Emergence

[chilehed]

That's not my understanding. To use another example, the property of wetness is an emergent property of water molecules. A single water molecule does not feel wet. However, a whole group of water molecules does produce the experience of wetness. Even more, we can explain and predict how a large group of water molecules will behave from the simple characteristics of a single water molecule.

To be wetted means that water has adhered to an object, due to the polarity of each individual water molecule. A single molecule might not feel subjectively wet but the object to which it clings has nonetheless been wetted by it.

 

[Derek Meyer]

Emergence is the idea that the behavior of a complex system cannot be predicted from the behavior of its parts.

No, it isn't. Emergence is the opposite.

Where do you get your wacky ideas from? YEC's?

 

[expos4ever]

No, it isn't. Emergence is the opposite.

Where do you get your wacky ideas from? YEC's?

I believe the definition provided in the OP is indeed put forward often as the definition of emergence. Why do you dispute the definition?

 

[durangodawood]

Every time I encounter the word emergence it's a way of saying nothing. Like when materialists say free will could be an "emergent phenomenon". It's saying "something emerged", but tells you nothing else.

The idea is that the behavior of the new/complex system follows rules that are different from (and not predictable from) those that govern its individual parts.

I view life itself as an emergent phenomenon, with new behaviors, capacities, possibilities, not predictable from the behavior of the plain chemistry that preceded it.

 

[Resha Caner]

I read Bedau in more depth today. In the end I have to agree with Chesterton's assessment. I think all he did was define "weak emergence" in such a way that it obviously follows the emergent properties are derived from properties of the parts.

As such, I guess I'm only interested in strong emergence. At least I did learn that the key concept currently attributed to strong emergence is downward causation, and that this has been established as logically possible (even if it feels like magic to Bedau).

I don't know that downward causation is the only possible form of strong emergence, but I don't have anything better to offer at the moment. All I can say is that it feels possible a system could exhibit properties that cannot be logically predicted from the properties of the parts. A non-existent argument, I know.

I'm not sure how much farther the thread can go unless someone has something to offer on downward causation or other strong ideas. My plan is to read O'Connor and some of the other stuff Bedau cites.

 

[expos4ever]

I don't know that downward causation is the only possible form of strong emergence, but I don't have anything better to offer at the moment. All I can say is that it feels possible a system could exhibit properties that cannot be logically predicted from the properties of the parts. A non-existent argument, I know.

For what it's worth, I suggest that we need to be aware of a very deep default inclination to simply assume "full reductionism" must be true - after all, how could it not be the case that the properties of a system can be completely explained in terms of the properties of its constituents?

Nature, however, is what it is and most certainly does not "care" about such assumptions on our part. Yes - the "reductionist" model has indeed worked well with respect to many things. But, of course, that is no reason to assume it works for everything.

 

[Loudmouth]

For what it's worth, I suggest that we need to be aware of a very deep default inclination to simply assume "full reductionism" must be true - after all, how could it not be the case that the properties of a system can be completely explained in terms of the properties of its constituents?

While true, we also have to be aware that we don't have to prove a universal negative in order to evidence the examples where reductionism has worked.

At one time, there was a battle between the ideas that nature worked through the supernatural and that nature worked through simple mechanisms that weren't supernatural. As time went on, we found millions and millions of natural explanations for what we saw in nature. To this point, we haven't found a single verified supernatural cause for anything in nature. This is why the first choice is to look for natural causes. We all know in the back of our minds that perhaps this time we will find that supernatural cause, but I don't those who work on questions of emergence are expecting to find a supernatural cause.

To use an analogy, no serious ecologist and forest biologists carry Bigfoot repellant with them, just in case they run into one.

 

[Resha Caner]

To use an analogy, no serious ecologist and forest biologists carry Bigfoot repellant with them, just in case they run into one.

That's a bad move if you're in Canada during mosquito season.

 

[Loudmouth]

That's a bad move if you're in Canada during mosquito season.

I didn't say anything about mosquito repellant.

 

[chilehed]

I didn't say anything about mosquito repellant.

Ever seen a Canada mosquito?

 

[Gene2memE]

Cellular automata provide a very good example of 'emergence' - very, very simple rules end up producing very complicated outcomes, including functional systems.

 

[timewerx]

Emergence is the idea that the behavior of a complex system cannot be predicted from the behavior of its parts.

Thoughts? Is this a real phenomena or just an artifact of our inability (at the present time) to accurately predict the behavior of the parts?

I don't believe it because I could predict something complex from the behavior of its parts.

We have the ability to discern. To see the big picture even if it is not there. To believe that we can't is the lie of the evil forces in our society.

 

[Resha Caner]

I just read an interesting essay on Connectionism at the Stanford Encyclopedia, which basically regards the conceptual issues surrounding neural networks. I didn't realize all of the challenges fuzzy logic is presenting to traditional computing.

