“Some problems are so complex that you

have to be highly intelligent and well informed

just to be undecided about them.”
--Laurence J. Peter

Wicked Problems
Social Complexity

by Jeff Conklin, Ph.D.

 Wicked Problems
Social Complexity1

This book is about collective incompatible tacit assumptions about the problem,
and each believes that his or her understandings are
intelligence: the creativity complete and shared by all.
and resourcefulness that a The antidote to fragmentation is shared understand-
group or team can bring to a ing and shared commitment. This book is about a
new way to create shared understanding, and this
collaborative problem. chapter sets the stage by exploring specific ways that
the forces of fragmentation work in organizations and
projects.
C ollective intelligence is a natural property of
socially shared cognition, a natural enabler of
collaboration. But there are also natural forces that
challenge collective intelligence, forces that doom
projects and make collaboration difficult or
impossible. These are forces of fragmentation.

The concept of fragmentation provides a name and an

image for a phenomenon that pulls apart something
which is potentially whole. Fragmentation suggests a
condition in which the people involved see themselves
as more separate than united, and in which informa-
tion and knowledge are chaotic and scattered. The
Fragmentation and Organizational Pain
fragmented pieces are, in essence, the perspectives, There is a subtle but pervasive kind of pain in our
understandings, and intentions of the collaborators. organizations. It is characterized by such frequently
Fragmentation, for example, is when the stakeholders heard complaints as “How am I supposed to get my
in a project are all convinced that their version of the work done with all of these meetings?” and “We
problem is correct. Fragmentation can be hidden, as always have time to do it over again, but never time to
when stakeholders don’t even realize that there are
This paper is Chapter 1 of Dialogue Mapping: Building Shared Understanding of Wicked Problems, by Jeff Conklin, Ph.D., Wiley, October 2005. For
more information see the CogNexus Institute website at http://www.cognexus.org. © 2001-2008 CogNexus Institute. Rev. Oct 2008.
 Wicked Problems and Social Complexity

do it right.” It is a sense of futility of expecting things in a project. The more parties involved in a collabora-
to be one way and repeatedly banging into a different tion, the more socially complex. The more different
reality. It is the dull ache of déjà-vu when you are those parties are, the more diverse, the more socially
handed an impossible deadline or a vague assignment. complex. The fragmenting force of social complexity
It is the frustration of calling a meeting to make a can make effective communication very difficult.
decision and watching the meeting unravel into a
Social complexity requires new understandings,
battle between rival departments, or get lost in a
processes, and tools that are attuned to the fundamen-
thicket of confusion over the meaning of a technical
tally social and conversational nature of work.
term. It is the frustration of finally achieving a hard-
won decision and then having it fall apart or get
For example, in a joint project involving several
“pocket vetoed” because there wasn’t really buy in. It
companies and government agencies, there was a
is the pain of fragmentation.
prolonged struggle over the mission statement simply
I was working late one evening when the because of a terminology difference: each sponsoring
janitor came in to vacuum the office. I noticed agency had their own term for the core concept 1 , and
that he was going back and forth over the to pick one term meant disenfranchising one of the
same areas without appearing to get the lint agencies. This is a very simple example of fragmenta-
up off the carpet. I smiled and shouted to him tion of meaning.
(the vacuum cleaner was a loud one) “It must
Social complexity means that a project team works in
be frustrating to have to use that vacuum
a social network, a network of controllers and influ-
cleaner.” He looked at me with a sad smile and
encers including individual stakeholders, other
said “Not as frustrating as being told to go
project teams, and other organizations. These
back and do it over!” It is that kind of pain,
relationships, whether they are with direct stake-
and it goes all the way up to the executive
holders or those more peripherally involved, must be
suite.
included in the project. For it is not whether the
Part of the pain is a misunderstanding of the nature of project team comes up with the right answer, but
the problems at hand. More precisely, the pain is whose buy-in they have that really matters. To put it
caused by working on a special class of problems – more starkly, without being included in the thinking
wicked problems – with thinking, tools, and methods and decision-making process, members of the social
that are useful only for simpler (“tame”) problems. network may seek to undermine or even sabotage the
Problem wickedness is a force of fragmentation. Most project if their needs are not considered. Social
projects today have a significant wicked component. complexity can be understood and used effectively,
Wicked problems are so commonplace that the chaos but it can be ignored only at great peril.
and futility that usually attend them are accepted as
My janitor friend had an advantage over the rest of us
inevitable. Failing to recognize the “wicked dynamics”
in the organization because he could clearly see that
in problems, we persist in applying inappropriate
his vacuum cleaner was not actually picking up the
methods and tools to them.

Another force of fragmentation is social complexity, 1

The project concerned “unmanned aerial vehicles” (UAVs),
the number and diversity of players who are involved also known as “remotely piloted aircraft” (RPAs).

Page 3
dirt. When we are working on wicked problems in a The analysis showed, not surprisingly, that these
socially complex environment, it is much harder to designers worked simultaneously on understanding the
notice that our tools are simply not “picking up the problem and formulating a solution. They exhibited
dirt.” two ways of trying to understand the problem:

