Ashby’s introduction to cybernetics

With respect to Enterprise Architecture

Copyright 2017-9 Graham Berrisford. One of a hundred papers on the System Theory page at Last updated 12/03/2019 21:17


TOGAF says EA "regards the enterprise as a system, or system of systems."

A general system theory was developed in the1950s by Bertalanffy, Wiener, Ashby and others.

This discussion includes several quotes from Ashby’s Introduction to Cybernetics (1956).


W. Ross Ashby (1903-1972) was a psychologist and systems theorist.

“Despite being widely influential within cybernetics, systems theory… Ashby is not as well known as many of the notable scientists his work influenced.”

Understanding Ashby’s ideas helps you to understand much else in the field of systems thinking.

The discussion that follows relates Ashby’s ideas to Enterprise Architecture (EA).


For example:

“Cybernetics was defined by Wiener as "the science of control and communication, in the animal and the machine" in a word, as the art of steersmanship.

Co-ordination, regulation and control are its themes, for these were of the greatest biological and practical interest.” (1956)

One may say coordination (or integration), regulation and control are the themes of Enterprise Architecture (EA).


This paper copies some sections from Ashby’s book.

It paraphrases some sentences, replacing Cybernetics by EA, and added comments presenting EA as a branch of Cybernetics.

Note that this is Cybernetics in its classical form - before sociologists messed with it by speaking of second-order cybernetics.



Chapter 1: WHAT IS NEW... 1

Chapter 2: CHANGE.. 3





Chapter 5: STABILITY (intro only) 10

Chapter 6: THE BLACK BOX (intro only) 11

Chapter 7: QUANTITY OF VARIETY (intro only) 11


Chapter 9: INCESSANT TRANSMISSION (intro only) 13

Chapter 10: REGULATION IN BIOLOGICAL SYSTEMS (intro only) 14

Chapter 11: REQUISITE VARIETY (intro only) 14

Chapter 12: THE ERROR-CONTROLLED REGULATOR (intro only) 14




Chapter 1: WHAT IS NEW

“The properties commonly ascribed to any object are, in last analysis, names for its behavior.” (Herrick)


1/1. Cybernetics was defined by Wiener as "the science of control and communication, in the animal and the machine" in a word, as the art of steersmanship, and it is to this aspect that the book will be addressed.


[EA is about the steersmanship of business actors and activities

It is about optimization, standardisation and integration of business processes.]


Co-ordination, regulation and control will be its themes, for these are of the greatest biological and practical interest.


[Coordination, integration, regulation of business processes are themes of enterprise architecture.]


1/2. The peculiarities of cybernetics.

Many a book has borne the title "Theory of Machines", but it usually contains information about mechanical things, about levers and cogs.

Cybernetics, too, is a "theory of machines", but it treats, not things but ways of behaving.

It does not ask "what is this thing?" but ''what does it do?" …

It is thus essentially functional and behaviouristic.


[EA does not begin by asking "what is this system?" but rather ''what does it do?" “what services does it offer?” and “what are processes are repeated”?

It takes an essentially functional and behaviouristic view of a business.]


Cybernetics started by being closely associated in many ways with physics, but it depends in no essential way on the laws of physics or on the properties of matter.

Cybernetics deals with all forms of behaviour in so far as they are regular, or determinate, or reproducible.


[EA deals with all forms of business process that are regular, or determinate or reproducible.]


1/5. cybernetics typically treats any given, particular, machine by asking not

·         "what individual act will it produce here and now?" but

·         "what are all the possible behaviours that it can produce ?"


[EA generalizes from individual acts to the general service types it offers, and the process types it repeats.

It almost always addresses the plural (sets of things) rather than the singular (individual thing).

It presumes a business process will follow the logical pattern of a general process type.

It classifies business entities and events into types.]


It is in this way that information theory comes to play an essential part in the subject; for information theory is characterised essentially by its dealing always with a set of possibilities;

both its primary data and its final statements are almost always about the set as such, and not about some individual element in the set.


[EA presume messages will convey information by giving values to the typical qualities of those business entities and events.

It is about those business entities and events that can be monitored and directed by information/data flows.]


In this discussion, questions of energy play almost no part; the energy is simply taken for granted.

