductive physicalist would pray accounted for all suspected cases of supervenience. Personally I dont believe there are cases of strong emergence in a sense that allows for Breaches of supervenience, but this might be unfounded and due to my own biases towards certain theories of properties (i.e. some causal subset account of higher order properties). Feel free to correct me if I missed the point.

I will take a look at her work, it sounds like a good empirical example.

Speaking of downward causation and emergence, is it correct that emergence is typically seen as relating properties, and that the causation problem involves a typical kimean event causation theory?

If so, then it seems that emergence is a problem only if causal description picks out properties instantiated in objects at the same level where one type is emergent relative to the other. For instance the Classic pain as Cause of behavior and c-fiber stimulation as the Cause of pain (and so on).

Now, one description of such cases would mention mental states of agents and physiological states of parts of agents, and the causal story would be about facts involving different objects. The other would either falsely attribute physiological states to agents, or else characterise them as states of a subsystem of a latter system (i.e. having c-fiber stimulation occuring in some part of ones body), which resembles the first option.

How much of the perceived trouble for emergentism hinges on translating the causal story into one of different property types instantiated in objects at the same level?

Another note: Chalmers mentions the view that some phenomena might be deducible from lower order FACTS, and not lower level laws and a lower level description. This is supposedly a middel position between acceptance of weak and strong emergence. It is something along these lines I have my own sympathies I guess. That emergence might be unproblematic in that sense. If so, then causal explanation involving a casually connected temporal ordering of "facts" could be emergent in something like that sense, without entailing the falsity of a broadly materialistic ontology. If anyone knows of some attempt at following up on a position like the one sketched by Chalmers, that would be great.

I have tried to deal with problems of emergence, novel causal power, downward causation, causal exclusion etc. with regard to the mind/body problem in a chapter from my 2013 book "The Ontology of Psychology" (Routledge). Kim's view, MR, and that of Searle figure prominently in the chapter. I attach it here, in case it might be helpful.

---(((LAWB)))

In a very brief way, I would like to point the idea that emergence evokes the awareness of alienation and subsequent atomizarion of  society members nevertheless they are necessarily intertwined and interdependent. Due division of labour also the representation and rationale toward reality reflects at most organic categories. From primacy of  exchange value comes primacy of tertium non datur principle and exclusion of contradiction. About that you can read Alfred Sohn Rethel. So unicity and totality are usually experienced and represented in mysthic ways, since they look like coming from outside and certainly  heteronomously, no produced by the own elements of the society (system, organism) so apparently as unexplained events that escape the usual rails of standard positive logic (which is based othe idea that society and community is just the random interchange between private elements) but perfectly coherent with the diverse forms of dialectic logic, so it is not a suprise that you can track the category of emergence and expressions such "more than the sum of its parts" in Hegel or in Marx texts. and I would like to venture a dialectic conclusion: in latus sensu, illness is the concrete category where decant all emergent events, it is the emergency par excellence and if it is understood as subjective alienation then it is materialization of alienation and differentia specifica: actual human condition:  lack of and actual human species which is the holistic entity that emergence evokes.

Thank you for your reply, Jorge! From the outset I must say that I am not very wellread or competent in the style of thinking you are. But I am aware of Marx and Hegel as proponents of strong emergence, and of exactly the kind of social ontology that I find intolerable as a foundation for social science in general. When that is said, I also think that there is much valuable in their criticism of so called methodological individualism (though I can't follow exactly what you are aiming at). I mainly know their views in a generic form, through the writings on the subject by Keynes, Menger and economists and sociologists of that breed - perhaps they give an unfair representation of "collectivism" or what we can Call it. 

Actually, one of my interests in the topic is the prospects of providing that individualistic account of the presupposed organic whole that seems to sneak in implicitly when we try to define social objects (Quinton adresses this in his 1975 article on social objects). I find Quinton's views close to the truth, and my interest in this thread is in the general type of concept of emergence/reduction that, in my opinion, once cleared up allows for a more palatable statement of an individualistic social ontology that does Justice to, amongst other things, the phenomena that might lead one to believe that the social phenomena are constitutive of the individual ones.

Take for instance the types of formal social objects, such as the class of unskilled workers. Singling them out involves an implicit reference to the employment relation, as well as the concept of goods, that itself needs social institutions mentioned in the final analysis. So the charge against individualism is, I think, that the social relations are presupposed in talk about types of individual objects used in analysis, and such relations presuppose a social whole, i.e. a proper social object. 

