Research 
 

                                                                                                                                  

Western Michigan university

Science Society

Truth is relative Wherever it is found

 

Alea Leibner

2/24/2014

“The question we ask today is not whether our government is too big or too small, but whether it works, whether it helps families find jobs at a decent wage, care they can afford, a retirement that is dignified.”

– Barack Obama

The world is in economic crisis bringing upheaval throughout the planet.  Experts disagree about the best ways to manage paths to stability and prosperity for global societies.  The severity of the crisis pressures policy makers toward pragmatism, whatever their ideologies.  The big question for every leader involves the effectiveness of their intended actions.  Will those actions work? The issues before world leaders range from short-term economic recovery necessitated by the failure of capital markets, to long term survival of humans on the planet earth challenged by climate change and ecological systems, natural resources, and population growth.  Potential consequences for world societies and civilizations are enormous.

World leaders need confidence that they can predict outcomes when they implement their plans.  They cannot manage their policies without prediction.  W. Edwards Deming tells us that management is prediction (Rienzo, 1993).  How does the human mind find confidence in predictions? From where does confidence come?

Confidence comes from knowing the systems we are attempting to manage.  The purest expressions of knowledge that we have as human beings are scientific laws.  Scientific laws allow scientists to predict outcomes with certainty when they engineer physical structures, mechanical technologies, or chemical/biological reactions.  Can political and business leaders use processes similar to the ones that produce scientific laws to address the most pressing issues of world societies?  The perspective of physicist Nancy Cartwright offers some insight.

Truth is Relative Wherever it is Found

Scientific laws come as close to uncontested truth as exists in the human experience.  They are often presented as true in all circumstances, but Nancy Cartwright (1999) argues against what she calls scientific fundamentalism.  Cartwright argues for more realism in pursuing scientific theories to solve human problems.  She is particularly concerned about unwarranted resources given to glamour theories that promise to produce universal laws governing the behavior of everything.

The pernicious effects of the belief in the single universal rule of law and the single scientific system are spread across the sciences.  Indeed, a good many physicists right now are in revolt.  Superstring theory is the new candidate for a theory of everything…The theory consumes resources and efforts that could go into the hundreds of other enterprises in physics that ask different kinds of questions and solve different kinds of problems (p. 16).

Cartwright is concerned that an unrealistic vision of what is possible poses a danger of much waste and little reward in the application of scientific and economic theories.  She questions scientific fundamentalism – the tendency to think that all facts, regimented into theoretical schemes which provide accurate predictions in highly structured environments, are exemplary of the way nature is supposed to work (p.25). She denies the universality of laws, advocating instead a dappled world governed by local realism and metaphysical pluralism in which reality is more like the outcome of negotiations between domains than the logical consequence of a system of order (p. 1).

Cartwright contends that well known laws, like Newton’s second law, F=ma, are successful only in nomological machines -- fixed (enough) arrangements of components, or factors, with stable (enough) capacities that in the right sort of stable (enough) environment will, with repeated operation, give rise to the kind of regular behavior that we represent in our scientific laws.  It is the design of nomological machines operating under specific conditions that produces scientific laws.

So here is my strong claim: look at any case where there is a regularity in the world (whether natural or constructed) that we judge to be highly reliable and which we understand – we can either explain the regularity or we believe it does not need explanation.  What you will find, I predict, is that the explanation provides what is clearly reasonable to label as a nomological machine.  And where there is no explanation needed you will still find a machine.  Sometimes for instance the whole situation is treated as one simple machine (like the lever), where the shielding conditions and the idea of repeated operation are so transparent that they go unnoted (Cartwright, 1999, pp. 58 – 59).

Nomological machines must be shielded, i.e. the nomological machine must operate as prescribed without interference.  Shielding is an important concept.  Its necessity for successful laws means that laws are true only under ceteris paribus conditions – all things being equal.  Failure to recognize the importance of shielding can result in mistaken notions of reality.

We tend to think that shielding does not matter to the laws we use.  The same laws apply both inside and outside the shields; the difference is that inside the shield we know how to calculate what the laws will produce, but outside, it is too complicated.  Holists are wary of these claims.  If the events we study are locked together and changes depend on the total structure rather than the arrangement of the pieces, we are likely to be very mistaken by looking at small chunks of special cases (Cartwright, 1999).

This systemic view is shared by Capra (1996) describing biological sciences.

