On human cognition

A cybernetic view of description and reality

Copyright 2014 Graham Berrisford. Now a chapter in “the book” at https://bit.ly/2yXGImr. Last updated 27/02/2021 22:56


The first of half the book addresses the questions: What is a system? How does system theory apply to the description of business and software systems? The second half addresses a more fundamental question: What is a description?


This chapter identifies three kinds of description, details the semantics of the epistemological triangle used to illustrate points in other chapters, and outlines a cybernetic view of human cognition.



The evolution of description. 1

Separating descriptions from described things. 2

Descriptions and described things in physics. 3

Descriptions and described things in mathematics. 4

Descriptions and described things in philosophy. 5

Our epistemological triangle – the semantics. 5

On human cognition (thinking) 8

Principles of description. 9

Postscript: on enterprise architecture. 12



The evolution of description

Descartes is famously said to have started his philosophy from the premise “I think therefore I am”. Psycho-biologists presume more. They presume that, in space and time, there exist animals that can perceive phenomena, remember them and communicate about them.


Implicit in those premises is the idea that there was no description of reality before life. The Darwinian view is that both memories and messages are biological phenomena that evolved to help animals survive. Remembering and sharing descriptions of things helps animals to understand, predict and manipulate things, and so, improves their chance of reproducing.


Highly relevant here is the good regulator theorem (Ashby and Conant), which declares that a good regulator has a description or model of what it regulates. Here, a regulator is an animal or a machine that either has, or has access to, a model of the target that it regulates. Read this triangle from left to right: regulators <have and use> models, which <represent> targets.


The good regulator


<have and use>           <represent>

Regulators    <monitor and regulate >   Targets


The question is not whether an animal (or a business) has a model of its environment; it is how complete and accurate is the model? To which the answers might be both “very incomplete and somewhat inaccurate” and “remarkably, complete and accurate enough”. Which leads us inexorably to the view of description and reality outlined in what follows.


Although animals demonstrably remember and share knowledge, like where food can be found, a description need not be complete or perfect. It need only represent things well enough. A cat remembers a mouse’s features well enough to spot and catch mice. A honey bee remembers the location of some pollen well enough it can direct other bees to that location.


A description must correlate in some way to the described phenomenon. An iconic description, like a statue or photograph, mimics features of the described phenomenon, and is recognisable using the basic senses. An indicative description, like the smoke of a fire, points to effects produced by a described phenomenon. A symbolic description encodes some features of the described phenomenon using some kind of code, which can only be recognised by an animal or machine that knows that code.


Our main interest is in symbolic descriptions, which are encoded using symbols (or signs) and in how animals and machines create and use them to describe things.

Separating descriptions from described things

Cartographers create maps (descriptions) of territories (described things). The map is not the territory, but the two must be correlated well enough to help map users find things. Read the triangle below from left to right: Mappers <observe and envisage> Territories. Mappers <create and use> Maps. Maps <represent> Territories.




<create and use>          <represent>

Mappers    <observe & envisage>   Territories


Hmm… Is the territory real? Or merely another description – a mental model - we form of the territory as we travel through it? It doesn’t matter to us, as travellers; it only matters that we find the map useful.


Cyberneticans create abstract systems (descriptions) to represent the behavior of physical systems (described things). A reality can be only be called a physical system in so far as it is correlated with an abstract system. The remainder of the reality is either beyond our ken or not described as a system.



Abstract systems

<create and use>          <represent>

Observers <observe & envisage> Physical systems


Building architects define the structures of buildings. A real-world building is infinitely complex and incomprehensible. The term architecture might be used for the visual sensation of a building (which may be seen from several viewpoints) and/or the architectural drawings of a building (of which several different sets might exist).


Architectures as sensed

Archiectures as drawn

 Visual sensations

<form and recall>         <represent>

Humans           <observe>        Buildings


<create and use>         <represent>

Architects <observe and envisage> Buildings


Whether the drawings of a building are best characterized as architectural, or blueprints, or detailed designs, is a subjective decision. Other chapters address the topic of system architecture in some depth.


Colors make for odd examples. Before life, there was light but no color. Experiments show animal brains manufacture the sensation of color from a mixture of the light they perceive and their experience. A color is a sensation in the neural system, which depends on the frequencies in the light radiation, and the context of the observation. Of course, we also give names to the sensations we form when looking at a rainbow or a color chart – and can define them in terms of wave lengths.


