Thinking Systemically

Achieving Sustainable Systems

Systems: Complex, Dynamic Processes
 Systems science provides a means of analyzing and
understanding complex processes based on a few basic
principles
 Complex systems behave in complex ways and may
change or adapt over time
 When problems emerge there is an underlying positive
feedback loop that may not be evident
 Such problems cannot be solved using linear causality
thinking
 True sustainability involves whole systems thinking and
design
Permaculture: A Systems Approach to
Sustainable Living
 The application of systems thinking and systems science
to the design of living arrangements
 Environment and climate
 Buildings, design, orientation, construction
 Water management
 Food production
 What is Systems Thinking?
 What is Systems Science?
 What is Sustainability?
 Then we can understand Permaculture
 Permaculture: The Systems Approach to
Sustainable Living

governance values

shelter education
A sustainable
minerals
social system

environment
water clothing
community

timber
fiber food
soil

fuel
What is Systems Thinking?
 Seeing the system as a functional whole and not just a
collection of parts
 Seeing relations among the components and between a
system and its environment as the basis of explanation
 Using various kinds of models to represent whole
systems and their dynamical behaviors
 Testing understanding by playing ‘what-if’ games with the
models to see:
 if they behave the same way the system does under similar
circumstances
 if there are leverage points that can be used to move a system
in a desired direction of behavior
What is Systems Science?
 The application of scientific thinking with systems thinking in
order to understand how the world works
 Systems science discovers and studies the common principles
that apply to all systems studied by specific sciences
 It defines a general property – systemness – that is found in
objects of interest regardless of scale and complexity, e.g.:
 Natural physical, chemical, biological systems
 Social systems
 Built systems, cultural artifacts
 It provides universal principles that can be applied to specific
sciences to help them develop better explanatory models
What Is Sustainability?
 A property of systems is longevity
 Systems are internally organized so as to maintain
themselves in spite of environmental contingencies
 Some kinds of systems are adaptable over time as the
environment changes
 What are the properties of system organization and
adaptation that lead to longevity?
 Real systems age 
 But complex, adaptive systems can give rise to newer
systems (e.g. living systems reproduce) 
 Life on Earth is a sustainable system as long as the Sun is
stable!
A First Look At Permaculture
 The term was devised by Bill Mollison and David
Holmgren (Australia) in the 1970s to describe their
application of systems ecology to designing sustainable
living arrangements
 Systems ecology is the science that looks at ecology from
a holistic perspective, but especially considers the flows of
energy and material resources through an ecosystem
 Designing local communities according to the principles
of systems that are at the heart of sustainability
 Combines organic food production, water management,
“green” building practices, and many other aspects of a
living system to achieve the goal
The Challenges for the 21st Century
 The growing evidence points to the possibility that we
are running out of critical resources if we continue to live
over-consuming lives
 Our reliance on high technology has been based on a
faith that might not be justified in practice – technology
alone cannot replenish dwindling non-renewable
resources
 Permaculture, and systems science in general, represents
a new way to look at technologies that have systemic
purposes
 The goal is to allow people to live comfortable, fulfilling
lives for many generations to come
The Principles of Systems Science
And the Keys to Sustainable Living
How Systems Science Works
 Survey models of specific systems, e.g. biological systems such
as cells and organisms or social systems such as communities
 Seek commonalities in terms of explanations of how systems
function and evolve
 Use analytical methods to find those commonalities
 Develop languages that can describe all systems regardless of
specific domains, e.g. whether biological or physical
 Develop general principles that provide causal explanations
regardless of the details of any specific system
 Develop mathematical descriptions of those principles such
that they can be employed to discover new aspects of specific
systems
What Principles are True of All Systems?
 Here are a set of principles that have been discovered to
operate over all knowledge domains
1. Systemness – the world is composed of systems of systems
2. Systems are organized in structural and functional hierarchies
3. Systems can be represented as abstract networks of relations
between components
4. Systems are dynamic processes on one or more time scales
5. Systems exhibit various kinds and levels of complexity
6. Systems emerge from proto-systems (unorganized, not complex)
and evolve over time to greater organization and complexity
7. Systems can encode knowledge and receive and send information
8. Systems evolve internal regulation subsystems to achieve stability
Additional Principles
 Several principles related to systems thinking, systems
science, and systems development
9. Systems can contain models of other systems
10. Sufficiently complex, adaptive systems can contain models of
themselves (brains and mental models)
11. Systems can be understood (a corollary of #9) – Science as
the building of models
12. Systems can be improved (a corollary of #6) – Engineering as
an evolutionary process
 Lets look at the details
Principle 1 – Systemness, or What Makes
Something a System
 A collection of many component parts (number and
types) – objects,
 That interact with one another through various
interconnections that have varying strengths
 That maintain structural integrity over time including:
 Maintaining a boundary that demarcates the system object from
the environment
 Maintaining the interconnections in stable configurations
 That perform an overall function by accepting inputs from
an environment and processing them into recognizable
outputs (to the environment) as a result of the internal
interconnections between the components.
Not a system

