pask, g.- the architectural relevance of cybernetics (article-1969)

3
7/24/2019 Pask, G.- The Architectural Relevance of Cybernetics (Article-1969) http://slidepdf.com/reader/full/pask-g-the-architectural-relevance-of-cybernetics-article-1969 1/3 sense as parrs of larger syste ms th a t in clude human com?oocnts and the :1rchitect is pnm:iril y E  ARCHITECTURAL concerned with these larger systems; t h> (not just the bricks and mo rtar pare) a rc what architects design. f shall dub this nori on architectural 'mutualism' meaning mu tu:tlism OF ordon ask is easy to argue char cybernetics is relevant co same way that it relevant a host of ocher professions; medicine, or law. PERT programming, for , is unequivocally a 'cybernetic' and it is commonly employed in nstructio n scheduling. Computer assisted is a 'cybernetic' method and there are eral instances of ics application to architcctute, example, the \Xi' SCC's planning scheme in the designer uses a graphic display to of scrucrural modules a grid and in wh i ch the computer s um m arizes cost eff o rc consequences of a proposed ) . Of these cases the first (PERT is a valuable but quite trivial o n of cybernetics; the second is likely have a far-reaching iniluencc upon But neither of th em more than a superficial bond between and archireccurc. f we leave the then architects di\"c into a i c bag of cricks and drnw out chose seem tC be appropriate. Thac is a perfectly thing co do, of course. But cybernetics architecture rcall}' enjoy a much more relatio nship; they share a common l osophy of architecture in the sense that Beer has shown ic to be the philosophy The argument rests upon the idea that . and foremost system designers have been forced. over the last roo yl·~rs r so, c o cake an increasing interest in the syste::n of development, communication and problems were coped with as cropped up, but for some rime it has been, vidcnc that an underpinning and unifying theor}' s required. Cybernetics is a discipline which the bill insofar as the abstract concepts of (and, where appropriate, identi fied with architectural systems), co form a lhtory architectural cybernetics, che cybernetic theory architcccurc). 1 roots! or before the early 1 Soos 'pure' architecture as an abstraction from the att of Its rules were essentially condensed o f what could be observed by looking c builders working on a site, and by looking t buildings constructed during different periods nd in different places. Architects added a modicum of engineering practice and of hiscorical r aeschecic sensibility to their discipline and created new structures with stability and sf.ylt. On the whole, their structures were judged, within 'pure' archicecturc, according ro these canons. fa·cn in those days, of course, architects were asked to solve problems entail ing the regulation and accommodation of human beings; hence ro design sysrcms. But, in a sense, their brief was quite narrow. The problems could all l C soh·ed by the judicious applicarion of pure ~rc u r a l rules. The form of the artefact ( house college or theatre) was largely determined by the quite rigid codes of architecture ( dictating, for example, ics acceptable whole part relationships) and by the conventions of society or the individual practitioner. Speaking 1cchoically there were well accepted communica tion media for conveying instructions, direccives : d ideas (style manuals and so on). Further, there was a / Jt/ala11g11agt for talking about these instructions, directives and ideas, for comparing chem, criticizing chem and e\·aluating them (as in statements of stability or style). Indeed, when interpreted, the body of metalinguistic statements formed the theory of pure architecture. Consequently, architects did not need co sec themse lves as systems designers, even though they designed systems, and the evidence suggests that they did not do so. 3 Instead the professional image was chat of a sophisticated house, college or theatre builder. In the course of the Victorian era new techniques were developed too rapidly to be assimilated inco pure architecture and new problems were posed and could no longer be sol ved by applying the rules of pure architecture, for example, make a 'railway sration' or make a 'great exhibition'. The sol uti on to such (in those days) ou tlandish problems clearly depends upon seeing rhe required building as a pare of the ecosystem of a human society. Of co u rse the problems were solved and the novel t e c h n i q u e ~ were mustered for this purpose (Temple Meads, the Tropical House at Kew, the Crystal Palace). To my own casce the solutions are exceptionally beautiful. 1 Nevertheless, they arc individual and idiosyncratic solutions because, in the nc\\ ' context, there was no way of carrying on a general and critical discussion. Let us be clear about this point. There obviously 11•a1 a great deal of discussion over I. K. Brunel, D. Burton and J. Paxton's use of glass and ironwork; technical discussion and aesthetic discussion. But nobody seems to have; appreciated the full significance of their scructures in the context of rhe arch i tec tural potentialities of the age, i.e. as examples of .J'Slt111 design. The reason is fairly obvious. \ Xfhercas the pure architecture of the ea r l y 1800s had a metalanguage, albeit a restrictive one which discouraged innovation, the new (augmented) architecLUre had nor yet developed on e . Another way of putting it is to say there was no theory of the new architecture.