My still underdeveloped suggestion is that these relations depend on social intentions of individuals, and that these can be given an individualistic analysis. I won't present it here, since it needs work and is perhaps a bit outside the topic. But if this can be done, and if one can provide a theory of emergence, capacities/properties and a correct principles of causal explanation, so that social objects are sui generis, yet constructible in individualistic terms - as multiple individuals Standing in certain social relations in virtue of their intentional attitudes - and these objects are then further analysable in their structure in functional terms, by means of a description of actual or intended capacities, then one would (this is what I think at least) have an individualistic ontology that can do without the type of tools you suggest. 

But I think the problems are very Real, and I must admit that my views are very, very sketchy, and that is precisely why I made this thread, so I think your input is still valuable, even if I suspect that it is (if I understand you correctly) what I more or less oppose. Besides, it actually seems like your proposal rests in a type of framework that is useful for understanding one possible reason for Holding that the social is emergent, so it is a fine special case that I will definitely have to consider and use for inspiration. The old collectivists are probably a rare bunch of proud hardcore emergentists that don't try to downplay their commitments, so they might serve as sources for an Extreme expression of emergentism.

Hello Luca—and hello to everyone following this question.

Emergence has to do with something coming newly into existence—with novelty—emergent products, emergent qualities, objects, and processes.

Emergent products of various sorts are observable almost anywhere one chooses to look. For my work I needed an understanding of the occurrence and roles of emergence in the subject matters of each of the disciplines. I looked at examples from various disciplines, searching for commonalities that were present in all cases. What I found was a great diversity of different types of emergence, and yet, there were a few features of emergence that occurred in every case.

(1)       Emergence has two parts, or stages.

First, there is the event of emergence, the event of novelty, of newness, of something coming newly into existence (the part we first notice). Second, there is an underlying process of emergence that results in the event of emergence. (As George Henry Lewes put it in 1891, in the paragraph where he coined the term emergent, “. . . each effect is the resultant of its components, the product of its factors, . . .”)

(2)       The underlying process part of emergence develops.

The emergence process occurs in relatively simple forms in simple situations where few other factors are present and playing roles in the nature of the situation. The process also occurs in increasingly more complex forms in situations that are increasingly more complex due to additional other factors being present and influencing the nature of the situation in which the process of emergence is occurring. Emergence takes on different qualities—it occurs in different forms—in the subject matters of the different disciplines. Therefore it is not possible to define emergence from the knowledge base of only one discipline.

(3)       The emergent product is always a newly occurring pattern-of-material-organization.

The underlying process of emergence is a process of reorganizing an existing pattern-of-material-organization into a new pattern, or a process of creating a newly existing pattern.

(4)       The emergence of new pattern-of-material-organization requires the motion of matter.

To reorganize an existing pattern, one or more of the components must move in relation to the other components of the pattern. To create a pattern, the components must come together—they must move toward each other into the pattern that is thus created.

(5)       Therefore, foundationally and universally, emergence is the coming into existence of new pattern-of-material-organization as a consequence of motion.

All cases of emergence in the subject matters of all the disciplines are based on the process of coming into existence of new pattern-of-material-organization as a consequence of motion. They differ due to the roles of a host of other factors, such as (a) the level of the hierarchic-organization of the material component of the universe at which the emergence occurs, (b) cause which plays an important role in the occurrence of process in general, and (c) the coherence of material units, one with another, which results in material structure.

The above is a very brief description of emergence. There are other significant factors that play roles in the intrinsic nature of emergence that were not discussed. For example that there are two foundationally different kinds of emergence process due to the difference between the roles of sequential-enhancement and combinatorial-enhancement. And then there is the role of transformation-point, another foundation factor determining yet another distinct type of emergence process.

All three factors, sequential-enhancement, combinatorial-enhancement, and transformation-point are based on the motion of matter, and just like emergence itself, they all develop, occurring in simple forms in simple situations and occurring in more complex forms in more complex situations, thus contributing to the development of emergence from simple to complex forms.

Ah, yes, there is also structural-logic. Structural-logic in the manner in which the existence and intrinsic qualities of that which goes before determine the existence and intrinsic qualities of that which follows. Structural-logic determines the nature of all cases of emergence.

There are a couple other points I should mention:

(a)       Emergent products—qualities, objects, processes—are predictable if there is adequate access to the necessary data, the relevant “components” and “factors.” The most significant obstacle to prediction is excessive complexity, great numbers of components and factors and great intricacy of interrelations of components and factors.

(b)       Emergent products can be reduced to the properties of their components and the interrelations between those components because the emergent product ontologically is the components and the interrelations. Achieving understanding of that is the purpose of scientific reduction—science’s divide and conquer methodology. Much of scientific research involves discovering the components and their “factors,” their qualities and interrelations. Once the components, their qualities, and their interrelations are known, it is possible to view simultaneously both levels, the whole and the parts, and thereby understand how and why the emergent product exists. The result is reproducible, reliable, scientific knowledge.

