p.15 Engineers make machines which control themselves and each other. These machines 'appreciate
the situation', measure the divergence of what is from what ought to be and adapt their behavior accordingly. In
doing so, they apply 'rules' which they have previously learned... In all this they behave very much like men,
who they excel in accuracy though not in variety of response.
Such machines do not need to be continually 'directed'... They will do these things largely for themselves...
Now this is precisely the goal of all managers of men.
p.16 Feedback plays a central part in the engineer's idea of control. Whenever purposeful action
is being taken, it can best be checked by comparing its result with the result it is intended to produce. Is it having
the desired effect? If not, in what way and to what extent is the result deviating from what was purposed? This deviation
is the measure of the action needed to correct it. If the deviation can be made to operate a mechanism which will 'automatically'
modify future action in the way needed to correct the error, then the circuit is closed and the system is set to control itself.
p.17 In devising controlled systems the engineer has two main problems. He has to devise indices
which will reflect with sufficient exactness the changing value of critical variables; and he has to arrange for these to
trigger off appropriate compensating action. In both fields he has great scope for inventive skill.
p.19 I commend to you the very basic idea that action should be controlled by the observed difference
between what is and what ought to be.
p.21 The theory of control is, in fact, an approach to the understanding of the laws which make it possible
for wholes to cohere and to act coherently.
p.27 Control serves three different purposes. It provides means to compare performance (a) with norms and
limits which are given by the nature of the operation, (b) with forecasts reflecting the expected results of policies, and
(c) with targets representing goals for attainment.
p.29 Control is used to guide action. It must therefore be available at each level where action
is taken and the guidance it supplies at each level must be relevant to the action which is taken at that level.
p.37-38-39 Most business decisions involve attaching degrees of probability to various hypotheses...
Expectation does not imply certainty.
But for purposes of action it is useful to be able to attach to some expectations such a
degree of assurance that we can act on the assumption that they will occur. I think it convenient to keep
the word 'confidence' for our attitude towards expectations of this kind, expectations which, though they are hypothetical,
do not require us to apply probability factors or to work out alternatives on the supposition that they will not be realized...
confidences are guides to action... The deliberate building and moulding of these structures of expectation form a
large part of business activity.
p.57-58 The mere multiplication of alternative means to an end might only make the choice harder,
unless it were accompanied by some gift which guides the problem-solver in the general direction of the still undiscovered
solution. The literature of problem solving, no less than common experience, attests our capacity for searching with
a lively sense of 'warm...warmer...warmer...', when we do not know what we are looking for.
p.58-59 what solutions are considered and in what order? ...Some mental process narrows the field rapidly
to a short list of alternatives, which alone are carefully considered.
p.68-69 organizations, like organisms and other open systems survive only so long
as they keep their constituent variables within critical limits. Shortage... beyond a critical point, creates a self-exciting
disturbance of relationships which overflows suddenly across all the activities of an undertaking, probably with irreversible
effects
p.69 A large number of inflows and outflows have to be kept at levels and within mutually related
limits which are necessary to survival. These are relations so like those which enable an animal to go on generating
the energy needed for survival that I will call them metabolic relations.
p.169 If the world will not give me what I want, I can react in any or all of four different
ways. I can try to alter my wants; I can try to alter the world; I can hunt for new ways of pursuing my wants; I can
bang my head on the floor and scream [JLJ - the wall is more convenient for this than the floor, which is often dirty
and harder to reach]. Any of the first three rank as adaptive behaviour and nearly every situation calls for a combination
of the three. Only the fourth is non-adaptive; among individuals it is called neurotic.
p.170 The higher animals have a host of such conditions which they have to maintain and their deviations
from these have thresholds which they must not transgress on pain of death; and they spend most of their time pursuing the
one and evading the other. I will call these positive governors 'norms' and the negative ones 'limits'. I shall suggest
that organizations also are controlled by the norms they pursue and the limits they avoid.
p.171 We have a troublesome habit of mind which makes us think of attainable ends as something final, whereas,
so long as we are alive, no attainable end can be final, since it gives no guidance as to where we go when we have attained
it. The only inexhaustible guides in a continuous, dynamic process are not goals but courses. Paradoxical though it
may seem, in any continuing activity courses are more fundamental than goals... business is governed fundamentally not by
the desire to attain specific goals and avoid specific threats, but by the need to maintain or avoid particular relationships.
p.171 the 'nature' of animals and men - and, I am suggesting, of organizations also - is defined
by stating their norms and limits, that is, the states, courses, conditions, which they must hold or must avoid, and the responses
and skills at their disposal in this endless task.
p.173-174 when we look for the norms and limits which define a business as a dynamic system we must seek
them also among the laws which govern growth and change in human relationships.
p.174 The process by which norms and limits control us is fairly familiar. They provide
a measure whereby we can tell where we ought to be - or, where the control is negative, where we ought not
to be - for comparison with where we are.
p.177 You will not find these norms and limits written down in full in any book;
the more important ones are unconsciously taken for granted. None the less, it is they which govern the behaviour of the undertaking;
and its success depends upon how closely these standards of what ought and ought not to be correspond with the world of reality
in which the undertaking works.
p.179 Adaptive responses are needed long before disaster is apparent, long before any one can prove with
certainty that the response commended is the only one or even the best.
p.180 To promote adaptation at every level, then, one must first understand what are the governors of activity
at the level concerned, the norms it tries to hold, the limits it strives to avoid. These must be won over to the change,
or adaptation at that level will be sluggish or disruptive or successfully resisted.
p.182 The governors of behaviour for men and societies, as for all other open systems,
are rather what I have called norms and limits - relationships between processes which must be maintained through
time and thresholds beyond which these relationships must not be allowed to stray.
p.182 management... Its function is not so much to adapt as to maintain an organization which will
be adaptive.