• Architectural sub-theories Tn place of a general theory there were su b thcorics dealing with isolated facets of the field; for example, theories of materials, of symmetry, of human commitment and responsibility, of craftsmanship and the like. But (it is probably fair to say) these sub-t heories developed more or less independently during the late 1800s. Naturally enough, each sub-theory fostered n certain sort of building or a certain sort of socio-architectural dogma; for example, futurism. I Tow evcr, the point of immediate interest is t hat many of the sub-theories were system orientated; although they antic ipat ed the invention of the word they were, in an embryonic sense, 'cybernetic' theories and the thinking behind them made a valuable contribution to the development of cybernccics as a formal science. Archi tectural functionalism and mutualism A structure exists chieAy to perform certain functi ons, fo r example, to shelter its occupants or to provide them with services. Ac this level, a 'functional' building is co n trasted with a 'decorative' building; it is an austere structure, stripped of excrescences. But, the concept of functionalism can be usefully refined in an humanistic direction. The functions, after all, are performed or hu m an beings or human societies. It follows that a building cannot he viewed simply in isolation. It is only meaningful as a human environment. It perpetually interacts with its inhabitants, on the one hand serving chem and on the other hand controlling their behaviour. Io other words structures make between structures and men or so cieties . One consequence o f functi o nali s m a n d mutualism is a shift o f emphasis t0 wards the form (rather rhan rhc material constituti l)n) of structures; materials and metho ds c o mi; mm prominence quite late in che design process. Another consequence is chat archit c ccs arc required to design d •11an1ic rather t han l  t u entities. Clearly, the human part of th e sy stem is dynamic. But it is equally true ( though less obvious) that the srrucrural parr muse bc as continuall; regulating its human inh abitants. Architectural holism Once a rudimcntarv version of th e fun criuna mutualistic hypothesis has been accepted , rhc integrity of any single system is questiona ble ~ o s e human jsuucrural systems rely up on o the r systems to which they are coupled via th e human componenrs. By hypo the sis, rhcr c a re organizational wholes which cannot be mean ingfully dissected into pares . Holism is of several types: a A functionally interpreted building can only be usefullv considered in the co nt e xt of a cit\ (notice th~t rhe city is also functi o nally imer~ pretcd and, as a result, is a dynamic entit y). b A (functionally interpreted) s tructure , e i the r a building or an entire city, can o nly be meaningfully conceived in the co nrcxt of ics temporal extension, i.e. its growth and development. c A (functionally inrerprcccd) structure exists as part o an intention, i.e. ;IS on e pr o d uct of a plan. If (assumed dogma) man should be aware of bis natural surroundings, then buildings should be wedded to o r arise from these sn rrNtndings (\Xfright's organic Lhcsb). It is a corollary of a, b and c that the stru ct u re of a city is not just the carapace of society. On the contrary, its structure acts as a symbo lic control programme on a par \\ ich the ritual constraints which arc known to regulat e the behaviour of various tribes and which ren d er rh ' V ery similar comments apply to engineering, since engineers, llke architects, prescribe artefacts. Surely, als some engineers make use of a cybernetic theory. But the requirement Is not so ubiquitous In engineering: nor Is the Impact of cybernetics so great because a creditable body of engineering theory, a predictive and explanatory theory existed long before the cybernetic concepts came along as daring Innovations. Moreover whilst all architects design systems that Interact closely with human beings and societies, most engineers (there are obvious exceptions) are not forced to do so . Human Interaction is a major source of difficulties which can only be overcome by cybernetic thinking. Th e choice of a historical origin Is somewhat arbitrary and depends upon the author's emphasis. For example, Al exander, preoccupied with the logic of form, traces essentially cybernetic concepts back to Lodoll and Laugler. In the present article I am anxious to follow the pragmatic development of cybernetic Ideas and to see them emerging In the history of modern architecture. •There are two Important sorts of exception: (i) Architects of genius, with a breadth of vision that impels them to see th i ngs in a systemic and Inter- disciplinary fashion. T hey have existed over the Sir Christopher Wren and Sir John Soane, for example (ii) Men llke John Nash, whose talents l ay In conceiving an urban development as a functional and aesthetic whole. But, within the tenets of the early 1800s such men are probably 'or gani zers with a vision', rather than 'architects'. I have chosen these examples partly because they are well known In the textbooks but mainly because I am Impressed by their systemi c qualities and the way In which they convey their designer's purpose to the occupant. Two of them still exist. I just recall the Palac Even In Its tawdry reincarnation It was a remarkable structure. Since It was one of the first Instances of a prefabricated building It also counts as a piece of syst em design at the engineering level. •Lack of an adequate metalanguage was not the only factor. As Prof. Nicolaus Pevsner points out the engine and the artists pursued divergent paths of developmen more or less in conflict with one another and this accounted for at least some of the architectural Idiosyncrasy. However, II a metalanguage h d existed, then the synthesis of the present century could have been achieved mJch earlier.