Vincent

Luca : I like the Chalmer's position too, but he didn't talk explicitly about emergence in the book I've read (The conscious mind). In my view, some positions were close to the emergence concept, though.

One told me he defends new ideas, but I haven't read about them. Have you a reference in mind?

Thank you for the long reply, Vincent! I will read it later and probably follow up with a couple of questions.

Lucas: I was thinking of the article 'Strong and Weak Emergence' (http://www.consc.net/papers/emergence.pdf), in which he sketches a middleposition that I find partially palatable. But I think I will have to go back a few steps in thinking it through, since my main issue lies with for instance the typical examples and how they are interpreted. Actually I find little flaw with the notions that has been layed out in this discussion thread, only in the perceived implications, such as strong supervenience entailing downward causation and so on. In my mind there is probably no real difference between non-reductive physicalism and emergentism, except when it comes to some prima facia consequences. Now, Kim makes the same charges against both views, as I understand him, and I think Kim's own event ontology and his views on causation and causal explanation are the core issue, and that, once cleared up, emergence becomes a clearer version of typical non-reductive physicalism without any troublesome causal problems.

Linda@ That was great! You actualy touch on all the topics that I am trying to join in my own project (at places my own formulations are embarrasingly close to yours! I just did an assignment, and your intro is almost word for word identical to mine - I didn't use that version, but if I had, it would have raised suspicions that I had plagiarized your chapter I think). Thank you very much, I will take a closer look at it and probably use it for support. It is very useful.

Thanks Luca! I do hope you find the chapter helpful.

(((LAWB)))

In Chalmers’ paper, Strong and Weak Emergence, the first few paragraphs reveal a distraction that has been plaguing the general discussion and understanding of emergence ever since George Henry Lewes coined the term emergent in 1891. Lewes wrote:

“Thus, although each effect is the resultant of its components, the product of its factors, we cannot always trace the steps of the process, so as to see in the product the mode of operation of each factor. In this latter case, I propose to call the effect an emergent.”

Lewes here introduces an anthropocentric aspect to the discussion of emergence.

The “effect is the resultant of its components, the product of its factors” is one of the most significant processes in the universe. It is a creative process, the origin of all newly existing pattern-of-material-organization. It is the origin of the hierarchic organization of material-reality, upon which the different disciplines of science are based, from physics, chemistry, geology, and planetology, from biology and psychology to ecology, and from astronomy to cosmology. Wherever there is matter, the process of emergence is there playing a role in the creation of new pattern-of-material-organization.

The problem is that when Lewes said “we cannot always trace the steps of the process” he was not really talking about emergence. He was talking about humans. His statement is about the state of human ability to identify and understand the process of emergence that results in the existence of the emergent product.

The consequence has been a seemingly endless discussion about whether humans are surprised by an emergent product, or if they feel it is unexpected, or whether humans can predict an upper level emergent product from a lower level, or reduce an upper level to a lower level, or if humans find the emergent product interesting, and on and on. But this is all about certain relations of humans with emergent products, and not really about the intrinsic nature of emergence itself.

When first encountering the terms, weak emergence and strong emergence, it is reasonable to think that they refer to two different types of emergence, and that Chalmers’ paper is about emergence. But when Chalmers defines these terms he is not really talking about emergence. He is talking about concepts and notions people have about how they feel about or relate to certain emergent products.

“We can say that a high-level phenomenon is strongly emergent with respect to a low-level domain when the high-level phenomenon arises from the low-level domain, but truths concerning that phenomenon are not deducible even in principle from truths in the low-level domain.”

“We can say that a high-level phenomenon is weakly emergent with respect to a low-level domain when the high-level phenomenon arises from the low-level domain, but truths concerning that phenomenon are unexpected given the principles governing the low-level domain.”

Chalmers’ statement, “the high-level phenomenon arises from the low-level domain,” is true. It is a vague statement about combinatorial-enhancement-based-emergence, which is one of several foundational forms of the process of emergence. (e.g. 1. sequential-enhancement-based-emergence, 2.emergence due to the roles of transformation-points, 3. cause-based emergence, or 4. coherence-based-emergence.) Combinatorial-enhancement-based-emergence is the form of emergence process that creates the levels, that creates the hierarchic-organization of material-reality from elementary particles to galactic clusters. The enhancement with this form of emergence is the coming into existence, the emergence, of the higher-level.

