Interpretation and Methodology


Interpretation is basic to all our endeavors  whether  as scientists or as indi- viduals  going  about  our  daily  lives. Interpretation is distinguished from inference. Inference draws valid conclusions from given premises while interpretation is never beyond  question. No final, absolutely  true interpre- tation  is ever proven: some conjecture  is inevitable  when facts are selected, connected  and put into a plausible pattern. Although  we recognize the role of interpretation and its importance, as when we say ‘it all depends on your interpretation’, we may fail to recognize how pervasive interpretation is, if we think  it is something  we only do when things are vague or ambiguous.

Every time we deliberate  on events or on our  experience,  we are inter- preting.  Interpretation is fundamental since  how  things  are  interpreted determines  what  actions  we consider.  But interpretations can be far from arbitrary. The better  interpretations will be consistent  with the commonly agreed  facts  and  account  for  the  facts  in a more  coherent  way:  bringing the  maximum number  of  facts  into  a  meaningful  relationship with  the minimum   of  conjecture.   Nonetheless,  disagreements over  interpretation will occur given that  the ‘facts’ to be interpreted are selected, ordered  and weighted  in accordance with  the  perspective  or  viewpoint  adopted. Few people  have a completely  open  mind  on an issue but  a point  of view that they prefer to have reinforced  rather  than challenged.

Understanding a person’s perspective  is a prerequisite to knowing  how a person  might be persuaded to another point  of view. Michael  Oakeshott saw this as a problem  for historians when  they impose  on the past  illicit patterns emanating from  the  perspective  of their  current  concerns.  And also  for  politicians  imposing  patterns on  the  future  to  fit a  perspective reflected in some grand scheme for ‘improving’ the lot of mankind (Franco,

2004).1 For Oakeshott, each of us has a governing perspective on the world whether theoretical or practical.  This is true for all scientists where relevant reality is viewed through the perspective of the discipline’s ‘paradigm’.

Scientific paradigms act  as conceptual lenses that  guide  research  and the interpretation of findings.  Although  the different  paradigms in social science are often viewed as competitors (e.g., behaviorism versus cognitive psychology),  more typically, they offer additional windows  onto a problem or seek answers to entirely different questions.


Pickstone (2000) in his history of science, technology  and medicine talks of three ‘ways of knowing’ in science; all involve interpretation:2

1. Natural  history  which  consists  of describing  and  classifying  things as they come to be. Pickstone  argues that  the more scientific inquiry is concerned  with  complexity  and/or  singularity,  the more  scientists tend  to  adopt  the  natural history  way  of knowing.  Zoologists and geneticists fit this category.  The human  genome sequence allows sci- entists to go back in history to infer the order, and even the timing, of each addition to our ancestral  genome. At a more pedestrian level, a good deal of marketing research is concerned  with the natural history way of knowing: doing surveys, describing trends or changes in values plus classifying and tabulating findings.

2. Analysis consists of seeking understanding by ‘dissection’, with things viewed  as a mix  of elements  or  a process  with  the  elements  ‘flow- ing’ through a system. Mathematical analysis belongs to this category which, while never creating knowledge  out of nothing, brings out the implications of data that would otherwise  be hidden.

3. Experimentation consists of tests with results that  are relevant  to the truth or falsity of some hypothesis or theory. But it is not just test results that  are in need of interpretation, for it cannot  just be assumed,  with- out checking, that subjects will interpret their task exactly as intended. Experiment is viewed as the scientific method, though perfectly respect- able sciences like geology and astronomy cannot  conduct  experiments. Pickstone  quotes  Rutherford’s well-known quip  that  science is either physics or stamp collecting to illustrate  the claim for the superiority of experimentation over analysis and natural history.

These  three  ways  of knowing  do  not  typically  address  the  same problems  or answer the same questions. When just one way of know- ing is adopted to  tackle  all the  questions  raised  in a discipline,  the result can be a deficiency in explanation. The three ways of knowing can complement each other.  As Pickstone  says, many  scientific proj- ects involve more than  one kind of knowing.  Thus experimentation, as a way of knowing,  may need to be supplemented by background history  and analysis. This is particularly so when we recall the prob- lem in social science of generalizing from an experiment.

Although  the three methods  embrace the traditional methods  used to gain knowledge  in science, there are advocates  of additional ways of knowing,  namely, intuition and tradition, while in this chapter  we add ‘interpretation’ itself as a sixth method  .

4. Intuition. In some circles, intuition carries the notion  of being a supe- rior  mode  of attaining knowledge  (Plato’s ‘eye of the soul’) or alter- natively as an unreflective inclination to believe something.  Intuition is also viewed as the delivery system for ideas that  reason  is used to defend.  It is generally  accepted  that  intuition is derived  from  non- conscious  knowledge.  Goldberg  (2000)  views intuition as the  con- densation of prior  experience  and  the  result  of condensed  analytic processes.3   The  expert,  using  intuition, bypasses  the  logical  steps precisely because  intuition is a condensation of the extensive  use of orderly logical steps in the past. The conventional view, from the study of adults with brain damage, is that the left side of the brain embraces language  functions  while the right  side embraces  visual-spatial rea- soning with the two hemispheres  communicating via the corpus  cal- losum.  But for  Goldberg  the  left hemisphere  is also  the  repository of compressed  knowledge  and  pattern recognition capacities,  allow- ing a person  to  deal with  familiar  situations, while  the  right  hemi- sphere is the novelty hemisphere,  the explorer of the unknown and the uncharted. He argues it is the right hemisphere  that is dominant when we are young but the right hemisphere loses out to the left hemisphere as we age since it is the left side that  accrues  an expanding ‘library’ of efficient pattern-recognition devices. This suggests the title of his book:  The Wisdom Paradox: How  the Mind  Can Grow  Stronger as Your Brain Grows  Older.

5. Tradition. Tradition in  the  interpretation of  sacred  texts  like  the Bible is for  some  the  foundation test  of truth. We are  all familiar with Galileo’s (1564 –1642)  confrontation with the Roman  Catholic Church  over  the  heliocentric  theory  that  the  earth  moves  in orbit around the sun and  spins about  its own  axis and  that,  in 1633,  the Inquisition coerced Galileo into recanting  the theory.  While it is true that  the heliocentric  theory  was considered  wrong  and  Galileo  was charged  with heresy, it might strike us as odd that  Galileo would  be so singled out, given that Galileo’s claim was simply a more grounded upholding of Copernicus (1473–1543), whose thesis was published  at the time of his death. And Copernicus himself had merely revived the essentially  heliocentric  view  of  Aristarchus (310 –230  BCE). What really incensed the Church  was Galileo’s refusal to acknowledge the

‘deeper truths’ of Church  tradition  over claims for his method as the way  to  establish  truth. As David  Deutsch  (1997)  argues,  ‘the real dispute was not about  whether the solar system had one layout rather than  another: it was about  Galileo’s brilliant  advocacy  of a new and dangerous way of thinking  about  reality’ (p. 74).4  Galileo implicitly claimed that  scientific reasoning  took  precedence  over religious tra- dition  and revelation:  it was this notion, not the heliocentric  theoryper  se, that  the  authorities considered  dangerous. Galileo  was  for- bidden  to  hold  and  defend  the  heliocentric  theory  as the  explana- tion of the appearance of the night sky. In denying the reliability  of scientific  knowledge,  it  was  the  explanatory part  that  the  Church rejected.  Galileo was not  forbidden from using or writing  about  his theory  or even defending  it as a method  of making  predictions. The Church  simply  believed  Church  tradition and  revelation  were  the source of true knowledge:  scripture  being but part,  not the whole, of that  living tradition. Indeed,  it was claimed  that  reading  the Bible, unaided, could  not  teach  doctrine  and  that  scripture  was  not  use- ful as a source  of direction  on  how  to  live in the  world  (Simpson, 2007).5  The  Church  could  point  out  that  no  final explanation can ever be proved  absolutely  as God could produce  the observed  effects in an infinite number  of ways. In today’s debate  over evolution  and creationism/intelligent design, we have a similar  clash between  one tradition of biblical interpretation and scientific claims.