Even whether the system is closed to energy or open is often irrelevant; what is important is the extent to which the system is subject to determining and controlling factors.

So no information or signal or determining factor may pass from part to part without its being recorded as a significant event.


[EA is concerned sometimes concerned with flows materials as well as information.

But the primary focus is on information: on

·         input data flows that describe the current state of external entities or request action, and

·         output data flows that inform or direct the behaviour of external entities.]


1/7. The complex system.

The second peculiar virtue of cybernetics is that it offers a method for the scientific treatment of the system in which complexity is outstanding and too important to be ignored.

Such systems are, as we well know, only too common in the biological world!

In the simpler systems, the methods of cybernetics sometimes show no obvious advantage over those that have long been known.

It is chiefly when the systems become complex that the new methods reveal their power.


[EA regards the enterprise as a large and complex system or system of systems.]


Chapter 2: CHANGE

2/1. The most fundamental concept in cybernetics is that of "difference", either that two things are recognisably different or that one thing has changed with time.

Its range of application need not be described now, for the subsequent chapters will illustrate the range abundantly.

All the changes that may occur with time are naturally included, for when plants grow and planets age and machines move, some change from one state to another is implicit.

So our first task will be to develop this concept of "change", not only making it more precise but making it richer.


[EA deals with regular processes that advance the state of a business, as recorded in its information systems.

It also deal with changes to the processes themselves - process reengineering.]


Often a change occurs continuously, that is, by infinitesimal steps, as when the earth moves through space, or a sunbather's skin darkens under exposure.

The consideration of steps that are infinitesimal, however, raises a number of purely mathematical difficulties, so we shall avoid their consideration entirely.

Instead, we shall assume in all cases that the changes occur by finite steps in time and that any difference is also finite.

We shall assume that the change occurs by a measurable jump, as the money in a bank account changes by at least a penny.


[EA is concerned with communications and behavior that is digitized in discrete event messages and process steps.]


Though this supposition may seem artificial in a world in which continuity is common, it is not as artificial as it seems.

As a simple trick, the discrete can often be carried over into the continuous, in a way suitable for practical purposes, by making a graph of the discrete, with the values shown as separate points.

It is then easy to see the form that the changes will take if the points were to become infinitely numerous and close together.

This, in fact, is the method that the mathematician always does use if he wants to be really sure of what happens when the changes are continuous.

Thus, consideration of the case in which all differences are finite loses nothing; it gives a clear and simple foundation; and it can always be converted to the continuous form if that is desired.



3/1. Having now established a clear set of ideas about transformations, we can turn to their first application: the establishment of an exact parallelism between

·         the properties of transformations, as developed here, and

·         the properties of machines and dynamic systems, as found in the real world.


About the best definition of "machine" there could of course be much dispute.

A determinate machine is defined as that which behaves in the same way as does a closed single-valued transformation.

The justification is simply that the definition works: that it gives us what we want, and nowhere runs grossly counter to what we feel intuitively to be reasonable.


[EA presumes that the responses of a business to events or service requests can be determined by rules - as in any determinate dynamic system.]


3/3. The discrete machine.

At this point it may be objected that most machines, whether man-made or natural, are smooth-working, while the transformations that have been discussed so far change by discrete jumps.

These discrete transformations are, however, the best introduction to the subject.

Their great advantage is their absolute freedom from subtlety and vagueness, for every one of their properties is unambiguously either present or absent.

This simplicity makes possible a security of deduction that is essential if further developments are to be reliable.


In any case the discrepancy is of no real importance.

The discrete change has only to become small enough in its jump to approximate as closely as is desired to the continuous change.

It must further be remembered that in natural phenomena the observations are almost invariably made at discrete intervals;

the "continuity" ascribed to natural events has often been put there by the observer's imagination, not by actual observation at each of an infinite number of points.

Thus the real truth is that the natural system is observed at discrete points, and our transformation represents it at discrete points.

There can, therefore, be no real incompatibility.


[Again: EA presumes continual communications and behaviour can be “digitized” into discrete event messages and process steps.

It treats the enterprise as a system in which all state changes are seen as discrete steps.]


3/4. Machine and transformation.

The parallelism between machine and transformation is shown most obviously when we compare

·         the machine's behaviour, as state succeeds state, with

·         the kinematic graph (S.2/17), as the arrows lead from element to element.