flows not
processed

A system boundary
no or few
interconnections

interconnections
no boundary

inputs and
outputs

Figure 2. A non-system vs. a system.

Definitional Problem
 It is hard to find a “non-system”!
 An arbitrarily chosen volume of outer space – no boundary
 An artificially contained volume of gas molecules – no function
 The difference between simple systems and non-systems
 Add an input of heat and a measuring sensor for pressure to
the contained volume of gas and it becomes a simple system
with a function of relating heat (temperature) and pressure and
producing information for an observer!
 Problem: If just about everything is a system then isn’t the
concept meaningless?
 Epistemological Answer: Because everything is a system,
systems thinking is a way to know the world.
 An Ontological Side Trip
 Four Aspects of the Physical Universe – What Exists
 Matter – Substance with attributes of mass and momentum
 Energy – That which changes the position/momentum of
matter over time; the capacity to do work
 Knowledge – Structure (of matter) that anticipates energy
flow
 Information – Message that conveys “a difference that makes
a difference”*
 All that exists involves these four aspects
 We understand systemness in terms of the interactions
between these four aspects.

* Gregory Bateson, (1972). Steps to an Ecology of Mind: Collected Essays in Anthropology, Psychiatry, Evolution, and
Epistemology. University Of Chicago Press.
The Systems Ontological Perspective
Mysterium
Physical Movement Ethereal
Transfers

Dark Quantum
Energy Uncertainty
Energy Information

Generates

Influences
Reveals
Forces

Stores

Constructed in

Dark Matter Knowledge Quantum

Matter Weirdness
Stores

Structure
Human
Largest Smallest
Scale
Relations of the Existential Aspects (I)
 The Physical
 Matter and Energy
 Matter can be converted to energy, E = mc 2
 Energy moves (forces) matter
 Matter is measurable directly
 Energy is measured by its effects on matter
 Matter can store energy (e.g. chemical bonds)
 The Ethereal
 Knowledge and Information
 Knowledge is embodied in the structural arrangements of matter
 Information is the measure of how a message receipt affects the
structural arrangements of matter – “news of difference”
 K = 1/I and I = -log (Pm): Both are probabilistic in nature
 Newly acquired knowledge generates information when it causes changes
in system behavior
Relations of the Existential Aspects (II)
 Structure
 Matter and Knowledge
 The structure of a system – how the components of matter are arranged
and exchange energy – encodes the history of exchanges with the
environment, the messages received, as memory
 The arrangement reflects a priori expectations of messages (conveyed by
energy), what the system expects in future exchanges
 Movement
 Energy and Information
 Energy flows from material node to node, transversing space
 Energy forces matter to move relative to other matter
 Modulated energy conveys messages from node to node
 Communication is the transport of energy through a channel conveying a
message
Relations of the Existential Aspects (III)
 Scale
 The four aspects are discerned at all scales of time and space
 At the largest scales of space and time the Universe is ruled by
the forces of gravity (dark matter) and dark energy
 At the smallest scales of space and time the Universe is ruled
by quantum effects
 Uncertainty
 Weirdness
 The Laws of Nature describe how all of these aspects
relate to one another on all of these scales
Principle 2 - Systems are Organized in a
Structural Hierarchy
 A system is a subsystem of a larger system
 A system contains components that may be, themselves,
subsystems
 Systems decompose through a subsystem tree
 Roughly akin to the material composition hierarchy, e.g.
organism – cells – molecules – atoms – subatomic
particles, etc.
Systems Composed of Subsystems

subsystems

inputs outputs

larger
system

Figure 3. Systems are comprised of subsystems.

