yaneer bar-yam

yaneer bar yam

first intro here (i think):

while reading (ch3 p 61 of Dave Elder-Vass‘ The Causal Power of Social Structure).. this tweet/reading.. fitting…


hierarchies have diminishing usefulness as complexity increases jarche.com/2013/08/the-so… pic.twitter.com/TtsHPA7s0k

textbook social systems that are engraved in stone”can be changed in one single generation. There may be hope for the human race, it seems.

Recent research shows that evolution is on the side of those who cooperate.

“We found evolution will punish you if you’re selfish and mean. For a short time and against a specific set of opponents, some selfish organisms may come out ahead. But selfishness isn’t evolutionarily sustainable.”


Hierarchical control structures are symptomatic of collective behavior that is no more complex than one individual.


As Yaneer Bar-Yam explains in Complexity Rising, hierarchies have diminishing usefulness as complexity increases.


Better social relationships (non-hierarchical and not based on the dominance of others) can make for healthier populations. In addition, networks are the only way our collective intelligence can be used to address increasing complexity

then this from feldercarb:

You need to follow @yaneerbaryam if you’re not already. He knows a thing or two about systems theory and problems of scale.


2015 – with Taleb on Peace for Syria




8 min video of highlights from above:


6 min – how to create a structure which is going to be robust and be a natural solution..

let’s try this: a nother way

8 min – my theory is.. we’ve got to move on.. the existing structure is not working..








6 min – there’s a lot of data in the world.. we don’t want to just look at how one thing correlates to another..

we really want to understand how all things work across all the info

8 min – really about understanding cascades, crisis, leverage points.. a complex dependent system..by understanding the dependencies.. we can apply .. patterns..

11 min – need to understand financials..

? deep enough..?

secret info.. don’t reveal till published.. this is privileged info

13 min – in order to implement policy.. have to get country needs.. can’t do policy till do the analysis..  this is just how econ flows work.. if money accumulates somewhere the econ doesn’t work..

is this complex..? or irrelevant..

made up money.. is the complexity we’re researching for a better world..?

15 min – we have to understand this structure (diagram w 3 parts: labor, capital, firms)



new trade opp – 2013





29 min – (this was on getting things into the market) a piece that is crucial.. recognizing that things work together and not separate.. core to thinking in complex systems sense.. guides understanding to what’s happening now.. what you want to be present and how to transition..

let’s transition like this.. short bp

44 min – i think a lot of decision now are being made locally.. i think the main thing is to show the effectiveness of global decision making

rather both.. at once.. ie: redefine decision making

using tools and science.. the fact that we can come up with better decisions and policies.. that will actually improve things ultimately will drive the fact that people will listen to those decisions..


not about creating global decisions.. if we can show decisions being made.. like here.. are effective.. i think people will listen..


public consensus always oppresses someone.. we have the means to a nother … much deeper way.. ie: mech to listen to all the voices.. 24/7..

let’s go there.. for (blank)’s sake

i think we’re looking at.. obsessing with too much data.. that doesn’t even include the data that could save us.. ie: self-talk as data

56 min – time scale.. how long we should expect something to take.. so have to understand from an expectations pov.. and gear toward that time plan.. (taking small holder farming)


he starts at 2:22 here (2014):




2:23 -we are no longer individuals acting on stage.. it’s us collectively.. ie: wikipedia



basis of convo has to be the matching..

org’s have to match the complexity of the challenge

ginorm small ness

very important statement.. ..in order to be successful an org structure has to match the structure of the challenge in scale and complexity

so what’s the task.. 2 diff parts.. one complex (content) and one large scale (structure/format).. the format that repeats itself across all scales.. uniform across the system..

so – 2 convos.. 2 needs..

2:24 – content is complex is quite clear..  but wikipedia has demo’s that even individual articles benefit from multiple contributing authors.. so complexity already exists at individual article.. and the format is large scale.. because it is uniform by its definition

2:25 – large scale required a specific time to be finished and therefore required and benefitted from central control… the internet set aside that second piece.. so more complexity possible.. but also meant challenges for large scale in terms of creating uniformity/conformity

he keeps saying.. is that clear.. no response.. so he says.. great..

