Analysis is a method of inquiry that proceeds by breaking a system down into its elementary parts, studying those parts in isolation and then forming a description of the whole system in terms of its basic components and interactions.1 Analysis is one of the two fundamental processes of inquiry and can be contrasted with synthesis.
Analysis is a process of inquiry based upon the reductionist paradigm. Whereas reductionism may be understood as a paradigm or worldview, analysis can be understood as the accompanying method of inquiry used within this paradigm. Reductionism posits that higher level phenomena within a system should be derivable from lower level of more elementary parts. Thus, analysis works to break systems down to derive these elementary components, their basic interactions and then tries to form an account of the whole as a product of some combination of the parts. Analysis attempts to explain complex phenomena with reference to the properties and interactions of individual constituent components. The reductionist paradigm can be summed up in the maxim that “the whole is equal to the sum of its parts”, and thus an account of the parts and their interactions is sufficed to form an account of the whole.
Process of Inquiry
The basic method of analysis involves a step-by-step breakdown of the phases to a process or components of a system, the study of those parts properties and operations in isolation and a recombination of the parts to form an account of the whole.2
Isolation: Analysis typically involves studying the parts of a system within a static and stable environment. To fully understand the parts functioning, the system must be isolated from its environment in order to remove auxiliary variables that may affect the behavior of the component parts. Thus, analysis involves isolating parts and performing controlled experiments where variables outside the boundary of the system are held constant, what is called “ceteris paribus,” a Latin phrase meaning “other things equal.”
Decomposition: Once a system has been identified and isolated, the next stage involves decomposition. Rene Descartes—arguably the main founder of the modern analytical approach—described this as such “reduce obscure and complex propositions step by step to those that are simpler.” Alternatively, as he writes in his book Discours de la Methode, “break down every problem into as many separate simple elements as might be possible.”
Identify Components Properties: The final stage in the process involves trying to understand the basic properties of these parts and the input/output functions that govern their operations. At this stage, we alter input variables to see how they change and from that formulate a function to describe the internal workings of the elementary component parts.
Recombination: Finally, we reassemble these elementary parts back into a whole system and form an account for the whole as some combination of the parts.
Throughout the modern era, from approximately the 1500s to the late 1900s, a reductionist philosophy has been dominant within virtually all areas of scientific inquiry, the expansion of human knowledge and most areas of its application, such as engineering and business management. Within many domains, it has been assumed that if a phenomenon was deconstructed sufficiently and the parts examined, understanding of the phenomenon was assured.3 This approach has proven highly successful in many ways, from the development of the large and sophisticated body of modern mathematics that provides a standardized set of solutions for solving problems analytically, to the understanding of biological systems in terms of DNA, to the development of large hierarchical modern corporations and bureaucracies. In all occasions when we are dealing with relatively simple systems with a low level of interconnectivity, interdependence, feedback and change analysis work well and it is a precondition to obtaining a full understanding of any system.
The analytical approach has many advantages and it represents the starting point to any full inquiry in that it provides an account of a system’s parts and their properties giving us a structured, and often detailed, understanding of the internal components of the system. Analysis allows us to approach complex phenomenon that otherwise may appear daunting, in that it allows us to break them down into more manageable parts that can be distributed out and tackled collectively or over a series of stages. The analytical approach renders control, predictability, and stability which can be very much desirable.
Since the Renaissance, analysis has been the foundation to our methods of problem solving. However, in the transition from a modern industrial society to a post-industrial society, it is becoming increasingly clearer that analysis alone is limited in its capacity to provide a complete way of understanding our world.4 In a recent article by H. William Dettmer on the topic of synthesis and analysis, the author describes the limitation to analysis as such: “Because early organizational (specifically, industrial) systems were simple, the analytic approach was much more effective than the alternative—which was more or less ad hoc. However, as the complexity of organizational systems increased to the point where no one person could have complete visibility on all components simultaneously, the ‘cracks in the plaster’ of the analytic approach began to show. Analysis could no longer explain the difference between the whole-equals-the-sum-of-the-parts and observed results that were disproportionately higher (or lower) than expected. In other words, the success of an analytical approach ‘topped out.'”5
The primary reason given for the limitations in the analytical approach being that complex phenomena are known to exhibit emergent properties, 6 where, when the parts are put together, new functions, properties and capabilities form on the level of the whole that are lost when the system is broken down into parts. Thus, these novel emergent features can only be understood by taking the system as a whole. Analytical reasoning will only tell us about those features to a system that are entailed within its parts, it does not tell us about what may emerge out of the interaction between those parts. Likewise, the analytical approach removes a system from its environment and studies its internal components. Because of this, it cannot describe how the system may change, adapt or evolve new states and properties due to its interaction with its environment and other systems. As such, standard reductionist science has had great success in studying lower level phenomena, such as physical and mechanical systems, but much less success in trying to understand more complex higher level phenomena like economies and societies.
Reductionism breaks systems down to their most elementary level and forms an account of the whole in terms of a hierarchy derived from this most basic level, with the different levels in the hierarchy seen to be fully derivable from the more elementary parts. However, “most philosophers would insist that this approach to conceptualizing reality—our need for a hierarchy of levels of understanding that can all be traced back to interactions on a lower level—does not change the fact that different levels of organization in reality do have different properties.”7
Over usage of reductionist, analytical methods can result in a fractured system or fracture understanding. This is manifest in the scientific community as a focus on specialized domains with limited overall cohesion to science as a whole enterprise, or of complete macro level descriptions to the world that cross and interrelate different domains. Within organizations, this fracturing is understood as the silo effect. The silo effect refers to the dividing up of organizations along lines of specialization and departments such as Finance, HR, IT, Strategy, Ops, etc. without proper interoperability between them, for them to function as a whole and the capacity to effectively respond to changes within the environment.8
Organizational silos are a classic manifestation of what Professor Russell Ackoff refers to as “Analytic thinking”. The belief that managing each department to optimize its performance in isolation, will contribute to the better performance of the organization as a whole, “In general, those who make public policy and engage in public decision-making do not understand that improvement in the performance of parts of a system taken separately may not, and usually does not improve performance of the system as a whole. In fact, it may make system performance worse or even destroy it.” – Russell Ackoff.9
Reductionist thinking by reducing its focus to inward pieces largely ignores the environment or context within which the system exists. This can lead to a number of problems, such as lack of adaptive capacity, failure to properly identify the overall functioning of the system and how that needs to change in response to changes within the environment and limited capacity to foresee and respond to major changes within the environment. Likewise, excessive use of reductionism within the design and management of a system can lead to unstainable behavior due to its lack of regard for the environment.
Analysis forms one part of a complete process of inquiry that involves both breaking things down to understand them as a function of their parts, but also understanding them in relation to other systems and their environment—what is called synthesis. Any effective inquiry, design or management effort will require a balanced emphasis on both the analytical and synthetic approach in order to counterbalance the negative effects of each and achieve an optimal overall outcome.