Although  we think of those seeking truth  by way of tradition as belong- ing to a religion,  in science there is the related  notion  of conventionalism, which regards  the truth  of some statement as determined not by empirical fact but by social usage or social agreement. For the conventionalist, once a law or method  is found  useful, its acceptance  becomes a pragmatic matter of convention. Paul Samuelson (1965) is a prominent conventionalist in the field of economics.6

All five ways of knowing  entail  interpretation as interpretation is part of any inquiry even in the natural sciences. Take, as illustration, a book on science I have in front  of me with the heading  “Tests  for Thinking  Rats”.7

A white rat is shown  leaping through one of three doors.  Two of the three doors  have  identical  horizontal stripes  but  these  doors  are  latched.  The third  door  has vertical  stripes  and  is unlatched, allowing  the rat  to jump through the  door.  The  rat  is shown  choosing  the  proper  door,  jumping through it to get a reward.  The caption  says this proves the rat has grasped the concept of ‘oddness’. This is one interpretation but not a defensible one; the rat has at best simply learned  to recognize the door  it would  be able to jump through. In fact, the grasping of concepts  presupposes language use.

Significant  innovations in the  natural sciences have  been  resisted  as a result  of dubious  interpretation. Thus  Eddington, whose  brilliant  experi- ment  tested  and  validated  Einstein’s general  theory  of relativity  in 1919, employed arguments based on a suspect interpretation of general relativity theory to undermine (and ridicule) the theory of a young Indian academic, Subrahmanyan Chandrasekhar, a colleague at Cambridge. This resulted in the search for black holes being held back for 40 years when Chandrasekhar came back  to work  on his original  discovery  (Miller,  2005).8  Fellow sci- entists  had  great  difficulty in accepting  Einstein’s general  theory  since its perspective  was so discontinuous with  Newtonian physics.  In his general


relativity  theory,  Einstein in 1916  was able to combine  gravity with space, time, matter  and energy; not bad for someone  whom  his calculus teacher, Hermann Minkowski, called a lazy dog! Einstein’s general theory changed the meaning,  conceptualization, and interpretation of gravity from a force to being viewed as the outcome  of the curved geometry of space and time.


6. Interpretation itself as a way of knowing. Interpretation itself can be regarded  as a way of knowing,  making  it the sixth way of knowing. Interpretation may be the methodology of interest,  not interpretation as something  just ongoing  to all methods  of inquiry.  As such, inter- pretation becomes a methodology or way of knowing  in its own right. In contrast to deductive  inference, interpretation is never guaranteed to  produce  valid  conclusions. This  does  not  mean  that  logic is not involved  in interpretation. Take  the  following  quote  from  the  first paragraph of an  Op-Ed  piece I have  just  been  reading,  written  by columnist  David Brooks (2005) in the New  York  Times:

Most  serious  people  who  spend  time  in  Iraq  report  that  reality there is contradictory and kaleidoscopic. The Sunnis are participating in the democratic process;  the Sunnis are supporting the insurgents. The Shiites are building  a national government; the Shiites are creat- ing death  squads.  The  Americans  are  securing  neighborhoods; the Americans are inciting violence. (Brooks, 2005)9

If we are to interpret this intelligently, we take account  of the logic. The  first  sentence  is true  only  if we  accept  the  author’s view  (not given) of what  constitutes ‘serious people’. And contrary to Brooks, his statements are not in contradiction: some Sunnis may participate in the democratic process and some may support the insurgents, while some Shiites may be building a national government while some may create death squads,  and some Americans may secure neighborhoods while at the same time be inciting violence. (In formal  logic, the con- trasting  propositions are not contraries but subcontraries.)




Each methodology used to obtain  knowledge  can be regarded  as a ‘technol- ogy’. Technology  is concerned  with  building  systems that  can succeed or fail, governed by rules that are not true or false but effective or ineffective. This  view of technology  as consisting  of rules  or  operational principles for achieving successful practical  performances is that  of Michael  Polanyi (1978).10  Toulmin  (1977)  similarly  defines technology  as a population of techniques, recipes, processes and procedures.11 Technology includes systems like telecommunications, computers, buildings, cars, trains, airplanes—but also all investigative  and  planning  systems. In contrast to technology,  the natural sciences like physics and chemistry  are concerned  with developing explanatory theory  that  aims at tracking  truth  in respect to things such as atoms,  heat, light, sound,  electricity, magnets,  forces and motion.

Polanyi points out that,  though  we can export  the objective fruits of sci- ence (like scientific explanations) throughout the world,  we cannot  export the skills of doing  good  research  since these skills require  practice  in the application of loosely textured rules, usually learned under the guidance of an expert.  In the recruitment of researchers,  the focus is typically on where someone was trained, by whom  and for how long. Every methodology is a skill and this implies that research methodology is not mastered  by reading books; books simply get us started  and help us avoid errors.

Interpretive methods  are a technology  as they are concerned  with devel- oping  systems of interpretation that  can be effective or ineffective.  There is no  single, unique  method  of interpreting. On  the  other  hand,  there  is no  unique  scientific method  for  the  natural sciences.  As Putnam  (1981) says, no philosopher of science today  accepts  that  there  is just one scien- tific method.12  Susan Haack  (2003)  agrees,  arguing  there  is no magic set of methods  we ‘baptize as scientific method’, distinct  from the intellectual tools  we employ  in our  daily  lives.13   As always,  the  method  employed  is determined by the kind  of understanding that  is sought  and/or  the ques- tions being addressed.


Technology  is governed  by rules that  are not  true  or false but  effective or ineffective. Marketing management, though, can never be a mechanical application of rules whether the rules are called rules, principles, heuristics, maxims or whatever.  They have to be interpreted in the light of situational factors to avoid putting  in standard solutions  when standard conditions do not exist. The trouble  with all rules or principles is that, when very general, they seem to have little applicability to the individual  case. On  the other hand,  the more specific the rule, the more it becomes like a recipe, with no room left for creativity.

Herbert Simon (1957) puts management principles into the category of proverbs,  essentially useless in that for almost every principle one can find an  equally  plausible  and  acceptable  contradictory principle  (‘too many cooks  spoil the  broth  versus  many  hands  make  light  work’).14  To  allow generality,  principles  assume  sameness  across  situations which  can  be denied.  But Simon’s is a wrong  perspective  on  the  nature  of principles. Principles  (like  proverbs)  fall  under  objective  relativism,  which  claims that,  while the valid application of a principle  is relative to the situation, a principle  can  still be objectively  right  or  objectively  wrong  as can  be the  case with  contradictory proverbs  (Putnam, 1981).15   In other  words, any contradiction is reconciled by recognizing that,  while the appropriate application is relative to circumstances, the application is objective  and not a subjective matter  in that  we have no problem  in saying which prov- erb applies in what  situation.

The appropriateness of a principle is tied to context,  that is, whether the principle is applicable  or not depends on the context  since context  suggests whether  it can be validly applied.  Principles of marketing or management emanate  from  the collective experience  of managers.  Interpreting a prin- ciple’s appropriateness is less a matter  of paying rigid attention to the rule so much as paying attention to the situation or circumstances to which it is to be applied.  Principles, like all rules, are guidelines not formulas  since there  is often  uncertainty as to  the  precise  circumstances to  which  they can  be  applied.  Sometimes  we  need  explanatory theory  to  justify  their appropriateness.