As we enter the new millennium the forces of frag- • efforts to understand the requirements for the
mentation appear to be increasing, and the increasing system (from a one page problem statement they
intensity of these forces causes more and more pro- were given at the beginning of the session); and
jects to flounder and fail. The bigger they are, the • mental simulations (e.g. “Let’s see, I’m on the
more intense the fragmenting forces, the more likely second floor and the elevator is on the third floor
the projects are to fail. and I push the ’Up’ button. That’s going to create
this situation....”).
Moreover, the situation is not that project teams are
aware of fragmentation and are taking appropriate On the solution side, their activities were classified
measures to deal with it – quite the opposite. Most into high, medium, and low levels of design, with
teams accept fragmentation as inevitable. Indeed, high-level design being general ideas, and low being
most people are unaware of some basic facts about details at the implementation level. These levels are
novel and complex problems. Managers, in particular, analogous to an architect’s sketch, working drawings,
seem to be unaware that linear processes are not and a detailed blueprint and materials list for a house.
effective with such problems.
Traditional thinking, cognitive studies, and the pre-
Opportunity Driven Problem Solving vailing design methods all predicted that the best way
to work on a problem like this was to follow an
A study in the 1980’s at the Microelectronics and
orderly and linear ‘top down’ process, working from
Computer Technology Corporation (MCC) looked
the problem to the solution. This logic is familiar to
into how people solve problems. The study focused
all of us. You begin by understanding the problem.
on design, but the results apply to virtually any other
This often includes gathering and analyzing ‘require-
kind of problem solving or decision-making activity –
ments’ from customers or users. Once you have the
the kinds projects are fraught with.
problem specified and the requirements analyzed, you
A number of designers participated in an experiment are ready to formulate a solution, and eventually to
in which the exercise was to design an elevator control implement that solution. This is illustrated by the
system for an office building. All of the participants in ‘waterfall’ line in Figure 1.
the study were experienced and expert integrated-
This is the pattern of thinking that everyone attempts
circuit designers, but they had never worked on eleva-
to follow when they are faced with a problem, and it is
tor systems before. Indeed, their only experience with
widely understood that the more complex the prob-
elevator systems came from riding in elevators. Each
lem is, the more important it is to follow this orderly
participant was asked to think out loud while they
flow. If you work in a large organization, you will
worked on the problem. The sessions were videotaped
recognize this linear pattern as being enshrined in
and analyzed in great detail.
policy manuals, textbooks, internal standards for

© 2001-2008 CogNexus Institute http://www.cognexus.org

 Wicked Problems and Social Complexity

project management, and even the most advanced elevators down to the first floor. But then what
tools and methods being used and taught in the happens in an emergency?” In other words, what is
organization. In the software industry it is known as driving the flow of thought is some marvelous inter-
the ‘waterfall model,’ because it suggests the image of a nal drive to make the most headway possible, regard-
waterfall as the project ‘flows’ down the steps towards less of where the headway happens, by making
completion. opportunity-driven leaps in the focus of attention. It
is precisely because these expert designers are being
creative and because they are learning rapidly that the
trace of their thinking pattern is full of unpredictable
leaps.

In particular, the experiment showed that, faced with

a novel and complex problem, human beings do not
simply start by gathering and analyzing data about the
problem. Cognition does not naturally form a pure
and abstract understanding of ‘the problem.’ The
Figure 1: Traditional wisdom for solving complex
subjects in the elevator experiment jumped immedi-
problems: the ‘waterfall’
ately into thinking about what kind of processors to
use in the elevator controller, and how to connect
However, the subjects in the elevator experiment did
them, and how to deal with unexpected situations,
not follow a waterfall. They would start by trying to
such as if one processor failed. These are detailed
understand the problem, but they would immediately
solution elements.
jump into formulating potential solutions. Then they
would jump back up to refining their understanding These experienced designers illustrated that problem
of the problem. Rather than being orderly and linear, understanding can only come from creating possible
the line plotting the course of their thinking looks solutions and considering how they might work.
more like a seismograph for a major earthquake, as Indeed, the problem often can best be described in
illustrated in Figure 2. We will refer to this jagged-line terms of solution elements. A requirement in the
pattern as opportunity-driven, because in each problem statement calling for ‘high reliability’ was
moment the designers are seeking the best opportu- quickly translated into the idea of using a network of
nity for progress toward a solution. distributed processors – a high-level solution that
drove the rest of the design process.
These designers are not being irrational. They are not
poorly trained or inexperienced. Their thought
process was something like: “Let’s see, idle elevators
should return to the first floor, but then, you only
need one elevator on the first floor, so the others
could move to an even distribution among the floors.
But the elevators need to be vacuumed regularly. I
suppose we could add a switch that brought idle

Page 5
 ment is significant because it gives us a real picture of
the process that people follow when they really think
about novel problems, and it is not the orderly and
linear process we have been taught is proper!

From another perspective, the jagged line of opportu-

nity-driven problem solving is a picture of learning.
The more novel the problem, the more the problem
solving process involves learning about the problem
Figure 2: Pattern of cognitive activity of one designer - domain. In this sense the waterfall is a picture of
the “jagged” line already knowing – you already know about the prob-
lem and its domain, you know about the right process
Figure 2 illustrates another striking observation: and tools to solve it, and you know what a solution
problem understanding continues to evolve until the will look like. As much as we might wish it were
very end of the experiment. Even late in the experi- otherwise, most projects in the knowledge economy
ments the designer subjects returned to problem operate much more in the realm of learning than
understanding, the upper part of the graph. Our already knowing. You still have experts, but it’s no
experience in observing individuals and groups longer possible for them to guide the project down
working on design and planning problems is that, the linear waterfall process. In the current business
indeed, their understanding of the problem continues environment, problem solving and learning are tightly
to evolve -- forever! Even well into the implementa- intertwined, and the flow of this learning process is
tion of the design or plan, the understanding of the opportunity-driven.
problem, the ‘real issue,’ is changing and growing.
Some readers might object to this claim. Perhaps most
The natural pattern of problem solving behavior may folks in their organization have a strong sense of
appear chaotic on the surface, but it is the chaos of an certainty about what is going on, a sense of confidence
earthquake or the breaking of an ocean wave – it and pride in their knowledge of their business, and a
reflects a deeper order in the cognitive process. The sense that the problems the business is confronted
non-linear pattern of activity that expert designers go with are quite manageable using the methodical
through gives us fresh insight into what is happening application of well known rules and linear process
when we are working on a complex and novel prob- logic. First, let me say, “Congratulations!” Certainly
lem. It reveals that the feeling that we are ‘wandering the modern economy is not all knowledge based, not
all over’ is not a mark of stupidity or lack of training. all problems are wicked, and there are many who still
This non-linear process is not a defect, but rather the enjoy a sense of quiet confidence and control in their
mark of an intelligent and creative learning process. professional lives. This book is not for them.