2:26 – so leads to tug of war between decentralization (for purposes of complexity) and centralization (for purposes of conformity)..

systems that are not controlled are… evolutionary

evolution in biology creates these incredibly complex amazing systems.. we can’t do this..

evolution is always a work in progress

2:28 – major.. most successful systems today.. almost all distributed thru method of evolutionary dynamics.. ie: markets; visa/mastercard (account for more than 10 trillion dollars of transactions global.. why you’re not aware as evolutionary system because process takes place at level of the card issuers.. nice book about this: birth of chaortic age by dee hock.. founder of visa international); internet; open source; wikipedia; app communities

2:29 – how these systems are structured.. *centralized.. framework consistent across system that enables.. individual actions which are distributed.. the combine to collective behaviors thru the mechs we often associate with evolution…persistence with variation.. selection with **competition

*with you there.. we just need to disengage from tons of stuff to get to a deep/simple/open enough framework for all of us..


contradictory things exist in this system.. and the fact that they are paradoxically coexistent means that there are some choices about balance/trade-offs that one has to understand..

2:30 – in order to think about this.. have to map wikipedia in to evolutionary process.. not an institution.. an evolution of pages..

the standard evolutionary dynamics.. set of orgs .. they reproduce.. creating more.. some selection process..  variation.. end up with diff things later than had at beginning..

2:31 – process of editing: edit; edit; edit.. doesn’t look like that.. variation variation variation is not evolution… the key step in the evolution (of wikipedia) is the reading.. and deciding whether to edit.. that’s a selection process.. a selective process..

issue: lack of parallelism.. we compare with prior versions.. and that’s not quite the same.. turns out.. mathematically still puts us in category of evolutionary dynamics.. but we’re concerned with what that should do..

2:32 – want to talk about some of the things evolution needs to accomplish..

2:33 – problem we have with evolutionary dynamics.. is one gets stuck at a local optima.. any time you have complex entities.. they’re going to have local optimas

what we want to do.. is jump to another one which is better

indeed.. for (blank)’s sake

problem is is that we can’t tell how.. we can only take small steps..

let’s try this: model a nother way – 6-12 months.. then 7 bn can leap.. w/in say a year

2:34 – so we have a situation where we take a step.. we’re going in wrong direction.. in terms of quality

this is why we have to leap.. we’ll never sync up..

so if we immediately select.. we will get stuck.. and this is a problem.. in any process that involves incremental changes that are trying to make improvements

so what we would really like to do.. is to take some set of steps.. and end up in the other one (local maxima).. now this is a problem.. because if we look more generally.. we’ll see.. it really looks like this.. (zoom out.. to see we’re really just at another optima.. maximas galore ahead)

sisyphus ness

zoom dance matters.. huge

there are all these local optima that are separated by a deep valleys .. and this is generally the way the world is.. you don’t have the control over this.. sorry..

what if the control over it.. is to let go of control over it.. what if that’s the complexity we’re missing.. trusting us.. all of us..  partial is what gets us out of sync.. clear vision of the zoom dance]

2:35 – so if you don’t have control over this.. how do you solve the problem.. and the answer is .. that we need balance of different processes.. particularly divergence (brainstorming).. and convergence (decision making).. di/con vergence

here’s the point.. it’s not exactly a balance.. because balance suggests that they are kind of happening at the same time.. and it may turn out that that’s not the best way to do it

2:36 – too much divergence.. end up in random places.. in entropy…. too much convergence.. you get stuck.. stasis

two kinds of people.. thick boundaried: everything has a place… thin boundaried: open to exploration.. so need a balance of *diff kinds of people.. and balance of **automation of people..

*how about all the kinds

**facil whimsy via self-talk as data

2:37 – automation works on the large scale part.. the structure.. the stuff that has to be done over and over

2 convos.. 3 and 30.. everyday.. et al.. ground hog ness

can’t automate the complex part

so what does framework look like.. can have central control of framework.. to create evolutionary context.. rules enable rather than restrict

2:38 – rules turn out to be huge between control and evolutionary process

in a controlled system rules are about what to do and what not to do .. in an evolutionary system rules are enabling and selection is the process that creates the improvement

2:39 – rules are white and black.. and selection is fundamentally recursive.. it’s not about where you are .. it’s whether you’re moving.. and that’s really important.. evolution is always a work in progress… always changing the organisms around us..

rules are about integrity of the system and the protection of the participants..