The other parts of these definitions, the “not deducible” and “unexpected” parts, are not about emergence itself. So why is Chalmers writing about them? He has a special purpose for the notion of strong emergence. He wants to use it in support of his idea that consciousness is not deducible from the lower-domains from which it arises.

As used here by Chalmers, these definitions need not be well grounded in reality. As he says, “Strong emergence, if it exists,..” He is not at all sure that there is any such thing as strong emergence. But he wants there to be strong emergence.

“We have seen that strong emergence, if it exists, has radical consequences. The question that immediately arises, then, is: are there strongly emergent phenomena?”

“My own view is that the answer to this question is yes. I think there is exactly one clear case of a strongly emergent phenomenon, and that is the phenomenon of consciousness.”

He repeats, “if it exists,” showing that he does not know if strong emergence is real. He then asks if there are “strongly emergent phenomena,” and says he thinks there is one, just one—which turns out to be his idea that consciousness is not deducible. To use a definition of something, the existence of which in his own mind is questionable, as a major support for an equally questionable notion is grasping at a straw.

The notion that consciousness is an example of strong emergence is both superfluous and false. The notion that consciousness, as a high-level phenomenon, is not deducible, even in principle, from the low-level domain from which it arises is unrealistic.

Chalmers states clearly that high-level phenomena arise from their lower-level domains, and he includes consciousness in this. “That is, consciousness still supervenes on the physical domain.”  High-level phenomena arise from their lower-level domains by way of the process of emergence. The way to deduce the higher-level from the lower-levels is to follow the path that process takes and observe and figure out how the one level arises from the lower levels.

In many situations there are factors that make it extremely difficult, or currently impossible, to follow the path of the process—such factors as high quantitative-complexity or high qualitative complexity. Nonetheless, the process of emergence does occur, and in principle could be followed if there was adequate access to the relevant data.

To understand the emergence relations of consciousness, it is not necessary to claim that consciousness is not deducible from the neurological levels. This claim is a superfluous notion.

Since the higher-level factor, consciousness, arises from lower-level factors by way of the processes of emergence, and is thereby deducible, at least in principle, from lower-level factors by developing an understanding of that process, the notion that consciousness is not deducible from lower-level factors is false.

No one knows how consciousness arises from the neurological level. Without that knowledge, it is not possible to know whether or not consciousness is deducible from the lower levels from which it arises. To say that it is not deducible is an unfounded speculation, an argument from ignorance.

The problem with the notion of weak emergence is that it is both superfluous, and, when applied to understanding emergence, it makes a peculiar anthropomorphic distinction between what is emergent and what is not emergent.

The notion of weak emergence is superfluous because it is extraneous. That an emergent product is unexpected is something that occurs within a human mind. ‘Unexpectedness’ is not intrinsic to those processes of emergence or to those events of emergence which occur outside the mind, throughout the universe. This notion is not needed for the description, understanding, or definition of emergence because it has nothing to do with the nature of emergence.

That the notion of weak emergence is a peculiar anthropomorphism is because it defines something as emergent at one time, when the observer finds it unexpected, and then defines it as not emergent a few minutes later when the observer learns more about how the upper level arises from the lower domain and the observer no longer considers it unexpected.

Chalmers writes, “I think that understanding weak emergence is vital for understanding all sorts of phenomena in nature, and in particular for understanding biological, cognitive, and social phenomena, . .” Then, referring to the notions he has been discussing, he ends the paper with, “. . . we can hope that they can play some role in understanding the phenomena to which the term [emergence] has been applied.”

The problem here is that because the notions of weak and strong emergence do not have anything to do with the intrinsic nature of the process or event of emergence, these notions and Chalmers’ discussion do not contribute to understanding the phenomena to which the term emergence is correctly, realistically, applied, or to understanding anything in the various disciplines of the sciences or the humanities.

Thanks Vincent! I have only skimmed your answer, but you point towards some relevant things. I will reply later, since you touch on some issues that I have (without really reflecting on it) assumed was obviously just "anthropocentric gloss", and automatically reinterpreted it in ontological terms. For instance the unpredictability-criterion. As you point out, emergence is supposed to pick out an ontological relation, not an epistemic one. I think most people would (or should) agree. You also note that, if so, then it must be in principle predictable/deducible. I will try to sketch an ontological reading, and why I find the "ontological concept of emergence" problematic. I think I agree with you, but perhaps not, since I think that kind of considerations suggest that "emergence" is a confused concept, and that you end up either with hardcore substance-pluralism OR a breed of materialism. But I will try to provide an argument later, and an indication of what I think is the correct story (still very sketchy, but it seems to cohere well with your discussion). I am in no way convinced of my own views, though, so if you see anything suspicious feel free to let me know.