The System Hierarchical Tree

System

Subsystems & interconnections

Components & interconnections

Figure 4. Systems, subsystems & components form a structural hierarchy.

Principle 3 - Systems Can Be Represented
in an Abstract Way – Networks of Networks
 Systems are, in reality, composed of networks of components
 Systems can be represented as a network comprised of:
 Nodes – representing components
 Links – representing the interconnections
 Network representations are powerful tools for analyzing and
modeling systems.
 Nodes have properties that can take on variable values over
time
 Some properties of nodes are exposed to other nodes and
constitute the “personality” of a component
 Links can be flows of matter, energy, or information, or forces
that bind or repel.
Abstract Network Representation

inputs
outputs

Graph Theory Mathematics Can

Be Used to Answer Structural and
Functional Questions About This
Network (a directed graph) decomposed node
Principle 4 – Systems are Dynamic
Processes
 Systems are always in motion:
 Relative to one another
 Internally
 Energy and Material Flows
 Principle 4.1 – Causal Relations
 Principle 4.2 – Multiple Time Scales
 Principle 4.3 – Radius of Effect
 Principle 4.4 – All Objects are Processes at Some Time
Scale
P4.1 – Causal Relations
 Processes appear to be continuous at larger scales of
time and space
 At a much finer scale we observe discrete events or
changes of state
 Processes can be described/characterized as state
changes
 Causality involves a change of state of the system based
on prior states and input events
 With no inputs processes decay or fall apart over time
 With inputs (esp. energy) processes proceed and may either
grow in size/complexity, or obtain a steady state
State Changes
 State Spaces
 Quantifiable parameters simultaneously measured at an instant
in time
 Causal relations between parameters constrain values to
certain levels relative to one another – the state of a system

 Phase Spaces
 A space in which all allowable states are represented by single
points
 A trajectory defines how a system moves from one state to
the next (or another)
 Example of State Space Representation
parameters

pl pl pl
inputs outputs inputs outputs inputs outputs
ik System oj ik System oj ik System oj
of of of
interest interest interest

sn so sp
time t0 t1 t2

At each interval of state

time (e.g. Δt = t1- t0)
the parameters and output
input = x A state transition diagram – the
outputs will be altered sr sn system will transition from state n
transition to so to state o on input x, otherwise it
by the inputs.
input ≠ x will remain in state n. State o will
have different parameter measures
sp and output.
sq
Causal Relations
 Event A (e.g. an input) causes event B (change of state) if:
 A precedes B by time Δt (on some scale)
 A never succeeds B except after some long time interval, nΔt
 A is connected to B by a force or a flow
 Event A occurring causes a change of state, event B, which
can act as an input to another system/process producing
a causal chain: A →
t B → t C, events separated by Δt

 Mutual or Circular Causality: A → B → C → A

 Multiple Causality: A → B → C → (B & Y or B OR Y)
X→Y
 Stochastic Causality: A → B → C (A causes B with probability, x)
P=x P=y
P4.2 – Multiple Time Scales
 Activity (movement and composition changes) occurs on
many time scales roughly correlated with the size of
components
 Atomic – on the order of attoseconds (10 -18sec.)
 Molecular – on the order of femtoseconds (10-15 sec.)
 Computer circuit switching – on the order of picoseconds
 Nerve impulses – on the order of milliseconds
 Digestion – on the order of hours
 Weather changes – on the order of days and weeks
 Human life span – on the order of tens of years
 Species longevity – on the order of hundreds of thousands of
years
 Continental drift – on the order of millions of years
P4.3 – Radius of Effect
 The system of interest, or agent, has a limited range of
perception (how far away it can be from an event that will
affect it)
 Causal chains can go back far in time and distant in space
and still have an impact
 Major source of uncertainty events from long
ago still able to
affect the agent

agent: system of radius

interest of
hidden
causal
chains
temporal scale radius of
perception
P4.4 – All Objects are Processes
 A rock at the atomic level of interaction is a chemical
process operating over very long time scales
 A rock, not having usable energy or material inputs is
undergoing a long process of decay
 Processes, if examined at a lower scale (higher resolution
in a microscope), are made up of objects that seem to
have form and solidity
 But at a yet higher resolution, these solid objects are
found to be processes themselves!
Systems Are Always in Flux
 Dynamics of the Environment
 Stochastic – unpredictable in detail
 Non-stationary – long term changes in statistical properties
 Chaotic – sensitive to initial conditions, no two systems follow
same trajectory
 Systems respond to their environments
 Environments respond to their component systems
 Adaptive systems are those that have complex, often
redundant mechanisms for dealing with changing
environments while maintaining a core constancy
 Life as the quintessential example of adaptive systems
 Homeostasis and Autopoiesis examples
Dynamics Continuing Over Time Tell a Story
 Dynamic behavior is described by a sequence of
statements about the changing state
 Input triggers cause state changes
 Subsequent states depend on prior states and current inputs
 The story line describes causal sequences
 Systems end up in a final state at the end of observation
 Complex systems tell complex stories
 All languages employ a lexicon of symbols to represent
things, events, causal relations, etc.
 In describing the on-going story of a system’s dynamics
we employ a formal language
System Dynamics Language
 A language consists of:
 Lexicon – words that serve purposes
 nouns – objects
 verbs – actions
 modifiers – descriptors, temporal, quantifiers, etc.
 Syntax – Grammar rules
 how symbols can be combined to convey thoughts
 Semantics – Meaning
 Pragmatics – Context
 The language can be used to express meaning in context
 There is a mapping from real things and behaviors:
 to an abstract verbal representation – text, narrative
 to an abstract graphical representation – pictures
 to an abstract numerical representation – mathematics and
computer programs
Describing the Dynamics of Systems
 Each process converts its inputs into outputs
 Each flow has a rate measure
 can vary over various time scales
 stochastic
 Each stock has a level measure
 Conversions can be represented by mathematical
functions
 Product per unit time = ft ( input1, input2, …, inputn )
 inputi = units of measure / unit of time
 wastej = units of measure / unit of time
 All equations have to conform to the physical laws of
conservation and decay
A Graphical Language for Describing
Systems Dynamics
Process Framework Flows Forces
low entropy
material flow repulsion
source process object sink

high entropy
material flow
attraction
stock System Dynamics high potential
energy flow
Language Lexicon –
Semantic Objects low potential
energy flow
(heat)
Flow Operators message flow

resistor
flow control Message Modulators
actuator
difference
A B
diode join split
stock level and flow sensors comparator
Example of Verbal Description
 A specific ‘firm’ is an economic process that manufactures
products using energy, labor, and material parts inputs. It
temporarily stores component parts as they are shipped
in. It temporarily stores finished product in preparation
for shipping to customers on demand. A manufacturing
control system keeps track of orders for product, the
state of product inventory, and issues orders to
manufacturing to replenish stocks as needed.
 The firm is comprised of a manufacturing sub-process, a
parts inventory and a product inventory, and three
management (information) processes, inventory,
manufacturing, and sales.
 Example of a Visual Description Language

energy
supplier
High Level View
Firm
labor manufacturing A process described as a flow network

shipping customer

parts parts production

supplier reports production
management
parts production
requirements product
schedule
requirements
inventory manufacturing
management

Adding Details sales

management
assembly kit
parts orders
order shipping
orders

waste
product customers
inventory
suppliers
parts
inventory

material losses
Example of a Numerical Representation
 A computer code for computing the behavior of the firm
Process: Firm {
time constant: 1 day;
inputs {
energy {
range: 4000 to 7000 kw;
type: electricity;
}
labor {
type: human;
range: 6 to 8 hrs/person;
}
:
: // etc.
Numerical Model cont.
Sub-process: Manufacturing {
inputs: labor, energy, parts;

function: manufacture (product_units_needed) {

while (product_count < product_units_needed) {
product_unit = .25 * labor + .75 * parts + .13 * electricity;
product_count = product_count + product_unit;
}
output product_count;
}
}
Sub-process: ProductionManagement {
inputs: product_inventory, sales;

function: product_inventory {
if product_inventory > 0 then {
output product_unit;
product_inventory = product_inventory – 1;
if product_inventory < 10 then production_management_notify(product);
}
}
Translations From One Language to Another
 All formal descriptions and stories should be inter-
convertible
 Stories about how a system behaves under various input
circumstances (i.e., the dynamics of the environment) are
scenarios
 Scenarios can be used to make predictions about system
behaviors
 The formal descriptions are called models – more on
those later
Thinking Systemically
 End of Part I
 Principles 5 – 12 in Part II