2:40 – rules are often about promoting change rather than preventing change..

evolution is a metaframework.. we can make it as a way to analyze what’s going on .. and that’s an opportunity..

human civilization is a complex collective.. you are all part of it.. wikipedia is a very important part of it today..

3:03 – what is a community.. i think the nature of collective is broader than the definition of .. common interest..

3:30 – i’m quite agnostic about structure of that decision making process… ie: www .. or carefully thought out.. i do know.. we’ve shown .. that very simple ideas can be transformational.. because of how people interact with them..

3:31 – on ai ideas and machines replacing people.. same fear as steam engines replacing people.. i think we’re discovering more and more that people are diff from machines.. and i do think.. that thinking about civilization as mostly about people and secondly about tech is really important..

opp.. is not to think about the code but to think about the people.. if we do that well and whoever does that well.. will create transformations..

hlb that io dance..

continuing opp to transform society..


find/follow Yaneer:

link twitter

Complexity scientist. Exploring the space of possibilities and its boundaries.

Working on solving our societal crises.

Team building for a better world.

yes.. let’s go deep Yaneer – short bit


NECSI has been instrumental in the development of complex systems science and its applications. We study how interactions within a system lead to its behavioral patterns, and how the system interacts with its environment. Our new tools overcome the limitations of classical approximations for the scientific study of complex systems, such as social organizations, biological organisms and ecological communities. NECSI’s unified mathematically-based approach transcends the boundaries of physical, biological and social sciences, as well as engineering, management, and medicine (see Complex Systems Resources).

NECSI research advances fundamental science and its applications to real world problems, including social policy matters. NECSI researchers study networks, agent-based modeling, multiscale analysis and complexity, chaos and predictability, evolution, ecology, biodiversity, altruism, systems biology, cellular response, health care, systems engineering, negotiation, military conflict, ethnic violence, and international development. (see NECSI Research).


wikipedia small

Yaneer Bar-Yam (born 1959) is an American physicist, systems scientist, and founding president of the New England Complex Systems Institute.

Yaneer Bar-Yam was born in Boston, Massachusetts in 1959. He received his B.S. degree in 1978 and his Ph.D. degree in 1984, both in physics from the Massachusetts Institute of Technology. He was a Bantrell PostDoctoral Fellow, and a joint postdoctoral fellow at MIT and IBM. In 1991, after a junior faculty appointment at the Weizmann Institute, he became an Associate Professor of Engineering at Boston University.

He left Boston University in 1997 to become president of the New England Complex Systems Institute. He is also an Associate of the Department of Molecular and Cellular Biology at Harvard University. He is chairman of the International Conference on Complex Systems and managing editor of InterJournal.

Bar-Yam studies the unified properties of complex systems as a systematic strategy for answering basic questions about the world. His research is focused both on formalizing complex systems concepts and relating them to everyday problems. In particular, he studies the relationship between observations at different scales, formal properties of descriptions of systems, the relationship of structure and function, the representation of information as a physical quantity, and quantitative properties of the complexity of real systems. Applications have been to physical, biological and social systems.

Bar-Yam has made further contributions to the theory of the structural and electronic dynamics of materials, the theory of polymer dynamics and protein folding, the theory of neural networks and structure-function relationships, the theory of quantitative multiscale complexity, and the theory of evolution.


Yaneer Bar-Yam (@yaneerbaryam) tweeted at 5:26 PM – 15 May 2017 :

@nntaleb @IGUZM4N @RonPaul In case you haven’t seen this https://t.co/UONPN01QEi(http://twitter.com/yaneerbaryam/status/864260624594939906?s=17)

Yaneer Bar-Yam (@yaneerbaryam) tweeted at 10:44 AM – 14 May 2017 :

@IGUZM4N @RonPaul @nntaleb There is a substantial literature. For example, markets are unstable to monopolization. We have other work on this https://t.co/nxr6AonogU (http://twitter.com/yaneerbaryam/status/863797289705504768?s=17)

people are unstable to marketization
we have other work on this

it is the internal structure of the market, not external news, which can cause a market to crash

The uptick rule was designed to limit the rapid selling of borrowed shares and was implemented after the crash of 1929 to prevent future crashes. After it was repealed in the summer of 2007 due to unsound interpretations of data, the market was left more vulnerable to spikes and drops. At NECSI recommendation, the SEC announced a return of the rule, which caused an immediate market response. However, the rule has been reinstated only in a limited form, leaving the market vulnerable.