Living systems are integrated wholes whose properties cannot be reduced to those of smaller parts.  Their essential, or “systemic,” properties are properties of the whole, which none of the parts have.  They arise from the “organizing relations” of the parts – that is, from a configuration of ordered relationships that is characteristic of that particular class of organisms, or systems.  Systemic properties are destroyed when a system is dissected into isolated elements (p. 36).

Emergentism in physics holds that there are macro-properties that do not supervene on micro-features, i.e. the nature of the parts does not define the properties of the whole.  Even in generally accepted scientific paradigms, reductionism and inductivism cannot adequately explain systemic reality, either in biology or physics.  Sciences are tied in application to the same material world, but scientific domains are not reducible to more fundamental domains.  Their boundaries are flexible, but real.  We cannot simply assume a universal cover of law.

Cartwright challenges the universality of F=ma by comparing the movement of a thousand dollar bill tossed in the wind with that of an airplane.  There is no adequate model for the movement of the paper bill in the wind.

I do not doubt that when the right questions can be asked and the right answers can be supplied, fluid dynamics can provide a practicable model.  But I do doubt that for every real case, or even for the majority, fluid dynamics has enough of the ‘right questions’. It does not have enough of the right concepts to allow it to model the full set of causes, or even all the dominant ones.  I am equally skeptical that the models which work will do so by legitimately bringing Newton’s laws (or Lagrange’s for that matter) into play (Cartwright, 1999, p. 27).

Adequate models for airplanes do exist.  The thousand dollar bill comes as it is.  The airplane is intentionally designed to fit a model in which Newton’s laws are known to be successful.  The design of the airplane assures its obedience to Newton’s laws.  Regimented behavior results from good engineering (p. 1).  Cartwright does not claim Newton’s Laws cannot be applied to the paper bill in the wind, but force is an abstract concept and there is no justification to assume it can be linked to the mass and acceleration of the thousand dollar bill until a working model is created.

When we have a good-fitting molecular model for the wind, and we have in our theory (either by composition from old principles or by the admission of new principles) systematic rules that assign force functions to the models, and the force functions assigned predict exactly the right motions, then we will have good scientific reason to maintain that the wind operates via a force.  Otherwise the assumption is another expression of fundamentalist faith (Cartwright, 1999, p. 28).

Cartwright claims that fundamentalists want true laws, but most of all they want their favorite laws to be in force everywhere.  Reality may be just a patchwork of laws (Cartwright, 1999, p. 34)


Engineering Systems: Natures and Capacities in Nomological Machines

Cartwright argues that discernment of the underlying reasons for reliable behavior is contained in natures and capacities of system components.  To understand the nature of system components is to understand their tendencies to behave in certain ways.  Components that are inclined to behave in certain ways have the capacity for those behaviors.  One of the most challenging activities of science is describing system capacities which can change in subtle, or even dramatic, ways with different component interactions and different environments.  Successful descriptions of natures and capacities allow scientists to manipulate the world to solve problems and achieve goals.  Cartwright (1999, p. 81) suggests that the Aristotelian concept of nature is a useful aid in grasping the significance of insightful engineering of components in systems.  Natures and capacities are inherent behavioral patterns of systems but their ultimate expression, at levels necessary for scientific utility, are context dependent.  Holistic thinking is needed, as well as a thorough understanding of inherent behaviors and the effect of environment.  Developing that awareness can be a tremendous amount of work.  Although Cartwright suggests that a modern understanding of natures does not exactly match that of Aristotle, the Aristotelian notion that natures describe inherent behaviors is critical to the design and maintenance of systems that produce reliable outcomes.  Differences between natures as understood by Aristotle and  

Table 1: Nature As Interpreted by Aristotle and Cartwright

Nature Characteristic

Aristotle

Cartwright

Association

Essence

Behaviors

Assignment

Substances

Systems and Configurations

Observation

Naked Eye

Special tools and instruments needed to detect behaviors

 

Figure 1 Nomological Machines

 
Science seeks to understand natures and capacities so that the propensity of systems under different environments and different stimuli can be recognized and exploited to address scientific problems or opportunities.  Natures are the more fundamental properties of system components, but holistic operation of the entire system depends upon the capacities of components, their tendencies to behave specific ways in an engineered system, when interacting with each other and the environments that surrounds them.  As shown in Figure 1, if capacities (C1, C2, and C3) are properly understood, they can be arranged in a nomological machine to achieve predictable outcomes.  Knowledge of natures and capacities builds upon itself.  New or improved understanding of capacities makes new nomological machines possible leading to new scientific laws.