Some academics argue that research in marketing should focus on devel- oping principles,  advocating effects application research,  problem  orienta- tion research etc., without being concerned with explanatory theory. But, as Robert  Merton (1968) points out, such naive empiricism is likely to lead to the chaotic accumulation of miscellaneous  empirical generalizations—as it has in marketing.16 This is because empirical research is blind without some guiding theory just as theory without empirical research can be empty.17



Interpretations are guided by perspective or purpose. An advertisement for The  Times  of London  points  to  this.  It shows  a banana on  a plate  with six plates  that  correspond to six different  perspectives:  (i) banana signal- ing fruit, (ii) banana signaling vitamins,  (iii) banana symbolizing  slapstick comedy,  (iv) banana as sexual  innuendo, (v) banana as symbolizing  trade wars, (vi) the banana as a racist weapon.  The ad caption  simply reads: “if you take things only at face value, you miss what is important”.

The indexicality of a word, phrase or sentence is that part of its meaning that  is specific to the context  in which  it occurs.  Language  interpretation is always tied to context. Thus ‘like’ can be used to mean ‘fond of’, ‘enjoy’,

‘feel’ and so on depending  on the context  or the word  ‘novel’ can be inter- preted as a work of fiction or as something  original.  Meaning  is indexed to context. The indexicality  of a word  is unknowable without knowledge  of context. It is this indexicality  that  rules out  replicating  the exact  findings of any study as contexts  are never exactly the same. An amusing  example of how the meaning  of a word  is tied to context  is provided  by someone’s e-mail  to  his local  authority protesting the  erection  of some  building.  It never reached  the official because all the computers had an anti-spammer which rejected any e-mail with offensive language!

Context can  change  expected  behavior,   ruling  out  law-like  general- izations.  Contexts change  interpretations and  the  weighting  of the  vari- ous considerations, just as the context  in which a wine will be consumed changes  the  weights  attached to  price,  type  of wine  and  brand  bought. We predict  within  a context. Thus  people  do all sorts  of things  to draw attention to themselves like acting silly, dressing oddly and so on but not in all situations (like a job interview)  since they are very much  aware  of what  contexts  are appropriate for what  behavior.  Zimbardo (2007)  dem- onstrates the power  of context  or situations in warping  people’s judgment and  channeling  behavior  in unexpected ways.18   In particular Zimbardo argues  that  situational factors  (peer pressure,  superior  demands)  are  far more likely to explain  abusive and cruel behavior  to others  (e.g., the Abu Ghraib  prison  case) than  dispositional states like attitudes.

But what  exactly  is context?  Scharfstein  (1989)  includes  under  context temporal, geographical, cultural,  cognitive,  emotion  . . . anything  at all in the  relevant  environment . . . and  argues  persuasively  that  no  reasoning or any action  can be fully understood outside  of its own  context.19  This definition  of context  includes  the conditions operating at the time. Politi- cians promise  to undertake certain  policies once in office but  fail to keep their  promises,  commonly  because,  on recognizing  the restraining condi- tions confronting them,  they are apt to re-think  the wisdom  of what  they promised.

Scharfstein  argues  that,  if the grasp  of context  is purely  cognitive,  this limits  understanding. Thus  understanding the  action  of others  is always held back if we have never shared  (experienced)  the relevant  context. Per- haps this is why the senior citizen market  is neglected as those actively in marketing are not senior citizens. People commonly  say “I know  how you feel” to those who have suffered but this is just an empty phrase unless they have experienced  the same tragedy  in a similar context.  This suggests that personal experience  of the various  contexts  in which a product is bought, consumed  or disposed of is needed to fully understand the customer.  Hence it helps a great deal for a product manager  to belong to the market  segment he or she caters to.

In endorsing  the  claim  that  no  one  can  distinguish  the  meaning  of a word divorced from the context, Scharfstein is also claiming that to under- stand  human  beings, there is a need to understand the various  contexts  in which  human  behavior  is manifested.  However,  the degree of contextual detail we amass will depend  on our aims, both  intellectual  and emotional. For  many  purposes, we can  think  of context  as embracing  the  medium of communication, time and  location. As for the medium  of communica- tion,  interpretations differ between  words  as spoken  and  the same words as written:  “There  is no god but  God,  and  Muhammad is the apostle  of God”.  Take the problem,  too, of how meaning can differ through time. The description of Ivan IV of Russia as ‘Ivan the Terrible’ has come to signify a cruel despot  but this was not what  historically  ‘Ivan the Terrible’ implied. In the early 17th  century,  when the label was first used, the meaning  con- noted ‘awe-inspiring’ or formidable (Madariaga, 2005).20

Interpretation takes  context   as  background information in  making things  intelligible.  We stress the word  ‘intelligible’  rather  than  rational (as per the canons  of rationality). An error  perhaps  made  with  Saddam Hussein was to assume he would act in what would be considered  a ratio- nal  way  by  American  and  UK politicians. His  conduct  was,  however, intelligible  in the  context  of Iraqi  culture  and  the  contextual pressures on him to avoid losing face.



Self-interest and the values reflecting our concerns influence interpretation. Livingstone  (2003)  illustrates  this in discussing  the reception  of Darwin’s Origin  of the Species in New Zealand and South Carolina.21  In New Zea- land,  the  book  had  an enthusiastic reception  as the  book  seemed to  jus- tify the colonists’ attempt to extirpate the native  Maoris,  while, in South Carolina, the book had a hostile reception  as it suggested the close kinship between the local plantation owners and their soon-to-be-freed slaves.

Methodological Constraints on Interpretation: Methodological Monism, Methodological Exclusivism, Methodological Pluralism and Positivism

Many  deny there  are any serious  problems  of interpretation in doing  sci- entific  research.  Those  who  claim  this  tend  to  endorse  methodological monism:  the  notion  that  any discipline  that  aspires  to  be a science must follow  the methodology of the natural sciences where interpretation does not loom large. Methodological monism  is a core thesis of positivism that all scientific inquiry must, to be called scientific, follow the methods  of the natural sciences.

As most disciplines promote themselves as ‘sciences’, there is inevitably debate  over  what  is science. Dennett  (2006)  rejects  as scientific evidence the mass of data contained in historical  narratives on the ground  that such cannot  be reproduced under  controlled conditions.22  This would  rule out

‘natural’ experiments and a good deal of what  we call sciences. It reminds us how often definitions  are adopted to fit a viewpoint,  in this case to dis- miss the visions of saints and mystics as worthless  since they are not repeat- able. In any case, all ways of understanding do not fall under the rubric  of science, for example,  art,  music and  literature. But even if we follow  the methods  of the natural sciences, interpretation of data  and  the results  of scientific inquiry can still be a problem.

The most  extreme  version  of methodological monism  is the twentieth- century brand  of positivism known  as ‘logical positivism’, a product of the so-called ‘Vienna Circle’ meetings in the 1930s.  Its tenets were:

Empiricism: positivists  confine  ‘reality’ to  that  revealed  by experience (mainly  sensory)  claiming  that  what  we know  we know  only because  the empirical evidence so far happens to point that way. Not appreciated was the fact that  this experience  needed to be interpreted and interpretation is tied to  the  scientist’s perspective  or scientific paradigm. Empiricism  contrasts with  rationalism,  which  claims that  the world  is knowable only through reason,  since sense data  need to be connected  (interpreted) in the light of reasons.  The rationalists deduce facts about  the world through the exercise of reason while the empiricists argue that the only way to an understanding of the world is by observation and experiment. Mathematics is the ideal for all rationalists, starting  with Descartes,  Spinoza and Leibniz.

