Lecture 1: Understanding

Complexity
CCST9048 Simplifying Complexity
University of Hong Kong
Dr. Tim Wotherspoon
Overview

• In this lecture we will try to introduce the idea of complexity and give
some examples of complex systems and look at some common
features.
• In order to do so, we will first look at the idea of reductionism which
is a key part of the modern approach to science.
• The study of complex systems arises from understanding places
where reductionism fails to give the desired results.
• We can think of systems where reductionism works well as the anti-
complex systems. (the opposite of complex is simple but I want to
avoid that word here).
A Very Brief
History of Science
Aristotle (384-322 BCE)

• Very influential still

• Increased our understanding
across many fields from politics
to astronomy
• Did early ecological research
• Attributed with the creation of
“empiricism”

Aristotle
“The phenomena of generation of the vole are most astonishing, both for the
number of the young and the rapidity of the recurrence in the births… Their
disappearance is unaccountable: in a few days not a mouse will be there to be seen.
And yet in the time before these few days men fail to keep down their numbers by
smoking them out and by unearthing them, or by regularly hunting them and
turning in swine upon them; for pigs turn up the vole-galleries by rooting with their
snouts. Foxes also hunt them, and in particular the ferocious ferrets, but they all
together make no way against the prolific qualities of the voles and the rapidity of
their breeding. When they are superabundant, nothing succeeds in thinning them
down except the rain; but after the heavy rains they disappear rapidly.”

Aristotle, History of Animals, 350 BC

Descartes (1596-1650)

• Commonly attributed with

developing the scientific
method
• Focused on using logic and
mathematical proofs to draw
conclusion (rationalism)
• An early proponent of
reductionism
Reductionism

• “To divide all the difficulties

under examination into as many
parts as possible, and as many
as were required to solve them
in the best way.”

• “to conduct my thoughts in a

given order, beginning with the
simplest and most easily
understood objects, and
gradually ascending, as it were
step by step, to the knowledge
of the most complex.”
Newton 1642-1726
• Newton developed classical
mechanics as the basis for
physics.
• Three laws of motion allowed for
the basis of understanding the
physical world.
Newton and Reductionism

• Even since early Greece, the

idea of a clockwork universe was
very common.
• Newton’s laws extended this
across the physical world.
• All phenomena can be
understood by reducing them to
simply fundamental laws.
Laplace

• This led to the idea of

determinism.
• Laplace’s demon describes a
powerful intelligence that
knows the position and speed of
every object in the universe.
• With this information, according
to determinism, it should be
possible to predict the future.
http://philosophicatz.wordpress.com/
“It seems probable that most of the grand underlying principles
have been firmly established and that further advances are to be
sought chiefly in the rigorous application of these principles to all
phenomena which come under our notice.”
– Albert Michelson 1894
Reductionism
Epistemological Reductionism

• One body of knowledge can be understood by reducing it through

the use of another.
• The tools for physics are mathematics…

http://xkcd.com/435/
 Methodological Reductionism
Reductionism is the most natural
thing in the world to grasp. It’s
simply the belief that ”a whole can
be understood completely if you
understand its parts, and the
nature of their ‘sum.’” No one in
her left brain could reject
reductionism.
Methodological Reductionism

• This has been extremely

successful in the physical
sciences.
• However it has major
limitations, particularly in
systems that have many
interactions between their
constituent simple parts.

http://xkcd.com/793/
 Methodological Reductionism
• Most students will come across • Assumptions:
the so-called Ideal Gas Law • Point particles
• 𝑃𝑉 = 𝑛𝑅𝑇 = 𝑁𝑘𝑇 • Elastic conditions with the
boundary
• This works well in many cases! • No interactions between particles

• None of these assumptions are true!

• Phase transition?!
• It works poorly in situations where the assumptions are not even approximately
true (high density, low temperature, high pressure, strong interactions).
• Modeling a real gas consists of treating it first as an ideal gas and then adding
correction terms corresponding to corrections to bad assumptions.
Reductionism and Homework

• Part a might help with part b!