Upload: theedarklord

Post on 21-Feb-2018

218 views

Category:

Documents


1 download

TRANSCRIPT

Page 1: Pask, G.- The Architectural Relevance of Cybernetics (Article-1969)

7/24/2019 Pask, G.- The Architectural Relevance of Cybernetics (Article-1969)

http://slidepdf.com/reader/full/pask-g-the-architectural-relevance-of-cybernetics-article-1969 1/3

sense as parrs

of

larger syste ms that include

human com?oocnts

and

the :1rchitect is pnm:irily

E

 ARCHITECTURAL

concerned with these

larger

systems; th>

(not

just the

bricks

and

mo

rtar

pare) a rc

what

architects design. f shall dub this norion

architectural

'mutualism'

meaning mu tu:tlism

OF

ordon ask

is easy to argue char cybernetics is relevant

co

same

way

that

it relevant

a host

of

ocher professions; medicine,

or

law. PERT

programming, for

, is unequivocally a 'cybernetic'

and it is

commonly

employed in

nstruction scheduling.

Computer

assisted

is

a

'cybernetic'

method and

there are

e r a l instances of ics application to architcctute,

example, the \Xi'SCC's

planning

scheme in

the designer uses a

graphic

display

to

of

scrucrural modules

a grid and in

wh

ich the computer sum marizes

cost efforc consequences of a proposed

). Of these cases the first (PERT

is

a valuable

but quite

trivial

on

of

cybernetics; the second is

likely

have a far-reaching iniluencc

upon

But

neither

of

them

more than

a superficial

bond

between

and

archireccurc.

f

we leave the

then

archite cts di\"c into a

ic

bag

of cricks

and drnw

out chose

seem tC be

appropriate.

Thac

is

a perfectly

thing

co

do, of

course.

But cybernetics

architecture rcall}'

enjoy

a

much

more

relatio

nship;

they

share

a

common

losophy

of architecture

in the sense that

Beer has shown ic to be the philosophy

The

argument

rests upon

the

idea that .

and

foremost

system

designers

have been forced.

over

the last

roo y l · ~ r s

r so, co cake

an

increasing interest in

the

syste::n

of development, communication and

problems were

coped

with as

cropped up, but for

some

rime it has been,

vidcnc

that

an

underpinning and

unifying theor}'

s

required. Cybernetics is a discipline

which

the bill insofar as the abstract concepts of

(and,

where appropriate,

identi

fied with

architectural systems), co

form

a

lhtory

architectural cybernetics, che cybernetic

theory

architcccurc).

1

roots!

or before the early

1

Soos 'pure' architecture

as

an

abstraction

from

the

att of

Its rules were essentially

condensed

of what

could

be

observed

by

looking

c builders working on a site, and by

looking

t

buildings

constructed during

different periods

nd in different places. Architects added a

modicum

of engineering

practice and of hiscorical

r aeschecic sensibility

to their

discipline

and

created new

structures with

stability and sf.ylt.

On

the whole,

their structures were

judged,

within 'pure' archicecturc,

according

ro these

canons.

fa·cn in those days, of

course,

architects

were

asked

to

solve problems entail ing the

regulation and

accommodation of

human beings;

hence

ro

design

sysrcms.

But,

in a sense,

their

brief was quite narrow. The problems could all

l C

soh·ed by

the

judicious applicarion

of pure

~ r c u r a l rules. The form

of

the artefact

(house college

or

theatre)

was

largely determined

by the quite rigid codes of

architecture

(dictating,

for

example, ics acceptable whole part

relationships) and by the

conventions of

society

or the individual practitioner. Speaking

1cchoically there were well accepted

communica

tion

media for

conveying

instructions, direccives

: d

ideas (style manua ls

and

so

on). Further,

there was a

/ Jt/ala11g11agt for talking

about these

instructions, directives and ideas,

for comparing

chem, criticizing chem and e\·aluating them (as

in statements

of

stability

or

style).

Indeed,

when

interpreted, the

body

of

metalinguistic statements

formed the theory

of

pure architecture.

Consequently, architects did not need

co

sec

themse lves as systems designers, even though

they designed systems,

and the

evidence

suggests

that

they did

not

do so.