In fact, this non-linear pattern does not come as a If your organization is a professional or consulting
surprise to most people. Anyone who has ever worked services business, or if there is a large information
on a complex project has the intuition that this jagged component to your organization’s products or
line process is what is really going on. But the experi- business process, then you are all too familiar with the

© 2001-2008 CogNexus Institute http://www.cognexus.org

 Wicked Problems and Social Complexity

roller coaster ride of opportunity-driven problem potential solutions. Indeed, there is no definitive
solving. There are many reasons for this state of statement of ‘The Problem.’ The problem is ill struc-
affairs, but one of the most important is that you are tured, an evolving set of interlocking issues and con-
operating in the realm of a special kind of problem: straints. Rittel said, “One cannot understand the
the wicked problem. Wicked problems are one of the problem without knowing about its context; one
fragmenting forces mentioned at the beginning of this cannot meaningfully search for information without
chapter, and it essential to understand the properties the orientation of a solution concept; one cannot first
of wicked problems in order to counter and manage understand, then solve.” Moreover, what ‘the Prob-
their fragmenting impact on projects. lem’ is depends on who you ask – different stake-
holders have different views about what the problem
Wicked Problems is and what constitutes an acceptable solution.

The man who coined the term ‘wicked problem,’ 2. Wicked problems have
Horst Rittel, was also the inventor of the Issue-Based no stopping rule.
Information System (IBIS) structure upon which
Dialogue Mapping is based. Rittel and his colleagues Since there is no definitive ‘The Problem’, there is also
perceived the limitations of the linear ‘systems no definitive ‘The Solution.’ The problem solving
approach’ of design and planning over 30 years ago, process ends when you run out of resources, such as
and their research provides a foundation for what time, money, or energy, not when some optimal or
Rittel termed a ‘second generation’ of systems analysis ‘final and correct’ solution emerges. Herb Simon,
methodology. Rittel invented IBIS because, as an Nobel laureate in economics, called this ‘satisficing’ --
urban planner and designer, he found traditional stopping when you have a solution that is ‘good
planning methods inadequate for the ill-structured enough’ (Simon 1969)
problems he encountered in city planning.
3. Solutions to wicked problems are
Rittel’s genius shines especially bright when we con- not right or wrong.
sider his solution for wicked problems: IBIS, a struc-
ture for rational dialogue among a set of diverse They are simply ‘better,’ ‘worse,’ ‘good enough,’ or
stakeholders. This is a perspective that puts human ‘not good enough.’ With wicked problems, the
relationships and social interactions at the center, a determination of solution quality is not objective and
perspective that is only now coming into vogue as a cannot be derived from following a formula. Solutions
key insight of post-modern thought. are assessed in a social context in which “many parties
are equally equipped, interested, and/or entitled to
As Rittel defined them, wicked problems are distin- judge [them],” and these judgements are likely to vary
guished by the following characteristics: widely and depend on the stakeholder’s independent
values and goals.
1. You don’t understand the problem
until you have developed a solution. 4. Every wicked problem is essentially
unique and novel.
Every solution that is offered exposes new aspects of
the problem, requiring further adjustments of the

Page 7
There are so many factors and conditions, all embed- • Whether to route the highway through our city or
ded in a dynamic social context, that no two wicked around it?
problems are alike, and the solutions to them will • How to deal with crime and violence in our
always be custom designed and fitted. Rittel: “The schools?
condition in a city constructing a subway may look • What to do when oil resources run out?
similar to the conditions in San Francisco, say, . . • What should our mission statement be?
differences in commuter habits or residential patterns • What features should be in our new product?
may far outweigh similarities in subway layout, down-
town layout, and the rest.” Over time one acquires While many of the problems that we will look at in
wisdom and experience about the approach to wicked this chapter are problems that occur in organizations,
problems, but one is always a beginner in the specifics the above list should make it clear that many of the
of a new wicked problem. social problems that we face in our communities are
also ‘wicked problems’.
5. Every solution to a wicked problem
is a ‘one-shot operation.’ Wicked Problem Example:
A New Car Design
Every attempt has consequences. As Rittel says, “One
cannot build a freeway to see how it works.” This is Let’s consider a potentially wicked problem in the
the “Catch 22” about wicked problems: you can’t design of a new car. Let’s imagine a project team that
learn about the problem without trying solutions, but has formed around a new assignment: the Marketing
every solution you try is expensive and has lasting department is asking for a design that emphasizes
unintended consequences which are likely to spawn side-impact safety – they want to promote a new ‘safe
new wicked problems. car’ to compete with Volvo. That is the problem to be
solved, that is the work of the project. There is a
6. Wicked problems have no given alterna- deadline and a budget and a senior executive that the
tive solutions. project reports to.
There may be no solutions, or there may be a host of Now consider the criteria for a wicked problem again:
potential solutions that are devised, and another host
that are never even thought of. Thus, it is a matter of 1. You don’t understand the problem until you have
creativity to devise potential solutions, and a matter of developed a solution. One approach to making a safer
judgement to determine which are valid, which car would be to add structural support in the doors to
should be pursued and implemented. make the car safer from side impact. It turns out that
the additional door structure doubles the cost of the
These criteria are more descriptive than definitional. door, makes the doors heavier and harder to open and
The point is not so much to be able to determine if a close, changes the fuel mileage and ride, and requires
given problem is wicked or not as to have a sense of an adjustment to the suspension and braking systems.
what contributes to the ‘wickedness’ of a problem. Making the doors stronger leads into other design
problems, but also bounces back into marketing
Here are a few examples of wicked problems:
problems such as “What should the price be?”, “How