Handy and Harwood, who are supporters of a strong positivist tradition, argue that rationalism is still the dominant orientation among formal model builders, giving rise to models like “game theory” and “utility  theory” that confuse  warranted assertions  about  the particular model  with  warranted assertions  about  some aspect of human  behavior.23 They take model build- ers  to  task  for  not  investigating  the  presumed  connections between  the model and observed behavior  with any degree of thoroughness: typically it is the internal aspects of the model that are examined  rather  than matching the model to actual behavior. This is still as true today as it was at the time Handy  and Harwood wrote it.

Handy  and Harwood point  out that  internal tests are seldom adequate since  assumptions can  often  seem unchallengeable, reasoning  absolutely sound,  and  conclusions  inescapable,  when  in instance  after  instance,  the assumptions are later shown to be unfounded, the facts proved wrong, and errors  in reasoning  detected.  They take  econometricians to task  for often “obtaining plausible  numbers  to  provide  ceremonial  adequacy  for  a the- ory.”  A little cleverness “will get you almost any result you want” and that is why “few econometricians have ever been forced by the facts to abandon any firmly held belief”. These criticisms are not easily dismissed.

Empiricism can equally be contested  on the ground  that even the natural sciences must make assumptions, like assuming  uniformity in nature  that cannot  be empirically  verified. But what  this debate  is about  is where the relative dominance lies since both inevitably play a part in scientific inquiry. Descartes undertook the most original experiments in optics though  believ- ing that  the way to understand Reality  was through mathematics. In any case, interpretation is at work whatever approach is used though interpreta- tion is more fundamental to empiricism.

The  logical  positivists  put  great  emphasis  on  ‘observables’ though  the interest today lies in the probability distributions associated  with the obser- vations, not in a single observation. This is what the statistical  revolution in the 20th century has been about  (Salsburg, 2001).24 The fact is that empiri- cal evidence for most decisions  is just not  there.  David Eddy, a pioneer  in the health-care quality  field and  in the application of statistical  modeling to medicine,  claims that  only about  15%  of what  doctors  do is backed  by hard  evidence; others  put it around 20%  to 25%  (Carey, 2006).25  This, of course, could be a reminder  of the need for more empirical support in justi- fying decisions, since the quality of decisions depends vitally on the quality of the information behind the decision.


Descriptive  laws:  science  to  the  logical  positivist  is the  search  for descriptive  laws,  e.g.,  ‘when metals  are  heated  they  expand.’ The- ory was viewed as systematizing  descriptive  laws and any theoretical entities not completely definable in observational terms were rejected. Abstract  concepts  like ‘attitude’, ‘motive’, ‘intention’ had to be given operational definitions  or operational measures  so as to have a con- crete, observational reference. But observations are not unproblematic since observations are interpreted in the light of some conceptual sche- mata,  perspective or scientific paradigm. The Vienna Circle members saw mathematics as essential to describing  physical laws and turned to  Bertrand  Russell’s program to  reduce  all mathematical concepts and truths  to pure logic. (The program never succeeded,  though  it is now agreed that  99.9% of mathematics follows from a small part  of the axiomatic theory of sets.)

  • Nominalis : logical  positivists  recognize  only  individual   particulars, denying that  general abstract concepts  like ‘society’ or ‘market’, ‘beauty’, ‘goodness’ offer any additional insight onto the world.  Mar- garet Thatcher, when the British prime minister,  seems to have been a nominalist in denying  there  was  any  such  thing  as ‘society’! For the logical positivists,  science starts  with direct observation of single facts as if the facts were out there like apples on a tree waiting  to be picked.
  • Teleological explanations, that  is,  explanations in  terms  of  functions,  goals, and  purposes  and  so on were considered  invalid  unless transformed into  non-teleological form.  In other  words,  science was to  avoid  interpreting things  in nature  or  social  life in terms  of the function  performed (as when we refer to someone fulfilling the role of buyer or researcher)  or in terms of purpose  (as when we say the con- sumer’s goal is to choose the cheapest  coffee from among  the brands available).  Such is not  acceptable  unless  translated into  a scientific (law-like)  format.  Teleological  explanations in practice  have  defied such translation.
  • Meaningful statements are  either  synthetic   or  analytic A  synthetic statement is an empirical  one (all buyers are risk-averse),  with observable  facts relevant to its truth. On the other  hand,  an analytic statement (a purchasing agent is someone  who buys on behalf of an organization) is true as a matter  of definition or just follows as a mat- ter of deduction from the meaning of the words used in the statement (e.g., a bachelor  is unmarried). Any denial  of an analytic  statement involves self-contradiction. The Austrian  school of economics,  asso- ciated  with  such luminaries  as Von Mises and  Hayek,  claims to be based  on analytic  propositions or self-evident  axioms  about  human behavior.  Synthetic  statements are  to  be tested  by verifying  them. This was enshrined  in the logical positivist’s verifiability  principle. Any assertion  not conforming to the verifiability principle was either analytic (not in need of any confirmation, being a conceptual or defi- nitional  truth)  or ‘nonsensical’ (just emotive  as in ethics). All scien- tific propositions state something  is or is not so. Propositions about ethics, religion, and aesthetics are in consequence  cast aside as unsci- entific. Whether  these topics are unscientific or not,  they are full of meaning  (significance) for the human  race and it seems an absurdity to attach  to them words  like ‘nonsensical’ (even if just non-sensical).

In logical positivism,  we are being asked as a first step to interpret whether a statement is analytic,  synthetic  or nonsensical  as these distinctions influ- ence all else. The logical positivists  aimed  to  dispense  with  metaphysics, but  an unintended consequence  has been to undermine the study  of phi- losophy  since most  of it fell into  the category  of the nonsensical. On  the other hand,  there is a reminder  here how common  it is to find some propo- sition  being paraded as empirical  (synthetic)  when  it is analytic,  simply a conceptual truth  like saying the stronger  the desire for some product, the more  the motivation to obtain  it. And it is equally  common  to find views expressed as ‘obviously true’ (analytic) when evidence is needed in support. In life generally,  it is impossible  to have empirical  support for everything we claim. If what  someone  says or writes ‘makes sense’, forms a coherent argument, then  others  are apt just to go along and demand  evidence only when their concerns  oblige them to do so. A.J. Ayer (1936)  recommended a weaker  version  of the  verifiability  principle,  namely,  that  a sentence  is factually  significant  to  someone  if, and  only  if, that  person  knows  what observations would  lead him or her, under  certain  conditions, to accept it as being true or reject it as being false.26  Ayer (1973)  was to claim that  his weaker principle made sense as it avoids condemning as nonsense scientific laws not reducible to descriptive experience.27

The distinction between  analytic  and  synthetic  statements is still com- monly (and usefully) made. We all need to be aware of what statements are true  as a matter  of logic and  what  statements require  empirical  support. Quine  (1951),  though, points  out that  analytic  statements are not immune to empirical  revision  as all beliefs are answerable to experience.28  Kripke (2004) adds the concept of necessary a posteriori truths, quoting  examples that  were neither  simply synthetic  nor  analytic.29  (Propositions are called

‘a priori’ or,  alternatively, ‘a posteriori ‘depending on  how  they  relate  to experience:  a priori  if they come before experience  and a posteriori if they put across experience.)