3

Instead the professional

image was chat

of

a sophisticated

house,

college

or

theatre builder.

In the

course of

the Victorian

era

new

techniques were developed too rapidly to be

assimilated inco pure

architecture

and

new

problems were posed and could no longer be

so

lved by

applying

the rules

of

pure architecture,

for

example,

make

a 'railway sration'

or make

a

'great

exhibition'.

The soluti on to such (in those

days) ou tla

ndish

problems clearly

depends upon

seeing

rhe required building as a pare of the

ecosystem

of

a

human

society. Of

cou

r

se

the

problems were solved and the novel t e c h n i q u e ~

were mustered

for

this

purpose (Temple

Meads,

the Tropical House at Kew,

the Crystal Palace).

To my own casce the

solutions are

exceptionally

beautiful.

1

Nevertheless, they

arc

individual and

idiosyncratic

solutions

because, in the

nc\\

'

context,

there was

no

way

of

carrying on a

general

and

critical discussion.

Let

us be

clear

about

this point.

There

obviously 11•a1

a great

deal of discussion over I. K. Brunel, D. Burton

and

J. Paxton's

use

of

glass and ironwork;

technical discussion

and

aesthetic discussion.

But nobody seems to have; appreciated the full

significance

of their

scructures in

the context

of

rhe

arch

itectural potentialities

of

the age,

i.e. as examples

of .J'Slt111

design.

The

reason is

fairly obvious. \Xfhercas the pure

architecture

of

the

ear

ly 1800s

had

a metalanguage,

albeit

a restrictive

one which

discouraged

innovation,

the new

(augmente

d) architecLUre

had

nor yet

developed

one. Another way of putting it is to

say there was

no theory of the

new

architecture.•

Architectural sub-theories

Tn place of a

general

theory there

were

sub

thcorics

dealing with

isolated facets

of the

field;

for example, theories

of

materials,

of symmetry,

of human commitment

and responsibility,

of

craftsmanship

and

the like. But (it is

probably

fair

to

say) these

sub-t

heories

developed more or

less independently during the late 1800s.

Naturally

enough,

each

sub-theory

fostered n

certain sort of

buildin

g or a certain sort of

socio-architectural dogma;

for

example, futurism.

I Towevcr, the point of immediate interest is that

many of

the

sub-theories were

system orientated;

although

they anticipated the

invention

of

the

word they were, in an

embryonic

sense,

'cybernetic'

theories

and the thinking behind

them made a valuable contribution to the

development of

cybernccics as a fo rmal science.

Architectural

functionalism

and mutualism

A structure exists chieAy to perform certain

functi

ons,

for example, to

shelter

its

occupants

or to provide them

with

services. Ac this level,

a

'functional' building

is

con

trasted with a

'decorative' building; it is an austere structure,

stripped of

excrescences.

Bu

t, the

concept of

functionalism can be usefully refined in an

humanistic direction. The functions,

after

all,

are performed

or

hum

an beings

or human

societies. It follows

that

a

building

cannot he

viewed simply in isolation.

It

is only meaningful

as a

human environment.

It

perpetually interacts

with

its inhabitants, on the one

hand

serving

chem and on the other hand controlling their

behaviour.

Io

other

words structures

make

between structures

and men or

societies.

One consequence o f functionalism a nd

mutualism is a shift o f emphasis t0wards the

form (rather

rhan rhc material constitutil)n)

of

structures;

materials

and meth

ods com

i; mm

prominence

quite late in che design

proc

ess.

Another consequence is chat

archit

c

ccs

arc

required

to design d •11an1ic rather t

han l

  t

u

entities. Clearly, the human part of the sys

tem

is

dynamic.

But

it

is

equally

true

(

thou

gh

less

obvious) that the srrucrural parr muse bc

as continuall;•

regulating

its human

inh

abitants.

Architectural

holism

Once a

rudimcntarv

version of the fun criuna

mutualistic hypothesis has been accepted, rhc

integrity of

any single syste m is

questiona

ble

~ o s e human jsuucrural systems rely upon other

systems to which they are coupled via the

human componenrs. By hypothesis, rhcrc are

organizational wholes which cannot be mean

ingfully dissected into pares.

Holism

is

of

several types:

a

A

functionally

interpreted

building

can

only

be usefullv considered in the co ntext of a cit\

(notice

t h ~ t

rhe city

is

also functionally i m e r ~

pretcd

and, as a result, is a

dynamic

entity).

b

A

(functionally interpreted) s

tructur

e, either

a building or an entire city, can o nly be

meaningfully conceived in the co nrcxt of ics

temporal

extension, i.e. its

growth and

development.

c A

(functionally inrerprcccd) structure

exists as part

o

an intention, i.e.