So, traditional computing is based on symbolic logic, and through the middle of the 20th century it was common to compare the mind to a computer ... in fact, the book I started reading on human memory for my summer class did just that. The idea is to represent a problem as a string of symbols and establish rules for manipulating those symbols.

However, that's not how neural nets (nor the brain) work. Instead, they create distributed and overlapping patterns that are referred to as "sub-symbolic". The operation of the individual neurons does not predict what the pattern will be, and therefore the emergent properties of the net don't "supervene" on the neurons ... i.e. the pattern can't be predicted from properties of the neurons.

However, I don't think that's strong emergence. It seems just a game of defining the system. If one includes the external influences as part of the "system", it's probably possible to predict the result.

Still, it introduces some very interesting concepts related to emergence. I think that if the training sets are fed to the net in a different order ... or if different subsets of training are used, the way the pattern is established in the net will be different ... even though the meta-result (the represented symbol) is essentially the same. Given this meta characteristic, the manner in which these sub-symbolic patterns are interacting feels as if it would have the potential to produce unpredictable results.

For example, one of the strengths of nets is their ability to deal with new situations. Traditional symbolic programming can only accomplish the tasks that are programmed. However, a net will respond with a "result" no matter the input. It will respond to situations it was not trained to handle. How the net responds, however, seems to have the potential to be unpredictable.

Maybe.

 

[Loudmouth]

However, that's not how neural nets (nor the brain) work. Instead, they create distributed and overlapping patterns that are referred to as "sub-symbolic". The operation of the individual neurons does not predict what the pattern will be, and therefore the emergent properties of the net don't "supervene" on the neurons ... i.e. the pattern can't be predicted from properties of the neurons.

We can't predict the pattern with modern technology, but is it only a problem of measurement? If we knew the exact state of all neurons at a certain point, and had all of the interactions worked out ahead of time, could we predict a pattern with a given stimuli? I suspect we could.

The real difficulty is mapping all of the neurons and knowing their current state. The amount of stimulus needed to make a neuron fire is dependent on many factors, and can change from hour to hour or even minute to minute. From what I can remember reading on the subject, serotonin levels and neurotransmitter receptor levels have a large influence on what happens. Anti-depressant drugs often work through changing serotonin levels, as an example of how the system can be manipulated.

However, at some point, we may run into observer/measurement problems. In order to accurate measure the state of neurons we may end up causing significant changes that influence the outcome. Perhaps this is another property of emergence in itself.

 

[durangodawood]

I don't believe it because I could predict something complex from the behavior of its parts.....

???

Emergence is about the cases where you cannot predict the behavior of a complex system based on the way its parts behave.

Are you saying you can always do this?

 

[timewerx]

???

Emergence is about the cases where you cannot predict the behavior of a complex system based on the way its parts behave.

Are you saying you can always do this?

You mean if I can do the opposite?

You can predict the overall personality of a person for example based on how they do certain things. Like how they drive for example. Whether their feet smells or not, what would they do to get what they want, if they make a lot of noise or not...

And then you can predict the "health" of a nation based on how their citizens behave and think on average.

You can also apply the same principles in engineering design.

My point is that nothing is without purpose (whether for evil or good) and everything is correlated. Like you'll find in places that are quite filthy, you can expect the same area to have high crime rates. If citizens on average have little concern for each other, and exploit one another, the nation is also in disarray and corruption is rampant.

It's rather useful to know these things. It will keep you out of trouble.

 

[Eudaimonist]

Emergence is the idea that the behavior of a complex system cannot be predicted from the behavior of its parts.

Emergence is the idea that the behavior of a complex system cannot be predicted simply from the behavior of its parts.

In other words, an understanding of the behavior of the parts considered in isolation doesn't reveal the behaviors when you see the system in operation. There are potential surprising behaviors that one might not have anticipated. However, I don't think that emergence insists that predictions are impossible.


eudaimonia,

Mark

 

[KCfromNC]

Emergence is the idea that the behavior of a complex system cannot be predicted from the behavior of its parts.

Thoughts?

Yep. I think you have the definition wrong. There's no requirement that an emergent behavior be unpredictable, just that it is different from the behavior of the individual parts.

Edit - oh well, should have read the other replies before responding. At least I'm not alone in seeing the problem.

 

[KCfromNC]

But does it explain how you get wetness from a whole group and not the single molecule

It explains at least some parts of it. Surface tension, for example.

 

[Chris B]

Cellular automata provide a very good example of 'emergence' - very, very simple rules end up producing very complicated outcomes, including functional systems.

That was the example that occurred to me, having tinkered in the past with Conway's game of life.

The initial rules are simple and do not suggest gliders, glider guns, organism arrangements to produce a series of glider guns...