Carnap, a prominent member  of the  Vienna  Circle,  substituted the  term logical  empiricism  for  logical  positivism.  Logical  empiricism  is a  much more  sophisticated version  of logical  positivism  with  the  goal  of science being explanation (not mere description) but continuing to insist that scien- tific hypotheses  be testable and potentially falsifiable. The particular brand of logical empiricism that presently seems to hold most sway is naturalism. For the naturalist, the only scientific explanation is the causal explanation. Naturalism in the philosophy of science has become the current  orthodoxy though  there are many critics (see Rea, 2003).30

Naturalism endorses  the methods  of the natural sciences in interpreting reality,  seeing the natural sciences as the authority on what  there is in the world and what the world is like. However,  while naturalism accepts meth- odological  monism it rejects the notion  that science must be built on direct sensory  experience,  never  going  beyond  what  is observable. It acknowl- edges that every scientific term cannot  always be defined operationally, that is, captured or  measured  in observational language.  This  is in line with modern  physics, which  no longer  insists on operational definitions  for all concepts  employed  in a theory,  since a concept  may be useful even if never observed,  like the  electron.  What  naturalism, however,  does  insist  on  is that scientific explanations be causal, acknowledging there are many kinds of causal explanations.


A parallel  claim  to  that  of methodological monism  (the  belief that  there is only one set of scientific methods  and these are the methods  used in the natural sciences) is that the study of human  beings requires a distinct meth- odology  of its own,  borrowing nothing  from  the methods  of the natural sciences. Both  methodological monism  and  the  counterclaim for  distinct methods  for studying  human  action,  Roth  categorizes  as methodological exclusivism.31  Methodological exclusivism is not just confined to positivist writers on social science like Rudner32 who are methodological monists but those like Winch who claim the social sciences require  a distinct  method- ology of their own.33  Winch (1958)  argues that,  if the objects of study are essentially sensory data, they can typically be studied via the methods of the natural sciences. But if the object of study is human  beings, acting in a way that  expresses a way of life, such a study  comes under  the heading  of the humanities and calls for methods  distinct  from the natural sciences. There is a danger  today  of replacing  methodological monism  with the claim that the social (human)  sciences require a unique  methodology of their own.


A complete  denial of methodological monism  is methodological pluralism that  rejects  the  claim  that  there  is any  one  set of methods  that  provides a privileged  access to reality  and  truth. Methodological pluralism  implies we can  be an  anti-positivist when  rejecting  methodological monism  but still access, when appropriate, the methods  of the natural sciences to study human  behavior.

Methodological pluralism  rejects any claim that  there is just one set of methods  that  gives privileged  access to  studying  and  explaining  human behavior.  Whatever  the controversy over Feyerabend’s (1977) book Against Method, with its anti-objectivism thesis, it has wide appeal in arguing that there  is no  one way to  conduct  successful  science and  science cannot  be restricted  to following  one set of rules, regardless  of subject matter;  there are just ‘different methods  for different topics’.34  Interpretation alone, with its focus on meanings  and intentions, will not answer  all questions  asked. As Fay (1996) says, social scientists ask questions  not only about  the mean- ings (significance) of various  acts but also want  to know  about  the causal factors  which  give rise to and  support the continuing existence  of certain meanings.  He or she will want to identify the causes of actions.

‘Critical pluralism’ is methodological pluralism  with the recognition of the need to subject  all theories,  models  or hypotheses  to critical  scrutiny. In philosophy, there  has been an undermining of faith  in universal  laws, absolute  proof  and disproof  and related  notions  such as empirical  verifica- tion, the possibility  of a neutral  observation language,  uninterpreted facts, value-free judgments  and the correspondence theory  of truth  (truth  as cor- responding to the objective facts in the world outside) as representing ratio- nality at its best. Even physicists are beginning to entertain the notion  that the laws of nature  might not be fundamental in that  they might not apply to other universes.

The attraction of methodological monism  (as opposed  to methodologi- cal pluralism)  is that,  in insisting  on the methods  used in the natural sci- ences, it dictates  what  type of evidence is acceptable  as ‘hard’ evidence. In a world  where  absolute  proof  is unobtainable, this seems important. Not surprisingly,  many  worry  about  the  relativist  slant  suggested  by an ‘any- thing goes’ position. Even if it is not exactly  a case of ‘anything goes’, the assertion  that  any  justification  procedure is simply  whatever  is accepted by the scientific community for that  discipline  (as suggested  by Kuhn35  ) seems to make the scientific review process sound  like a ‘popularity’ con- test. Hence some writers argue there must be universal, objective standards or rules for the conduct  of science and scientific thinking,  just as there are rules for valid deductive arguments. In a deductive argument we infer from premises to conclusion  as in the syllogism so beloved in logic texts: All men are fallible, Socrates is a man, and therefore  Socrates is fallible. The prem- ises logically entail  the conclusion, making  the argument a valid one. But only if the premises are true is the conclusion  also true.  But an alternative position  is that there can be premature closure on methods  with the danger of rationally defensible methods  being excluded.

Although  ‘anything goes’  was  the  slogan  Feyerabend  (1977)  used  to sum up his position  on choosing  a methodology, Feyerabend  was not  (as commonly  claimed)  saying rationality should  be abandoned but  insisting that  methods  be evaluated  by results  and  not  by their  adherence  to some set of dogmatic  guidelines.36  He was not  recommending that  scientists  or researchers proceed without rules but that they should expand the inventory of rules, with  the recognition that  there  are standards operating ‘locally’, tied to a specific research process: his intention was not to reject rationality but to recognize it takes many forms

Neither   the  methods   used  in  the  natural  sciences  nor   interpretive approaches are certain to yield true knowledge.  Quine (1970),  the philoso- pher,  talks  of the underdetermination of theories  in that  it is possible  to formulate scientific theories  that  are  empirically  equivalent  but  logically incompatible.37  But what  is logically possible need not be probable. In any case, underdetermination is not universal.  Kitcher (2001) illustrates  this by pointing  out that we still seem unable to think of a rival hypothesis  to that which  states  that  the typical  structure of the DNA  is a double  helix with sugar-phosphate backbones and bases jutting  inwards.38

For Quine,  theories  in the natural sciences are not  a mirror  of reality as there  is ‘no unvarnished news of the world’. Quine  sees knowledge  as a combination of sensory  evidence  and  subjective  creation  (construction) and denies we can distinguish  these two elements in any analysis of knowl- edge. Quine’s  (1970)  ‘indeterminacy of translation  thesis’ maintains that there  are  no  universal  meanings  or  logical  standards through which  we can arrive at some uniquely  correct  interpretation of the utterances of oth- ers. He stresses his ‘indeterminacy of translation’ applies to all psychologi- cal theories  that  rely on the interpretation of verbal behavior  as data  (e.g., answers  to  a questionnaire).39  This  has  relevance  to  marketing research. He  shows  that  researchers  can  never  be absolutely  sure  their  interpreta- tions  reflect the structure and  meaning  of the thought which  the speaker intended  to communicate. But then no scientists can be absolutely sure their theories reflect absolute  truth.

There are no impartial observers of behavior;  we deceive ourselves if we think there are. We are not even sure of the truth  conditions for employing the  concept  of impartiality. All interpretations possess  a quality  shaped by past experiences,  interests,  and what  things mean to us: we are not just cameras  selecting and recording  various  scenes but infuse the scenes with something  of ourselves. It is not just the Eiffel Tower  that  registers but my Eiffel Tower colored  by my own past and its meanings.