;IS

one pro duct

of

a

plan.

If (assumed dogma)

man

should be

awar

e

of

bis

natural surroundings, then

buildings

should be wedded to o r arise

from

these

sn rrNtndings (\Xfright's

organic

Lhcsb).

It is a corollary of a, b and c that the structure

of a city is not just the carapace of society. On

the

contrary,

its

structure

acts as a

symb

olic

control

programme

on a par \\•ich the ritual

constraints which

arc known

to

regulat

e the

behaviour of various tribes and which render rh

' Very similar comments apply to engineering, since

engineers, llke architects, prescribe artefacts. Surely,

als

some engineers make use of a cybernetic theory. But

the requirement Is

not

so ubiquitous In engineering:

nor

Is the Impact of cybernetics

so

great because a

creditable body of engineering theory, a predictive and

explanatory theor

y

existed

long

before the cybernetic

concepts came along as daring Innovations. Moreover

whilst

all architects design systems that Interact closely

with human beings and societies,

most

engineers

(there are obvious exceptions) are

not

forced to do

so

.

Human Interaction is a

major

source of difficulties which

can

only

be overcome by cybernetic thinking.

• The choice of a historical origin Is somewhat arbitrary

and

depends upon the

author's

emphasis. For example,

Al exander, preoccupied with the

logic

of form, traces

essentially cybernetic concepts back

to

Lodoll and

Laugler.

In

the present article I am anxious to follow the

pragmatic development of cybernetic Ideas and

to

see

them emerging

In

the history of modern architecture.

•There are

two

Important

sorts

of exception:

(i) Architects

of genius,

with

a breadth of vision

that

impels them to

see th

i

ngs

in a systemic and Inter-

disciplinary fashion. T hey have existed over the

Sir Christopher Wren

and

Sir

John

Soane,

for

example

(ii)

Men llke John Nash, whose talents l

ay

In conceiving

an urban development as a functional and aesthetic

whole. But, within the tenets of the early

1800s such

men are probably

'or

ganizers with a vision', rather than

'architects'.

• I have chosen these examples partly because they are

well known In the textbooks

but

mainly because I am

Impressed by their systemic qualities and the way In

which they convey their designer's purpose

to

the

occupant.

Two

of them

still

exist. I

just

recall

the

Palac

Even

In Its tawdry reincarnation

It

was a remarkable

structure. Since It was

one of

the

first

Instances of

a

prefabricated building

It

also counts as a piece of

system design at the engineering leve

l.

•Lack of an adequate metalanguage was

not

the

only

factor. As Prof. Nicolaus Pevsner

points

out

the

engine

and the

artists

pursued divergent paths of developmen

more

or less

in

conflict

with one

another and this

accounted for at least some of the architectural

Idiosyncrasy. However,

II

a metalanguage

h d

existed,

then

the

synthesis of

the

present century could have

been achieved mJch earlier.

Page 2: Pask, G.- The Architectural Relevance of Cybernetics (Article-1969)

7/24/2019 Pask, G.- The Architectural Relevance of Cybernetics (Article-1969)

http://slidepdf.com/reader/full/pask-g-the-architectural-relevance-of-cybernetics-article-1969 2/3

behaviour homeostatic rather than divergent.

Hence, the architect is responsible for building

conventions and shaping the development

of

traditions (rhis comment simply elevates the idea

that a building concro ls its inhabitants to a

higher level

of

organization).

Evolutionary ideas in

architecture

Systems, notably cities, grow and develop and,

in

general

ev

olve. Clearly, this co ncept is

contingent

upon

the functionalist/mutualisc

hypothesis (without which

it

is difficult to see in

what sense the system itself

doe.r

grow) though the

dependency is often unstated. n immediate

practical consequence

of

the evolutionary

point

of view is that archfrectural designs should have

rules

for

evolution built into them i their

growth is to be healthy rather than cancerous.

In other words, a responsible architect must be

concerned with evolutionary properties; he

cannot

merely stand back and observe evolution

as something that happens to his structures. The

cvolu,tionary thesis is closely related to holism,

type

c,

but it is a carefully specialized version

of

c as manifest in the work

of

the Japanese.

Svm b

ol i

c en vironmen ts in arch ite

ctu

re

Many human activities are symbolic in character.

Us ing visual, verbal

or

tactile symbols, man

talks with his surroundings. These consist

in

other men, information systems such as

libraries, computers or works of art and also,

of

course, the

st r

u

ct

ures

around

him.

Buildings have always been classified s works of

art. The

novel

sub-theory is that structures

may be designed (as well s intuited) to foster a

productive and pleasurable dialogue. This way

of thinking is most clearly manifest in con

nection with the literary art forms, notably

surrealism

wh ich relics upon a variety (novelty)

produc

i

ng

juxtaposition of releas.ers and

supernormal

stimuli (evoking inbuilt emotive

responses)

within

a thematic matrix.

At the

architectuxal level, this type

of

design appears in

t.he vegetable surrealism of some of

the Art

Nouveau. But it reaches maturity in Gaudi s

work, especially the

Parque

Guell (right) which, at a

symbolic level, is one

of

the most cybernetic

structures in existence. As you explore the

piece, statements are made

in

terms

of

releasers,

your

exploration is

guided

by specially contrived

feedback,

and

variety (surprise value) is introduced

at

appropriate points to make you explore.

It

is interesting

that

Gaucli s

work

is

often

&ontrasted with functionalism. Systemically

it

is functionalism pure and simple, though

it

is

aimed

at

satisfying

only the

symbolic

and

informational needs of man.

6

he machinery of architectural production

Just as a functionally interpreted building

so

also the construction of

this building is a system. The new techniques

in the last century and the general

of production facilities led to

b-theories concerned with the achievement of

most important centred around the

) and these, in turn, restricted the forms

could

be

pr

o

duc

ed.

widening

brief

a result of these, essen tially cybernetic,

b-theoretical developments, many architects

co

design systems

but,

on the whole they

expected to

design buildings. To a l

arge

chis is still (quite reasonably) true.

All

the

there is a sense

in which

the brief given

to

architect has widened during the last decades.

In

part

this is

due to

a spate of problems for

no conventional solutio n exists (structures

industry,

entertainment, the use of oceans, etc.).

rbc architect is in much the same position

Victori

an

predecessor when asked to build

In part, however, the restraints

been relaxed because of the greater pre

of

system orientated thinking

amongst

It is, nowadays,

to enter the design process much

even for a convention  l project. For

J   n a/AO

Guadi s

Parq11e

G11ell- 011e o} lb

»l(JS/

r,;•bernetic

.rtrucl11ru

i

exiflt11ce.

Photographs: Leopoldo Pomes

Joan Prats and Joaquim Gomts

Page 3: Pask, G.- The Architectural Relevance of Cybernetics (Article-1969)

7/24/2019 Pask, G.- The Architectural Relevance of Cybernetics (Article-1969)

http://slidepdf.com/reader/full/pask-g-the-architectural-relevance-of-cybernetics-article-1969 3/3

e, it is commonplace to design

at least co plan) ciries as a w hole with

vision for their evolution. A University

not be conceived as a set of bu ildings

a cour tyard with li ving ac

co

mmodation

ecture thcarre. The educational system

ht in certain circumst:1nces, be spatially

ocalized. In any case,

s a rc positively encouraged w

ate trends such as the development of

ona l technology and to prm·ide for their

ct upon whateve r structure is erected. By

of

this the a rchi tect quite o ften comes into

picture at the time when a higher

onal system is being contemplated,

o

ut

commitment co whether o r noc

it

is

l

ed a university.

The

Fun

Palace project, by

Cedric Price, was

an

ear ly

pr

oject of this type in the field of

e

nt and

it is nor difficult to f ind

es in areas ranging from exhibition

ign to factory building.

he point I wish to establish is thar nowadays

is

a de a d for system oriencared thinking

in the past, there was only a more

or

esoteric

desire

fo r

it.

Decause

of

this demand,

s w o

rth

while collecting the isolated sub

together by forming a generalization

their common con

stituents.

As

we

hav

e

argued,

the comm

on constituents

ar

c

th

e

of concrol, communication and system.

the

generalization is no

more

nor less

abstract cybernetics inccrprctcd as

an

archi tectural theory.

would be

prematur

e to suggest t

hat

the

ssary interpretation :rnd consolidati

on

is

But a creditable start has been made

number

of people; citing only those with

I have personal contact, Christopher

Nicholas

Negr

oponce, many srudcms

ex-students from the AA School of

and from cwcasrle.

t us of t

he

n ew theory

common with the pu re architecture of the

,

cybernetics provides a metalanguage for

cal discussion. But the cybernetic theory is

an

extension of 'pure' architecture.

somewhat earlier, pure archirccturc

descriptive (a taxonomy of buildings

and

prescriptive (as in

the

preparation

plans) bu t it did little to predict or explain.

contra

st, the

cybernetic theory has

an

eciable predictive power.

7

Por example,

development can be modelled as self

system (a formal statement of

ideas in architecture

')

and in these

it

is possible to predict the extent to which

growth

of a ciry

will

be chao tic o r ordered

differentiation.