The nomological machine itself is an abstract entity involving abstract capacities arranged by scientists in a particular way to produce abstract laws.  The abstract machine requires a concrete representation that makes it relatable to the way we think.  The concrete instantiation of abstract capacities and laws is accomplished through representative, interpretive models.  Abstract fields, forces, and fluxes are given the context of weights, springs, switches, electrical charges, distances, accelerations, and fulcrums.  Cartwright likens abstract laws instantiated by representative models to abstract morals instantiated by fables that give the morals context in human living.  Fables “fit” morals to our human experience.  Models of nomological machines “fit” laws to scientific experience.  Nomological machines offer science two major benefits:

·         They provide absolutely predictable and reliable outputs as long as their systems are permitted to operate according to their designs, i.e. the machines must be shielded to produce predictable outcomes.

·         The understanding of natures and capacities needed to create one nomological machine can provide clues to the creation of new nomological machines, providing new ways to manipulate the world.

Although scientific laws necessarily involve regularities, and regularity is essential for accurate prediction, knowledge of capacities is more widely useful than knowledge of regularities alone (Menzies, 2002).  Knowledge of capacities creates the theories that explain regularities.  Capacities allow us to speculate about what is possible, but we cannot predict what a system will do simply by knowing its capacities.  Prediction depends upon the very special structures and regularities of nomological machines.  In nature, these regularities are rare.  If we want predictable situations, we must carefully engineer them (Cartwright 1999, p. 73.) Even with carefully engineered nomological machines, we can only be confident of our predictions if the machines are shielded.  Shielding can often be effectively imposed in the laboratory, but recognition of its presence or absence can be elusive beyond the walls of the laboratory.

Nomological Machines and Social Systems

Nomological machines are not exclusive to the natural sciences.  They can also work for social sciences if appropriate system capacities are known, the machines are well engineered, and they are shielded.  Shielding can be particularly problematic for socio-economic nomological machines because there is no controlled laboratory equivalent to that of the natural sciences.  It may be difficult to tell when ceteris paribus conditions are in effect and when they are not.  Cartwright bridges nomological machines of physics and economics by representing the ability of the banking system to multiply money by means of reserve ratios with the ability of a rigid rod to multiply force by mechanical advantage (Cartwright 1999, p. 142).  Just as the position of the fulcrum determines the multiplication of force, the required reserve ratio determines the multiplication of money.  Both economists and physicists create models which serve as blueprints for their respective nomological machines.  Cartwright claims that three theses follow from the fact that socio-economic laws are created by socio-economic machines

1.      Natures are primary, and behaviors are derivative

2.      Organic analogies suggest a kind of irreducible holism.

3.      Ordinary machines do not evolve.  They have to be assembled and the assembly has to be carefully engineered. (pp. 149 - 150)

All three theses have been addressed earlier in the consideration of the nomological machines of natural science.  Cartwright does point out that evidence is divided about thesis three.  Economic historians tend to rate the designed International Monetary Fund (IMF) as less effective than the self-organized gold standard (p. 150).  Perhaps the designers of the IMF lacked a sufficient understanding of complex economic capacities.  Peter Drucker (1993) claims that any attempt to create a large centralized economic system (nomological machine) is doomed to failure.  He offers the continual and progressive failure of the Soviet central planning system from the 1950s to its complete demise in 1991 as a primary example (p. 189).  There is room, and perhaps a necessity, for self- organizing economic systems.


Implications for Industrial Policy

Cartwright’s focus on behaviors derived from capacities of natures, and regular behaviors resulting from appropriately engineered and properly shielded nomological machines has implications for private industry.  When industry transforms laboratory science into commercial technology it must take special care to insure the nomological machine producing regular behavior remains properly shielded.  Most often, this is accomplished industrially through process analysis and quality control techniques.  Statistical quality control tells system operators when the system is changed (shielding compromised) by common or special causes.  The control chart in Figure 2 represents red beads removed from a large container by a dimpled paddle in experiments conducted during seminars by quality management expert W. Edwards Deming (1991).  Each participant in the experiment used one paddle containing 50 concave dimples to draw beads, on four different occasions, from a fish tank containing a mixture of white and red beads.  The mixture contained 80 percent white beads and 20 percent red.  Paddle operators were not permitted to sort or filter beads that filled paddle dimples in any way, yet they were told that they were expected to produce only white beads when they drew beads from the fish tank.  During his long career as a statistical consultant, Dr. Deming used these exercises to demonstrate predictable outcomes from a stable system that were beyond the capabilities of operators to change.  The heavy red line represents the average number of red beads accumulated by each participant.  The broken red lines represent upper and lower control limits of the paddle system calculated from experimental outcomes.  As long as the system was shielded (no change in percentage of beads in the tank, no change in paddle dipping technique, no change in beads that would affect affect bead accumulation characteristics) it would produce between 2 and 19 red beads each time the paddle was dipped into the tank with an average of 10.25.