Colors as sensed

Colors as named

 Color sensations

<form and recall>      <represent>

Animals           <observe>    Light waves

Seven color names

<named>               <represent>

Isaac Newton <observed> Color sensations

Descriptions and described things in physics

Scientific advances are made by imaginatively envisaging changes or alternatives to existing knowledge, then verifying what is described. Have you heard that Einstein didn’t shine at school? The truth is, his insights were deeply informed by extensive study of mathematics, which he mastered at a young age.


"Max Talmud says that after he had given the 12-year-old Einstein a geometry textbook, after a short time he “had worked through the whole book. He thereupon devoted himself to higher mathematics... Soon the flight of his mathematical genius was so high I could not follow." His passion for geometry and algebra led the 12-year-old to become convinced nature could be understood as a "mathematical structure". Einstein started teaching himself calculus at 12, and as a 14-year-old he says he had mastered integral and differential calculus". Wikipedia



Mathematical structures

<create and use>           <represent>

Physicists     <observe and envisage>     Nature


Throughout his life. Einstein distinguished his mathematical descriptions from the reality of the universe, and declared that it is impossible to directly comprehend the reality.


“There are several kinds of theory in physics. Most of them are constructive…When we say that we understand a group of natural phenomena, we mean that we have found a constructive theory which embraces them.Albert Einstein, The Times (28 Nov 1919)


“As far as the laws of mathematics refer to reality, they are not certain; and as far as they are certain, they do not refer to reality.” Albert Einstein, Sidelights on Relativity (1920), 28.


“Physical concepts are free creations of the human mind, and are not, however it may seem, uniquely determined by the external world. In our endeavour to understand reality we are somewhat like a man trying to understand the mechanism of a closed watch. He sees the face and the moving hands, even hears its ticking, but he has no way of opening the case. If he is ingenious he may form some picture of a mechanism which could be responsible for all the things he observes, but he may never be quite sure his picture is the only one which could explain his observations. He will never be able to compare his picture with the real mechanism and he cannot even imagine the possibility or the meaning of such a comparison.” Albert Einstein, The Evolution of Physics (1938), 33

Descriptions and described things in mathematics

Mathematicians define a type (like “point”, “line” and “triangle”) to describe the properties of a thing that instantiates that type.


Type name

Type elaboration

“Even number”

a number divisible by two”.


A type is a pattern for (one member of) a collection or set. An instance is an exhibition or embodiment of a type in a described thing.




<create and use>       <are instantiated in>

Mathematicians <observe and envisage> Set members


Two mathematicians had a famous debate about the meaning of descriptive types or axiomatic assertions. The debate might be distilled as follows.


Frege believed mathematics is carried out at the level of thoughts about real things, rather than descriptions of them. He believed descriptions are imperfect representations of our thoughts. We “know” what a geometric entity (like “point”, “line” and “triangle”) is without reference to a description of it, or how it relates to other entities. And every detail of it may be relevant to answering questions about it.

Hilbert said that, even if we do know other details of an entity (say, its color), they are entirely irrelevant to understanding geometry, which merely defines relations between entities. Geometric descriptions are holistic in the sense that they define how geometric entities are related, rather than what those entities are or contain; they can be anything (large or small) that is relatable as described. Mathematics is carried out at the level of descriptions, regardless of what the entities are in reality.

According to the Stanford Encyclopedia of Philosophy, Hilbert is now regarded as the winner of the debate. Geometry does not address the whole of a thing (Frege); rather, it addresses only those features of it that are describable by geometry (Hilbert).


In short, holism is not wholeism. Similarly, enterprise architecture does not address the whole of an enterprise; rather, it addresses how selected features of a business are related in regular activity systems.

Descriptions and described things in philosophy

Philosophy has moved over centuries from a position in which descriptions perfectly capture the real nature of things, to positions in which descriptions are, at best, approximations to things. Some propose descriptions are convenient fictions to organise sense data. Others propose all descriptions are untethered from external reality, and equally justified. To decide where you sit in this mess leads you into the endless morass of philosophical debate.


This book may be read as endorsing three philosophical positions. Instrumentalists say models are instruments of prediction. Pragmatists say models are concepts or artifacts used in producing scientific knowledge. Constructive empiricists say models are symbolic representations of empirical phenomena.


Constructive empiricism

Symbolic representations

<create and use>          <represent>

Describers    <observe & envisage>    Phenomena


Some classify these three positions as “anti-realist”, meaning they deny that models give a true description of reality. Here, there is no doubt that a) reality does exist b) a description can be true empirically (enough to be useful), or true logically (a consequence that follows from some axiomatic assertion) and c) we can share descriptions and so share some knowledge of reality. To question any of those seems futile sophistry.