If a particular machine and a particular graph show full correspondence it will be found that:

(1)   Each possible state of the machine corresponds uniquely to a particular element in the graph, and vice versa. The correspondence is one-one.

(2)   Each succession of states that the machine passes through because of its inner dynamic nature corresponds to an unbroken chain of arrows through the corresponding elements.

(3)   If the machine goes to a state and remains there (a state of equilibrium, S.5/3) the element that corresponds to the state will have no arrow leaving it (or a re-entrant one, S.2/17).

(4)   If the machine passes into a regularly recurring cycle of states, the graph will show a circuit of arrows passing through the corresponding elements.

(5)   The stopping of a machine by the experimenter, and its re- starting from some new, arbitrarily selected, state corresponds, in the graph, to a movement of the representative point from one element to another when the movement is due to the arbitrary action of the mathematician and not to an arrow.

When a real machine and a transformation are so related, the transformation is the canonical representation of the machine, and the machine is said to embody the transformation.



3/5. In the previous sections a machine's "state" has been regarded as something that is known as a whole, not requiring more detailed specification.

States of this type are particularly common in biological systems where, for instance, characteristic postures or expressions or patterns can be recognised with confidence though no analysis of their components has been made.

Nevertheless, systems often have states whose specification demands (for whatever reason) further analysis.

Thus suppose a news item over the radio were to give us the "state", at a certain hour, of a Marathon race now being run; it would proceed to give, for each runner, his position on the road at that hour.

These positions, as a set, specify the "state" of the race.

So the "state" of the race as a whole is given by the various states (positions) of the various runners, taken simultaneously.

Such "compound" states are extremely common, and the rest of the book will be much concerned with them.


[Business information systems record the complex current state of business operations in the values of many (thousands) of state variable types.

This state vector is not only very large, but also distributed across scores or hundreds of business data stores.

EA has to study disintegrity between discrete databases, and consider where and how to reduce it.]


It should be noticed that we are now beginning to consider the relation, most important in machinery, that exists between the whole and the parts.

Thus, it often happens that the state of the whole is given by a list of the states taken, at that moment, by each of the parts.

Such a quantity is a vector, which is defined as a compound entity, having a definite number of components.


[Business information systems often record the state vectors of many (thousands or millions) of entity instances.]


3/11. What is a ''system'"?

In S.3/1 it was stated that every real determinate machine or dynamic system corresponds to a closed, single-valued transformation; and the intervening sections have illustrated the thesis with many examples.

It does not, however, follow that the correspondence is always obvious; on the contrary, any attempt to apply the thesis generally will soon encounter certain difficulties, which must now be considered.


Our difficulty is typical in scientific investigation and is fundamental: we want the transformation to be single-valued but it will not come so.

We cannot give up the demand for singleness, for to do so would be to give up the hope of making single-valued predictions.

Fortunately, experience has long since shown what is to be done: the system must be re-defined.


At this point we must be clear about how a "system" is to be defined.

Our first impulse is to point at the pendulum and to say "the system is that thing there".

This method, however, has a fundamental disadvantage: every material object contains no less than an infinity of variables and therefore of possible systems.

The real pendulum, for instance, has not only length and position; it has also mass, temperature, electric conductivity, crystalline structure,  chemical impurities, some radio-activity, velocity, reflecting power,  tensile strength, a surface film of moisture, bacterial contamination, an optical absorption, elasticity, shape, specific gravity, and so on and on.


[We should be clear about what a "system” is.

Our first impulse is to point at an operational business and say "the system is that thing there".

This method has a fundamental disadvantage: every real-world entity contains no less than an infinity of variables.

A real employee has not only a name, salary and role, but also a nervous system, bodily organs, food tastes, musical aptitude, and so on.]


Any suggestion that we should study "all" the facts is unrealistic, and actually the attempt is never made.

What is necessary is that we should pick out and study the facts that are relevant to some main interest that is already given.


[Any suggestion that architects should study "all" the facts is unrealistic, and actually the attempt is never made.

What is necessary is that architects select and study facts that can be usefully monitored or directed by the business.]


The truth is that in the world around us only certain sets of facts are capable of yielding transformations that are closed and single-valued.