• The separate problems might not
seem related but usually they
are!
• Asking the big problem would
overwhelm students so we try
and guide you through it from
the simple scenarios to the most
difficult.
• Or, we ask you to do each step
separately before asking you to
combine the results to answer a
difficult problem.

From Introduction to Quantum Mechanics by David J. Griffiths. Note:

the answer to Problem 1.5 is 2/π
Limits of Reductionism

“The ability to reduce everything to simple

fundamental laws does not imply the
ability to start from those laws and
reconstruct the universe. In fact, the more
the elementary particle physicists tell us
about the nature of fundamental laws, the
less revelance they seem to have to the
very real problems of the rest of the
science, much less to those of society.”

Anderson, P. W. (1972). More is different. Science,

177(4047), 393–396.
https://doi.org/10.1126/science.177.4047.393
Philip W. Anderson
 Limits of Reductionism

The Human Genome Project

may stand as an archetypical
example of a reductionist
approach.
• Human physiology is determined by
genes. Therefore (understanding the
genes) = (understand human
physiology)
• Led to hopes of “O-GOD” (One gene-
one disease)
• However, genes do not act
independently. Although the number
of genes is around 20,000, the Genomic medicine: the sorcerer's new broom?
number of proteins may be much A Challenge to Gene Theory, a Tougher Look at Biotech
higher.
Complexity Definition
 Complexity etymology

• The word complexity

comes from this latin word
that is related to weaving.
• Please note that the word
“complicated” has a
different origin.
• Examples of “complicated
systems”:
• Mechanical watch
• Rube Goldberg
Machine

http://en.wiktionary.org/wiki/plectere
 Definition

“Complex system: a system in which large networks of

components with no central control and simple rules of operation
give rise to complex collective behavior, sophisticated information
processing and adaptation via learning or evolution.”

Mitchell 2011
Defining Complexity

• The above is only one possible definition for complexity, but others
may fit better to certain complex systems.
• Some authors distinguish between complex systems and complex adaptive
systems.
• A hurricane is not adaptive, but it fits other properties of a complex system
• An alternative definition:
• A system that exhibits nontrivial emergent and self-organizing behaviors.
Defining Complexity

• A complex system is difficult to

define.
• We have a list of typical
features.
• If a system has some but not all
those features is it a complex
system?
• If a system has all those features
is it a complex system?
• Researchers in this field typically
gain a “I know it when I see it”
attitude about defining Complex
Systems
Examples:
Insect colonies
Insect Colonies as Complex Systems

• Individual ants are not sophisticated yet the collective behaviors of

ant colonies are.
• Individual ants have an instinct for food and can exchange and
respond to chemical signals with other ants.
• How do those signals lead to the collective behaviors?
• How does the colony adapt to changing environment?
• How did such colonies emerge in the biological evolution of diverse
species?
Examples:
Neural networks
The Brain as a Complex System

• Brains are made up of cells called neurons

• Like in the example of an ant colony, the working of individual
neurons is unsophisticated and easily understood.
• All of your large-scale brain activities such as perception, emotion,
behavior, intellectual thought are believed to emerge from patterns
of connections and signals amongst groups of these simple
components.
• How does this work?
• How does learning occur?
• How did this network emerge?
The Immune System

• The individual components of immune systems: lymphocytes,

antibodies, macrophages, cytokines… are also well understood in
their individual functions.
• Yet were it not for the interactions or signals between these cells and
different types of cells, your body would be unable to identify and
attack disease. (none of the cells can operate without the others)
• What keeps this from going wrong? (sometimes it does!)
• How did this emerge?
 Examples:
The global economy

Vitali S, Glattfelder JB, Battiston S (2011) The Network of Global Corporate Control. PLoS ONE 6(10): e25995.
doi:10.1371/journal.pone.0025995
http://www.plosone.org/article/info:doi/10.1371/journal.pone.0025995
Economies

• An economy consist of players such as companies or individuals.