better up stick rule: quit measuring transactions


Regulatory capture necsi.edu/research/econo…

“It’s hard to maintain integrity when the monetary benefits are so high, especially when the harmful consequences are not obvious even when they are severe, as happens often in the financial sector,” says Bar-Yam. “We need to make the consequences more clear and the process of collusion more difficult.”

hard to maintain integrity when measuring transactions..need to disengage

NassimNicholasTaleb (@nntaleb) tweeted at 11:07 AM – 14 May 2017 :

@yaneerbaryam @IGUZM4N @RonPaul Well functioning markets REQUIRE absence of monopolies. Monopolies are often the result of government cronies/patronage

well functioning people require absence of markets ..markets are often result of measure/compare


Yaneer Bar-Yam (@yaneerbaryam) tweeted at 9:12 AM on Tue, Aug 15, 2017:
Question: I am interested in suggestions about how to teach the concept of “space of possibilities”

depends..do you want people who can 1\ discuss math or 2\ imagine a nother way to live (aka: the seemingly impossible)

How about both possibilities ;)

Suggestions for two possibilities welcome–

perhaps watch i am .. [https://www.youtube.com/watch?v=iYtfnONazTU]

a dose of reality.. elements that undermine everything we’ve been told.. wake up..

we are far grander than we have been told

[your sec 3.1 suggested movies as helping with description: @yaneerbaryam No great reference. Space of possibilities is a set of possible states (or time histories). Related: Sec 3.1 in necsi.edu/projects/yanee…]

and perhaps.. let’s try this .. for space of possibilities:

ie: hlb via 2 convos that io dance.. as the day..[aka: not part\ial.. for (blank)’s sake…]..  a nother way



Yaneer Bar-Yam (@yaneerbaryam) tweeted at 10:19 AM on Tue, Oct 17, 2017:
How can we fix the economy? Our just released analysis https://t.co/6iqKmrcmNJhttps://t.co/F8oXYXl8bX

Our analysis supports advocates of greater income and / or government support for the poor who use a larger fraction of income for consumption. This promotes investment due to the growth in expenditures. Otherwise, investment has limited opportunities to gain returns above inflation so capital remains uninvested, and does not contribute to the growth of economic activity.

wow. contribute to growth of econ? that’s how you fix it..? more money circulated.

twitter bio:

Exploring the space of possibilities and its boundaries.

Working on solving our societal crises.

this (what you suggest) is not that


Yaneer Bar-Yam (@yaneerbaryam) tweeted at 10:31 AM on Wed, Oct 18, 2017:
Wealth Redistribution is Essential to Stabilize Economy Says Math
by Daniel Starkey and https://t.co/MVvU1UM4N4 Original paper https://t.co/6iqKmrcmNJ

Look, it’s been pretty clear that “Reaganomics” wasn’t really a viable tactic for some time now. We haven’t been lead to any new economic promised land..Now there’s another bit of evidence to back up the idea that other economic tools — like wealth redistribution — are more effective for achieving tangible goals. *It’s called math. 

A new paper by the New England Complex Systems Institute uses mathematics to find a solution one of society’s most pernicious problems — income inequality..t

oy.. of math and men

math isn’t going to bring us equity (everyone getting a go everyday)

*“We need a very measured, but definite shift in direction that will address the economic problems and also address economic inequality problems,” Yaneer Bar-Yam, physicist and founder of the New England Complex Systems Institute, told Motherboard. “We went too far with Reaganomics, and now we have to go back in order to have healthy economic growth.”

rather.. *we need a very un measured.. shift.. that will address equity.. everyone getting a go everyday

krishnamurti measure law

Without enough cash in the hands of consumers, they won’t have resources to buy the things manufacturers make

affluence w/o abundance.. and sans measuring/money

Citizens have to be able to buy things *for this economic systems to work, and right now many of the poorest among us just don’t have the ability to buy anything. By giving them money through government programs and wage hikes and the like, they’ll have money to spend which will then go back into the hands of the wealthy to encourage more investment and production.

*that’s our goal..? in 2017.. that’s our goal..?

People want less inequality. Wealth redistribution does that. And we can prove it mathematically.

of course