The discovery of these sets is sometimes easy, sometimes difficult.

The history of science, and even of any single investigation, abounds in examples.

Usually the discovery involves the other method for the defining of a system, that of listing the variables that are to be taken into account.

The system now means, not a thing, but a list of variables.


[A system describer describes things in terms of variable types; a system in operation gives values to those variable types.]



4/1. In the previous chapter we studied the relation between transformation and machine, regarding the latter simply as a unit.

We now proceed to find, in the world of transformations, what corresponds to the fact that every ordinary machine can be acted on by various conditions, and thereby made to change its behaviour, as a crane can be controlled by a driver or a muscle controlled by a nerve.


It will be seen that the word "change" if applied to such a machine can refer to two very different things.

·         change from state to state, which is the machine's behaviour, and which occurs under its own internal drive, and

·         change from transformation to transformation, which is a change of its way of behaving, and occurs at the whim of the experimenter or some other outside factor.

The distinction is fundamental and must on no account be slighted.


[EA is concerned with two completely different kinds of change:

·         state (adaptive) changes to an operational system state, made by processes, which occur in the everyday behavior of a business, under its own business rules.

·         generational (evolutionary) changes to a system description, made by designers, which define a new system generation.]


4/4. Input and output.

The word "transducer" is used by the physicist, and especially by the electrical engineer, to describe any determinate physical system that has certain defined places of input, at which the experimenter may enforce changes that affect its behaviour, and certain defined places of output, at which he observes the changes of certain variables, either directly or through suitable instruments.

It will now be clear that the mathematical system described in S.4/1 is the natural representation of such a material system.

It will also be clear that the machine's "input" corresponds to the set of states provided by its parameters; for as the parameters or input are altered so is the machine's or transducer's behaviour affected.


[EA is concerned with business systems where the inputs and outputs can include materials as well as information.

But the primary focus is on information: on input data flows that describe the current state of external entities or request action, and output data flows that inform or direct the behaviour of external entities.]


With an electrical system, the input is usually obvious and restricted to a few terminals.

In biological systems, however, the number of parameters is commonly very large and the whole set of them is by no means obvious.

It is, in fact, co-extensive with the set of "all variables whose change directly affects the organism".

The parameters thus include the conditions in which the organism lives.

In the chapters that follow, the reader must therefore be prepared to interpret the word "input" to mean either the few parameters appropriate to a simple mechanism or the many parameters appropriate to the free-living organism in a complex environment.

(The increase in the number of parameters does not necessarily imply any diminution in the rigour of the argument, for all the quantities concerned can be measured with an accuracy that is bounded only by the experimenter's resources of time and money.)


4/6. A fundamental property of machines is that they can be coupled.

Two or more whole machines can be coupled to form one machine; and any one machine can be regarded as formed by the coupling of its parts, which can themselves be thought of as small, sub-machines.

The coupling is of profound importance in science, for when the experimenter runs an experiment he is coupling himself temporarily to the system that he is studying.


[EA is concerned with how the enterprise system is coupled to its environment, and with how its subsystems are coupled to each other.]


To what does this process, the joining of machine to machine or of part to part, correspond in the symbolic form of transformations ?

Of what does the operation of "coupling" consist?

Before proceeding to the answer we must notice that there is more than one answer.


One way is to force them roughly together,

So that they become "coupled" as two automobiles may be locked together after an accident.

This form, however, is of little interest to us, for the automobiles are too much changed by the process.


[Another way is] coupling that does no violence to each machine's inner working, so that after the coupling each machine is still the same machine that it was before.

For this to be so, the coupling must be arranged so that, in principle, each machine affects the other only by affecting its conditions, i.e. by affecting its input.

Thus, if the machines are to retain their individual natures after being coupled to form a whole, the coupling must be between the (given) inputs and outputs, other parts being left alone no matter how readily accessible they may be.


4/10. parts can be coupled in different ways to form a whole.

The defining of the component parts does not determine the way of coupling.

From this follows an important corollary.

That a whole machine should be built of parts of given behaviour is not sufficient to determine its behaviour as a whole.

Only when the details of coupling are added does the whole's behaviour become determinate.


[EA takes the holistic view; it looks to integrate subsystems to the benefit of the wider business system.]