The best defense of methodological pluralism  or critical  pluralism  rests on the observation that  different  methods  address  different  questions  and that  different  methodologies go with  different  explanatory systems.  If we insist  on  a methodology that  is quantitative, this  limits  the  questions  we are able to ask. There  is the inherent  danger  that  the questions  addressed will be those that  fit some favored  technique;  the researcher  acting like the little boy with a hammer  who finds everything needs pounding (or it may be that,  when all you have is a hammer,  everything looks like a nail). Different explanatory systems  or  paradigms  represent   different  conceptual  lenses through which  to view the world  and  may seek to answer  different  ques- tions.  What  caused  A to do B? What  function  was performed by A doing B? What meaning does doing B have for A? If we are interested  in questions about  inner mental states, we do not go to radical behaviorism for answers. If we are interested  in cultural,  social and emotional influences on behavior, we are unlikely to look to cognitive psychology and so on.

Krausz (1993) illustrates  how the particular explanatory system adopted determines  interests  addressed.  Thus  he argues that  a Marxist interpreta- tion  of Van  Gogh’s Potato  Eaters  would  be superior  to  a psychological interpretation in terms of its power to reveal the relations  of economic insti- tutions, but a psychological  interpretation would  be superior  to a Marxist interpretation in its power to reveal the character of its leading figures.40


Popper (1959) substituted falsifiability for verifiability as the necessary con- dition  for any hypothesis  to be interpreted as potentially scientific on the ground  that scientific theories or hypotheses  can be falsified but never com- pletely proven.41 This claim by Popper is still quoted  as orthodoxy by many in social science, though  Duhem,  a French  physicist,  early on in the 20th century demonstrated that the falsifiability of a scientific law in an absolute sense is also not demonstrable.42 In any case, social science theories  do not commonly  come  along  with  obvious  ways  of testing  them.  Determining how  to  test a theory  may require  considerable ingenuity,  more  than  that needed to think  of the theory itself.


The verifiability principle  of the logical positivists  went hand  in hand  with the correspondence theory  of truth.  This asserts that  something  is true or can be interpreted as true, if it corresponds with the ‘facts’: the idea of the world consisting of unambiguous facts to be objectively observed  and gen- eralized  about  was a central  tenet  of positivism.  However,  the correspon- dence theory  of truth  is less operational than  it seems, once we recognize that  the notion  of truth  is semantic  in that  it depends,  first and foremost, on the interpretation of the meaning of the expression  whose truth  is being determined. (Hence that  favorite  retort:  “It all depends  on what  you mean by . . .”).  In any  case, if we were  to  fully accept  all the  tenets  of logical positivism today,  there would be little in social science that would pass the logical positivists’ criteria for being a science.

Anyone who doubts this might consult the (already cited) Handy and Harwood’s (1973)  review of the social sciences from  the point  of view of logical positivism.43  There is not much they acknowledge as science among the so-called social sciences. However,  when we speak of positivists  today we are not  talking  about  those  who  subscribe  to the doctrines  of logical positivism  but to those who focus on empirical  observation, causal expla- nation, experimentation, measurement and testing. There is the assumption we can avoid interpretive bias stemming from preconceptions, self-interests and  sympathies.  But bias is a problem  for all ‘impartial’ inquiries.  Kagan (2006),  a Harvard psychologist, says it took  him  years  to  shake  off the prejudices  against  biology  acquired  from behaviorism and  psychoanalytic theory.44 As he says:

The ideas indoctrinated during graduate training  can limit the concep- tions the mature  investigator entertains. I used to begin the first meet- ing of my graduate seminar  by telling  the  dozen  or  so students  that much of what  I had been taught  at Yale turned  out to be mistaken,  so they remained  skeptical of everything I said over the next four months. (Kagan, 2006,  p. 112 )

The indoctrination to which Kagan refers mainly applies to his training  in behaviorism and  psychoanalytic theory.  His  verdict  seems a little  harsh. Every paradigm like behaviorism in the social sciences is a way of seeing but also a way of not-seeing;  answering  different  questions  or offering dif- ferent  windows  onto  a problem,  with  some windows  clearer  than  others, depending  on the questions  being addressed.

In life generally,  we have perspectives  that  lead to bias. Judson  (2005), challenging alleged differences in the sexes, points out that when American symphony  orchestras introduced blind  auditions in the 1970s,  where  the musicians  being evaluated  played behind  a screen so gender was invisible, the number  of women  offered  jobs in professional orchestras increased.45

Posner (2004), in discussing the International Court of Justice in The Hague, maintains the judges, 90%  of the time, vote in favor  of their  countries  if they  are  parties  to  the  litigation;  vote  for  states  that  are  more  like their home  states;  favor  wealthy  states  if their  home  states  are  wealthy;  favor poor  states  if their  home  states  are poor,  while judges from  democracies appear  to favor  democracies  and  judges from  authoritarian states  appear to favor authoritarian states.46  We would  find analogous (if less apparent) biases among judges elsewhere, which is why there are appeal courts. In the news media political bias is pervasive. What The Economist (May 5, 2007, p. 11) says about  the motto  of the pro-Republican Fox News (‘We report, You decide’) is “about as convincing  as an anchorman’s suntan.” Regular viewers of Fox News  claim it is the least biased  of the news channels.  A news report  may not obviously lie but simply ignore contextual factors that would induce a different interpretation.

In academia there is a good deal of groupthink, with faculties often selected on the basis of likeness in perspective.  Thus the mathematical model build- ers may determine  such skills to be the basic criteria for selection. The result is a sort of intellectual  incest prone to the PLU syndrome;  only ‘People Like Us’ should be considered.  One former president of Yale doubted whether the truly innovative  could overcome the collective bias to get tenure.  I hope this is not true but perhaps  it needed to be said to remind us of how perspectives grounded in our commitments guide our judgments.

Can We Interpret Statements as Simply True or False? Austin’s

Performatives and the Analytic/Synthetic Distinction

John L. Austin, a linguistic philosopher, pointed  out that  there are classes of utterances that  are perfectly meaningful  but cannot  be said to be either empirically  true or simply analytic.  To say, for example,  “I promise  I will buy you that bicycle tomorrow” is neither true nor false, neither describing nor evaluating  but simply doing or acting.  Only humans  can make prom- ises and assume sufficient commitment to retain  that  promise  in memory. This does not rule out interpretation and no statement is immune from mis- interpretation. Austin  called such utterances ‘performatives’ (though  they never quite lived up to his claim about  them never being true or false).

Austin’s (1962) book How to Do Things with Words has become a major source for speech act theory, with its distinctions between locutionary, illo- cutionary and perlocutionary ‘forces’ in speech acts or utterances.47  These distinctions are useful in the interpretation of speech acts.

All three speech acts are present in a speech act. Locutionary utterances say something  with  an inherent,  public  meaning  (that  is, meaning  appar- ent to users of the language,  e.g., “I will be there’). The expression  of any proper  sentence  is a locutionary utterance. Locutionary speech acts con- tain illocutionary acts since what  one does in saying something  like ‘I will be there’ is carry  out  a speech act that  declares  intent.  Illocutionary acts have intentional meaning with the force of affirming, promising  , denying, vowing,  diagnosing,  suggesting,  thanking, appointing etc., and  entail  the execution of some  recognized  type  of action  of intent  such  as “I  do”  at a wedding  ceremony.  In other  words,  they involve interpreting  intention, purpose, reason  or motive.  Thus  choosing  a product and  walking  toward the checkout  register signifies intent  to buy.

Correctly  interpreting an act as illocutionary implies the total  context points  to intention without the need to ask about  desires and beliefs. Thus the interpretation of intentions is a matter  of knowing  the context  in which the speech act occurred.  Context is all important. It is common  in politics (and elsewhere) to take an opponent’s remarks  out of context,  which can be damning  when subtracted from context.  The combination of the locution- ary and the illocutionary speech acts gives rise to perlocutionary utterances. The perlocutionary act is the act you succeed in performing by means  of the two preceding speech acts. Consequential meaning is involved because the utterance has consequences.  Thus “I will be there”  has the consequence of getting the hearer to take this into account.

Austin’s  distinctions stress  the  importance of performative utterances in interpretation and draw  attention to the fact that  interpreting for truth and falsity characterize but a relatively small set of utterances. The larger set is made up of performative utterances like promising,  consenting,  veto- ing,  approving etc.,  where  the  utterance itself is the  performance of the language act and not a report  of that performance.

Treating  utterances as intentional acts, we can ask “How many kinds of illocutionary acts are there?”  Austin posited  five basic types of illocution- ary act which provide categories for interpreting speech of any sort:

(i)  Assertions or intention to tell people how things are, as happens when salespeople talk about  their product.

(ii)  Directives or intention to get people to do things, as when salespeople try to close the sale.

(iii)  Commissives or intention to commit ourselves to do things, as occurs with promises to buy.

(iv)  Expressives or intention to express our feelings and attitudes, as when we comment  on the service in a restaurant.

(v)  Declarations  or intention to seek to bring about  changes in the world through our utterances so things are changed in line with the content  of the utterance, as occurs in registering a complaint about  the service.

Austin’s work  has instigated  two  forms  of analysis,  namely,  conversation analysis and discourse analysis.

Conversation analysis (CA) links not only to Austin but to Garfinkel’s ethnomethodology48  in that  it aims to describe  how  people  produce orderly  social interaction or how  conversations are coordinated as a basis for interpreting what is going on.

Discourse  analysis  (DA)  focuses  on  the  analysis  of  recorded  talk, going beyond  ordinary conversation since discourses  can take  place in institutional settings.

Austin was fond of noting  features  of language  that  surprise  us. In one lec- ture at Columbia University he pointed  out that,  while a double  negative is equivalent  to a positive, never does a double  positive amount to a negative. From  the audience  the familiar  voice of philosopher Sidney Morgenbesser dismissively called out,  “Yeah,  yeah”  (Ryerson,  2004).49  This is an exem- plary illustration of how something  that  is spoken,  as opposed  to written, can alter an interpretation. ‘Yeah, yeah’ could in fact have been said in a way that  signified agreement  but instead  was said in a highly skeptical  manner leading us to acknowledge that a double positive can amount to a negative.


If we could  explain  human  behavior  as falling under  some law, we could simply infer the behavior from the law. No conjecture as with interpretation as it falls under a nomothetic (law-like) explanation. A nomothetic explana- tion invokes universal laws for explaining  repeatable events and processes. Although  we talk of laws in the natural sciences as being unconditionally universal, in many branches  of the natural sciences, laws are stated as being universally valid only under certain ‘ideal’ conditions, for pure cases of the phenomena being discussed. The discrepancies,  however, between what the scientific law asserts and what  observation discloses can be accounted for by well-authenticated discrepancies  between  the ideal conditions and  the actual  conditions being observed.  This is where the natural sciences score over the social sciences. Even in economics,  the discrepancy  between  the assumed ideal conditions for an economic law to apply and the actual con- ditions in the market  are usually so huge and the postulations needed to fill the gap are so tricky and complicated that  the strategy used by the natural sciences is infeasible.

Ideographic  disciplines seek to understand the unique  event, as history sometimes  claims it does. Ideographic explanation is associated  with pro- cess tracing or genetic explanation which  traces  the set of causal  factors giving rise to the situation. It is an explanation that  links to the historian but it can be used in tracing  the historical  (causal) antecedents in reaching any present situation (see Chapter 6).

If we  think  consumers  for  some  purposes  are  tokens  of  each  other who  operate  in  strictly  defined  contexts,   we  may  adopt  a  nomothetic approach and  seek universal  law-like  findings. If we believe each of our subjects  is essentially  different,  we adopt  a more  ideographic approach. The  adoption of a nomothetic approach does  not  necessarily  imply  we actually  subscribe  to  the  notion   that  people  are  exact  tokens  of  each other but simply that the approach may provide the best working  hypoth- esis for the research  at hand.  Consumers are not tokens  of each other  in wants  or behavior.  Even in eating  the same food  two  people  experience profoundly different  sensations.

There are no universal laws that apply to purposive  behavior.  If there are no universal laws on buying behavior or elsewhere in social science, it is pre- sumptuous to give advice as if there were. Specific advice depends  not just on knowing  social science findings but on knowing  contexts.  This is where experience comes in: the manager’s knowledge  of contexts  is all-important if ‘expert’ advice is to have relevance. What  every manager  has to avoid is being manipulated by pseudo profundity where the advisor starts by slowly asserting some truism (‘Quality decisions presuppose quality information’) and  moves on to claims couched  in jargon,  suggesting  both  sets of state- ments can be equally accepted as true.


The  question   arises  as  to  whether   in  interpreting or  indeed  explaining human  behavior  in markets,  families,  businesses,  social  classes,  or  deci- sion-making units requires  anything  more than  aggregating  the individual interpretations or  explanations of action.  This  question  is answered  dif- ferently  by “methodological individualists” and  “methodological holists” who represent  two contrasting perspectives in social science.

Methodological holism focuses on social wholes, not individuals, as the building  units of social science. Emile Durkheim (1858–1917), the French sociologist,  saw  holism  as support for  the  distinctness  of sociology  as a social science and claimed that,  while social forces work through individu- als, social facts influence and constrain individual  behavior.  Methodologi- cal individualism, in contrast, focuses on the individual  agent.

Methodological holism has attractions for social scientists in that  their interest  generally lies in the behavior  of groups,  not the differences among members,  while causes in social science apply more to groups than individ- uals. On the other hand,  people are not tokens of each other,  nor do group norms enforce complete conformity.

If  methodological  individualists discount   the  scientific  usefulness  of social wholes, methodological holists discount  the influence of individuals when considering  social behavior.  Holists  regard  social groups  ‘as if’ inde- pendent  of their  members  as individuals. It asserts  that  theories  of social behavior  are not reducible  to the behavior  of individuals  and those collec- tive entities  like “social  group” are not  specifiable in terms  of individual behavior.  Methodological holism postulates that  social wholes, like firms, competitors, and  society, have functions, can cause events to happen  and can cause individual wants, beliefs and actions to change. It does not regard social phenomena as reducible to individual  psychology.  Holists reject such “reductionism” entirely.

Schumpeter  and  Hayek,  the  economists,  and  Popper,  the  philosopher, claim  (as did  John  Stuart  Mill  in the  19th  century)  that  social  phenom- ena are wholly explainable in terms of facts about  individuals. Schumpeter referred  to this as methodological individualism. Watkins  neatly puts for- ward the claims of methodological individualism that:50

• The ultimate  constituents of the social world  are individuals  who act more or less appropriately in the light of their wants and beliefs given the situation.

• Every event or institution is the result of a particular configuration of individuals, their  dispositions, situations, beliefs, physical  resources and environment. All interpretations or explanations of human actions, achievements, etc., stem from the goals, wants,  beliefs, resources and the interrelations of individuals. Social institutions and social change are explained  by showing how they come into being as a result of the actions  and  interactions of individuals. For  example,  whenever  we refer to diffusion theory  in marketing, based as it is on the behavior of individuals, we assume methodological individualism.