Even if

the necessary data

is unavailable we can,

at

least,

e and test rational hypothesis. Much the same

ents apply to predictions in which time is

of

primary .imporcancc; for instance, in

cting the influence of spatial and

e constraints upon the s tability of a

onally interpreted) s tructure.

cybernetic theory can also claim some

ory power insofar as

it

is possible to

certain aspects of a rchi tectural design by

al intelligence

computer programs

8

ovided, incidentally, that the program is able

earn about and f rom architects and by

t i n g in

the language

of

architects,

by explo ring plans, material specifications,

ensed versions

of

clients' comments, etc.).

rograms

are

clearly of value in their own

They are potential aids to design; acting

ntell igent extensions of the tool-like

at

the outset. Further, they

r a means for integrating the constructional

(the

'machinery of production') wich the

design process sinct..

it

is quite cas} tt•

the consrrainrs of current technology

part of rbe simulation. l l0\"1."Cver, J

these programs arc of far greater

as evidencing out theoretical

of

what

architecture is about. Insofar

program

can

be

wrirtcn, the cybernetic

Speculations

It seems

Likely

that rapid advances \\ill be made

in a t least live areas guided by

the

cybernetic

rhcory

of

architecture.

i

Va

rious computer-assisted

(or

even

computer-

di r

ccred) desi

gn

procedures will

be

developed into usefu l instruments.

z. Concepts

in ve

ry different disc iplines

(no rably social a

nthr

opology, psychology,

sociology, ecology and economics) will be

un ified with the concep ts

of

architecture to

yield an adequately broad view of such entities

as 'civilizarion', 'city '

or

'educational system'.

3. There will be a proper and systematic

formulation of chc sense in which archi tecture

acts as a social cont ro l (i.e. the germ of an idea,

mentioned under 'Ho lism', will be elaborated).

4. Th e high po int of functionalism is the

concept of a house as a 'machine fo r living in '.

But chc bias is towards a machine that aces as a

tool serving the inhabitant. This no tion will,

I belieYc, be relined inco the conce

pt

of an

cnvironmc

nc

1  ith which the inhabirnnc

cooperates and i which he can externalize his

mental processes, i.e. murualism will be

emp hasized as compared wirh mere

functional ism. For example, the machine for

Jiving in will re

li

eve the inhabitant o f th e need

co store informarion in memory and the need to

perform ca lculations as well as helping out with

more o bvious c hores like garbage disposal and

washing

up

dishes. Further,

it

will elicit his

interest as well as simply answering his enquiries.

5 Gaudi (intentionally

or

not) achieved a

dial

ogue

between his cnvironmenc and its

inhabitants. H e did

so

using physically static

structures ( the dynamic processes depending

upon the movement of people

or

shifts in their

attention).

Th

e dialo

gue

can

be

relined

and

extended with

ch

e a id

of

modern techniques

which allow us co weave the same parre

ro

in

terms of a reactive environment. If,

in

addition,

the

environment

is malleable

and

adaptive the

results can

be

very

potent

indeed. I have

experimented along these lines mysclf

9

but the

work

of Brodey and his group at the

environmental ecol

og

y laboracory is a project

on

a

much

more impressive scale.

As

a

br

oad

statement of what is going on, a computer

controls

th

e v isual

and

tactile

pr

o perties o f

env

ironmental materials (which arc available

in

sufficient

di

vcrsity for

most

architectural

purpo

ses) . These materials contain sensors,

tactile or visual as the case may be , which

return messages to the computer at several

levels of generality. lo the absence of a human

inhabitant, the feedback leads to stabilization

with respect co certain pre-programmed

invariants (for example, that a body of material

shall maintain mechanical stability and occupy

a prescribed value), and to a search process

in

which the mate rial actively looks for signs

of

a

human being in contact with it. If there ls a

human being in the environment, computer,

material and all , engages him in dial

og

ue and ,

within quite wide limits, is able to learn about

and adapt to his beba viour

pattern

. T here is thus

one sense in which the rcaccive environment is a

fo11Jro/ler

and ano ther in which

it

is con

tr

olled

by

its inhabita

nt

s.

A s

imple

cybernetic

de

sign

paradigm

In

th

e c

ontext

of a reactive and adapti ve

environment, architectural design takes place in

several

inte

rdependent stages.

i. Specification of the purpose or

goal

of the

system (with respect

to the

human inhabitants).

It

should be emphasized that che goal

may

be and

nearly always 11•i// be underspecified, 1e. the

architect will 110 more

know

the purpose of the

system than he rtt1//y knows Lhe purpose of a

convenrional house.