Our epistemological triangle – the semantics

Epistemology is about what we know of reality, through observation, description, testing, reasoning and learning from others. The constructivist epistemologist distinguishes descriptions (constructed) from described things (observed or envisaged).


We have no direct knowledge of things in reality. We know them only via descriptions we construct (in the mind, in speech, writing, mathematics, whatever) to describe them, which are (ultimately) associated with our sensations of the world.


We (describers) can only understand physical reality in so far as it is correlated with a description we have access to – which we can correlate well enough with real-world phenomena for practical use.


Read the triangle from left to right:

·       Describers <observe and envisage> Phenomena

·       Describers <create and use> Descriptions

·       Descriptions <represent> Phenomena





<create and use>          <represent>

Describers <observe & envisage> Phenomena


A phenomenon is any aspect or part of reality that a describer can observe or envisage. A description is a representation (in mind, speech, writing or other form) that is correlatable with a phenomenon. A describer is any animal or machine that can observe or envisage something, and create or use (encode or decode) a description of it. (AI software can now abstract types/patterns from things it observes as sharing features).


Note that each pair of concepts is related in a many-to-many way. A describer can create several descriptions of the same thing. Two descriptions may be compatible or in conflict. Is light rightly described as waves or particles? Physicists do not say either model is the “true” model, they say only that each can be useful.


Conversely, several describers can contribute to one recorded description of the same thing. It may well be that none of those describers knows whole description, which can be larger, more complex, consistent and informative than any human memory can be.


The view of description and reality above seems obvious to some and obscure to others. The semantics of the triangle are explored below.

Descriptions <represent> phenomena

Descriptions embrace every kind of model an animal or machine can make of the world, be it a mental or documented model, a memory or message, in speech or writing, a 2D picture or 3D model, in the head or in stone. Here, we mostly discuss descriptions expressed symbolically, and verbally.


Phenomena include everything that can be observed or envisaged, including descriptions and describers.




<create and use>    <represent>

Describers <observe and envisage> Phenomena

Describers <create and use> descriptions

Describers are animals or machines that can encode and decode descriptive models of phenomena. Descriptions are given meanings only in the acts of encoding or decoding.




<create and use>   <represent>

Describers <observe and envisage> Phenomena


We speak of the describer in the triangle as an animal or a machine. Strictly, it that actor’s cognitive processes – its ability to observe and envisage, and create and use descriptions at the apex of the triangle.

In short, to create a description is to encode a model that represents some feature(s) of a phenomenon. To use a description is to decode it, then use it to respond to or manipulate whatever is described.

Describers <observe and envisage> phenomena

Again, describers are animals or machines that can encode and decode descriptive models of phenomena. And a phenomenon is anything that can be observed or envisaged, including descriptions and describers. Note that both describers and descriptions are physical phenomena (matter/energy structures) and so, are describable.




<create and use>   <represent>

Describers <observe and envisage> Phenomena


Describers may observe a house, which exists in a physical form. Abstract descriptions of that house also exist in physical forms – in minds and records, in messages and memories.


Describers may observe the largest known prime number, or least part of it. It exists only in records, because it is far too large for a human to remember. The abstract description of “the “largest known prime” also exists in a physical form – in minds and in records.


Describers can envisage a thing that is purely fantastic, such as a flying elephant. To envisage it is to describe it in some physical form.


Describers can envisage a thing that might exist in the future, such as the next prime number beyond today’s largest known prime. Again, to envisage it is to describe it in some physical form – in mind or writing.


In short, observing and envisaging are processes that create, use and compare descriptions.


On human cognition (thinking)

This section observes and envisages what thinking is. You may see this section as a collection of unsubstantiated claims. I hope you will see it also as insightful about what cognition is.


In the triangle, suppose we replace the describer (animal or machine) by cognition. The proposal here is that observing and envisaging is creating and using descriptions.




<creates and uses>   <represent>

Cognition <observes and envisages> Phenomena


Below, distilled from chapter 7b, are some staging posts in the evolution of cognition.