4/16. Up till now, the systems considered have all seemed fairly simple, and it has been assumed that at all times we have understood them in all detail.

Cybernetics, however, looks forward to being able to handle systems of vastly greater complexity: computing machines, nervous systems, societies.


Chapter 5: STABILITY (intro only)

5/1. The word "stability" is apt to occur frequently in discussions of machines, but is not always used with precision.

Bellman refers to it as ". . . stability, that much overburdened word with an unstabilised definition".

Since the ideas behind the word are of great practical importance, we shall examine the subject with some care, distinguishing the various types that occur.


Chapter 6: THE BLACK BOX (intro only)

6/1. The methods developed in the previous chapters now enable us to undertake a study of the Problem of the Black Box; and the study will provide an excellent example of the use of the methods.

The Problem of the Black Box arose in electrical engineering.

The engineer is given a sealed box that has terminals for input, to which he may bring any voltages, shocks, or other disturbances he pleases, and terminals for output, from which he may observe what he can.

He is to deduce what he can of its contents.


[EA is concerned with abstraction of system description from operational system contents and workings.

Most obviously, it encapsulates systems and subsystems; it defines system behavior in the form of interfaces and service contracts.]


Chapter 7: QUANTITY OF VARIETY (intro only)

7/1. In Part I we considered the main properties of the machine, usually with the assumption that we had before us the actual thing, about which we would make some definite statement, with reference to what it is doing here and now.

To progress in cybernetics, however, we shall have to extend our range of consideration.

The fundamental questions in regulation and control can be answered only when we are able to consider the broader set of what it might do, when "might" is given some exact specification.

Throughout Part II, therefore, we shall be considering always a set of possibilities.

The study will lead us into the subjects of information and communication, and how they are coded in their passages through mechanism.

This study is essential for the thorough understanding of regulation and control.

We shall start from the most elementary or basic considerations possible.


[EA presumes senders encode meaning in messages, and receivers decode the same meanings from those messages.

So business information is encoded in data flows (messages) by senders and decoded from data flows by receivers.

Enterprise data architects are concerned with standardization of how information is encoded in data types.]



8/1. The previous chapter has introduced the concept of "variety", a concept inseparable from that of "information", and we have seen how important it is, in some problems, to recognise that we are dealing with a set of possibilities.

In the present chapter we shall study how such possibilities are transmitted through a machine, in the sense of studying the relation that exists between the set that occurs at the input and the consequent set that occurs, usually in somewhat coded form, at the output.

We shall see that the transmission is, if the machine is determinate, perfectly orderly and capable of rigorous treatment.

Our aim will be to work towards an understanding good enough to serve as a basis for considering the extremely complex codings used by the brain.


[EA is concerned with systems that are not only large but variegated and complex; the challenges lie as much in the complexity as in the size.]


8/2. Ubiquity of coding.

To get a picture of the amount of coding that goes on during the ordinary interaction between organism and environment;

let us consider, in some detail, the comparatively simple sequence of events that occurs when a "Gale warning" is  broadcast.

·         It starts as some patterned process in the nerve cells of the meteorologist, and then becomes a pattern of muscle-movements as he writes or types it, thereby making it a pattern of ink marks on paper.

·         From here it becomes a pattern of light and dark on the announcer's retina, then a pattern of retinal excitation, then a pattern of nerve impulses in the optic nerve, and so on through his nervous system. It emerges, while he is reading the warning, as a pattern of lip and tongue movements, and then travels as a pattern of waves in the air.

·         Reaching the microphone it becomes  a pattern of variations of electrical potential, and then goes through further changes as it is amplified, modulated, and broadcast.

·         Now it is a pattern of waves in the ether, and next a pattern in the receiving set.

·         Back again to the pattern of waves in the air, it then becomes a pattern of vibrations traversing the listener's ear-drums, ossicles, cochlea, and then becomes a pattern of nerve-impulses moving up the auditory nerve.

Here we can leave it, merely noticing that this very brief account mentions no less than sixteen major transformations through all of which something has been preserved, though the superficial appearances have changed almost out of recognition.


[EA presumes that information is coded/decoded in layers up/down a communication stack.

At the top are human thoughts; at the bottom are binary digits represented in some kind of physical matter or energy.