Watkins acknowledges that interpreting group behavior presupposes under- standing  social facts like a society’s institutions or its bureaucratic struc- tures but argues such understanding is also needed to interpret individual action  since actions  are  taken  within  a temporal and  contextual frame- work.  While agreeing that the behavior  that characterizes a group may not necessarily  be a simple summation of individual  behavior  and  that  group concepts and explanations may on occasions be useful, he argues that they are never “rock  bottom” unless built  up from  explanations of individual behavior.  Watkins  concedes  that  some social regularity  is inexplicable  in individualistic terms  if they are the outcome  of a large number  of “acci- dents”  or simply the “automatic” group  behavior  such as that  which  fol- lows an earthquake. Nozick (1974) endorses this view in claiming it makes no sense to think  of society making  choices, as only individuals  can make choices.51  And Kenneth  Arrow  (1963),  the economist  shows,  on the most plausible  of assumptions, that  there  is no  rule  for  combining  individual preferences  into a social choice that does not generate paradoxes.52 But the debate  continues  as is apparent in Amartya  Sen’s (2003) recent book.53

For  methodological individualism, all  social,  political,  economic  and marketing behaviors  are capable  of being interpreted in terms of the unin- tended/intended  consequences   of  the  actions  of  individuals. Even  large social  processes  like inflation  and  the  trade  cycle are viewed in terms  of individual  behavior. Methodological individualists consider  concepts  like “group mind”  “national mood”, “institutions” or “classes”  as either reified nonsense  or analyzable  into  the actions  of individuals. A related  concept, “psychologistic individualism”, identifies each individual  with a given psy- chological  state.  Thus  neoclassical  economics  is based  on  psychologistic individualism  in  that  it  identifies  every  individual   with  his  or  her  util- ity  function. However,   methodological  individualism  need  not  embrace psychologistic  individualism. Thus, while Popper (1972) subscribes to methodological individualism, he identifies individuals  with their problem- situation and not with a psychological  state.54  Yet Popper  believes institu- tions are never entirely explained  in terms of individuals  though  regarding all institutions as the creation  of individual  decision makers.  This position is highly defensible.


The reason that prediction is successful in the natural sciences is that scien- tific laws working  at the level of the individual  atoms can be integrated into new laws as we move up to more complex  systems. For example,  the laws of electrical charges bring about  those of thermodynamics and chemistry. It is interesting  to reflect on the physical sciences, as opposed  to the social sciences. Thus  all species of elementary  particles,  like electrons,  manifest absolutely  no individuality but  are completely  identical.  The result  is that elementary  particles  give rise to an unusual  interdependence, as described in quantum theory.  As Pesic (2002)  points  out, chemistry  like physics also depends  on this loss of individuality.55  But consumers,  as people,  are not tokens  of each other but possess individuality.

Reductionism, as advocated by the  logical  positivists,  seeks to  reduce all  the  macro-sciences  to  the  micro-sciences  so  that  psychology  (say)  is finally explainable in terms  of neurobiology, which,  in turn,  is explain- able in terms of physics. Reductionism in psychology is the conviction  that mental states, events and processes can be shown to be neural occurrences. Unfortunately, we do not have any cluster of laws in psychology  to reduce and, even if we did, we do not know  how such laws could be derived from biology  or physics. But, nonetheless,  there  is still this pursuit  of reducing mental states to neural occurrences.  But we are a long way from being able to interpret mental happenings in terms of neurobiology.

What  about  the natural sciences? Even in biology,  few believe that  bio- logical processes can be understood by just studying  genes and molecules. A major  unsolved  problem  in physics  is how  Einstein’s theory  of gravity (general relativity  theory)  can be united  with quantum theory.  String the- ory (now  called M-theory) is the most  popular approach to this problem at  present.  It is promoted as something  that  can  explain  all the  laws  of physics and all the forces of nature:  a reductionist dream.  String theory  is pursued  as a ‘theory of everything’, as advocates  of string theory claim that string theory  embraces  both  gravity  and  quantum mechanics.  The theory posits  that  the basic constituents of matter  and  energy are not  point-like particles but infinitesimally tiny wriggling strings and loops that vibrate in 10 dimensions.  What appear  to be different particles representing electrons and  quarks  are simply different  ways for the strings to vibrate:  vibrations that give rise to all the forces and particles in the physical realm.

String  theory  is a visionary  interpretation of reductionism in physics. Those  advocates  of string  theory  are  ideologically  followers  of Einstein, who in his later life sought unified field theories while in disagreement with those  like Niels Bohr,  who  embraced  the quantum revolution in physics. Einstein  rejected  the quantum revolution, which  started  with  the sugges- tions  that  light  and  heat  radiation are  emitted  in small  packages  called quanta, and instead  fell back on his imagination and reason  alone to start a second revolution.

String theory has not yielded to empirical verification but the sheer reach of the theory,  its beauty and elegance makes it too promising  to let go. The absence  of experimental evidence in support is not  decisive (after  all, no hard  evidence is just that—no hard  evidence) for a theory  that  eventually might reconcile quantum mechanics (which governs all particles) with gen- eral relativity  theory  (which describes how matter  and gravity interact  on the larger scale). But Laughlin  (2005)  is skeptical,  regarding  string theory as without practical  utility other than to sustain the myth of some ultimate theory of everything.56 For him, string theory is an exemplar  of the ‘Deceit- ful Turkey’; a beautiful  set of ideas that always remain out of reach.

From  the point  of view of this book,  string  theory  represents  a revolu- tionary  new perspective for interpreting reality which has given rise to lots of misunderstandings and  detractors (Horgan, 1996).57  But ultimately,  as a claim in physics,  string  theory  must  be subject  to experimental testing. Physicists know this all too well, since a single experiment, the Michelson- Morley  experiment carried  out in 1887,  dispelled the notion  of the ‘ether’ (that hypothetical medium assumed to be a necessary condition to support the propagation of electromagnetic radiation). But this will not be easy, as string  theory  does not  throw  up a single model  of physics but  trillions  of models,  with Susskind (2005),  in a book  on the basic concepts  of particle physics, viewing each potential model as corresponding to another universe as real as our own!58

Many in social science find it inconceivable  that the concepts and expla- nations  of physics will be able to capture  the whole of reality as suggested by reductionism. But eliminative  materialists  claim that,  at least in prin- ciple, it is possible  to  explain  all behavior  without reference  to  anything happening in the  mind;  reducing  everything  to  neurology.  Critics  argue that  explanations of behavior  that  confine  themselves  to  physiology  and neurology  are  capable  of explaining  only  involuntary behavior  (physical movements)  but not intentional action; a difference often illustrated by the difference between the ‘blink’ and the ‘wink’. There may be only one expla- nation  of a reflex movement  but human  action is interpreted in terms of the context, so a raised hand  can be to call attention, or be an attempt to get goods on the top shelf or an attempt to hit someone!  For in science, many concepts  only apply  to macro-phenomena in that,  say, ‘temperature’ and ‘pressure’ only make any sense at the macro  level since an atom alone can- not have pressure or temperature.

Methodological holism is anti-reductionist on the ground  that  there are autonomous levels in science that are not reducible to explanations at lower levels. While methodological individualism in social science could be inter- preted  as reductionist, it implicitly  recognizes  the  problems  of excessive reductionism by staying at the level of the individual.

The debate between holism and individualism has a long history in soci- ology (e.g., the individualistic approach of Max  Weber  versus the holistic approach of Durkheim). In a way, it is a reflection of a modern  version of a still older debate  between  ‘realism’ and ‘nominalism’. Nominalists claim that only individual  things are real, that universal categories like man, soci- ety, and  market  are mere names  applied  to classes of things.  Those  who stress  context  and  symbolism  and  underwrite uniqueness  come  close  to supporting nominalism. Today, the debate is less heated as few of us believe we must  take  sides.  To  many  social  scientists,  the  basic  question  is not whether  group  concepts  or group  explanations are replaceable  with  con- cepts or explanations at the individual  level but whether  something  is lost if interpretations or explanations of social behavior  are couched  purely in terms of individual  psychology.

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: