Chapter 1
INTRODUCTION AND OVERVIEW

Addressing complex systems such as health-care delivery, sustainable energy, financial systems, urban infrastructures, and national security requires knowledge and skills from many disciplines, including systems science and engineering, behavioral and social science, policy and political science, economics and finance, and so on. These disciplines have a wide variety of views of the essential phenomena underlying such complex systems. Great difficulties are frequently encountered when interdisciplinary teams attempt to bridge and integrate these often-disparate views.

This book is intended to be a valuable guide to all the disciplines involved in such endeavors. The central construct in this guide is the notion of phenomena, particularly the essential phenomena that different disciplines address in complex systems. Phenomena are observed or observable events or chains of events. Examples include the weather, climate change, traffic congestion, aggressive behaviors, and cultural compliance. A team asked to propose policies to address the problem of overly aggressive motorist behaviors during inclement weather in the evening rush hour might have to consider the full range of these phenomena.

Traditionally, such problems would be decomposed into their constituent phenomena, appropriate disciplines would each be assigned one piece of the puzzle, and each disciplinary team would return from their deliberations with insights into their assigned phenomena and possibly elements of solutions. This reductionist approach often leads to inferior solutions compared to what might be achieved with a more holistic approach that explicitly addresses the interactions among phenomena and central trade-offs underlying truly creative solutions. This book is intended to enable such holistic problem solving.

Five themes are woven throughout this book.

• Understanding the essential phenomena underlying the overall behaviors of complex systems and enterprises can enable improving these systems.
• These phenomena range from physical, behavioral, and organizational, to economic and social, all of which involve significant human components.
• Specific phenomena of interest and how they are represented depend on the questions of interest and the relevant domains or contexts.
• Visualization of phenomena and relationships among phenomena can provide the basis for understanding where deeper exploration is warranted.
• Mathematical and computational models, defined very broadly across disciplines, can enable the necessary deeper understanding.

This chapter proceeds as follows. We first consider the nature of a range of perspectives on systems. This begins with an exploration of historical perspectives, drawing upon several disciplines. We then consider the nature of complexity and complex systems. This leads to elaboration of the contrast between complex and complicated systems and the notion of complex adaptive systems. We then consider systems practice over the past century. This background is intended to provide a well-informed foundation that will enable digesting the material discussed in later chapters.

SYSTEMS PERSPECTIVES

It is useful to reflect on the roots of systems thinking. This section begins with a discussion of the systems movement. We then elaborate the philosophical underpinnings of systems thinking. Finally, we review a range of seminal concepts. Brief sketches of these concepts are presented here; they are elaborated in greater depth in later chapters.

Systems Movement

The systems movement emerged from the formalization of systems theory as an area of study during and following World War II, although it can be argued that the physicists and chemists of the 19th century contributed to the foundations of this area. Before delving into the ideas emerging in the 1940s and beyond, it is important to distinguish four aspects of the systems movement:

• Systems Thinking is the process of understanding how things influence one another within a whole and represents an approach to problem solving that views "problems" as components of an overall system.
• Systems Philosophy is the study of systems, with an emphasis on causality and design. The most fundamental property of any system is the arbitrary boundary that humans create to suit their own purposes.
• Systems Science is an interdisciplinary field that studies the characteristics of complex systems in nature and society, to develop interdisciplinary foundations, which are applicable in a variety of areas, such as engineering, biology, medicine, and economics.
• Systems Engineering is an interdisciplinary field focused on identifying how complex engineering undertakings should be designed, developed, and managed over their life cycles.

Contrasting these four aspects of systems, it is important to recognize that different disciplines tend to see "systems" quite differently, for the most part due to the varying contexts of interest (Adams et al., 2014). Thus, a systems scientist studying marsh ecosystems and a systems engineer designing and developing the next fighter aircraft will, from a practical perspective at least, have much less in common than the term "system" might lead one to expect. The key point is that systems exist in contexts and different contexts may (and do) involve quite disparate phenomena.

Philosophical Background

There are many interpretations of what system thinking means and the nature of systems thinkers. Some are inclined toward model-based deduction, while others are oriented toward data-driven inference. The former extol the deductive powers of Newton and Einstein, while the latter are enamored with the inferential capabilities of Darwin. These different perspectives reflect different epistemologies.

The study of epistemology involves the questions of what is knowledge, how can it be acquired, and what can be known. The empiricism branch of epistemology emphasizes the value of experience. The idealism branch sees knowledge as innate. The rationalism branch relies on reason. The constructivism branch seeks knowledge in terms of creation. These branches differ in terms of how they represent knowledge, in particular how this knowledge is best modeled and simulated (Tolk, 2013).

There are many possible ways of thinking about complex systems and enterprises (Rouse, 2005, 2007). Systems paradigms for representation of knowledge include hierarchical mappings, state equations, nonlinear mechanisms, and autonomous agents (Rouse, 2003). For hierarchical mappings, complexity is typically due to large numbers of interacting elements. With uncertain state equations, complexity is due to large numbers of interacting state variables and significant levels of uncertainty. Discontinuous, nonlinear mechanisms attribute complexity to departures from the expectations stemming from continuous, linear phenomena. Finally, autonomous agents generate complexity via the reactions of agents to each other's behavior and lead to emergent phenomena. The most appropriate choice among these representations depends on how the boundaries of the system of interest are defined (Robinson et al., 2011).

Horst Rittel argued that the choice of representation is particularly difficult for "wicked problems" (Rittel & Webber, 1973). There is no definitive formulation of a wicked problem. Wicked problems have no stopping rule - there is always a better solution, for example, "fair" taxation and "just" legal systems. Solutions to wicked problems are not true or false, but good or bad. There is no immediate or ultimate test of a solution to a wicked problem. Wicked problems are not amenable to trial-and-error solutions. There is no innumerable (or an exhaustively describable) set of potential solutions and permissible operations. Every wicked problem is essentially unique. Every wicked problem can be considered a symptom of another problem. Discrepancies in representations can be explained in numerous ways - the choice of explanation determines the nature of problem's resolution. Problem solvers are liable for the consequences of the actions their solutions generate. Many real-world problems have the aforementioned characteristics.

The notion of wicked problems raises the possibility of system paradoxes (Baldwin et al., 2010). Classic paradoxes include whether light is a particle or a wave. Contemporary paradoxes include both collaborating and competing with the same organization. The conjunction paradox relates to the system including element A and element not A. The biconditional paradox holds if A implies B and B implies A. For the equivalence paradox, system elements have contradictory qualities. With the implication paradox, one or more system elements lead to its own contradiction. The disjunction paradox involves systems that are more than the sum of their parts. Finally, the perceptual paradox reflects perceptions of a system that are other than reality.

Finally, there are fundamental theoretical limits as to what we can know about a system and its properties (Rouse and Morris, 1986; Rouse et al., 1989; Rouse and Hammer, 1991). There are limits of system information processing capabilities (Chaitin, 1974), limits to identifying signal processing and symbol processing models, limits of validating knowledge bases underlying intelligent systems, and limits of accessibility of mental models in terms of forms and content of representations. The...