The primary concern is with the meaning understood by human senders and receivers of information.

Though it could be argued that every machine lower down a communication stack works at its own level of meaning.]


8/5. Coding by machine.

Next we can consider what happens when a message becomes coded by being passed through a machine.

That such questions are of importance in the study of the brain needs no elaboration.

Among their other applications are those pertaining to "instrumentation" the science of getting information from some more or less inaccessible variable or place, such as the interior of a furnace or of a working heart, to the observer.

The transmission of such information almost always involves some intermediate stage of coding, and this must be selected suitably.

Until recently, each such instrument was designed simply on the principles peculiar to the particular branch of science; today, however, it is known, after the pioneer work of Shannon and Wiener, that certain general laws hold over all such instruments.

What they are will be described below.


A "machine" was defined in S.3/4 as any set of states whose changes in time corresponded to a closed single-valued transformation.

This definition applies to the machine that is totally isolated, i.e. in constant conditions; it is identical with the absolute system defined in Design ...

In S.4/1 the machine with input was defined as a system that has a closed single-valued transformation for each one of the possible states of a set of parameters.

This is identical with the "transducer" of Shannon, which is defined as a system whose next state is determined by its present state and the present values of its parameters.

(He also assumes that it can have a finite internal memory, but we shall ignore this for the moment, returning to it in S.9/8.)


Assume then that we have before us a transducer M that can be in some one of the states… which will be assumed here to be finite in number.

It has one or more parameters that can take, at each moment, some one of a set of values…

Each of these values will define a transformation of the states.

We now find that such a system can accept a message, can code it, and can emit the coded form.

By "message" I shall mean simply some succession of states that is, by the coupling between two systems, at once the output of one system and the input of the other.

Often the state will be a vector.

I shall omit consideration of any "meaning" to be attached to the message and shall consider simply what will happen in these determinate systems.


[EA presumes senders encode meaning in messages, and receivers decode meaning from messages.

Our primary concern is with the meaning understood by human senders and receivers of information.

Though every machine lower down a communication stack works at its own level of meaning.]


Chapter 9: INCESSANT TRANSMISSION (intro only)

9/1. The present chapter will continue the theme of the previous, and will study variety and its transmission, but will be concerned rather with the special case of the transmission that is sustained for an indefinitely long time.

This is the case of the sciatic nerve, or the telephone cable, that goes on incessantly carrying messages, unlike the transmissions of the previous chapter, which were studied for only a few steps in time.


Incessant transmission has been specially studied by Shannon, and this chapter will, in fact, be devoted chiefly to introducing his Mathematical Theory of Communication, with special emphasis on how it is related to the other topics in this Introduction.


What is given in this chapter is, however, a series of notes, intended to supplement Shannon's masterly work, rather than a description that is complete in itself.

Shannon's book must be regarded as the primary source, and should be consulted first. I assume that the reader has it available.


[Again: EA presumes continual communications and behaviour can be “digitized” into discrete event messages and process steps.

It treats the enterprise as a system in which all state changes are seen as discrete steps.]



10/1. The two previous Parts have treated of Mechanism (and the processes within the system) and Variety (and the processes of communication between system and system).

These two subjects had to be studied first, as they are fundamental.

Now we shall use them, and in Part III we shall study what is the central theme of cybernetics - regulation and control.


[EA presumes that business information systems monitor and direct the behaviour of business actors and activities.]


Chapter 11: REQUISITE VARIETY (intro only)

11/1. In the previous chapter we considered regulation from the biological point of view, taking it as something sufficiently well understood.

In this chapter we shall examine the process of regulation itself, with the aim of finding out exactly what is involved and implied.

In particular we shall develop ways of measuring the amount or degree of regulation achieved, and we shall show that this amount has an upper limit.


[EA assumes what Ashby’s law of requisite variety implies. In short, in my words:

1.      The more different ways a system can deviate from its normal state, the more different control actions a homeostatic control system will need.

2.      A controlling system cannot be less complex (in its variable qualities and quantities) than the selected behaviour of the real machine to be controlled.



12/1. In the previous chapter we studied the nature of regulation, and showed that certain relations and laws must hold if regulation is to be achieved.

There we assumed that regulation was achieved, and then studied what was necessary.

This point of view, however, though useful, hardly corresponds with that commonly used in practice.

Let us change to a new point of view.


In practice, the question of regulation usually arises in this way:

The essential variables E are given, and also given is the set of states T in which they must be maintained if the organism is to survive (or the industrial plant to run satisfactorily).

These two must be given before all else.

Before any regulation can be undertaken or even discussed, we must know what is important and what is wanted.

Any particular species has its requirements given: the cat must keep itself dry, the fish must keep itself wet.

A servo-mechanism has its aim given by other considerations: one must keep an incubating room hot, another must keep a refrigerating room cold.

Throughout this book it is assumed that outside considerations have already determined what is to be the goal, i.e. what are the acceptable states 7].

Our concern, within the book, is solely with the problem of how to achieve the goal in spite of disturbances and difficulties.



13/1. Regulation and control in the very large system is of peculiar interest to the worker in any of the biological sciences, for most of the systems he deals with are complex and composed of almost uncountably many parts.

The ecologist may want to regulate the incidence of an infection in a biological system of great size and complexity, with climate, soil, host's reactions, predators, competitors, and many other factors playing a part.

The economist may want to regulate against a tendency to slump in a system in which prices, availability of labour, consumer's demands, costs of raw materials, are only a few of the factors that play some part.

The sociologist faces a similar situation.

And the psychotherapist attempts to regulate the working of a sick brain that is of the same order of size as his own, and of fearful complexity.

These regulations are obviously very different from those considered in the simple mechanisms of the previous chapter.

At first sight they look so different that one may well wonder whether what has been said so far is not essentially inapplicable.


13/2. … many of the propositions established earlier are stated in a form that leaves the size of the system irrelevant.

(Sometimes the number of states or the number of variables may be involved, but in such a way that the proposition remains true whatever the actual number.)

Regulation in biological systems certainly raises difficult problems that can be admitted freely.

But let us be careful, in admitting this, not to attribute the difficulty to the wrong source.

Largeness in itself is not the source; it tends to be so regarded partly because its obviousness makes it catch the eye and partly because variations in size tend to be correlated with variations in the source of the real difficulty.

What is usually the main cause of difficulty is the variety in the disturbances that must be regulated against.


[EA is concerned with systems that are not only large but complex, and challenges lie as much in the complexity as in the size.]


13/3. Before we proceed we should notice that when the system is very large the distinction between D, the source of the disturbances, and T, the system that yields the outcome, may be somewhat vague,

in the sense that the boundary can often be drawn in a variety of ways that are equally satisfactory.


This flexibility is particularly well-marked among the systems that occur on this earth (for the terrestrial systems tend markedly to have certain general characteristics).

On this earth, the whole dynamic biological and ecological system tends to consist of many sub- systems loosely coupled (S.4/20);

and the sub-systems themselves tend to consist of yet smaller systems, again more closely coupled internally yet less closely coupled between one another; and so on.

Thus in a herd of cattle, the coupling between members is much looser than the couplings within one member and between its parts (e.g. between its four limbs);

and the four limbs are not coupled as closely to one another as are the molecules within one bone.

Thus if some portion of the totality is marked out as T, the chief source D of disturbance is often other systems that are loosely coupled to r, and often sufficiently similar to those in T that they might equally reasonably have been included in it.

In the discussion that follows, through the rest of the book, this fact must be borne in mind: that sometimes an equally reasonable demarcation of 7" and D might have drawn the boundary differently, without the final conclusions being affected significantly.

Arbitrary or not, however, some boundary must always be drawn, at least in practical scientific work, for otherwise no definite statement can be made.


[EA is concerned with systems whose boundaries are determined by observers, and therefore somewhat arbitrary.

But there is no escaping the need to bound the system of interest as best we can.]



“An Introduction to Cybernetics” W. Ross Ashby M.A., M.D.(Cantab.), D.P.M. Director of Research, Barnwood House, Gloucester

First published1956 by Chapman & Hall; later in New York, by John Wiley and son. 440 Fourth Avenue; and by William Clowes and Sons, London and Beccles.


Design for a Brain: The Origin of Adaptive Behaviour”. Ross Ashby, 1952, Chapman & Hall.

Also available as pdf. Even: