Week 3 (emergence)

Paul Grobstein's picture

Some interesting discussions last week. And some interesting/productive confusions, with thoughts from other courses intersecting this one. Emergence? Let's though try and keep things a little orderly around here. This is the forum for Emergence; those posting for other courses should ... notice and be sure postings go in intended places.

And let's try and post by the end of the weekend each week? So we can discuss what people have been thinking on Mondays.

Thoughts about your experiences with the Game of Life and Langton's Ant? About modelling, and being modelers yourselves? And/or anything else on your mind this week.

asmoser's picture

Form, Function and Object-Oriented Living

I think Rob’s comments about the difference between purpose and reflex are interesting, but I think he misses a vital piece. E. Coli or any other bacterium placed on a suitable medium will replicate and grow without further stimuli. It is the function they are designed to fulfill. However, they do this because they need to, because this function is existence in its entirety for a bacterium. Calling it reflex seems to me to credit the life form with a level of complexity it does not have. The E. coli don’t have a large range of possible responses to stimuli, they either have the materials necessary to reproduce or they do not and either live or die in response to these conditions. Knowing very little about cellular biology, I can’t comment whether in the absence of food the e. coli search for it, but certainly there are bacteria that do. So we have a simple set of instructions that each bacteria follows. Is that purpose? Maybe not, but it doesn’t strike me as reflex. In fact, these concepts seem difficult to apply to something so simple. Bacteria does not have purpose or reflexes. Its form is its function, built to survive and reproduce.

 

I spent a decent amount of time thinking about the connection between form and function in living organisms and it struck me that life has some interesting parallels with classes in object oriented programming languages. Each successively more complex class inherits the functions of the classes it is based on. In life, the most basic bacteria could be said to have a single, combined function based on asexual reproduction through division. We can write pseudocode to reflect this single combined function:

(Imagine each bacteria is an organism, o, of a class that has a size attribute and a location attribute)

BacteriaLife(size, location){

If (size == 8)

o.divide(); //if the organism (o) is of an arbitrary size, it will divide into 2

elseif (current location == food!)

size = size+1;

eat(o.currentlocation) //this should make the current location no longer //food

BacteriaLife(o.size, o.location+1);

}

 

When this simple function begins to require sexual reproduction (first in gamete clouds, then in mating) the functions required of an organism become increasingly more complex, and with the need to prioritize between functions, relatively simple behaviors should give rise to increasingly complicated patterns of behavior. I’d like to write more on this, but class theoretically starts in four minutes, so I’ll try to post again later.

 

samkaplan's picture

A Response

I don't see how this is any different from the way humanity operates. We have one function: to procreate--just like the bacteria. Whether you want to call it a purpose or a reflex just depends on how much credit you give the human race.

Lauren's picture

What Langston's Ant Tells Us About Science

I believe a duality exists in science, where its true nature may be rendered either quantitative or qualitative depending on the perspective from which it is viewed. Convention states that for something to be considered science, it must be quantitative with measurements and equations that will, in theory, demystify the scenarios and questions that surface in any scientific field and explain phenomena like Langston's Ant. In reality, if every issue could be reduced to a simple calculation, most of Man’s enduring questions about himself and the world around him would have, in all likelihood, been put to rest in the last few centuries of unprecedented inquiry and "progress". Therefore, science must exist in a more qualitative nature and should be approached instead by studying the logical contexts of these computations--through a context like emergence.

In many ways, the scientific drive to gain and possess knowledge lies at the heart of human nature. Man seeks to exert dominance and to demonstrate mastery of the world around him, and it is in this drive for physical and psychological control that the fundamental “problem” occurs. Mark Twain once said: "There is something fascinating about science. One gets such wholesome returns of conjecture out of such a trifling investment of fact". As this statement insinuates, scientists frequently spend a short amount of time experimenting and calculating before arriving at a conclusion, which is often incomplete but accepted as universal fact. If one studies the progression of leading scientific theory over the centuries, it becomes painfully evident that one is only “true” as long as it has not been disproved by another. Regardless if one is talking about the makeup of the solar system or the anatomy of a body, humans generate volumes of conjecture based on incomplete evidence and guesswork in an attempt to fully understand the dynamics of a system. To the human mind, this understanding represents knowledge. In turn, knowledge is viewed as a kind of ownership and mastery over information; this ownership only further imparts to Man a sense of control that may not be entirely stable in its foundation. The only thing that humans can be sure of is that they are only just beginning to understand how and why things work. Langston's Ant is proof of this. In fact, there may be many more questions posed before Man's ability to fully and thoughtfully answer them catches up. Overall, there continues to be new discoveries because there still remains that much more that Man does not even realize he does not know.

 

Paul Grobstein's picture

the nature of science

see http://serendip.brynmawr.edu/sci_cult/stanford

samkaplan's picture

Science And Langston's Ant

I can see what you're saying about humanity's relationship with science. However, I don't see Langston's Ant as a good example of that relationship.

It's not like there's some kind of meaning hidden deep within the Langston's Ant code that we simply can't understand at this time. By making that assertion, you cross the line from scientific inquiry to, say, "Gattaca"- or "Pi"-style pseudoscience.

Langston's Ant is much more a demonstration of degrees of complexity and computation than an example of our limited understanding of the world around us.

rob's picture

What I'm trying to say...

In class on Wendesday we started talking about the concept of purpose in a system, so i did some thinking about the meaning of purpose. to do so i came up with an example situation and thought about it.

Here it is:
John felt bored and wanted to get high, so he bought a bag of pot, rolled a joint and smoked it.

Let’s break this down:
Subject: John
Present / Current State: bored
Purpose / Desired State: stoned
Actions: buying a bag of pot, rolling a joint, smoking it

Several things become evident in this breakdown:

1. In order to have a purpose, there must be a subject (john) with consciousness who is having the purpose. If you hit me in the knee with one of those little rubber hammers, my leg will fly up and smack you upside the face, but that doesn’t mean that my leg’s “purpose” is to spazz out. I'd say that without consciousness, an agent can have “reflexes” but not “purposes.” If you put an E Coli colony on a rotten T-Bone steak, the bacteria will replicate, and the population will grow, but it’s not a conscious action, so I’d be reluctant to say that the E Coli’s “purpose” is to replicate. I’m more inclined to call the bacteria’s growing a reflex to the stimulus of rotting animal cells.

2. So why is consciousness essential to purpose? I’d argue that purpose is essentially about having a mental image of things being different than they are and designing a series of actions to achieve that end state. In the original example, John’s purpose in doing the actions that he does is based on a sense of how he’d like things to be NOT on his sense of how things already are. If he couldn’t deal with the concept of things being different, he couldn’t possibly construct a plan for a series of actions to change things from their present state. If the actions that the agent takes to accomplish the purpose lead to an unforeseen result, the actions were still “on purpose” (taken to achieve the imagined state) but the actual end state is “accidental / unintentional” (something other than the imagined state).

3. Because the intrinsic value of the endstate gives meaning to the series of actions, the actions themselves need only be of instrumental value. Buying a bag of dope is nothing in of itself: if you don’t smoke it, it’s just a waste of cash, but in order for getting high to be a purpose in all of this, it must be something John wants in of itself. This then constructs a heirarchy of purpose because while John might peel carrots to make dinner or pack a bowl to get high; the meaning in those actions come from the fact that he wants to make dinner to fuel his body and wants to get high to have fun.

But this too is problematic, because why does he want to survive and enjoy himself? I don’t feel like fleshing this out now, but I would argue that all actions (verbs) are of instrumental value, and the only things that are of intrinsic value are states (adjectives): John may peel carrots, to make dinner, to fuel his body, but ultimately it is his desire TO BE alive that drives the rest of it. While this difference can be semantic, it feels right to me. I’m not sure quite what it means yet, but if it’s true then it means that if we could automatically be healthy and happy, there wouldn’t be any value in doing anything. If that’s true, then there’s something in human nature that ultimately wants things to be static and, fundamentally, we’re all lazy and obsessed with death, which probably we are.

Jessica B's picture

Perspective and assumptions

While I agree in general, I think you're forgetting that the scenario in class involved taking an outside perspective.

We know (hopefully) that we have purpose because we can see it in ourselves. I eat because I'm hungry. I do my homework because I want a good grade in the class and, in general, because I want to graduate.

However, how do we know that other people also have purpose to their actions? Ultimately, we don't. However, we can assume that they do, based on their actions. Sarah is in college, so I can assume she wants to graduate college. She is going her homework which, from my experience, helps your grade. Getting good grades means you can graduate. Therefore, I can assume that Sarah does her homework for the same reason I do. She has purpose.

There is slightly less assumption looking at another person, because we can relatively safely assume that since people are ultimately the same, they must all act with purpose. It's a little trickier when we look at something that is not human. I take the dog's food away so he snaps at me. Did he do this for a reason or is it just a reflexive reaction to having his food taken away? Scientific neutrality aside, most people would opt for the former explanation. If somebody took my food away while I was eating it, I would say something sharp to them or grab at them. My purpose in doing this is to get my food back (and to express my displeasure in having it taken away in the first place). I would assume that the dog had generally the same purpose.

That's the point of the ant, I think. An alien watching this thing moving around can't know if it has a purpose or not. But assuming an intelligent alien, when it moves in a certain pattern, it's doing it for a reason. So it would very likely assume that the ant also has a purpose. Obviously, since this thing is a program, we know there is know true purpose, but it appears to have one. I think that's the point.

samkaplan's picture

Purpose And Consciousness

I think I agree with you that consciousness is required for purpose—let's just call it free will, since that's what we're really talking about.

In which case, though I agree that the one requires the other, I would first like you to prove to me that we are conscious and that the decisions we make are conscious decisions.

biophile's picture

Patterns and Consciousness

This is a bit random, but I was working at the Franklin Institute today and one of the exhibits reminded me of what we've been talking about very strongly. Basically, this particular exhibit is about the intersection of physics and art. In one part, there were many pendulums set up containing sand that slowly leaked out so that they made patterns as they moved. Each stand was different, but each had only one pendulum (for example, one had a conveyor belt under the pendulum that complicated the pattern that the sand made.)

There was also a strange structure next to them that was composed of three cogs that one could turn individually or in different combinations; as each cog was turned, a luminescent tracing was left by a needle. The bottom line is that these showed how "complicated things can get when you combine two or more simple motions." It's mind-blowing to think that this concept can also explain the evolution of consciousness.

On a slight tangent, I was reading about Bergson today and thought about emergence yet again. It may not seem terribly related, but the author was talking about how the philosopher Bergson disagreed with Darwin's theory of evolution in that it did not explain the evolution of instinct or of consciousness. They can't be the inherited accumulation of acquired characteristic habits; it's generally agreed that that's not how evolution works. So how did these higher functions come to be? Maybe both instinct and reason emerged fully formed, but that makes little sense. Perhaps it makes more sense to think of instinct and reason not as discrete, heritable characteristics but as emergent properties which are intrinsic to the human brain and are acted upon by internal interactions and external forces throughout life. Instinct and reason didn't emerge spontaneously; perhaps they were inevitable byproducts given the evolving structure of the brain.

And while I'm on a roll with the general incoherence, does anyone think that the emergence of life is inevitable given reasonable conditions? It seems as if life exists everywhere on earth: beneath the earth, in deep sea vents, in the Arctic... For years scientists thought that life could only exist within very narrow parameters; however, now it's known that microorganisms exist everywhere on our planet. It seems as if life is destined to exist, as if chemical components are innately compelled to interact and form more complex structures. It just never seems to stop.

shikha's picture

I dont think that the

I dont think that the emergence of life is inevitable given reasonable circumstances. i think the outcomes of processes that have resulted in life, which we think are so many, arn't really that many when one thinks of all the possible processes that may have been or are going on. we can't really see the outcomes of processes that do not lead to the emergence of life, so maybe there isn't really a way of knowing?

on another topic, i was wondering about an interesting idea that was brought up in class, on how religion fits into emergence. As most religions would like us to believe, life was planned by its architect, God. However this does not fit in with the definition of emergence. If we are a part of an emerging system, then we cannot have this intended planning. I was interested in knowing how religion has responded to this, if at all, but haven't really been able to find anything. any ideas?

mgupta's picture

Langton's Ant

So this with regard to the excitement while trying to make Langton's Ant do something "surprising" and "unusual".

I was able to make the turtles create symmetry as they moved. I noticed that only an even number of turtles can lead to symmetry, and that is quite understandable. However, the surprising thing is that each even number would not lead to symmetry if moved a certain number of steps - for example, 2 turtles would not make symmetry if they were moved 8 steps forward, 10 turtles wouldnt work if moved 2 steps and so on and so forth. Another thing to notice was that 6 turtles never made symmetry, no matter how many number of steps and 8 turtles always made symmetry no matter how many number of steps. I was trying to look for a pattern in this pattern of symmetry, but couldnt find one.

Any thoughts?

Frank's picture

Langton's Ant and Complexity

I found that as I increased the complexity of the code for Langton's ant, the effects it produced were generally less interesting than those produced by the original, simpler code. More often that not, it would instantly start in a pattern which would degrade into "randomness." But how could the inclusion of interacting agents affect things? I've been plagued by this idea of free will in an emergent system, and one idea is to have interacting agents on the computer reproducing and making basic "decisions." Is it plausible to think that interacting agents would be more likely to create interesting effects with each other than the ones these same agents make on the world around them?

Paul Grobstein's picture

thursday class

Thanks all for interesting/rich conversation. I've added some notes about things it made me think about at the end of http://serendip.brynmawr.edu/complexity/course/emergence07/7feb07.html. Delighted to have corrections, additions, further thoughts here.

samkaplan's picture

On The Subject Of Emergence And Language...

Here's a cool website that has at least something to do with emergence:

www.languageisavirus.com

The quotation "Language is a virus..." comes from William S. Burroughs, whose cut-up techniques could almost be seen as examples of emergence. (Meaning arises from seemingly random arrangements of sentences and phrases that were rearranged according to simple rules.)

The site is also associated with the writer Jeff Noon, who writes computer science-informed science fiction that deals with a lot issues we've talked about in Emergence. (In one of his books, "Nymphomation," hackers are trying to figure out the secrets behind a national lottery based on the arrangements simple dot patterns of dominoes.)

Jeff Noon is also just really awesome. (en.wikipedia.org/wiki/Jeff_Noon)

samkaplan's picture

Emergent Music And Douglas Adams

I've been listening to a lot of 20th century classical music lately, which all of a sudden gave me this idea: can emergent systems be used to compose—or, less controversially, produce—music? (People like John Cage have in the past exploited randomness and chance to "compose" music.)

Obviously, the answer to this question is yes, supposing that we define music simply as organized sound. But what if, for example, you rigged up Langston's Ant so that every step the ant made was linked to a specific note?—or something like that

This idea also reminded me of Douglast Adams' book "Dirk Gently's Holistic Detective Agency." More specifically, one of the characters, Gordon Way, gets rich by writing a program that produces music based on a company's financial statements, which is a pretty wild idea.

On the subject of Douglas Adams, it would seem that the Dirk Gently detective novels have much in common with Emergence. For example, Dirk Gently solves his cases based on his belief in "the fundamental interconnectedness of everything."

In conclusion, I will endeavor to somehow modify Langston's ant to produce music—if this is even possible. I think it will be very interesting to hear the exact point when pattern arises out of chaos.

natsu's picture

patterns in music?

Don't you think that whether a certain peice of music sounds like it has a pattern or not depends on the person that is listening to it, at least to some degree?I'm just interested about what you define as chaos or pattern in music...

samkaplan's picture

Patterns In Music

Well, I mean, objectively speaking, something can have a pattern or something can not have a pattern. Certainly, different people "enjoy" or "understand" different kinds of music, but that doesn't mean that I, as, for example, someone who doesn't like Akon can't still tell that his music is obviously not pure chaos.

natsu's picture

Emerging Creativity?

For those of you who are like me and cannot help looking up information on the authors that have written the books you read, I would strongly recommend that you have a look at a clip you can find here:

 

http://www.ted.com/tedtalks/tedtalksplayer.cfm?key=m_gladwell

 

Gladwell is the author of two books listed on the suggested reading list for this course. Although I am unfortunately still unsure about how exactly his ideas apply to the concept of emergence, I found his ideas about the society's opinions very interesting.

 

I don't know how familiar everyone is with the TED Talk clips, but many of them are actually very interesting. One that I found quite relevant to our discussions on emergence is a very new talk by Charles Leadbeater and can be found here:

 

http://www.ted.com/tedtalks/tedtalksplayer.cfm?key=c_leadbeater

 

Although the clip itself is not exactly as entertaining as the others (personal opinion) his ideas that creative ideas can emerge from the society and that you don't need an organization to innovate made me think. It actually made me think of all the moments when we were working with Net Logo in the classroom, and somebody in the room finds a new fascinating model. The patterns that people see are quite simple (both the set of rules and the actually patterns) but something about the beauty in the simplicity makes us feel like we are seeing something so interesting, and maybe even complex, that we can't help saying "Wow, that is so cool...!"

samkaplan's picture

Malcolm Gladwell

Yeah, Gladwell's great. He seems to overlap with this course quite frequently in that most of his work combines pretty distinct disciplines and subjects in exciting ways that are usually pretty counterintuitive.

On the subject of emerging creativity (kind of), here's a link to a recent article Gladwell wrote for The New Yorker about using computer programs to predict whether a movie will be commercially succesful, and using that information to produce more commercially viable movies:

http://www.newyorker.com/fact/content/articles/061016fa_fact6

randomness