Reductionism has always been the problem solving technique in sciences and associated disciplines. In reality solving a problem by reduction means transforming the problem into simpler problems and constructing or deducing solution of the original problem from the solution of the new problem. For example a student is given a problem to calculate the sum of all numbers between 1 and 101 which are not divisible by 3. Working a direct solution for this will be tedious so it is solved reductively. The required sum is the difference between sum of all numbers between 3 and 99 which are divisible by 3. Both sums can be calculated as arithmetic series. Thus the original problem is subdivided into sub problems of calculating arithmetic series which is actually nothing but reductionism (Armoni, Gal-Ezer & Tirosh 2005, p. 114). On the hand the term systems approach derives its base from holism or holistic thinking. The term holism was coined by South African philosopher Jaan Christiaan Smuts. He argued (Anderson, H 2001, p. 155) that “a unity of the parts could be so close and intense as to be more than the sum of its parts.” Smut explains the philosophy behind holism and systems thinking. Systems thinking usually deals with complex systems. Now a complex system unlike a conventional single feedback system comprises of numerous subsystems and the overall behavior of the system relies on the interaction of these systems. So the system output is not as simplistic and is not merely a sum total of the constituent entities.
Professor Yaneer Bar-Yam President of the New England Complex Systems Institute (Bar Yam, Y 2000) defines reductionism as “an approach to building description of systems out of the description of the subsystems that a system is composed of and ignoring the relationships between them.” That means any reduction thinking is bound to overlook certain aspects of engineering system. Systems approach on the other hand is the tendency to look at an object as a whole. During the mid 19th century the British philosopher John Stuart Mill argued that the properties of molecule could not be derived from the properties of the constituent elements. He (Anderson, H 2001, p. 153) postulates “…. the chemical combination of two substances produces, as is well known, a third substance with properties different from those of either of the two substances separately or of both of them taken together. Not a trace of the properties of hydrogen or of oxygen is observable in those of their compound, water.” Stuart Mill speaks of what later came to be known as emergence.
In systems thinking the problem is analyzed by its face value. That is if there is patient suffering from difficulty in breathing then instead of checking his/her respiratory organs we check for his ability to inhale and exhale. In this way all the causative agents will be taken into account. Reductionist might advance into inspecting internal organs while the problem might lie in some allergic substance in the environment or it can be a common case of influenza. Therefore the system approach has a broad spectrum which might help anticipate emergent behavior or at least take every precaution against any undesired emerging behavior.
One of the most peculiar attribute of scientific reductionism is the way in which it proceeds to diagnose a problem. A reductionist would go about gathering data using a response or feedback from source. So in scientific reduction a stimuli-response is used for data collection in research. A similar set of rules are applied to every source to extract a response which reduces the time to extract data if the number of observations are large. The problem with reductionist approach (Verschuren, P 2001, p. 391) is that there is a tendency to “prestructure”. That means for every problem a system encounters a standardized and structured methodology is adopted. On the other hand a systems approach would be to look at collective characteristics. System thinker would look at culture, team spirit, and ethics especially in behavioral sciences. Similarly in engineering problem a systems thinker will look for aggregated characteristics like engine throughput, intake, output power rather than efficiency, input voltage and engine-shaft torque. The guiding force behind system thinking is holism. In contrast, reductionism is a pursuit to identify sub problems to the central problem and finding solutions to these sub problems.
A systems approach encourages inferences from previously encountered problems. A systems approach is to draw parallels between problem and the reductionist has the proclivity to proceed serially (Verschuren, P 2001, p. 392). The iterative-parallel way has an inverse relation between the research and the theory (Creswell, J 1994) . Systems approach complement, adds and challenges the existing theory. On the other hand scientific reductionism forms a basis for formulating the research. However in a systems approach the iterative and parallel thinking gives rise to theory and in turn gives rise to generation and analysis of new research material.
There are several reasons why reductionism has been successful in science and engineering. Quantification and quest for precision is one reason (Verschuren, P 2001, p. 399). Since all the physical forces are quantified engineers require measurable parameters and want that the underlying entities add to a sum total. The insistence quantification is the belief of objectivity of figures and it is a common refrain “figures can’t be wrong”. A lot of trust is attached to raw data, facts and figures. For instance, ( Verschuren, P 2001, p. 400 )consider a researcher formulating a hypothesis that there are two groups the members within group 1 interact significantly more than those of group 2. Every day interactions are observed for each group and the number of interaction per day are recorded. This is done repeated and data is recorded for a period. Now a collected data of these interactions among individuals will be more convincing than an overall impression of a researcher who would observe the whole group entities such as trust, friendship and compatibility.