• Even if we treat individuals under the simple assumptions such as
that they act in their rational self-interest, macroscopic properties
such as price and market efficiency emerge.
• How does efficiency come about?
• Why does it fail in real-life markets?
• How are market bubbles/crashes created?
Examples:
The world wide web

http://www.independent.co.uk/life-style/gadgets-and-tech/facebook-maps-111-billion-
friendships-to-show-off-social-networks-global-reach-8839116.html
Social networks and the World Wide Web

• As a social network user, you make independent choices about who

you interact with. Webpage designers make independent choices
about which webpages to link to.
• (With notable exceptions), there is little oversight as to how internet
connections are formed.
• These networks reliably form organizational structures.
• How do properties like virality form?
• How is it possible for the internet and search engines to adapt so
quickly to changing interests of users?
Features of Complexity
Collective complex behavior

Although the behaviors of the individual

components may be ruled by simple rules and
absent of central control, the collective behavior
may be difficult to predict.
 Signal and information processing

These systems produce and respond to

information internally between
components and exchange information
with their environment.

http://www.gizmodo.com.au/2014/03/what-happens-when-you-throw-four-sharks-into-a-giant-school-of-fish/
Adaptation

Over time, the system may react very differently to

similar inputs or may be able to produce new
behaviors to react to new inputs, through learning or
evolutionary processes.

Photograph by Bruce Robison/Corbis

Self-organization Organized behavior that arises without the
presence of leadership

Emergence The macroscopic behaviors of a complex-

system; used when those behaviors are not
associated with the components individually
Understanding Ecology as a Complex System
Yellowstone and Wolves

• Yellowstone National Park is the first nature conservancy of its kind in

the world and a wonderful place, est. 1872.
• At the time of its opening, hunting was a permitted and unregulated
activity in the park.
• This led to the removal of wolf populations in the park.
• This removal led to an explosion in the elk population which park
rangers then also attempted to manage.
• Several negative ecological impacts were observed.
• Wolves were later reintroduced in hopes of repairing the damage.
• Did it work?
http://www.cof.orst.edu/wolves/
https://www.nps.gov/yell/learn/nature/images/crop38401254631_8b63d73741_6k.jpg?maxwidth=1200&maxheight=1200&autorotate=false

“So far, data suggest wolves are contributing to decreased numbers of elk calves
surviving to adulthood and decreased survival of adult elk. Wolves may also be
affecting where and how elk use the habitat. Some of these effects were predictable
but were based on research in relatively simple systems of one to two predator and
prey species. Such is not the case in Yellowstone, where four other large predators
(black bears, grizzly bears, coyotes, and cougars) prey on elk—and people hunt the
elk outside the park. Thus, interactions of wolves with elk and other ungulates have
created a new degree of complexity that makes it difficult to project long-term
population trends.” (https://www.nps.gov/yell/learn/nature/wolf-restoration.htm)
Simplifying Complexity
Complexity and Reductionism: Two opposing
ideas.
• Reductionism
• To understand a complicated whole, understand its constituent parts.
• If those parts are difficult to understand, break them into smaller parts.
• Works well in systems where interactions between parts are few or are
unimportant.
• Complexity
• Reductionism fails at times because understanding the parts may do little for
understanding the whole.
• In these cases, we need to discover the rules of the interactions.
• Luckily, we will often see that the macroscopic behaviors and/or structures
might arise from very simple rules of interaction.
Simplifying Complexity

• Throughout this course, we will look at different methods and

strategies for modeling complex systems.
• Recall… “Complex system: a system in which large networks of
components with no central control and simple rules of operation
give rise to complex collective behavior, sophisticated information
processing and adaptation via learning or evolution.”
• As we will see, often in modeling complex systems, the rules of
operation are very simple.
• Complexity often arises from the repeated application of rules that
are intuitive and easy-to-understand.