1.     Molecular memory: organisms sense and respond to molecular structures.

2.     Neural memory: animals remember things they have perceived through their senses

3.     Social interaction: animals use fixed format messages, like alarm calls

4.     Speech: humans encode messages in words

5.     Writing: humans record more complex descriptions in a persistent form.

6.     Science: humans refine how to form theories, predict outcomes and test them.

7.     Machine learning: humans create machines that abstract descriptions from phenomena.


Human and computer actors do process data differently. In humans, memories and messages take different physical forms; and a memory, being encapsulated in a brain, is inaccessible to others (unless translated into a message). By contrast, in software, memories and messages are merely varieties of digital data storage; both can read/written by many software components.


Note also that software can read a memory without creating one. By contrast, to read a human a memory is also to create a new one. (Which is one reason why human memory is less reliable, as in “false memory syndrome”.)


Nevertheless, both human and computer actors can be reasonably be thought of as writing and reading – creating and using memories and messages. Consider the evolution of intelligence, which is discussed at length in the next chapter.



The psycho-biological view of thinking is called “cognitive embodiment”, meaning that the brain is inseparable from the rest of the nervous system. Thinking is not done in the brain’s grey matter alone.


In primitive animals, sense-response thinking is an end-to-end input-to-output process (cf. a value chain in business architecture). It begins in perception by a bodily sensor, where the state of things is first described by encoding a neural message; and ends with a direction to the body’s motors and organs, where a message is decoded and used in actions to change the state of things.



In higher animals, with memories, perception is a mix of observation (sensing) and envisaging (guessing) at what there is out in reality. Both observation and envisaging are processes that create (encode) and use (decode) descriptions in memory.


Social interaction

More advanced animals have a “theory of mind” – meaning they can attribute mental states to other animals. This is a foundation for social interaction. It is important because it enables animals to interpret and predict the behavior of others. And it enables animals to envisage the consequences of actions they may take.



Every remembered description of the past serves as a type that defines (one member of) a set that may contain more members in future. Envisaging the future involves creating and playing with descriptions possible futures. Consciousness enables an animal to compare descriptions of past, present and possible future phenomena.



In humans, memories are translated into and out of verbal messages for communication. Ashby observed that in thought and communication “coding is ubiquitous”. The multiple translation steps involved in social communication are illustrated in the next chapter.



The proposal here is that thinking is a process that encodes, decodes and compares descriptions. Moreover, in humans at least, this is process is not entirely internal.


I observe my own thinking process is partly externalised via writing. It involves a circular feedback loop in which I encode some half-baked thoughts in written words, then read back what has been written, test it for consistency and coherence, and correct or clarify it.  Writing makes it possible to document a description of the world that is more complete, consistent and coherent than any I can hold in mind.


Consider writing and reading as a circular feedback loop. Moving data from the brain, to the hand, to the keyboard, to the screen, to the eyes and back to the brain, is a succession of processes that translate data encoded in one form to data encoded in another.


In short, observing and envisaging is creating and use descriptions, and cognition is a process that encodes, decodes and compares descriptions, both internal and external.

Principles of description

Implications of what is proposed in this chapter are pursued in the following chapters. This section is a brief abstract of principles to be explained and explored.

The evolution of intelligence and civilization

“A biological approach to human knowledge naturally gives emphasis to the pragmatist view that theories [descriptions of reality] function as instruments of survival.” Stanford Encyclopedia of Philosophy


Chapter 7b looks at description from the viewpoint of Darwinian biology. It starts with the idea that knowledge is a biological phenomenon. It describes the emergence of human intelligence and civilization from the biological evolution of animals, with reference to symbolic languages and the sharing of knowledge in writing. It discusses the following three principles.


Knowledge and description evolved in biological organisms

The "second order cyberneticians" claimed that

·       knowledge is a biological phenomenon (Maturana, 1970), that

·       each individual constructs his or her own "reality" (Foerster, 1973) and that

·       knowledge "fits" but does not "match" the world of experience (von Glasersfeld, 1987).

Stuart A. Umpleby (1994) The Cybernetics of Conceptual Systems. p. 3.


Consciousness enables us to compare the past, present and future

Consciousness is a process that, among other things, enables us to compare descriptions of past, present and possible future phenomena.

On description as typification

Chapter7c may be seen as foundational to how we describe things. Alternatively, you can see it as an academic aside of interest to those with a mathematical bent. This chapter declares three principles (below) related to the creation and use of types, and discusses fuzziness in how well real-world things and phenomena instantiate types.


To describe a thing is to typify it in terms of types already understood

“No statement which refers to a ‘reality’ transcending the limits of all sense-experience can possibly have any literal significance” Chapter 1 of “Language truth and logic” A J Ayer.



Every type is a description

There are several set and type theories. In our type theory, a type is an intensional definition; it is a description of (a member of) a set.


Type name

Type elaboration

“Even number”

a number divisible by two”.

“Triangle corner”

“an angle between the two lines in a corner of a triangle”


“an animal with feathers and a beak”

“Bird of prey”

“a bird which feeds on other animals”


“a bird of prey with a wide wing span”


Together, the type name and elaboration make an intensional definition or predicate statement of the form illustrated below.


Intensional definition pattern

Predicate statement type

Intensional definition example

Predicate statement instance

A thing

of the named type

is a thing of a more general type

with these particular features.

A thing

of the even number type

is a number

which is divisible by two.


Every description is a type

“The fact is that one cannot in language point to an object without describing it… And in describing a situation, one is not merely ‘registering’ a sense-content; one is classifying it in some way or other, and this means going beyond what is immediately given.” Chapter 5 of “Language truth and logic” A J Ayer.


A description is not categorical – it cannot pin down a single object – since it applies equally to any object in universe that shares the same description. E.g. Physicists say there is nothing in their description of the universe that prevents parallel universes from existing. Think of any particular thing; a molecule, a game of chess, a galaxy, whatever. Write down a description of it. Perhaps the thing you have described is unique. But there is nothing to prevent your description being realized in more than one particular thing. To describe one thing is to create a type to which other things might conform.

Communicating information

Chapter 7d outlines some information and communication theory. It features a WKID hierarchy, and the principle that in symbolic communication, coding is ubiquitous. It goes on to describe how message senders communicate, and so share knowledge, with message receivers. It discusses the following two principles.


A description is meaningful to an actor only in the process of creating or using it

The table below helps us to discuss a social system of communicating actors.





the ability to apply knowledge in new situations.


information that is accurate enough to be useful.


meaning created/encoded or found/decoded in data by an actor.


a structure of matter/energy in which information has been created/encoded or found/decoded


Coding is ubiquitous in the creation, sharing and use of symbolic descriptions

A brain can encode some information in the structure of its memory, which becomes useful later, when the brain decodes it, by reversing the encoding process. Similarly, a brain can direct the mouth to encode some information in spoken words, which become useful when a receiver hears and decodes the message (using the language it was encoded in). The information/meaning in the structure of memory or message exists in the processes of encoding and decoding it.


Ashby observed that coding is ubiquitous in thought and communication. To create a description is to encode a model that represents some feature(s) of a phenomenon. To use a description is to decode it, then use it to respond to or manipulate whatever is described.

How we share knowledge and verify truth

Chapter 7e discusses how we clarify information by reducing noise and ambiguity, and verify the truth, or at least the usefulness, of information by and empirical, logical and social means. In doing so, it rejects extreme interpretations of relativism and perspectivism.


We share knowledge by verifying descriptions we share

We clarify the information in a description by reducing noise and ambiguity, and verify the truth of information by empirical, logical and social means.


In short, the principles of description introduced above are.

·       Knowledge and description evolved in biological organisms

·       A good regulator has a description of the target regulates

·       Consciousness is a process that enables us to compare the past, present and future.

·       To describe a thing is to typify it in terms of types already understood

·       Every type is a description

·       Every description is a type

·       A description is meaningful to an actor only in the process of creating or using it

·       Coding is ubiquitous in the creation, use and sharing of symbolic descriptions

·       We share knowledge by verifying descriptions we share


Many copies of a description can be created and used. If all copies are deleted then the description disappears from the cosmos. In other words, there is no ethereal description aside from what exists in one or more copies of it.


Chapter 8 summarizes some implications of these principles for

·       system architecture – as defined in ISO/IEC 42010.

·       semiotics – notably Peirce and Popper.

·       philosophy – including the problem of universals

·       mathematics – did numbers exist before life?

Postscript: on enterprise architecture

Enterprise architecture is concerned with regular business activity systems in which business roles and processes create and use business data. The architects observe baseline business systems and envisage target business systems They create and use architectural descriptions of those systems.


Enterprise architecture

Abstract systems

<create and use>         <represent>

Architects <observe and envisage> Physical systems


Architecting is a process that encodes, decodes and compares descriptions of baseline and target systems. At design time, general entity types like customer, event types like order and payment, and process types like billing are defined. These types are related to each other in logical data structures and process structures. E.g. order value = order amount * unit price.


At run time, physical systems consume inputs that carry information about entities and events, record those entities and events with attribute values, and use remembered information to decide how to respond to events. In all these ways, an enterprise can be seen as imitating what comes naturally to animals.