In spite of such widespread success of scientific reductionism systems approach is now being preferred in sciences and in solving practical problems as well (Ackoff, 1974; Esterby-Smith et al., 1991) because the modern problems are more complex and dynamic for any reduction to be possible. Dividing and giving each quantity a unit and variable might not be feasible. It can be tedious, expensive and might not give a clear picture. The wisdom of scientific reduction might not be true. A good example is the performance of a football team. There are teams consisting of excellent athletes and players having high number of championship goals yet they are ranked low because they lose crucial matches. In general the quantity of goals individual footballer has scored is not a definitive indicator of the character of the team.
Reductionism at times can give a reduced picture of a problem and would rather ignore any emergent property. Another example, between physical between physical and social reality is the aesthetic quality of an object (Verschuren, P 2001, p. 401). For instance, a single object all by itself may not be of any aesthetic value. But the combination of objects may constitute an exceptional aesthetic appeal. This phenomenon is too abstract to define and scientific reductionism has no answer to it.
Scientific reductionism suffers from observational bias and the observation is tunneled in a sense that an object is perceived to be isolated from its physical and social surroundings. The problem can be looked upon after being isolated from the whole which it belongs. Focusing on rudimentary parts of the object/problem in reductionism one can forget its functionality, interconnectedness and context dependency (Verschuren, P 2001, p. 401).
A system approach utilizes analogous thinking and inductive reasoning (Verschuren, P 2001, p. 397). Analogous thinking is based on perception of similarity and inductive reasoning deals with reasoning that moves away from specificity to generality. Given below is a table that compares reductionist versus systems approach. The difference in perception is given with respect to object, observation and strategy (Verschuren, P 2001, p. 398).
The development in science and technology after the industrial revolution has been fast and unrestrained. The research domains have expanded and the mechanism of breaking down the problem via scientific reduction has become redundant. To counter the modern problems certain existing theories and beliefs must be changed (Wulun, J 2007,p. 395).
- All things cannot be decomposed or reduced to its elements and elements cannot be replaced always.
- All the elements never always add up to the original entire body. The whole is not always equivalent but larger or smaller than the sum of its part.
- The whole can not always be understood by being dividing it into or reducing it into its parts and understanding those parts (Wulun, J 2007,p. 397).
In my view it is not constructive to discuss what is better reductionism or systems approach. It is concerned with the nature of the problem, interests and priorities of the stakeholder. Newtonian sciences have led to the development of ‘constitutive theory’ (Wulun, J 2007, p. 399). According to constitutive pattern of thinking, development and change are caused by division and combination of unchangeable constitutive elements, which are made of even smaller elements. The current sciences base themselves heavily on Newtonian physics and its philosophy. Albert Einstein contribution to the quantum physics gave the sciences a new breath of life. Biology, behavioral science and especially systems sciences realized that the natural world and human society are inseparable, interacted whole. To understand the whole system thinking is required.
REFERENCES
Ackoff, R. L 1974, ‘Redesigning the Future; a Systems Approach to Societal Problems, New York, London, Sydney, Toronto; john Wiley and Sons
Andersen, H 2001, ‘The history of reductionism versus holistic approaches to scientific research’, Endeavour, vol. 25(4), pp. 153-156
Armoni, M, Gal-Ezer, J & Tirosh, D 2005, ‘Solving Problems Reductively’, J. Educational Computing Research, vol. 32(2), pp. 113-129.
Bar Yam, Y 2000, Reductionism, Concepts in Complex Systems, viewed 6 May 2008,
<>.
Creswell, J. W 1994, Research Design: Qualitative and Quantitative Approaches. Thousand Oaks, London, New Delhi: Sage.
Verschuren, P 2001, ‘Holism versus Reductionism in Modern Social Science Research’, Quality and Quantity, vol35, pp. 389-405.
Wulun, J 2007, ‘Understanding Complexity, Challenging Traditional Ways of Thinking’, Systems Research and Behavioral Sciences, Sys. Res 24, pp. 393-402.