I

!is aim is

to pro\

ide a

s

of conscraints that allo\\

for

ccrtam,

presumably

dcsirahk,

modes of evolution.

ii. Choice of the basic environmental matcriah.

iii. SclccLion

of

the invariants "hich arc to be

programmed

into

the system.

Panly

:it this stage

and partly in ii above, the architect determines

what

propcnics

will be relevant in the man

iv. Specification of what the envir

onment

\\'ill

learn about and bow it will adapt.

v. Choice of a plan for adaptation and

development.

n

case the goal of the system is

de

r1pecified (as

in

i) the plan will chiefly consist

m a number of evolutionary principles.

Of course, this paradigm applies co systems

whi ch adapt over rath

er

short time interva ls

(mi nutes

or

hours).

In

contrast, the adaptation

in a project such as the Fun Palace system took

place over much longer cimc intervals (for

instance, an 8-hourly cycle and a weekly cycle

formed

part of

the proposal). Depending upon

the time consrraincs and the degree of

flex

ib ili ty required , it is more

or

less conveni ent

co use a

comput

er (fo r exa

mp

le, the weekly

cycle is

mor

e economica

ll

y

programmed

by a

llcxible o ffice procedure) .But exactly the s

am

e

principles a rc involved.

Urban planning usually extends over time

pe

riods of years o r d ecades

an

d, as

curr

ently

conceived, the plan is qui re an inflexible

spccific.'l

tion

. Ho wever, the a rgument just

pr

esented suggests that

it

need not be inflexible

and that urban development could, perhaps with

advantage, be governed by a process like that

in the dialogue of a reactive environment

(physical contact with chc inhabitants giving

place to ao awareness of their preferences and

predilections ; the inflexible plan to the e nv

ir

on -

mental computing machine).

If

so, the same

design paradigm applies, since in all of the cases

so far considered the p rimary decisions are

systemic in character, i.e. they amount

co

the

deline

ation

or

the

m

od

ification

of

a

control

program. This universality is typical of the

cybernetic approach.

One final manoeuvre will indicate the flavour

of a cybernetic theo ry. Lee us turn the design

paradigm in

up

on irsclf; let us apply

i t

to the

interac

tion be t

ween

th

e

de

sig

ner

and the

system he designs, rather than the interaction

between

th

e s ystem and

th

e

pe

ople

who

inhabit

it. The glove firs, almost perfectly

in

the case

when

the

designer uses a com

puter

as his

assistant. Jn oth

er

words, the relation

'co

ntrollrr

/ r.ontrollecl en riry' is preserved when

these omnibus words arc replaced either by

'designer/sys

tem

being designed' or by

'sys temic env

ironment

/ inhabitants' or by

'urban plan/cicy'. But notice the trick the

des igner is controlling chc construction of

contro

l systems and consequently design is

contro l o control, i.e. the designer docs much

the same job as bis system, b J be operates

at

a

higher level

in

the organizational hierarchy.

Further the design goal is nearly always

und erspecified and the 'controller' is no longe r th

authoritarian appararus which this purely

technical name commonly brings to mind. In

contrast the controller is an

odd

mixture of

catalyst, crutch , memory and arbiter. These, I

be lieve, are the dispositions a designer should

bring to bear upon his wo rk {when he

professionally plays the part of a controller) and

these are the qualities he should

embed

in the

systems (control systems) which be des igns.

10  

0

• Clearly, In other respects, It would be

uncomfortably

prickly to

li

ve In.

1

The

I

mpact of

cybernetics upon architecture i s

considerable Just because the theory does have muc ll

more predicti

ve powe

r than pure arch itectu re had.

Cybernetics

did

relatively li ttle

to

alter the shape of

biochemistry for instance, because although these

concepts are bound up w

ith

everything from entyme

organltatlon to

mol

ecular

biology, the d iscipline

of

biochemistry already had

predlcl/ve

and

explanatory

theory of its own. I made the same

poi

nt for engineering

In an earlier footnote.

I have

the

work

of

N

egroponte s group

(see p. 509-514}

chiefly In mind, though there are other exemplars.

• For example,

the

colloquy

of

moblies project

and the

muslcolour

system.

A comment a c se history and

a

plan In Computer Art

Editor Jalsha Reichardt.

••The

cybernetic

notions

moo ted In this article, are

discussed In

An approach to cybernetics Hutchinson,

1961 (paperback 1968) and, In a lighter vein,

in

My

predictions

for

19

84

 

In

Prospect

Tho

Schweppes

Book

is explanatory.

environment dialogue

of

the

New Generation,

Hutchinson. London,

196

2