Ashby

1
Ashbys Law of Requisite Variety

• Only variety can destroy variety
• For a system to adapt and survive, therefore, its
  response must be adequate to the variety of the
  stimuli that confront it
• The response might be cognitive, behavioral, or
  both

2
Ashbys Law of Requisite Variety

• But what, in fact, do we mean by requisite?
• Something that steers a path between
  fossilization and disintegration

3
Ashbys Law of Requisite Variety
Fossilization
High
Variety of Stimuli
Disintegration
Low
Low
High
Variety of Responses
? Max Boisot
Ref. 2.4
4
Reducing Response Variety
5
Reducing Stimulus Variety
Ref. 1.2.1.6 ? Max Boisot
6
Knowledge and Complexity

• Both strategies face an epistemological challenge
  the more you understand, the less you need to
  waste energy reacting.
• Variety is the surface manifestation of
  complexity and stochastic processes working in
  tandem.
• In order to distinguish what is the part played
  by complexity and what is the part played by
  chance (noise), one has to go beneath the surface
  and achieve understanding of the processes at
  play.

7
Complexity

• A function of the number of entities in
  interaction where these are varied and non-linear
• Where the entities are numerous and exhibit
  collective intelligence we can talk of complex
  adaptive systems

Ref. 2.4 ? Max Boisot
8
Measuring Complexity

• Kolmogorov complexity and Chaitins Algorithmic
  Information Complexity (AIC) both measure the
  complexity of a phenomenon by the amount of data
  processing required to reproduce it.
• This gives us an ontological definition of
  complexity.

9
Crude Versus Effective Complexity

• Gell-Mann tells us that AIC is crude complexity
  indistinguishable from randomness
• He contrasts it with effective complexity the
  amount of data processing required to reproduce
  the regularities present in a given phenomenon.

10
But

• But we can never be sure that we have found the
  shortest program ie, that we have exhausted our
  understanding of the phenomenon in question.
• And we may be interested in the shortest program
  that yields a representation of the phenomenon in
  question ie, cognitive (epistemological) rather
  than ontological complexity

11
The Cognitive Aspects of Complexity
High
Facing a Major Challenge
Learning a new skill
Subjective Complexity
Mastered Routine
Competence
Low
Low
High
Objective Complexity
Ref. 2.1
? Max Boisot 1999
12
Complexity Reduction versus Complexity Absorption

• In dealing with the law of requisite variety, a
  Complex Adaptive System either absorbs complexity
  or reduces it
• Complexity absorption, pushed to extremes, leads
  to excessive dissipation of energy and eventually
  to chaos and disintegration
• Complexity reduction, pushed to extremes, leads
  to a loss of context (oversimplification) and
  eventually to fossilisation

13
Representation versus Reproduction

• But what is the it that constitutes the
  phenomenon in question? The amount of data
  processing required to represent it is likely to
  be lower than the amount required to exactly
  reproduce it (the sum of all conceivable
  representations). The adequacy of a
  representation is related to its purpose.
• This would give us an epistemological reduction
  of complexity.

14
The Managerial Challenge

• How might we go about developing an epistemology
  of complexity and how would this help us to
  manage it?

15
Knowledge as Justified True Belief (Plato)

• Justified what procedure did you follow to
  acquire it?
• True what does it correspond to in the world?
  And how coherent is this?
• Belief are you willing to act upon it?

16
Three Worlds

• Actual worlds - certainty
• Probable worlds measurable risk
• Possible worlds - uncertainty

17
Three Worlds

• Actual Worlds Justified True Belief (JTB)
• Probable Worlds Justified Belief (JB)
• Possible Worlds Belief (B)

18
Actual , Probable, and Possible Worlds
Possible worlds (B)
Probable worlds (JB)
Actual Worlds (JTB)
Ref. 2.7.1 ? Max Boisot 2002
19
Cash, NPV, and Option Value three ways of
managing under the three different epistemologies.

• Justified true beliefs have cash value
  Managing under conditions of certainty
•  
• Justified beliefs have a Net Present Value
  Managing under conditions of measurable risk
•  
• Beliefs have an option value Managing under
  conditions of uncertainty

20
Actual , Probable, and Possible Worlds
Possible worlds (options)
Probable worlds (NPV)
Actual Worlds (Cash)
Ref. 2.7.1 ? Max Boisot 2002
21
Epistemological Transitions

• The challenge to knowledge management is to
  understand the conditions under which one passes
  from one of the three worlds to another and hence
  from one epistemology to another.
•  
• Anxiety makes us move faster towards the inner
  circle the world of certainty - than we ought
  to (example of the lost elephants at BP).
•  
• There is an inevitable tension between
  exploration and exploitation

22
Managing Complexity

• Managers inhabit the world of NPVs they are
  analytically trained to converts probable worlds
  into actual worlds, risks into certainties.
• Entrepreneurs inhabit the world of options they
  convert possible worlds into probable worlds,
  uncertainties into measurable risks (which they
  can then sell on!)

23
Construction Projects versus Entrepreneurial
Projects
Ref. 2.7.1 ? Max Boisot 2002
24
Weick and Ashbys Law

• The law of requisite variety trades in variety of
  response for variety of representation
• In possible worlds, the Weickean concept of
  enactment can be applied but under what
  conditions?
• At bifurcation points, where choice becomes
  possible and thus requires a minimum of energy
  and a maximum of information

25
Organizational Implications

• Current governance structures favour management
  the stewardship of resources - at the expense of
  entrepreneurship the creation of resources
• Current educational provisions favour management
  at the expense of entrepreneurship ie,
  analysing risk rather than exploring uncertainty

26
Conclusion

• With increasing complexity and uncertainty,
  knowing how to convert uncertainty into risk will
  become more important than knowing how to convert
  risk into certainty.
• Our institutionalized epistemologies, however,
  focus us on the latter at the expense of the
  former.
• For this reason, entrepreneurial epistemologies
  go begging