© 2001-2008 CogNexus Institute http://www.cognexus.org

 Wicked Problems and Social Complexity

much do people really care about side impact surviv- 5. Every solution to a wicked problem is a ‘one-shot
ability?”, “What do customers really want in a car?” operation’. The creation of a safer car is a one-shot
All of these problems interact with each other. And at operation. When the new safer car finally reaches the
the senior executive level, the real question is “Should market, it may be a flop, or it may change the safety
we continue this project to produce this new car?” standards for the whole industry. The design team can
build prototypes of the car and test them, but there is
2. Wicked problems have no stopping rule. When no way to anticipate the unintended consequences of
does the car become ‘safe’? There is no natural stop- producing and selling the new vehicle.
ping point in working out the tradeoffs among safety,
performance, appearance, and cost. At some point, 6. Wicked problems have no given alternative solu-
the design team will be forced to make a decision. If it tions. The safe door problem does not have a few
were not for project deadlines, the team would swirl discrete possible solutions from which to choose.
indefinitely in ‘analysis paralysis’. There is an immense space of options in terms of
structural reinforcement, materials, cushioning,
3. Solutions to wicked problems are not right or window design, hinge placement, and how the door
wrong. No amount of study, laboratory experiments, latches and opens. The design team cannot select
or market surveys will establish that that project from a few options – it must collectively exercise
team’s solution is ‘correct.’ Ironically, when the car creativity and judgement about an elegant resolution
gets produced, there will be reviews pointing out that of all of the design priorities.
the doors are heavy and difficult to open when
parking on a hill, mixed with law suits from people The design of a new ‘safe car’ is an example of a
who were injured in side-impact accidents despite the wicked problem. It cannot be solved by engineers
stronger doors. alone, nor is there any way of determining that any
given solution is ‘correct’ or even optimal. It all
4. Every wicked problem is essentially unique and depends on where you stand.
novel. Even if the project team has several successful
car designs under its belt, the ‘safe door’ problem is Coping with Wicked Problems
essentially unique and novel, because of the configura-
tion of issues and stakeholders. First, a recent study by Not all problems are wicked. In contrast, a ‘tame
a consumer safety organization suggests that side problem’ is one for which the traditional linear proc-
impact injuries would be reduced by side air bags, ess is sufficient to produce a workable solution in an
which are not a part of the design. Second, a side- acceptable time frame. A tame problem:
impact injury lawsuit has been filed against the com-
1. has a well-defined and stable problem statement,
pany – if the new design is announced now, it may
2. has a definite stopping point, i.e. when the solu-
look like an acknowledgement of prior unsafe designs.
tion is reached,
Moreover, federal legislation is emerging that may put
3. has a solution which can be objectively evaluated
legal constraints on the strength of the doors. The
as right or wrong,
design of safer doors is not merely a technical prob-
4. belongs to a class of similar problems which are all
lem: it is a political and PR problem as well.
solved in the same similar way,

Page 9
5. has solutions which can be easily tried and approach they require. There is a psychological
abandoned, dimension here – a shift from denial to acceptance.
6. comes with a limited set of alternative solutions.
The command and control paradigm of management
Finding the square root of 7358 is a tame problem, as reinforces blindness about the true nature of the
is finding the shortest route from A to B on a map. problem. Inherent in this paradigm is the idea that a
Repairing a computer, raising $10,000, and selecting a person in charge gives the solution (the right solution,
new doctor when you move to a new city are all tame, the only solution) to other people, who are in charge
if complex and difficult, problems. Tame does not of implementing it. To function in such a hierarchy
mean simple – a tame problem can be very technically often means to collude in systematic denial of the
complex. complex and ill-structured dynamics of wicked prob-
lems, a phenomenon dubbed ‘skilled incompetence’
A problem doesn’t have to possess all six characteris- by Chris Argyris (e.g. Argyris and Schön 1996).
tics in order to be wicked. Putting a man on the moon
was a problem with a lot of wickedness, for example, As a result, there are two common organizational
but also with some tame elements. There were cer- coping mechanisms that are routinely applied to
tainly some wicked sub-problems. But notice that the wicked problems: studying the problem, and taming
main problem statement, putting a man on the moon it.
and returning him safely, did not change over time
While studying a novel and complex problem is
(criterion 1). There was a definite ‘stopping point’ at
natural and important, it is an approach that will run
which we could say we had solved that problem
out of gas quickly if the problem is wicked. Pure study
(criterion 2). And the solution could be clearly evalu-
amounts to procrastination, because little can be
ated as having succeeded or failed (criterion 3). It may
learned about a wicked problem by objective data
be convenient to describe a problem as wicked or
gathering and analysis. Wicked problems demand an
tame, but it’s not binary – most problems have
opportunity-driven approach; they require making
degrees of wickedness.
decisions, doing experiments, launching pilot pro-
You also can’t tell from the outside if a problem is grams, testing prototypes, and so on. Study alone leads
going to be wicked. Like the safe car design example, to more study, and results in the condition known as
many problems appear tame on the surface, but are ‘analysis paralysis,’ a Catch 22 in which we can’t take
indeed wicked once you get into them. There seems to action until we have more information, but we can’t
be a natural inclination to see problems as tame, and get more information until someone takes action.
to avoid the wicked ones. Who wants to take on a One corporation I worked with, struggling to decide
problem that, by definition, can’t be solved!? between two very different strategic paths for the
future, studied and discussed the two options for so
The first step in coping with a wicked problem is to long that, by the time they had decided and imple-
recognize its nature. There is a tendency to treat all mented their choice, the chosen option was no longer
problems as tame, perhaps because tame problems are viable.
easier to solve, reinforced by the lack of understanding
about wicked problem dynamics and the tools and

© 2001-2008 CogNexus Institute http://www.cognexus.org

 Wicked Problems and Social Complexity

Taming a wicked problem is a very natural and com- number of deaths and injuries on school property – if
mon way of coping with it. Instead of dealing with the this measure drops to zero, then we have solved the
full wickedness of the problem, one simplifies it in problem. This taming approach amounts to locking
various ways to make it more manageable – to make it the problem down (point 1), however, because what is
solvable! There are (at least) six ways to tame wicked measured becomes, officially and by definition, the
problems, corresponding to the six criteria for wick- problem. Whatever is not measured is then free to
edness: absorb the real problem. With intense enough focus,
we might reduce the number of violent incidents on
1. Lock down the problem definition. Develop a the school grounds to zero … problem solved! … but
description of a related problem or a sub-problem that overlook new problems that had been created, such as
you can solve, and declare that to be the problem. a sharp rise in violent incidents just off of the school
Resist all efforts to expand or modify the problem grounds.
definition. For example, if the problem is how to
reduce violence in schools, you could focus on the 4. Cast the problem as ‘just like’ a previous problem
much more tractable problem of how to install metal that has been solved. Ignore or filter out evidence that
detectors in all school entrances. As another example, complicates the picture. Refer to the previous solution
in the software field, one learns to ‘freeze the of the related problem: “It’s just like that problem.
requirements’ as a way to lock down the problem. Just do the same thing again.” For example, there is a
saying in military circles that “we always fight the last
2. Assert that the problem is solved. Since a wicked war,” meaning the tendency to assume that the enemy
problem has no definitive solution, the whole point of will behave as he did in the last war.
attempting to tame it is so that a solution can be
reached. Usually this step requires locking the prob- 5. Give up on trying to get a good solution to the
lem down (see point 1), though it is possible to simply problem. Just follow orders, do your job, and try not
assert that the problem is ‘solved’ without clarity to get in trouble. Maybe the organization will fix the
about what the problem was. Such assertions, how- serious problems in the current solution in a revised
ever, generally require considerable authority to version or release next year.
appear successful, such as in an autocratic organiza-
tion or a dictatorship. As an example illustrates, one 6. Declare that there are just a few possible solutions,
way of dealing with a United Nations resolution and focus on selecting from among these options. A
demanding that you destroy all weapons of mass specific way to do this is to frame the problem in
destruction in your country is to simply assert that ‘either/or’ terms, e.g. “Should we attack Iraq or let the
you have done so. It should be clear that this approach terrorists take over the world?”
to taming a problem depends critically on the
Different people prefer different coping mechanisms
evidence that you offer that the problem is solved.
– some would rather study the problem until they
3. Specify objective parameters by which to measure really understand it; others, impatient with sitting
the solution’s success. This is the measurement around, would rather tame the problem to something
approach. For example, to find out if we have solved manageable and jump into action.
the problem of school violence, we might count the

Page 11
However, attempting to tame a wicked problem, ‘we all pretty much think and act the same way’ just
while appealing in the short run, fails in the long run. doesn’t hold any more. In addition, organizations are
The wicked problem simply reasserts itself, perhaps in flattening, opening up, and moving toward increased
a different guise, as if nothing had been done. Or, workplace democracy. More disciplines, departments,
worse, sometimes the tame solution exacerbates the and dogmas are represented on the typical project
problem. team.

Social Complexity The jagged line graph from the MCC elevator study
can help us visualize the impact of social complexity
At the beginning of the chapter we asserted that the on a project. Imagine adding a second designer, repre-
two most intensely fragmenting forces impacting sented by the dotted orange line in Figure 3, to help
projects today are wicked problems and social com- solve the elevator design problem.
plexity. These forces tend to co-occur. Can a socially
complex group have a tame problem? Probably so.
Can an individual have a wicked problem? Certainly.
Yet the concepts are distinct: while wickedness is a
property of the problem/solution space and the
cognitive dynamics of exploring that space, social
complexity is a property of the social network that is
engaging with the problem.

Social complexity is a function of the number and

diversity of players who are involved in a project. The
more parties involved in a collaboration, the more Figure 3: A wicked project with a second designer
socially complex it is. The more different those parties working on the problem
are, the more socially complex.
Notice that the second designer, like the first designer,
Projects and problem solving have always been social goes through an opportunity-driven process between
in nature. Project success has always depended on the problem and solution spaces, but the new
collaborative skills and collective intelligence. But in designer’s thinking process is quite different in the
earlier times of greater social homogeneity, the particulars from the first designer’s. Since she has a
collaborative skills picked up on the playground were different background and training, the pattern of her
sufficient. The rules of engagement of family dynam- cognitive flow will also differ.
ics held in project meetings, and hierarchical author-
ity could always be used to sort out the hardest parts. Let’s imagine you are the project leader. You are the
Now, in the ‘knowledge workforce’, more democratic one who is responsible for the project being on time,
models of decision-making are being used. Also, in budget, and meeting all of its requirements. Even if
women have a far stronger role, often playing leader- you understand that the process is going to be oppor-
ship roles. Minorities and foreign nationals are almost tunity-driven, you must still make plans, create sched-
always present on the team. The old assumption that ules, allocate resources, and commit to milestones.

© 2001-2008 CogNexus Institute http://www.cognexus.org

 Wicked Problems and Social Complexity

You can’t ‘plan’ for the process to be opportunity- to done. I was with the client yesterday, and it turns
driven! Thus, in effect, you are officially in charge of out that there is a transaction that the system needs to
keeping the project on the waterfall line. handle that they never even told us about. They said it
didn’t have anything to do with our system. But it
Now let’s consider two project team meetings, occur- turns out it has a lot to do with our system. We’ve got
ring at different points along the time line. At meet- to back to square one and start over!”
ing A in Figure 4, you are a happy project leader
because everyone is in synch, focused on the same Neither of these key players is where you need them
activity, analyzing the requirements just like it says in to be, according to the linear plan you created at the
‘the book.’ Your prospects for bringing your project in beginning of the project. You can feel chaos rising and
on time and in budget look good. control slipping away. You desperately plead with
them to refocus on the high-level design, because,
according to the calendar, that’s where the project
needs to be.

Perhaps you turn to the first designer and say some-

thing like “That’s a good idea, Henry – but we really
need to finish the high-level design. Can you hang on
to that code for a while?” Turning to the other
designer, you beg, “Look, Sally, we already have gotten
those requirements signed off. We can’t go back.
Figure 4: Two team meetings, A and B, during the We’ll just have to take care of that new transaction in
project. the next release of the system.”

This scenario exemplifies a tiny slice of the tension

Some time later, at meeting B in Figure 4, the team
that is inherently part of working in a socially com-
has finished up with the data analysis and is now in
plex environment. Despite the most carefully
the next phase, high-level design. But there are signs of
thought-out plans, wicked problem dynamics and the
trouble. Designer 2 looks tired but radiant. He says, “I
different jobs and orientations of the two fragment
was driving home last night and I had an idea. I stayed
the project team and its process. The above scenario is
up all night programming, and … you won’t believe
mild – there are only three stakeholders involved in
this, but … I put together a program that does the
the project. As projects grow in size and organizations
whole thing. Sure, it still needs a little work, but, hey,
grow flatter, social complexity increases. Large
we’re practically done! Way ahead of schedule! I can’t
projects typically have dozens of stakeholders, repre-
wait to show it to you!” In his personal opportunity-
senting the project team, other departments, and
driven process, he has made a major leap, all the way
other organizations. And not only does each one have
to bottom of the chart, to the final solution.
their own ‘jagged line,’ they are likely to have different
There is a long pause. Designer 1 also looks tired, but ideas about what the real issues are, and what the
not so radiant. Holding up the long-finished require- criteria for success are.
ments document she says, “Sorry, we’re not even close

Page 13
Consider the safe car design team. Bob, from speaking a different language. Thus, achieving shared
Marketing, has been conducting studies and focus meaning and shared context is especially difficult.
groups that indicate a lot of interest in cars that are
safer in a collision. He is concerned with how to pack- Moreover, social complexity goes beyond individual
age a new ‘safe car’ in a way that is positive, sexy, and diversity and diversity among disciplines. The real
up-beat. Christine, from Engineering, is very con- corker is that these players represent different organi-
cerned about making the doors too heavy, but she has zations. Each organization has its own function and
worked on structural integrity in the past and is charter, its own goals, and is managed by its own
excited about new technologies that, while expensive, executive director. These organizations often have
could make the doors both stronger and lighter. divergent goals. Marketing is trying to make its sales
Harry, the representative from the Management numbers, while Engineering is trying to win the
Team, sees the big issue as cost – top management is Baldridge quality award. When the members of a
pushing affordability and value as the new strategy to project team come together to collaborate, they repre-
increase sales. Alan, from IT, has a mandate from his sent not only themselves but also their respective
management to get this team to use the new CAD management chain in the hierarchy. Ideally, everyone
(Computer Aided Design) system on this project. in the organization is committed to the same thing,
There are team members who represent Regulatory but operationally goals and agendas can be quite
Affairs, Finance, Graphic Design, Power Train, and fragmented.
Quality Assurance, as well as team members from
Thus, social complexity makes wicked problems even
several major suppliers, including electronics and
more wicked, raising the bar of collaborative success
interior materials.
higher than ever.
Each player has their own individual experience, per-
Recall that the main feature of a wicked problem is
sonality type, and style of thinking and learning. Each
that you don’t understand the problem until you have
player adds a new jagged line to the graph. The indi-
a solution. But with social complexity, ‘not under-
vidual diversity among these players will make collec-
standing the problem’ does not show up as innocent
tive intelligence a challenge, and will make consensus
wonder about the mystery of the problem, nor does it
virtually impossible to achieve.
usually occur as a thoughtful collective inquiry into
But social complexity doesn’t stop with individual the deeper nature of the problem.
diversity – each of these players comes from a differ-
Rather, ‘not understanding the problem’ shows up as
ent discipline, with its own specialized language and
different stakeholders who are certain that their ver-
culture. When Bob is among his colleagues in
sion of the problem is correct. In severe cases, such as
Marketing, they share common knowledge, a com-
many political situations, each stakeholder’s position
mon set of concerns, and shared ways of thinking
about what the problem is reflects the mission and
about and dealing with those concerns. However,
objectives of the organization (or region) they repre-
when Bob tries to talk to Christine, from Engineer-
sent. In such cases there is a fine line between collabo-
ing, he finds that she has little knowledge of basic
ration and colluding with the enemy. How can you
marketing concepts, and seems to be uninterested in
them. It’s as if she were from a different country,

© 2001-2008 CogNexus Institute http://www.cognexus.org

 Wicked Problems and Social Complexity

make headway on a mutually acceptable solution if Virtually all creative work is a process of design. To
the stakeholders cannot agree on what the problem is? design simply means ‘to formulate a plan for,’ ‘to plan
out in systematic, usually graphic form,’ and ‘to create
The answer to this question – and the Holy Grail of or contrive for a particular purpose or effect’ (Ameri-
effective collaboration – is in creating shared under- can Heritage Dictionary, 2000). All problems call for
standing about the problem, and shared commitment designing a solution. All projects are essentially
to the possible solutions. Shared understanding does designing something. Design, in both the technical
not mean we necessarily agree on the problem, and artistic sense, is the process of creating something
although that is a good thing when it happens. Shared new – e.g., a new car, a strategic plan, a software pro-
understanding means that the stakeholders under- gram, a stickier web site, next year’s budget, a new
stand each other’s positions well enough to have environmental policy.
intelligent dialogue about the different interpreta-
tions of the problem, and to exercise collective intelli- Any design problem is a problem of resolving tension
gence about how to solve it. between what is needed and what can be done. On
the one hand, the process of design is driven by some
Because of social complexity, solving a wicked desire or need – someone wants or needs something
problem is fundamentally a social process. new. The need might be expressed by a customer, or it
Having a few brilliant people or the latest may be a guess about what the market wants. The
project management technology is no longer need or want is expressed in the language of what
sufficient. ought to be – what should be done, what should be
built, what should be written. On the other hand, the
This book offers a practical approach for creating
process of design is constrained by resources – what
shared understanding and shared commitment in a
can be done given the available resources such as time
complex social network, and explores the underlying
and money and the constraints imposed by the envi-
principles that make this approach so effective. But
ronment and the laws of science.
before we can get into the ‘solution’ offered in this
book, there is a little bit more to understand about the Every need has a price tag – the process of design is
collaborative challenge posed by wicked problems and about devising solutions that are feasible and cost
social complexity. effective. Going back to the safe car design, the need
might be quite specific, e.g., the car must protect the
The Polarity of Design
occupants from harm if it is struck from the side by
Most projects wrestle with large social networks and another vehicle of similar weight traveling 30 miles an
their attendant complexity. It would be a mistake, hour. It may turn out that such a car would cost twice
however, to think that small project teams can escape as much as a normally safe car. It may turn out to be
fragmentation. Design possesses a fundamental prop- impossible at any cost. Perhaps we have to change the
erty that can make a team of two socially complex. All need: reduce the required speed of safe impact to 10
that is needed is a representative from each of the two miles an hour, because then it only increases the cost
polarities of design: what is needed (marketing), and of the car by 15%.
what can be built (engineering).

Page 15
Thus, in a very basic way, every project is about recon- The world of What-Is-Needed is the domain of the
ciling the fundamental polarity between the world of Marketing and Sales department, and sometimes
What-Is-Needed and the world of What-Can-Be- upper Management, whereas the world of What-Can-
Built. These two worlds correspond to the upper and Be-Built is territory that belongs to the Engineering
lower halves of the MCC elevator study diagram. In (or Manufacturing, Software Development, IT, etc.)
Figure 5, the upper half, being about understanding department.
the problem, is focused out in the world on a specific
client or user or market. There is always someone who The inherent unity of the design process turns into a
has a need or a desire, and the task in the Problem or battle between departments. The world of What-Is-
What-Is-Needed aspect of design is to specify that Needed, claimed by the Marketing department,
need. The lower half, being about the Solution, is becomes a self-referential world with its own culture
focused ‘in the shop’ on What-Can-Be-Built – what and customs and language. The world of What-Can-
do we have the resources and skills and tools to Be-Built is claimed by the technologists, the nerds and
actually make, and what will it cost and how long will hackers who actually build things, with its own
it take. culture and customs and language. When they sit
down together on a project, the polarity of design
turns into an inter-cultural war that is expensive,
wasteful and ineffective, a war frequently featured in
Scott Adams “Dilbert” cartoons.

Thus social complexity is not just a function of the

number stakeholders – it is also a function of struc-
tural relationships among the stakeholders. While
large projects have an increasing number and diversity
of stakeholders, it only takes one player from each side
of the polarity of design – one from Marketing and
Figure 5: The two parts of the world of design
one from Engineering – to cause the collaborative
As you can see, there is an immense difference gears to grind to a halt.
between these two worlds. When an individ-
ual does design, she stands with one foot in Technical Complexity
each world. Moving back and forth between In addition to wicked problems and social complexity,
the two worlds, she tries to create a solution technical complexity is a potentially fragmenting
that joins the two polarities of design in an force. It includes the number of technologies that are
elegant way. involved in a project, the immense number of possible
interactions among them, and the rate of technical
Design teams have a bigger challenge. While it is pos-
change. For example, to be a serious player in the
sible for each person on a project team to be standing
software industry today, your software must run on a
in both worlds, the tendency is for the polarity of
variety of types of computers. Each type (or ‘plat-
design to be reflected in a polarity of roles.
form’) has several operating systems, and each oper-

© 2001-2008 CogNexus Institute http://www.cognexus.org

 Wicked Problems and Social Complexity

ating system has many versions that are currently in Moreover, the physics of fragmentation is obscured by
the field and must be supported. You must choose a cultural condition of resignation, denial, and grim
what language your software will be written in: Java, determination that has grown up around it. In my
C, C++, Cobol, Fortran, etc. Each of these pro- consulting and facilitation experience I have met this
gramming languages has a variety of supported ver- condition in organizations and on project teams. I
sions (compilers); for example, Microsoft and Sun have seen it manifest in many forms, sometimes as
each have a major version of the popular Java language outright panic, sometimes as plodding determination,
used in World Wide Web applications. Then you sometimes as a vague sense of futility. This condition
must choose the set of utilities (‘library’) you will use of organizational pain is so chronic, however, that,
for creating your user interface. There are dozens of like low-grade back pain, it has faded into the back-
other choices, and all of these options interact with ground of organizational experience and is taken for
each other. Moreover, the field is changing so fast that granted, assumed to be normal and inevitable.
new options become available, and others drop into
oblivion, almost every day. The condition is not wicked problems, nor social
complexity – these are causes of the condition. Once
As much as technical complexity raises the risk of this chronic condition is seen and understood, in my
project failure, it is also the most well recognized experience, a huge compassion emerges for what we
fragmenting force. So much has been written about are up against when we go to work. For what we do
technical complexity and how to deal with it, so many accomplish and the courage that it takes. A whole new
tools and methods are available, that there is little to perspective about work and life opens up.
add here. The Dialogue Mapping approach presented
in this book excels at dealing with complex technical This is why it’s so useful to distinguish the common
element of fragmentation. Wicked problems frag-
information, but the real power of Dialogue Mapping,
ment the process of project work, especially when the
and the point of this book, is to provide an approach
problem is misdiagnosed as tame. Wicked problems
and a set of tools for dealing with the nontechnical
also fragment direction and mission – if you can’t
side of fragmentation: wicked problem dynamics and agree on what the problem is, how can you be aligned
social complexity. on a solution? Social complexity fragments team iden-
tity – the ideal of team unity is compromised by the
Fragmentation and Coherence dynamics of competing interests and hidden agendas.
The duality of design tends to divide allegiances
We have described wicked problems, social complex-
between requirements and implementation. Social
ity, and technical complexity as forces that fragment
complexity also fragments meaning – key terms and
projects, causing them to fail. It is important to rec-
concepts are used in different ways by the different
ognize that these forces are not due to incompetence, stakeholders. Project teams are often geographically
poor management, or any human failing. They are distributed, further fragmenting relationships and
part of the ‘physics’ of projects. There is no quick fix communications. Participants in a modern project
for the phenomenon of wicked problems. No glib team are pulled in a thousand different directions by
formula about “Seven Steps to Crush Social Com- the centrifugal forces of wicked problems, social com-
plexity” or “Tame Your Way to the Top”. plexity, and technical complexity (see Figure 6).

Page 17
 and create chaos. Without understanding the
cause, there is finger-pointing instead of
learning.

Not so long ago most human illness was regarded as

the result of evil spirits, so, when people got sick, the
fix was to let the evil spirits out, by, for example,
drilling holes in their heads. It wasn’t very effective,
but – within that system of thought – it was rational.
Figure 6: The ‘centrifugal’ fragmenting forces These days, when big projects run into problems, we
pulling a project apart hold emergency meetings, then fire the consultants or
rearrange the org chart. It isn’t very effective, but it’s a
The notion of fragmentation points to all of these
rational response to fragmentation … if you believe
problems, but it is pretty abstract. Because it points
that the problems result from human failing, i.e. poor
deep into the culture and practices of project work, it
performance or incompetence.
is difficult to observe fragmentation directly. There is,
however, a more observable indicator of fragmenta- I was doing some training with a management team at
tion: blame. Instead of seeing the systemic nature of a utility company several years ago. The Human
project challenges and the value of social diversity, we Resources (HR) department had recently announced
tend to see a big mess, to view it as the result of a new policy regarding the way employee performance
incompetence, and to blame each other for it. We would be evaluated and reported in the future, and
blame upper management for sending mixed signals these managers were very upset because the policy was
or for lack of direction. We blame HR for poor hiring so obviously flawed, and it had a direct impact on
practices and lack of training. We blame the ‘bean them. “What were they thinking?!?” and “Those
counters’ for over-tight budgets and lack of fiscal morons in HR!” they exclaimed. As an exercise we
flexibility. We blame IT for confusion and the lack of decided to design a better policy. After an hour and a
stable infrastructure. We blame our customers for not half we reviewed our solutions, and what do you
knowing what they really want. We blame each other suppose they realized? That it was a very hard prob-
because we have different personalities and learning lem, given the organizational and legal constraints in
styles. (How many conversations do you notice in the system, and that, all things considered, HR had
your organization that involve placing blame?) come up with a pretty good approach! They shifted
from blame to deeper understanding of the problem.
In times of stress the natural human tendency
is to find fault with someone else. We tend to If we step back and take a systemic view, we can see
take the problem personally, at an organiza- that the issue is not whose fault the mess is – the issue
tional level, and assume that the chaos we see is our collective failure to recognize the recurring and
is a result of incompetence or, worse, insincere inevitable dynamics of the mess. If we take a systemic
leadership. Since our education and experi- view, we turn away from blame and away from easy
ence have prepared us to see and solve tame technical fixes, and to look in the social domain – in
problems, wicked problems sneak up on us

© 2001-2008 CogNexus Institute http://www.cognexus.org

 Wicked Problems and Social Complexity

building capacity to collaborate effectively on wicked Summary

problems.
This chapter has been about laying a foundation that
As Rittel said, “We are now sensitized to the waves of identifies the ‘problem’ which Dialogue Mapping
repercussions generated by a problem-solving decision addresses. This problem is:
directed to any one node in the network, and we are
The powerful fragmenting forces of wicked problems,
no longer surprised to find it inducing problems of
social complexity, and technical complexity;
greater severity at some other node.” The confusion, chaos, and blame created by failing to
distinguish these forces;
The antidote for fragmentation is coherence. How,
The lack of tools and techniques for ‘defragmenting’
then, do we create coherence? In organizations and
project dynamics.
project teams – in situations where collaboration is
the life blood of success – coherence amounts to The process of Dialogue Mapping is a powerful
shared understanding and shared commitment. approach for addressing the problem of fragmenta-
tion, as it allows a diverse group of people to generate
Shared understanding of meaning and context, and of coherence around wicked problems. This group
the dimensions and issues in the problem. Shared coherence is a necessary step toward addressing frag-
commitment to the processes of project work and to mentation, yet it is neither a silver bullet nor a cure-
the emergent solution matrix. all. Given the complex nature of organizations, it is
not sufficient for a single team or even multiple teams
Coherence means that stakeholders have shared to achieve coherence; the organization as a whole
meaning for key terms and concepts, that they are needs to become a knowledge organization, and gain a
clear about their role in the effort, that together they kind of ‘literacy’ or ‘fluency’ in the language of coher-
have a shared understanding of the background for ence: distinctions, tools, methods, and practices for
the project and what the issues are, and that they have crafting shared understanding and shared commit-
a shared commitment to how the project will reach its ment. Dialogue Mapping is a first step toward that
objectives and achieve success. Coherence means that kind of literacy.
the project team understands and is aligned with the
goals of the project and how to reach them. Coher-
Acknowledgements
ence means that a wicked problem is recognized as Many people have contributed to this paper over its
such, and appropriate tools and processes are con- years of evolution. I am immensely grateful to Horst
stantly used to ‘defragment’ the project. With Rittel for his groundbreaking work in this area—this
increased coherence, more collective intelligence paper is dedicated to him. William Weil co-authored
becomes available to deal with change and complexity. earlier versions of this paper, and his contribution still
Coherence means that despite social complexity there shines through in the current version. Kim Salins has
is a sense of ability and confidence in crafting shared been a stalwart ally for the past few years of evolution
understanding and negotiating shared meaning. of this work. I’m also grateful for constructive reviews
from Bill Pardee, Rosa Zubizarreta, Jack Park, and
Eugene Eric Kim.

Page 19
References 159. Also Reprint No. 86, The Institute of Urban and
Regional Development, University of California,
American Heritage® Dictionary of the English Berkeley, California, Tel: (510) 642-4874, email:
Language, 4th edn (2000), Houghton Mifflin, [email protected].
Boston, MA.
Rittel, Horst, and Douglas Noble (1989) “Issue-
Argyris, Chris, and Donald Schön, Organizational Based Information Systems for Design,” Working
Learning II: Theory, Method, and Practice, Addison- Paper 492, The Institute of Urban and Regional
Wesley, Reading, Massachusetts. Development, University of California, Berkeley,
Guindon, Raymonde (1990) Designing the Design California, Tel: (510) 642-4874, email:
Process: Exploiting Opportunistic Thoughts. [email protected].
Human-Computer Interaction, Vol. 5, pp. 305-344. Simon, Herbert A. (1969) The Sciences of the
Kunz, Werner, and Horst Rittel (1970) “Issues as Artificial, Second Edition, MIT Press, Cambridge,
Elements of Information Systems,” Working Paper Mass.
131, The Institute of Urban and Regional
Development, University of California, Berkeley,
Original design of this white paper by Chris Conerly,
California, Tel: (510) 642-4874, email:
[email protected].
[email protected].
Rittel, Horst (1969) “Reflections on the Scientific
and Political Significance of Decision Theory,”
Working Paper 115, The Institute of Urban and
Regional Development, University of California,
Berkeley, California, Tel: (510) 642-4874, email:
[email protected].
Rittel, Horst (1972) “On the Planning Crisis:
Systems Analysis of the ’First and Second
Generations’,” Bedriftsøkonomen, Nr. 8. Also
Reprint No. 107, The Institute of Urban and
Regional Development, University of California,
Berkeley, California, Tel: (510) 642-4874, email:
[email protected].
Rittel, Horst (1972) “Structure and Usefulness of
Planning Information Systems,” Bedriftsøkonomen,
Nr. 8. Also Reprint No. 108, The Institute of Urban
and Regional Development, University of California,
Berkeley, California, Tel: (510) 642-4874, email:
[email protected].
Rittel, Horst and Melvin Webber (1973) “Dilemmas
in a General Theory of Planning,” Policy Sciences 4,
Elsevier Scientific Publishing, Amsterdam, pp. 155

© 2001-2008 CogNexus Institute http://www.cognexus.org