 


Dr. Deming tells us all of management is prediction (Rienzo, 1993).  Accurate prediction comes from an appropriate understanding of capacities and nomological machines.  Deming himself suggests that the benefits of understanding systems and processes extend well beyond the shop floor.  He claims only 3 percent of the benefits of understanding systems and processes are located on the shop floor (Deming, 1991).  Most of the potential benefit lies in overall business strategy and companywide business systems in sales, marketing, human resources, training, logistics, and service.  Why are these areas left unexploited? The answer lies in the confidence of management in the nomological machines of manufacturing.  Businesses believe in the models that are used to design the nomological machines of the shop floor.  They are confident that they will be able to achieve predicted outcomes if the shielding conditions of the machine are maintained through quality control.  And they are confident they are capable of maintaining shielding conditions.  Few managers would argue with the expected red bead delivery of between 2 and 19 beads with every scoop of the paddle.  Simply put, manufacturing systems readily lend themselves to mechanistic models.  Many service and logistics activities are concerned with process timing, and they also readily translate to a machine perspective.  More could be done in logistics and service activities with the tools of manufacturing quality.  But strategy and company-wide business systems are driven by semantic processing of data and information to make choices that fulfill business aims.  They are oriented by information, communication, and human judgment.  They exist in a dappled business environment accurately described by the same vocabulary that Cartwright’s uses to portray scientific events: local realism and metaphysical pluralism in which reality is more like the outcome of negotiations between domains than the logical consequence of a system of order.  Universally recognized reliable models for business processes other than manufacturing and time-dependent service and logistics activities are virtually non-existent.  When (and if) reliable models are demonstrated, permitting engineered processes that lend themselves to reliable prediction, new nomological machines will insure that the remaining portion of Deming’s quality transformation will virtually explode into businesses.

Implications for Civilization and Social Policy

The global financial crisis begun in the United States during Fall 2008 has created tremendous challenges for global economic systems.  Calls for bold action ring throughout the world.  U.S. President Barack Obama has received calls to be bolder than Franklin D. Roosevelt during the depression of the 1930s.  During times of crisis, pragmatism overwhelms ideology in all but the most dogmatic perspectives.  The almost exclusive focus of pragmatism is effectiveness.  A plan for effectiveness depends upon the quality of prediction, and prediction depends on the regularity of engineered machines.  Obama himself has set a data driven decision making process based on scientific evidence and effectiveness (Obama, 2009).  But the big issues of government policies and global ecosystems do not lend themselves well to engineering machine models because they are:

v  processes that have tremendous degrees of complexity, making fine grain analysis extraordinarily challenging

v  impossible to shield.

Mechanistic models have been used to examine civilizations.  Blaha (2006) investigated the health of historic societies through a model derived from statistical mechanics.  He proposed a “Perfect Civilization/Society” sufficiently isolated (sufficient shielding) to make predictions over long periods of time.  He found societal strength and state of mind rising and falling in three and a half beat cyclical patterns through thousands of years.  While the recognition of these patterns makes it possible to make predictions on a macro scale, they are not relevant to politicians concerned about the next election cycle, or policy decisions that affect the immediate future.

The short-term complex behavior of human beings may be possible to characterize with large generalized “laws” (e.g., people tend to act in ways they believe to be in their self interest) but filling in detail is a formidable task.  Even if detailed models could be created, laws are only valid when they operate in shielded nomological machines, and there is tremendous complexity in examining shielding in social settings.  Laws have been proposed for information itself.  In his study of information technology and civilizations, Targowski (2009) proposes five laws of information:

1.      Complexity grows proportionately to the level of existing information.

2.      Information generates consequences which it cannot foresee.

3.      Precision and certainty are proportional to simplicity and inversely proportional to complexity.

4.      Effective use of information promotes globalism and is necessary to succeed in a global economy

5.      Information has no saturation point

A Cartwright perspective of Targowski’s third law suggests that precision and certainty are based upon an understanding of system natures and capacities rather than level of complexity.  Of course, it is reasonable to argue that understanding the natures and capacities of simple systems is more readily accomplished than understanding those of complex systems.  And the second law depends upon system boundaries and shields.  If natures and capacities can be completely determined in a shielded system, there are no unforeseen consequences within that system, although consequences may appear outside of the system.

The predictive value of scientific laws depends upon “good enough” environments, i.e. environments that are aligned with the conditions that were present when the nomological machines that created the laws were formulated.  The great complexity of civilizations challenges the stability that is required for prediction.  Even the scientific problems of climate change and ecology are extremely complex processes with many unknowns.  There is a need for holist expertise, but some concern that sufficient expertise does not even exist.  A. F. Chalmers (1999) describes the advantages of expertise backed by established scientific theory.  He postulates that a walk through a field would be very different if done by a trained botanist compared with someone with little botany education.  The botanist would know what to expect, what was normal, and what was out of place.  Most people would not have the background to make judgments with any degree of confidence that they were sufficiently aware of the natures and capacities of ecosystems to make confident predictions.

So leaders of society are caught in the dilemma of tremendous challenges with enormous potential consequences and no holistic engineered models to give them certainty about their predictions.  Cartwright’s concern about realism in pursuing scientific theories to solve human problems is relevant, and her perspective about the development of scientific laws can be useful to policy makers.  Leaders have to create social and scientific models using the best judgment of a consortium of experts knowing that no single expertise will be sufficiently insightful, and wondering if even the collective wisdom of the society will produce “good enough” models.  Understanding natures and capacities of proposed solution models is no trivial task, and will require significant private and public investment.  Shielding conditions must also be carefully considered when applying “laws” developed in social or economic nomological machines to different environments.  Cartwright’s work should encourage people involved with government policy to be particularly concerned with shielding conditions when evaluating pilot programs.  Can a pilot economic nomological machine be properly shielded if the same machine is implemented elsewhere?  Shielding defines the ceteris paribus conditions of the nomological machine, and it is critical to its success.

Cartwright’s contention that reality is more like the outcome of negotiations between domains than the logical consequence of a system of order suggests the following rules of engagement when dealing with the enormous economic and social problems of our times:

1.      Even the most reliable predictions of science are not necessarily universal.  Context is critical even in science.  And its importance is magnified when dealing with social models.  Be careful about assuming one glamour theory provides the only path to solutions.

2.      Work required to understand the natures and capacities of systems is very demanding, and most readily applied to mechanistic processes.  Natures are primary, and behaviors are derivative.  Research involving natures and capacities of complex human and climate problems must be continuous, and models are needed to make systems comprehensible.  A thorough understanding of the natures and capacities of one system can help researchers understand other systems.

3.      Organic analogies suggest a kind of irreducible holism.  Every proposed solution must be “humble”, recognizing that it is part of a larger whole.

4.      Ordinary machines do not evolve.  They have to be assembled and the assembly has to be carefully engineered.  Engineer we must, but we need flexibility and frequent review built in the process.  Particular attention must be paid to shielding.

5.      So much cannot be known in complex societies that faith is needed, since belief is the only way we can deal with what we cannot know.  But faith must always be subject to scrutiny from revelations presented by ongoing research.

Nancy Cartwright does not remove the difficulty and complexity that both natural science and social science face in wrestling with metaphysics, but she advocates practical approaches that recognize the rich diversity of systems and an awareness of the limitations of human engineering.  Policy makers should pay attention.

 

References

Blaha, S. (2006). A Unified Quantitative Theory of Civilizations and Societies 9600 BC – 2100 AD. Auburn, NH: Pingree-Hill Publishing.

Capra, F. (1996). The Web of Life. New York, NY: Doubleday.

Cartwright, N. (1999). The Dappled World. Cambridge, England: Cambridge University Press.

Chalmers, A. F. (1999). What Is This Thing Called Science: An Assessment of the Nature and Status of Science and Its Methods. New York, NY: Open University Press.

Deming, W. E. (1991). Seminar Notes. Transformation for Management of Quality and Productivity. Philadelphia, PA.

Drucker, P. F. (1993). Post-Capitalist Society. New York, NY: Harper Collins.

Menzies, P. (2002). Capacities, Natures, and Pluralism: A New Metaphysics for Science? Retrieved from http://www.phil.mq.edu.au/staff/pmenzies/cartwright.pdf

Obama, B. H. (2009). Inaugural Address. Retrieved from http://www.nytimes.com/2009/01/20/us/politics/20text-obama.html

Rienzo, T. (1993). Planning Deming Management for Service Organizations. 19-29.

Targowski, A. (2009). Information Technology and Societal Development. Hershey, PA: IGI Global.